CME

A highly active, water- and alcohol-soluble, basic pressor substance is formed when renin and renin-activator interact, for which we suggest the name “angiotonin.”

—Irvine H. Page and O.M. Helmer, 1940.¹

The renin-angiotensin-aldosterone system regulates salt and, in part, water homeostasis, and therefore blood pressure and fluid balance through its actions on the heart, kidneys, and blood vessels.² Drugs that target this system—angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs)—are used primarily to treat hypertension and also to treat chronic kidney disease and heart failure with reduced ejection fraction.

See related editorial

Controlling blood pressure is important, as hypertension increases the risk of myocardial infarction, cerebrovascular events, and progression of chronic kidney disease, which itself is a risk factor for cardiovascular disease. However, the benefit of these drugs is only partly due to their effect on blood pressure. They also reduce proteinuria, which is a graded risk factor for progression of kidney disease as well as morbidity and death from vascular events.³

Despite the benefits of ACE inhibitors and ARBs, concern about their adverse effects—especially hyperkalemia and a decline in renal function—has led to their underuse in patients likely to derive the greatest benefit.³

ACE INHIBITORS AND ARBs

momoniat_aceinbitorsarbspotassiumrenalfunction_f1.jpg

The renin-angiotensin-aldosterone system is activated when hypoperfusion to the glomerular afferent arteriole, reduced sodium delivery to the distal convoluted tubule, or increased sympathetic activity stimulates the renal juxtaglomerular apparatus to produce renin (Figure 1). This leads to a cascade of effects culminating in sodium retention and potassium excretion, thus increasing blood pressure.

ACE inhibitors, as their name indicates, inhibit conversion of angiotensin I to angiotensin II by ACE, resulting in vasodilation of the efferent arteriole and a drop in blood pressure. Inhibition of ACE, a kininase, also results in a rise in kinins. One of these, bradykinin, is associated with some of the side effects of this class of drugs such as cough, which affects 5% to 20% of patients.⁴ Elevation of bradykinin is also believed to account for ACE inhibitor-induced angioedema, an uncommon but potentially serious side effect. Kinins are also associated with desirable effects such as lowering blood pressure, increasing insulin sensitivity, and dilating blood vessels.

ARBs were developed as an alternative for patients unable to tolerate the adverse effects of ACE inhibitors. While ACE inhibitors reduce the activity of angiotensin II at both the AT1 and AT2 receptors, ARBs block only the AT1 receptors, thereby inhibiting their vasoconstricting activity on smooth muscle. ARBs also raise the levels of renin, angiotensin I, and angiotensin II as a result of feedback inhibition. Angiotensin II is associated with release of inflammatory mediators such as tumor necrosis factor alpha, cytokines, and chemokines, the consequences of which are also inhibited by ARBs, further preventing renal fibrosis and scarring from chronic inflammation.³

What is the evidence supporting the use of ACE inhibitors and ARBs?

ACE inhibitors and ARBs, used singly, reduce blood pressure and proteinuria, slow progression of kidney disease, and improve outcomes in patients who have heart failure, diabetes mellitus, or a history of myocardial infarction.^5–11 

While dual blockade with the combination of an ACE inhibitor and an ARB lowers blood pressure and proteinuria to a greater degree than monotherapy, dual blockade has been associated with higher rates of complications, including hyperkalemia.^12–17

RISK FACTORS FOR HYPERKALEMIA

ACE inhibitors and ARBs raise potassium, especially when used in combination. Other risk factors for hyperkalemia include the following—and note that some of them are also indications for ACE inhibitors and ARBs:

Renal insufficiency. The kidneys are responsible for over 90% of potassium removal in healthy individuals,^18,19 and the lower the GFR, the higher the risk of hyperkalemia.^3,20,21

Heart failure

Diabetes mellitus^6,21–23

Endogenous potassium load due to hemolysis, rhabdomyolysis, insulin deficiency, lactic acidosis, or gastrointestinal bleeding

Exogenous potassium load due to dietary consumption or blood products

Other medications, eg, sacubitril-valsartan, aldosterone antagonists, mineralocorticoid receptor antagonists, potassium-sparing diuretics, beta-adrenergic antagonists, nonsteroidal anti-inflammatory drugs, heparin, cyclosporine, trimethoprim, digoxin

Hypertension

Hypoaldosteronism (including type 4 renal tubular acidosis)

Addison disease

Advanced age

Lower body mass index.

Both hypokalemia and hyperkalemia are associated with a higher risk of death,^20,21,24  but in patients with heart failure, the survival benefit from ACE inhibitors, ARBs, and mineralocorticoid receptor antagonists outweighs the risk of hyperkalemia.^25–27 Weir and Rolfe²⁸ concluded that patients with heart failure and chronic kidney disease are at greatest risk of hyperkalemia from renin-angiotensin-aldosterone system inhibition, but the increases in potassium levels are small (about 0.1 to 0.3 mmol/L) and unlikely to be clinically significant.

Hyperkalemia tends to recur. Einhorn et al²⁰ found that nearly half of patients with chronic kidney disease who had an episode of hyperkalemia had 1 or more recurrent episodes within a year.

 

 

ACE INHIBITORS, ARBs, ABD RENAL FUNCTION

Another concern about using ACE inhibitors and ARBs, especially in patients with chronic kidney disease, is that the serum creatinine level tends to rise when starting these drugs,²⁹ although several studies have shown that an acute rise in creatinine may demonstrate that the drug is actually protecting the kidney.^30,31 Hirsch³² described this phenomenon as “prerenal success,” proposing that the decline in GFR is hemodynamic, secondary to a fall in intraglomerular pressure as a result of efferent vasodilation, and therefore should not be reversed.

Schmidt et al,^33,34 in a study in 122,363 patients who began ACE inhibitor or ARB therapy, found that cardiorenal outcomes were worse, with higher rates of end-stage renal disease, myocardial infarction, heart failure, and death, in those in whom creatinine rose by 30% or more since starting treatment. This trend was also seen, to a lesser degree, in those with a smaller increase in creatinine, suggesting that even this group of patients should receive close monitoring.

Whether renin-angiotensin-aldosterone system inhibitors provide a benefit in advanced progressive chronic kidney disease remains unclear.^35–37  The Angiotensin Converting Enzyme Inhibitor (ACEi)/Angiotensin Receptor Blocker (ARB) Withdrawal in Advanced Renal Disease trial (STOP-ACEi),³⁸ currently under way, will provide valuable data to help close this gap in our knowledge. This open-label randomized controlled trial is testing the hypothesis that stopping ACE inhibitor or ARB treatment, or a combination of both, compared with continuing these treatments, will improve or stabilize renal function in patients with progressive stage 4 or 5 chronic kidney disease.

NEED FOR MONITORING

Taken together, the above data suggest close and regular monitoring is required in patients receiving these drugs. However, monitoring tends to be lax.^34,37,39 A 2017 study of adherence to the guidelines for monitoring serum creatinine and potassium after starting an ACE inhibitor or ARB and subsequent discontinuation found that fewer than 10% of patients had follow-up within the recommended 2 weeks after starting these drugs.³⁴ Most patients with a creatinine rise of 30% or more or a potassium level higher than 6.0 mmol/L continued treatment. There was also no evidence of increased monitoring in those deemed at higher risk of these complications.

WHAT DO THE GUIDELINES SUGGEST?

ACE inhibitors and ARBs in chronic kidney disease and hypertension

Target blood pressures vary in guidelines from different organizations.^4,40–45 The 2017 joint guidelines of the American College of Cardiology and American Heart Association (ACC/AHA)⁴⁰ recommend a target blood pressure of 130/80 mm Hg or less in all patients irrespective of the level of proteinuria and whether they have diabetes mellitus, based on several studies.^46–48 In the elderly, other factors such as the risk of hypotension and falls must be taken into consideration in establishing the most appropriate blood pressure target.

In general, a renin-angiotensin-aldosterone system inhibitor is recommended if the patient has diabetes, stage 1, 2, or 3 chronic kidney disease, or proteinuria. For example, the guidelines recommend a renin-angiotensin-aldosterone system inhibitor in diabetic patients with albuminuria.

None of the guidelines recommend routine use of combination therapy.

ACE inhibitors and ARBs in heart failure

The 2017 ACC/AHA and Heart Failure Society of America (HFSA) guidelines for heart failure⁴⁹ recommend an ACE inhibitor or ARB for patients with stage C (symptomatic) heart failure with reduced ejection fraction, in view of the known cardiovascular morbidity and mortality benefits.

The European Society of Cardiology⁵⁰ recommends ACE inhibitors for patients with symptomatic heart failure with reduced ejection fraction, as well as those with asymptomatic left ventricular systolic dysfunction. In patients with stable coronary artery disease, an ACE inhibitor should be considered even with normal left ventricular function.

ARBs should be used as alternatives in those unable to tolerate ACE inhibitors.

Combination therapy should be avoided due to the increased risk of renal impairment and hyperkalemia but may be considered in patients with heart failure and reduced ejection fraction in whom other treatments are unsuitable. These include patients on beta-blockers who cannot tolerate mineralocorticoid receptor antagonists such as spironolactone. Combination therapy should be done only under strict supervision.⁵⁰

 

 

Starting ACE or ARB therapy

Close monitoring of serum potassium is recommended during ACE inhibitor or ARB use. Those at greatest risk of hyperkalemia include elderly patients, those taking other medications associated with hyperkalemia, and diabetic patients, because of their higher risk of renovascular disease.

Caution is advised when starting ACE inhibitor or ARB therapy in these high-risk groups as well as in patients with potassium levels higher than 5.0 mmol/L at baseline, at high risk of prerenal acute kidney injury, with known renal insufficiency, and with previous deterioration in renal function on these medications.^3,41,51

Before starting therapy, ensure that patients are volume-replete and measure baseline serum electrolytes and creatinine.^41,51

The ACC/AHA and HFSA recommend starting at a low dose and titrating upward slowly. If maximal doses are not tolerated, then a lower dose should be maintained.⁴⁹ The European Society of Cardiology guidelines⁵² suggest increasing the dose at no less than every 2 weeks unless in an inpatient setting. Blood testing should be done 7 to 14 days after starting therapy, after any titration in dosage, and every 4 months thereafter.⁵³

The guidelines generally agree that a rise in creatinine of up to 30% and a fall in eGFR of up to 25% is acceptable, with the need for regular monitoring, particularly in high-risk groups.^{40–42,51,52}

What if serum potassium or creatinine rises during treatment?

If hyperkalemia arises or renal function declines by a significant amount, one should first address contributing factors. If no improvement is seen, then the dose of the ACE inhibitor or ARB should be reduced by 50% and blood work repeated in 1 to 2 weeks. If the laboratory values do not return to an acceptable level, reducing the dose further or stopping the drug is advised.

Give dietary advice to all patients with chronic kidney disease being considered for a renin-angiotensin-aldosterone system inhibitor or for an increase in dose with a potassium level higher than 4.5 mmol/L. A low-potassium diet should aim for potassium intake of less than 50 or 75 mmol/day and sodium intake of less than 60 mmol/day for hypertensive patients with chronic kidney disease.

Review the patient’s medications if the baseline potassium level is higher than 5.0 mmol/L. Consider stopping potassium-sparing agents, digoxin, trimethoprim, and nonsteroidal anti-inflammatory drugs. Also think about starting a non–potassium-sparing diuretic as well as sodium bicarbonate to reduce potassium levels. Blood work should be repeated within 2 weeks after these changes.

Do not start a renin-angiotensin-aldosterone system inhibitor, or do not increase the dose, if the potassium level is elevated until measures have been taken to reduce the degree of hyperkalemia.⁵¹

In renal transplant recipients, renin-angiotensin-aldosterone system inhibitors are often preferred to manage hypertension in those who have proteinuria or cardiovascular disease. However, the risk of hyperkalemia is also greater with concomitant use of immunosuppressive drugs such as tacrolimus and cyclosporine. Management of complications should be approached according to guidelines discussed above.⁵¹

Monitor renal function, potassium. The National Institute for Health and Care Excellence guideline⁵⁴ advocates that baseline renal function testing should be followed by repeat blood testing 1 to 2 weeks after starting renin-angiotensin-aldosterone system inhibitors in patients with ischemic heart disease. The advice is similar when starting therapy in patients with chronic heart failure, emphasizing the need to monitor after each dose increment and to use clinical judgment when deciding to start treatment. The AHA advises caution in patients with renal insufficiency or a potassium level above 5.0 mmol/L.⁴⁹

Sick day rules. The National Institute for Health and Care Excellence encourages discussing “sick day rules” with patients starting renin-angiotensin-aldosterone system inhibitors. This means patients should be advised to temporarily stop taking nephrotoxic medications, including over-the-counter nonsteroidal anti-inflammatory drugs, in any potential state of illness or dehydration, such as diarrhea and vomiting. There is, however, little evidence that this advice can actually reduce the incidence of acute kidney injury.^55,56

momoniat_aceinbitorsarbspotassiumrenalfunction_t1.jpg

Potassium-lowering agents. Evidence is emerging to support the use of potassium-lowering agents to manage hyperkalemia. New compounds such as patiromer and zirconium cyclosilicate bind potassium in the gastrointestinal tract so it is excreted fecally. Meaney et al⁵⁶ performed a systematic review and meta-analysis of current phase 2 and 3 trials and concluded that these drugs lowered serum potassium levels by up to 0.70 mmol/L. There may be a significant role for these novel agents in diseases such as chronic kidney disease and heart failure, in which hyperkalemia is the limiting factor in the use of renin-angiotensin-aldosterone system inhibitors.⁵⁷

OUR RECOMMENDATIONS

Our advice for managing patients receiving ACE inhibitors or ARBs is summarized in Table 1.

A highly active, water- and alcohol-soluble, basic pressor substance is formed when renin and renin-activator interact, for which we suggest the name “angiotonin.”

—Irvine H. Page and O.M. Helmer, 1940.¹

The renin-angiotensin-aldosterone system regulates salt and, in part, water homeostasis, and therefore blood pressure and fluid balance through its actions on the heart, kidneys, and blood vessels.² Drugs that target this system—angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs)—are used primarily to treat hypertension and also to treat chronic kidney disease and heart failure with reduced ejection fraction.

See related editorial

Controlling blood pressure is important, as hypertension increases the risk of myocardial infarction, cerebrovascular events, and progression of chronic kidney disease, which itself is a risk factor for cardiovascular disease. However, the benefit of these drugs is only partly due to their effect on blood pressure. They also reduce proteinuria, which is a graded risk factor for progression of kidney disease as well as morbidity and death from vascular events.³

Despite the benefits of ACE inhibitors and ARBs, concern about their adverse effects—especially hyperkalemia and a decline in renal function—has led to their underuse in patients likely to derive the greatest benefit.³

ACE INHIBITORS AND ARBs

momoniat_aceinbitorsarbspotassiumrenalfunction_f1.jpg

The renin-angiotensin-aldosterone system is activated when hypoperfusion to the glomerular afferent arteriole, reduced sodium delivery to the distal convoluted tubule, or increased sympathetic activity stimulates the renal juxtaglomerular apparatus to produce renin (Figure 1). This leads to a cascade of effects culminating in sodium retention and potassium excretion, thus increasing blood pressure.

ACE inhibitors, as their name indicates, inhibit conversion of angiotensin I to angiotensin II by ACE, resulting in vasodilation of the efferent arteriole and a drop in blood pressure. Inhibition of ACE, a kininase, also results in a rise in kinins. One of these, bradykinin, is associated with some of the side effects of this class of drugs such as cough, which affects 5% to 20% of patients.⁴ Elevation of bradykinin is also believed to account for ACE inhibitor-induced angioedema, an uncommon but potentially serious side effect. Kinins are also associated with desirable effects such as lowering blood pressure, increasing insulin sensitivity, and dilating blood vessels.

ARBs were developed as an alternative for patients unable to tolerate the adverse effects of ACE inhibitors. While ACE inhibitors reduce the activity of angiotensin II at both the AT1 and AT2 receptors, ARBs block only the AT1 receptors, thereby inhibiting their vasoconstricting activity on smooth muscle. ARBs also raise the levels of renin, angiotensin I, and angiotensin II as a result of feedback inhibition. Angiotensin II is associated with release of inflammatory mediators such as tumor necrosis factor alpha, cytokines, and chemokines, the consequences of which are also inhibited by ARBs, further preventing renal fibrosis and scarring from chronic inflammation.³

What is the evidence supporting the use of ACE inhibitors and ARBs?

ACE inhibitors and ARBs, used singly, reduce blood pressure and proteinuria, slow progression of kidney disease, and improve outcomes in patients who have heart failure, diabetes mellitus, or a history of myocardial infarction.^5–11 

While dual blockade with the combination of an ACE inhibitor and an ARB lowers blood pressure and proteinuria to a greater degree than monotherapy, dual blockade has been associated with higher rates of complications, including hyperkalemia.^12–17

RISK FACTORS FOR HYPERKALEMIA

ACE inhibitors and ARBs raise potassium, especially when used in combination. Other risk factors for hyperkalemia include the following—and note that some of them are also indications for ACE inhibitors and ARBs:

Renal insufficiency. The kidneys are responsible for over 90% of potassium removal in healthy individuals,^18,19 and the lower the GFR, the higher the risk of hyperkalemia.^3,20,21

Heart failure

Diabetes mellitus^6,21–23

Endogenous potassium load due to hemolysis, rhabdomyolysis, insulin deficiency, lactic acidosis, or gastrointestinal bleeding

Exogenous potassium load due to dietary consumption or blood products

Other medications, eg, sacubitril-valsartan, aldosterone antagonists, mineralocorticoid receptor antagonists, potassium-sparing diuretics, beta-adrenergic antagonists, nonsteroidal anti-inflammatory drugs, heparin, cyclosporine, trimethoprim, digoxin

Hypertension

Hypoaldosteronism (including type 4 renal tubular acidosis)

Addison disease

Advanced age

Lower body mass index.

Both hypokalemia and hyperkalemia are associated with a higher risk of death,^20,21,24  but in patients with heart failure, the survival benefit from ACE inhibitors, ARBs, and mineralocorticoid receptor antagonists outweighs the risk of hyperkalemia.^25–27 Weir and Rolfe²⁸ concluded that patients with heart failure and chronic kidney disease are at greatest risk of hyperkalemia from renin-angiotensin-aldosterone system inhibition, but the increases in potassium levels are small (about 0.1 to 0.3 mmol/L) and unlikely to be clinically significant.

Hyperkalemia tends to recur. Einhorn et al²⁰ found that nearly half of patients with chronic kidney disease who had an episode of hyperkalemia had 1 or more recurrent episodes within a year.

 

 

ACE INHIBITORS, ARBs, ABD RENAL FUNCTION

Another concern about using ACE inhibitors and ARBs, especially in patients with chronic kidney disease, is that the serum creatinine level tends to rise when starting these drugs,²⁹ although several studies have shown that an acute rise in creatinine may demonstrate that the drug is actually protecting the kidney.^30,31 Hirsch³² described this phenomenon as “prerenal success,” proposing that the decline in GFR is hemodynamic, secondary to a fall in intraglomerular pressure as a result of efferent vasodilation, and therefore should not be reversed.

Schmidt et al,^33,34 in a study in 122,363 patients who began ACE inhibitor or ARB therapy, found that cardiorenal outcomes were worse, with higher rates of end-stage renal disease, myocardial infarction, heart failure, and death, in those in whom creatinine rose by 30% or more since starting treatment. This trend was also seen, to a lesser degree, in those with a smaller increase in creatinine, suggesting that even this group of patients should receive close monitoring.

Whether renin-angiotensin-aldosterone system inhibitors provide a benefit in advanced progressive chronic kidney disease remains unclear.^35–37  The Angiotensin Converting Enzyme Inhibitor (ACEi)/Angiotensin Receptor Blocker (ARB) Withdrawal in Advanced Renal Disease trial (STOP-ACEi),³⁸ currently under way, will provide valuable data to help close this gap in our knowledge. This open-label randomized controlled trial is testing the hypothesis that stopping ACE inhibitor or ARB treatment, or a combination of both, compared with continuing these treatments, will improve or stabilize renal function in patients with progressive stage 4 or 5 chronic kidney disease.

NEED FOR MONITORING

Taken together, the above data suggest close and regular monitoring is required in patients receiving these drugs. However, monitoring tends to be lax.^34,37,39 A 2017 study of adherence to the guidelines for monitoring serum creatinine and potassium after starting an ACE inhibitor or ARB and subsequent discontinuation found that fewer than 10% of patients had follow-up within the recommended 2 weeks after starting these drugs.³⁴ Most patients with a creatinine rise of 30% or more or a potassium level higher than 6.0 mmol/L continued treatment. There was also no evidence of increased monitoring in those deemed at higher risk of these complications.

WHAT DO THE GUIDELINES SUGGEST?

ACE inhibitors and ARBs in chronic kidney disease and hypertension

Target blood pressures vary in guidelines from different organizations.^4,40–45 The 2017 joint guidelines of the American College of Cardiology and American Heart Association (ACC/AHA)⁴⁰ recommend a target blood pressure of 130/80 mm Hg or less in all patients irrespective of the level of proteinuria and whether they have diabetes mellitus, based on several studies.^46–48 In the elderly, other factors such as the risk of hypotension and falls must be taken into consideration in establishing the most appropriate blood pressure target.

In general, a renin-angiotensin-aldosterone system inhibitor is recommended if the patient has diabetes, stage 1, 2, or 3 chronic kidney disease, or proteinuria. For example, the guidelines recommend a renin-angiotensin-aldosterone system inhibitor in diabetic patients with albuminuria.

None of the guidelines recommend routine use of combination therapy.

ACE inhibitors and ARBs in heart failure

The 2017 ACC/AHA and Heart Failure Society of America (HFSA) guidelines for heart failure⁴⁹ recommend an ACE inhibitor or ARB for patients with stage C (symptomatic) heart failure with reduced ejection fraction, in view of the known cardiovascular morbidity and mortality benefits.

The European Society of Cardiology⁵⁰ recommends ACE inhibitors for patients with symptomatic heart failure with reduced ejection fraction, as well as those with asymptomatic left ventricular systolic dysfunction. In patients with stable coronary artery disease, an ACE inhibitor should be considered even with normal left ventricular function.

ARBs should be used as alternatives in those unable to tolerate ACE inhibitors.

Combination therapy should be avoided due to the increased risk of renal impairment and hyperkalemia but may be considered in patients with heart failure and reduced ejection fraction in whom other treatments are unsuitable. These include patients on beta-blockers who cannot tolerate mineralocorticoid receptor antagonists such as spironolactone. Combination therapy should be done only under strict supervision.⁵⁰

 

 

Starting ACE or ARB therapy

Close monitoring of serum potassium is recommended during ACE inhibitor or ARB use. Those at greatest risk of hyperkalemia include elderly patients, those taking other medications associated with hyperkalemia, and diabetic patients, because of their higher risk of renovascular disease.

Caution is advised when starting ACE inhibitor or ARB therapy in these high-risk groups as well as in patients with potassium levels higher than 5.0 mmol/L at baseline, at high risk of prerenal acute kidney injury, with known renal insufficiency, and with previous deterioration in renal function on these medications.^3,41,51

Before starting therapy, ensure that patients are volume-replete and measure baseline serum electrolytes and creatinine.^41,51

The ACC/AHA and HFSA recommend starting at a low dose and titrating upward slowly. If maximal doses are not tolerated, then a lower dose should be maintained.⁴⁹ The European Society of Cardiology guidelines⁵² suggest increasing the dose at no less than every 2 weeks unless in an inpatient setting. Blood testing should be done 7 to 14 days after starting therapy, after any titration in dosage, and every 4 months thereafter.⁵³

The guidelines generally agree that a rise in creatinine of up to 30% and a fall in eGFR of up to 25% is acceptable, with the need for regular monitoring, particularly in high-risk groups.^{40–42,51,52}

What if serum potassium or creatinine rises during treatment?

If hyperkalemia arises or renal function declines by a significant amount, one should first address contributing factors. If no improvement is seen, then the dose of the ACE inhibitor or ARB should be reduced by 50% and blood work repeated in 1 to 2 weeks. If the laboratory values do not return to an acceptable level, reducing the dose further or stopping the drug is advised.

Give dietary advice to all patients with chronic kidney disease being considered for a renin-angiotensin-aldosterone system inhibitor or for an increase in dose with a potassium level higher than 4.5 mmol/L. A low-potassium diet should aim for potassium intake of less than 50 or 75 mmol/day and sodium intake of less than 60 mmol/day for hypertensive patients with chronic kidney disease.

Review the patient’s medications if the baseline potassium level is higher than 5.0 mmol/L. Consider stopping potassium-sparing agents, digoxin, trimethoprim, and nonsteroidal anti-inflammatory drugs. Also think about starting a non–potassium-sparing diuretic as well as sodium bicarbonate to reduce potassium levels. Blood work should be repeated within 2 weeks after these changes.

Do not start a renin-angiotensin-aldosterone system inhibitor, or do not increase the dose, if the potassium level is elevated until measures have been taken to reduce the degree of hyperkalemia.⁵¹

In renal transplant recipients, renin-angiotensin-aldosterone system inhibitors are often preferred to manage hypertension in those who have proteinuria or cardiovascular disease. However, the risk of hyperkalemia is also greater with concomitant use of immunosuppressive drugs such as tacrolimus and cyclosporine. Management of complications should be approached according to guidelines discussed above.⁵¹

Monitor renal function, potassium. The National Institute for Health and Care Excellence guideline⁵⁴ advocates that baseline renal function testing should be followed by repeat blood testing 1 to 2 weeks after starting renin-angiotensin-aldosterone system inhibitors in patients with ischemic heart disease. The advice is similar when starting therapy in patients with chronic heart failure, emphasizing the need to monitor after each dose increment and to use clinical judgment when deciding to start treatment. The AHA advises caution in patients with renal insufficiency or a potassium level above 5.0 mmol/L.⁴⁹

Sick day rules. The National Institute for Health and Care Excellence encourages discussing “sick day rules” with patients starting renin-angiotensin-aldosterone system inhibitors. This means patients should be advised to temporarily stop taking nephrotoxic medications, including over-the-counter nonsteroidal anti-inflammatory drugs, in any potential state of illness or dehydration, such as diarrhea and vomiting. There is, however, little evidence that this advice can actually reduce the incidence of acute kidney injury.^55,56

momoniat_aceinbitorsarbspotassiumrenalfunction_t1.jpg

Potassium-lowering agents. Evidence is emerging to support the use of potassium-lowering agents to manage hyperkalemia. New compounds such as patiromer and zirconium cyclosilicate bind potassium in the gastrointestinal tract so it is excreted fecally. Meaney et al⁵⁶ performed a systematic review and meta-analysis of current phase 2 and 3 trials and concluded that these drugs lowered serum potassium levels by up to 0.70 mmol/L. There may be a significant role for these novel agents in diseases such as chronic kidney disease and heart failure, in which hyperkalemia is the limiting factor in the use of renin-angiotensin-aldosterone system inhibitors.⁵⁷

OUR RECOMMENDATIONS

Our advice for managing patients receiving ACE inhibitors or ARBs is summarized in Table 1.

A highly active, water- and alcohol-soluble, basic pressor substance is formed when renin and renin-activator interact, for which we suggest the name “angiotonin.”

—Irvine H. Page and O.M. Helmer, 1940.¹

The renin-angiotensin-aldosterone system regulates salt and, in part, water homeostasis, and therefore blood pressure and fluid balance through its actions on the heart, kidneys, and blood vessels.² Drugs that target this system—angiotensin-converting enzyme (ACE) inhibitors and angiotensin II receptor blockers (ARBs)—are used primarily to treat hypertension and also to treat chronic kidney disease and heart failure with reduced ejection fraction.

See related editorial

Controlling blood pressure is important, as hypertension increases the risk of myocardial infarction, cerebrovascular events, and progression of chronic kidney disease, which itself is a risk factor for cardiovascular disease. However, the benefit of these drugs is only partly due to their effect on blood pressure. They also reduce proteinuria, which is a graded risk factor for progression of kidney disease as well as morbidity and death from vascular events.³

Despite the benefits of ACE inhibitors and ARBs, concern about their adverse effects—especially hyperkalemia and a decline in renal function—has led to their underuse in patients likely to derive the greatest benefit.³

ACE INHIBITORS AND ARBs

momoniat_aceinbitorsarbspotassiumrenalfunction_f1.jpg

The renin-angiotensin-aldosterone system is activated when hypoperfusion to the glomerular afferent arteriole, reduced sodium delivery to the distal convoluted tubule, or increased sympathetic activity stimulates the renal juxtaglomerular apparatus to produce renin (Figure 1). This leads to a cascade of effects culminating in sodium retention and potassium excretion, thus increasing blood pressure.

ACE inhibitors, as their name indicates, inhibit conversion of angiotensin I to angiotensin II by ACE, resulting in vasodilation of the efferent arteriole and a drop in blood pressure. Inhibition of ACE, a kininase, also results in a rise in kinins. One of these, bradykinin, is associated with some of the side effects of this class of drugs such as cough, which affects 5% to 20% of patients.⁴ Elevation of bradykinin is also believed to account for ACE inhibitor-induced angioedema, an uncommon but potentially serious side effect. Kinins are also associated with desirable effects such as lowering blood pressure, increasing insulin sensitivity, and dilating blood vessels.

ARBs were developed as an alternative for patients unable to tolerate the adverse effects of ACE inhibitors. While ACE inhibitors reduce the activity of angiotensin II at both the AT1 and AT2 receptors, ARBs block only the AT1 receptors, thereby inhibiting their vasoconstricting activity on smooth muscle. ARBs also raise the levels of renin, angiotensin I, and angiotensin II as a result of feedback inhibition. Angiotensin II is associated with release of inflammatory mediators such as tumor necrosis factor alpha, cytokines, and chemokines, the consequences of which are also inhibited by ARBs, further preventing renal fibrosis and scarring from chronic inflammation.³

What is the evidence supporting the use of ACE inhibitors and ARBs?

ACE inhibitors and ARBs, used singly, reduce blood pressure and proteinuria, slow progression of kidney disease, and improve outcomes in patients who have heart failure, diabetes mellitus, or a history of myocardial infarction.^5–11 

While dual blockade with the combination of an ACE inhibitor and an ARB lowers blood pressure and proteinuria to a greater degree than monotherapy, dual blockade has been associated with higher rates of complications, including hyperkalemia.^12–17

RISK FACTORS FOR HYPERKALEMIA

ACE inhibitors and ARBs raise potassium, especially when used in combination. Other risk factors for hyperkalemia include the following—and note that some of them are also indications for ACE inhibitors and ARBs:

Renal insufficiency. The kidneys are responsible for over 90% of potassium removal in healthy individuals,^18,19 and the lower the GFR, the higher the risk of hyperkalemia.^3,20,21

Heart failure

Diabetes mellitus^6,21–23

Endogenous potassium load due to hemolysis, rhabdomyolysis, insulin deficiency, lactic acidosis, or gastrointestinal bleeding

Exogenous potassium load due to dietary consumption or blood products

Other medications, eg, sacubitril-valsartan, aldosterone antagonists, mineralocorticoid receptor antagonists, potassium-sparing diuretics, beta-adrenergic antagonists, nonsteroidal anti-inflammatory drugs, heparin, cyclosporine, trimethoprim, digoxin

Hypertension

Hypoaldosteronism (including type 4 renal tubular acidosis)

Addison disease

Advanced age

Lower body mass index.

Both hypokalemia and hyperkalemia are associated with a higher risk of death,^20,21,24  but in patients with heart failure, the survival benefit from ACE inhibitors, ARBs, and mineralocorticoid receptor antagonists outweighs the risk of hyperkalemia.^25–27 Weir and Rolfe²⁸ concluded that patients with heart failure and chronic kidney disease are at greatest risk of hyperkalemia from renin-angiotensin-aldosterone system inhibition, but the increases in potassium levels are small (about 0.1 to 0.3 mmol/L) and unlikely to be clinically significant.

Hyperkalemia tends to recur. Einhorn et al²⁰ found that nearly half of patients with chronic kidney disease who had an episode of hyperkalemia had 1 or more recurrent episodes within a year.

 

 

ACE INHIBITORS, ARBs, ABD RENAL FUNCTION

Another concern about using ACE inhibitors and ARBs, especially in patients with chronic kidney disease, is that the serum creatinine level tends to rise when starting these drugs,²⁹ although several studies have shown that an acute rise in creatinine may demonstrate that the drug is actually protecting the kidney.^30,31 Hirsch³² described this phenomenon as “prerenal success,” proposing that the decline in GFR is hemodynamic, secondary to a fall in intraglomerular pressure as a result of efferent vasodilation, and therefore should not be reversed.

Schmidt et al,^33,34 in a study in 122,363 patients who began ACE inhibitor or ARB therapy, found that cardiorenal outcomes were worse, with higher rates of end-stage renal disease, myocardial infarction, heart failure, and death, in those in whom creatinine rose by 30% or more since starting treatment. This trend was also seen, to a lesser degree, in those with a smaller increase in creatinine, suggesting that even this group of patients should receive close monitoring.

Whether renin-angiotensin-aldosterone system inhibitors provide a benefit in advanced progressive chronic kidney disease remains unclear.^35–37  The Angiotensin Converting Enzyme Inhibitor (ACEi)/Angiotensin Receptor Blocker (ARB) Withdrawal in Advanced Renal Disease trial (STOP-ACEi),³⁸ currently under way, will provide valuable data to help close this gap in our knowledge. This open-label randomized controlled trial is testing the hypothesis that stopping ACE inhibitor or ARB treatment, or a combination of both, compared with continuing these treatments, will improve or stabilize renal function in patients with progressive stage 4 or 5 chronic kidney disease.

NEED FOR MONITORING

Taken together, the above data suggest close and regular monitoring is required in patients receiving these drugs. However, monitoring tends to be lax.^34,37,39 A 2017 study of adherence to the guidelines for monitoring serum creatinine and potassium after starting an ACE inhibitor or ARB and subsequent discontinuation found that fewer than 10% of patients had follow-up within the recommended 2 weeks after starting these drugs.³⁴ Most patients with a creatinine rise of 30% or more or a potassium level higher than 6.0 mmol/L continued treatment. There was also no evidence of increased monitoring in those deemed at higher risk of these complications.

WHAT DO THE GUIDELINES SUGGEST?

ACE inhibitors and ARBs in chronic kidney disease and hypertension

Target blood pressures vary in guidelines from different organizations.^4,40–45 The 2017 joint guidelines of the American College of Cardiology and American Heart Association (ACC/AHA)⁴⁰ recommend a target blood pressure of 130/80 mm Hg or less in all patients irrespective of the level of proteinuria and whether they have diabetes mellitus, based on several studies.^46–48 In the elderly, other factors such as the risk of hypotension and falls must be taken into consideration in establishing the most appropriate blood pressure target.

In general, a renin-angiotensin-aldosterone system inhibitor is recommended if the patient has diabetes, stage 1, 2, or 3 chronic kidney disease, or proteinuria. For example, the guidelines recommend a renin-angiotensin-aldosterone system inhibitor in diabetic patients with albuminuria.

None of the guidelines recommend routine use of combination therapy.

ACE inhibitors and ARBs in heart failure

The 2017 ACC/AHA and Heart Failure Society of America (HFSA) guidelines for heart failure⁴⁹ recommend an ACE inhibitor or ARB for patients with stage C (symptomatic) heart failure with reduced ejection fraction, in view of the known cardiovascular morbidity and mortality benefits.

The European Society of Cardiology⁵⁰ recommends ACE inhibitors for patients with symptomatic heart failure with reduced ejection fraction, as well as those with asymptomatic left ventricular systolic dysfunction. In patients with stable coronary artery disease, an ACE inhibitor should be considered even with normal left ventricular function.

ARBs should be used as alternatives in those unable to tolerate ACE inhibitors.

Combination therapy should be avoided due to the increased risk of renal impairment and hyperkalemia but may be considered in patients with heart failure and reduced ejection fraction in whom other treatments are unsuitable. These include patients on beta-blockers who cannot tolerate mineralocorticoid receptor antagonists such as spironolactone. Combination therapy should be done only under strict supervision.⁵⁰

 

 

Starting ACE or ARB therapy

Close monitoring of serum potassium is recommended during ACE inhibitor or ARB use. Those at greatest risk of hyperkalemia include elderly patients, those taking other medications associated with hyperkalemia, and diabetic patients, because of their higher risk of renovascular disease.

Caution is advised when starting ACE inhibitor or ARB therapy in these high-risk groups as well as in patients with potassium levels higher than 5.0 mmol/L at baseline, at high risk of prerenal acute kidney injury, with known renal insufficiency, and with previous deterioration in renal function on these medications.^3,41,51

Before starting therapy, ensure that patients are volume-replete and measure baseline serum electrolytes and creatinine.^41,51

The ACC/AHA and HFSA recommend starting at a low dose and titrating upward slowly. If maximal doses are not tolerated, then a lower dose should be maintained.⁴⁹ The European Society of Cardiology guidelines⁵² suggest increasing the dose at no less than every 2 weeks unless in an inpatient setting. Blood testing should be done 7 to 14 days after starting therapy, after any titration in dosage, and every 4 months thereafter.⁵³

The guidelines generally agree that a rise in creatinine of up to 30% and a fall in eGFR of up to 25% is acceptable, with the need for regular monitoring, particularly in high-risk groups.^{40–42,51,52}

What if serum potassium or creatinine rises during treatment?

If hyperkalemia arises or renal function declines by a significant amount, one should first address contributing factors. If no improvement is seen, then the dose of the ACE inhibitor or ARB should be reduced by 50% and blood work repeated in 1 to 2 weeks. If the laboratory values do not return to an acceptable level, reducing the dose further or stopping the drug is advised.

Give dietary advice to all patients with chronic kidney disease being considered for a renin-angiotensin-aldosterone system inhibitor or for an increase in dose with a potassium level higher than 4.5 mmol/L. A low-potassium diet should aim for potassium intake of less than 50 or 75 mmol/day and sodium intake of less than 60 mmol/day for hypertensive patients with chronic kidney disease.

Review the patient’s medications if the baseline potassium level is higher than 5.0 mmol/L. Consider stopping potassium-sparing agents, digoxin, trimethoprim, and nonsteroidal anti-inflammatory drugs. Also think about starting a non–potassium-sparing diuretic as well as sodium bicarbonate to reduce potassium levels. Blood work should be repeated within 2 weeks after these changes.

Do not start a renin-angiotensin-aldosterone system inhibitor, or do not increase the dose, if the potassium level is elevated until measures have been taken to reduce the degree of hyperkalemia.⁵¹

In renal transplant recipients, renin-angiotensin-aldosterone system inhibitors are often preferred to manage hypertension in those who have proteinuria or cardiovascular disease. However, the risk of hyperkalemia is also greater with concomitant use of immunosuppressive drugs such as tacrolimus and cyclosporine. Management of complications should be approached according to guidelines discussed above.⁵¹

Monitor renal function, potassium. The National Institute for Health and Care Excellence guideline⁵⁴ advocates that baseline renal function testing should be followed by repeat blood testing 1 to 2 weeks after starting renin-angiotensin-aldosterone system inhibitors in patients with ischemic heart disease. The advice is similar when starting therapy in patients with chronic heart failure, emphasizing the need to monitor after each dose increment and to use clinical judgment when deciding to start treatment. The AHA advises caution in patients with renal insufficiency or a potassium level above 5.0 mmol/L.⁴⁹

Sick day rules. The National Institute for Health and Care Excellence encourages discussing “sick day rules” with patients starting renin-angiotensin-aldosterone system inhibitors. This means patients should be advised to temporarily stop taking nephrotoxic medications, including over-the-counter nonsteroidal anti-inflammatory drugs, in any potential state of illness or dehydration, such as diarrhea and vomiting. There is, however, little evidence that this advice can actually reduce the incidence of acute kidney injury.^55,56

momoniat_aceinbitorsarbspotassiumrenalfunction_t1.jpg

Potassium-lowering agents. Evidence is emerging to support the use of potassium-lowering agents to manage hyperkalemia. New compounds such as patiromer and zirconium cyclosilicate bind potassium in the gastrointestinal tract so it is excreted fecally. Meaney et al⁵⁶ performed a systematic review and meta-analysis of current phase 2 and 3 trials and concluded that these drugs lowered serum potassium levels by up to 0.70 mmol/L. There may be a significant role for these novel agents in diseases such as chronic kidney disease and heart failure, in which hyperkalemia is the limiting factor in the use of renin-angiotensin-aldosterone system inhibitors.⁵⁷

OUR RECOMMENDATIONS

Our advice for managing patients receiving ACE inhibitors or ARBs is summarized in Table 1.

Prompt diagnois of infective endocarditis is critical. Potential consequences of missed or delayed diagnosis, including heart failure, stroke, intracardiac abscess, conduction delays, prosthesis dysfunction, and cerebral emboli, are often catastrophic. Echocardiography is the test used most frequently to evaluate for infective endocarditis, but it misses the diagnosis in almost one-third of cases, and even more often if the patient has a prosthetic valve.

mgbojikwe_infectiveendocarditis_t1.jpg

But now, several sophisticated imaging tests are available that complement echocardiography in diagnosing and assessing infective endocarditis; these include 4-dimensional computed tomography (4D CT), fluorodeoxyglucose positron emission tomography (FDG-PET), and leukocyte scintigraphy. These tests have greatly improved our ability not only to diagnose infective endocarditis, but also to determine the extent and spread of infection, and they aid in perioperative assessment. Abnormal findings on these tests have been incorporated into the European Society of Cardiology’s 2015 modified diagnostic criteria for infective endocarditis.¹

This article details the indications, advantages, and limitations of the various imaging tests for diagnosing and evaluating infective endocarditis (Table 1).

INFECTIVE ENDOCARDITIS IS DIFFICULT TO DIAGNOSE AND TREAT

Infective endocarditis is difficult to diagnose and treat. Clinical and imaging clues can be subtle, and the diagnosis requires a high level of suspicion and visualization of cardiac structures.

Further, the incidence of infective endocarditis is on the rise in the United States, particularly in women and young adults, likely due to intravenous drug use.^2,3

ECHOCARDIOGRAPHY HAS AN IMPORTANT ROLE, BUT IS LIMITED

Echocardiography remains the most commonly performed study for diagnosing infective endocarditis, as it is fast, widely accessible, and less expensive than other imaging tests.

Transthoracic echocardiography (TTE) is often the first choice for testing. However, its sensitivity is only about 70% for detecting vegetations on native valves and 50% for detecting vegetations on prosthetic valves.¹ It is inherently constrained by the limited number of views by which a comprehensive external evaluation of the heart can be achieved. Using a 2-dimensional instrument to view a 3-dimensional object is difficult, and depending on several factors, it can be hard to see vegetations and abscesses that are associated with infective endocarditis. Further, TTE is impeded by obesity and by hyperinflated lungs from obstructive pulmonary disease or mechanical ventilation. It has poor sensitivity for detecting small vegetations and for detecting vegetations and paravalvular complications in patients who have a prosthetic valve or a cardiac implanted electronic device.

Transesophageal echocardiography (TEE) is the recommended first-line imaging test for patients with prosthetic valves and no contraindications to the test. Otherwise, it should be done after TTE if the results of TTE are negative but clinical suspicion for infective endocarditis remains high (eg, because the patient uses intravenous drugs). But although TEE has a higher sensitivity than TTE (up to 96% for vegetations on native valves and 92% for those on prosthetic valves, if performed by an experienced sonographer), it can still miss infective endocarditis. Also, TEE does not provide a significant advantage over TTE in patients who have a cardiac implanted electronic device.^1,4,5

Regardless of whether TTE or TEE is used, they are estimated to miss up to 30% of cases of infective endocarditis and its sequelae.⁴ False-negative findings are likelier in patients who have preexisting severe valvular lesions, prosthetic valves, cardiac implanted electronic devices, small vegetations, or abscesses, or if a vegetation has already broken free and embolized. Furthermore, distinguishing between vegetations and thrombi, cardiac tumors, and myxomatous changes using echocardiography is difficult.

 

 

CARDIAC CT

For patients who have inconclusive results on echocardiography, contraindications to TEE, or poor sonic windows, cardiac CT can be an excellent alternative. It is especially useful in the setting of a prosthetic valve.

Synchronized (“gated”) with the patient’s heart rate and rhythm, CT machines can acquire images during diastole, reducing motion artifact, and can create 3D images of the heart. In addition, newer machines can acquire several images at different points in the heart cycle to add a fourth dimension—time. The resulting 4D images play like short video loops of the beating heart and allow noninvasive assessment of cardiac anatomy with remarkable detail and resolution.

4D CT is increasingly being used in infective endocarditis, and growing evidence indicates that its accuracy is similar to that of TEE in the preoperative evaluation of patients with aortic prosthetic valve endocarditis.⁶ In a study of 28 patients, complementary use of CT angiography led to a change in treatment strategy in 7 (25%) compared with routine clinical workup.⁷ Several studies have found no difference between 4D CT and preoperative TEE in detecting pseudoaneurysm, abscess, or valve dehiscence. TEE and 4D CT also have similar sensitivities for detecting infective endocarditis in native and prosthetic valves.^8,9

mgbojikwe_infectiveendocarditis_f1a.jpg

mgbojikwe_infectiveendocarditis_f1b.jpg

mgbojikwe_infectiveendocarditis_f1c.jpg

Coupled with CT angiography, 4D CT is also an excellent noninvasive way to perioperatively evaluate the coronary arteries without the risks associated with catheterization in those requiring nonemergency surgery (Figure 1A, B, and C).

4D CT performs well for detecting abscess and pseudoaneurysm but has slightly lower sensitivity for vegetations than TEE (91% vs 99%).⁹

Gated CT, PET, or both may be useful in cases of suspected prosthetic aortic valve endocarditis when TEE is negative. Pseudoaneurysms are not well visualized with TEE, and the atrial mitral curtain area is often thickened on TEE in cases of aortic prosthetic valve infective endocarditis that do not definitely involve abscesses. Gated CT and PET show this area better.⁸ This information is important in cases in which a surgeon may be unconvinced that the patient has prosthetic valve endocarditis.

Limitations of 4D cardiac CT

4D CT with or without angiography has limitations. It requires a wide-volume scanner and an experienced reader.

Patients with irregular heart rhythms or uncontrolled tachycardia pose technical problems for image acquisition. Cardiac CT is typically gated (ie, images are obtained within a defined time period) to acquire images during diastole. Ideally, images are acquired when the heart is in mid to late diastole, a time of minimal cardiac motion, so that motion artifact is minimized. To estimate the timing of image acquisition, the cardiac cycle must be predictable, and its duration should be as long as possible. Tachycardia or irregular rhythms such as frequent ectopic beats or atrial fibrillation make acquisition timing difficult, and thus make it nearly impossible to accurately obtain images when the heart is at minimum motion, limiting assessment of cardiac structures or the coronary tree.^4,10

Extensive coronary calcification can hinder assessment of the coronary tree by CT coronary angiography.

Contrast exposure may limit the use of CT in some patients (eg, those with contrast allergies or renal dysfunction). However, modern scanners allow for much smaller contrast boluses without decreasing sensitivity.

4D CT involves radiation exposure, especially when done with angiography, although modern scanners have greatly reduced exposure. The average radiation dose in CT coronary angiography is 2.9 to 5.9 mSv¹¹ compared with 7 mSv in diagnostic cardiac catheterization (without angioplasty or stenting) or 16 mSv in routine CT of the abdomen and pelvis with contrast.^12,13 In view of the morbidity and mortality risks associated with infective endocarditis, especially if the diagnosis is delayed, this small radiation exposure may be justifiable.

Bottom line for cardiac CT

4D CT is an excellent alternative to echocardiography for select patients. Clinicians should strongly consider this study in the following situations:

• Patients with a prosthetic valve
• Patients who are strongly suspected of having infective endocarditis but who have a poor sonic window on TTE or TEE, as can occur with chronic obstructive lung disease, morbid obesity, or previous thoracic or cardiovascular surgery
• Patients who meet clinical indications for TEE, such as having a prosthetic valve or a high suspicion for native valve infective endocarditis with negative TTE, but who have contraindications to TEE
• As an alternative to TEE for preoperative evaluation in patients with known infective endocarditis.

Patients with tachycardia or irregular heart rhythms are not good candidates for this test.

FDG-PET AND LEUKOCYTE SCINTIGRAPHY

FDG-PET and leukocyte scintigraphy are other options for diagnosing infective endocarditis and determining the presence and extent of intra- and extracardiac infection. They are more sensitive than echocardiography for detecting infection of cardiac implanted electronic devices such as ventricular assist devices, pacemakers, implanted cardiac defibrillators, and cardiac resynchronization therapy devices.^14–16

The utility of FDG-PET is founded on the uptake of ¹⁸F-fluorodeoxyglucose by cells, with higher uptake taking place in cells with higher metabolic activity (such as in areas of inflammation). Similarly, leukocyte scintigraphy relies on the use of radiolabeled leukocytes (ie, leukocytes previously extracted from the patient, labelled, and re-introduced into the patient) to allow for localization of inflamed tissue.

The most significant contribution of FDG-PET may be the ability to detect infective endocarditis early, when echocardiography is initially negative. When abnormal FDG uptake was included in the modified Duke criteria, it increased the sensitivity to 97% for detecting infective endocarditis on admission, leading some to propose its incorporation as a major criterion.¹⁷ In patients with prosthetic valves and suspected infective endocarditis, FDG-PET was found in one study to have a sensitivity of up to 91% and a specificity of up to 95%.¹⁸

Both FDG-PET and leukocyte scintigraphy have a high sensitivity, specificity, and negative predictive value for cardiac implanted electronic device infection, and should be strongly considered in patients in whom it is suspected but who have negative or inconclusive findings on echocardiography.^14,15

In addition, a common conundrum faced by clinicians with use of echocardiography is the difficulty of differentiating thrombus from infected vegetation on valves or device lead wires. Some evidence indicates that FDG-PET may help to discriminate between vegetation and thrombus, although more rigorous studies are needed before its use for that purpose can be recommended.¹⁹

 

 

Limitations of nuclear studies

Both FDG-PET and leukocyte scintigraphy perform poorly for detecting native-valve infective endocarditis. In a study in which 90% of the patients had native-valve infective endocarditis according to the Duke criteria, FDG-PET had a specificity of 93% but a sensitivity of only 39%.²⁰

Both studies can be cumbersome, laborious, and time-consuming for patients. FDG-PET requires a fasting or glucose-restricted diet before testing, and the test itself can be complicated by development of hyperglycemia, although this is rare.

While FDG-PET is most effective in detecting infections of prosthetic valves and cardiac implanted electronic devices, the results can be falsely positive in patients with a history of recent cardiac surgery (due to ongoing tissue healing), as well as maladies other than infective endocarditis that lead to inflammation, such as vasculitis or malignancy. Similarly, for unclear reasons, leukocyte scintigraphy can yield false-negative results in patients with enterococcal or candidal infective endocarditis.²¹

FDG-PET and leukocyte scintigraphy are more expensive than TEE and cardiac CT²² and are not widely available.

Both tests entail radiation exposure, with the average dose ranging from 7 to 14 mSv. However, this is less than the average amount acquired during percutaneous coronary intervention (16 mSv), and overlaps with the amount in chest CT with contrast when assessing for pulmonary embolism (7 to 9 mSv). Lower doses are possible with optimized protocols.^12,13,15,23

Bottom line for nuclear studies

mgbojikwe_infectiveendocarditis_f2.jpg

FDG-PET and leukocyte scintigraphy are especially useful for patients with a prosthetic valve or cardiac implanted electronic device. However, limitations must be kept in mind.

A suggested algorithm for testing with nuclear imaging is shown in Figure 2.^1,4

CEREBRAL MAGNETIC RESONANCE IMAGING

Cerebral magnetic resonance imaging (MRI) is more sensitive than cerebral CT for detecting emboli in the brain. According to American Heart Association guidelines, cerebral MRI should be done in patients with known or suspected infective endocarditis and neurologic impairment, defined as headaches, meningeal symptoms, or neurologic deficits. It is also often used in neurologically asymptomatic patients with infective endocarditis who have indications for valve surgery to assess for mycotic aneurysms, which are associated with increased intracranial bleeding during surgery.

MRI use in other asymptomatic patients remains controversial.²⁴ In cases with high clinical suspicion for infective endocarditis and no findings on echocardiography, cerebral MRI can increase the sensitivity of the Duke criteria by adding a minor criterion. Some have argued that, in patients with definite infective endocarditis, detecting silent cerebral complications can lead to management changes. However, more studies are needed to determine if there is indeed a group of neurologically asymptomatic infective endocarditis patients for whom cerebral MRI leads to improved outcomes.

Limitations of cerebral MRI

Cerebral MRI cannot be used in patients with non-MRI-compatible implanted hardware.

Gadolinium, the contrast agent typically used, can cause nephrogenic systemic fibrosis in patients who have poor renal function. This rare but serious adverse effect is characterized by irreversible systemic fibrosis affecting skin, muscles, and even visceral tissue such as lungs. The American College of Radiology allows for gadolinium use in patients without acute kidney injury and patients with stable chronic kidney disease with a glomerular filtration rate of at least 30 mL/min/1.73 m². Its use should be avoided in patients with renal failure on replacement therapy, with advanced chronic kidney disease (glomerular filtration rate < 30 mL/min/1.73 m²), or with acute kidney injury, even if they do not need renal replacement therapy.²⁵

Concerns have also been raised about gadolinium retention in the brain, even in patients with normal renal function.^26–28 Thus far, no conclusive clinical adverse effects of retention have been found, although more study is warranted. Nevertheless, the US Food and Drug Administration now requires a black-box warning about this possibility and advises clinicians to counsel patients appropriately.

Bottom line on cerebral MRI

Cerebral MRI should be obtained when a patient presents with definite or possible infective endocarditis with neurologic impairment, such as new headaches, meningismus, or focal neurologic deficits. Routine brain MRI in patients with confirmed infective endocarditis without neurologic symptoms, or those without definite infective endocarditis, is discouraged.

CARDIAC MRI

Cardiac MRI, typically obtained with gadolinium contrast, allows for better 3D assessment of cardiac structures and morphology than echocardiography or CT, and can detect infiltrative cardiac disease, myopericarditis, and much more. It is increasingly used in the field of structural cardiology, but its role for evaluating infective endocarditis remains unclear.

Cardiac MRI does not appear to be better than echocardiography for diagnosing infective endocarditis. However, it may prove helpful in the evaluation of patients known to have infective endocarditis but who cannot be properly evaluated for disease extent because of poor image quality on echocardiography and contraindications to CT.^1,29 Its role is limited in patients with cardiac implanted electronic devices, as most devices are incompatible with MRI use, although newer devices obviate this concern. But even for devices that are MRI-compatible, results are diminished due to an eclipsing effect, wherein the device parts can make it hard to see structures clearly because the “brightness” basically eclipses the surrounding area.⁴

Concerns regarding use of gadolinium as described above need also be considered.

The role of cardiac MRI in diagnosing and managing infective endocarditis may evolve, but at present, the 2017 American College of Cardiology and American Heart Association appropriate-use criteria discourage its use for these purposes.¹⁶

Bottom line for cardiac MRI

Cardiac MRI to evaluate a patient for suspected infective endocarditis is not recommended due to lack of superiority compared with echocardiography or CT, and the risk of nephrogenic systemic fibrosis from gadolinium in patients with renal compromise.

Prompt diagnois of infective endocarditis is critical. Potential consequences of missed or delayed diagnosis, including heart failure, stroke, intracardiac abscess, conduction delays, prosthesis dysfunction, and cerebral emboli, are often catastrophic. Echocardiography is the test used most frequently to evaluate for infective endocarditis, but it misses the diagnosis in almost one-third of cases, and even more often if the patient has a prosthetic valve.

mgbojikwe_infectiveendocarditis_t1.jpg

But now, several sophisticated imaging tests are available that complement echocardiography in diagnosing and assessing infective endocarditis; these include 4-dimensional computed tomography (4D CT), fluorodeoxyglucose positron emission tomography (FDG-PET), and leukocyte scintigraphy. These tests have greatly improved our ability not only to diagnose infective endocarditis, but also to determine the extent and spread of infection, and they aid in perioperative assessment. Abnormal findings on these tests have been incorporated into the European Society of Cardiology’s 2015 modified diagnostic criteria for infective endocarditis.¹

This article details the indications, advantages, and limitations of the various imaging tests for diagnosing and evaluating infective endocarditis (Table 1).

INFECTIVE ENDOCARDITIS IS DIFFICULT TO DIAGNOSE AND TREAT

Infective endocarditis is difficult to diagnose and treat. Clinical and imaging clues can be subtle, and the diagnosis requires a high level of suspicion and visualization of cardiac structures.

Further, the incidence of infective endocarditis is on the rise in the United States, particularly in women and young adults, likely due to intravenous drug use.^2,3

ECHOCARDIOGRAPHY HAS AN IMPORTANT ROLE, BUT IS LIMITED

Echocardiography remains the most commonly performed study for diagnosing infective endocarditis, as it is fast, widely accessible, and less expensive than other imaging tests.

Transthoracic echocardiography (TTE) is often the first choice for testing. However, its sensitivity is only about 70% for detecting vegetations on native valves and 50% for detecting vegetations on prosthetic valves.¹ It is inherently constrained by the limited number of views by which a comprehensive external evaluation of the heart can be achieved. Using a 2-dimensional instrument to view a 3-dimensional object is difficult, and depending on several factors, it can be hard to see vegetations and abscesses that are associated with infective endocarditis. Further, TTE is impeded by obesity and by hyperinflated lungs from obstructive pulmonary disease or mechanical ventilation. It has poor sensitivity for detecting small vegetations and for detecting vegetations and paravalvular complications in patients who have a prosthetic valve or a cardiac implanted electronic device.

Transesophageal echocardiography (TEE) is the recommended first-line imaging test for patients with prosthetic valves and no contraindications to the test. Otherwise, it should be done after TTE if the results of TTE are negative but clinical suspicion for infective endocarditis remains high (eg, because the patient uses intravenous drugs). But although TEE has a higher sensitivity than TTE (up to 96% for vegetations on native valves and 92% for those on prosthetic valves, if performed by an experienced sonographer), it can still miss infective endocarditis. Also, TEE does not provide a significant advantage over TTE in patients who have a cardiac implanted electronic device.^1,4,5

Regardless of whether TTE or TEE is used, they are estimated to miss up to 30% of cases of infective endocarditis and its sequelae.⁴ False-negative findings are likelier in patients who have preexisting severe valvular lesions, prosthetic valves, cardiac implanted electronic devices, small vegetations, or abscesses, or if a vegetation has already broken free and embolized. Furthermore, distinguishing between vegetations and thrombi, cardiac tumors, and myxomatous changes using echocardiography is difficult.

 

 

CARDIAC CT

For patients who have inconclusive results on echocardiography, contraindications to TEE, or poor sonic windows, cardiac CT can be an excellent alternative. It is especially useful in the setting of a prosthetic valve.

Synchronized (“gated”) with the patient’s heart rate and rhythm, CT machines can acquire images during diastole, reducing motion artifact, and can create 3D images of the heart. In addition, newer machines can acquire several images at different points in the heart cycle to add a fourth dimension—time. The resulting 4D images play like short video loops of the beating heart and allow noninvasive assessment of cardiac anatomy with remarkable detail and resolution.

4D CT is increasingly being used in infective endocarditis, and growing evidence indicates that its accuracy is similar to that of TEE in the preoperative evaluation of patients with aortic prosthetic valve endocarditis.⁶ In a study of 28 patients, complementary use of CT angiography led to a change in treatment strategy in 7 (25%) compared with routine clinical workup.⁷ Several studies have found no difference between 4D CT and preoperative TEE in detecting pseudoaneurysm, abscess, or valve dehiscence. TEE and 4D CT also have similar sensitivities for detecting infective endocarditis in native and prosthetic valves.^8,9

mgbojikwe_infectiveendocarditis_f1a.jpg

mgbojikwe_infectiveendocarditis_f1b.jpg

mgbojikwe_infectiveendocarditis_f1c.jpg

Coupled with CT angiography, 4D CT is also an excellent noninvasive way to perioperatively evaluate the coronary arteries without the risks associated with catheterization in those requiring nonemergency surgery (Figure 1A, B, and C).

4D CT performs well for detecting abscess and pseudoaneurysm but has slightly lower sensitivity for vegetations than TEE (91% vs 99%).⁹

Gated CT, PET, or both may be useful in cases of suspected prosthetic aortic valve endocarditis when TEE is negative. Pseudoaneurysms are not well visualized with TEE, and the atrial mitral curtain area is often thickened on TEE in cases of aortic prosthetic valve infective endocarditis that do not definitely involve abscesses. Gated CT and PET show this area better.⁸ This information is important in cases in which a surgeon may be unconvinced that the patient has prosthetic valve endocarditis.

Limitations of 4D cardiac CT

4D CT with or without angiography has limitations. It requires a wide-volume scanner and an experienced reader.

Patients with irregular heart rhythms or uncontrolled tachycardia pose technical problems for image acquisition. Cardiac CT is typically gated (ie, images are obtained within a defined time period) to acquire images during diastole. Ideally, images are acquired when the heart is in mid to late diastole, a time of minimal cardiac motion, so that motion artifact is minimized. To estimate the timing of image acquisition, the cardiac cycle must be predictable, and its duration should be as long as possible. Tachycardia or irregular rhythms such as frequent ectopic beats or atrial fibrillation make acquisition timing difficult, and thus make it nearly impossible to accurately obtain images when the heart is at minimum motion, limiting assessment of cardiac structures or the coronary tree.^4,10

Extensive coronary calcification can hinder assessment of the coronary tree by CT coronary angiography.

Contrast exposure may limit the use of CT in some patients (eg, those with contrast allergies or renal dysfunction). However, modern scanners allow for much smaller contrast boluses without decreasing sensitivity.

4D CT involves radiation exposure, especially when done with angiography, although modern scanners have greatly reduced exposure. The average radiation dose in CT coronary angiography is 2.9 to 5.9 mSv¹¹ compared with 7 mSv in diagnostic cardiac catheterization (without angioplasty or stenting) or 16 mSv in routine CT of the abdomen and pelvis with contrast.^12,13 In view of the morbidity and mortality risks associated with infective endocarditis, especially if the diagnosis is delayed, this small radiation exposure may be justifiable.

Bottom line for cardiac CT

4D CT is an excellent alternative to echocardiography for select patients. Clinicians should strongly consider this study in the following situations:

• Patients with a prosthetic valve
• Patients who are strongly suspected of having infective endocarditis but who have a poor sonic window on TTE or TEE, as can occur with chronic obstructive lung disease, morbid obesity, or previous thoracic or cardiovascular surgery
• Patients who meet clinical indications for TEE, such as having a prosthetic valve or a high suspicion for native valve infective endocarditis with negative TTE, but who have contraindications to TEE
• As an alternative to TEE for preoperative evaluation in patients with known infective endocarditis.

Patients with tachycardia or irregular heart rhythms are not good candidates for this test.

FDG-PET AND LEUKOCYTE SCINTIGRAPHY

FDG-PET and leukocyte scintigraphy are other options for diagnosing infective endocarditis and determining the presence and extent of intra- and extracardiac infection. They are more sensitive than echocardiography for detecting infection of cardiac implanted electronic devices such as ventricular assist devices, pacemakers, implanted cardiac defibrillators, and cardiac resynchronization therapy devices.^14–16

The utility of FDG-PET is founded on the uptake of ¹⁸F-fluorodeoxyglucose by cells, with higher uptake taking place in cells with higher metabolic activity (such as in areas of inflammation). Similarly, leukocyte scintigraphy relies on the use of radiolabeled leukocytes (ie, leukocytes previously extracted from the patient, labelled, and re-introduced into the patient) to allow for localization of inflamed tissue.

The most significant contribution of FDG-PET may be the ability to detect infective endocarditis early, when echocardiography is initially negative. When abnormal FDG uptake was included in the modified Duke criteria, it increased the sensitivity to 97% for detecting infective endocarditis on admission, leading some to propose its incorporation as a major criterion.¹⁷ In patients with prosthetic valves and suspected infective endocarditis, FDG-PET was found in one study to have a sensitivity of up to 91% and a specificity of up to 95%.¹⁸

Both FDG-PET and leukocyte scintigraphy have a high sensitivity, specificity, and negative predictive value for cardiac implanted electronic device infection, and should be strongly considered in patients in whom it is suspected but who have negative or inconclusive findings on echocardiography.^14,15

In addition, a common conundrum faced by clinicians with use of echocardiography is the difficulty of differentiating thrombus from infected vegetation on valves or device lead wires. Some evidence indicates that FDG-PET may help to discriminate between vegetation and thrombus, although more rigorous studies are needed before its use for that purpose can be recommended.¹⁹

 

 

Limitations of nuclear studies

Both FDG-PET and leukocyte scintigraphy perform poorly for detecting native-valve infective endocarditis. In a study in which 90% of the patients had native-valve infective endocarditis according to the Duke criteria, FDG-PET had a specificity of 93% but a sensitivity of only 39%.²⁰

Both studies can be cumbersome, laborious, and time-consuming for patients. FDG-PET requires a fasting or glucose-restricted diet before testing, and the test itself can be complicated by development of hyperglycemia, although this is rare.

While FDG-PET is most effective in detecting infections of prosthetic valves and cardiac implanted electronic devices, the results can be falsely positive in patients with a history of recent cardiac surgery (due to ongoing tissue healing), as well as maladies other than infective endocarditis that lead to inflammation, such as vasculitis or malignancy. Similarly, for unclear reasons, leukocyte scintigraphy can yield false-negative results in patients with enterococcal or candidal infective endocarditis.²¹

FDG-PET and leukocyte scintigraphy are more expensive than TEE and cardiac CT²² and are not widely available.

Both tests entail radiation exposure, with the average dose ranging from 7 to 14 mSv. However, this is less than the average amount acquired during percutaneous coronary intervention (16 mSv), and overlaps with the amount in chest CT with contrast when assessing for pulmonary embolism (7 to 9 mSv). Lower doses are possible with optimized protocols.^12,13,15,23

Bottom line for nuclear studies

mgbojikwe_infectiveendocarditis_f2.jpg

FDG-PET and leukocyte scintigraphy are especially useful for patients with a prosthetic valve or cardiac implanted electronic device. However, limitations must be kept in mind.

A suggested algorithm for testing with nuclear imaging is shown in Figure 2.^1,4

CEREBRAL MAGNETIC RESONANCE IMAGING

Cerebral magnetic resonance imaging (MRI) is more sensitive than cerebral CT for detecting emboli in the brain. According to American Heart Association guidelines, cerebral MRI should be done in patients with known or suspected infective endocarditis and neurologic impairment, defined as headaches, meningeal symptoms, or neurologic deficits. It is also often used in neurologically asymptomatic patients with infective endocarditis who have indications for valve surgery to assess for mycotic aneurysms, which are associated with increased intracranial bleeding during surgery.

MRI use in other asymptomatic patients remains controversial.²⁴ In cases with high clinical suspicion for infective endocarditis and no findings on echocardiography, cerebral MRI can increase the sensitivity of the Duke criteria by adding a minor criterion. Some have argued that, in patients with definite infective endocarditis, detecting silent cerebral complications can lead to management changes. However, more studies are needed to determine if there is indeed a group of neurologically asymptomatic infective endocarditis patients for whom cerebral MRI leads to improved outcomes.

Limitations of cerebral MRI

Cerebral MRI cannot be used in patients with non-MRI-compatible implanted hardware.

Gadolinium, the contrast agent typically used, can cause nephrogenic systemic fibrosis in patients who have poor renal function. This rare but serious adverse effect is characterized by irreversible systemic fibrosis affecting skin, muscles, and even visceral tissue such as lungs. The American College of Radiology allows for gadolinium use in patients without acute kidney injury and patients with stable chronic kidney disease with a glomerular filtration rate of at least 30 mL/min/1.73 m². Its use should be avoided in patients with renal failure on replacement therapy, with advanced chronic kidney disease (glomerular filtration rate < 30 mL/min/1.73 m²), or with acute kidney injury, even if they do not need renal replacement therapy.²⁵

Concerns have also been raised about gadolinium retention in the brain, even in patients with normal renal function.^26–28 Thus far, no conclusive clinical adverse effects of retention have been found, although more study is warranted. Nevertheless, the US Food and Drug Administration now requires a black-box warning about this possibility and advises clinicians to counsel patients appropriately.

Bottom line on cerebral MRI

Cerebral MRI should be obtained when a patient presents with definite or possible infective endocarditis with neurologic impairment, such as new headaches, meningismus, or focal neurologic deficits. Routine brain MRI in patients with confirmed infective endocarditis without neurologic symptoms, or those without definite infective endocarditis, is discouraged.

CARDIAC MRI

Cardiac MRI, typically obtained with gadolinium contrast, allows for better 3D assessment of cardiac structures and morphology than echocardiography or CT, and can detect infiltrative cardiac disease, myopericarditis, and much more. It is increasingly used in the field of structural cardiology, but its role for evaluating infective endocarditis remains unclear.

Cardiac MRI does not appear to be better than echocardiography for diagnosing infective endocarditis. However, it may prove helpful in the evaluation of patients known to have infective endocarditis but who cannot be properly evaluated for disease extent because of poor image quality on echocardiography and contraindications to CT.^1,29 Its role is limited in patients with cardiac implanted electronic devices, as most devices are incompatible with MRI use, although newer devices obviate this concern. But even for devices that are MRI-compatible, results are diminished due to an eclipsing effect, wherein the device parts can make it hard to see structures clearly because the “brightness” basically eclipses the surrounding area.⁴

Concerns regarding use of gadolinium as described above need also be considered.

The role of cardiac MRI in diagnosing and managing infective endocarditis may evolve, but at present, the 2017 American College of Cardiology and American Heart Association appropriate-use criteria discourage its use for these purposes.¹⁶

Bottom line for cardiac MRI

Cardiac MRI to evaluate a patient for suspected infective endocarditis is not recommended due to lack of superiority compared with echocardiography or CT, and the risk of nephrogenic systemic fibrosis from gadolinium in patients with renal compromise.

Prompt diagnois of infective endocarditis is critical. Potential consequences of missed or delayed diagnosis, including heart failure, stroke, intracardiac abscess, conduction delays, prosthesis dysfunction, and cerebral emboli, are often catastrophic. Echocardiography is the test used most frequently to evaluate for infective endocarditis, but it misses the diagnosis in almost one-third of cases, and even more often if the patient has a prosthetic valve.

mgbojikwe_infectiveendocarditis_t1.jpg

But now, several sophisticated imaging tests are available that complement echocardiography in diagnosing and assessing infective endocarditis; these include 4-dimensional computed tomography (4D CT), fluorodeoxyglucose positron emission tomography (FDG-PET), and leukocyte scintigraphy. These tests have greatly improved our ability not only to diagnose infective endocarditis, but also to determine the extent and spread of infection, and they aid in perioperative assessment. Abnormal findings on these tests have been incorporated into the European Society of Cardiology’s 2015 modified diagnostic criteria for infective endocarditis.¹

This article details the indications, advantages, and limitations of the various imaging tests for diagnosing and evaluating infective endocarditis (Table 1).

INFECTIVE ENDOCARDITIS IS DIFFICULT TO DIAGNOSE AND TREAT

Infective endocarditis is difficult to diagnose and treat. Clinical and imaging clues can be subtle, and the diagnosis requires a high level of suspicion and visualization of cardiac structures.

Further, the incidence of infective endocarditis is on the rise in the United States, particularly in women and young adults, likely due to intravenous drug use.^2,3

ECHOCARDIOGRAPHY HAS AN IMPORTANT ROLE, BUT IS LIMITED

Echocardiography remains the most commonly performed study for diagnosing infective endocarditis, as it is fast, widely accessible, and less expensive than other imaging tests.

Transthoracic echocardiography (TTE) is often the first choice for testing. However, its sensitivity is only about 70% for detecting vegetations on native valves and 50% for detecting vegetations on prosthetic valves.¹ It is inherently constrained by the limited number of views by which a comprehensive external evaluation of the heart can be achieved. Using a 2-dimensional instrument to view a 3-dimensional object is difficult, and depending on several factors, it can be hard to see vegetations and abscesses that are associated with infective endocarditis. Further, TTE is impeded by obesity and by hyperinflated lungs from obstructive pulmonary disease or mechanical ventilation. It has poor sensitivity for detecting small vegetations and for detecting vegetations and paravalvular complications in patients who have a prosthetic valve or a cardiac implanted electronic device.

Transesophageal echocardiography (TEE) is the recommended first-line imaging test for patients with prosthetic valves and no contraindications to the test. Otherwise, it should be done after TTE if the results of TTE are negative but clinical suspicion for infective endocarditis remains high (eg, because the patient uses intravenous drugs). But although TEE has a higher sensitivity than TTE (up to 96% for vegetations on native valves and 92% for those on prosthetic valves, if performed by an experienced sonographer), it can still miss infective endocarditis. Also, TEE does not provide a significant advantage over TTE in patients who have a cardiac implanted electronic device.^1,4,5

Regardless of whether TTE or TEE is used, they are estimated to miss up to 30% of cases of infective endocarditis and its sequelae.⁴ False-negative findings are likelier in patients who have preexisting severe valvular lesions, prosthetic valves, cardiac implanted electronic devices, small vegetations, or abscesses, or if a vegetation has already broken free and embolized. Furthermore, distinguishing between vegetations and thrombi, cardiac tumors, and myxomatous changes using echocardiography is difficult.

 

 

CARDIAC CT

For patients who have inconclusive results on echocardiography, contraindications to TEE, or poor sonic windows, cardiac CT can be an excellent alternative. It is especially useful in the setting of a prosthetic valve.

Synchronized (“gated”) with the patient’s heart rate and rhythm, CT machines can acquire images during diastole, reducing motion artifact, and can create 3D images of the heart. In addition, newer machines can acquire several images at different points in the heart cycle to add a fourth dimension—time. The resulting 4D images play like short video loops of the beating heart and allow noninvasive assessment of cardiac anatomy with remarkable detail and resolution.

4D CT is increasingly being used in infective endocarditis, and growing evidence indicates that its accuracy is similar to that of TEE in the preoperative evaluation of patients with aortic prosthetic valve endocarditis.⁶ In a study of 28 patients, complementary use of CT angiography led to a change in treatment strategy in 7 (25%) compared with routine clinical workup.⁷ Several studies have found no difference between 4D CT and preoperative TEE in detecting pseudoaneurysm, abscess, or valve dehiscence. TEE and 4D CT also have similar sensitivities for detecting infective endocarditis in native and prosthetic valves.^8,9

mgbojikwe_infectiveendocarditis_f1a.jpg

mgbojikwe_infectiveendocarditis_f1b.jpg

mgbojikwe_infectiveendocarditis_f1c.jpg

Coupled with CT angiography, 4D CT is also an excellent noninvasive way to perioperatively evaluate the coronary arteries without the risks associated with catheterization in those requiring nonemergency surgery (Figure 1A, B, and C).

4D CT performs well for detecting abscess and pseudoaneurysm but has slightly lower sensitivity for vegetations than TEE (91% vs 99%).⁹

Gated CT, PET, or both may be useful in cases of suspected prosthetic aortic valve endocarditis when TEE is negative. Pseudoaneurysms are not well visualized with TEE, and the atrial mitral curtain area is often thickened on TEE in cases of aortic prosthetic valve infective endocarditis that do not definitely involve abscesses. Gated CT and PET show this area better.⁸ This information is important in cases in which a surgeon may be unconvinced that the patient has prosthetic valve endocarditis.

Limitations of 4D cardiac CT

4D CT with or without angiography has limitations. It requires a wide-volume scanner and an experienced reader.

Patients with irregular heart rhythms or uncontrolled tachycardia pose technical problems for image acquisition. Cardiac CT is typically gated (ie, images are obtained within a defined time period) to acquire images during diastole. Ideally, images are acquired when the heart is in mid to late diastole, a time of minimal cardiac motion, so that motion artifact is minimized. To estimate the timing of image acquisition, the cardiac cycle must be predictable, and its duration should be as long as possible. Tachycardia or irregular rhythms such as frequent ectopic beats or atrial fibrillation make acquisition timing difficult, and thus make it nearly impossible to accurately obtain images when the heart is at minimum motion, limiting assessment of cardiac structures or the coronary tree.^4,10

Extensive coronary calcification can hinder assessment of the coronary tree by CT coronary angiography.

Contrast exposure may limit the use of CT in some patients (eg, those with contrast allergies or renal dysfunction). However, modern scanners allow for much smaller contrast boluses without decreasing sensitivity.

4D CT involves radiation exposure, especially when done with angiography, although modern scanners have greatly reduced exposure. The average radiation dose in CT coronary angiography is 2.9 to 5.9 mSv¹¹ compared with 7 mSv in diagnostic cardiac catheterization (without angioplasty or stenting) or 16 mSv in routine CT of the abdomen and pelvis with contrast.^12,13 In view of the morbidity and mortality risks associated with infective endocarditis, especially if the diagnosis is delayed, this small radiation exposure may be justifiable.

Bottom line for cardiac CT

4D CT is an excellent alternative to echocardiography for select patients. Clinicians should strongly consider this study in the following situations:

• Patients with a prosthetic valve
• Patients who are strongly suspected of having infective endocarditis but who have a poor sonic window on TTE or TEE, as can occur with chronic obstructive lung disease, morbid obesity, or previous thoracic or cardiovascular surgery
• Patients who meet clinical indications for TEE, such as having a prosthetic valve or a high suspicion for native valve infective endocarditis with negative TTE, but who have contraindications to TEE
• As an alternative to TEE for preoperative evaluation in patients with known infective endocarditis.

Patients with tachycardia or irregular heart rhythms are not good candidates for this test.

FDG-PET AND LEUKOCYTE SCINTIGRAPHY

FDG-PET and leukocyte scintigraphy are other options for diagnosing infective endocarditis and determining the presence and extent of intra- and extracardiac infection. They are more sensitive than echocardiography for detecting infection of cardiac implanted electronic devices such as ventricular assist devices, pacemakers, implanted cardiac defibrillators, and cardiac resynchronization therapy devices.^14–16

The utility of FDG-PET is founded on the uptake of ¹⁸F-fluorodeoxyglucose by cells, with higher uptake taking place in cells with higher metabolic activity (such as in areas of inflammation). Similarly, leukocyte scintigraphy relies on the use of radiolabeled leukocytes (ie, leukocytes previously extracted from the patient, labelled, and re-introduced into the patient) to allow for localization of inflamed tissue.

The most significant contribution of FDG-PET may be the ability to detect infective endocarditis early, when echocardiography is initially negative. When abnormal FDG uptake was included in the modified Duke criteria, it increased the sensitivity to 97% for detecting infective endocarditis on admission, leading some to propose its incorporation as a major criterion.¹⁷ In patients with prosthetic valves and suspected infective endocarditis, FDG-PET was found in one study to have a sensitivity of up to 91% and a specificity of up to 95%.¹⁸

Both FDG-PET and leukocyte scintigraphy have a high sensitivity, specificity, and negative predictive value for cardiac implanted electronic device infection, and should be strongly considered in patients in whom it is suspected but who have negative or inconclusive findings on echocardiography.^14,15

In addition, a common conundrum faced by clinicians with use of echocardiography is the difficulty of differentiating thrombus from infected vegetation on valves or device lead wires. Some evidence indicates that FDG-PET may help to discriminate between vegetation and thrombus, although more rigorous studies are needed before its use for that purpose can be recommended.¹⁹

 

 

Limitations of nuclear studies

Both FDG-PET and leukocyte scintigraphy perform poorly for detecting native-valve infective endocarditis. In a study in which 90% of the patients had native-valve infective endocarditis according to the Duke criteria, FDG-PET had a specificity of 93% but a sensitivity of only 39%.²⁰

Both studies can be cumbersome, laborious, and time-consuming for patients. FDG-PET requires a fasting or glucose-restricted diet before testing, and the test itself can be complicated by development of hyperglycemia, although this is rare.

While FDG-PET is most effective in detecting infections of prosthetic valves and cardiac implanted electronic devices, the results can be falsely positive in patients with a history of recent cardiac surgery (due to ongoing tissue healing), as well as maladies other than infective endocarditis that lead to inflammation, such as vasculitis or malignancy. Similarly, for unclear reasons, leukocyte scintigraphy can yield false-negative results in patients with enterococcal or candidal infective endocarditis.²¹

FDG-PET and leukocyte scintigraphy are more expensive than TEE and cardiac CT²² and are not widely available.

Both tests entail radiation exposure, with the average dose ranging from 7 to 14 mSv. However, this is less than the average amount acquired during percutaneous coronary intervention (16 mSv), and overlaps with the amount in chest CT with contrast when assessing for pulmonary embolism (7 to 9 mSv). Lower doses are possible with optimized protocols.^12,13,15,23

Bottom line for nuclear studies

mgbojikwe_infectiveendocarditis_f2.jpg

FDG-PET and leukocyte scintigraphy are especially useful for patients with a prosthetic valve or cardiac implanted electronic device. However, limitations must be kept in mind.

A suggested algorithm for testing with nuclear imaging is shown in Figure 2.^1,4

CEREBRAL MAGNETIC RESONANCE IMAGING

Cerebral magnetic resonance imaging (MRI) is more sensitive than cerebral CT for detecting emboli in the brain. According to American Heart Association guidelines, cerebral MRI should be done in patients with known or suspected infective endocarditis and neurologic impairment, defined as headaches, meningeal symptoms, or neurologic deficits. It is also often used in neurologically asymptomatic patients with infective endocarditis who have indications for valve surgery to assess for mycotic aneurysms, which are associated with increased intracranial bleeding during surgery.

MRI use in other asymptomatic patients remains controversial.²⁴ In cases with high clinical suspicion for infective endocarditis and no findings on echocardiography, cerebral MRI can increase the sensitivity of the Duke criteria by adding a minor criterion. Some have argued that, in patients with definite infective endocarditis, detecting silent cerebral complications can lead to management changes. However, more studies are needed to determine if there is indeed a group of neurologically asymptomatic infective endocarditis patients for whom cerebral MRI leads to improved outcomes.

Limitations of cerebral MRI

Cerebral MRI cannot be used in patients with non-MRI-compatible implanted hardware.

Gadolinium, the contrast agent typically used, can cause nephrogenic systemic fibrosis in patients who have poor renal function. This rare but serious adverse effect is characterized by irreversible systemic fibrosis affecting skin, muscles, and even visceral tissue such as lungs. The American College of Radiology allows for gadolinium use in patients without acute kidney injury and patients with stable chronic kidney disease with a glomerular filtration rate of at least 30 mL/min/1.73 m². Its use should be avoided in patients with renal failure on replacement therapy, with advanced chronic kidney disease (glomerular filtration rate < 30 mL/min/1.73 m²), or with acute kidney injury, even if they do not need renal replacement therapy.²⁵

Concerns have also been raised about gadolinium retention in the brain, even in patients with normal renal function.^26–28 Thus far, no conclusive clinical adverse effects of retention have been found, although more study is warranted. Nevertheless, the US Food and Drug Administration now requires a black-box warning about this possibility and advises clinicians to counsel patients appropriately.

Bottom line on cerebral MRI

Cerebral MRI should be obtained when a patient presents with definite or possible infective endocarditis with neurologic impairment, such as new headaches, meningismus, or focal neurologic deficits. Routine brain MRI in patients with confirmed infective endocarditis without neurologic symptoms, or those without definite infective endocarditis, is discouraged.

CARDIAC MRI

Cardiac MRI, typically obtained with gadolinium contrast, allows for better 3D assessment of cardiac structures and morphology than echocardiography or CT, and can detect infiltrative cardiac disease, myopericarditis, and much more. It is increasingly used in the field of structural cardiology, but its role for evaluating infective endocarditis remains unclear.

Cardiac MRI does not appear to be better than echocardiography for diagnosing infective endocarditis. However, it may prove helpful in the evaluation of patients known to have infective endocarditis but who cannot be properly evaluated for disease extent because of poor image quality on echocardiography and contraindications to CT.^1,29 Its role is limited in patients with cardiac implanted electronic devices, as most devices are incompatible with MRI use, although newer devices obviate this concern. But even for devices that are MRI-compatible, results are diminished due to an eclipsing effect, wherein the device parts can make it hard to see structures clearly because the “brightness” basically eclipses the surrounding area.⁴

Concerns regarding use of gadolinium as described above need also be considered.

The role of cardiac MRI in diagnosing and managing infective endocarditis may evolve, but at present, the 2017 American College of Cardiology and American Heart Association appropriate-use criteria discourage its use for these purposes.¹⁶

Bottom line for cardiac MRI

Cardiac MRI to evaluate a patient for suspected infective endocarditis is not recommended due to lack of superiority compared with echocardiography or CT, and the risk of nephrogenic systemic fibrosis from gadolinium in patients with renal compromise.

Autism spectrum disorder (ASD) has increased significantly over the past 40 years. Even in the past 2 decades, the prevalence increased from 6.7 per 1,000 in 2000¹ to 14.6 per 1,000 in 2012—1 in 59 people.² Of those with ASD, 46% have an intelligence quotient (IQ) greater than 85, meaning they are of average or above-average intelligence.¹

See related editorial

As more children with autism become adults, understanding this condition across the life span grows paramount. While many studies have focused on understanding how diagnosis and treatment can help young children, few have focused on adults with autism and how primary care teams can better assist these individuals. However, this is changing, with studies of the benefits of employment programs and pharmacologic treatment, and reproductive health needs of adults with ASD. Here we provide an updated review of ASD in adult patients.

NO MORE ASPERGER SYNDROME— IT’S ON THE SPECTRUM NOW

As the scientific understanding of autism has expanded, revisions in the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5),³ published in 2013, have paralleled these advances. For many adult patients with autism who were evaluated as children, these revisions have led to changes in diagnosis and available services.

swetlik_adultautismspectrumdisorder_t1.jpg

In the previous edition (DSM-IV-TR, published in 2000),⁴ autistic disorder and Asperger syndrome were separate (Table 1). However, DSM-5 lumped autistic disorder and Asperger disorder together under the diagnosis of ASD; this leaves it to the clinician to specify whether the patient with ASD has accompanying intellectual or language impairment and to assign a level of severity based on communication deficits and restrictive behaviors.

The shift in diagnosis was worrisome for some, particularly for clinicians treating patients with DSM-IV Asperger syndrome, who lost this diagnostic label. Concerns that patients with Asperger syndrome may not meet the DSM-5 criteria for ASD were validated by a systematic review showing that only 50% to 75% of patients with DSM-IV autistic disorder, Asperger syndrome, or pervasive developmental disorder not otherwise specified (PDD-NOS) met the DSM-5 criteria for ASD.⁵ Most of those who no longer met the criteria for ASD carried a DSM-IV diagnosis of Asperger syndrome or PDD-NOS or had an IQ over 70.⁵ Nevertheless, these individuals may struggle with impairing symptoms related to repetitive behaviors or communication or may be affected by learning or social-emotional disabilities. Additionally, even if they meet the criteria for ASD, some may identify with the Asperger syndrome label and fear they will be stigmatized should they be classified as having the more general ASD.^6,7

Although future revisions to the DSM may include further changes in classification, grouping adults with ASD according to their functional and cognitive ability may allow for pragmatic characterization of their needs. At least 3 informal groupings of autistic adults have been described that integrate cognitive ability and independence⁸:

• Those with low cognitive and social abilities, who need lifelong support
• Those with midrange cognitive and social limitations but who can complete their work in special education classes; they often find employment in supervised workshops or other work with repetitive tasks
• Those who have greater cognitive ability and some social skills; they may proceed to college and employment and live independently.

UNCERTAIN PROGNOSIS

Prognostication for people with ASD remains an area of research. Some adults experience a reduction in symptoms as they age, with significant improvements in speech and, sometimes, modest improvements in restrictive and repetitive behaviors.^9,10

Nevertheless, autism remains a lifelong disorder for many. Adults may still require significant support and may experience impairment, particularly in social interaction.¹⁰ In longitudinal studies, only 15% to 27% of patients with ASD are characterized as having a positive outcome (often defined as variables related to independent function, near-normal relationships, employment, or a quantified reduction in core symptoms), and many experience significant dependency into adulthood.^10–13

IQ has been cited as a possible prognostic factor,^10,13 with an IQ below 70 associated with poorer outcome, although an IQ above 70 does not necessarily confer a positive outcome. Less-severe impairment in speech at baseline in early childhood also suggests better outcomes in adulthood.¹⁰

As we see more adults with autism, studies that include both children and adults, such as the Longitudinal European Autism Cohort, will be important to characterize the natural history, comorbidities, and genetics of ASD and may help provide more specific predictors of disease course into adulthood.¹⁴

 

 

ACHIEVING A DIAGNOSIS FOR ADULT PATIENTS WITH SUSPECTED AUTISM

While many patients are recognized as having autism in early to mid-childhood, some adults may not receive a formal diagnosis until much later in life. Those with fluent language and normal-range IQ are likely to be overlooked.¹⁵ People with ASD may have had mild symptoms during childhood that did not impair their functioning until demands of daily life exceeded their capacities in adulthood. Alternatively, parents of a child with newly diagnosed ASD may realize that they themselves or another adult family member also show signs of it.

The UK National Institute of Health and Care Excellence suggests that assessment should be considered if the patient meets psychiatric diagnostic criteria and one of the following:

• Difficulty obtaining or sustaining employment or education
• Difficulty initiating or sustaining social relationships
• Past or current contact with mental health or learning disability services
• History of a neurodevelopmental or mental health disorder.^15,16

Currently, diagnosis typically involves a multidisciplinary approach, with psychiatric assessment, neuropsychological testing, and speech and language evaluation.¹⁷ Providers may need to refer patients for these services, sometimes at the patient’s request, if previous mental health misdiagnoses are suspected, if patients report symptoms or impairment consistent with ASD, or if benefits, services, or accommodations, such as a coach in the workplace, are needed.

Diagnosing ASD in adults can be difficult, given that the gold-standard diagnostic tests such as the Autism Diagnostic Observation Schedule-2 (ADOS-2)¹⁸ and the Autism Diagnostic Interview-Revised (ADI-R)¹⁹ are typically used to diagnose autism in children. However, Module 4 in the ADOS-2 was developed for adolescents and older patients with fluent language and has shown at least moderate power to distinguish adults with ASD from those without ASD.^18,20

An initial psychiatric assessment should include a thorough history taken from the patient and, if applicable, the patient’s caregiver, as well as a psychiatric interview of the patient. Neuropsychological testing should include evaluation of cognitive function, social functioning (using the ADOS-2 for adults without intellectual disability, the ADI-R, or both), and adaptive functioning (using the Vineland Adaptive Behavior Scales, second edition²¹).

Evaluation of speech and language is particularly important in patients with limited language ability and should include both expressive and receptive language abilities. Serial testing every few years, as is often recommended in childhood, may help establish the pattern of impairment over time.

Comorbid psychiatric disorders are common

Many people with ASD also have other psychiatric disorders,^17,22 which clinicians should keep in mind when seeing an adult seeking evaluation for ASD.

Attention-deficit/hyperactivity disorder is present at higher rates in patients of average intellectual function with ASD than in the general population.²³

Anxiety disorders, including obsessive-compulsive disorder, were found to often coexist with autism in a sample of adults with autism without intellectual disability,^24,25 and approximately 40% of youths with ASD have at least 1 comorbid anxiety disorder.²⁶

Mood disorders are also prevalent in adults with ASD, with a small study showing that 70% of adults with DSM-IV Asperger syndrome had at least 1 depressive episode in their lifetime.²⁷

BEHAVIORAL AND PHARMACOLOGIC THERAPIES FOR THE ADULT PATIENT

swetlik_adultautismspectrumdisorder_t2.jpg

Treatment of adults with ASD should be individualized based on the challenges they are facing. Many, including those with average or above-average intelligence, struggle with interpersonal relationships, employment, housing, other health conditions, and quality of life.²⁸ Thus, behavioral services and programs should be tailored to help the patient with current challenges (Table 2).

Services and medications for adults with ASD are discussed below. These will vary by individual, and services available may vary by region.

Historically, vocational and social outcomes have been poor for adults with ASD. It is estimated that most larger universities may be home to 100 to 300 students with ASD. To combat isolation, the University of California, Los Angeles, the University of Alabama, and others provide special support services, including group social activities such as board games and individual coaching.⁸ Nevertheless, half of the students with autism who attend institutions of higher learning leave without completing their intended degree.²⁹ Many still struggle to establish meaningful friendships or romantic relationships.²⁹

Planning for a transition of care

Healthcare transition planning is important  but is strikingly underused.³⁰ Individual providers, including adult psychiatrists, vary in their level of training and comfort in diagnosing, treating, and monitoring adults with autism. Youths with ASD are half as likely to receive healthcare transition services as other youths with special healthcare needs.³¹

Pediatric providers, including pediatric psychiatrists, developmental behavioral specialists, and pediatric neurologists, may be best equipped to treat young adult patients or to refer patients to appropriate generalists and specialists comfortable with autism-specific transition of care. The question of eligibility for services is important to patients and families during the transition period, with many parents and professionals unaware of services available to them.³² Receiving adequate transition services is enabled by having a medical home during childhood—that is, a comprehensive, centralized medical record, culturally competent care, interaction with schools, and patient access to clear, unbiased information.³¹

Ideally, in our experience, transitioning should be discussed well before the child ages out of the pediatric provider’s practice. If necessary, healthcare transition services should include 4 components:

• Discussing the switch to a new physician who treats adults
• Discussing changing healthcare needs as an adult
• Planning insurance coverage as an adult
• Encouragement by the physician for the child to take age-appropriate responsibility for his or her healthcare.^31,33

Tools such as the Got Transition checklist from the National Health Care Transition Center can provide support during this process.³⁴

 

Other services

Other services provided as an extension or adjunct to the medical home in early adulthood may include customized vocational or employment training, specialized mentorship or support in a college setting, housing support, and psychological services.³⁵

Community-based programs that emphasize leisure have been shown to improve participants’ independence and quality of life.³⁶ Similarly, participants in programs that emphasized supported employment, with a job coach, on-the-job support, collaboration with the participant’s larger social support network, and selection of tasks to match an individual’s abilities and strengths, demonstrated improved cognitive performance, particularly executive functioning,³⁷ and employment.^38,39 These programs work best for patients who have mild to moderate symptoms.^37,39

Patients with symptoms that are more severe may do better in a residential program. Many of these programs maintain an emphasis on vocational and social skills development. One such long-standing program is Bittersweet Farms, a rural farming community in Ohio for adults with ASD, where individuals with moderate to low function live in a group setting, with emphasis on scheduled, meaningful work including horticulture, animal care, carpentry; and activities of daily living.⁴⁰

Studies of patients across the autism spectrum have generally found better outcomes when vocational support is given, but larger and randomized studies are needed to characterize how to best support these individuals after they leave high school.⁴¹

Psychological services such as applied behavioral therapy, social cognition training, cognitive behavioral therapy, and mindfulness training may be particularly useful in adults.^42–44

Some versions of applied behavioral therapy, such as the Early Start Denver Model,⁴⁵ have been found to be cost-effective and offset some expenses in the care of children with autism, using play-based and relationship-based interventions to promote development across domains while reducing symptoms.

In randomized controlled trials, modified cognitive behavioral therapy⁴³ and mindfulness⁴⁴ were shown to reduce symptoms of anxiety, obsessive-compulsive disorder, and depression.

Dialectical behavior therapy, used to find a balance between accepting oneself and desiring to change, may help in some circumstances to regulate emotions and reduce reactivity and lability, although large randomized clinical trials have not been conducted in the ASD population.⁴⁶

Drug therapy

Medications may be appropriate to manage symptoms or comorbid conditions in adults with ASD. Over 75% adults with ASD have been found to use psychotropic medications.⁴⁷ However, although these drugs have been approved for treating behaviors commonly associated with ASD, none of them provide definitive treatment for this disorder, and they have not been rigorously tested or approved for use in adults with ASD.⁴⁸

Irritability and aggression associated with ASD can be treated with risperidone (approved for children over age 5), aripiprazole (approved for children ages 6–17), clozapine, or haloperidol.⁴⁹

Aberrant social behavior can be treated with risperidone.⁵⁰ Treatments under investigation include oxytocin and secretin.⁴⁹

While no approved drug has been shown to improve social communication,⁵¹ balovaptan, a vasopressin V1a agonist, has shown potential and has been granted breakthrough status by the US Food and Drug Administration for treating challenging behaviors in adults, with additional studies ongoing in children.^52,53

Repetitive behaviors, if the patient finds them impairing, can be managed with selective serotonin reuptake inhibitors.⁴⁹

Much more study of drug therapy in adults with ASD is needed to fully understand the best approaches to psychotropic medication use, including appropriate classes and effective dosage, in this population.

SEX: UNEXPLORED TERRITORY

The reproductive health needs of people with autism remain largely underexplored.⁵⁴ Historically, individuals with ASD were thought to have little interest in sexual activity or parenthood, owing to the nature of the core symptoms of the disorder. This has been shown to be untrue, particularly as studies on this topic began to engage in direct interviews with people with ASD, rather than solely gathering information from caregivers or parents. The findings reinforce the importance of broaching this component of health in this population, for the following reasons:

Adults with ASD are at increased risk of sexual victimization, with nearly 4 out of 5 reporting unwanted sexual advances, coercion, or rape.⁵⁵

They have a smaller pool of knowledge with respect to sexual health. They report⁵⁶ that they learned about sex from television and from “making mistakes.” They use fewer sources. They are less likely to speak to peers and figures of authority to gain knowledge about sexually transmitted infections, sexual behaviors, and contraception. And they are more likely to use forms of nonsocial media, such as television, for information.⁵⁵

They report more concerns about the future with respect to sexual behavior, suggesting the need for targeted sexual education programs.⁵⁶

College-age young adults with ASD who misread communication may be particularly affected by Title IX, which requires schools to promptly investigate reports of sexual harassment and sexual assault, should they struggle to comport themselves appropriately.⁵⁷ Early and frank conversations about issues of consent and appropriate displays of interest and affection may better equip youth to navigate new social scenarios as they plan to leave a supervised home environment for college or the workforce.

Gender identification: Male, female, other

In one study, 77.8% of birth-sex males with ASD said they identified as men, and 67.1% of birth-sex females identified as women,  compared with 93.1% of birth-sex males and 87.3% of birth-sex females without ASD. Many of the remaining individuals with ASD reported a transgender, genderqueer, or other gender identity.⁵⁸ Some studies have found females with ASD report a gay or bisexual orientation more often than males with ASD.^59–61

Adolescents and young adults may be exploring their changing bodies, sexual preferences, and gender roles, and as for all people at this age, these roles emerge against a backdrop of familial and societal expectations that may or may not be concordant with their own projected path regarding sexuality and reproductive health.⁶²

Having the conversation

As with non-ASD patients, a thorough sexual history should be collected via open-ended questions when possible to determine types of sexual activity and partners.

Education of the patient, alongside caregivers and parents, about healthy and safe sexual practices, screening for sexual violence, and hormonal and nonhormonal contraception options are important components of care for this population.

 

CAREGIVER STRESS MAY PERSIST INTO PATIENT’S ADULTHOOD

Caregiver burden is a monumental concern for parents or others who may have lifelong primary responsibility for these neurodiverse adults.⁶³ Family members may feel isolated and may feel they have encountered many barriers to services.⁶⁴ Remaining sensitive, knowledgeable, and inquisitive about the types of support that are needed may help forge a trusting relationship between the provider and the family.

Parents of children with ASD have been reported to experience worse physical and emotional health than parents whose children do not have developmental disabilities.^63,65 These disparities have been found to persist  as their children enter adolescence and young adulthood.^66,67 Parents of children with ASD report more anxiety, depression, and distress compared with parents of children without ASD,⁶³ and parents themselves may be affected by ASD symptoms, which has been linked to increased parenting stress.⁶⁸ Some studies have found blunted cortisol responses,^63,69,70 and some,⁷¹ but not all,⁶³ have found elevated blood pressure in caregivers of children with developmental disabilities. Headache, backache, muscle soreness, and fatigue may also be commonly reported.⁶⁷

In our experience, caregivers are tremendously appreciative when provided connections to adult ASD services and support systems as their child ages. The school system and other formal support systems often assist until the time of transition into adulthood. This transition can be stressful for the adolescent and family alike, and informal support systems such as friends and family may become increasingly crucial, particularly if the adolescent still lives at home.^72,73

The affected young adult’s unmet needs, as perceived by the caregiver, have been found to be significantly associated with caregiver burden, whereas the severity of the adult patient’s ASD symptoms has not.⁶⁶ Therefore, it may be helpful to ask caregivers whether they perceive any unmet needs, regardless of the clinician’s perception of the severity of the patient’s ASD symptoms. Providing support to address these needs, particularly those relating to the child’s mood disorders, communication, social needs, safety, and daytime activities, may be the domains of support that most effectively reduce the caregiver burden in this population.⁶⁶

Caregiver positivity, lower stress levels, and increased social support, particularly in the form of friends and family members providing no-cost assistance to caregivers whose children do not live independently,⁷⁴ have been linked to better outcomes for caregivers.^70,74,75 Rigorous studies that examine caregiver burden as individuals with ASD enter mid- and late-adulthood are limited.

THE ROLE OF THE INTERNIST IN CARING FOR ADULTS WITH AUTISM

A major challenge for many adults with ASD is the transition from services provided during childhood to those provided in adulthood. While children with autism have subspecialty providers who diagnose and manage their condition, including developmental-behavioral pediatricians, pediatric neurologists, and child psychiatrists, adults with autism may have fewer options.

Autism centers are becoming more available across the nation, and many provide care across the life span. However, depending on a patient’s needs, the primary care provider may need to manage residual symptoms as the patient transitions from pediatric to adult care, ultimately deciding when and where to refer the patient.

The patient’s family should pay close attention to function and mood around the time the patient leaves the structure of high school, and they should build rapport with a primary care provider they can turn to if problems persist or arise. Referrals for behavioral therapy and for social work, job training, and vocational support can greatly benefit patients as they transition to young adulthood. Referrals and suggestions for social support can also help caregivers.

Medical care

Deciding when and how to medicate the patient for symptoms of autism and related behaviors necessitates consideration of the patient’s impairment, side effects of the medication, and the impact medications may have on the patient’s other conditions. Disordered eating, mood problems, anxiety, and attention-deficit/hyperactivity disorder should be considered, and, as in all patients, regular screenings of mental health status should be conducted.^76,77

Comorbid medical conditions may cause worsening of a patient’s known behavioral symptoms or may precipitate new behaviors or aggression as a result of pain or discomfort, particularly in patients with limited speech. A change in stereotypes or increased irritability warrants a thoughtful investigation for a cause other than ASD before adding or increasing behavioral medications. Common comorbid conditions include gastrointestinal distress, most commonly constipation and diarrhea in an idiopathic ASD population, with increasing ASD symptom severity correlating with increased odds of a gastrointestinal problem.⁷⁸ Allergies, sleep disorders, seizures, and other psychiatric conditions are also frequent.⁷⁹

Preventive care, including vaccinations, should be given as scheduled. Caregivers and patients can be reminded if needed that vaccines do not cause or worsen autism, and vaccination is intended to improve the safety of the patient and those around them, protecting against potentially life-threatening disease. Regular dental care visits, particularly for patients who are using medications that may affect tooth or gingival health,⁸⁰ and regular visits to an optometrist or ophthalmologist for screening of vision are also advised.

Adverse effects. Weight gain and metabolic syndrome are common adverse effects of medications used for behavioral management, and the primary care physician may uncover diabetes, cardiac disorders, and hyperlipidemia. Patients with ASD may be particularly sensitive to the effects of medications and therefore may require a lower dose or a slower titration than other patients. Working with a behavioral team, careful weaning of psychiatric medications to the minimum needed is strongly recommended whenever possible.⁸¹

TAKE-HOME POINTS

As more adults with autism enter society, they may require varying levels of support from the healthcare community to ensure that therapeutic gains from childhood persist, allowing them to achieve maximal functional potential.

Adults with ASD may have a high, normal, or low IQ and intellectual capability. Knowledge of this and of the patient’s symptom severity and presence of comorbid psychiatric and other health conditions can help the clinician guide the patient to appropriate social services and pharmacologic treatments.

Individualized support in the workplace, as well as education regarding sexual health, can help improve outcomes for affected individuals.

Caregiver burden for individuals with autism can be high, but it can be mitigated by social support.

Further research regarding appropriate diagnostic instruments in adulthood and appropriate treatments for impairing autism-related symptoms across the life span may be particularly helpful in supporting this patient population.

Autism spectrum disorder (ASD) has increased significantly over the past 40 years. Even in the past 2 decades, the prevalence increased from 6.7 per 1,000 in 2000¹ to 14.6 per 1,000 in 2012—1 in 59 people.² Of those with ASD, 46% have an intelligence quotient (IQ) greater than 85, meaning they are of average or above-average intelligence.¹

See related editorial

As more children with autism become adults, understanding this condition across the life span grows paramount. While many studies have focused on understanding how diagnosis and treatment can help young children, few have focused on adults with autism and how primary care teams can better assist these individuals. However, this is changing, with studies of the benefits of employment programs and pharmacologic treatment, and reproductive health needs of adults with ASD. Here we provide an updated review of ASD in adult patients.

NO MORE ASPERGER SYNDROME— IT’S ON THE SPECTRUM NOW

As the scientific understanding of autism has expanded, revisions in the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5),³ published in 2013, have paralleled these advances. For many adult patients with autism who were evaluated as children, these revisions have led to changes in diagnosis and available services.

swetlik_adultautismspectrumdisorder_t1.jpg

In the previous edition (DSM-IV-TR, published in 2000),⁴ autistic disorder and Asperger syndrome were separate (Table 1). However, DSM-5 lumped autistic disorder and Asperger disorder together under the diagnosis of ASD; this leaves it to the clinician to specify whether the patient with ASD has accompanying intellectual or language impairment and to assign a level of severity based on communication deficits and restrictive behaviors.

The shift in diagnosis was worrisome for some, particularly for clinicians treating patients with DSM-IV Asperger syndrome, who lost this diagnostic label. Concerns that patients with Asperger syndrome may not meet the DSM-5 criteria for ASD were validated by a systematic review showing that only 50% to 75% of patients with DSM-IV autistic disorder, Asperger syndrome, or pervasive developmental disorder not otherwise specified (PDD-NOS) met the DSM-5 criteria for ASD.⁵ Most of those who no longer met the criteria for ASD carried a DSM-IV diagnosis of Asperger syndrome or PDD-NOS or had an IQ over 70.⁵ Nevertheless, these individuals may struggle with impairing symptoms related to repetitive behaviors or communication or may be affected by learning or social-emotional disabilities. Additionally, even if they meet the criteria for ASD, some may identify with the Asperger syndrome label and fear they will be stigmatized should they be classified as having the more general ASD.^6,7

Although future revisions to the DSM may include further changes in classification, grouping adults with ASD according to their functional and cognitive ability may allow for pragmatic characterization of their needs. At least 3 informal groupings of autistic adults have been described that integrate cognitive ability and independence⁸:

• Those with low cognitive and social abilities, who need lifelong support
• Those with midrange cognitive and social limitations but who can complete their work in special education classes; they often find employment in supervised workshops or other work with repetitive tasks
• Those who have greater cognitive ability and some social skills; they may proceed to college and employment and live independently.

UNCERTAIN PROGNOSIS

Prognostication for people with ASD remains an area of research. Some adults experience a reduction in symptoms as they age, with significant improvements in speech and, sometimes, modest improvements in restrictive and repetitive behaviors.^9,10

Nevertheless, autism remains a lifelong disorder for many. Adults may still require significant support and may experience impairment, particularly in social interaction.¹⁰ In longitudinal studies, only 15% to 27% of patients with ASD are characterized as having a positive outcome (often defined as variables related to independent function, near-normal relationships, employment, or a quantified reduction in core symptoms), and many experience significant dependency into adulthood.^10–13

IQ has been cited as a possible prognostic factor,^10,13 with an IQ below 70 associated with poorer outcome, although an IQ above 70 does not necessarily confer a positive outcome. Less-severe impairment in speech at baseline in early childhood also suggests better outcomes in adulthood.¹⁰

As we see more adults with autism, studies that include both children and adults, such as the Longitudinal European Autism Cohort, will be important to characterize the natural history, comorbidities, and genetics of ASD and may help provide more specific predictors of disease course into adulthood.¹⁴

 

 

ACHIEVING A DIAGNOSIS FOR ADULT PATIENTS WITH SUSPECTED AUTISM

While many patients are recognized as having autism in early to mid-childhood, some adults may not receive a formal diagnosis until much later in life. Those with fluent language and normal-range IQ are likely to be overlooked.¹⁵ People with ASD may have had mild symptoms during childhood that did not impair their functioning until demands of daily life exceeded their capacities in adulthood. Alternatively, parents of a child with newly diagnosed ASD may realize that they themselves or another adult family member also show signs of it.

The UK National Institute of Health and Care Excellence suggests that assessment should be considered if the patient meets psychiatric diagnostic criteria and one of the following:

• Difficulty obtaining or sustaining employment or education
• Difficulty initiating or sustaining social relationships
• Past or current contact with mental health or learning disability services
• History of a neurodevelopmental or mental health disorder.^15,16

Currently, diagnosis typically involves a multidisciplinary approach, with psychiatric assessment, neuropsychological testing, and speech and language evaluation.¹⁷ Providers may need to refer patients for these services, sometimes at the patient’s request, if previous mental health misdiagnoses are suspected, if patients report symptoms or impairment consistent with ASD, or if benefits, services, or accommodations, such as a coach in the workplace, are needed.

Diagnosing ASD in adults can be difficult, given that the gold-standard diagnostic tests such as the Autism Diagnostic Observation Schedule-2 (ADOS-2)¹⁸ and the Autism Diagnostic Interview-Revised (ADI-R)¹⁹ are typically used to diagnose autism in children. However, Module 4 in the ADOS-2 was developed for adolescents and older patients with fluent language and has shown at least moderate power to distinguish adults with ASD from those without ASD.^18,20

An initial psychiatric assessment should include a thorough history taken from the patient and, if applicable, the patient’s caregiver, as well as a psychiatric interview of the patient. Neuropsychological testing should include evaluation of cognitive function, social functioning (using the ADOS-2 for adults without intellectual disability, the ADI-R, or both), and adaptive functioning (using the Vineland Adaptive Behavior Scales, second edition²¹).

Evaluation of speech and language is particularly important in patients with limited language ability and should include both expressive and receptive language abilities. Serial testing every few years, as is often recommended in childhood, may help establish the pattern of impairment over time.

Comorbid psychiatric disorders are common

Many people with ASD also have other psychiatric disorders,^17,22 which clinicians should keep in mind when seeing an adult seeking evaluation for ASD.

Attention-deficit/hyperactivity disorder is present at higher rates in patients of average intellectual function with ASD than in the general population.²³

Anxiety disorders, including obsessive-compulsive disorder, were found to often coexist with autism in a sample of adults with autism without intellectual disability,^24,25 and approximately 40% of youths with ASD have at least 1 comorbid anxiety disorder.²⁶

Mood disorders are also prevalent in adults with ASD, with a small study showing that 70% of adults with DSM-IV Asperger syndrome had at least 1 depressive episode in their lifetime.²⁷

BEHAVIORAL AND PHARMACOLOGIC THERAPIES FOR THE ADULT PATIENT

swetlik_adultautismspectrumdisorder_t2.jpg

Treatment of adults with ASD should be individualized based on the challenges they are facing. Many, including those with average or above-average intelligence, struggle with interpersonal relationships, employment, housing, other health conditions, and quality of life.²⁸ Thus, behavioral services and programs should be tailored to help the patient with current challenges (Table 2).

Services and medications for adults with ASD are discussed below. These will vary by individual, and services available may vary by region.

Historically, vocational and social outcomes have been poor for adults with ASD. It is estimated that most larger universities may be home to 100 to 300 students with ASD. To combat isolation, the University of California, Los Angeles, the University of Alabama, and others provide special support services, including group social activities such as board games and individual coaching.⁸ Nevertheless, half of the students with autism who attend institutions of higher learning leave without completing their intended degree.²⁹ Many still struggle to establish meaningful friendships or romantic relationships.²⁹

Planning for a transition of care

Healthcare transition planning is important  but is strikingly underused.³⁰ Individual providers, including adult psychiatrists, vary in their level of training and comfort in diagnosing, treating, and monitoring adults with autism. Youths with ASD are half as likely to receive healthcare transition services as other youths with special healthcare needs.³¹

Pediatric providers, including pediatric psychiatrists, developmental behavioral specialists, and pediatric neurologists, may be best equipped to treat young adult patients or to refer patients to appropriate generalists and specialists comfortable with autism-specific transition of care. The question of eligibility for services is important to patients and families during the transition period, with many parents and professionals unaware of services available to them.³² Receiving adequate transition services is enabled by having a medical home during childhood—that is, a comprehensive, centralized medical record, culturally competent care, interaction with schools, and patient access to clear, unbiased information.³¹

Ideally, in our experience, transitioning should be discussed well before the child ages out of the pediatric provider’s practice. If necessary, healthcare transition services should include 4 components:

• Discussing the switch to a new physician who treats adults
• Discussing changing healthcare needs as an adult
• Planning insurance coverage as an adult
• Encouragement by the physician for the child to take age-appropriate responsibility for his or her healthcare.^31,33

Tools such as the Got Transition checklist from the National Health Care Transition Center can provide support during this process.³⁴

 

Other services

Other services provided as an extension or adjunct to the medical home in early adulthood may include customized vocational or employment training, specialized mentorship or support in a college setting, housing support, and psychological services.³⁵

Community-based programs that emphasize leisure have been shown to improve participants’ independence and quality of life.³⁶ Similarly, participants in programs that emphasized supported employment, with a job coach, on-the-job support, collaboration with the participant’s larger social support network, and selection of tasks to match an individual’s abilities and strengths, demonstrated improved cognitive performance, particularly executive functioning,³⁷ and employment.^38,39 These programs work best for patients who have mild to moderate symptoms.^37,39

Patients with symptoms that are more severe may do better in a residential program. Many of these programs maintain an emphasis on vocational and social skills development. One such long-standing program is Bittersweet Farms, a rural farming community in Ohio for adults with ASD, where individuals with moderate to low function live in a group setting, with emphasis on scheduled, meaningful work including horticulture, animal care, carpentry; and activities of daily living.⁴⁰

Studies of patients across the autism spectrum have generally found better outcomes when vocational support is given, but larger and randomized studies are needed to characterize how to best support these individuals after they leave high school.⁴¹

Psychological services such as applied behavioral therapy, social cognition training, cognitive behavioral therapy, and mindfulness training may be particularly useful in adults.^42–44

Some versions of applied behavioral therapy, such as the Early Start Denver Model,⁴⁵ have been found to be cost-effective and offset some expenses in the care of children with autism, using play-based and relationship-based interventions to promote development across domains while reducing symptoms.

In randomized controlled trials, modified cognitive behavioral therapy⁴³ and mindfulness⁴⁴ were shown to reduce symptoms of anxiety, obsessive-compulsive disorder, and depression.

Dialectical behavior therapy, used to find a balance between accepting oneself and desiring to change, may help in some circumstances to regulate emotions and reduce reactivity and lability, although large randomized clinical trials have not been conducted in the ASD population.⁴⁶

Drug therapy

Medications may be appropriate to manage symptoms or comorbid conditions in adults with ASD. Over 75% adults with ASD have been found to use psychotropic medications.⁴⁷ However, although these drugs have been approved for treating behaviors commonly associated with ASD, none of them provide definitive treatment for this disorder, and they have not been rigorously tested or approved for use in adults with ASD.⁴⁸

Irritability and aggression associated with ASD can be treated with risperidone (approved for children over age 5), aripiprazole (approved for children ages 6–17), clozapine, or haloperidol.⁴⁹

Aberrant social behavior can be treated with risperidone.⁵⁰ Treatments under investigation include oxytocin and secretin.⁴⁹

While no approved drug has been shown to improve social communication,⁵¹ balovaptan, a vasopressin V1a agonist, has shown potential and has been granted breakthrough status by the US Food and Drug Administration for treating challenging behaviors in adults, with additional studies ongoing in children.^52,53

Repetitive behaviors, if the patient finds them impairing, can be managed with selective serotonin reuptake inhibitors.⁴⁹

Much more study of drug therapy in adults with ASD is needed to fully understand the best approaches to psychotropic medication use, including appropriate classes and effective dosage, in this population.

SEX: UNEXPLORED TERRITORY

The reproductive health needs of people with autism remain largely underexplored.⁵⁴ Historically, individuals with ASD were thought to have little interest in sexual activity or parenthood, owing to the nature of the core symptoms of the disorder. This has been shown to be untrue, particularly as studies on this topic began to engage in direct interviews with people with ASD, rather than solely gathering information from caregivers or parents. The findings reinforce the importance of broaching this component of health in this population, for the following reasons:

Adults with ASD are at increased risk of sexual victimization, with nearly 4 out of 5 reporting unwanted sexual advances, coercion, or rape.⁵⁵

They have a smaller pool of knowledge with respect to sexual health. They report⁵⁶ that they learned about sex from television and from “making mistakes.” They use fewer sources. They are less likely to speak to peers and figures of authority to gain knowledge about sexually transmitted infections, sexual behaviors, and contraception. And they are more likely to use forms of nonsocial media, such as television, for information.⁵⁵

They report more concerns about the future with respect to sexual behavior, suggesting the need for targeted sexual education programs.⁵⁶

College-age young adults with ASD who misread communication may be particularly affected by Title IX, which requires schools to promptly investigate reports of sexual harassment and sexual assault, should they struggle to comport themselves appropriately.⁵⁷ Early and frank conversations about issues of consent and appropriate displays of interest and affection may better equip youth to navigate new social scenarios as they plan to leave a supervised home environment for college or the workforce.

Gender identification: Male, female, other

In one study, 77.8% of birth-sex males with ASD said they identified as men, and 67.1% of birth-sex females identified as women,  compared with 93.1% of birth-sex males and 87.3% of birth-sex females without ASD. Many of the remaining individuals with ASD reported a transgender, genderqueer, or other gender identity.⁵⁸ Some studies have found females with ASD report a gay or bisexual orientation more often than males with ASD.^59–61

Adolescents and young adults may be exploring their changing bodies, sexual preferences, and gender roles, and as for all people at this age, these roles emerge against a backdrop of familial and societal expectations that may or may not be concordant with their own projected path regarding sexuality and reproductive health.⁶²

Having the conversation

As with non-ASD patients, a thorough sexual history should be collected via open-ended questions when possible to determine types of sexual activity and partners.

Education of the patient, alongside caregivers and parents, about healthy and safe sexual practices, screening for sexual violence, and hormonal and nonhormonal contraception options are important components of care for this population.

 

CAREGIVER STRESS MAY PERSIST INTO PATIENT’S ADULTHOOD

Caregiver burden is a monumental concern for parents or others who may have lifelong primary responsibility for these neurodiverse adults.⁶³ Family members may feel isolated and may feel they have encountered many barriers to services.⁶⁴ Remaining sensitive, knowledgeable, and inquisitive about the types of support that are needed may help forge a trusting relationship between the provider and the family.

Parents of children with ASD have been reported to experience worse physical and emotional health than parents whose children do not have developmental disabilities.^63,65 These disparities have been found to persist  as their children enter adolescence and young adulthood.^66,67 Parents of children with ASD report more anxiety, depression, and distress compared with parents of children without ASD,⁶³ and parents themselves may be affected by ASD symptoms, which has been linked to increased parenting stress.⁶⁸ Some studies have found blunted cortisol responses,^63,69,70 and some,⁷¹ but not all,⁶³ have found elevated blood pressure in caregivers of children with developmental disabilities. Headache, backache, muscle soreness, and fatigue may also be commonly reported.⁶⁷

In our experience, caregivers are tremendously appreciative when provided connections to adult ASD services and support systems as their child ages. The school system and other formal support systems often assist until the time of transition into adulthood. This transition can be stressful for the adolescent and family alike, and informal support systems such as friends and family may become increasingly crucial, particularly if the adolescent still lives at home.^72,73

The affected young adult’s unmet needs, as perceived by the caregiver, have been found to be significantly associated with caregiver burden, whereas the severity of the adult patient’s ASD symptoms has not.⁶⁶ Therefore, it may be helpful to ask caregivers whether they perceive any unmet needs, regardless of the clinician’s perception of the severity of the patient’s ASD symptoms. Providing support to address these needs, particularly those relating to the child’s mood disorders, communication, social needs, safety, and daytime activities, may be the domains of support that most effectively reduce the caregiver burden in this population.⁶⁶

Caregiver positivity, lower stress levels, and increased social support, particularly in the form of friends and family members providing no-cost assistance to caregivers whose children do not live independently,⁷⁴ have been linked to better outcomes for caregivers.^70,74,75 Rigorous studies that examine caregiver burden as individuals with ASD enter mid- and late-adulthood are limited.

THE ROLE OF THE INTERNIST IN CARING FOR ADULTS WITH AUTISM

A major challenge for many adults with ASD is the transition from services provided during childhood to those provided in adulthood. While children with autism have subspecialty providers who diagnose and manage their condition, including developmental-behavioral pediatricians, pediatric neurologists, and child psychiatrists, adults with autism may have fewer options.

Autism centers are becoming more available across the nation, and many provide care across the life span. However, depending on a patient’s needs, the primary care provider may need to manage residual symptoms as the patient transitions from pediatric to adult care, ultimately deciding when and where to refer the patient.

The patient’s family should pay close attention to function and mood around the time the patient leaves the structure of high school, and they should build rapport with a primary care provider they can turn to if problems persist or arise. Referrals for behavioral therapy and for social work, job training, and vocational support can greatly benefit patients as they transition to young adulthood. Referrals and suggestions for social support can also help caregivers.

Medical care

Deciding when and how to medicate the patient for symptoms of autism and related behaviors necessitates consideration of the patient’s impairment, side effects of the medication, and the impact medications may have on the patient’s other conditions. Disordered eating, mood problems, anxiety, and attention-deficit/hyperactivity disorder should be considered, and, as in all patients, regular screenings of mental health status should be conducted.^76,77

Comorbid medical conditions may cause worsening of a patient’s known behavioral symptoms or may precipitate new behaviors or aggression as a result of pain or discomfort, particularly in patients with limited speech. A change in stereotypes or increased irritability warrants a thoughtful investigation for a cause other than ASD before adding or increasing behavioral medications. Common comorbid conditions include gastrointestinal distress, most commonly constipation and diarrhea in an idiopathic ASD population, with increasing ASD symptom severity correlating with increased odds of a gastrointestinal problem.⁷⁸ Allergies, sleep disorders, seizures, and other psychiatric conditions are also frequent.⁷⁹

Preventive care, including vaccinations, should be given as scheduled. Caregivers and patients can be reminded if needed that vaccines do not cause or worsen autism, and vaccination is intended to improve the safety of the patient and those around them, protecting against potentially life-threatening disease. Regular dental care visits, particularly for patients who are using medications that may affect tooth or gingival health,⁸⁰ and regular visits to an optometrist or ophthalmologist for screening of vision are also advised.

Adverse effects. Weight gain and metabolic syndrome are common adverse effects of medications used for behavioral management, and the primary care physician may uncover diabetes, cardiac disorders, and hyperlipidemia. Patients with ASD may be particularly sensitive to the effects of medications and therefore may require a lower dose or a slower titration than other patients. Working with a behavioral team, careful weaning of psychiatric medications to the minimum needed is strongly recommended whenever possible.⁸¹

TAKE-HOME POINTS

As more adults with autism enter society, they may require varying levels of support from the healthcare community to ensure that therapeutic gains from childhood persist, allowing them to achieve maximal functional potential.

Adults with ASD may have a high, normal, or low IQ and intellectual capability. Knowledge of this and of the patient’s symptom severity and presence of comorbid psychiatric and other health conditions can help the clinician guide the patient to appropriate social services and pharmacologic treatments.

Individualized support in the workplace, as well as education regarding sexual health, can help improve outcomes for affected individuals.

Caregiver burden for individuals with autism can be high, but it can be mitigated by social support.

Further research regarding appropriate diagnostic instruments in adulthood and appropriate treatments for impairing autism-related symptoms across the life span may be particularly helpful in supporting this patient population.

Autism spectrum disorder (ASD) has increased significantly over the past 40 years. Even in the past 2 decades, the prevalence increased from 6.7 per 1,000 in 2000¹ to 14.6 per 1,000 in 2012—1 in 59 people.² Of those with ASD, 46% have an intelligence quotient (IQ) greater than 85, meaning they are of average or above-average intelligence.¹

See related editorial

As more children with autism become adults, understanding this condition across the life span grows paramount. While many studies have focused on understanding how diagnosis and treatment can help young children, few have focused on adults with autism and how primary care teams can better assist these individuals. However, this is changing, with studies of the benefits of employment programs and pharmacologic treatment, and reproductive health needs of adults with ASD. Here we provide an updated review of ASD in adult patients.

NO MORE ASPERGER SYNDROME— IT’S ON THE SPECTRUM NOW

As the scientific understanding of autism has expanded, revisions in the Diagnostic and Statistical Manual of Mental Disorders, fifth edition (DSM-5),³ published in 2013, have paralleled these advances. For many adult patients with autism who were evaluated as children, these revisions have led to changes in diagnosis and available services.

swetlik_adultautismspectrumdisorder_t1.jpg

In the previous edition (DSM-IV-TR, published in 2000),⁴ autistic disorder and Asperger syndrome were separate (Table 1). However, DSM-5 lumped autistic disorder and Asperger disorder together under the diagnosis of ASD; this leaves it to the clinician to specify whether the patient with ASD has accompanying intellectual or language impairment and to assign a level of severity based on communication deficits and restrictive behaviors.

The shift in diagnosis was worrisome for some, particularly for clinicians treating patients with DSM-IV Asperger syndrome, who lost this diagnostic label. Concerns that patients with Asperger syndrome may not meet the DSM-5 criteria for ASD were validated by a systematic review showing that only 50% to 75% of patients with DSM-IV autistic disorder, Asperger syndrome, or pervasive developmental disorder not otherwise specified (PDD-NOS) met the DSM-5 criteria for ASD.⁵ Most of those who no longer met the criteria for ASD carried a DSM-IV diagnosis of Asperger syndrome or PDD-NOS or had an IQ over 70.⁵ Nevertheless, these individuals may struggle with impairing symptoms related to repetitive behaviors or communication or may be affected by learning or social-emotional disabilities. Additionally, even if they meet the criteria for ASD, some may identify with the Asperger syndrome label and fear they will be stigmatized should they be classified as having the more general ASD.^6,7

Although future revisions to the DSM may include further changes in classification, grouping adults with ASD according to their functional and cognitive ability may allow for pragmatic characterization of their needs. At least 3 informal groupings of autistic adults have been described that integrate cognitive ability and independence⁸:

• Those with low cognitive and social abilities, who need lifelong support
• Those with midrange cognitive and social limitations but who can complete their work in special education classes; they often find employment in supervised workshops or other work with repetitive tasks
• Those who have greater cognitive ability and some social skills; they may proceed to college and employment and live independently.

UNCERTAIN PROGNOSIS

Prognostication for people with ASD remains an area of research. Some adults experience a reduction in symptoms as they age, with significant improvements in speech and, sometimes, modest improvements in restrictive and repetitive behaviors.^9,10

Nevertheless, autism remains a lifelong disorder for many. Adults may still require significant support and may experience impairment, particularly in social interaction.¹⁰ In longitudinal studies, only 15% to 27% of patients with ASD are characterized as having a positive outcome (often defined as variables related to independent function, near-normal relationships, employment, or a quantified reduction in core symptoms), and many experience significant dependency into adulthood.^10–13

IQ has been cited as a possible prognostic factor,^10,13 with an IQ below 70 associated with poorer outcome, although an IQ above 70 does not necessarily confer a positive outcome. Less-severe impairment in speech at baseline in early childhood also suggests better outcomes in adulthood.¹⁰

As we see more adults with autism, studies that include both children and adults, such as the Longitudinal European Autism Cohort, will be important to characterize the natural history, comorbidities, and genetics of ASD and may help provide more specific predictors of disease course into adulthood.¹⁴

 

 

ACHIEVING A DIAGNOSIS FOR ADULT PATIENTS WITH SUSPECTED AUTISM

While many patients are recognized as having autism in early to mid-childhood, some adults may not receive a formal diagnosis until much later in life. Those with fluent language and normal-range IQ are likely to be overlooked.¹⁵ People with ASD may have had mild symptoms during childhood that did not impair their functioning until demands of daily life exceeded their capacities in adulthood. Alternatively, parents of a child with newly diagnosed ASD may realize that they themselves or another adult family member also show signs of it.

The UK National Institute of Health and Care Excellence suggests that assessment should be considered if the patient meets psychiatric diagnostic criteria and one of the following:

• Difficulty obtaining or sustaining employment or education
• Difficulty initiating or sustaining social relationships
• Past or current contact with mental health or learning disability services
• History of a neurodevelopmental or mental health disorder.^15,16

Currently, diagnosis typically involves a multidisciplinary approach, with psychiatric assessment, neuropsychological testing, and speech and language evaluation.¹⁷ Providers may need to refer patients for these services, sometimes at the patient’s request, if previous mental health misdiagnoses are suspected, if patients report symptoms or impairment consistent with ASD, or if benefits, services, or accommodations, such as a coach in the workplace, are needed.

Diagnosing ASD in adults can be difficult, given that the gold-standard diagnostic tests such as the Autism Diagnostic Observation Schedule-2 (ADOS-2)¹⁸ and the Autism Diagnostic Interview-Revised (ADI-R)¹⁹ are typically used to diagnose autism in children. However, Module 4 in the ADOS-2 was developed for adolescents and older patients with fluent language and has shown at least moderate power to distinguish adults with ASD from those without ASD.^18,20

An initial psychiatric assessment should include a thorough history taken from the patient and, if applicable, the patient’s caregiver, as well as a psychiatric interview of the patient. Neuropsychological testing should include evaluation of cognitive function, social functioning (using the ADOS-2 for adults without intellectual disability, the ADI-R, or both), and adaptive functioning (using the Vineland Adaptive Behavior Scales, second edition²¹).

Evaluation of speech and language is particularly important in patients with limited language ability and should include both expressive and receptive language abilities. Serial testing every few years, as is often recommended in childhood, may help establish the pattern of impairment over time.

Comorbid psychiatric disorders are common

Many people with ASD also have other psychiatric disorders,^17,22 which clinicians should keep in mind when seeing an adult seeking evaluation for ASD.

Attention-deficit/hyperactivity disorder is present at higher rates in patients of average intellectual function with ASD than in the general population.²³

Anxiety disorders, including obsessive-compulsive disorder, were found to often coexist with autism in a sample of adults with autism without intellectual disability,^24,25 and approximately 40% of youths with ASD have at least 1 comorbid anxiety disorder.²⁶

Mood disorders are also prevalent in adults with ASD, with a small study showing that 70% of adults with DSM-IV Asperger syndrome had at least 1 depressive episode in their lifetime.²⁷

BEHAVIORAL AND PHARMACOLOGIC THERAPIES FOR THE ADULT PATIENT

swetlik_adultautismspectrumdisorder_t2.jpg

Treatment of adults with ASD should be individualized based on the challenges they are facing. Many, including those with average or above-average intelligence, struggle with interpersonal relationships, employment, housing, other health conditions, and quality of life.²⁸ Thus, behavioral services and programs should be tailored to help the patient with current challenges (Table 2).

Services and medications for adults with ASD are discussed below. These will vary by individual, and services available may vary by region.

Historically, vocational and social outcomes have been poor for adults with ASD. It is estimated that most larger universities may be home to 100 to 300 students with ASD. To combat isolation, the University of California, Los Angeles, the University of Alabama, and others provide special support services, including group social activities such as board games and individual coaching.⁸ Nevertheless, half of the students with autism who attend institutions of higher learning leave without completing their intended degree.²⁹ Many still struggle to establish meaningful friendships or romantic relationships.²⁹

Planning for a transition of care

Healthcare transition planning is important  but is strikingly underused.³⁰ Individual providers, including adult psychiatrists, vary in their level of training and comfort in diagnosing, treating, and monitoring adults with autism. Youths with ASD are half as likely to receive healthcare transition services as other youths with special healthcare needs.³¹

Pediatric providers, including pediatric psychiatrists, developmental behavioral specialists, and pediatric neurologists, may be best equipped to treat young adult patients or to refer patients to appropriate generalists and specialists comfortable with autism-specific transition of care. The question of eligibility for services is important to patients and families during the transition period, with many parents and professionals unaware of services available to them.³² Receiving adequate transition services is enabled by having a medical home during childhood—that is, a comprehensive, centralized medical record, culturally competent care, interaction with schools, and patient access to clear, unbiased information.³¹

Ideally, in our experience, transitioning should be discussed well before the child ages out of the pediatric provider’s practice. If necessary, healthcare transition services should include 4 components:

• Discussing the switch to a new physician who treats adults
• Discussing changing healthcare needs as an adult
• Planning insurance coverage as an adult
• Encouragement by the physician for the child to take age-appropriate responsibility for his or her healthcare.^31,33

Tools such as the Got Transition checklist from the National Health Care Transition Center can provide support during this process.³⁴

 

Other services

Other services provided as an extension or adjunct to the medical home in early adulthood may include customized vocational or employment training, specialized mentorship or support in a college setting, housing support, and psychological services.³⁵

Community-based programs that emphasize leisure have been shown to improve participants’ independence and quality of life.³⁶ Similarly, participants in programs that emphasized supported employment, with a job coach, on-the-job support, collaboration with the participant’s larger social support network, and selection of tasks to match an individual’s abilities and strengths, demonstrated improved cognitive performance, particularly executive functioning,³⁷ and employment.^38,39 These programs work best for patients who have mild to moderate symptoms.^37,39

Patients with symptoms that are more severe may do better in a residential program. Many of these programs maintain an emphasis on vocational and social skills development. One such long-standing program is Bittersweet Farms, a rural farming community in Ohio for adults with ASD, where individuals with moderate to low function live in a group setting, with emphasis on scheduled, meaningful work including horticulture, animal care, carpentry; and activities of daily living.⁴⁰

Studies of patients across the autism spectrum have generally found better outcomes when vocational support is given, but larger and randomized studies are needed to characterize how to best support these individuals after they leave high school.⁴¹

Psychological services such as applied behavioral therapy, social cognition training, cognitive behavioral therapy, and mindfulness training may be particularly useful in adults.^42–44

Some versions of applied behavioral therapy, such as the Early Start Denver Model,⁴⁵ have been found to be cost-effective and offset some expenses in the care of children with autism, using play-based and relationship-based interventions to promote development across domains while reducing symptoms.

In randomized controlled trials, modified cognitive behavioral therapy⁴³ and mindfulness⁴⁴ were shown to reduce symptoms of anxiety, obsessive-compulsive disorder, and depression.

Dialectical behavior therapy, used to find a balance between accepting oneself and desiring to change, may help in some circumstances to regulate emotions and reduce reactivity and lability, although large randomized clinical trials have not been conducted in the ASD population.⁴⁶

Drug therapy

Medications may be appropriate to manage symptoms or comorbid conditions in adults with ASD. Over 75% adults with ASD have been found to use psychotropic medications.⁴⁷ However, although these drugs have been approved for treating behaviors commonly associated with ASD, none of them provide definitive treatment for this disorder, and they have not been rigorously tested or approved for use in adults with ASD.⁴⁸

Irritability and aggression associated with ASD can be treated with risperidone (approved for children over age 5), aripiprazole (approved for children ages 6–17), clozapine, or haloperidol.⁴⁹

Aberrant social behavior can be treated with risperidone.⁵⁰ Treatments under investigation include oxytocin and secretin.⁴⁹

While no approved drug has been shown to improve social communication,⁵¹ balovaptan, a vasopressin V1a agonist, has shown potential and has been granted breakthrough status by the US Food and Drug Administration for treating challenging behaviors in adults, with additional studies ongoing in children.^52,53

Repetitive behaviors, if the patient finds them impairing, can be managed with selective serotonin reuptake inhibitors.⁴⁹

Much more study of drug therapy in adults with ASD is needed to fully understand the best approaches to psychotropic medication use, including appropriate classes and effective dosage, in this population.

SEX: UNEXPLORED TERRITORY

The reproductive health needs of people with autism remain largely underexplored.⁵⁴ Historically, individuals with ASD were thought to have little interest in sexual activity or parenthood, owing to the nature of the core symptoms of the disorder. This has been shown to be untrue, particularly as studies on this topic began to engage in direct interviews with people with ASD, rather than solely gathering information from caregivers or parents. The findings reinforce the importance of broaching this component of health in this population, for the following reasons:

Adults with ASD are at increased risk of sexual victimization, with nearly 4 out of 5 reporting unwanted sexual advances, coercion, or rape.⁵⁵

They have a smaller pool of knowledge with respect to sexual health. They report⁵⁶ that they learned about sex from television and from “making mistakes.” They use fewer sources. They are less likely to speak to peers and figures of authority to gain knowledge about sexually transmitted infections, sexual behaviors, and contraception. And they are more likely to use forms of nonsocial media, such as television, for information.⁵⁵

They report more concerns about the future with respect to sexual behavior, suggesting the need for targeted sexual education programs.⁵⁶

College-age young adults with ASD who misread communication may be particularly affected by Title IX, which requires schools to promptly investigate reports of sexual harassment and sexual assault, should they struggle to comport themselves appropriately.⁵⁷ Early and frank conversations about issues of consent and appropriate displays of interest and affection may better equip youth to navigate new social scenarios as they plan to leave a supervised home environment for college or the workforce.

Gender identification: Male, female, other

In one study, 77.8% of birth-sex males with ASD said they identified as men, and 67.1% of birth-sex females identified as women,  compared with 93.1% of birth-sex males and 87.3% of birth-sex females without ASD. Many of the remaining individuals with ASD reported a transgender, genderqueer, or other gender identity.⁵⁸ Some studies have found females with ASD report a gay or bisexual orientation more often than males with ASD.^59–61

Adolescents and young adults may be exploring their changing bodies, sexual preferences, and gender roles, and as for all people at this age, these roles emerge against a backdrop of familial and societal expectations that may or may not be concordant with their own projected path regarding sexuality and reproductive health.⁶²

Having the conversation

As with non-ASD patients, a thorough sexual history should be collected via open-ended questions when possible to determine types of sexual activity and partners.

Education of the patient, alongside caregivers and parents, about healthy and safe sexual practices, screening for sexual violence, and hormonal and nonhormonal contraception options are important components of care for this population.

 

CAREGIVER STRESS MAY PERSIST INTO PATIENT’S ADULTHOOD

Caregiver burden is a monumental concern for parents or others who may have lifelong primary responsibility for these neurodiverse adults.⁶³ Family members may feel isolated and may feel they have encountered many barriers to services.⁶⁴ Remaining sensitive, knowledgeable, and inquisitive about the types of support that are needed may help forge a trusting relationship between the provider and the family.

Parents of children with ASD have been reported to experience worse physical and emotional health than parents whose children do not have developmental disabilities.^63,65 These disparities have been found to persist  as their children enter adolescence and young adulthood.^66,67 Parents of children with ASD report more anxiety, depression, and distress compared with parents of children without ASD,⁶³ and parents themselves may be affected by ASD symptoms, which has been linked to increased parenting stress.⁶⁸ Some studies have found blunted cortisol responses,^63,69,70 and some,⁷¹ but not all,⁶³ have found elevated blood pressure in caregivers of children with developmental disabilities. Headache, backache, muscle soreness, and fatigue may also be commonly reported.⁶⁷

In our experience, caregivers are tremendously appreciative when provided connections to adult ASD services and support systems as their child ages. The school system and other formal support systems often assist until the time of transition into adulthood. This transition can be stressful for the adolescent and family alike, and informal support systems such as friends and family may become increasingly crucial, particularly if the adolescent still lives at home.^72,73

The affected young adult’s unmet needs, as perceived by the caregiver, have been found to be significantly associated with caregiver burden, whereas the severity of the adult patient’s ASD symptoms has not.⁶⁶ Therefore, it may be helpful to ask caregivers whether they perceive any unmet needs, regardless of the clinician’s perception of the severity of the patient’s ASD symptoms. Providing support to address these needs, particularly those relating to the child’s mood disorders, communication, social needs, safety, and daytime activities, may be the domains of support that most effectively reduce the caregiver burden in this population.⁶⁶

Caregiver positivity, lower stress levels, and increased social support, particularly in the form of friends and family members providing no-cost assistance to caregivers whose children do not live independently,⁷⁴ have been linked to better outcomes for caregivers.^70,74,75 Rigorous studies that examine caregiver burden as individuals with ASD enter mid- and late-adulthood are limited.

THE ROLE OF THE INTERNIST IN CARING FOR ADULTS WITH AUTISM

A major challenge for many adults with ASD is the transition from services provided during childhood to those provided in adulthood. While children with autism have subspecialty providers who diagnose and manage their condition, including developmental-behavioral pediatricians, pediatric neurologists, and child psychiatrists, adults with autism may have fewer options.

Autism centers are becoming more available across the nation, and many provide care across the life span. However, depending on a patient’s needs, the primary care provider may need to manage residual symptoms as the patient transitions from pediatric to adult care, ultimately deciding when and where to refer the patient.

The patient’s family should pay close attention to function and mood around the time the patient leaves the structure of high school, and they should build rapport with a primary care provider they can turn to if problems persist or arise. Referrals for behavioral therapy and for social work, job training, and vocational support can greatly benefit patients as they transition to young adulthood. Referrals and suggestions for social support can also help caregivers.

Medical care

Deciding when and how to medicate the patient for symptoms of autism and related behaviors necessitates consideration of the patient’s impairment, side effects of the medication, and the impact medications may have on the patient’s other conditions. Disordered eating, mood problems, anxiety, and attention-deficit/hyperactivity disorder should be considered, and, as in all patients, regular screenings of mental health status should be conducted.^76,77

Comorbid medical conditions may cause worsening of a patient’s known behavioral symptoms or may precipitate new behaviors or aggression as a result of pain or discomfort, particularly in patients with limited speech. A change in stereotypes or increased irritability warrants a thoughtful investigation for a cause other than ASD before adding or increasing behavioral medications. Common comorbid conditions include gastrointestinal distress, most commonly constipation and diarrhea in an idiopathic ASD population, with increasing ASD symptom severity correlating with increased odds of a gastrointestinal problem.⁷⁸ Allergies, sleep disorders, seizures, and other psychiatric conditions are also frequent.⁷⁹

Preventive care, including vaccinations, should be given as scheduled. Caregivers and patients can be reminded if needed that vaccines do not cause or worsen autism, and vaccination is intended to improve the safety of the patient and those around them, protecting against potentially life-threatening disease. Regular dental care visits, particularly for patients who are using medications that may affect tooth or gingival health,⁸⁰ and regular visits to an optometrist or ophthalmologist for screening of vision are also advised.

Adverse effects. Weight gain and metabolic syndrome are common adverse effects of medications used for behavioral management, and the primary care physician may uncover diabetes, cardiac disorders, and hyperlipidemia. Patients with ASD may be particularly sensitive to the effects of medications and therefore may require a lower dose or a slower titration than other patients. Working with a behavioral team, careful weaning of psychiatric medications to the minimum needed is strongly recommended whenever possible.⁸¹

TAKE-HOME POINTS

As more adults with autism enter society, they may require varying levels of support from the healthcare community to ensure that therapeutic gains from childhood persist, allowing them to achieve maximal functional potential.

Adults with ASD may have a high, normal, or low IQ and intellectual capability. Knowledge of this and of the patient’s symptom severity and presence of comorbid psychiatric and other health conditions can help the clinician guide the patient to appropriate social services and pharmacologic treatments.

Individualized support in the workplace, as well as education regarding sexual health, can help improve outcomes for affected individuals.

Caregiver burden for individuals with autism can be high, but it can be mitigated by social support.

Further research regarding appropriate diagnostic instruments in adulthood and appropriate treatments for impairing autism-related symptoms across the life span may be particularly helpful in supporting this patient population.

A 61-year-old woman presented to our hematology clinic for evaluation of multiple episodes of bruising. The first episode occurred 8 months earlier, when she developed a large bruise after water skiing. Two months before coming to us, she went to her local emergency room because of new bruising and was found to have a prolonged activated partial thromboplastin time (aPTT) of 60 seconds (reference range 23.3–34.9), but she underwent no further testing at that time.

At presentation to our clinic, she reported having no fevers, night sweats, unintentional weight loss, swollen lymph nodes, joint pain, rashes, mouth sores, nosebleeds, or blood in the urine or stool. Her history was notable only for hypothyroidism, which was diagnosed in the previous year. Her medications included levothyroxine, vitamin D₃, and vitamin C. She had been taking a baby aspirin daily for the past 10 years but had stopped 1 month earlier because of the bruising.

atieh_bruising_t1.jpg

On examination, she had a single small hematoma on her right thigh. She had no ecchymoses, petechiae, or adenopathy, and her spleen was nonpalpable.

Ten years earlier she had been evaluated for a possible transient ischemic attack; laboratory results at that time included a normal aPTT of 25.1 seconds and a normal factor VIII level of 153% (reference range 50%–173%).

atieh_bruising_t2.jpg

Laboratory testing at our clinic showed a normal complete blood cell count (Table 1); the coagulation factor assay confirmed that her aPTT was elevated (prolonged), but other values were normal (Table 2).

atieh_bruising_t3.jpg

Causes of an isolated prolonged aPTT include medications (eg, heparin), inherited factor deficiencies, acquired inhibitors of coagulation factors, and inherited or acquired von Willebrand disease. Lupus anticoagulant can prolong the aPTT but is usually associated with thrombosis rather than bleeding.¹ The differential diagnoses for this and other patterns of coagulation assay abnormalities are listed in Table 3.

EVALUATION FOR AN ISOLATED PROLONGED aPTT

1. What is the appropriate next test to evaluate this patient’s prolonged aPTT?

• Lupus anticoagulant panel
• Coagulation factor levels
• Mixing studies
• Bethesda assay

Mixing studies

Once a prolonged aPTT is confirmed, the appropriate next step is a mixing study. This involves mixing the patient’s plasma with pooled normal plasma in a 1-to-1 ratio, then repeating the aPTT test immediately, and again after 1 hour of incubation at 37°C. If the patient does not have enough of one of the coagulation factors, the aPTT immediately returns to the normal range when plasma is mixed with the pooled plasma because the pooled plasma contains the factor that is lacking. If this happens, then factor assays should be performed to identify the deficient factor.¹

Various antibodies that inhibit coagulation factors can also affect the aPTT. There are 2 general types: immediate-acting and delayed.

With an immediate-acting inhibitor, the aPTT does not correct into the normal range with initial mixing. Immediate-acting inhibitors are often seen together with lupus anticoagulants, which are nonspecific phospholipid antibodies. If an immediate-acting inhibitor is detected, further testing should focus on evaluation for lupus anticoagulant, including phospholipid-dependency studies.

With a delayed inhibitor, the aPTT initially comes down, but subsequently goes back up after incubation. Acquired factor VIII inhibitor is a classic delayed-type inhibitor and is also the most common factor inhibitor.¹ If a delayed-acting inhibitor is found, specific intrinsic factor levels should be measured (factors VIII, IX, XI, and XII),² and testing should also be done for lupus anticoagulant, as these inhibitors may occur together.

Bethesda assay

atieh_bruising_t4.jpg

If factor levels are decreased, a Bethesda assay should be performed to differentiate a specific factor inhibitor from a lupus anticoagulant. In the case of a factor VIII inhibitor, serial dilutions of patient plasma are incubated at 37°C with pooled normal plasma for 2 hours, then factor VIII activity is measured. The reciprocal dilution of patient plasma that results in 50% of factor VIII activity in the control plasma is 1 Bethesda unit (BU). The stronger the inhibitor in the patient’s sample, the more dilutions are required to measure factor VIII activity, and thus the higher the Bethesda titer.³

Case continued: Results of mixing and Bethesda studies

atieh_bruising_t5.jpg

Results of the mixing study (Table 4) showed an initial correction of the aPTT in a 1-to-1 mix, but after 1 hour of incubation, the aPTT was again prolonged at 42 seconds (reference range < 37.3). Further testing revealed very low levels of factor VIII (< 1%), and the presence of a factor VIII inhibitor, quantified at 5.8 BU (reference range < 0.5) (Table 5). Additional coagulation tests, including von Willebrand factor testing and a lupus anticoagulation panel, were negative.

 

 

FACTOR VIII INHIBITOR EVALUATION

2. What is the most likely underlying condition associated with this patient’s factor VIII inhibitor?

• Autoimmune disease
• Malignancy
• A medication
• Unknown (idiopathic)

Acquired hemophilia A (AHA) is a rare disorder caused by autoantibodies against factor VIII. Its estimated incidence is about 1 person per million per year.⁴ It usually presents as unexplained bruising or bleeding and is only rarely diagnosed by an incidentally noted prolonged aPTT. The severity of bleeding is variable and can include subcutaneous, soft-tissue, retroperitoneal, gastrointestinal, and intracranial hemorrhage.⁵

AHA is considered idiopathic in more than half of cases. A study based on a European registry⁵ of 501 patients with AHA and a UK study⁶ of 172 patients found no underlying disease in 52% and 65% of patients, respectively. For patients with an identified cause, the most common causes were malignancy (12%⁵ and 15%⁶) and autoimmune disease (12%⁵ and 17%⁶).

Drugs have rarely been associated with factor VIII inhibitors. Such occurrences have been reported with interferon, blood thinners, antibiotics, and psychiatric medications, but no study yet has indicated causation. However, patients with congenital hemophilia A treated with factor VIII preparations have about a 15% chance of developing factor VIII inhibitors. In this setting, inhibitors develop in response to recombinant factor VIII exposure, unlike the autoimmune phenomena seen in AHA.

TREATMENT OF ACQUIRED HEMOPHILIA A

3. What is the most appropriate treatment for AHA?

• Desmopressin and prednisone
• Recombinant porcine factor VIII and prednisone plus cyclophosphamide
• Recombinant factor VIIa and rituximab
• Any of the above

Any of the above regimens can be used. In general, treatment of AHA has two purposes: to stop acute hemorrhage, and to reduce the level of factor VIII inhibitor. No standard treatment guidelines are available; evidence of the effectiveness of different drugs is based largely on data on congenital hemophilia A.³

Acute treatment to stop bleeding

Initial treatment of AHA often focuses on stopping an acute hemorrhage by either raising circulating levels of factor VIII or bypassing it in the coagulation cascade.

Desmopressin can temporarily raise factor VIII levels, but it is often ineffective in AHA unless the patient has very low inhibitor titers.³

Factor VIII concentrate (human or recombinant porcine factor VIII) may be effective in patients with low inhibitor titers (< 5 BU). Higher doses are often required than those used in congenital hemophilia A. Factor VIII concentrate is usually combined with immunosuppressive treatment to lower the factor VIII inhibitor level (described below).³

If these methods are ineffective or the patient has high inhibitor titers (> 5 BU), activated prothrombin complex concentrates, known as FEIBA (factor eight inhibitor bypassing activity), or recombinant factor VIIa is available. These agents bypass factor VIII in the clotting cascade.

Immunosuppression to reduce factor VIII inhibitor

Immunosuppressive agents are the mainstay of AHA treatment to lower the inhibitor level.

Regimens vary. A 2003 meta-analysis⁴ including 249 patients found that prednisone alone resulted in complete response in about 30% of patients, and the addition of cyclophosphamide increased the response rate to 60% to 100%. High-dose intravenous immunoglobulin led to conflicting results. Conclusions were limited by the variability of dosing and duration in treatment regimens among the 20 different studies included.

An analysis of 331 patients in the European Acquired Hemophilia Registry (EACH2)⁷ found that steroids alone produced remission in 48% of patients, while steroids combined with cyclophosphamide raised the rate to 70%. Rituximab-based regimens were successful in 59% but required twice as long to achieve remission as steroid or cyclophosphamide-based regimens. No benefit was noted from intravenous immunoglobulin.

Risks of disease and treatment

AHA is associated with significant risk of morbidity and death related to bleeding, complications of treatment, and underlying disease.

In EACH2, 16 of the 331 patients died of bleeding, 16 died of causes related to immunosuppression, and 45 died of causes related to the underlying condition.⁵ In the UK registry of 172 patients, 13 patients died of bleeding, and 12 died of sepsis related to immunosuppression.⁶

The factor VIII level and inhibitor titer are not necessarily useful in stratifying bleeding risk, as severe and fatal bleeding can occur at variable levels and patients remain at risk of bleeding as long as the inhibitor persists.^6,7

 

 

CASE CONTINUED: TREATMENT, LYMPHOCYTOSIS

The patient was started on 60 mg daily of prednisone, resulting in a decrease in her aPTT, increase in factor VIII level, and lower Bethesda titer. On a return visit, her absolute lymphocyte count was 7.04 × 10⁹/L (reference range 1.0–4.0). She reported no fevers, chills, or recent infections.  

EVALUATING LYMPHOCYTOSIS

Lymphocytosis is defined in most laboratories as an absolute lymphocyte count greater than 4.0 × 10⁹/L for adults. Normally, T cells (CD3+) make up 60% to 80% of lymphocytes, B cells (CD20+) 10% to 20%, and natural killer (NK) cells (CD3–, CD56+) 5% to 10%. Lymphocytosis is usually caused by infection, but it can have other causes, including malignancy.

Peripheral blood smear. If there is no clear cause of lymphocytosis, a peripheral blood smear can be used to assess lymphocyte morphology, providing clues to the underlying etiology. For example, atypical lymphocytes are often seen in infectious mononucleosis, while “smudge” lymphocytes are characteristic of chronic lymphocytic leukemia. If a peripheral smear shows abnormal morphology, further workup should include establishing whether the lymphocytes are polyclonal or clonal.⁸

CASE CONTINUED: LARGE GRANULAR LYMPHOCYTES

atieh_bruising_f1.jpg

On the patient’s peripheral smear, 54% of lymphocytes were large lymphocytes with moderate amounts of pale cytoplasm filled with azurophilic granules, consistent with large granular lymphocytes (LGLs) (Figure 1).

4. What is the next step to evaluate the patient’s lymphocytosis?

• Bone marrow biopsy
• Karyotype analysis
• Flow cytometry
• Fluorescence in situ hybridization

Flow cytometry with V-beta analysis is the best first test to determine the cause of lymphocytosis after review of the peripheral smear. For persistent lymphocytosis, flow cytometry should be done even if a peripheral smear shows normal lymphocyte morphology.

Most T cells possess receptors composed of alpha and beta chains, each encoded by variable (V), diversity (D), joining (J), and constant (C) gene segments. The V, D, and J segments undergo rearrangement during T-cell development in the thymus based on antigen exposure, producing a diverse T-cell receptor population.

In a polyclonal population of lymphocytes, the T-cell receptors have a variety of gene segment arrangements, indicating normal T-cell development. But in a clonal population of lymphocytes, the T-cell receptors have a single identical gene segment arrangement, indicating they all originated from a single clone.⁹ Lymphocytosis in response to an infection is typically polyclonal, while malignant lymphocytosis is clonal. 

Monoclonal antibodies against many of the variable regions of the beta chain (V-beta) of T-cell receptors have been developed, enabling flow cytometry to establish clonality.

T-cell receptor gene rearrangement studies can also be performed using polymerase chain reaction and Southern blot techniques.⁹

Karyotype analysis is usually not performed for the finding of LGLs, because most leukemias (eg, T-cell and NK-cell leukemias) have cells with a normal karyotype. 

Bone marrow biopsy is invasive and usually not required to evaluate LGLs. It can be especially risky for a patient with a bleeding disorder such as a factor VIII inhibitor.¹⁰

Case continued: Flow cytometry confirms clonality

Subsequent flow cytometry found that more than 50% of the patient’s lymphocytes were LGLs that co-expressed CD3+, CD8+, CD56+, and CD57+, with aberrantly decreased CD7 expression. T-cell V-beta analysis demonstrated an expansion of the V-beta 17 family, and T-cell receptor gene analysis with polymerase chain reaction confirmed the presence of a clonal rearrangement.

LGL LEUKEMIA: CLASSIFICATION AND MANAGEMENT

LGLs normally account for 10% to 15% of peripheral mononuclear cells.¹¹ LGL leukemia is caused by a clonal population of cytotoxic T cells or NK cells and involves an increased number of LGLs (usually > 2 × 10⁹/L).¹⁰

LGL leukemia is divided into 3 categories according to the most recent World Health Organization classification^10,12:

T-cell LGL leukemia (about 85% of cases) is considered indolent but can cause significant cytopenias and is often associated with autoimmune disease.¹³ Cells usually express a CD3+, CD8+, CD16+, and CD57+ phenotype. Survival is about 70% at 10 years.

Chronic NK-cell lymphocytosis (about 10%) also tends to have an indolent course with cytopenia and an autoimmune association, and with a similar prognosis to T-cell LGL leukemia. Cells express a CD3–, CD16+, and CD56+ phenotype.

Aggressive NK-cell LGL leukemia (about 5%) is associated with Epstein-Barr virus infection and occurs in younger patients. It is characterized by severe cytopenias, “B symptoms” (ie, fever, night sweats, weight loss), and has a very poor prognosis. Like chronic NK-cell lymphocytosis, cells express a CD3–, CD16+, and CD56+ phenotype. Fas (CD95) and Fas-ligand (CD178) are strongly expressed.^10,13

Most cases of LGL leukemia can be diagnosed on the basis of classic morphology on peripheral blood smear and evidence of clonality on flow cytometry or gene rearrangement studies. T-cell receptor gene studies cannot be used to establish clonality in the NK subtypes, as NK cells do not express T-cell receptors.¹¹

Case continued: Diagnosis, continued course 

In our patient, T-cell LGL leukemia was diagnosed on the basis of the peripheral smear, flow cytometry results, and positive T-cell receptor gene studies for clonal rearrangement in the T-cell receptor beta region.

While her corticosteroid therapy was being tapered, her factor III inhibitor level increased, and she had a small episode of bleeding, prompting the start of cyclophosphamide 50 mg daily with lower doses of prednisone.

atieh_bruising_f2.jpg

She then developed elevated liver enzymes, prompting discontinuation of cyclophosphamide. Rituximab was started and continued for 4 weekly doses, resulting in normalization of aPTT and factor VIII level with undetectable Bethesda titers (Figure 2).

 

 

LGL LEUKEMIA AND AUTOIMMUNE DISEASE

Patients with LGL leukemia commonly have or develop autoimmune conditions. Immune-mediated cytopenias including pure red cell aplasia, aplastic anemia, and autoimmune hemolytic anemias can occur. Neutropenia, the most common cytopenia in LGL leukemia, is thought to be at least partly autoimmune, as the degree of neutropenia is often worse than would be expected solely from bone-marrow infiltration of LGL cells.^10,14,15

Rheumatoid arthritis is the most common autoimmune condition associated with LGL leukemia, with a reported incidence between 11% and 36%.^13–15

Felty syndrome (rheumatoid arthritis, splenomegaly, and neutropenia) is often associated with LGL leukemia and is thought by some to be part of the same disease process.¹⁵

Treat with immunosuppressives if needed

Indications for treating LGL leukemia include the development of cytopenias and associated autoimmune diseases. Immunosuppressive agents, such as methotrexate, cyclophosphamide, and cyclosporine, are commonly used.^10,11,14 Most evidence of treatment efficacy is from retrospective studies and case reports, with widely variable response rates that overall are around 50%.¹⁰

ACQUIRED HEMOPHILIA A AND HEMATOLOGIC MALIGNANCY

A systematic review found 30 cases of AHA associated with hematologic malignancies.¹⁶ The largest case series¹⁷ in this analysis had 8 patients, and included diagnoses of chronic lymphocytic leukemia, erythroleukemia, myelofibrosis, multiple myeloma, and myelodysplastic syndrome. In 3 of these patients, the appearance of the inhibitor preceded the diagnosis of the underlying malignancy by an average of 3.5 months. In 1 patient with erythroleukemia and another with multiple myeloma, the activity of the inhibitor could be clearly correlated with the underlying malignancy. In the other 6 patients, no association between the two could be made.

In the same series, complete resolution of the inhibitor was related only to the level of Bethesda titer present at diagnosis, with those who achieved resolution having lower mean Bethesda titers.¹⁷ Similarly, in EACH2, lower inhibitor Bethesda titers and higher factor VIII levels at presentation were associated with faster inhibitor eradication and normalization of factor VIII levels.⁷

Murphy et al¹⁸ described a 62-year-old woman with Felty syndrome who developed a factor VIII inhibitor and was subsequently given a diagnosis of LGL leukemia. Treatment with immunosuppressive agents, including cyclophosphamide, azathioprine, and rituximab, successfully eradicated her factor VIII inhibitor, although the LGL leukemia persisted.

Case conclusion: Eradication of factor VIII inhibitor

Our patient, similar to the patient described by Murphy et al¹⁸ above, had eradication of the factor VIII inhibitor despite persistence of LGL leukemia. Between the time of diagnosis at our clinic, when she had 54% LGLs, and eradication of the inhibitor 3 months later, the LGL percentage ranged from 45% to 89%. No clear direct correlation between LGL and factor VIII inhibitor levels could be detected.

Given the strong association of LGL leukemia with autoimmune disease, it is tempting to believe that her factor VIII inhibitor was somehow related to her malignancy, although the exact mechanism remained unclear. The average age at diagnosis is 60 for LGL leukemia¹¹ and over 70 for AHA,^5,6 so advanced age may be the common denominator. Whether or not our patient will have recurrence of her factor VIII inhibitor or the development of other autoimmune diseases with the persistence of her LGL leukemia remains to be seen.

At last follow-up, our patient was off all therapy and continued to have normal aPTT and factor VIII levels. Repeat flow cytometry after treatment of her factor VIII inhibitor showed persistence of a clonal T-cell population, although reduced from 72% to 60%. It may be that the 2 entities were unrelated, and the clonal T-cell population was simply fluctuating over time. This can be determined only with further observation. As the patient had no symptoms from her LGL leukemia, she continued to be observed without treatment.

TAKE-HOME POINTS

• The coagulation assay is key to initially assessing a bleeding abnormality; whether the prothrombin time and aPTT are normal or prolonged narrows the differential diagnosis and determines next steps in evaluation.
• Mixing studies can help pinpoint the responsible deficient factor.
• Acquired factor VIII deficiency, also known as AHA, may be caused by autoimmune disease, malignancy, or medications, but it is usually idiopathic.
• AHA treatment is focused on achieving hemostasis and reducing factor VIII inhibitor.
• Lymphocytosis should be evaluated with a peripheral blood smear and flow cytometry to determine if the population is polyclonal (associated with infection) or clonal (associated with malignancy).
• LGL leukemia is usually a chronic, indolent disease, although an uncommon subtype has an aggressive course.
• The association between AHA and LGL leukemia is unclear, and both conditions must be monitored and managed.

A 61-year-old woman presented to our hematology clinic for evaluation of multiple episodes of bruising. The first episode occurred 8 months earlier, when she developed a large bruise after water skiing. Two months before coming to us, she went to her local emergency room because of new bruising and was found to have a prolonged activated partial thromboplastin time (aPTT) of 60 seconds (reference range 23.3–34.9), but she underwent no further testing at that time.

At presentation to our clinic, she reported having no fevers, night sweats, unintentional weight loss, swollen lymph nodes, joint pain, rashes, mouth sores, nosebleeds, or blood in the urine or stool. Her history was notable only for hypothyroidism, which was diagnosed in the previous year. Her medications included levothyroxine, vitamin D₃, and vitamin C. She had been taking a baby aspirin daily for the past 10 years but had stopped 1 month earlier because of the bruising.

atieh_bruising_t1.jpg

On examination, she had a single small hematoma on her right thigh. She had no ecchymoses, petechiae, or adenopathy, and her spleen was nonpalpable.

Ten years earlier she had been evaluated for a possible transient ischemic attack; laboratory results at that time included a normal aPTT of 25.1 seconds and a normal factor VIII level of 153% (reference range 50%–173%).

atieh_bruising_t2.jpg

Laboratory testing at our clinic showed a normal complete blood cell count (Table 1); the coagulation factor assay confirmed that her aPTT was elevated (prolonged), but other values were normal (Table 2).

atieh_bruising_t3.jpg

Causes of an isolated prolonged aPTT include medications (eg, heparin), inherited factor deficiencies, acquired inhibitors of coagulation factors, and inherited or acquired von Willebrand disease. Lupus anticoagulant can prolong the aPTT but is usually associated with thrombosis rather than bleeding.¹ The differential diagnoses for this and other patterns of coagulation assay abnormalities are listed in Table 3.

EVALUATION FOR AN ISOLATED PROLONGED aPTT

1. What is the appropriate next test to evaluate this patient’s prolonged aPTT?

• Lupus anticoagulant panel
• Coagulation factor levels
• Mixing studies
• Bethesda assay

Mixing studies

Once a prolonged aPTT is confirmed, the appropriate next step is a mixing study. This involves mixing the patient’s plasma with pooled normal plasma in a 1-to-1 ratio, then repeating the aPTT test immediately, and again after 1 hour of incubation at 37°C. If the patient does not have enough of one of the coagulation factors, the aPTT immediately returns to the normal range when plasma is mixed with the pooled plasma because the pooled plasma contains the factor that is lacking. If this happens, then factor assays should be performed to identify the deficient factor.¹

Various antibodies that inhibit coagulation factors can also affect the aPTT. There are 2 general types: immediate-acting and delayed.

With an immediate-acting inhibitor, the aPTT does not correct into the normal range with initial mixing. Immediate-acting inhibitors are often seen together with lupus anticoagulants, which are nonspecific phospholipid antibodies. If an immediate-acting inhibitor is detected, further testing should focus on evaluation for lupus anticoagulant, including phospholipid-dependency studies.

With a delayed inhibitor, the aPTT initially comes down, but subsequently goes back up after incubation. Acquired factor VIII inhibitor is a classic delayed-type inhibitor and is also the most common factor inhibitor.¹ If a delayed-acting inhibitor is found, specific intrinsic factor levels should be measured (factors VIII, IX, XI, and XII),² and testing should also be done for lupus anticoagulant, as these inhibitors may occur together.

Bethesda assay

atieh_bruising_t4.jpg

If factor levels are decreased, a Bethesda assay should be performed to differentiate a specific factor inhibitor from a lupus anticoagulant. In the case of a factor VIII inhibitor, serial dilutions of patient plasma are incubated at 37°C with pooled normal plasma for 2 hours, then factor VIII activity is measured. The reciprocal dilution of patient plasma that results in 50% of factor VIII activity in the control plasma is 1 Bethesda unit (BU). The stronger the inhibitor in the patient’s sample, the more dilutions are required to measure factor VIII activity, and thus the higher the Bethesda titer.³

Case continued: Results of mixing and Bethesda studies

atieh_bruising_t5.jpg

Results of the mixing study (Table 4) showed an initial correction of the aPTT in a 1-to-1 mix, but after 1 hour of incubation, the aPTT was again prolonged at 42 seconds (reference range < 37.3). Further testing revealed very low levels of factor VIII (< 1%), and the presence of a factor VIII inhibitor, quantified at 5.8 BU (reference range < 0.5) (Table 5). Additional coagulation tests, including von Willebrand factor testing and a lupus anticoagulation panel, were negative.

 

 

FACTOR VIII INHIBITOR EVALUATION

2. What is the most likely underlying condition associated with this patient’s factor VIII inhibitor?

• Autoimmune disease
• Malignancy
• A medication
• Unknown (idiopathic)

Acquired hemophilia A (AHA) is a rare disorder caused by autoantibodies against factor VIII. Its estimated incidence is about 1 person per million per year.⁴ It usually presents as unexplained bruising or bleeding and is only rarely diagnosed by an incidentally noted prolonged aPTT. The severity of bleeding is variable and can include subcutaneous, soft-tissue, retroperitoneal, gastrointestinal, and intracranial hemorrhage.⁵

AHA is considered idiopathic in more than half of cases. A study based on a European registry⁵ of 501 patients with AHA and a UK study⁶ of 172 patients found no underlying disease in 52% and 65% of patients, respectively. For patients with an identified cause, the most common causes were malignancy (12%⁵ and 15%⁶) and autoimmune disease (12%⁵ and 17%⁶).

Drugs have rarely been associated with factor VIII inhibitors. Such occurrences have been reported with interferon, blood thinners, antibiotics, and psychiatric medications, but no study yet has indicated causation. However, patients with congenital hemophilia A treated with factor VIII preparations have about a 15% chance of developing factor VIII inhibitors. In this setting, inhibitors develop in response to recombinant factor VIII exposure, unlike the autoimmune phenomena seen in AHA.

TREATMENT OF ACQUIRED HEMOPHILIA A

3. What is the most appropriate treatment for AHA?

• Desmopressin and prednisone
• Recombinant porcine factor VIII and prednisone plus cyclophosphamide
• Recombinant factor VIIa and rituximab
• Any of the above

Any of the above regimens can be used. In general, treatment of AHA has two purposes: to stop acute hemorrhage, and to reduce the level of factor VIII inhibitor. No standard treatment guidelines are available; evidence of the effectiveness of different drugs is based largely on data on congenital hemophilia A.³

Acute treatment to stop bleeding

Initial treatment of AHA often focuses on stopping an acute hemorrhage by either raising circulating levels of factor VIII or bypassing it in the coagulation cascade.

Desmopressin can temporarily raise factor VIII levels, but it is often ineffective in AHA unless the patient has very low inhibitor titers.³

Factor VIII concentrate (human or recombinant porcine factor VIII) may be effective in patients with low inhibitor titers (< 5 BU). Higher doses are often required than those used in congenital hemophilia A. Factor VIII concentrate is usually combined with immunosuppressive treatment to lower the factor VIII inhibitor level (described below).³

If these methods are ineffective or the patient has high inhibitor titers (> 5 BU), activated prothrombin complex concentrates, known as FEIBA (factor eight inhibitor bypassing activity), or recombinant factor VIIa is available. These agents bypass factor VIII in the clotting cascade.

Immunosuppression to reduce factor VIII inhibitor

Immunosuppressive agents are the mainstay of AHA treatment to lower the inhibitor level.

Regimens vary. A 2003 meta-analysis⁴ including 249 patients found that prednisone alone resulted in complete response in about 30% of patients, and the addition of cyclophosphamide increased the response rate to 60% to 100%. High-dose intravenous immunoglobulin led to conflicting results. Conclusions were limited by the variability of dosing and duration in treatment regimens among the 20 different studies included.

An analysis of 331 patients in the European Acquired Hemophilia Registry (EACH2)⁷ found that steroids alone produced remission in 48% of patients, while steroids combined with cyclophosphamide raised the rate to 70%. Rituximab-based regimens were successful in 59% but required twice as long to achieve remission as steroid or cyclophosphamide-based regimens. No benefit was noted from intravenous immunoglobulin.

Risks of disease and treatment

AHA is associated with significant risk of morbidity and death related to bleeding, complications of treatment, and underlying disease.

In EACH2, 16 of the 331 patients died of bleeding, 16 died of causes related to immunosuppression, and 45 died of causes related to the underlying condition.⁵ In the UK registry of 172 patients, 13 patients died of bleeding, and 12 died of sepsis related to immunosuppression.⁶

The factor VIII level and inhibitor titer are not necessarily useful in stratifying bleeding risk, as severe and fatal bleeding can occur at variable levels and patients remain at risk of bleeding as long as the inhibitor persists.^6,7

 

 

CASE CONTINUED: TREATMENT, LYMPHOCYTOSIS

The patient was started on 60 mg daily of prednisone, resulting in a decrease in her aPTT, increase in factor VIII level, and lower Bethesda titer. On a return visit, her absolute lymphocyte count was 7.04 × 10⁹/L (reference range 1.0–4.0). She reported no fevers, chills, or recent infections.  

EVALUATING LYMPHOCYTOSIS

Lymphocytosis is defined in most laboratories as an absolute lymphocyte count greater than 4.0 × 10⁹/L for adults. Normally, T cells (CD3+) make up 60% to 80% of lymphocytes, B cells (CD20+) 10% to 20%, and natural killer (NK) cells (CD3–, CD56+) 5% to 10%. Lymphocytosis is usually caused by infection, but it can have other causes, including malignancy.

Peripheral blood smear. If there is no clear cause of lymphocytosis, a peripheral blood smear can be used to assess lymphocyte morphology, providing clues to the underlying etiology. For example, atypical lymphocytes are often seen in infectious mononucleosis, while “smudge” lymphocytes are characteristic of chronic lymphocytic leukemia. If a peripheral smear shows abnormal morphology, further workup should include establishing whether the lymphocytes are polyclonal or clonal.⁸

CASE CONTINUED: LARGE GRANULAR LYMPHOCYTES

atieh_bruising_f1.jpg

On the patient’s peripheral smear, 54% of lymphocytes were large lymphocytes with moderate amounts of pale cytoplasm filled with azurophilic granules, consistent with large granular lymphocytes (LGLs) (Figure 1).

4. What is the next step to evaluate the patient’s lymphocytosis?

• Bone marrow biopsy
• Karyotype analysis
• Flow cytometry
• Fluorescence in situ hybridization

Flow cytometry with V-beta analysis is the best first test to determine the cause of lymphocytosis after review of the peripheral smear. For persistent lymphocytosis, flow cytometry should be done even if a peripheral smear shows normal lymphocyte morphology.

Most T cells possess receptors composed of alpha and beta chains, each encoded by variable (V), diversity (D), joining (J), and constant (C) gene segments. The V, D, and J segments undergo rearrangement during T-cell development in the thymus based on antigen exposure, producing a diverse T-cell receptor population.

In a polyclonal population of lymphocytes, the T-cell receptors have a variety of gene segment arrangements, indicating normal T-cell development. But in a clonal population of lymphocytes, the T-cell receptors have a single identical gene segment arrangement, indicating they all originated from a single clone.⁹ Lymphocytosis in response to an infection is typically polyclonal, while malignant lymphocytosis is clonal. 

Monoclonal antibodies against many of the variable regions of the beta chain (V-beta) of T-cell receptors have been developed, enabling flow cytometry to establish clonality.

T-cell receptor gene rearrangement studies can also be performed using polymerase chain reaction and Southern blot techniques.⁹

Karyotype analysis is usually not performed for the finding of LGLs, because most leukemias (eg, T-cell and NK-cell leukemias) have cells with a normal karyotype. 

Bone marrow biopsy is invasive and usually not required to evaluate LGLs. It can be especially risky for a patient with a bleeding disorder such as a factor VIII inhibitor.¹⁰

Case continued: Flow cytometry confirms clonality

Subsequent flow cytometry found that more than 50% of the patient’s lymphocytes were LGLs that co-expressed CD3+, CD8+, CD56+, and CD57+, with aberrantly decreased CD7 expression. T-cell V-beta analysis demonstrated an expansion of the V-beta 17 family, and T-cell receptor gene analysis with polymerase chain reaction confirmed the presence of a clonal rearrangement.

LGL LEUKEMIA: CLASSIFICATION AND MANAGEMENT

LGLs normally account for 10% to 15% of peripheral mononuclear cells.¹¹ LGL leukemia is caused by a clonal population of cytotoxic T cells or NK cells and involves an increased number of LGLs (usually > 2 × 10⁹/L).¹⁰

LGL leukemia is divided into 3 categories according to the most recent World Health Organization classification^10,12:

T-cell LGL leukemia (about 85% of cases) is considered indolent but can cause significant cytopenias and is often associated with autoimmune disease.¹³ Cells usually express a CD3+, CD8+, CD16+, and CD57+ phenotype. Survival is about 70% at 10 years.

Chronic NK-cell lymphocytosis (about 10%) also tends to have an indolent course with cytopenia and an autoimmune association, and with a similar prognosis to T-cell LGL leukemia. Cells express a CD3–, CD16+, and CD56+ phenotype.

Aggressive NK-cell LGL leukemia (about 5%) is associated with Epstein-Barr virus infection and occurs in younger patients. It is characterized by severe cytopenias, “B symptoms” (ie, fever, night sweats, weight loss), and has a very poor prognosis. Like chronic NK-cell lymphocytosis, cells express a CD3–, CD16+, and CD56+ phenotype. Fas (CD95) and Fas-ligand (CD178) are strongly expressed.^10,13

Most cases of LGL leukemia can be diagnosed on the basis of classic morphology on peripheral blood smear and evidence of clonality on flow cytometry or gene rearrangement studies. T-cell receptor gene studies cannot be used to establish clonality in the NK subtypes, as NK cells do not express T-cell receptors.¹¹

Case continued: Diagnosis, continued course 

In our patient, T-cell LGL leukemia was diagnosed on the basis of the peripheral smear, flow cytometry results, and positive T-cell receptor gene studies for clonal rearrangement in the T-cell receptor beta region.

While her corticosteroid therapy was being tapered, her factor III inhibitor level increased, and she had a small episode of bleeding, prompting the start of cyclophosphamide 50 mg daily with lower doses of prednisone.

atieh_bruising_f2.jpg

She then developed elevated liver enzymes, prompting discontinuation of cyclophosphamide. Rituximab was started and continued for 4 weekly doses, resulting in normalization of aPTT and factor VIII level with undetectable Bethesda titers (Figure 2).

 

 

LGL LEUKEMIA AND AUTOIMMUNE DISEASE

Patients with LGL leukemia commonly have or develop autoimmune conditions. Immune-mediated cytopenias including pure red cell aplasia, aplastic anemia, and autoimmune hemolytic anemias can occur. Neutropenia, the most common cytopenia in LGL leukemia, is thought to be at least partly autoimmune, as the degree of neutropenia is often worse than would be expected solely from bone-marrow infiltration of LGL cells.^10,14,15

Rheumatoid arthritis is the most common autoimmune condition associated with LGL leukemia, with a reported incidence between 11% and 36%.^13–15

Felty syndrome (rheumatoid arthritis, splenomegaly, and neutropenia) is often associated with LGL leukemia and is thought by some to be part of the same disease process.¹⁵

Treat with immunosuppressives if needed

Indications for treating LGL leukemia include the development of cytopenias and associated autoimmune diseases. Immunosuppressive agents, such as methotrexate, cyclophosphamide, and cyclosporine, are commonly used.^10,11,14 Most evidence of treatment efficacy is from retrospective studies and case reports, with widely variable response rates that overall are around 50%.¹⁰

ACQUIRED HEMOPHILIA A AND HEMATOLOGIC MALIGNANCY

A systematic review found 30 cases of AHA associated with hematologic malignancies.¹⁶ The largest case series¹⁷ in this analysis had 8 patients, and included diagnoses of chronic lymphocytic leukemia, erythroleukemia, myelofibrosis, multiple myeloma, and myelodysplastic syndrome. In 3 of these patients, the appearance of the inhibitor preceded the diagnosis of the underlying malignancy by an average of 3.5 months. In 1 patient with erythroleukemia and another with multiple myeloma, the activity of the inhibitor could be clearly correlated with the underlying malignancy. In the other 6 patients, no association between the two could be made.

In the same series, complete resolution of the inhibitor was related only to the level of Bethesda titer present at diagnosis, with those who achieved resolution having lower mean Bethesda titers.¹⁷ Similarly, in EACH2, lower inhibitor Bethesda titers and higher factor VIII levels at presentation were associated with faster inhibitor eradication and normalization of factor VIII levels.⁷

Murphy et al¹⁸ described a 62-year-old woman with Felty syndrome who developed a factor VIII inhibitor and was subsequently given a diagnosis of LGL leukemia. Treatment with immunosuppressive agents, including cyclophosphamide, azathioprine, and rituximab, successfully eradicated her factor VIII inhibitor, although the LGL leukemia persisted.

Case conclusion: Eradication of factor VIII inhibitor

Our patient, similar to the patient described by Murphy et al¹⁸ above, had eradication of the factor VIII inhibitor despite persistence of LGL leukemia. Between the time of diagnosis at our clinic, when she had 54% LGLs, and eradication of the inhibitor 3 months later, the LGL percentage ranged from 45% to 89%. No clear direct correlation between LGL and factor VIII inhibitor levels could be detected.

Given the strong association of LGL leukemia with autoimmune disease, it is tempting to believe that her factor VIII inhibitor was somehow related to her malignancy, although the exact mechanism remained unclear. The average age at diagnosis is 60 for LGL leukemia¹¹ and over 70 for AHA,^5,6 so advanced age may be the common denominator. Whether or not our patient will have recurrence of her factor VIII inhibitor or the development of other autoimmune diseases with the persistence of her LGL leukemia remains to be seen.

At last follow-up, our patient was off all therapy and continued to have normal aPTT and factor VIII levels. Repeat flow cytometry after treatment of her factor VIII inhibitor showed persistence of a clonal T-cell population, although reduced from 72% to 60%. It may be that the 2 entities were unrelated, and the clonal T-cell population was simply fluctuating over time. This can be determined only with further observation. As the patient had no symptoms from her LGL leukemia, she continued to be observed without treatment.

TAKE-HOME POINTS

• The coagulation assay is key to initially assessing a bleeding abnormality; whether the prothrombin time and aPTT are normal or prolonged narrows the differential diagnosis and determines next steps in evaluation.
• Mixing studies can help pinpoint the responsible deficient factor.
• Acquired factor VIII deficiency, also known as AHA, may be caused by autoimmune disease, malignancy, or medications, but it is usually idiopathic.
• AHA treatment is focused on achieving hemostasis and reducing factor VIII inhibitor.
• Lymphocytosis should be evaluated with a peripheral blood smear and flow cytometry to determine if the population is polyclonal (associated with infection) or clonal (associated with malignancy).
• LGL leukemia is usually a chronic, indolent disease, although an uncommon subtype has an aggressive course.
• The association between AHA and LGL leukemia is unclear, and both conditions must be monitored and managed.

A 61-year-old woman presented to our hematology clinic for evaluation of multiple episodes of bruising. The first episode occurred 8 months earlier, when she developed a large bruise after water skiing. Two months before coming to us, she went to her local emergency room because of new bruising and was found to have a prolonged activated partial thromboplastin time (aPTT) of 60 seconds (reference range 23.3–34.9), but she underwent no further testing at that time.

At presentation to our clinic, she reported having no fevers, night sweats, unintentional weight loss, swollen lymph nodes, joint pain, rashes, mouth sores, nosebleeds, or blood in the urine or stool. Her history was notable only for hypothyroidism, which was diagnosed in the previous year. Her medications included levothyroxine, vitamin D₃, and vitamin C. She had been taking a baby aspirin daily for the past 10 years but had stopped 1 month earlier because of the bruising.

atieh_bruising_t1.jpg

On examination, she had a single small hematoma on her right thigh. She had no ecchymoses, petechiae, or adenopathy, and her spleen was nonpalpable.

Ten years earlier she had been evaluated for a possible transient ischemic attack; laboratory results at that time included a normal aPTT of 25.1 seconds and a normal factor VIII level of 153% (reference range 50%–173%).

atieh_bruising_t2.jpg

Laboratory testing at our clinic showed a normal complete blood cell count (Table 1); the coagulation factor assay confirmed that her aPTT was elevated (prolonged), but other values were normal (Table 2).

atieh_bruising_t3.jpg

Causes of an isolated prolonged aPTT include medications (eg, heparin), inherited factor deficiencies, acquired inhibitors of coagulation factors, and inherited or acquired von Willebrand disease. Lupus anticoagulant can prolong the aPTT but is usually associated with thrombosis rather than bleeding.¹ The differential diagnoses for this and other patterns of coagulation assay abnormalities are listed in Table 3.

EVALUATION FOR AN ISOLATED PROLONGED aPTT

1. What is the appropriate next test to evaluate this patient’s prolonged aPTT?

• Lupus anticoagulant panel
• Coagulation factor levels
• Mixing studies
• Bethesda assay

Mixing studies

Once a prolonged aPTT is confirmed, the appropriate next step is a mixing study. This involves mixing the patient’s plasma with pooled normal plasma in a 1-to-1 ratio, then repeating the aPTT test immediately, and again after 1 hour of incubation at 37°C. If the patient does not have enough of one of the coagulation factors, the aPTT immediately returns to the normal range when plasma is mixed with the pooled plasma because the pooled plasma contains the factor that is lacking. If this happens, then factor assays should be performed to identify the deficient factor.¹

Various antibodies that inhibit coagulation factors can also affect the aPTT. There are 2 general types: immediate-acting and delayed.

With an immediate-acting inhibitor, the aPTT does not correct into the normal range with initial mixing. Immediate-acting inhibitors are often seen together with lupus anticoagulants, which are nonspecific phospholipid antibodies. If an immediate-acting inhibitor is detected, further testing should focus on evaluation for lupus anticoagulant, including phospholipid-dependency studies.

With a delayed inhibitor, the aPTT initially comes down, but subsequently goes back up after incubation. Acquired factor VIII inhibitor is a classic delayed-type inhibitor and is also the most common factor inhibitor.¹ If a delayed-acting inhibitor is found, specific intrinsic factor levels should be measured (factors VIII, IX, XI, and XII),² and testing should also be done for lupus anticoagulant, as these inhibitors may occur together.

Bethesda assay

atieh_bruising_t4.jpg

If factor levels are decreased, a Bethesda assay should be performed to differentiate a specific factor inhibitor from a lupus anticoagulant. In the case of a factor VIII inhibitor, serial dilutions of patient plasma are incubated at 37°C with pooled normal plasma for 2 hours, then factor VIII activity is measured. The reciprocal dilution of patient plasma that results in 50% of factor VIII activity in the control plasma is 1 Bethesda unit (BU). The stronger the inhibitor in the patient’s sample, the more dilutions are required to measure factor VIII activity, and thus the higher the Bethesda titer.³

Case continued: Results of mixing and Bethesda studies

atieh_bruising_t5.jpg

Results of the mixing study (Table 4) showed an initial correction of the aPTT in a 1-to-1 mix, but after 1 hour of incubation, the aPTT was again prolonged at 42 seconds (reference range < 37.3). Further testing revealed very low levels of factor VIII (< 1%), and the presence of a factor VIII inhibitor, quantified at 5.8 BU (reference range < 0.5) (Table 5). Additional coagulation tests, including von Willebrand factor testing and a lupus anticoagulation panel, were negative.

 

 

FACTOR VIII INHIBITOR EVALUATION

2. What is the most likely underlying condition associated with this patient’s factor VIII inhibitor?

• Autoimmune disease
• Malignancy
• A medication
• Unknown (idiopathic)

Acquired hemophilia A (AHA) is a rare disorder caused by autoantibodies against factor VIII. Its estimated incidence is about 1 person per million per year.⁴ It usually presents as unexplained bruising or bleeding and is only rarely diagnosed by an incidentally noted prolonged aPTT. The severity of bleeding is variable and can include subcutaneous, soft-tissue, retroperitoneal, gastrointestinal, and intracranial hemorrhage.⁵

AHA is considered idiopathic in more than half of cases. A study based on a European registry⁵ of 501 patients with AHA and a UK study⁶ of 172 patients found no underlying disease in 52% and 65% of patients, respectively. For patients with an identified cause, the most common causes were malignancy (12%⁵ and 15%⁶) and autoimmune disease (12%⁵ and 17%⁶).

Drugs have rarely been associated with factor VIII inhibitors. Such occurrences have been reported with interferon, blood thinners, antibiotics, and psychiatric medications, but no study yet has indicated causation. However, patients with congenital hemophilia A treated with factor VIII preparations have about a 15% chance of developing factor VIII inhibitors. In this setting, inhibitors develop in response to recombinant factor VIII exposure, unlike the autoimmune phenomena seen in AHA.

TREATMENT OF ACQUIRED HEMOPHILIA A

3. What is the most appropriate treatment for AHA?

• Desmopressin and prednisone
• Recombinant porcine factor VIII and prednisone plus cyclophosphamide
• Recombinant factor VIIa and rituximab
• Any of the above

Any of the above regimens can be used. In general, treatment of AHA has two purposes: to stop acute hemorrhage, and to reduce the level of factor VIII inhibitor. No standard treatment guidelines are available; evidence of the effectiveness of different drugs is based largely on data on congenital hemophilia A.³

Acute treatment to stop bleeding

Initial treatment of AHA often focuses on stopping an acute hemorrhage by either raising circulating levels of factor VIII or bypassing it in the coagulation cascade.

Desmopressin can temporarily raise factor VIII levels, but it is often ineffective in AHA unless the patient has very low inhibitor titers.³

Factor VIII concentrate (human or recombinant porcine factor VIII) may be effective in patients with low inhibitor titers (< 5 BU). Higher doses are often required than those used in congenital hemophilia A. Factor VIII concentrate is usually combined with immunosuppressive treatment to lower the factor VIII inhibitor level (described below).³

If these methods are ineffective or the patient has high inhibitor titers (> 5 BU), activated prothrombin complex concentrates, known as FEIBA (factor eight inhibitor bypassing activity), or recombinant factor VIIa is available. These agents bypass factor VIII in the clotting cascade.

Immunosuppression to reduce factor VIII inhibitor

Immunosuppressive agents are the mainstay of AHA treatment to lower the inhibitor level.

Regimens vary. A 2003 meta-analysis⁴ including 249 patients found that prednisone alone resulted in complete response in about 30% of patients, and the addition of cyclophosphamide increased the response rate to 60% to 100%. High-dose intravenous immunoglobulin led to conflicting results. Conclusions were limited by the variability of dosing and duration in treatment regimens among the 20 different studies included.

An analysis of 331 patients in the European Acquired Hemophilia Registry (EACH2)⁷ found that steroids alone produced remission in 48% of patients, while steroids combined with cyclophosphamide raised the rate to 70%. Rituximab-based regimens were successful in 59% but required twice as long to achieve remission as steroid or cyclophosphamide-based regimens. No benefit was noted from intravenous immunoglobulin.

Risks of disease and treatment

AHA is associated with significant risk of morbidity and death related to bleeding, complications of treatment, and underlying disease.

In EACH2, 16 of the 331 patients died of bleeding, 16 died of causes related to immunosuppression, and 45 died of causes related to the underlying condition.⁵ In the UK registry of 172 patients, 13 patients died of bleeding, and 12 died of sepsis related to immunosuppression.⁶

The factor VIII level and inhibitor titer are not necessarily useful in stratifying bleeding risk, as severe and fatal bleeding can occur at variable levels and patients remain at risk of bleeding as long as the inhibitor persists.^6,7

 

 

CASE CONTINUED: TREATMENT, LYMPHOCYTOSIS

The patient was started on 60 mg daily of prednisone, resulting in a decrease in her aPTT, increase in factor VIII level, and lower Bethesda titer. On a return visit, her absolute lymphocyte count was 7.04 × 10⁹/L (reference range 1.0–4.0). She reported no fevers, chills, or recent infections.  

EVALUATING LYMPHOCYTOSIS

Lymphocytosis is defined in most laboratories as an absolute lymphocyte count greater than 4.0 × 10⁹/L for adults. Normally, T cells (CD3+) make up 60% to 80% of lymphocytes, B cells (CD20+) 10% to 20%, and natural killer (NK) cells (CD3–, CD56+) 5% to 10%. Lymphocytosis is usually caused by infection, but it can have other causes, including malignancy.

Peripheral blood smear. If there is no clear cause of lymphocytosis, a peripheral blood smear can be used to assess lymphocyte morphology, providing clues to the underlying etiology. For example, atypical lymphocytes are often seen in infectious mononucleosis, while “smudge” lymphocytes are characteristic of chronic lymphocytic leukemia. If a peripheral smear shows abnormal morphology, further workup should include establishing whether the lymphocytes are polyclonal or clonal.⁸

CASE CONTINUED: LARGE GRANULAR LYMPHOCYTES

atieh_bruising_f1.jpg

On the patient’s peripheral smear, 54% of lymphocytes were large lymphocytes with moderate amounts of pale cytoplasm filled with azurophilic granules, consistent with large granular lymphocytes (LGLs) (Figure 1).

4. What is the next step to evaluate the patient’s lymphocytosis?

• Bone marrow biopsy
• Karyotype analysis
• Flow cytometry
• Fluorescence in situ hybridization

Flow cytometry with V-beta analysis is the best first test to determine the cause of lymphocytosis after review of the peripheral smear. For persistent lymphocytosis, flow cytometry should be done even if a peripheral smear shows normal lymphocyte morphology.

Most T cells possess receptors composed of alpha and beta chains, each encoded by variable (V), diversity (D), joining (J), and constant (C) gene segments. The V, D, and J segments undergo rearrangement during T-cell development in the thymus based on antigen exposure, producing a diverse T-cell receptor population.

In a polyclonal population of lymphocytes, the T-cell receptors have a variety of gene segment arrangements, indicating normal T-cell development. But in a clonal population of lymphocytes, the T-cell receptors have a single identical gene segment arrangement, indicating they all originated from a single clone.⁹ Lymphocytosis in response to an infection is typically polyclonal, while malignant lymphocytosis is clonal. 

Monoclonal antibodies against many of the variable regions of the beta chain (V-beta) of T-cell receptors have been developed, enabling flow cytometry to establish clonality.

T-cell receptor gene rearrangement studies can also be performed using polymerase chain reaction and Southern blot techniques.⁹

Karyotype analysis is usually not performed for the finding of LGLs, because most leukemias (eg, T-cell and NK-cell leukemias) have cells with a normal karyotype. 

Bone marrow biopsy is invasive and usually not required to evaluate LGLs. It can be especially risky for a patient with a bleeding disorder such as a factor VIII inhibitor.¹⁰

Case continued: Flow cytometry confirms clonality

Subsequent flow cytometry found that more than 50% of the patient’s lymphocytes were LGLs that co-expressed CD3+, CD8+, CD56+, and CD57+, with aberrantly decreased CD7 expression. T-cell V-beta analysis demonstrated an expansion of the V-beta 17 family, and T-cell receptor gene analysis with polymerase chain reaction confirmed the presence of a clonal rearrangement.

LGL LEUKEMIA: CLASSIFICATION AND MANAGEMENT

LGLs normally account for 10% to 15% of peripheral mononuclear cells.¹¹ LGL leukemia is caused by a clonal population of cytotoxic T cells or NK cells and involves an increased number of LGLs (usually > 2 × 10⁹/L).¹⁰

LGL leukemia is divided into 3 categories according to the most recent World Health Organization classification^10,12:

T-cell LGL leukemia (about 85% of cases) is considered indolent but can cause significant cytopenias and is often associated with autoimmune disease.¹³ Cells usually express a CD3+, CD8+, CD16+, and CD57+ phenotype. Survival is about 70% at 10 years.

Chronic NK-cell lymphocytosis (about 10%) also tends to have an indolent course with cytopenia and an autoimmune association, and with a similar prognosis to T-cell LGL leukemia. Cells express a CD3–, CD16+, and CD56+ phenotype.

Aggressive NK-cell LGL leukemia (about 5%) is associated with Epstein-Barr virus infection and occurs in younger patients. It is characterized by severe cytopenias, “B symptoms” (ie, fever, night sweats, weight loss), and has a very poor prognosis. Like chronic NK-cell lymphocytosis, cells express a CD3–, CD16+, and CD56+ phenotype. Fas (CD95) and Fas-ligand (CD178) are strongly expressed.^10,13

Most cases of LGL leukemia can be diagnosed on the basis of classic morphology on peripheral blood smear and evidence of clonality on flow cytometry or gene rearrangement studies. T-cell receptor gene studies cannot be used to establish clonality in the NK subtypes, as NK cells do not express T-cell receptors.¹¹

Case continued: Diagnosis, continued course 

In our patient, T-cell LGL leukemia was diagnosed on the basis of the peripheral smear, flow cytometry results, and positive T-cell receptor gene studies for clonal rearrangement in the T-cell receptor beta region.

While her corticosteroid therapy was being tapered, her factor III inhibitor level increased, and she had a small episode of bleeding, prompting the start of cyclophosphamide 50 mg daily with lower doses of prednisone.

atieh_bruising_f2.jpg

She then developed elevated liver enzymes, prompting discontinuation of cyclophosphamide. Rituximab was started and continued for 4 weekly doses, resulting in normalization of aPTT and factor VIII level with undetectable Bethesda titers (Figure 2).

 

 

LGL LEUKEMIA AND AUTOIMMUNE DISEASE

Patients with LGL leukemia commonly have or develop autoimmune conditions. Immune-mediated cytopenias including pure red cell aplasia, aplastic anemia, and autoimmune hemolytic anemias can occur. Neutropenia, the most common cytopenia in LGL leukemia, is thought to be at least partly autoimmune, as the degree of neutropenia is often worse than would be expected solely from bone-marrow infiltration of LGL cells.^10,14,15

Rheumatoid arthritis is the most common autoimmune condition associated with LGL leukemia, with a reported incidence between 11% and 36%.^13–15

Felty syndrome (rheumatoid arthritis, splenomegaly, and neutropenia) is often associated with LGL leukemia and is thought by some to be part of the same disease process.¹⁵

Treat with immunosuppressives if needed

Indications for treating LGL leukemia include the development of cytopenias and associated autoimmune diseases. Immunosuppressive agents, such as methotrexate, cyclophosphamide, and cyclosporine, are commonly used.^10,11,14 Most evidence of treatment efficacy is from retrospective studies and case reports, with widely variable response rates that overall are around 50%.¹⁰

ACQUIRED HEMOPHILIA A AND HEMATOLOGIC MALIGNANCY

A systematic review found 30 cases of AHA associated with hematologic malignancies.¹⁶ The largest case series¹⁷ in this analysis had 8 patients, and included diagnoses of chronic lymphocytic leukemia, erythroleukemia, myelofibrosis, multiple myeloma, and myelodysplastic syndrome. In 3 of these patients, the appearance of the inhibitor preceded the diagnosis of the underlying malignancy by an average of 3.5 months. In 1 patient with erythroleukemia and another with multiple myeloma, the activity of the inhibitor could be clearly correlated with the underlying malignancy. In the other 6 patients, no association between the two could be made.

In the same series, complete resolution of the inhibitor was related only to the level of Bethesda titer present at diagnosis, with those who achieved resolution having lower mean Bethesda titers.¹⁷ Similarly, in EACH2, lower inhibitor Bethesda titers and higher factor VIII levels at presentation were associated with faster inhibitor eradication and normalization of factor VIII levels.⁷

Murphy et al¹⁸ described a 62-year-old woman with Felty syndrome who developed a factor VIII inhibitor and was subsequently given a diagnosis of LGL leukemia. Treatment with immunosuppressive agents, including cyclophosphamide, azathioprine, and rituximab, successfully eradicated her factor VIII inhibitor, although the LGL leukemia persisted.

Case conclusion: Eradication of factor VIII inhibitor

Our patient, similar to the patient described by Murphy et al¹⁸ above, had eradication of the factor VIII inhibitor despite persistence of LGL leukemia. Between the time of diagnosis at our clinic, when she had 54% LGLs, and eradication of the inhibitor 3 months later, the LGL percentage ranged from 45% to 89%. No clear direct correlation between LGL and factor VIII inhibitor levels could be detected.

Given the strong association of LGL leukemia with autoimmune disease, it is tempting to believe that her factor VIII inhibitor was somehow related to her malignancy, although the exact mechanism remained unclear. The average age at diagnosis is 60 for LGL leukemia¹¹ and over 70 for AHA,^5,6 so advanced age may be the common denominator. Whether or not our patient will have recurrence of her factor VIII inhibitor or the development of other autoimmune diseases with the persistence of her LGL leukemia remains to be seen.

At last follow-up, our patient was off all therapy and continued to have normal aPTT and factor VIII levels. Repeat flow cytometry after treatment of her factor VIII inhibitor showed persistence of a clonal T-cell population, although reduced from 72% to 60%. It may be that the 2 entities were unrelated, and the clonal T-cell population was simply fluctuating over time. This can be determined only with further observation. As the patient had no symptoms from her LGL leukemia, she continued to be observed without treatment.

TAKE-HOME POINTS

• The coagulation assay is key to initially assessing a bleeding abnormality; whether the prothrombin time and aPTT are normal or prolonged narrows the differential diagnosis and determines next steps in evaluation.
• Mixing studies can help pinpoint the responsible deficient factor.
• Acquired factor VIII deficiency, also known as AHA, may be caused by autoimmune disease, malignancy, or medications, but it is usually idiopathic.
• AHA treatment is focused on achieving hemostasis and reducing factor VIII inhibitor.
• Lymphocytosis should be evaluated with a peripheral blood smear and flow cytometry to determine if the population is polyclonal (associated with infection) or clonal (associated with malignancy).
• LGL leukemia is usually a chronic, indolent disease, although an uncommon subtype has an aggressive course.
• The association between AHA and LGL leukemia is unclear, and both conditions must be monitored and managed.

Thanks to advances in medical science and our understanding of inherited and acquired liver disease, many more children with acquired or congenital liver disease survive into adulthood than they did 2 decades ago. Improvements in immunosuppression and surgery have increased the chances of pediatric liver transplant recipients reaching adulthood, with a survival rate of 75% at 15 to 20 years.¹

selvakumar_pediatriccholestaticliverdisease_f1.jpg

The causes of cholestasis in children are different from those in adults (Figure 1). Pediatric cholestasis is caused either by genetic defects that affect the process of synthesis, processing, or secreting bile, or by mechanical obstruction of the biliary tract, the classic example being biliary atresia.

With the growing number of adult patients with pediatric-onset liver disease, internists and adult hepatologists need to be aware of these liver diseases and develop expertise to manage this challenging group of patients. Moreover, young adults with pediatric-onset chronic liver disease pose distinct challenges such as pregnancy, adherence to medical regimens, and psychosocial changes in life.

These patients need a “transition of care” rather than a “transfer of care.” Transition of care is a multifaceted process that takes the medical, educational, and psychosocial needs of the patient into consideration before switching their care to adult care physicians, whereas transfer of care is simply an administrative process of change to adult care without previous knowledge of the patients.²

selvakumar_pediatriccholestaticliverdisease_t1.jpg

In this article, we discuss relatively common types of inherited childhood-onset cholestatic disease and their implications and management in adulthood (Table 1). Other acquired or rare cholestatic diseases are beyond the scope of this article.

BILIARY ATRESIA

Biliary atresia is a progressive inflammatory fibrosclerosing cholangiopathy of unknown cause. Its prevalence varies with geographic location, ranging from 1 in 6,000 to 1 in 19,000, with the highest prevalence in Taiwan.³

Biliary atresia usually presents within the first few weeks of life, with progressive cholestasis leading to failure to thrive and to fat-soluble vitamin deficiency. Approximately 20% of patients have congenital splenic, gastrointestinal, genitourinary, cardiac, and venous malformations.^4,5 Untreated, biliary atresia progresses to end-stage liver disease and death within 2 years.

The first-line treatment for biliary atresia is to establish biliary outflow with the Kasai procedure (hepatic portoenterostomy), in which a jejunal limb is anastomosed in a Roux-en-Y with the liver. The outcomes of the Kasai procedure depend on the timing of surgery, so timely diagnosis of biliary atresia is crucial. When the Kasai procedure is performed within 60 days of birth, biliary flow is achieved in up to 70% of patients; but if performed after 90 days, biliary flow is achieved in fewer than 25%.⁶

Long-term outcomes of biliary atresia in patients with their native liver have been reported in a few studies.

In a French study,⁷ 743 patients with biliary atresia underwent the Kasai procedure at a median age of 60 days. Survival rates were 57.1% at 2 years, 37.9% at 5 years, 32.4% at 10 years, and 28.5% at 15 years. In other studies,^4–9 the 20-year transplant-free survival rate ranged from 23% to 46%. Therefore, at least one-third of children with biliary atresia survive to adulthood with their native liver.

Implications of biliary atresia in adulthood

Although the Kasai procedure improves biliary outflow, up to 70% of patients develop complications of biliary atresia such as progressive fibrosis, cirrhosis, portal hypertension, cholangitis, and hepatocellular carcinoma, even after a successful Kasai procedure.¹⁰

Portal hypertension with evidence of splenomegaly, thrombocytopenia, or ascites is found in two-thirds of long-term survivors of biliary atresia with a native liver, with variceal hemorrhage occurring in 30%.¹¹ Therefore, patients with biliary atresia who have evidence of portal hypertension should be screened for varices with upper endoscopy on an annual basis. Management of variceal hemorrhage in these patients includes the use of octreotide, antibiotics, variceal ligation, and sclerotherapy; primary prophylaxis can be achieved with beta-blockers and endoscopic variceal ligation.¹²

Cholangitis is frequent, occurring in 40% to 60% of biliary atresia patients after the Kasai procedure, and about one-fourth of these patients have multiple episodes.¹³ The number of episodes of cholangitis negatively affects transplant-free survival.¹⁴ Patients with cholangitis should be adequately treated with oral or intravenous antibiotics depending on the severity of presentation. The role of prophylaxis with antibiotics remains unclear.¹⁵

Pulmonary complications such as hepato­pulmonary syndrome and portopulmonary hypertension can also occur in biliary atresia patients with a native liver. It is important for physicians to be aware of these complications and to screen for them, for example, with agitated saline echocardiography for hepatopulmonary syndrome and with echocardiography for portopulmonary hypertension. Timely screening is crucial, as the outcome of liver transplant depends on the severity at the time of transplant in these conditions, especially portopulmonary hypertension.

Hepatocellular carcinoma has been rarely reported in children with biliary atresia,¹⁶ so well-defined guidelines for screening in young adults with biliary atresia are lacking. Most centers recommend screening with ultrasonography of the abdomen and alpha-fetoprotein measurement every 6 months or annually starting soon after the Kasai procedure, since hepatocellular carcinoma has been reported in children as young as age 2.¹⁶

Transplant. Adult hepatologists are faced with the challenging task of deciding when it is time for transplant, balancing the long-term complications of biliary atresia with the risk of long-term immunosuppression after transplant. In addition, young adults with these complications may have preserved synthetic function, resulting in low Model for End-Stage Liver Disease (MELD) scores, which may complicate the process of listing for transplant.

Neurocognitive deficits are reported in children with biliary atresia,¹⁷ but young adults with biliary atresia generally have reasonable cognitive function and prospects for education and employment.

Pregnancy with successful outcomes has been reported.⁸

 

 

ALAGILLE SYNDROME

Alagille syndrome is an autosomal-dominant multisystemic disease caused by mutations in the JAG1 gene (accounting for > 95% of cases) and the NOTCH2 gene, with highly variable expression.¹⁸

Extrahepatic manifestations include butterfly vertebral defects, facial dysmorphism (eg, deep-set and low-set eyes, with characteristic “triangular” facies), posterior embryotoxon (a congenital defect of the eye characterized by an opaque ring around the margin of the cornea), peripheral pulmonary stenosis, renal abnormalities, and vascular malformations.

Hepatic manifestations vary from asymptomatic laboratory abnormalities to progressive cholestasis starting in early infancy with intractable pruritus, xanthomas, failure to thrive, and end-stage liver disease requiring liver transplant in childhood in 15% to 20% of patients.¹⁹

Implications of Alagille syndrome in adulthood

Transplant. Interestingly, the phenotype of hepatic disease is already established in childhood, with minimal or no progression in adulthood. Most children with minimal liver disease experience spontaneous resolution, whereas those with significant cholestasis might ultimately develop progressive liver fibrosis or cirrhosis requiring liver transplant in childhood. Only a small subset of children with minimal cholestasis progress to end-stage liver disease in late childhood or early adulthood.²⁰ Therefore, liver transplant for progressive liver disease from significant cholestasis almost always occurs in childhood, usually between ages 1 and 4.²¹

In a retrospective study comparing posttransplant outcomes in children with Alagille syndrome and biliary atresia, 1-year patient survival was excellent overall in children with Alagille syndrome, although slightly lower than in children with biliary atresia, most likely owing to extrahepatic morbidities of Alagille syndrome and especially the use of immunosuppression in those with renal disease.²¹ Similarly, 1- and 5-year patient and graft survival outcomes of liver transplant in adults with Alagille syndrome were also excellent compared with those who received a liver transplant in childhood for Alagille syndrome or in adulthood for biliary atresia.²²

Hepatocellular carcinoma has occurred in these patients in the absence of cirrhosis, which makes implementation of prognostic and surveillance strategies almost impossible to design for them. Annual ultrasonography with alpha-fetoprotein testing might be applicable in Alagille syndrome patients. However, deciding which patients should undergo this testing and when it should start will be challenging, given the paucity of data.

Cardiovascular disease. Cardiac phenotype is also mostly established in childhood, with the pulmonary vasculature being most commonly involved.¹⁹ In contrast, renal and other vascular abnormalities can manifest in adulthood. Renal manifestations vary and include structural anomalies such as hyperechoic kidneys or renal cysts, which can manifest in childhood, and some abnormalities such as hypertension and renal artery stenosis that can manifest in adulthood.^23,24

Vasculopathy is reported to involve the intracranial, renal, and intra-abdominal blood vessels.²⁵ Neurovascular accidents such as stroke and intracranial hemorrhage can occur at any age, with significant rates of morbidity and death.²⁶ Therefore, some experts recommend magnetic resonance angiography every 5 years and before any major intervention to prevent these devastating complications.²⁰

Pregnancy. Successful pregnancies have been reported. Preexisting cardiac and hepatic disease can complicate pregnancy depending on the severity of the disease. Because of the autosomal-dominant pattern of inheritance, infants have a 50% risk of the disease, so genetic counseling should be seriously considered before conception.²⁷ Prenatal diagnosis is possible, but the lack of genotype-phenotype correlation precludes its use in clinical practice.

PROGRESSIVE FAMILIAL INTRAHEPATIC CHOLESTASIS

Progressive familial intrahepatic cholestasis (PFIC) is a heterogeneous group of autosomal-recessive conditions associated with disruption of bile formation causing cholestatic liver disease in infants and young children. Three types have been described, depending on the genetic mutation in the hepatobiliary transport pathway:

• PFIC 1 (Byler disease) is caused by impaired bile salt secretion due to mutations in the ATP8B1 gene encoding for the familial intrahepatic cholestasis 1 (FIC 1) protein
• PFIC 2 is caused by impaired bile salt secretion due to mutations in the ABCB11 gene encoding for the bile salt export pump (BSEP) protein
• PFIC 3 is caused by impaired biliary phospholipid secretion due to a defect in ABCB4 encoding for multidrug resistance 3 (MDR3) protein.²⁸

PFIC 1 and 2 manifest with low gamma-glutamyl transferase (GGT) cholestasis, whereas PFIC 3 presents with high GGT cholestasis.

PFIC 1 and PFIC 2 usually cause cholestasis in early infancy, but PFIC 3 can cause cholestasis in late infancy, childhood, and even adulthood.

Because ATP8B1 is expressed in other tissues, PFIC 1 is characterized by extrahepatic manifestations such as sensorineural hearing loss, growth failure, severe diarrhea, and pancreatic insufficiency.

 

 

Implications of PFIC in adulthood

PFIC 1 and 2 (low-GGT cholestasis) are usually progressive and often lead to end-stage liver disease and cirrhosis before adulthood. Therefore, almost all patients with PFIC 1 and 2 undergo liver transplant or at least a biliary diversion procedure before reaching adulthood. Intractable pruritus is one of the most challenging clinical manifestations in patients with PFIC.

First-line management is pharmacologic and includes ursodeoxycholic acid, antihistamines (eg, hydroxyzine), bile acid sequestrants (eg, cholestyramine, colestipol), naltrexone, and rifampin, but these have limited efficacy.¹⁰

Most patients, especially those with PFIC 1 and 2, undergo a biliary diversion procedure such as partial external biliary diversion (cholecystojejunocutaneostomy), ileal exclusion, or partial internal biliary diversion (cholecystojejunocolic anastomosis) to decrease enterohepatic circulation of bile salts. The efficacy of these procedures is very limited in patients with established cirrhosis. Excessive losses of bile can occur through the biliary stoma, leading to dehydration in patients with external biliary diversion. In patients who are not candidates for biliary diversion, endoscopic nasobiliary drainage of pancreatobiliary secretions could be achieved by placing a catheter in the common bile duct; this has been reported to be effective in relieving cholestasis in a few cases.²⁹

Liver transplant is needed in patients with progressive liver disease and intractable pruritus despite medical management and biliary diversion. Unlike in biliary atresia, liver transplant is not curative in PFIC 1, due to extrahepatic manifestations: patients with PFIC 1 can still have intractable diarrhea and pancreatitis after liver transplant. More importantly, allograft steatohepatitis with further progression to cirrhosis can occur after liver transplant in patients with PFIC 1. Interestingly, biliary diversion has been reported to improve graft steatosis and diarrhea after liver transplant.³⁰

Although graft survival after transplant is good in patients with PFIC 2, recurrence of low-GGT cholestasis has been reported and is believed to be due to the formation of anti-bile salt export pump (anti-BSEP) antibodies by the host immune system in response to exposure to new proteins from the transplant graft.³¹

Cancer. The risk of malignancy, especially hepatocellular carcinoma, is also increased in PFIC 2, affecting nearly 15% of patients. Therefore, standard hepatocellular carcinoma surveillance with ultrasonography or alpha-fetoprotein testing or both is recommended in patients with PFIC 2. Cholangiocarcinoma and pancreatic adenocarcinoma have also been reported in patients with PFIC 2.²⁰

Incomplete penetrance of mutations in ATP8B1 and ABCB11 can cause recurrent episodes of cholestasis and pruritus with asymptomatic periods between episodes, referred to as benign recurrent intrahepatic cholestasis. Prognosis is usually good, with no progression to cirrhosis.³²

Pregnancy. In contrast to FIC 1 and BSEP deficiency, MDR3 defects lead to a wide phenotypic spectrum depending on the type of mutation. Heterozygous mutation is associated with increased risk of development of cholestasis during pregnancy, which typically presents with generalized pruritus in the third trimester and is associated with adverse fetal outcomes. Intrahepatic cholestasis of pregnancy is usually treated with ursodeoxycholic acid, with reported improvement in pruritus, liver function, and pregnancy outcomes.³³

In adults, drug-induced liver injury and idiopathic cirrhosis have also been described with MDR3 defects. Intrahepatic lithiasis and cholesterol gallstones can also occur with MDR3 defects as a result of impaired secretion of biliary phospholipid.³² Despite intrahepatic cholestasis of pregnancy, successful outcomes have been reported in women with PFIC.²⁰

OTHER CHILDHOOD-ONSET INHERITED CHOLESTATIC DISEASES

Cystic fibrosis-associated liver disease

Nearly 40% of patients with cystic fibrosis develop liver disease.³⁴ Cystic fibrosis-associated liver disease encompasses a broad clinical spectrum including asymptomatic elevation of aminotransferases, neonatal cholestasis, hepatic steatosis, focal biliary cirrhosis, and multilobar cirrhosis. Cirrhosis and portal hypertension can occur in 5% to 10% of patients and is the third-leading cause of death in patients with cystic fibrosis.³⁵

Risk factors for cystic fibrosis-associated liver disease include male sex, meconium ileus, and severe CFTR gene mutation (class I–III) with pancreatic insufficiency. Cystic fibrosis-related cirrhosis is more frequent in children and adolescents, whereas noncirrhotic portal hypertension and intrahepatic cholangiopathies are more common in adults.³⁶

Limited available studies support treatment with ursodeoxycholic acid in patients with cholestasis to delay the progression of liver disease, but the impact of this drug on long-term outcome is unknown.²⁹

Most patients remain in compensated cirrhosis for many years before progressing to decompensated cirrhosis requiring liver transplant. Other indications for liver transplant include recurrent intractable variceal bleeding, hepatopulmonary syndrome, and portopulmonary hypertension. Combined liver and lung transplant may be considered in patients with advanced liver and lung disease. Outcomes after isolated liver or liver-lung transplant in cystic fibrosis patients have been comparable to those in patients with other liver diseases.³⁷

Defects in bile acid synthesis

Inherited defects of enzymes required for the synthesis of primary bile acids from cholesterol can cause cholestasis from impaired bile flow and production of hepatotoxic aberrant bile acids. The clinical presentation varies depending on the enzymatic defect and can range from liver disease of varying severity to neurologic manifestations. Idiopathic late-onset cholestasis and cirrhosis of unknown etiology have been reported in adults with bile acid synthesis defects.^38,39 Therefore, this diagnosis should be considered in cases of cryptogenic cirrhosis and other cholestatic features.

Treatment with primary bile acids (cholic acid) has been effective in most patients with defective bile acid synthesis.

Primary sclerosing cholangitis

Primary sclerosing cholangitis is characterized by progressive obliteration of intrahepatic and extrahepatic bile ducts and is most commonly seen in patients with inflammatory bowel disease. Sclerosing cholangitis can also be secondary to other diseases in children such as immunodeficiency syndromes, Langerhans cell histiocytosis, cystic fibrosis, or sickle cell anemia.⁴⁰ Neonatal sclerosing cholangitis is a rare autosomal-recessive disease characterized by a severe form of cholangiopathy in neonates and young infants requiring transplant. It can be associated with Kabuki syndrome and neonatal ichthyosis-sclerosing cholangitis syndrome.

Treatment options are limited. Ursodeoxycholic acid and oral vancomycin have variable efficacy. Liver transplant is needed in patients with decompensated cirrhosis. Patients with primary sclerosing cholangitis, especially adults, are at higher risk of developing cholangiocarcinoma, and therefore screening with ultrasonography or magnetic resonance imaging every 6 to 12 months is recommended.

The risk of preterm and cesarean deliveries may be elevated in women with primary sclerosing cholangitis, though data are limited.³³

 

 

PEDIATRIC LIVER TRANSPLANT RECIPIENTS WHO SURVIVE INTO ADULTHOOD

Adolescent rebellion poses risks

Outcomes of liver transplant in children and adolescents have improved tremendously in the past 2 decades with advances in surgical techniques, pre- and postoperative management, organ preservation, and immunosuppression. Now, most pediatric liver transplant recipients survive into adulthood, creating a unique challenge for internists and adult  care hepatologists.⁴¹

In rebellious adolescents and young adults, risk-taking behavior, nonadherence to immunosuppressive medications, alcohol intake, and substance abuse increase the risk of graft rejection and loss. Current immunosuppressive drugs such as calcineurin inhibitors (tacrolimus, cyclosporine), mycophenolate mofetil, sirolimus, and corticosteroids have drastically decreased rejection rates in compliant patients.⁴¹ Educating patients on the importance of taking their medications and avoiding alcohol and drug abuse is especially important for adolescents and young adults, as rates of nonadherence are high in these age groups.

Although pregnancy is usually successful after liver transplant, it should be considered high-risk due to reported complications such as graft rejection, diabetes, preeclampsia, sepsis, prematurity, and low birth weight. Conception should be avoided for at least 1 year after transplant.⁴² Appropriate counseling with regard to pregnancy and contraception is important.

There is no consensus on breastfeeding, but it is considered safe in women on low-dose calcineurin inhibitors.⁴³

Life is better with a new liver, but patients have special needs

Liver transplant is life-saving and improves quality of life. However, long-term pediatric liver transplant recipients face challenges such as strict adherence to medications and follow-up visits, avoiding exposure to infections, and fear of graft rejection.

Chronic liver disease in children leads to failure to thrive, growth failure, and even delayed puberty, which resolve in most patients after liver transplant before adulthood in the absence of other comorbidities.⁴⁴ However, these patients are reported to have lower psychosocial functioning and more psychiatric disorders such as anxiety or posttraumatic disorder.^41,44

Therefore, a psychologist or other mental health professional should be part of the management team from the time of pretransplant assessment to identify mental health problems and the need for adjustments before liver transplant. Ongoing psychosocial assessment after liver transplant is equally important to identify risks such as drug or alcohol abuse, depression, posttraumatic stress disorder, and medication nonadherence, all of which can negatively affect posttransplant outcome.⁴⁵

In addition, assessment of family functioning and structure is important for good long-term outcomes posttransplant; therefore, a social worker should also be a part of the transplant team. Psyschosocial assessment tools can identify high-risk candidates who would benefit from earlier intervention to avoid any negative impact posttransplant.

Neurocognitive development can be delayed in children with chronic liver disease, and the delay may persist even after liver transplant, with reported impairments in intellectual ability, language, verbal, and visuospatial functioning skills.⁴¹ In spite of this, a recent study found that more than half the study patients were employed at a median follow-up of 24 years from liver transplant and a median age of 27.⁴⁶

Remarkably, pediatric liver transplant recipients have reported quality of life comparable to that in the general population,⁴⁷ and even better than in patients with other chronic illnesses.⁴⁸

Long-term medical comorbidities in pediatric liver transplant recipients

Favorable outcomes such as long-term survival and good quality of life in pediatric liver transplant recipients are lessened by late complications such as portal vein thrombosis or biliary strictures needing interventions, chronic graft rejection, adverse effects of immunosuppression, and recurrence of the disease.

Split-liver transplant—splitting a deceased-donor allograft to provide grafts for 2 recipients—has revolutionized liver transplant by increasing the donor pool and thereby decreasing waitlist mortality rates, especially in pediatric candidates. Despite this advantage, split-liver transplant is technically challenging and associated with increased perioperative complications compared with whole-liver transplant, especially in adult recipients. Recently, experienced centers have reported favorable outcomes with split-liver transplant comparable to those with whole-liver transplant; therefore, split-liver transplant should be considered after careful evaluation of donor organ and recipient clinical status.⁴⁹

Old age in the recipient can also adversely affect liver transplant outcomes.⁵⁰

Interestingly, even in patients whose clinical course is unremarkable and biochemical values are normal, graft hepatitis or fibrosis of unknown cause with progression to cirrhosis has been described in the decade after transplant.⁴¹

Chronic rejection with eventual graft loss may be related to nonadherence in adolescents and can be reduced with use of an additional immunosuppressant such as sirolimus or mycophenolate. Chronic kidney disease can occur in about one-third of liver transplant recipients secondary to renal disease associated with primary disease (like Alagille syndrome), hepatorenal syndrome, and most importantly, use of calcineurin inhibitors.⁴⁵

Components of the metabolic syndrome such as type 2 diabetes, obesity, nonalcoholic fatty liver disease, hypertension, and dyslipidemia are also seen in long-term pediatric liver transplant survivors. Internists are advised to screen for these comorbidities so that interventions can be applied early to improve long-term health outcomes and graft survival.

Of importance, multiple studies have shown a 2-fold increase in the rates of de novo malignancy in liver transplant recipients, including solid-organ and lymphoproliferative cancers, probably due to long-term immunosuppression. Posttransplant lymphoproliferative disorder occurs at lower rates than with other solid-organ transplants; its incidence is greatest in pediatric patients and in the first 12 to 18 months after transplant.⁵¹

 

 

TRANSITION TO ADULT CARE

While the number of patients with childhood-onset liver disease and pediatric liver transplant recipients who survive into adulthood is increasing, there are no established guidelines or formal models for transitioning these patients into adult care. Consequently, studies on transitional process have examined various issues such as patient and parent frustration, poor medical knowledge among patients during transition, lack of parental facilitation, and inadequate knowledge on disease process among adult-care hepatologists.^52–54

selvakumar_pediatriccholestaticliverdisease_f2.jpg

A position paper on transitional care from Europe has put forth important strategies.⁵⁵ A multidisciplinary team including a pediatric provider, nurse practitioner, transplant coordinator, psychologist, social worker, and adult provider is vital for a successful transition. Key requirements for an adolescent patient include knowledge about the disease, demonstration of responsibility toward disease management, and self-management of healthcare needs such as scheduling appointments or insurance needs. This demanding task could be streamlined if pediatric liver transplant programs could adopt a formal transitional model (Figure 2). The transition could start as early as age 10 to 12, with gradual transfer of health management responsibilities from parents to adolescents from age 13 to 17.⁵⁶

A prolonged period of transition up to age 25 is preferred in complicated cases. Distinctive consideration for transition should include those with neurocognitive developmental delay from underlying disease or hepatic encephalopathy before transplant. These patients need additional support and time to achieve independence in health management before transition.⁵⁷ Validated questionnaires are available to assess readiness to transition into adult care,⁵⁸ implying that the decision to transition should not be based solely on age. 

Thanks to advances in medical science and our understanding of inherited and acquired liver disease, many more children with acquired or congenital liver disease survive into adulthood than they did 2 decades ago. Improvements in immunosuppression and surgery have increased the chances of pediatric liver transplant recipients reaching adulthood, with a survival rate of 75% at 15 to 20 years.¹

selvakumar_pediatriccholestaticliverdisease_f1.jpg

The causes of cholestasis in children are different from those in adults (Figure 1). Pediatric cholestasis is caused either by genetic defects that affect the process of synthesis, processing, or secreting bile, or by mechanical obstruction of the biliary tract, the classic example being biliary atresia.

With the growing number of adult patients with pediatric-onset liver disease, internists and adult hepatologists need to be aware of these liver diseases and develop expertise to manage this challenging group of patients. Moreover, young adults with pediatric-onset chronic liver disease pose distinct challenges such as pregnancy, adherence to medical regimens, and psychosocial changes in life.

These patients need a “transition of care” rather than a “transfer of care.” Transition of care is a multifaceted process that takes the medical, educational, and psychosocial needs of the patient into consideration before switching their care to adult care physicians, whereas transfer of care is simply an administrative process of change to adult care without previous knowledge of the patients.²

selvakumar_pediatriccholestaticliverdisease_t1.jpg

In this article, we discuss relatively common types of inherited childhood-onset cholestatic disease and their implications and management in adulthood (Table 1). Other acquired or rare cholestatic diseases are beyond the scope of this article.

BILIARY ATRESIA

Biliary atresia is a progressive inflammatory fibrosclerosing cholangiopathy of unknown cause. Its prevalence varies with geographic location, ranging from 1 in 6,000 to 1 in 19,000, with the highest prevalence in Taiwan.³

Biliary atresia usually presents within the first few weeks of life, with progressive cholestasis leading to failure to thrive and to fat-soluble vitamin deficiency. Approximately 20% of patients have congenital splenic, gastrointestinal, genitourinary, cardiac, and venous malformations.^4,5 Untreated, biliary atresia progresses to end-stage liver disease and death within 2 years.

The first-line treatment for biliary atresia is to establish biliary outflow with the Kasai procedure (hepatic portoenterostomy), in which a jejunal limb is anastomosed in a Roux-en-Y with the liver. The outcomes of the Kasai procedure depend on the timing of surgery, so timely diagnosis of biliary atresia is crucial. When the Kasai procedure is performed within 60 days of birth, biliary flow is achieved in up to 70% of patients; but if performed after 90 days, biliary flow is achieved in fewer than 25%.⁶

Long-term outcomes of biliary atresia in patients with their native liver have been reported in a few studies.

In a French study,⁷ 743 patients with biliary atresia underwent the Kasai procedure at a median age of 60 days. Survival rates were 57.1% at 2 years, 37.9% at 5 years, 32.4% at 10 years, and 28.5% at 15 years. In other studies,^4–9 the 20-year transplant-free survival rate ranged from 23% to 46%. Therefore, at least one-third of children with biliary atresia survive to adulthood with their native liver.

Implications of biliary atresia in adulthood

Although the Kasai procedure improves biliary outflow, up to 70% of patients develop complications of biliary atresia such as progressive fibrosis, cirrhosis, portal hypertension, cholangitis, and hepatocellular carcinoma, even after a successful Kasai procedure.¹⁰

Portal hypertension with evidence of splenomegaly, thrombocytopenia, or ascites is found in two-thirds of long-term survivors of biliary atresia with a native liver, with variceal hemorrhage occurring in 30%.¹¹ Therefore, patients with biliary atresia who have evidence of portal hypertension should be screened for varices with upper endoscopy on an annual basis. Management of variceal hemorrhage in these patients includes the use of octreotide, antibiotics, variceal ligation, and sclerotherapy; primary prophylaxis can be achieved with beta-blockers and endoscopic variceal ligation.¹²

Cholangitis is frequent, occurring in 40% to 60% of biliary atresia patients after the Kasai procedure, and about one-fourth of these patients have multiple episodes.¹³ The number of episodes of cholangitis negatively affects transplant-free survival.¹⁴ Patients with cholangitis should be adequately treated with oral or intravenous antibiotics depending on the severity of presentation. The role of prophylaxis with antibiotics remains unclear.¹⁵

Pulmonary complications such as hepato­pulmonary syndrome and portopulmonary hypertension can also occur in biliary atresia patients with a native liver. It is important for physicians to be aware of these complications and to screen for them, for example, with agitated saline echocardiography for hepatopulmonary syndrome and with echocardiography for portopulmonary hypertension. Timely screening is crucial, as the outcome of liver transplant depends on the severity at the time of transplant in these conditions, especially portopulmonary hypertension.

Hepatocellular carcinoma has been rarely reported in children with biliary atresia,¹⁶ so well-defined guidelines for screening in young adults with biliary atresia are lacking. Most centers recommend screening with ultrasonography of the abdomen and alpha-fetoprotein measurement every 6 months or annually starting soon after the Kasai procedure, since hepatocellular carcinoma has been reported in children as young as age 2.¹⁶

Transplant. Adult hepatologists are faced with the challenging task of deciding when it is time for transplant, balancing the long-term complications of biliary atresia with the risk of long-term immunosuppression after transplant. In addition, young adults with these complications may have preserved synthetic function, resulting in low Model for End-Stage Liver Disease (MELD) scores, which may complicate the process of listing for transplant.

Neurocognitive deficits are reported in children with biliary atresia,¹⁷ but young adults with biliary atresia generally have reasonable cognitive function and prospects for education and employment.

Pregnancy with successful outcomes has been reported.⁸

 

 

ALAGILLE SYNDROME

Alagille syndrome is an autosomal-dominant multisystemic disease caused by mutations in the JAG1 gene (accounting for > 95% of cases) and the NOTCH2 gene, with highly variable expression.¹⁸

Extrahepatic manifestations include butterfly vertebral defects, facial dysmorphism (eg, deep-set and low-set eyes, with characteristic “triangular” facies), posterior embryotoxon (a congenital defect of the eye characterized by an opaque ring around the margin of the cornea), peripheral pulmonary stenosis, renal abnormalities, and vascular malformations.

Hepatic manifestations vary from asymptomatic laboratory abnormalities to progressive cholestasis starting in early infancy with intractable pruritus, xanthomas, failure to thrive, and end-stage liver disease requiring liver transplant in childhood in 15% to 20% of patients.¹⁹

Implications of Alagille syndrome in adulthood

Transplant. Interestingly, the phenotype of hepatic disease is already established in childhood, with minimal or no progression in adulthood. Most children with minimal liver disease experience spontaneous resolution, whereas those with significant cholestasis might ultimately develop progressive liver fibrosis or cirrhosis requiring liver transplant in childhood. Only a small subset of children with minimal cholestasis progress to end-stage liver disease in late childhood or early adulthood.²⁰ Therefore, liver transplant for progressive liver disease from significant cholestasis almost always occurs in childhood, usually between ages 1 and 4.²¹

In a retrospective study comparing posttransplant outcomes in children with Alagille syndrome and biliary atresia, 1-year patient survival was excellent overall in children with Alagille syndrome, although slightly lower than in children with biliary atresia, most likely owing to extrahepatic morbidities of Alagille syndrome and especially the use of immunosuppression in those with renal disease.²¹ Similarly, 1- and 5-year patient and graft survival outcomes of liver transplant in adults with Alagille syndrome were also excellent compared with those who received a liver transplant in childhood for Alagille syndrome or in adulthood for biliary atresia.²²

Hepatocellular carcinoma has occurred in these patients in the absence of cirrhosis, which makes implementation of prognostic and surveillance strategies almost impossible to design for them. Annual ultrasonography with alpha-fetoprotein testing might be applicable in Alagille syndrome patients. However, deciding which patients should undergo this testing and when it should start will be challenging, given the paucity of data.

Cardiovascular disease. Cardiac phenotype is also mostly established in childhood, with the pulmonary vasculature being most commonly involved.¹⁹ In contrast, renal and other vascular abnormalities can manifest in adulthood. Renal manifestations vary and include structural anomalies such as hyperechoic kidneys or renal cysts, which can manifest in childhood, and some abnormalities such as hypertension and renal artery stenosis that can manifest in adulthood.^23,24

Vasculopathy is reported to involve the intracranial, renal, and intra-abdominal blood vessels.²⁵ Neurovascular accidents such as stroke and intracranial hemorrhage can occur at any age, with significant rates of morbidity and death.²⁶ Therefore, some experts recommend magnetic resonance angiography every 5 years and before any major intervention to prevent these devastating complications.²⁰

Pregnancy. Successful pregnancies have been reported. Preexisting cardiac and hepatic disease can complicate pregnancy depending on the severity of the disease. Because of the autosomal-dominant pattern of inheritance, infants have a 50% risk of the disease, so genetic counseling should be seriously considered before conception.²⁷ Prenatal diagnosis is possible, but the lack of genotype-phenotype correlation precludes its use in clinical practice.

PROGRESSIVE FAMILIAL INTRAHEPATIC CHOLESTASIS

Progressive familial intrahepatic cholestasis (PFIC) is a heterogeneous group of autosomal-recessive conditions associated with disruption of bile formation causing cholestatic liver disease in infants and young children. Three types have been described, depending on the genetic mutation in the hepatobiliary transport pathway:

• PFIC 1 (Byler disease) is caused by impaired bile salt secretion due to mutations in the ATP8B1 gene encoding for the familial intrahepatic cholestasis 1 (FIC 1) protein
• PFIC 2 is caused by impaired bile salt secretion due to mutations in the ABCB11 gene encoding for the bile salt export pump (BSEP) protein
• PFIC 3 is caused by impaired biliary phospholipid secretion due to a defect in ABCB4 encoding for multidrug resistance 3 (MDR3) protein.²⁸

PFIC 1 and 2 manifest with low gamma-glutamyl transferase (GGT) cholestasis, whereas PFIC 3 presents with high GGT cholestasis.

PFIC 1 and PFIC 2 usually cause cholestasis in early infancy, but PFIC 3 can cause cholestasis in late infancy, childhood, and even adulthood.

Because ATP8B1 is expressed in other tissues, PFIC 1 is characterized by extrahepatic manifestations such as sensorineural hearing loss, growth failure, severe diarrhea, and pancreatic insufficiency.

 

 

Implications of PFIC in adulthood

PFIC 1 and 2 (low-GGT cholestasis) are usually progressive and often lead to end-stage liver disease and cirrhosis before adulthood. Therefore, almost all patients with PFIC 1 and 2 undergo liver transplant or at least a biliary diversion procedure before reaching adulthood. Intractable pruritus is one of the most challenging clinical manifestations in patients with PFIC.

First-line management is pharmacologic and includes ursodeoxycholic acid, antihistamines (eg, hydroxyzine), bile acid sequestrants (eg, cholestyramine, colestipol), naltrexone, and rifampin, but these have limited efficacy.¹⁰

Most patients, especially those with PFIC 1 and 2, undergo a biliary diversion procedure such as partial external biliary diversion (cholecystojejunocutaneostomy), ileal exclusion, or partial internal biliary diversion (cholecystojejunocolic anastomosis) to decrease enterohepatic circulation of bile salts. The efficacy of these procedures is very limited in patients with established cirrhosis. Excessive losses of bile can occur through the biliary stoma, leading to dehydration in patients with external biliary diversion. In patients who are not candidates for biliary diversion, endoscopic nasobiliary drainage of pancreatobiliary secretions could be achieved by placing a catheter in the common bile duct; this has been reported to be effective in relieving cholestasis in a few cases.²⁹

Liver transplant is needed in patients with progressive liver disease and intractable pruritus despite medical management and biliary diversion. Unlike in biliary atresia, liver transplant is not curative in PFIC 1, due to extrahepatic manifestations: patients with PFIC 1 can still have intractable diarrhea and pancreatitis after liver transplant. More importantly, allograft steatohepatitis with further progression to cirrhosis can occur after liver transplant in patients with PFIC 1. Interestingly, biliary diversion has been reported to improve graft steatosis and diarrhea after liver transplant.³⁰

Although graft survival after transplant is good in patients with PFIC 2, recurrence of low-GGT cholestasis has been reported and is believed to be due to the formation of anti-bile salt export pump (anti-BSEP) antibodies by the host immune system in response to exposure to new proteins from the transplant graft.³¹

Cancer. The risk of malignancy, especially hepatocellular carcinoma, is also increased in PFIC 2, affecting nearly 15% of patients. Therefore, standard hepatocellular carcinoma surveillance with ultrasonography or alpha-fetoprotein testing or both is recommended in patients with PFIC 2. Cholangiocarcinoma and pancreatic adenocarcinoma have also been reported in patients with PFIC 2.²⁰

Incomplete penetrance of mutations in ATP8B1 and ABCB11 can cause recurrent episodes of cholestasis and pruritus with asymptomatic periods between episodes, referred to as benign recurrent intrahepatic cholestasis. Prognosis is usually good, with no progression to cirrhosis.³²

Pregnancy. In contrast to FIC 1 and BSEP deficiency, MDR3 defects lead to a wide phenotypic spectrum depending on the type of mutation. Heterozygous mutation is associated with increased risk of development of cholestasis during pregnancy, which typically presents with generalized pruritus in the third trimester and is associated with adverse fetal outcomes. Intrahepatic cholestasis of pregnancy is usually treated with ursodeoxycholic acid, with reported improvement in pruritus, liver function, and pregnancy outcomes.³³

In adults, drug-induced liver injury and idiopathic cirrhosis have also been described with MDR3 defects. Intrahepatic lithiasis and cholesterol gallstones can also occur with MDR3 defects as a result of impaired secretion of biliary phospholipid.³² Despite intrahepatic cholestasis of pregnancy, successful outcomes have been reported in women with PFIC.²⁰

OTHER CHILDHOOD-ONSET INHERITED CHOLESTATIC DISEASES

Cystic fibrosis-associated liver disease

Nearly 40% of patients with cystic fibrosis develop liver disease.³⁴ Cystic fibrosis-associated liver disease encompasses a broad clinical spectrum including asymptomatic elevation of aminotransferases, neonatal cholestasis, hepatic steatosis, focal biliary cirrhosis, and multilobar cirrhosis. Cirrhosis and portal hypertension can occur in 5% to 10% of patients and is the third-leading cause of death in patients with cystic fibrosis.³⁵

Risk factors for cystic fibrosis-associated liver disease include male sex, meconium ileus, and severe CFTR gene mutation (class I–III) with pancreatic insufficiency. Cystic fibrosis-related cirrhosis is more frequent in children and adolescents, whereas noncirrhotic portal hypertension and intrahepatic cholangiopathies are more common in adults.³⁶

Limited available studies support treatment with ursodeoxycholic acid in patients with cholestasis to delay the progression of liver disease, but the impact of this drug on long-term outcome is unknown.²⁹

Most patients remain in compensated cirrhosis for many years before progressing to decompensated cirrhosis requiring liver transplant. Other indications for liver transplant include recurrent intractable variceal bleeding, hepatopulmonary syndrome, and portopulmonary hypertension. Combined liver and lung transplant may be considered in patients with advanced liver and lung disease. Outcomes after isolated liver or liver-lung transplant in cystic fibrosis patients have been comparable to those in patients with other liver diseases.³⁷

Defects in bile acid synthesis

Inherited defects of enzymes required for the synthesis of primary bile acids from cholesterol can cause cholestasis from impaired bile flow and production of hepatotoxic aberrant bile acids. The clinical presentation varies depending on the enzymatic defect and can range from liver disease of varying severity to neurologic manifestations. Idiopathic late-onset cholestasis and cirrhosis of unknown etiology have been reported in adults with bile acid synthesis defects.^38,39 Therefore, this diagnosis should be considered in cases of cryptogenic cirrhosis and other cholestatic features.

Treatment with primary bile acids (cholic acid) has been effective in most patients with defective bile acid synthesis.

Primary sclerosing cholangitis

Primary sclerosing cholangitis is characterized by progressive obliteration of intrahepatic and extrahepatic bile ducts and is most commonly seen in patients with inflammatory bowel disease. Sclerosing cholangitis can also be secondary to other diseases in children such as immunodeficiency syndromes, Langerhans cell histiocytosis, cystic fibrosis, or sickle cell anemia.⁴⁰ Neonatal sclerosing cholangitis is a rare autosomal-recessive disease characterized by a severe form of cholangiopathy in neonates and young infants requiring transplant. It can be associated with Kabuki syndrome and neonatal ichthyosis-sclerosing cholangitis syndrome.

Treatment options are limited. Ursodeoxycholic acid and oral vancomycin have variable efficacy. Liver transplant is needed in patients with decompensated cirrhosis. Patients with primary sclerosing cholangitis, especially adults, are at higher risk of developing cholangiocarcinoma, and therefore screening with ultrasonography or magnetic resonance imaging every 6 to 12 months is recommended.

The risk of preterm and cesarean deliveries may be elevated in women with primary sclerosing cholangitis, though data are limited.³³

 

 

PEDIATRIC LIVER TRANSPLANT RECIPIENTS WHO SURVIVE INTO ADULTHOOD

Adolescent rebellion poses risks

Outcomes of liver transplant in children and adolescents have improved tremendously in the past 2 decades with advances in surgical techniques, pre- and postoperative management, organ preservation, and immunosuppression. Now, most pediatric liver transplant recipients survive into adulthood, creating a unique challenge for internists and adult  care hepatologists.⁴¹

In rebellious adolescents and young adults, risk-taking behavior, nonadherence to immunosuppressive medications, alcohol intake, and substance abuse increase the risk of graft rejection and loss. Current immunosuppressive drugs such as calcineurin inhibitors (tacrolimus, cyclosporine), mycophenolate mofetil, sirolimus, and corticosteroids have drastically decreased rejection rates in compliant patients.⁴¹ Educating patients on the importance of taking their medications and avoiding alcohol and drug abuse is especially important for adolescents and young adults, as rates of nonadherence are high in these age groups.

Although pregnancy is usually successful after liver transplant, it should be considered high-risk due to reported complications such as graft rejection, diabetes, preeclampsia, sepsis, prematurity, and low birth weight. Conception should be avoided for at least 1 year after transplant.⁴² Appropriate counseling with regard to pregnancy and contraception is important.

There is no consensus on breastfeeding, but it is considered safe in women on low-dose calcineurin inhibitors.⁴³

Life is better with a new liver, but patients have special needs

Liver transplant is life-saving and improves quality of life. However, long-term pediatric liver transplant recipients face challenges such as strict adherence to medications and follow-up visits, avoiding exposure to infections, and fear of graft rejection.

Chronic liver disease in children leads to failure to thrive, growth failure, and even delayed puberty, which resolve in most patients after liver transplant before adulthood in the absence of other comorbidities.⁴⁴ However, these patients are reported to have lower psychosocial functioning and more psychiatric disorders such as anxiety or posttraumatic disorder.^41,44

Therefore, a psychologist or other mental health professional should be part of the management team from the time of pretransplant assessment to identify mental health problems and the need for adjustments before liver transplant. Ongoing psychosocial assessment after liver transplant is equally important to identify risks such as drug or alcohol abuse, depression, posttraumatic stress disorder, and medication nonadherence, all of which can negatively affect posttransplant outcome.⁴⁵

In addition, assessment of family functioning and structure is important for good long-term outcomes posttransplant; therefore, a social worker should also be a part of the transplant team. Psyschosocial assessment tools can identify high-risk candidates who would benefit from earlier intervention to avoid any negative impact posttransplant.

Neurocognitive development can be delayed in children with chronic liver disease, and the delay may persist even after liver transplant, with reported impairments in intellectual ability, language, verbal, and visuospatial functioning skills.⁴¹ In spite of this, a recent study found that more than half the study patients were employed at a median follow-up of 24 years from liver transplant and a median age of 27.⁴⁶

Remarkably, pediatric liver transplant recipients have reported quality of life comparable to that in the general population,⁴⁷ and even better than in patients with other chronic illnesses.⁴⁸

Long-term medical comorbidities in pediatric liver transplant recipients

Favorable outcomes such as long-term survival and good quality of life in pediatric liver transplant recipients are lessened by late complications such as portal vein thrombosis or biliary strictures needing interventions, chronic graft rejection, adverse effects of immunosuppression, and recurrence of the disease.

Split-liver transplant—splitting a deceased-donor allograft to provide grafts for 2 recipients—has revolutionized liver transplant by increasing the donor pool and thereby decreasing waitlist mortality rates, especially in pediatric candidates. Despite this advantage, split-liver transplant is technically challenging and associated with increased perioperative complications compared with whole-liver transplant, especially in adult recipients. Recently, experienced centers have reported favorable outcomes with split-liver transplant comparable to those with whole-liver transplant; therefore, split-liver transplant should be considered after careful evaluation of donor organ and recipient clinical status.⁴⁹

Old age in the recipient can also adversely affect liver transplant outcomes.⁵⁰

Interestingly, even in patients whose clinical course is unremarkable and biochemical values are normal, graft hepatitis or fibrosis of unknown cause with progression to cirrhosis has been described in the decade after transplant.⁴¹

Chronic rejection with eventual graft loss may be related to nonadherence in adolescents and can be reduced with use of an additional immunosuppressant such as sirolimus or mycophenolate. Chronic kidney disease can occur in about one-third of liver transplant recipients secondary to renal disease associated with primary disease (like Alagille syndrome), hepatorenal syndrome, and most importantly, use of calcineurin inhibitors.⁴⁵

Components of the metabolic syndrome such as type 2 diabetes, obesity, nonalcoholic fatty liver disease, hypertension, and dyslipidemia are also seen in long-term pediatric liver transplant survivors. Internists are advised to screen for these comorbidities so that interventions can be applied early to improve long-term health outcomes and graft survival.

Of importance, multiple studies have shown a 2-fold increase in the rates of de novo malignancy in liver transplant recipients, including solid-organ and lymphoproliferative cancers, probably due to long-term immunosuppression. Posttransplant lymphoproliferative disorder occurs at lower rates than with other solid-organ transplants; its incidence is greatest in pediatric patients and in the first 12 to 18 months after transplant.⁵¹

 

 

TRANSITION TO ADULT CARE

While the number of patients with childhood-onset liver disease and pediatric liver transplant recipients who survive into adulthood is increasing, there are no established guidelines or formal models for transitioning these patients into adult care. Consequently, studies on transitional process have examined various issues such as patient and parent frustration, poor medical knowledge among patients during transition, lack of parental facilitation, and inadequate knowledge on disease process among adult-care hepatologists.^52–54

selvakumar_pediatriccholestaticliverdisease_f2.jpg

A position paper on transitional care from Europe has put forth important strategies.⁵⁵ A multidisciplinary team including a pediatric provider, nurse practitioner, transplant coordinator, psychologist, social worker, and adult provider is vital for a successful transition. Key requirements for an adolescent patient include knowledge about the disease, demonstration of responsibility toward disease management, and self-management of healthcare needs such as scheduling appointments or insurance needs. This demanding task could be streamlined if pediatric liver transplant programs could adopt a formal transitional model (Figure 2). The transition could start as early as age 10 to 12, with gradual transfer of health management responsibilities from parents to adolescents from age 13 to 17.⁵⁶

A prolonged period of transition up to age 25 is preferred in complicated cases. Distinctive consideration for transition should include those with neurocognitive developmental delay from underlying disease or hepatic encephalopathy before transplant. These patients need additional support and time to achieve independence in health management before transition.⁵⁷ Validated questionnaires are available to assess readiness to transition into adult care,⁵⁸ implying that the decision to transition should not be based solely on age. 

Thanks to advances in medical science and our understanding of inherited and acquired liver disease, many more children with acquired or congenital liver disease survive into adulthood than they did 2 decades ago. Improvements in immunosuppression and surgery have increased the chances of pediatric liver transplant recipients reaching adulthood, with a survival rate of 75% at 15 to 20 years.¹

selvakumar_pediatriccholestaticliverdisease_f1.jpg

The causes of cholestasis in children are different from those in adults (Figure 1). Pediatric cholestasis is caused either by genetic defects that affect the process of synthesis, processing, or secreting bile, or by mechanical obstruction of the biliary tract, the classic example being biliary atresia.

With the growing number of adult patients with pediatric-onset liver disease, internists and adult hepatologists need to be aware of these liver diseases and develop expertise to manage this challenging group of patients. Moreover, young adults with pediatric-onset chronic liver disease pose distinct challenges such as pregnancy, adherence to medical regimens, and psychosocial changes in life.

These patients need a “transition of care” rather than a “transfer of care.” Transition of care is a multifaceted process that takes the medical, educational, and psychosocial needs of the patient into consideration before switching their care to adult care physicians, whereas transfer of care is simply an administrative process of change to adult care without previous knowledge of the patients.²

selvakumar_pediatriccholestaticliverdisease_t1.jpg

In this article, we discuss relatively common types of inherited childhood-onset cholestatic disease and their implications and management in adulthood (Table 1). Other acquired or rare cholestatic diseases are beyond the scope of this article.

BILIARY ATRESIA

Biliary atresia is a progressive inflammatory fibrosclerosing cholangiopathy of unknown cause. Its prevalence varies with geographic location, ranging from 1 in 6,000 to 1 in 19,000, with the highest prevalence in Taiwan.³

Biliary atresia usually presents within the first few weeks of life, with progressive cholestasis leading to failure to thrive and to fat-soluble vitamin deficiency. Approximately 20% of patients have congenital splenic, gastrointestinal, genitourinary, cardiac, and venous malformations.^4,5 Untreated, biliary atresia progresses to end-stage liver disease and death within 2 years.

The first-line treatment for biliary atresia is to establish biliary outflow with the Kasai procedure (hepatic portoenterostomy), in which a jejunal limb is anastomosed in a Roux-en-Y with the liver. The outcomes of the Kasai procedure depend on the timing of surgery, so timely diagnosis of biliary atresia is crucial. When the Kasai procedure is performed within 60 days of birth, biliary flow is achieved in up to 70% of patients; but if performed after 90 days, biliary flow is achieved in fewer than 25%.⁶

Long-term outcomes of biliary atresia in patients with their native liver have been reported in a few studies.

In a French study,⁷ 743 patients with biliary atresia underwent the Kasai procedure at a median age of 60 days. Survival rates were 57.1% at 2 years, 37.9% at 5 years, 32.4% at 10 years, and 28.5% at 15 years. In other studies,^4–9 the 20-year transplant-free survival rate ranged from 23% to 46%. Therefore, at least one-third of children with biliary atresia survive to adulthood with their native liver.

Implications of biliary atresia in adulthood

Although the Kasai procedure improves biliary outflow, up to 70% of patients develop complications of biliary atresia such as progressive fibrosis, cirrhosis, portal hypertension, cholangitis, and hepatocellular carcinoma, even after a successful Kasai procedure.¹⁰

Portal hypertension with evidence of splenomegaly, thrombocytopenia, or ascites is found in two-thirds of long-term survivors of biliary atresia with a native liver, with variceal hemorrhage occurring in 30%.¹¹ Therefore, patients with biliary atresia who have evidence of portal hypertension should be screened for varices with upper endoscopy on an annual basis. Management of variceal hemorrhage in these patients includes the use of octreotide, antibiotics, variceal ligation, and sclerotherapy; primary prophylaxis can be achieved with beta-blockers and endoscopic variceal ligation.¹²

Cholangitis is frequent, occurring in 40% to 60% of biliary atresia patients after the Kasai procedure, and about one-fourth of these patients have multiple episodes.¹³ The number of episodes of cholangitis negatively affects transplant-free survival.¹⁴ Patients with cholangitis should be adequately treated with oral or intravenous antibiotics depending on the severity of presentation. The role of prophylaxis with antibiotics remains unclear.¹⁵

Pulmonary complications such as hepato­pulmonary syndrome and portopulmonary hypertension can also occur in biliary atresia patients with a native liver. It is important for physicians to be aware of these complications and to screen for them, for example, with agitated saline echocardiography for hepatopulmonary syndrome and with echocardiography for portopulmonary hypertension. Timely screening is crucial, as the outcome of liver transplant depends on the severity at the time of transplant in these conditions, especially portopulmonary hypertension.

Hepatocellular carcinoma has been rarely reported in children with biliary atresia,¹⁶ so well-defined guidelines for screening in young adults with biliary atresia are lacking. Most centers recommend screening with ultrasonography of the abdomen and alpha-fetoprotein measurement every 6 months or annually starting soon after the Kasai procedure, since hepatocellular carcinoma has been reported in children as young as age 2.¹⁶

Transplant. Adult hepatologists are faced with the challenging task of deciding when it is time for transplant, balancing the long-term complications of biliary atresia with the risk of long-term immunosuppression after transplant. In addition, young adults with these complications may have preserved synthetic function, resulting in low Model for End-Stage Liver Disease (MELD) scores, which may complicate the process of listing for transplant.

Neurocognitive deficits are reported in children with biliary atresia,¹⁷ but young adults with biliary atresia generally have reasonable cognitive function and prospects for education and employment.

Pregnancy with successful outcomes has been reported.⁸

 

 

ALAGILLE SYNDROME

Alagille syndrome is an autosomal-dominant multisystemic disease caused by mutations in the JAG1 gene (accounting for > 95% of cases) and the NOTCH2 gene, with highly variable expression.¹⁸

Extrahepatic manifestations include butterfly vertebral defects, facial dysmorphism (eg, deep-set and low-set eyes, with characteristic “triangular” facies), posterior embryotoxon (a congenital defect of the eye characterized by an opaque ring around the margin of the cornea), peripheral pulmonary stenosis, renal abnormalities, and vascular malformations.

Hepatic manifestations vary from asymptomatic laboratory abnormalities to progressive cholestasis starting in early infancy with intractable pruritus, xanthomas, failure to thrive, and end-stage liver disease requiring liver transplant in childhood in 15% to 20% of patients.¹⁹

Implications of Alagille syndrome in adulthood

Transplant. Interestingly, the phenotype of hepatic disease is already established in childhood, with minimal or no progression in adulthood. Most children with minimal liver disease experience spontaneous resolution, whereas those with significant cholestasis might ultimately develop progressive liver fibrosis or cirrhosis requiring liver transplant in childhood. Only a small subset of children with minimal cholestasis progress to end-stage liver disease in late childhood or early adulthood.²⁰ Therefore, liver transplant for progressive liver disease from significant cholestasis almost always occurs in childhood, usually between ages 1 and 4.²¹

In a retrospective study comparing posttransplant outcomes in children with Alagille syndrome and biliary atresia, 1-year patient survival was excellent overall in children with Alagille syndrome, although slightly lower than in children with biliary atresia, most likely owing to extrahepatic morbidities of Alagille syndrome and especially the use of immunosuppression in those with renal disease.²¹ Similarly, 1- and 5-year patient and graft survival outcomes of liver transplant in adults with Alagille syndrome were also excellent compared with those who received a liver transplant in childhood for Alagille syndrome or in adulthood for biliary atresia.²²

Hepatocellular carcinoma has occurred in these patients in the absence of cirrhosis, which makes implementation of prognostic and surveillance strategies almost impossible to design for them. Annual ultrasonography with alpha-fetoprotein testing might be applicable in Alagille syndrome patients. However, deciding which patients should undergo this testing and when it should start will be challenging, given the paucity of data.

Cardiovascular disease. Cardiac phenotype is also mostly established in childhood, with the pulmonary vasculature being most commonly involved.¹⁹ In contrast, renal and other vascular abnormalities can manifest in adulthood. Renal manifestations vary and include structural anomalies such as hyperechoic kidneys or renal cysts, which can manifest in childhood, and some abnormalities such as hypertension and renal artery stenosis that can manifest in adulthood.^23,24

Vasculopathy is reported to involve the intracranial, renal, and intra-abdominal blood vessels.²⁵ Neurovascular accidents such as stroke and intracranial hemorrhage can occur at any age, with significant rates of morbidity and death.²⁶ Therefore, some experts recommend magnetic resonance angiography every 5 years and before any major intervention to prevent these devastating complications.²⁰

Pregnancy. Successful pregnancies have been reported. Preexisting cardiac and hepatic disease can complicate pregnancy depending on the severity of the disease. Because of the autosomal-dominant pattern of inheritance, infants have a 50% risk of the disease, so genetic counseling should be seriously considered before conception.²⁷ Prenatal diagnosis is possible, but the lack of genotype-phenotype correlation precludes its use in clinical practice.

PROGRESSIVE FAMILIAL INTRAHEPATIC CHOLESTASIS

Progressive familial intrahepatic cholestasis (PFIC) is a heterogeneous group of autosomal-recessive conditions associated with disruption of bile formation causing cholestatic liver disease in infants and young children. Three types have been described, depending on the genetic mutation in the hepatobiliary transport pathway:

• PFIC 1 (Byler disease) is caused by impaired bile salt secretion due to mutations in the ATP8B1 gene encoding for the familial intrahepatic cholestasis 1 (FIC 1) protein
• PFIC 2 is caused by impaired bile salt secretion due to mutations in the ABCB11 gene encoding for the bile salt export pump (BSEP) protein
• PFIC 3 is caused by impaired biliary phospholipid secretion due to a defect in ABCB4 encoding for multidrug resistance 3 (MDR3) protein.²⁸

PFIC 1 and 2 manifest with low gamma-glutamyl transferase (GGT) cholestasis, whereas PFIC 3 presents with high GGT cholestasis.

PFIC 1 and PFIC 2 usually cause cholestasis in early infancy, but PFIC 3 can cause cholestasis in late infancy, childhood, and even adulthood.

Because ATP8B1 is expressed in other tissues, PFIC 1 is characterized by extrahepatic manifestations such as sensorineural hearing loss, growth failure, severe diarrhea, and pancreatic insufficiency.

 

 

Implications of PFIC in adulthood

PFIC 1 and 2 (low-GGT cholestasis) are usually progressive and often lead to end-stage liver disease and cirrhosis before adulthood. Therefore, almost all patients with PFIC 1 and 2 undergo liver transplant or at least a biliary diversion procedure before reaching adulthood. Intractable pruritus is one of the most challenging clinical manifestations in patients with PFIC.

First-line management is pharmacologic and includes ursodeoxycholic acid, antihistamines (eg, hydroxyzine), bile acid sequestrants (eg, cholestyramine, colestipol), naltrexone, and rifampin, but these have limited efficacy.¹⁰

Most patients, especially those with PFIC 1 and 2, undergo a biliary diversion procedure such as partial external biliary diversion (cholecystojejunocutaneostomy), ileal exclusion, or partial internal biliary diversion (cholecystojejunocolic anastomosis) to decrease enterohepatic circulation of bile salts. The efficacy of these procedures is very limited in patients with established cirrhosis. Excessive losses of bile can occur through the biliary stoma, leading to dehydration in patients with external biliary diversion. In patients who are not candidates for biliary diversion, endoscopic nasobiliary drainage of pancreatobiliary secretions could be achieved by placing a catheter in the common bile duct; this has been reported to be effective in relieving cholestasis in a few cases.²⁹

Liver transplant is needed in patients with progressive liver disease and intractable pruritus despite medical management and biliary diversion. Unlike in biliary atresia, liver transplant is not curative in PFIC 1, due to extrahepatic manifestations: patients with PFIC 1 can still have intractable diarrhea and pancreatitis after liver transplant. More importantly, allograft steatohepatitis with further progression to cirrhosis can occur after liver transplant in patients with PFIC 1. Interestingly, biliary diversion has been reported to improve graft steatosis and diarrhea after liver transplant.³⁰

Although graft survival after transplant is good in patients with PFIC 2, recurrence of low-GGT cholestasis has been reported and is believed to be due to the formation of anti-bile salt export pump (anti-BSEP) antibodies by the host immune system in response to exposure to new proteins from the transplant graft.³¹

Cancer. The risk of malignancy, especially hepatocellular carcinoma, is also increased in PFIC 2, affecting nearly 15% of patients. Therefore, standard hepatocellular carcinoma surveillance with ultrasonography or alpha-fetoprotein testing or both is recommended in patients with PFIC 2. Cholangiocarcinoma and pancreatic adenocarcinoma have also been reported in patients with PFIC 2.²⁰

Incomplete penetrance of mutations in ATP8B1 and ABCB11 can cause recurrent episodes of cholestasis and pruritus with asymptomatic periods between episodes, referred to as benign recurrent intrahepatic cholestasis. Prognosis is usually good, with no progression to cirrhosis.³²

Pregnancy. In contrast to FIC 1 and BSEP deficiency, MDR3 defects lead to a wide phenotypic spectrum depending on the type of mutation. Heterozygous mutation is associated with increased risk of development of cholestasis during pregnancy, which typically presents with generalized pruritus in the third trimester and is associated with adverse fetal outcomes. Intrahepatic cholestasis of pregnancy is usually treated with ursodeoxycholic acid, with reported improvement in pruritus, liver function, and pregnancy outcomes.³³

In adults, drug-induced liver injury and idiopathic cirrhosis have also been described with MDR3 defects. Intrahepatic lithiasis and cholesterol gallstones can also occur with MDR3 defects as a result of impaired secretion of biliary phospholipid.³² Despite intrahepatic cholestasis of pregnancy, successful outcomes have been reported in women with PFIC.²⁰

OTHER CHILDHOOD-ONSET INHERITED CHOLESTATIC DISEASES

Cystic fibrosis-associated liver disease

Nearly 40% of patients with cystic fibrosis develop liver disease.³⁴ Cystic fibrosis-associated liver disease encompasses a broad clinical spectrum including asymptomatic elevation of aminotransferases, neonatal cholestasis, hepatic steatosis, focal biliary cirrhosis, and multilobar cirrhosis. Cirrhosis and portal hypertension can occur in 5% to 10% of patients and is the third-leading cause of death in patients with cystic fibrosis.³⁵

Risk factors for cystic fibrosis-associated liver disease include male sex, meconium ileus, and severe CFTR gene mutation (class I–III) with pancreatic insufficiency. Cystic fibrosis-related cirrhosis is more frequent in children and adolescents, whereas noncirrhotic portal hypertension and intrahepatic cholangiopathies are more common in adults.³⁶

Limited available studies support treatment with ursodeoxycholic acid in patients with cholestasis to delay the progression of liver disease, but the impact of this drug on long-term outcome is unknown.²⁹

Most patients remain in compensated cirrhosis for many years before progressing to decompensated cirrhosis requiring liver transplant. Other indications for liver transplant include recurrent intractable variceal bleeding, hepatopulmonary syndrome, and portopulmonary hypertension. Combined liver and lung transplant may be considered in patients with advanced liver and lung disease. Outcomes after isolated liver or liver-lung transplant in cystic fibrosis patients have been comparable to those in patients with other liver diseases.³⁷

Defects in bile acid synthesis

Inherited defects of enzymes required for the synthesis of primary bile acids from cholesterol can cause cholestasis from impaired bile flow and production of hepatotoxic aberrant bile acids. The clinical presentation varies depending on the enzymatic defect and can range from liver disease of varying severity to neurologic manifestations. Idiopathic late-onset cholestasis and cirrhosis of unknown etiology have been reported in adults with bile acid synthesis defects.^38,39 Therefore, this diagnosis should be considered in cases of cryptogenic cirrhosis and other cholestatic features.

Treatment with primary bile acids (cholic acid) has been effective in most patients with defective bile acid synthesis.

Primary sclerosing cholangitis

Primary sclerosing cholangitis is characterized by progressive obliteration of intrahepatic and extrahepatic bile ducts and is most commonly seen in patients with inflammatory bowel disease. Sclerosing cholangitis can also be secondary to other diseases in children such as immunodeficiency syndromes, Langerhans cell histiocytosis, cystic fibrosis, or sickle cell anemia.⁴⁰ Neonatal sclerosing cholangitis is a rare autosomal-recessive disease characterized by a severe form of cholangiopathy in neonates and young infants requiring transplant. It can be associated with Kabuki syndrome and neonatal ichthyosis-sclerosing cholangitis syndrome.

Treatment options are limited. Ursodeoxycholic acid and oral vancomycin have variable efficacy. Liver transplant is needed in patients with decompensated cirrhosis. Patients with primary sclerosing cholangitis, especially adults, are at higher risk of developing cholangiocarcinoma, and therefore screening with ultrasonography or magnetic resonance imaging every 6 to 12 months is recommended.

The risk of preterm and cesarean deliveries may be elevated in women with primary sclerosing cholangitis, though data are limited.³³

 

 

PEDIATRIC LIVER TRANSPLANT RECIPIENTS WHO SURVIVE INTO ADULTHOOD

Adolescent rebellion poses risks

Outcomes of liver transplant in children and adolescents have improved tremendously in the past 2 decades with advances in surgical techniques, pre- and postoperative management, organ preservation, and immunosuppression. Now, most pediatric liver transplant recipients survive into adulthood, creating a unique challenge for internists and adult  care hepatologists.⁴¹

In rebellious adolescents and young adults, risk-taking behavior, nonadherence to immunosuppressive medications, alcohol intake, and substance abuse increase the risk of graft rejection and loss. Current immunosuppressive drugs such as calcineurin inhibitors (tacrolimus, cyclosporine), mycophenolate mofetil, sirolimus, and corticosteroids have drastically decreased rejection rates in compliant patients.⁴¹ Educating patients on the importance of taking their medications and avoiding alcohol and drug abuse is especially important for adolescents and young adults, as rates of nonadherence are high in these age groups.

Although pregnancy is usually successful after liver transplant, it should be considered high-risk due to reported complications such as graft rejection, diabetes, preeclampsia, sepsis, prematurity, and low birth weight. Conception should be avoided for at least 1 year after transplant.⁴² Appropriate counseling with regard to pregnancy and contraception is important.

There is no consensus on breastfeeding, but it is considered safe in women on low-dose calcineurin inhibitors.⁴³

Life is better with a new liver, but patients have special needs

Liver transplant is life-saving and improves quality of life. However, long-term pediatric liver transplant recipients face challenges such as strict adherence to medications and follow-up visits, avoiding exposure to infections, and fear of graft rejection.

Chronic liver disease in children leads to failure to thrive, growth failure, and even delayed puberty, which resolve in most patients after liver transplant before adulthood in the absence of other comorbidities.⁴⁴ However, these patients are reported to have lower psychosocial functioning and more psychiatric disorders such as anxiety or posttraumatic disorder.^41,44

Therefore, a psychologist or other mental health professional should be part of the management team from the time of pretransplant assessment to identify mental health problems and the need for adjustments before liver transplant. Ongoing psychosocial assessment after liver transplant is equally important to identify risks such as drug or alcohol abuse, depression, posttraumatic stress disorder, and medication nonadherence, all of which can negatively affect posttransplant outcome.⁴⁵

In addition, assessment of family functioning and structure is important for good long-term outcomes posttransplant; therefore, a social worker should also be a part of the transplant team. Psyschosocial assessment tools can identify high-risk candidates who would benefit from earlier intervention to avoid any negative impact posttransplant.

Neurocognitive development can be delayed in children with chronic liver disease, and the delay may persist even after liver transplant, with reported impairments in intellectual ability, language, verbal, and visuospatial functioning skills.⁴¹ In spite of this, a recent study found that more than half the study patients were employed at a median follow-up of 24 years from liver transplant and a median age of 27.⁴⁶

Remarkably, pediatric liver transplant recipients have reported quality of life comparable to that in the general population,⁴⁷ and even better than in patients with other chronic illnesses.⁴⁸

Long-term medical comorbidities in pediatric liver transplant recipients

Favorable outcomes such as long-term survival and good quality of life in pediatric liver transplant recipients are lessened by late complications such as portal vein thrombosis or biliary strictures needing interventions, chronic graft rejection, adverse effects of immunosuppression, and recurrence of the disease.

Split-liver transplant—splitting a deceased-donor allograft to provide grafts for 2 recipients—has revolutionized liver transplant by increasing the donor pool and thereby decreasing waitlist mortality rates, especially in pediatric candidates. Despite this advantage, split-liver transplant is technically challenging and associated with increased perioperative complications compared with whole-liver transplant, especially in adult recipients. Recently, experienced centers have reported favorable outcomes with split-liver transplant comparable to those with whole-liver transplant; therefore, split-liver transplant should be considered after careful evaluation of donor organ and recipient clinical status.⁴⁹

Old age in the recipient can also adversely affect liver transplant outcomes.⁵⁰

Interestingly, even in patients whose clinical course is unremarkable and biochemical values are normal, graft hepatitis or fibrosis of unknown cause with progression to cirrhosis has been described in the decade after transplant.⁴¹

Chronic rejection with eventual graft loss may be related to nonadherence in adolescents and can be reduced with use of an additional immunosuppressant such as sirolimus or mycophenolate. Chronic kidney disease can occur in about one-third of liver transplant recipients secondary to renal disease associated with primary disease (like Alagille syndrome), hepatorenal syndrome, and most importantly, use of calcineurin inhibitors.⁴⁵

Components of the metabolic syndrome such as type 2 diabetes, obesity, nonalcoholic fatty liver disease, hypertension, and dyslipidemia are also seen in long-term pediatric liver transplant survivors. Internists are advised to screen for these comorbidities so that interventions can be applied early to improve long-term health outcomes and graft survival.

Of importance, multiple studies have shown a 2-fold increase in the rates of de novo malignancy in liver transplant recipients, including solid-organ and lymphoproliferative cancers, probably due to long-term immunosuppression. Posttransplant lymphoproliferative disorder occurs at lower rates than with other solid-organ transplants; its incidence is greatest in pediatric patients and in the first 12 to 18 months after transplant.⁵¹

 

 

TRANSITION TO ADULT CARE

While the number of patients with childhood-onset liver disease and pediatric liver transplant recipients who survive into adulthood is increasing, there are no established guidelines or formal models for transitioning these patients into adult care. Consequently, studies on transitional process have examined various issues such as patient and parent frustration, poor medical knowledge among patients during transition, lack of parental facilitation, and inadequate knowledge on disease process among adult-care hepatologists.^52–54

selvakumar_pediatriccholestaticliverdisease_f2.jpg

A position paper on transitional care from Europe has put forth important strategies.⁵⁵ A multidisciplinary team including a pediatric provider, nurse practitioner, transplant coordinator, psychologist, social worker, and adult provider is vital for a successful transition. Key requirements for an adolescent patient include knowledge about the disease, demonstration of responsibility toward disease management, and self-management of healthcare needs such as scheduling appointments or insurance needs. This demanding task could be streamlined if pediatric liver transplant programs could adopt a formal transitional model (Figure 2). The transition could start as early as age 10 to 12, with gradual transfer of health management responsibilities from parents to adolescents from age 13 to 17.⁵⁶

A prolonged period of transition up to age 25 is preferred in complicated cases. Distinctive consideration for transition should include those with neurocognitive developmental delay from underlying disease or hepatic encephalopathy before transplant. These patients need additional support and time to achieve independence in health management before transition.⁵⁷ Validated questionnaires are available to assess readiness to transition into adult care,⁵⁸ implying that the decision to transition should not be based solely on age. 

Giant cell arteritis (GCA) is a systemic vasculitis involving medium-sized and large arteries, most commonly the temporal, ophthalmic, occipital, vertebral, posterior ciliary, and proximal vertebral arteries. Moreover, involvement of the ophthalmic artery and its branches results in loss of vision. GCA can also involve the aorta and its proximal branches, especially in the upper extremities.

GCA is the most common systemic vasculitis in adults. It occurs almost exclusively in patients over age 50 and affects women more than men. It is most frequent in populations of northern European ancestry, especially Scandinavian. In a retrospective cohort study in Norway, the average annual cumulative incidence rate of GCA was 16.7 per 100,000 people over age 50.¹ Risk factors include older age, history of smoking, current smoking, early menopause, and, possibly, stress-related disorders.²

PATHOGENESIS IS NOT COMPLETELY UNDERSTOOD

The pathogenesis of GCA is not completely understood, but there is evidence of immune activation in the arterial wall leading to activation of macrophages and formation of multinucleated giant cells (which may not always be present in biopsies).

The most relevant cytokines in the ongoing pathogenesis are still being defined, but the presence of interferon gamma and interleukin 6 (IL-6) seem to be critical for the expression of the disease. The primary immunogenic triggers for the elaboration of these cytokines and the arteritis remain elusive.

A SPECTRUM OF PRESENTATIONS

The initial symptoms of GCA may be vague, such as malaise, fever, and night sweats, and are likely due to systemic inflammation. Features of vascular involvement include headache, scalp tenderness, and jaw claudication (cramping pain in the jaw while chewing).

A less common but serious feature associated with GCA is partial or complete vision loss affecting 1 or both eyes.³ Some patients suddenly go completely blind without any visual prodrome.

Overlapping GCA phenotypes exist, with a spectrum of presentations that include classic cranial arteritis, extracranial GCA (also called large-vessel GCA), and polymyalgia rheumatica.²

Cranial GCA, the best-characterized clinical presentation, causes symptoms such as headache or signs such as tenderness of the temporal artery. On examination, the temporal arteries may be tender or nodular, and the pulses may be felt above the zygomatic arch, above and in front of the tragus of the ear. About two-thirds of patients with cranial GCA present with new-onset headache, most often in the temporal area, but possibly anywhere throughout the head.

Visual disturbance, jaw claudication, and tongue pain are less common but, if present, increase the likelihood of this diagnosis.²

Large-vessel involvement in GCA is common and refers to involvement of the aorta and its proximal branches. Imaging methods used in diagnosing large-vessel GCA include color Doppler ultrasonography, computed tomography with angiography, magnetic resonance imaging with angiography, and positron emission tomography. In some centers, such imaging is performed in all patients diagnosed with GCA to survey for large-vessel involvement.

Depending on the imaging study, large-vessel involvement has been found in 30% to 80% of cases of GCA.^4,5 It is often associated with nonspecific symptoms such as fever, weight loss, chills, and malaise, but it can also cause more specific symptoms such as unilateral extremity claudication. In contrast to patients with cranial GCA, patients with large-vessel GCA were younger at onset, less likely to have headaches, and more likely to have arm claudication at presentation.⁶ Aortitis of the ascending aorta can occur with a histopathologic pattern of GCA but without the clinical stigmata of GCA.

The finding of aortitis should prompt the clinician to question the patient about other symptoms of GCA and to order imaging of the whole vascular tree. Ultrasonography and biopsy of the temporal arteries can be considered. Whether idiopathic aortitis is part of the GCA spectrum remains to be seen.

Laboratory tests often show anemia, leukocytosis, and thrombocytosis. Acute-phase reactants such as C-reactive protein and the erythrocyte sedimentation rate are often elevated. The sedimentation rate often exceeds 50 mm/hour and sometimes 100 mm/hour.

In 2 retrospective studies, the number of patients with GCA whose sedimentation rate was less than 50 mm/hour ranged between 5% and 11%.^7,8 However, a small percentage of patients with GCA have normal inflammatory markers. Therefore, if the suspicion for GCA is high, treatment should be started and biopsy pursued.⁹ In patients with paraproteinemia or other causes of a spuriously elevated or low erythrocyte sedimentation rate, C-reactive protein is a more reliable test.

Polymyalgia rheumatica is another rheumatologic condition that can occur independently or in conjunction with GCA. It is characterized by stiffness and pain in the proximal joints such as the hips and shoulders, typically worse in the morning and better with activity. Although the patient may subjectively feel weak, a close neurologic examination will reveal normal muscle strength.

Polymyalgia rheumatica is observed in 40% to 60% of patients with GCA at the time of diagnosis; 16% to 21% of patients with polymyalgia rheumatica may develop GCA, especially if untreated.^2,10

Differential diagnosis

Other vasculitides (eg, Takayasu arteritis) can also present with unexplained fever, anemia, and constitutional symptoms.

Infection should be considered if fever is present. An infectious disease accompanied by fever, headache, and elevated inflammatory markers can mimic GCA.

Nonarteritic anterior ischemic optic neuropathy can present with sudden vision loss, prompting concern for underlying GCA. Risk factors include hypertension and diabetes mellitus; other features of GCA, including elevated inflammatory markers, are generally absent.

 

 

TEMPORAL ARTERY BIOPSY: THE GOLD STANDARD FOR DIAGNOSIS

Temporal artery biopsy remains the standard to confirm the diagnosis. However, because inflammation in the temporal arteries can affect some segments but not others, biopsy results on conventional hematoxylin and eosin staining can be falsely negative in patients with GCA. In one study,¹¹ the mean sensitivity of unilateral temporal artery biopsy was 86.9%.

Typical positive histologic findings are inflammation with panarteritis, CD4-positive lymphocytes, macrophages, giant cells, and fragmentation of the internal elastic lamina.¹²

When GCA is suspected, treatment with glucocorticoids should be started immediately and biopsy performed as soon as possible. Delaying biopsy for 14 days or more may not affect the accuracy of biopsy study.¹³ Treatment should never be withheld while awaiting the results of biopsy study.

Biopsy is usually performed unilaterally, on the same side as the symptoms or abnormal findings on examination. Bilateral temporal artery biopsy is also performed and compared with unilateral biopsy; this approach increases the diagnostic yield by about 5%.¹⁴

IMAGING

In patients with suspected GCA, imaging is recommended early to complement the clinical criteria for the diagnosis of GCA.¹⁵ Positron emission tomography, computed tomography angiography, magnetic resonance angiography, or Doppler ultrasonography can reveal inflammation of the arteries in the proximal upper or lower limbs or the aorta.²

In patients with suspected cranial GCA, ultrasonography of the temporal and axillary arteries is recommended first. If ultrasonography is not available or is inconclusive, high-resolution magnetic resonance imaging of the cranial arteries can be used as an alternative. Computed tomography and positron emission tomography of the cranial arteries are not recommended.

In patients with suspected large-vessel GCA, ultrasonography, positron emission tomography, computed tomography, and magnetic resonance imaging may be used to screen for vessel wall inflammation, edema, and luminal narrowing in extracranial arteries. Ultrasonography is of limited value in assessing aortitis.

Color duplex ultrasonography can be applied to assess for vascular inflammation of the temporal or large arteries. The typical finding of the “halo” sign, a hypoechoic ring around the arterial lumen, represents the inflammation-induced thickening of the arterial wall. The “compression sign,” the persistence of the “halo” during compression of the vessel lumen by the ultrasound probe, has high specificity for the diagnosis.¹⁶

Ultrasonography of suspected GCA has yielded sensitivities of 55% to 100% and specificities of 78% to 100%. However, its sensitivity depends on the user’s level of expertise, so it should be done only in medical centers with a high number of GCA cases and with highly experienced sonographers. High-resolution magnetic resonance imaging is an alternative to ultrasonography and has shown similar sensitivity and specificity.³

TREATMENT WITH GLUCOCORTICOIDS

Glucocorticoids remain the standard for treatment of GCA. The therapeutic effect of glucocorticoids in GCA has been established by years of clinical experience, but has never been proven in a placebo-controlled trial. When started appropriately and expeditiously, glucocorticoids produce exquisite resolution of signs and symptoms and prevent the serious complication of vision loss. Rapid resolution of symptoms is so typical of GCA that if the patient’s symptoms persist more than a few days after starting a glucocorticoid, the diagnosis of GCA should be reconsidered.

In a retrospective study of 245 patients with biopsy-proven GCA treated with glucocorticoids, 34 had permanent loss of sight.¹⁷ In 32 (94%) of the 34, the vision loss occurred before glucocorticoids were started. Of the remaining 2 patients, 1 lost vision 8 days into treatment, and the other lost vision 3 years after diagnosis and 1 year after discontinuation of glucocorticoids.

In a series of 144 patients with biopsy-proven GCA, 51 had no vision loss at presentation and no vision loss after starting glucocorticoids, and 93 had vision loss at presentation. In the latter group, symptoms worsened within 5 days of starting glucocorticoids in 9 patients.¹⁸ If vision was intact at the time of presentation, prompt initiation of glucocorticoids reduced the risk of vision loss to less than 1%.

High doses, slowly tapered

The European League Against Rheumatism recommends early initiation of high-dose glucocorticoids for patients with large-vessel vasculitis,¹⁹ and it also recommends glucocorticoids for patients with polymyalgia rheumatica.²⁰ The optimal initial and tapering dosage has never been formally evaluated, but regimens have been devised on the basis of expert opinion.²¹

For patients with GCA who do not have vision loss at the time of diagnosis, the initial dose is prednisone 1 mg/kg or its equivalent daily for 2 to 4 weeks, after which it is tapered.²¹ If the initial dosage is prednisone 60 mg orally daily for 2 to 4 weeks, our practice is to taper it to 50 mg daily for 2 weeks, then 40 mg daily for 2 weeks. Then, it  is decreased by 5 mg every 2 weeks until it is 20 mg daily, and then by 2.5 mg every 2 weeks until it is 10 mg orally daily. Thereafter, the dosage is decreased by 1 mg every 2 to 4 weeks.

For patients with GCA who experience transient vision loss or diplopia at the time of diagnosis, intravenous pulse glucocorticoid therapy should be initiated to reduce the risk of vision loss as rapidly as possible.²² A typical pulse regimen is methylprednisolone 1 g intravenously daily for 3 days. Though not rigorously validated in studies, such an approach is used to avoid vision impairment due to GCA, which is rarely reversible.

 

 

RELAPSE OF DISEASE

Suspect a relapse of GCA if the patient’s initial symptoms recur, if inflammatory markers become elevated, or if classic symptoms of GCA or polymyalgia rheumatica occur. Elevations in inflammatory markers do not definitely indicate a flare of GCA, but they should trigger close monitoring of the patient’s symptoms.

Relapse is treated by increasing the glucocorticoid dosage as appropriate to the nature of the relapse. If vision is affected or the patient has symptoms of GCA, then increments of 30 to 60 mg of prednisone are warranted, whereas if the patient has symptoms of polymyalgia rheumatica, then increments of 5 to 10 mg of prednisone are usually used.

The incidence of relapses of GCA in multiple tertiary care centers has been reported to vary between 34% and 75%.^23,24 Most relapses occur at prednisone dosages of less than 20 mg orally daily and within the first year after diagnosis. The most common symptoms are limb ischemia, jaw claudication, constitutional symptoms, headaches, and polymyalgia rheumatica. In a review of 286 patients,²⁵ 213 (74%) had at least 1 relapse. The first relapse occurred in the first year in 50%, by 2 years in 68%, and by 5 years in 79%.

ADVERSE EFFECTS OF GLUCOCORTICOIDS

In high doses, glucocorticoids have well-known adverse effects. In a population-based study of 120 patients, each patient treated with glucocorticoids experienced at least 1 adverse effect (cataract, fracture, infection, osteonecrosis, diabetes, hypertension, weight gain, capillary fragility, or hair loss).²⁶ The effects were related to aging and cumulative dosage of prednisone but not to the initial dosage.

Glucocorticoids can affect many organs and systems:

• Eyes (cataracts, increased intraocular pressure, exophthalmos)
• Heart (premature atherosclerotic disease, hypertension, fluid retention, hyperlipidemia, arrhythmias)
• Gastrointestinal system (ulcer, gastrointestinal bleeding, gastritis, visceral perforation, hepatic steatosis, acute pancreatitis)
• Bone and muscle (osteopenia, osteoporosis, osteonecrosis, myopathy)
• Brain (mood disorder, psychosis, memory impairment)
• Endocrine system (hyperglycemia, hypothalamic-pituitary-adrenal axis suppression)
• Immune system (immunosuppression, leading to infection and leukocytosis).

Patients receiving a glucocorticoid dose equivalent to 20 mg or more of prednisone daily for 1 month or more who also have another cause of immunocompromise need prophylaxis against Pneumocystis jirovecii pneumonia.²⁷ They should also receive appropriate immunizations before starting glucocorticoids. Live-virus vaccines should not be given to these patients until they have been off glucocorticoids for 1 month.

Glucocorticoids and bone loss

Glucocorticoids are associated with bone loss and fracture, which can occur within the first few months of use and with dosages as low as 2.5 to 7.5 mg orally daily.²⁸ Therefore, glucocorticoid-induced bone loss has to be treated aggressively, particularly in patients who are older and have a history of fragility fracture.

For patients with GCA who need glucocorticoids in doses greater than 5 mg orally daily for more than 3 months, the following measures are advised to decrease the risk of bone loss:

• Weight-bearing exercise
• Smoking cessation
• Moderation in alcohol intake
• Measures to prevent falls²⁹
• Supplementation with 1,200 mg of calcium and 800 IU of vitamin D.³⁰

Pharmacologic therapy should be initiated in men over age 50 who have established osteoporosis and in postmenopausal women with established osteoporosis or osteopenia. For men over age 50 with established osteopenia, risk assessment with the glucocorticoid-corrected FRAX score (www.sheffield.ac.uk/FRAX) should be performed to identify those at high risk in whom pharmacologic therapy is warranted.³¹

Bisphosphonates are the first-line therapy for glucocorticoid-induced osteoporosis.³²

Teriparatide is the second-line therapy and is used in patients who cannot tolerate bis­phosphonates or other osteoporosis therapies, and in those who have severe osteoporosis, with T scores of –3.5 and below if they have not had a fracture, and –2.5 and below if they have had a fragility fracture.³³

Denosumab, a monoclonal antibody to an osteoclast differentiating factor, may be beneficial for some patients with glucocorticoid-induced osteoporosis.³⁴

To assess the efficacy of therapy, measuring bone mineral density at baseline and at 1 year of therapy is recommended. If density is stable or improved, then repeating the measurement at 2- to 3-year intervals is suggested.

 

 

TOCILIZUMAB: A STEROID-SPARING MEDICATION

Due to the adverse effects of long-term use of glucocorticoids and high rates of relapse, there is a pressing need for medications that are more efficacious and less toxic to treat GCA.

The European League Against Rheumatism, in its 2009 management guidelines for large-vessel vasculitis, recommend using an adjunctive immunosuppressant agent.¹⁹ In the case of GCA, they recommend using methotrexate 10 to 15 mg/week, which has shown modest evidence of reducing the relapse rate and lowering the cumulative doses of glucocorticoids needed.^35,36

Studies of tumor necrosis factor inhibitors and abatacept have not yielded significant reductions in the relapse rate or decreased cumulative doses of prednisone.^37,38

Advances in treatment for GCA have stagnated, but recent trials^39,40 have evaluated the IL-6 receptor alpha inhibitor tocilizumab, given the central role of IL-6 in the pathogenesis of GCA. Case reports have revealed rapid induction and maintenance of remission in GCA using tocilizumab.^41,42

Villiger et al³⁹ performed a randomized, placebo-controlled trial to study the efficacy and safety of tocilizumab in induction and maintenance of disease remission in 30 patients with newly diagnosed GCA. The primary outcome, complete remission at 12 weeks, was achieved in 85% of patients who received tocilizumab plus tapered prednisolone, compared with 40% of patients who received placebo plus tapering prednisolone. The tocilizumab group also had favorable results in secondary outcomes including relapse-free survival at 52 weeks, time to first relapse after induction of remission, and cumulative dose of prednisolone.

The GiACTA trial. Stone et al⁴⁰ studied the effect of tocilizumab on rates of relapse during glucocorticoid tapering in 251 GCA patients over the course of 52 weeks. Patients were randomized in a 2:1:1:1 ratio to 4 treatment groups:

• Tocilizumab weekly plus prednisone, with prednisone tapered over 26 weeks
• Tocilizumab every other week plus prednisone tapered over 26 weeks
• Placebo plus prednisone tapered over 26 weeks
• Placebo plus prednisone tapered over 52 weeks.

The primary outcome was the rate of sustained glucocorticoid-free remission at 52 weeks. Secondary outcomes included the remission rate, the cumulative glucocorticoid dose, and safety measures. At 52 weeks, the rates of sustained remission were:

• 56% with tocilizumab weekly
• 53% with tocilizumab every other week
• 14% with placebo plus 26-week prednisone taper
• 18% with placebo plus 52-week taper.

Differences between the active treatment groups and the placebo groups were statistically significant (P < .001).

The cumulative dose of prednisone in tocilizumab recipients was significantly less than in placebo recipients. Rates of adverse events were similar. Ultimately, the study showed that tocilizumab, either weekly or every other week, was more effective than prednisone alone at sustaining glucocorticoid-free remission in patients with GCA.

However, the study also raised questions about tocilizumab’s toxic effect profile and its  long-term efficacy, as well as who are the optimal candidates for this therapy. Data on long-term use of tocilizumab are primarily taken from its use in rheumatoid arthritis.⁴³ As of this writing, Stone et al are conducting an open-label trial to help provide long-term safety and efficacy data in patients with GCA. In the meantime, we must extrapolate data from the long-term use of tocilizumab in rheumatoid arthritis.

Tocilizumab and lower gastrointestinal tract perforation

One of the major adverse effects of long-term use of tocilizumab is lower gastrointestinal tract perforation.

Xie et al,⁴⁴ in 2016, reported that the risk of perforation in patients on tocilizumab for rheumatoid arthritis was more than 2 times higher than in patients taking a tumor necrosis factor inhibitor. However, the absolute rates of perforation were low overall,  roughly 1 to 3 per 1,000 patient-years in the tocilizumab group. Risk factors for perforation included older age, history of diverticulitis or other gastrointestinal tract condition, and prednisone doses of 7.5 mg or more a day.

Does tocilizumab prevent blindness?

Another consideration is that tocilizumab may not prevent optic neuropathy. In the GiACTA trial, 1 patient in the group receiving tocilizumab every other week developed optic neuropathy.⁴⁰ Prednisone had been completely tapered off at the time, and the condition resolved when glucocorticoids were restarted. Thus, it is unknown if tocilizumab would be effective on its own without concomitant use of glucocorticoids.

Vision loss is one of the most severe complications of GCA, and it is still unclear whether tocilizumab can prevent vision loss in GCA. Also, we still have no data on the effect of tocilizumab on histopathologic findings, and whether biopsy yield diminishes over time. We hope future studies will help guide us in this regard.

No guidelines on tocilizumab yet

Clinical guidelines on the appropriate use of tocilizumab in GCA are lacking. The American College of Rheumatology and the European League Against Rheumatism have yet to publish updated guidelines with comments on use of tocilizumab. Therefore, it is unclear if tocilizumab is a first-line treatment in GCA, as its efficacy alone without glucocorticoids and its long-term safety in GCA patients have not been studied.

Treatment with tocilizumab should be individualized; it should be considered in patients who have had adverse effects from glucocorticoids, and in patients who experience a flare or cannot have their glucocorticoid dose lowered to an appropriate range.

The optimal duration of tocilizumab therapy is also unknown. However, using the GiACTA study as a rough guide, we try to limit its use to 1 year until additional data are available.

Patients on IL-6 inhibition may have suppressed C-reactive protein regardless of disease activity.⁴³ Therefore, this laboratory value may not be reliable in determining active disease in patients on tocilizumab.

The GiACTA trial has shown an impressive improvement in the relapse-free remission period in patients with GCA taking tocilizumab. However, much work needs to be done to define the safety of this medication and determine which patients should be started on it. In the meantime, we recommend starting high-dose glucocorticoid therapy as soon as the diagnosis of GCA is suspected. In patients who do not tolerate glucocorticoids or whose disease flares during glucocorticoid taper, we recommend starting treatment with tocilizumab either once a week or every other week for at least 1 year.

Giant cell arteritis (GCA) is a systemic vasculitis involving medium-sized and large arteries, most commonly the temporal, ophthalmic, occipital, vertebral, posterior ciliary, and proximal vertebral arteries. Moreover, involvement of the ophthalmic artery and its branches results in loss of vision. GCA can also involve the aorta and its proximal branches, especially in the upper extremities.

GCA is the most common systemic vasculitis in adults. It occurs almost exclusively in patients over age 50 and affects women more than men. It is most frequent in populations of northern European ancestry, especially Scandinavian. In a retrospective cohort study in Norway, the average annual cumulative incidence rate of GCA was 16.7 per 100,000 people over age 50.¹ Risk factors include older age, history of smoking, current smoking, early menopause, and, possibly, stress-related disorders.²

PATHOGENESIS IS NOT COMPLETELY UNDERSTOOD

The pathogenesis of GCA is not completely understood, but there is evidence of immune activation in the arterial wall leading to activation of macrophages and formation of multinucleated giant cells (which may not always be present in biopsies).

The most relevant cytokines in the ongoing pathogenesis are still being defined, but the presence of interferon gamma and interleukin 6 (IL-6) seem to be critical for the expression of the disease. The primary immunogenic triggers for the elaboration of these cytokines and the arteritis remain elusive.

A SPECTRUM OF PRESENTATIONS

The initial symptoms of GCA may be vague, such as malaise, fever, and night sweats, and are likely due to systemic inflammation. Features of vascular involvement include headache, scalp tenderness, and jaw claudication (cramping pain in the jaw while chewing).

A less common but serious feature associated with GCA is partial or complete vision loss affecting 1 or both eyes.³ Some patients suddenly go completely blind without any visual prodrome.

Overlapping GCA phenotypes exist, with a spectrum of presentations that include classic cranial arteritis, extracranial GCA (also called large-vessel GCA), and polymyalgia rheumatica.²

Cranial GCA, the best-characterized clinical presentation, causes symptoms such as headache or signs such as tenderness of the temporal artery. On examination, the temporal arteries may be tender or nodular, and the pulses may be felt above the zygomatic arch, above and in front of the tragus of the ear. About two-thirds of patients with cranial GCA present with new-onset headache, most often in the temporal area, but possibly anywhere throughout the head.

Visual disturbance, jaw claudication, and tongue pain are less common but, if present, increase the likelihood of this diagnosis.²

Large-vessel involvement in GCA is common and refers to involvement of the aorta and its proximal branches. Imaging methods used in diagnosing large-vessel GCA include color Doppler ultrasonography, computed tomography with angiography, magnetic resonance imaging with angiography, and positron emission tomography. In some centers, such imaging is performed in all patients diagnosed with GCA to survey for large-vessel involvement.

Depending on the imaging study, large-vessel involvement has been found in 30% to 80% of cases of GCA.^4,5 It is often associated with nonspecific symptoms such as fever, weight loss, chills, and malaise, but it can also cause more specific symptoms such as unilateral extremity claudication. In contrast to patients with cranial GCA, patients with large-vessel GCA were younger at onset, less likely to have headaches, and more likely to have arm claudication at presentation.⁶ Aortitis of the ascending aorta can occur with a histopathologic pattern of GCA but without the clinical stigmata of GCA.

The finding of aortitis should prompt the clinician to question the patient about other symptoms of GCA and to order imaging of the whole vascular tree. Ultrasonography and biopsy of the temporal arteries can be considered. Whether idiopathic aortitis is part of the GCA spectrum remains to be seen.

Laboratory tests often show anemia, leukocytosis, and thrombocytosis. Acute-phase reactants such as C-reactive protein and the erythrocyte sedimentation rate are often elevated. The sedimentation rate often exceeds 50 mm/hour and sometimes 100 mm/hour.

In 2 retrospective studies, the number of patients with GCA whose sedimentation rate was less than 50 mm/hour ranged between 5% and 11%.^7,8 However, a small percentage of patients with GCA have normal inflammatory markers. Therefore, if the suspicion for GCA is high, treatment should be started and biopsy pursued.⁹ In patients with paraproteinemia or other causes of a spuriously elevated or low erythrocyte sedimentation rate, C-reactive protein is a more reliable test.

Polymyalgia rheumatica is another rheumatologic condition that can occur independently or in conjunction with GCA. It is characterized by stiffness and pain in the proximal joints such as the hips and shoulders, typically worse in the morning and better with activity. Although the patient may subjectively feel weak, a close neurologic examination will reveal normal muscle strength.

Polymyalgia rheumatica is observed in 40% to 60% of patients with GCA at the time of diagnosis; 16% to 21% of patients with polymyalgia rheumatica may develop GCA, especially if untreated.^2,10

Differential diagnosis

Other vasculitides (eg, Takayasu arteritis) can also present with unexplained fever, anemia, and constitutional symptoms.

Infection should be considered if fever is present. An infectious disease accompanied by fever, headache, and elevated inflammatory markers can mimic GCA.

Nonarteritic anterior ischemic optic neuropathy can present with sudden vision loss, prompting concern for underlying GCA. Risk factors include hypertension and diabetes mellitus; other features of GCA, including elevated inflammatory markers, are generally absent.

 

 

TEMPORAL ARTERY BIOPSY: THE GOLD STANDARD FOR DIAGNOSIS

Temporal artery biopsy remains the standard to confirm the diagnosis. However, because inflammation in the temporal arteries can affect some segments but not others, biopsy results on conventional hematoxylin and eosin staining can be falsely negative in patients with GCA. In one study,¹¹ the mean sensitivity of unilateral temporal artery biopsy was 86.9%.

Typical positive histologic findings are inflammation with panarteritis, CD4-positive lymphocytes, macrophages, giant cells, and fragmentation of the internal elastic lamina.¹²

When GCA is suspected, treatment with glucocorticoids should be started immediately and biopsy performed as soon as possible. Delaying biopsy for 14 days or more may not affect the accuracy of biopsy study.¹³ Treatment should never be withheld while awaiting the results of biopsy study.

Biopsy is usually performed unilaterally, on the same side as the symptoms or abnormal findings on examination. Bilateral temporal artery biopsy is also performed and compared with unilateral biopsy; this approach increases the diagnostic yield by about 5%.¹⁴

IMAGING

In patients with suspected GCA, imaging is recommended early to complement the clinical criteria for the diagnosis of GCA.¹⁵ Positron emission tomography, computed tomography angiography, magnetic resonance angiography, or Doppler ultrasonography can reveal inflammation of the arteries in the proximal upper or lower limbs or the aorta.²

In patients with suspected cranial GCA, ultrasonography of the temporal and axillary arteries is recommended first. If ultrasonography is not available or is inconclusive, high-resolution magnetic resonance imaging of the cranial arteries can be used as an alternative. Computed tomography and positron emission tomography of the cranial arteries are not recommended.

In patients with suspected large-vessel GCA, ultrasonography, positron emission tomography, computed tomography, and magnetic resonance imaging may be used to screen for vessel wall inflammation, edema, and luminal narrowing in extracranial arteries. Ultrasonography is of limited value in assessing aortitis.

Color duplex ultrasonography can be applied to assess for vascular inflammation of the temporal or large arteries. The typical finding of the “halo” sign, a hypoechoic ring around the arterial lumen, represents the inflammation-induced thickening of the arterial wall. The “compression sign,” the persistence of the “halo” during compression of the vessel lumen by the ultrasound probe, has high specificity for the diagnosis.¹⁶

Ultrasonography of suspected GCA has yielded sensitivities of 55% to 100% and specificities of 78% to 100%. However, its sensitivity depends on the user’s level of expertise, so it should be done only in medical centers with a high number of GCA cases and with highly experienced sonographers. High-resolution magnetic resonance imaging is an alternative to ultrasonography and has shown similar sensitivity and specificity.³

TREATMENT WITH GLUCOCORTICOIDS

Glucocorticoids remain the standard for treatment of GCA. The therapeutic effect of glucocorticoids in GCA has been established by years of clinical experience, but has never been proven in a placebo-controlled trial. When started appropriately and expeditiously, glucocorticoids produce exquisite resolution of signs and symptoms and prevent the serious complication of vision loss. Rapid resolution of symptoms is so typical of GCA that if the patient’s symptoms persist more than a few days after starting a glucocorticoid, the diagnosis of GCA should be reconsidered.

In a retrospective study of 245 patients with biopsy-proven GCA treated with glucocorticoids, 34 had permanent loss of sight.¹⁷ In 32 (94%) of the 34, the vision loss occurred before glucocorticoids were started. Of the remaining 2 patients, 1 lost vision 8 days into treatment, and the other lost vision 3 years after diagnosis and 1 year after discontinuation of glucocorticoids.

In a series of 144 patients with biopsy-proven GCA, 51 had no vision loss at presentation and no vision loss after starting glucocorticoids, and 93 had vision loss at presentation. In the latter group, symptoms worsened within 5 days of starting glucocorticoids in 9 patients.¹⁸ If vision was intact at the time of presentation, prompt initiation of glucocorticoids reduced the risk of vision loss to less than 1%.

High doses, slowly tapered

The European League Against Rheumatism recommends early initiation of high-dose glucocorticoids for patients with large-vessel vasculitis,¹⁹ and it also recommends glucocorticoids for patients with polymyalgia rheumatica.²⁰ The optimal initial and tapering dosage has never been formally evaluated, but regimens have been devised on the basis of expert opinion.²¹

For patients with GCA who do not have vision loss at the time of diagnosis, the initial dose is prednisone 1 mg/kg or its equivalent daily for 2 to 4 weeks, after which it is tapered.²¹ If the initial dosage is prednisone 60 mg orally daily for 2 to 4 weeks, our practice is to taper it to 50 mg daily for 2 weeks, then 40 mg daily for 2 weeks. Then, it  is decreased by 5 mg every 2 weeks until it is 20 mg daily, and then by 2.5 mg every 2 weeks until it is 10 mg orally daily. Thereafter, the dosage is decreased by 1 mg every 2 to 4 weeks.

For patients with GCA who experience transient vision loss or diplopia at the time of diagnosis, intravenous pulse glucocorticoid therapy should be initiated to reduce the risk of vision loss as rapidly as possible.²² A typical pulse regimen is methylprednisolone 1 g intravenously daily for 3 days. Though not rigorously validated in studies, such an approach is used to avoid vision impairment due to GCA, which is rarely reversible.

 

 

RELAPSE OF DISEASE

Suspect a relapse of GCA if the patient’s initial symptoms recur, if inflammatory markers become elevated, or if classic symptoms of GCA or polymyalgia rheumatica occur. Elevations in inflammatory markers do not definitely indicate a flare of GCA, but they should trigger close monitoring of the patient’s symptoms.

Relapse is treated by increasing the glucocorticoid dosage as appropriate to the nature of the relapse. If vision is affected or the patient has symptoms of GCA, then increments of 30 to 60 mg of prednisone are warranted, whereas if the patient has symptoms of polymyalgia rheumatica, then increments of 5 to 10 mg of prednisone are usually used.

The incidence of relapses of GCA in multiple tertiary care centers has been reported to vary between 34% and 75%.^23,24 Most relapses occur at prednisone dosages of less than 20 mg orally daily and within the first year after diagnosis. The most common symptoms are limb ischemia, jaw claudication, constitutional symptoms, headaches, and polymyalgia rheumatica. In a review of 286 patients,²⁵ 213 (74%) had at least 1 relapse. The first relapse occurred in the first year in 50%, by 2 years in 68%, and by 5 years in 79%.

ADVERSE EFFECTS OF GLUCOCORTICOIDS

In high doses, glucocorticoids have well-known adverse effects. In a population-based study of 120 patients, each patient treated with glucocorticoids experienced at least 1 adverse effect (cataract, fracture, infection, osteonecrosis, diabetes, hypertension, weight gain, capillary fragility, or hair loss).²⁶ The effects were related to aging and cumulative dosage of prednisone but not to the initial dosage.

Glucocorticoids can affect many organs and systems:

• Eyes (cataracts, increased intraocular pressure, exophthalmos)
• Heart (premature atherosclerotic disease, hypertension, fluid retention, hyperlipidemia, arrhythmias)
• Gastrointestinal system (ulcer, gastrointestinal bleeding, gastritis, visceral perforation, hepatic steatosis, acute pancreatitis)
• Bone and muscle (osteopenia, osteoporosis, osteonecrosis, myopathy)
• Brain (mood disorder, psychosis, memory impairment)
• Endocrine system (hyperglycemia, hypothalamic-pituitary-adrenal axis suppression)
• Immune system (immunosuppression, leading to infection and leukocytosis).

Patients receiving a glucocorticoid dose equivalent to 20 mg or more of prednisone daily for 1 month or more who also have another cause of immunocompromise need prophylaxis against Pneumocystis jirovecii pneumonia.²⁷ They should also receive appropriate immunizations before starting glucocorticoids. Live-virus vaccines should not be given to these patients until they have been off glucocorticoids for 1 month.

Glucocorticoids and bone loss

Glucocorticoids are associated with bone loss and fracture, which can occur within the first few months of use and with dosages as low as 2.5 to 7.5 mg orally daily.²⁸ Therefore, glucocorticoid-induced bone loss has to be treated aggressively, particularly in patients who are older and have a history of fragility fracture.

For patients with GCA who need glucocorticoids in doses greater than 5 mg orally daily for more than 3 months, the following measures are advised to decrease the risk of bone loss:

• Weight-bearing exercise
• Smoking cessation
• Moderation in alcohol intake
• Measures to prevent falls²⁹
• Supplementation with 1,200 mg of calcium and 800 IU of vitamin D.³⁰

Pharmacologic therapy should be initiated in men over age 50 who have established osteoporosis and in postmenopausal women with established osteoporosis or osteopenia. For men over age 50 with established osteopenia, risk assessment with the glucocorticoid-corrected FRAX score (www.sheffield.ac.uk/FRAX) should be performed to identify those at high risk in whom pharmacologic therapy is warranted.³¹

Bisphosphonates are the first-line therapy for glucocorticoid-induced osteoporosis.³²

Teriparatide is the second-line therapy and is used in patients who cannot tolerate bis­phosphonates or other osteoporosis therapies, and in those who have severe osteoporosis, with T scores of –3.5 and below if they have not had a fracture, and –2.5 and below if they have had a fragility fracture.³³

Denosumab, a monoclonal antibody to an osteoclast differentiating factor, may be beneficial for some patients with glucocorticoid-induced osteoporosis.³⁴

To assess the efficacy of therapy, measuring bone mineral density at baseline and at 1 year of therapy is recommended. If density is stable or improved, then repeating the measurement at 2- to 3-year intervals is suggested.

 

 

TOCILIZUMAB: A STEROID-SPARING MEDICATION

Due to the adverse effects of long-term use of glucocorticoids and high rates of relapse, there is a pressing need for medications that are more efficacious and less toxic to treat GCA.

The European League Against Rheumatism, in its 2009 management guidelines for large-vessel vasculitis, recommend using an adjunctive immunosuppressant agent.¹⁹ In the case of GCA, they recommend using methotrexate 10 to 15 mg/week, which has shown modest evidence of reducing the relapse rate and lowering the cumulative doses of glucocorticoids needed.^35,36

Studies of tumor necrosis factor inhibitors and abatacept have not yielded significant reductions in the relapse rate or decreased cumulative doses of prednisone.^37,38

Advances in treatment for GCA have stagnated, but recent trials^39,40 have evaluated the IL-6 receptor alpha inhibitor tocilizumab, given the central role of IL-6 in the pathogenesis of GCA. Case reports have revealed rapid induction and maintenance of remission in GCA using tocilizumab.^41,42

Villiger et al³⁹ performed a randomized, placebo-controlled trial to study the efficacy and safety of tocilizumab in induction and maintenance of disease remission in 30 patients with newly diagnosed GCA. The primary outcome, complete remission at 12 weeks, was achieved in 85% of patients who received tocilizumab plus tapered prednisolone, compared with 40% of patients who received placebo plus tapering prednisolone. The tocilizumab group also had favorable results in secondary outcomes including relapse-free survival at 52 weeks, time to first relapse after induction of remission, and cumulative dose of prednisolone.

The GiACTA trial. Stone et al⁴⁰ studied the effect of tocilizumab on rates of relapse during glucocorticoid tapering in 251 GCA patients over the course of 52 weeks. Patients were randomized in a 2:1:1:1 ratio to 4 treatment groups:

• Tocilizumab weekly plus prednisone, with prednisone tapered over 26 weeks
• Tocilizumab every other week plus prednisone tapered over 26 weeks
• Placebo plus prednisone tapered over 26 weeks
• Placebo plus prednisone tapered over 52 weeks.

The primary outcome was the rate of sustained glucocorticoid-free remission at 52 weeks. Secondary outcomes included the remission rate, the cumulative glucocorticoid dose, and safety measures. At 52 weeks, the rates of sustained remission were:

• 56% with tocilizumab weekly
• 53% with tocilizumab every other week
• 14% with placebo plus 26-week prednisone taper
• 18% with placebo plus 52-week taper.

Differences between the active treatment groups and the placebo groups were statistically significant (P < .001).

The cumulative dose of prednisone in tocilizumab recipients was significantly less than in placebo recipients. Rates of adverse events were similar. Ultimately, the study showed that tocilizumab, either weekly or every other week, was more effective than prednisone alone at sustaining glucocorticoid-free remission in patients with GCA.

However, the study also raised questions about tocilizumab’s toxic effect profile and its  long-term efficacy, as well as who are the optimal candidates for this therapy. Data on long-term use of tocilizumab are primarily taken from its use in rheumatoid arthritis.⁴³ As of this writing, Stone et al are conducting an open-label trial to help provide long-term safety and efficacy data in patients with GCA. In the meantime, we must extrapolate data from the long-term use of tocilizumab in rheumatoid arthritis.

Tocilizumab and lower gastrointestinal tract perforation

One of the major adverse effects of long-term use of tocilizumab is lower gastrointestinal tract perforation.

Xie et al,⁴⁴ in 2016, reported that the risk of perforation in patients on tocilizumab for rheumatoid arthritis was more than 2 times higher than in patients taking a tumor necrosis factor inhibitor. However, the absolute rates of perforation were low overall,  roughly 1 to 3 per 1,000 patient-years in the tocilizumab group. Risk factors for perforation included older age, history of diverticulitis or other gastrointestinal tract condition, and prednisone doses of 7.5 mg or more a day.

Does tocilizumab prevent blindness?

Another consideration is that tocilizumab may not prevent optic neuropathy. In the GiACTA trial, 1 patient in the group receiving tocilizumab every other week developed optic neuropathy.⁴⁰ Prednisone had been completely tapered off at the time, and the condition resolved when glucocorticoids were restarted. Thus, it is unknown if tocilizumab would be effective on its own without concomitant use of glucocorticoids.

Vision loss is one of the most severe complications of GCA, and it is still unclear whether tocilizumab can prevent vision loss in GCA. Also, we still have no data on the effect of tocilizumab on histopathologic findings, and whether biopsy yield diminishes over time. We hope future studies will help guide us in this regard.

No guidelines on tocilizumab yet

Clinical guidelines on the appropriate use of tocilizumab in GCA are lacking. The American College of Rheumatology and the European League Against Rheumatism have yet to publish updated guidelines with comments on use of tocilizumab. Therefore, it is unclear if tocilizumab is a first-line treatment in GCA, as its efficacy alone without glucocorticoids and its long-term safety in GCA patients have not been studied.

Treatment with tocilizumab should be individualized; it should be considered in patients who have had adverse effects from glucocorticoids, and in patients who experience a flare or cannot have their glucocorticoid dose lowered to an appropriate range.

The optimal duration of tocilizumab therapy is also unknown. However, using the GiACTA study as a rough guide, we try to limit its use to 1 year until additional data are available.

Patients on IL-6 inhibition may have suppressed C-reactive protein regardless of disease activity.⁴³ Therefore, this laboratory value may not be reliable in determining active disease in patients on tocilizumab.

The GiACTA trial has shown an impressive improvement in the relapse-free remission period in patients with GCA taking tocilizumab. However, much work needs to be done to define the safety of this medication and determine which patients should be started on it. In the meantime, we recommend starting high-dose glucocorticoid therapy as soon as the diagnosis of GCA is suspected. In patients who do not tolerate glucocorticoids or whose disease flares during glucocorticoid taper, we recommend starting treatment with tocilizumab either once a week or every other week for at least 1 year.

Giant cell arteritis (GCA) is a systemic vasculitis involving medium-sized and large arteries, most commonly the temporal, ophthalmic, occipital, vertebral, posterior ciliary, and proximal vertebral arteries. Moreover, involvement of the ophthalmic artery and its branches results in loss of vision. GCA can also involve the aorta and its proximal branches, especially in the upper extremities.

GCA is the most common systemic vasculitis in adults. It occurs almost exclusively in patients over age 50 and affects women more than men. It is most frequent in populations of northern European ancestry, especially Scandinavian. In a retrospective cohort study in Norway, the average annual cumulative incidence rate of GCA was 16.7 per 100,000 people over age 50.¹ Risk factors include older age, history of smoking, current smoking, early menopause, and, possibly, stress-related disorders.²

PATHOGENESIS IS NOT COMPLETELY UNDERSTOOD

The pathogenesis of GCA is not completely understood, but there is evidence of immune activation in the arterial wall leading to activation of macrophages and formation of multinucleated giant cells (which may not always be present in biopsies).

The most relevant cytokines in the ongoing pathogenesis are still being defined, but the presence of interferon gamma and interleukin 6 (IL-6) seem to be critical for the expression of the disease. The primary immunogenic triggers for the elaboration of these cytokines and the arteritis remain elusive.

A SPECTRUM OF PRESENTATIONS

The initial symptoms of GCA may be vague, such as malaise, fever, and night sweats, and are likely due to systemic inflammation. Features of vascular involvement include headache, scalp tenderness, and jaw claudication (cramping pain in the jaw while chewing).

A less common but serious feature associated with GCA is partial or complete vision loss affecting 1 or both eyes.³ Some patients suddenly go completely blind without any visual prodrome.

Overlapping GCA phenotypes exist, with a spectrum of presentations that include classic cranial arteritis, extracranial GCA (also called large-vessel GCA), and polymyalgia rheumatica.²

Cranial GCA, the best-characterized clinical presentation, causes symptoms such as headache or signs such as tenderness of the temporal artery. On examination, the temporal arteries may be tender or nodular, and the pulses may be felt above the zygomatic arch, above and in front of the tragus of the ear. About two-thirds of patients with cranial GCA present with new-onset headache, most often in the temporal area, but possibly anywhere throughout the head.

Visual disturbance, jaw claudication, and tongue pain are less common but, if present, increase the likelihood of this diagnosis.²

Large-vessel involvement in GCA is common and refers to involvement of the aorta and its proximal branches. Imaging methods used in diagnosing large-vessel GCA include color Doppler ultrasonography, computed tomography with angiography, magnetic resonance imaging with angiography, and positron emission tomography. In some centers, such imaging is performed in all patients diagnosed with GCA to survey for large-vessel involvement.

Depending on the imaging study, large-vessel involvement has been found in 30% to 80% of cases of GCA.^4,5 It is often associated with nonspecific symptoms such as fever, weight loss, chills, and malaise, but it can also cause more specific symptoms such as unilateral extremity claudication. In contrast to patients with cranial GCA, patients with large-vessel GCA were younger at onset, less likely to have headaches, and more likely to have arm claudication at presentation.⁶ Aortitis of the ascending aorta can occur with a histopathologic pattern of GCA but without the clinical stigmata of GCA.

The finding of aortitis should prompt the clinician to question the patient about other symptoms of GCA and to order imaging of the whole vascular tree. Ultrasonography and biopsy of the temporal arteries can be considered. Whether idiopathic aortitis is part of the GCA spectrum remains to be seen.

Laboratory tests often show anemia, leukocytosis, and thrombocytosis. Acute-phase reactants such as C-reactive protein and the erythrocyte sedimentation rate are often elevated. The sedimentation rate often exceeds 50 mm/hour and sometimes 100 mm/hour.

In 2 retrospective studies, the number of patients with GCA whose sedimentation rate was less than 50 mm/hour ranged between 5% and 11%.^7,8 However, a small percentage of patients with GCA have normal inflammatory markers. Therefore, if the suspicion for GCA is high, treatment should be started and biopsy pursued.⁹ In patients with paraproteinemia or other causes of a spuriously elevated or low erythrocyte sedimentation rate, C-reactive protein is a more reliable test.

Polymyalgia rheumatica is another rheumatologic condition that can occur independently or in conjunction with GCA. It is characterized by stiffness and pain in the proximal joints such as the hips and shoulders, typically worse in the morning and better with activity. Although the patient may subjectively feel weak, a close neurologic examination will reveal normal muscle strength.

Polymyalgia rheumatica is observed in 40% to 60% of patients with GCA at the time of diagnosis; 16% to 21% of patients with polymyalgia rheumatica may develop GCA, especially if untreated.^2,10

Differential diagnosis

Other vasculitides (eg, Takayasu arteritis) can also present with unexplained fever, anemia, and constitutional symptoms.

Infection should be considered if fever is present. An infectious disease accompanied by fever, headache, and elevated inflammatory markers can mimic GCA.

Nonarteritic anterior ischemic optic neuropathy can present with sudden vision loss, prompting concern for underlying GCA. Risk factors include hypertension and diabetes mellitus; other features of GCA, including elevated inflammatory markers, are generally absent.

 

 

TEMPORAL ARTERY BIOPSY: THE GOLD STANDARD FOR DIAGNOSIS

Temporal artery biopsy remains the standard to confirm the diagnosis. However, because inflammation in the temporal arteries can affect some segments but not others, biopsy results on conventional hematoxylin and eosin staining can be falsely negative in patients with GCA. In one study,¹¹ the mean sensitivity of unilateral temporal artery biopsy was 86.9%.

Typical positive histologic findings are inflammation with panarteritis, CD4-positive lymphocytes, macrophages, giant cells, and fragmentation of the internal elastic lamina.¹²

When GCA is suspected, treatment with glucocorticoids should be started immediately and biopsy performed as soon as possible. Delaying biopsy for 14 days or more may not affect the accuracy of biopsy study.¹³ Treatment should never be withheld while awaiting the results of biopsy study.

Biopsy is usually performed unilaterally, on the same side as the symptoms or abnormal findings on examination. Bilateral temporal artery biopsy is also performed and compared with unilateral biopsy; this approach increases the diagnostic yield by about 5%.¹⁴

IMAGING

In patients with suspected GCA, imaging is recommended early to complement the clinical criteria for the diagnosis of GCA.¹⁵ Positron emission tomography, computed tomography angiography, magnetic resonance angiography, or Doppler ultrasonography can reveal inflammation of the arteries in the proximal upper or lower limbs or the aorta.²

In patients with suspected cranial GCA, ultrasonography of the temporal and axillary arteries is recommended first. If ultrasonography is not available or is inconclusive, high-resolution magnetic resonance imaging of the cranial arteries can be used as an alternative. Computed tomography and positron emission tomography of the cranial arteries are not recommended.

In patients with suspected large-vessel GCA, ultrasonography, positron emission tomography, computed tomography, and magnetic resonance imaging may be used to screen for vessel wall inflammation, edema, and luminal narrowing in extracranial arteries. Ultrasonography is of limited value in assessing aortitis.

Color duplex ultrasonography can be applied to assess for vascular inflammation of the temporal or large arteries. The typical finding of the “halo” sign, a hypoechoic ring around the arterial lumen, represents the inflammation-induced thickening of the arterial wall. The “compression sign,” the persistence of the “halo” during compression of the vessel lumen by the ultrasound probe, has high specificity for the diagnosis.¹⁶

Ultrasonography of suspected GCA has yielded sensitivities of 55% to 100% and specificities of 78% to 100%. However, its sensitivity depends on the user’s level of expertise, so it should be done only in medical centers with a high number of GCA cases and with highly experienced sonographers. High-resolution magnetic resonance imaging is an alternative to ultrasonography and has shown similar sensitivity and specificity.³

TREATMENT WITH GLUCOCORTICOIDS

Glucocorticoids remain the standard for treatment of GCA. The therapeutic effect of glucocorticoids in GCA has been established by years of clinical experience, but has never been proven in a placebo-controlled trial. When started appropriately and expeditiously, glucocorticoids produce exquisite resolution of signs and symptoms and prevent the serious complication of vision loss. Rapid resolution of symptoms is so typical of GCA that if the patient’s symptoms persist more than a few days after starting a glucocorticoid, the diagnosis of GCA should be reconsidered.

In a retrospective study of 245 patients with biopsy-proven GCA treated with glucocorticoids, 34 had permanent loss of sight.¹⁷ In 32 (94%) of the 34, the vision loss occurred before glucocorticoids were started. Of the remaining 2 patients, 1 lost vision 8 days into treatment, and the other lost vision 3 years after diagnosis and 1 year after discontinuation of glucocorticoids.

In a series of 144 patients with biopsy-proven GCA, 51 had no vision loss at presentation and no vision loss after starting glucocorticoids, and 93 had vision loss at presentation. In the latter group, symptoms worsened within 5 days of starting glucocorticoids in 9 patients.¹⁸ If vision was intact at the time of presentation, prompt initiation of glucocorticoids reduced the risk of vision loss to less than 1%.

High doses, slowly tapered

The European League Against Rheumatism recommends early initiation of high-dose glucocorticoids for patients with large-vessel vasculitis,¹⁹ and it also recommends glucocorticoids for patients with polymyalgia rheumatica.²⁰ The optimal initial and tapering dosage has never been formally evaluated, but regimens have been devised on the basis of expert opinion.²¹

For patients with GCA who do not have vision loss at the time of diagnosis, the initial dose is prednisone 1 mg/kg or its equivalent daily for 2 to 4 weeks, after which it is tapered.²¹ If the initial dosage is prednisone 60 mg orally daily for 2 to 4 weeks, our practice is to taper it to 50 mg daily for 2 weeks, then 40 mg daily for 2 weeks. Then, it  is decreased by 5 mg every 2 weeks until it is 20 mg daily, and then by 2.5 mg every 2 weeks until it is 10 mg orally daily. Thereafter, the dosage is decreased by 1 mg every 2 to 4 weeks.

For patients with GCA who experience transient vision loss or diplopia at the time of diagnosis, intravenous pulse glucocorticoid therapy should be initiated to reduce the risk of vision loss as rapidly as possible.²² A typical pulse regimen is methylprednisolone 1 g intravenously daily for 3 days. Though not rigorously validated in studies, such an approach is used to avoid vision impairment due to GCA, which is rarely reversible.

 

 

RELAPSE OF DISEASE

Suspect a relapse of GCA if the patient’s initial symptoms recur, if inflammatory markers become elevated, or if classic symptoms of GCA or polymyalgia rheumatica occur. Elevations in inflammatory markers do not definitely indicate a flare of GCA, but they should trigger close monitoring of the patient’s symptoms.

Relapse is treated by increasing the glucocorticoid dosage as appropriate to the nature of the relapse. If vision is affected or the patient has symptoms of GCA, then increments of 30 to 60 mg of prednisone are warranted, whereas if the patient has symptoms of polymyalgia rheumatica, then increments of 5 to 10 mg of prednisone are usually used.

The incidence of relapses of GCA in multiple tertiary care centers has been reported to vary between 34% and 75%.^23,24 Most relapses occur at prednisone dosages of less than 20 mg orally daily and within the first year after diagnosis. The most common symptoms are limb ischemia, jaw claudication, constitutional symptoms, headaches, and polymyalgia rheumatica. In a review of 286 patients,²⁵ 213 (74%) had at least 1 relapse. The first relapse occurred in the first year in 50%, by 2 years in 68%, and by 5 years in 79%.

ADVERSE EFFECTS OF GLUCOCORTICOIDS

In high doses, glucocorticoids have well-known adverse effects. In a population-based study of 120 patients, each patient treated with glucocorticoids experienced at least 1 adverse effect (cataract, fracture, infection, osteonecrosis, diabetes, hypertension, weight gain, capillary fragility, or hair loss).²⁶ The effects were related to aging and cumulative dosage of prednisone but not to the initial dosage.

Glucocorticoids can affect many organs and systems:

• Eyes (cataracts, increased intraocular pressure, exophthalmos)
• Heart (premature atherosclerotic disease, hypertension, fluid retention, hyperlipidemia, arrhythmias)
• Gastrointestinal system (ulcer, gastrointestinal bleeding, gastritis, visceral perforation, hepatic steatosis, acute pancreatitis)
• Bone and muscle (osteopenia, osteoporosis, osteonecrosis, myopathy)
• Brain (mood disorder, psychosis, memory impairment)
• Endocrine system (hyperglycemia, hypothalamic-pituitary-adrenal axis suppression)
• Immune system (immunosuppression, leading to infection and leukocytosis).

Patients receiving a glucocorticoid dose equivalent to 20 mg or more of prednisone daily for 1 month or more who also have another cause of immunocompromise need prophylaxis against Pneumocystis jirovecii pneumonia.²⁷ They should also receive appropriate immunizations before starting glucocorticoids. Live-virus vaccines should not be given to these patients until they have been off glucocorticoids for 1 month.

Glucocorticoids and bone loss

Glucocorticoids are associated with bone loss and fracture, which can occur within the first few months of use and with dosages as low as 2.5 to 7.5 mg orally daily.²⁸ Therefore, glucocorticoid-induced bone loss has to be treated aggressively, particularly in patients who are older and have a history of fragility fracture.

For patients with GCA who need glucocorticoids in doses greater than 5 mg orally daily for more than 3 months, the following measures are advised to decrease the risk of bone loss:

• Weight-bearing exercise
• Smoking cessation
• Moderation in alcohol intake
• Measures to prevent falls²⁹
• Supplementation with 1,200 mg of calcium and 800 IU of vitamin D.³⁰

Pharmacologic therapy should be initiated in men over age 50 who have established osteoporosis and in postmenopausal women with established osteoporosis or osteopenia. For men over age 50 with established osteopenia, risk assessment with the glucocorticoid-corrected FRAX score (www.sheffield.ac.uk/FRAX) should be performed to identify those at high risk in whom pharmacologic therapy is warranted.³¹

Bisphosphonates are the first-line therapy for glucocorticoid-induced osteoporosis.³²

Teriparatide is the second-line therapy and is used in patients who cannot tolerate bis­phosphonates or other osteoporosis therapies, and in those who have severe osteoporosis, with T scores of –3.5 and below if they have not had a fracture, and –2.5 and below if they have had a fragility fracture.³³

Denosumab, a monoclonal antibody to an osteoclast differentiating factor, may be beneficial for some patients with glucocorticoid-induced osteoporosis.³⁴

To assess the efficacy of therapy, measuring bone mineral density at baseline and at 1 year of therapy is recommended. If density is stable or improved, then repeating the measurement at 2- to 3-year intervals is suggested.

 

 

TOCILIZUMAB: A STEROID-SPARING MEDICATION

Due to the adverse effects of long-term use of glucocorticoids and high rates of relapse, there is a pressing need for medications that are more efficacious and less toxic to treat GCA.

The European League Against Rheumatism, in its 2009 management guidelines for large-vessel vasculitis, recommend using an adjunctive immunosuppressant agent.¹⁹ In the case of GCA, they recommend using methotrexate 10 to 15 mg/week, which has shown modest evidence of reducing the relapse rate and lowering the cumulative doses of glucocorticoids needed.^35,36

Studies of tumor necrosis factor inhibitors and abatacept have not yielded significant reductions in the relapse rate or decreased cumulative doses of prednisone.^37,38

Advances in treatment for GCA have stagnated, but recent trials^39,40 have evaluated the IL-6 receptor alpha inhibitor tocilizumab, given the central role of IL-6 in the pathogenesis of GCA. Case reports have revealed rapid induction and maintenance of remission in GCA using tocilizumab.^41,42

Villiger et al³⁹ performed a randomized, placebo-controlled trial to study the efficacy and safety of tocilizumab in induction and maintenance of disease remission in 30 patients with newly diagnosed GCA. The primary outcome, complete remission at 12 weeks, was achieved in 85% of patients who received tocilizumab plus tapered prednisolone, compared with 40% of patients who received placebo plus tapering prednisolone. The tocilizumab group also had favorable results in secondary outcomes including relapse-free survival at 52 weeks, time to first relapse after induction of remission, and cumulative dose of prednisolone.

The GiACTA trial. Stone et al⁴⁰ studied the effect of tocilizumab on rates of relapse during glucocorticoid tapering in 251 GCA patients over the course of 52 weeks. Patients were randomized in a 2:1:1:1 ratio to 4 treatment groups:

• Tocilizumab weekly plus prednisone, with prednisone tapered over 26 weeks
• Tocilizumab every other week plus prednisone tapered over 26 weeks
• Placebo plus prednisone tapered over 26 weeks
• Placebo plus prednisone tapered over 52 weeks.

The primary outcome was the rate of sustained glucocorticoid-free remission at 52 weeks. Secondary outcomes included the remission rate, the cumulative glucocorticoid dose, and safety measures. At 52 weeks, the rates of sustained remission were:

• 56% with tocilizumab weekly
• 53% with tocilizumab every other week
• 14% with placebo plus 26-week prednisone taper
• 18% with placebo plus 52-week taper.

Differences between the active treatment groups and the placebo groups were statistically significant (P < .001).

The cumulative dose of prednisone in tocilizumab recipients was significantly less than in placebo recipients. Rates of adverse events were similar. Ultimately, the study showed that tocilizumab, either weekly or every other week, was more effective than prednisone alone at sustaining glucocorticoid-free remission in patients with GCA.

However, the study also raised questions about tocilizumab’s toxic effect profile and its  long-term efficacy, as well as who are the optimal candidates for this therapy. Data on long-term use of tocilizumab are primarily taken from its use in rheumatoid arthritis.⁴³ As of this writing, Stone et al are conducting an open-label trial to help provide long-term safety and efficacy data in patients with GCA. In the meantime, we must extrapolate data from the long-term use of tocilizumab in rheumatoid arthritis.

Tocilizumab and lower gastrointestinal tract perforation

One of the major adverse effects of long-term use of tocilizumab is lower gastrointestinal tract perforation.

Xie et al,⁴⁴ in 2016, reported that the risk of perforation in patients on tocilizumab for rheumatoid arthritis was more than 2 times higher than in patients taking a tumor necrosis factor inhibitor. However, the absolute rates of perforation were low overall,  roughly 1 to 3 per 1,000 patient-years in the tocilizumab group. Risk factors for perforation included older age, history of diverticulitis or other gastrointestinal tract condition, and prednisone doses of 7.5 mg or more a day.

Does tocilizumab prevent blindness?

Another consideration is that tocilizumab may not prevent optic neuropathy. In the GiACTA trial, 1 patient in the group receiving tocilizumab every other week developed optic neuropathy.⁴⁰ Prednisone had been completely tapered off at the time, and the condition resolved when glucocorticoids were restarted. Thus, it is unknown if tocilizumab would be effective on its own without concomitant use of glucocorticoids.

Vision loss is one of the most severe complications of GCA, and it is still unclear whether tocilizumab can prevent vision loss in GCA. Also, we still have no data on the effect of tocilizumab on histopathologic findings, and whether biopsy yield diminishes over time. We hope future studies will help guide us in this regard.

No guidelines on tocilizumab yet

Clinical guidelines on the appropriate use of tocilizumab in GCA are lacking. The American College of Rheumatology and the European League Against Rheumatism have yet to publish updated guidelines with comments on use of tocilizumab. Therefore, it is unclear if tocilizumab is a first-line treatment in GCA, as its efficacy alone without glucocorticoids and its long-term safety in GCA patients have not been studied.

Treatment with tocilizumab should be individualized; it should be considered in patients who have had adverse effects from glucocorticoids, and in patients who experience a flare or cannot have their glucocorticoid dose lowered to an appropriate range.

The optimal duration of tocilizumab therapy is also unknown. However, using the GiACTA study as a rough guide, we try to limit its use to 1 year until additional data are available.

Patients on IL-6 inhibition may have suppressed C-reactive protein regardless of disease activity.⁴³ Therefore, this laboratory value may not be reliable in determining active disease in patients on tocilizumab.

The GiACTA trial has shown an impressive improvement in the relapse-free remission period in patients with GCA taking tocilizumab. However, much work needs to be done to define the safety of this medication and determine which patients should be started on it. In the meantime, we recommend starting high-dose glucocorticoid therapy as soon as the diagnosis of GCA is suspected. In patients who do not tolerate glucocorticoids or whose disease flares during glucocorticoid taper, we recommend starting treatment with tocilizumab either once a week or every other week for at least 1 year.

For millions of couples, a primary care physician may be the first point of contact for fertility concerns. Statistics from the US Centers for Disease Control and Prevention indicate that 12% of women ages 15 to 44 received fertility services from 2006 to 2010.¹ Despite seeking services, most couples requested only advice or testing rather than treatments such as ovulation-inducing medications, surgery, or, rarely, assisted reproductive technologies including in vitro fertilization. Based on these data, primary care physicians are in a unique position to offer guidance and provide fertility services in most circumstances without the need for referral.

This article reviews the answers to questions patients frequently ask, and outlines a practical framework for the evaluation and management of the infertile couple.

MANY PATIENTS SEEK INFORMATION

At least 1 million medical visits per year are for women seeking help in becoming pregnant, with the number increasing over the last several decades.¹ Reasons for the increase include delayed childbearing and the effects of aging on the female reproductive system (“female reproductive aging”), as well as the availability of increasingly effective treatments for infertility.

While the prevalence of infertility in US couples is widely quoted as 10% to 15%,² there is no estimate for the number of fertility-related questions patients routinely pose to care providers. These questions often relate to coital timing, use of lubricants, positioning, and the use of fertility trackers and ovulation predictors.

A 2017 study of women with 12 months of infertility found that only 8% sought subspecialist care vs care from a general physician or provider, indicating that generalists are most often the first point of contact.³ The majority (92%) of women responding to a survey regarding fertility-awareness education indicated a preference for immediate counseling from their general practitioner.⁴

Although some healthcare providers may consider infertility simply a quality-of-life issue, the World Health Organization classifies it as a disease, and as such it warrants identification, assessment, and intervention.⁵ Further, patients with infertility are known to experience considerable psychological distress related to their condition. In a comparison study, women with infertility experienced levels of psychological distress similar to the level in patients with cancer and patients with chronic medical illness.⁶

In the current era, general practitioners and women’s health specialists may also now address patients’ questions about reproductive aging and egg-freezing, which is now an established technology.⁷

FAILURE TO CONCEIVE AFTER 1 YEAR

flyckt_infertility_t1.jpg

The American Society of Reproductive Medicine (ASRM) defines infertility as failure to conceive after 1 year of appropriately timed unprotected intercourse⁸; 85% of couples will have achieved a pregnancy within this time period.⁹ In practice, some women are evaluated sooner if they are of advanced maternal age (> age 35) or report a history of chemotherapy or radiation exposure, anovulation, or risk factors for obstructed fallopian tubes (ie, endometriosis, fibroids, or pelvic inflammatory disease). Common causes of infertility are listed in Table 1.

As women approach age 40, the potential for fertility decreases rapidly and significantly. Women in their later 30s have only half the fertility of women in their early 20s.¹⁰ Misperceptions of aging and female fertility have been fueled by widely publicized celebrity births from women in their 40s and even 50s, without disclosing the use of frozen or donor eggs. This unfortunate fact affects women actively trying to conceive as well as women who wish to delay childbearing due to lack of a partner or for personal or professional reasons. Primary care physicians should be able to provide counseling relevant to female reproductive aging and make suitable and timely referrals for fertility preservation if indicated.

AN EMOTIONAL ISSUE

In approaching the couple with infertility, it is important to proceed with great sensitivity for the socioemotional context of this diagnosis. For both the male and female partner, infertility can be highly stigmatizing, and can be viewed as a personal or relationship failure.

Couples should be encouraged to ask embarrassing or uncomfortable questions. Although this may not be feasible in many circumstances, interviews should ideally be conducted with both partners individually as well as together, to allow sensitive issues to be shared. In some cases, a partner may be unaware of a history of a sexually transmitted infection, a prior abortion, the use of testosterone supplements or medications to enhance male sexual performance, or a vasectomy or tubal ligation during a previous relationship.

It is not unusual that the anxiety of infertility can cause decreased libido and sexual and erectile dysfunction. These issues can further complicate the problem of conceiving, and couples counseling is not uncommonly required.11 Patients are often reassured to know that they are not alone in their diagnosis.

 

 

LOOK FOR CLUES

Before embarking on a series of tests, the primary care physician can carefully evaluate for clues that may guide the diagnostic evaluation. The approach can be individualized based on the patient’s age, duration of subfertility (ie, how long they have been trying to become pregnant), and risk factors. But as a general rule, regardless of age, couples who have been trying to conceive for more than 1 year should be encouraged to pursue additional testing.

Because each month presents a new cycle of hope (often followed by intense disappointment), the prevailing sentiment to “just give it a little more time” must be countered by education and counseling. The primary care physician must increase awareness that lack of pregnancy in the stated time periods is a compelling reason for evaluation.

History-taking in the infertile couple should include a complete gynecologic and menstrual history. A history of sexually transmitted diseases that can cause tubal disease, such as gonorrhea and Chlamydia, is significant. Both partners should be assessed for a history of prior conceptions, past medical or surgical problems, medications, and exposures to environmental toxins including alcohol, tobacco, and drugs.

A detailed physical examination can provide clues to the cause of subfertility, especially if signs of obesity, androgen excess, or insulin resistance are present.

QUESTIONS OFTEN ASKED BY COUPLES TRYING TO CONCEIVE

Clinicians are frequently asked questions related to sexual practices and lifestyle in relation to fertility and should be comfortable responding to questions in these areas.

Does frequent ejaculation ‘use up’ my sperm?

Men should be reassured that frequent ejaculations do not decrease sperm counts; even daily ejaculation does not deplete the concentration of sperm. Male partners can be reassured that “saving up” is not an effective strategy; in fact, abstinence periods of greater than 5 days can adversely affect semen parameters.¹²

How often should we have sex?

Infrequent intercourse (< 1 time per week) reduces the monthly chance of conceiving.¹³ There does not seem to be a significant improvement in fecundity with daily intercourse vs intercourse on alternate days. Strict schedules surrounding intercourse may increase stress, and reassurance should be offered that intercourse need not be regimented. Every 1 to 2 days should suffice.

Are any sexual positions better for conception?

There is no evidence that particular coital positioning or remaining supine after intercourse improves fertility. Sperm can be found within the endocervix within seconds of ejaculation, irrespective of sexual position.

What is the window of fertility?

There is good evidence that the fertile window lasts approximately 6 days and closes after ovulation.^13,14 Women with regular cycles can determine their typical day of ovulation based on menstrual tracking. Intercourse should begin about 6 days before ovulation and should continue every 1 to 2 days for 1 week to fully capture this window.

Should we change our lifestyle?

Couples seeking pregnancy should be advised to limit alcohol and caffeine use, completely abstain from cigarette smoking or illicit drug use, and maintain a healthy body mass index.

Very few data exist to support particular diets or supplements to promote fertility, including antioxidants and herbal remedies. Folic acid supplementation is recommended in all women attempting to conceive to reduce the incidence of birth defects.

Do lubricants reduce fertility?

Although there seem to be no differences in fecundity rates in couples using commercial lubricants, most water-based lubricants are best avoided in couples with infertility, as adverse effects on sperm have been demonstrated in vitro.¹⁵ If lubrication is needed, couples may try mineral oil, canola oil, or hydroxyetyl­cellulose-based lubricants (eg, Pre-seed).

Do fertility trackers work?

Many couples with primary infertility perceive that coital timing is critical and worry that their infertility is due to poorly timed intercourse; in fact, this is seldom the case.

Despite widespread marketing of urinary luteinizing hormone (LH) detection kits and electronic trackers and monitors, there is no clear evidence that these methods improve monthly rates of conception.

Women with a regular menstrual cycle should be encouraged to take notice when their cervical mucus appears clear and slippery (a sign of ovulation). Not all women are able to detect these fluctuations; however, for those who can, observing cervical mucus changes appears to be equivalent or superior to predictor kits in predicting conception.¹⁶

A PRACTICAL FRAMEWORK FOR EVALUATING THE INFERTILE COUPLE

To assess for the common factors identified in Table 1, the essential investigation of the infertile couple includes:

• Semen analysis
• Confirmation of ovulation
• Hysterosalpingography.

Consideration can also be given to ovarian reserve testing in women at risk of diminished ovarian reserve. The above investigation can be performed simultaneously to allow for prompt identification of any issues. Further, infertility is often a combination of problems (eg, anovulation in the woman together with a problem in the man), so an incomplete evaluation may overlook a coexisting diagnosis and lead to delays in treatment and pregnancy.

flyckt_infertility_t2.jpg

Although abnormal results from this screening will likely prompt referral to a fertility specialist, most patients seeking management from their primary care physician simply want assessment and education.

Tests that are no longer typically used in clinical practice are outlined in Table 2.

 

 

OVARIAN RESERVE TESTING AND FEMALE REPRODUCTIVE AGING

Ovarian reserve refers to the number of fertilizable oocytes that remain in the ovary. This reserve changes over time, and changes occur rapidly as women approach and enter their 30s. Though not the case in men, the age of the female partner is an independent risk factor for infertility. This discrepancy is due to loss of ovarian reserve, chromosome abnormalities in embryos, and the development of medical conditions with age that affect fertility.

Testing for ovarian reserve does not necessarily predict an overall inability to achieve a live birth,¹⁷ but it can predict response to exogenous gonadotropins and, to some degree, the chance for successful pregnancy with assisted reproductive technology.¹⁸

The ASRM states that testing for diminished ovarian reserve may provide useful information in women who have had a previous poor response to gonadotropins and in women planning assisted reproductive technology.¹⁹ The ASRM also indicates that the following are risk factors for diminished ovarian reserve, and clinicians may target the assessment accordingly¹⁹:

• Age 35 or older
• History of exposure to chemotherapy or pelvic radiation
• Family history of early menopause (age < 40)
• History of ovarian surgery
• Unexplained or idiopathic fertility.

Although several tests of ovarian reserve exist, either an antimullerian hormone (AMH) test or a combined cycle day-3 follicle-stimulating hormone (FSH) and estradiol level are the 2 tests commonly used in clinical practice. Antral follicle counts are an ultrasonographic measure used by infertility specialists but rarely by primary care physicians. Assays such as inhibin are rarely ordered and have limited clinical utility.

The AMH test

Many reproductive endocrinologists rely on the AMH level as a single test of ovarian reserve as it is easy to obtain, has a relatively low cost, and offers stable results. AMH is produced by the granulosa cells of the ovarian antral follicles and is readily detected in serum samples.

Conveniently for the clinician, levels of this hormone remain stable throughout the menstrual cycle and therefore can be tested on any day and at any time of day. Lower serum AMH levels (< 1 ng/mL) have been shown to correspond to diminished ovarian stimulation with gonadotropins as well as decreased embryo quality and poor pregnancy outcomes with assisted reproductive technology.¹⁹

Nevertheless, despite overall stability, AMH levels can be falsely lowered in women using exogenous hormones or with a diagnosis of hypogonadotropic hypogonadism. Levels may be higher than expected in women with polycystic ovary syndrome due to higher numbers of antral and preantral follicles in the polycystic ovary.

The day-3 follicle-stimulating hormone test

FSH and 17-beta estradiol testing can be ordered in combination to assess function of the hypothalamic-pituitary-ovarian axis on day 3 of the menstrual cycle. There is some flexibility, however, and testing obtained on cycle day 2, 3, or 4 yields equivalent results.

Although there are no strict cutoffs, FSH levels that appear elevated (> 10–20 IU/L) are associated with lower chances of conceiving with in vitro fertilization in multiple studies.²⁰

The test is limited by levels that may fluctuate cycle to cycle, and reassuring test results do not necessarily indicate that a woman will achieve a pregnancy. Although a serum estradiol value alone is not a useful test, it can be used in combination with day-3 FSH to screen for diminished ovarian reserve.

As premature recruitment of a follicle can cause an early follicular rise in estradiol, FSH may be falsely suppressed on day 3. For example, a “normal” day-3 FSH combined with an elevated day-3 17-beta estradiol level of 60 to 80 pg/mL is associated with a poor response to medical treatments for infertility.

Female reproductive aging

Aging of the female reproductive system is a central threat to fertility, and prompt assessment and referral are warranted for women age 35 or older who have been trying to conceive for more than 6 months. The ASRM recommends that women over age 40 be evaluated immediately.²¹

A prevailing misconception is that regular menstrual cycles correspond with normal fertility. In reality, women lose their ability to achieve a healthy live birth in the 5 to 10 years preceding menopause. Although all women who do not desire pregnancy should still use appropriate contraception to avoid unintended pregnancy, women who do desire pregnancy should be aware of these physiologic changes.

Classic age-related changes in ovarian reserve are accompanied by a steep rise in aneuploidy and miscarriage risk.²² This is particularly relevant as women increasingly delay childbearing in modern society. Loss of fertility begins at 32 and abruptly accelerates at age 37²¹; this fact is poorly communicated to and understood by patients. In a 2018 study of highly educated women, most respondents failed to identify that 45-year-old women can only rarely achieve a successful pregnancy.²³

In recent decades, the percentage of women who delay childbearing until after age 35 has steadily increased. There is a widespread misconception that fertility treatments and assisted reproductive technology can compensate for female reproductive aging. Primary care physicians can play a central role in reminding couples that age remains the single greatest predictor of natural fertility and the chance of success with assisted reproduction.

Further, for women who desire future fertility and are without a partner, primary care physicians can counsel them regarding the availability of donor insemination or egg freezing. Studies confirm that women want clinicians to initiate information on reproductive health, and 80% of women undergoing elective egg-freezing for fertility preservation wished that they had done so at an earlier age.^24,25

 

 

FEMALE PERITONEAL AND STRUCTURAL CAUSES

Women with endometriosis, fibroids, or a history of tubal disease have impaired fecundity. Pelvic imaging is an essential component of their evaluation. Although hysterosalpingography is the mainstay of tubal assessment, in select cases ultrasonography or hysteroscopy may be indicated.

Tubal disease and hysterosalpingography

Tubal disease remains one of the most common causes of infertility in the US females. In most cases, tubal damage is secondary to pelvic inflammatory disease from infection with gonorrhea or Chlamydia, or both.

Rates of confirmed tubal-factor infertility have been shown to increase with both the severity of the infection and the number of past infections.²⁶ In a landmark study, 1 episode of pelvic inflammatory disease was associated with a 12% risk of tubal-factor infertility, whereas 3 infections carried a risk as high as 54%. Pelvic inflammatory disease is also known to increase the risk of ectopic pregnancy.

To assess tubal patency, hysterosalpingography, a radiographic procedure, is typically performed using fluoroscopy and injected contrast material. Some centers may offer sonohysterography as a radiation-free alternative, depending on sonographic skill and experience. Both tests are best scheduled in the window between the end of menstrual bleeding and ovulation. In practice, patients with regular cycles can typically schedule hysterosalpingography between cycle days 5 and 12.

In patients with known hydrosalpinx (a distended fallopian tube due to blockage) or a history of pelvic infection, doxycycline should be given before the procedure.²⁷ Patients with demonstrated hydrosalpinx on hysterosalpingography should receive doxycycline 100 mg twice daily for 5 days to prevent posthysterosalpingography pelvic inflammatory disease.²⁷ Patients with active pelvic or cervical infection should not undergo hysterosalpingography .

Women with confirmed hydrosalpinx or tubal obstruction can be referred for laparoscopy. Gynecologic surgeons will plan their approach based on whether the obstruction is proximal (near the uterus) or distal (near the ovary) as well as whether hydrosalpinx, abnormal tubal architecture, salpingitis isthmica nodosa, or peritubal adhesions are noted. Tubal surgery can be effective in mild cases of tubal disease; however, as in vitro fertilization is becoming more effective, patients with moderate or severe tubal disease are increasingly being referred directly for assisted reproductive technology. Before undergoing assisted reproductive technology, hydrosalpinx will need to be addressed, as it can decrease clinical pregnancy rates with in vitro fertilization.

Endometriosis

Endometriosis is found in 21% to 47% of women with subfertility²⁸ and commonly causes pain, ovarian cysts, and tubal disease. There is often a delay of 7 to 8 years for diagnosis due to the misapprehension that severe dysmenorrhea is normal. Women with an affected first-degree family member are at substantially increased risk.

Although endometriosis is commonly thought to result from reflux of endometrial tissue into the peritoneal cavity with menses, there are multiple proposed mechanisms for the disease.²⁹ The pathogenesis of endometriosis is enigmatic, and there are likely as yet undetermined immunologic and genetic predispositions that confer increased risk.

Common symptoms of endometriosis are dysmenorrhea, dyspareunia, and pelvic pain, and these are sometimes accompanied by bowel and bladder symptoms. Pelvic examination classically demonstrates an immobile uterus and uterosacral nodularity; palpation of these nodules can elicit pain. On laparoscopy, endometriosis can range from minimal to severe; however, stage of endometriosis correlates poorly with reported symptoms.³⁰

Consideration of surgery is based on clinical history, results of the pelvic examination, and possible findings on ultrasonography or hysterosalpingography. Although positive findings on imaging can support a plan for intervention, endometriosis is largely a peritoneal disease, and evidence of tubal damage or ovarian cysts is rarely evident on ultrasonography. In women with menstrual complaints (eg, dysmenorrhea, heavy menstrual bleeding, abnormal uterine bleeding) and a history of infertility, ultrasonography may be useful in determining the presence of uterine pathology such as ovarian cyst or endometrioma, large hydrosalpinx, polyp, or substantial fibroid burden—any of which may have a significant impact on female fertility.

In the absence of a reliable blood test or imaging study, the gold standard for the diagnosis of endometriosis continues to be laparoscopic surgery. Hormonal treatments for endometriosis symptoms are not effective in improving infertility and will preclude pregnancy. Laparoscopic surgery is more successful in improving pregnancy rates in women with advanced disease: pregnancy rates after surgery can be as high as 60% in women with ovarian endometriomas but are significantly lower in women with removal of minimal to mild disease.^30,31 Women over age 35 or who present with low ovarian reserve and whose male partner has semen abnormalities should consider moving directly to assisted reproductive technology rather than pursuing endometriosis surgery.

 

 

MALE FACTOR INFERTILITY

Although male partners are often highly engaged in and supportive of the fertility evaluation, some are reluctant to undergo testing, and some wish to undergo semen analysis only after female factors have been ruled out. Our practice is to evaluate male factors immediately, due to the high contribution of male factors (up to 40% of cases) either alone or in combination with female factors.³²

Men at particularly increased risk of semen abnormalities include those with a history of chemotherapy or radiation or exposure to toxins (eg, environmental exposures, alcohol, tobacco, illicit substances) and prescribed medications.

At a minimum, for the male partner, a reproductive history should be taken and a semen analysis ordered. Men should be directly queried about testosterone use, as this often-used anabolic steroid hormone can severely impair sperm production.

flyckt_infertility_t3.jpg

Normal semen parameters as designated by the World Health Organization³³ are listed in Table 3. Home collection can be offered at some centers to allay any uneasiness associated with the procedure. Although frequent ejaculation does not appear to affect sperm counts, the ASRM recommends performing formal semen analysis after a window of abstinence of 2 to 5 days.³⁴ The test should be repeated if the result is abnormal, as transient influences such as recent illness may manifest in the sperm parameters for up to 3 months after recovery; this extended effect is related to the duration of normal germ cell maturation. Although there are some differences in sperm parameters of older men, reproductive success does not seem significantly diminished.

Men who have low sperm counts, motility, or morphology scores based on World Health Organization criteria should not be deemed “infertile,” as there is significant variation from one analysis to the next, and normal fertility has been reported in men with notably low sperm counts. Particular caution should be exercised in interpreting low morphology scores in men with normal counts and motility, as this parameter appears to have the least prognostic value in this context. Men with abnormal semen analyses should be referred to a specialist for further urologic evaluation and treatment.

Treatments for male factor infertility include surgery, steroid hormones, and possibly intrauterine insemination or assisted reproductive technology. In even the most challenging cases, male infertility is now largely treatable with intracytoplasmic sperm injection with assisted reproductive technology. While most advances in in vitro fertilization have been evolutionary, intracytoplasmic sperm injection was revolutionary. This breakthrough technology allows a single sperm to be injected directly into the oocyte. Sperm for this procedure can be obtained either from the ejaculate or from microsurgical testicular sperm extraction.

ANOVULATION

flyckt_infertility_t4.jpg

Anovulation manifests with oligo- or amenorrhea and may explain up to 40% of female infertility.² There are myriad causes of anovulation (Table 4); however, polycystic ovary syndrome is the most common.

A thorough menstrual history can be informative, as most females of reproductive age have a fairly predictable 25-to-35-day monthly menstrual cycle. Women presenting with menstrual charting with this pattern do not require laboratory confirmation of ovulation. Basal body temperatures are rarely used currently, as they are time-consuming, can induce stress, and are confirmatory rather than predictive of ovulation. Endometrial biopsy for endometrial “dating” is no longer performed in infertile women.

If laboratory confirmation is desired, LH kit testing with a commercially available test or a luteal phase serum progesterone obtained 7 days after suspected ovulation can be obtained. A serum progesterone level higher than 3 ng/mL is indicative of ovulation.¹⁹ Due to the notable fluctuations in ovulatory-appearing progesterone levels over several hours, caution must be taken in interpreting a lower-normal level as indicative of a luteal phase insufficiency.

Polycystic ovary syndrome

Polycystic ovary syndrome is important to understand because it is a metabolic condition that predisposes patients to a variety of health risks. Along with gynecologic consequences such as infertility, abnormal uterine bleeding, and endometrial pathology, it is often accompanied by alterations in glucose and lipid metabolism, obesity, hypertension, and cardiovascular disease.³⁵

Despite its name, the syndrome does not involve the presence of classic ovarian cysts. In fact, the cysts associated with polycystic ovary syndrome are dense accumulations of antral follicles arranged peripherally in the ovarian cortex; they should not be removed surgically as they represent the ovarian reserve.

Although ovaries that appear polycystic on transvaginal ultrasonography are often associated with the syndrome, they are not invariably present and are not absolutely required for the diagnosis of polycystic ovary syndrome based on the most commonly used criteria.³⁵ Several diagnostic criteria have been proposed for polycystic ovary syndrome and its phenotypes. The 2003 revised Rotterdam criteria require 2 out of the following 3 features:

• Oligo-ovulation or anovulation
• Evidence of hyperandrogenism, whether clinical (eg, acne or hirsutism) or based on laboratory testing
• Polycystic-appearing ovaries on ultrasonography.

There is no single test that can diagnose the disease. Although polycystic ovary syndrome is often characterized by elevated LH levels, LH–FSH ratios, and fasting insulin levels, these are not diagnostic criteria. The diagnosis hinges on excluding other causes of anovulation such as thyroid disease, hyperprolactinemia, 21-hydroxylase deficiency, androgen-producing neoplasms, and Cushing syndrome. In addition to checking serum testosterone levels, irregular menstrual cycles and infertility should be assessed at minimum with measurement of TSH, prolactin, and day-3 FSH. Obese women should be screened for metabolic syndrome, which should include an assessment of impaired glucose tolerance with a 2-hour oral glucose tolerance test.³⁶

Women with polycystic ovary syndrome are known to have insulin resistance, which is difficult to assess and is independent of their body mass index.³⁷ They often report a family history of diabetes or a personal history of gestational diabetes or giving birth to infants who are large for gestational age. Although most women diagnosed with insulin resistance and anovulatory infertility will not yet have a diagnosis of diabetes, women with polycystic ovary syndrome are 3 to 7 times more likely to develop type 2 diabetes later in life³⁷ and are at increased risk of lipid abnormalities, cardiovascular disease, and stroke. Therefore, interventions to address the compounding influences of polycystic ovary syndrome and obesity can improve fertility outcomes and help prevent long-term sequelae that accompany the syndrome.

Treatment for women with polycystic ovary syndrome attempting conception includes lifestyle modifications, medications for ovulation induction, and possible use of insulin sensitizers. Metformin alone is not effective as a single agent for achieving pregnancy.³⁸ Diet, weight loss, and exercise can have dramatic effects on ovulation and pregnancy and should be highly encouraged.

Ovulation induction is often required in anovulatory women, either in combination with lifestyle modifications or used subsequently if modifications are not successful. Letrozole is advised as the initial agent in women with obesity and anovulatory infertility rather than clomiphene citrate; a side-by-side comparison demonstrated increased rates of ovulation and live birth with letrozole.³⁹

Once-daily letrozole 2.5 mg or clomiphene 50 mg can be prescribed for 5 days, from cycle days 3 through 7 to cycle days 5 through 9. If this initial dosing fails to result in ovulation, the dose can be increased. Known adverse effects are hot flashes, headaches, ovarian cysts, and increased risk of multiple gestation.

Metformin should be considered as an adjunct to fertility treatments in women with polycystic ovary syndrome, especially those with obesity or impaired glucose tolerance, or if there is no response to standard ovulation induction.

Ovarian hyperstimulation syndrome (cystic enlargement of the ovaries with potentially dangerous fluid and electrolyte imbalances) can occur in women with polycystic ovary syndrome; however, it rarely occurs with oral medications.

For millions of couples, a primary care physician may be the first point of contact for fertility concerns. Statistics from the US Centers for Disease Control and Prevention indicate that 12% of women ages 15 to 44 received fertility services from 2006 to 2010.¹ Despite seeking services, most couples requested only advice or testing rather than treatments such as ovulation-inducing medications, surgery, or, rarely, assisted reproductive technologies including in vitro fertilization. Based on these data, primary care physicians are in a unique position to offer guidance and provide fertility services in most circumstances without the need for referral.

This article reviews the answers to questions patients frequently ask, and outlines a practical framework for the evaluation and management of the infertile couple.

MANY PATIENTS SEEK INFORMATION

At least 1 million medical visits per year are for women seeking help in becoming pregnant, with the number increasing over the last several decades.¹ Reasons for the increase include delayed childbearing and the effects of aging on the female reproductive system (“female reproductive aging”), as well as the availability of increasingly effective treatments for infertility.

While the prevalence of infertility in US couples is widely quoted as 10% to 15%,² there is no estimate for the number of fertility-related questions patients routinely pose to care providers. These questions often relate to coital timing, use of lubricants, positioning, and the use of fertility trackers and ovulation predictors.

A 2017 study of women with 12 months of infertility found that only 8% sought subspecialist care vs care from a general physician or provider, indicating that generalists are most often the first point of contact.³ The majority (92%) of women responding to a survey regarding fertility-awareness education indicated a preference for immediate counseling from their general practitioner.⁴

Although some healthcare providers may consider infertility simply a quality-of-life issue, the World Health Organization classifies it as a disease, and as such it warrants identification, assessment, and intervention.⁵ Further, patients with infertility are known to experience considerable psychological distress related to their condition. In a comparison study, women with infertility experienced levels of psychological distress similar to the level in patients with cancer and patients with chronic medical illness.⁶

In the current era, general practitioners and women’s health specialists may also now address patients’ questions about reproductive aging and egg-freezing, which is now an established technology.⁷

FAILURE TO CONCEIVE AFTER 1 YEAR

flyckt_infertility_t1.jpg

The American Society of Reproductive Medicine (ASRM) defines infertility as failure to conceive after 1 year of appropriately timed unprotected intercourse⁸; 85% of couples will have achieved a pregnancy within this time period.⁹ In practice, some women are evaluated sooner if they are of advanced maternal age (> age 35) or report a history of chemotherapy or radiation exposure, anovulation, or risk factors for obstructed fallopian tubes (ie, endometriosis, fibroids, or pelvic inflammatory disease). Common causes of infertility are listed in Table 1.

As women approach age 40, the potential for fertility decreases rapidly and significantly. Women in their later 30s have only half the fertility of women in their early 20s.¹⁰ Misperceptions of aging and female fertility have been fueled by widely publicized celebrity births from women in their 40s and even 50s, without disclosing the use of frozen or donor eggs. This unfortunate fact affects women actively trying to conceive as well as women who wish to delay childbearing due to lack of a partner or for personal or professional reasons. Primary care physicians should be able to provide counseling relevant to female reproductive aging and make suitable and timely referrals for fertility preservation if indicated.

AN EMOTIONAL ISSUE

In approaching the couple with infertility, it is important to proceed with great sensitivity for the socioemotional context of this diagnosis. For both the male and female partner, infertility can be highly stigmatizing, and can be viewed as a personal or relationship failure.

Couples should be encouraged to ask embarrassing or uncomfortable questions. Although this may not be feasible in many circumstances, interviews should ideally be conducted with both partners individually as well as together, to allow sensitive issues to be shared. In some cases, a partner may be unaware of a history of a sexually transmitted infection, a prior abortion, the use of testosterone supplements or medications to enhance male sexual performance, or a vasectomy or tubal ligation during a previous relationship.

It is not unusual that the anxiety of infertility can cause decreased libido and sexual and erectile dysfunction. These issues can further complicate the problem of conceiving, and couples counseling is not uncommonly required.11 Patients are often reassured to know that they are not alone in their diagnosis.

 

 

LOOK FOR CLUES

Before embarking on a series of tests, the primary care physician can carefully evaluate for clues that may guide the diagnostic evaluation. The approach can be individualized based on the patient’s age, duration of subfertility (ie, how long they have been trying to become pregnant), and risk factors. But as a general rule, regardless of age, couples who have been trying to conceive for more than 1 year should be encouraged to pursue additional testing.

Because each month presents a new cycle of hope (often followed by intense disappointment), the prevailing sentiment to “just give it a little more time” must be countered by education and counseling. The primary care physician must increase awareness that lack of pregnancy in the stated time periods is a compelling reason for evaluation.

History-taking in the infertile couple should include a complete gynecologic and menstrual history. A history of sexually transmitted diseases that can cause tubal disease, such as gonorrhea and Chlamydia, is significant. Both partners should be assessed for a history of prior conceptions, past medical or surgical problems, medications, and exposures to environmental toxins including alcohol, tobacco, and drugs.

A detailed physical examination can provide clues to the cause of subfertility, especially if signs of obesity, androgen excess, or insulin resistance are present.

QUESTIONS OFTEN ASKED BY COUPLES TRYING TO CONCEIVE

Clinicians are frequently asked questions related to sexual practices and lifestyle in relation to fertility and should be comfortable responding to questions in these areas.

Does frequent ejaculation ‘use up’ my sperm?

Men should be reassured that frequent ejaculations do not decrease sperm counts; even daily ejaculation does not deplete the concentration of sperm. Male partners can be reassured that “saving up” is not an effective strategy; in fact, abstinence periods of greater than 5 days can adversely affect semen parameters.¹²

How often should we have sex?

Infrequent intercourse (< 1 time per week) reduces the monthly chance of conceiving.¹³ There does not seem to be a significant improvement in fecundity with daily intercourse vs intercourse on alternate days. Strict schedules surrounding intercourse may increase stress, and reassurance should be offered that intercourse need not be regimented. Every 1 to 2 days should suffice.

Are any sexual positions better for conception?

There is no evidence that particular coital positioning or remaining supine after intercourse improves fertility. Sperm can be found within the endocervix within seconds of ejaculation, irrespective of sexual position.

What is the window of fertility?

There is good evidence that the fertile window lasts approximately 6 days and closes after ovulation.^13,14 Women with regular cycles can determine their typical day of ovulation based on menstrual tracking. Intercourse should begin about 6 days before ovulation and should continue every 1 to 2 days for 1 week to fully capture this window.

Should we change our lifestyle?

Couples seeking pregnancy should be advised to limit alcohol and caffeine use, completely abstain from cigarette smoking or illicit drug use, and maintain a healthy body mass index.

Very few data exist to support particular diets or supplements to promote fertility, including antioxidants and herbal remedies. Folic acid supplementation is recommended in all women attempting to conceive to reduce the incidence of birth defects.

Do lubricants reduce fertility?

Although there seem to be no differences in fecundity rates in couples using commercial lubricants, most water-based lubricants are best avoided in couples with infertility, as adverse effects on sperm have been demonstrated in vitro.¹⁵ If lubrication is needed, couples may try mineral oil, canola oil, or hydroxyetyl­cellulose-based lubricants (eg, Pre-seed).

Do fertility trackers work?

Many couples with primary infertility perceive that coital timing is critical and worry that their infertility is due to poorly timed intercourse; in fact, this is seldom the case.

Despite widespread marketing of urinary luteinizing hormone (LH) detection kits and electronic trackers and monitors, there is no clear evidence that these methods improve monthly rates of conception.

Women with a regular menstrual cycle should be encouraged to take notice when their cervical mucus appears clear and slippery (a sign of ovulation). Not all women are able to detect these fluctuations; however, for those who can, observing cervical mucus changes appears to be equivalent or superior to predictor kits in predicting conception.¹⁶

A PRACTICAL FRAMEWORK FOR EVALUATING THE INFERTILE COUPLE

To assess for the common factors identified in Table 1, the essential investigation of the infertile couple includes:

• Semen analysis
• Confirmation of ovulation
• Hysterosalpingography.

Consideration can also be given to ovarian reserve testing in women at risk of diminished ovarian reserve. The above investigation can be performed simultaneously to allow for prompt identification of any issues. Further, infertility is often a combination of problems (eg, anovulation in the woman together with a problem in the man), so an incomplete evaluation may overlook a coexisting diagnosis and lead to delays in treatment and pregnancy.

flyckt_infertility_t2.jpg

Although abnormal results from this screening will likely prompt referral to a fertility specialist, most patients seeking management from their primary care physician simply want assessment and education.

Tests that are no longer typically used in clinical practice are outlined in Table 2.

 

 

OVARIAN RESERVE TESTING AND FEMALE REPRODUCTIVE AGING

Ovarian reserve refers to the number of fertilizable oocytes that remain in the ovary. This reserve changes over time, and changes occur rapidly as women approach and enter their 30s. Though not the case in men, the age of the female partner is an independent risk factor for infertility. This discrepancy is due to loss of ovarian reserve, chromosome abnormalities in embryos, and the development of medical conditions with age that affect fertility.

Testing for ovarian reserve does not necessarily predict an overall inability to achieve a live birth,¹⁷ but it can predict response to exogenous gonadotropins and, to some degree, the chance for successful pregnancy with assisted reproductive technology.¹⁸

The ASRM states that testing for diminished ovarian reserve may provide useful information in women who have had a previous poor response to gonadotropins and in women planning assisted reproductive technology.¹⁹ The ASRM also indicates that the following are risk factors for diminished ovarian reserve, and clinicians may target the assessment accordingly¹⁹:

• Age 35 or older
• History of exposure to chemotherapy or pelvic radiation
• Family history of early menopause (age < 40)
• History of ovarian surgery
• Unexplained or idiopathic fertility.

Although several tests of ovarian reserve exist, either an antimullerian hormone (AMH) test or a combined cycle day-3 follicle-stimulating hormone (FSH) and estradiol level are the 2 tests commonly used in clinical practice. Antral follicle counts are an ultrasonographic measure used by infertility specialists but rarely by primary care physicians. Assays such as inhibin are rarely ordered and have limited clinical utility.

The AMH test

Many reproductive endocrinologists rely on the AMH level as a single test of ovarian reserve as it is easy to obtain, has a relatively low cost, and offers stable results. AMH is produced by the granulosa cells of the ovarian antral follicles and is readily detected in serum samples.

Conveniently for the clinician, levels of this hormone remain stable throughout the menstrual cycle and therefore can be tested on any day and at any time of day. Lower serum AMH levels (< 1 ng/mL) have been shown to correspond to diminished ovarian stimulation with gonadotropins as well as decreased embryo quality and poor pregnancy outcomes with assisted reproductive technology.¹⁹

Nevertheless, despite overall stability, AMH levels can be falsely lowered in women using exogenous hormones or with a diagnosis of hypogonadotropic hypogonadism. Levels may be higher than expected in women with polycystic ovary syndrome due to higher numbers of antral and preantral follicles in the polycystic ovary.

The day-3 follicle-stimulating hormone test

FSH and 17-beta estradiol testing can be ordered in combination to assess function of the hypothalamic-pituitary-ovarian axis on day 3 of the menstrual cycle. There is some flexibility, however, and testing obtained on cycle day 2, 3, or 4 yields equivalent results.

Although there are no strict cutoffs, FSH levels that appear elevated (> 10–20 IU/L) are associated with lower chances of conceiving with in vitro fertilization in multiple studies.²⁰

The test is limited by levels that may fluctuate cycle to cycle, and reassuring test results do not necessarily indicate that a woman will achieve a pregnancy. Although a serum estradiol value alone is not a useful test, it can be used in combination with day-3 FSH to screen for diminished ovarian reserve.

As premature recruitment of a follicle can cause an early follicular rise in estradiol, FSH may be falsely suppressed on day 3. For example, a “normal” day-3 FSH combined with an elevated day-3 17-beta estradiol level of 60 to 80 pg/mL is associated with a poor response to medical treatments for infertility.

Female reproductive aging

Aging of the female reproductive system is a central threat to fertility, and prompt assessment and referral are warranted for women age 35 or older who have been trying to conceive for more than 6 months. The ASRM recommends that women over age 40 be evaluated immediately.²¹

A prevailing misconception is that regular menstrual cycles correspond with normal fertility. In reality, women lose their ability to achieve a healthy live birth in the 5 to 10 years preceding menopause. Although all women who do not desire pregnancy should still use appropriate contraception to avoid unintended pregnancy, women who do desire pregnancy should be aware of these physiologic changes.

Classic age-related changes in ovarian reserve are accompanied by a steep rise in aneuploidy and miscarriage risk.²² This is particularly relevant as women increasingly delay childbearing in modern society. Loss of fertility begins at 32 and abruptly accelerates at age 37²¹; this fact is poorly communicated to and understood by patients. In a 2018 study of highly educated women, most respondents failed to identify that 45-year-old women can only rarely achieve a successful pregnancy.²³

In recent decades, the percentage of women who delay childbearing until after age 35 has steadily increased. There is a widespread misconception that fertility treatments and assisted reproductive technology can compensate for female reproductive aging. Primary care physicians can play a central role in reminding couples that age remains the single greatest predictor of natural fertility and the chance of success with assisted reproduction.

Further, for women who desire future fertility and are without a partner, primary care physicians can counsel them regarding the availability of donor insemination or egg freezing. Studies confirm that women want clinicians to initiate information on reproductive health, and 80% of women undergoing elective egg-freezing for fertility preservation wished that they had done so at an earlier age.^24,25

 

 

FEMALE PERITONEAL AND STRUCTURAL CAUSES

Women with endometriosis, fibroids, or a history of tubal disease have impaired fecundity. Pelvic imaging is an essential component of their evaluation. Although hysterosalpingography is the mainstay of tubal assessment, in select cases ultrasonography or hysteroscopy may be indicated.

Tubal disease and hysterosalpingography

Tubal disease remains one of the most common causes of infertility in the US females. In most cases, tubal damage is secondary to pelvic inflammatory disease from infection with gonorrhea or Chlamydia, or both.

Rates of confirmed tubal-factor infertility have been shown to increase with both the severity of the infection and the number of past infections.²⁶ In a landmark study, 1 episode of pelvic inflammatory disease was associated with a 12% risk of tubal-factor infertility, whereas 3 infections carried a risk as high as 54%. Pelvic inflammatory disease is also known to increase the risk of ectopic pregnancy.

To assess tubal patency, hysterosalpingography, a radiographic procedure, is typically performed using fluoroscopy and injected contrast material. Some centers may offer sonohysterography as a radiation-free alternative, depending on sonographic skill and experience. Both tests are best scheduled in the window between the end of menstrual bleeding and ovulation. In practice, patients with regular cycles can typically schedule hysterosalpingography between cycle days 5 and 12.

In patients with known hydrosalpinx (a distended fallopian tube due to blockage) or a history of pelvic infection, doxycycline should be given before the procedure.²⁷ Patients with demonstrated hydrosalpinx on hysterosalpingography should receive doxycycline 100 mg twice daily for 5 days to prevent posthysterosalpingography pelvic inflammatory disease.²⁷ Patients with active pelvic or cervical infection should not undergo hysterosalpingography .

Women with confirmed hydrosalpinx or tubal obstruction can be referred for laparoscopy. Gynecologic surgeons will plan their approach based on whether the obstruction is proximal (near the uterus) or distal (near the ovary) as well as whether hydrosalpinx, abnormal tubal architecture, salpingitis isthmica nodosa, or peritubal adhesions are noted. Tubal surgery can be effective in mild cases of tubal disease; however, as in vitro fertilization is becoming more effective, patients with moderate or severe tubal disease are increasingly being referred directly for assisted reproductive technology. Before undergoing assisted reproductive technology, hydrosalpinx will need to be addressed, as it can decrease clinical pregnancy rates with in vitro fertilization.

Endometriosis

Endometriosis is found in 21% to 47% of women with subfertility²⁸ and commonly causes pain, ovarian cysts, and tubal disease. There is often a delay of 7 to 8 years for diagnosis due to the misapprehension that severe dysmenorrhea is normal. Women with an affected first-degree family member are at substantially increased risk.

Although endometriosis is commonly thought to result from reflux of endometrial tissue into the peritoneal cavity with menses, there are multiple proposed mechanisms for the disease.²⁹ The pathogenesis of endometriosis is enigmatic, and there are likely as yet undetermined immunologic and genetic predispositions that confer increased risk.

Common symptoms of endometriosis are dysmenorrhea, dyspareunia, and pelvic pain, and these are sometimes accompanied by bowel and bladder symptoms. Pelvic examination classically demonstrates an immobile uterus and uterosacral nodularity; palpation of these nodules can elicit pain. On laparoscopy, endometriosis can range from minimal to severe; however, stage of endometriosis correlates poorly with reported symptoms.³⁰

Consideration of surgery is based on clinical history, results of the pelvic examination, and possible findings on ultrasonography or hysterosalpingography. Although positive findings on imaging can support a plan for intervention, endometriosis is largely a peritoneal disease, and evidence of tubal damage or ovarian cysts is rarely evident on ultrasonography. In women with menstrual complaints (eg, dysmenorrhea, heavy menstrual bleeding, abnormal uterine bleeding) and a history of infertility, ultrasonography may be useful in determining the presence of uterine pathology such as ovarian cyst or endometrioma, large hydrosalpinx, polyp, or substantial fibroid burden—any of which may have a significant impact on female fertility.

In the absence of a reliable blood test or imaging study, the gold standard for the diagnosis of endometriosis continues to be laparoscopic surgery. Hormonal treatments for endometriosis symptoms are not effective in improving infertility and will preclude pregnancy. Laparoscopic surgery is more successful in improving pregnancy rates in women with advanced disease: pregnancy rates after surgery can be as high as 60% in women with ovarian endometriomas but are significantly lower in women with removal of minimal to mild disease.^30,31 Women over age 35 or who present with low ovarian reserve and whose male partner has semen abnormalities should consider moving directly to assisted reproductive technology rather than pursuing endometriosis surgery.

 

 

MALE FACTOR INFERTILITY

Although male partners are often highly engaged in and supportive of the fertility evaluation, some are reluctant to undergo testing, and some wish to undergo semen analysis only after female factors have been ruled out. Our practice is to evaluate male factors immediately, due to the high contribution of male factors (up to 40% of cases) either alone or in combination with female factors.³²

Men at particularly increased risk of semen abnormalities include those with a history of chemotherapy or radiation or exposure to toxins (eg, environmental exposures, alcohol, tobacco, illicit substances) and prescribed medications.

At a minimum, for the male partner, a reproductive history should be taken and a semen analysis ordered. Men should be directly queried about testosterone use, as this often-used anabolic steroid hormone can severely impair sperm production.

flyckt_infertility_t3.jpg

Normal semen parameters as designated by the World Health Organization³³ are listed in Table 3. Home collection can be offered at some centers to allay any uneasiness associated with the procedure. Although frequent ejaculation does not appear to affect sperm counts, the ASRM recommends performing formal semen analysis after a window of abstinence of 2 to 5 days.³⁴ The test should be repeated if the result is abnormal, as transient influences such as recent illness may manifest in the sperm parameters for up to 3 months after recovery; this extended effect is related to the duration of normal germ cell maturation. Although there are some differences in sperm parameters of older men, reproductive success does not seem significantly diminished.

Men who have low sperm counts, motility, or morphology scores based on World Health Organization criteria should not be deemed “infertile,” as there is significant variation from one analysis to the next, and normal fertility has been reported in men with notably low sperm counts. Particular caution should be exercised in interpreting low morphology scores in men with normal counts and motility, as this parameter appears to have the least prognostic value in this context. Men with abnormal semen analyses should be referred to a specialist for further urologic evaluation and treatment.

Treatments for male factor infertility include surgery, steroid hormones, and possibly intrauterine insemination or assisted reproductive technology. In even the most challenging cases, male infertility is now largely treatable with intracytoplasmic sperm injection with assisted reproductive technology. While most advances in in vitro fertilization have been evolutionary, intracytoplasmic sperm injection was revolutionary. This breakthrough technology allows a single sperm to be injected directly into the oocyte. Sperm for this procedure can be obtained either from the ejaculate or from microsurgical testicular sperm extraction.

ANOVULATION

flyckt_infertility_t4.jpg

Anovulation manifests with oligo- or amenorrhea and may explain up to 40% of female infertility.² There are myriad causes of anovulation (Table 4); however, polycystic ovary syndrome is the most common.

A thorough menstrual history can be informative, as most females of reproductive age have a fairly predictable 25-to-35-day monthly menstrual cycle. Women presenting with menstrual charting with this pattern do not require laboratory confirmation of ovulation. Basal body temperatures are rarely used currently, as they are time-consuming, can induce stress, and are confirmatory rather than predictive of ovulation. Endometrial biopsy for endometrial “dating” is no longer performed in infertile women.

If laboratory confirmation is desired, LH kit testing with a commercially available test or a luteal phase serum progesterone obtained 7 days after suspected ovulation can be obtained. A serum progesterone level higher than 3 ng/mL is indicative of ovulation.¹⁹ Due to the notable fluctuations in ovulatory-appearing progesterone levels over several hours, caution must be taken in interpreting a lower-normal level as indicative of a luteal phase insufficiency.

Polycystic ovary syndrome

Polycystic ovary syndrome is important to understand because it is a metabolic condition that predisposes patients to a variety of health risks. Along with gynecologic consequences such as infertility, abnormal uterine bleeding, and endometrial pathology, it is often accompanied by alterations in glucose and lipid metabolism, obesity, hypertension, and cardiovascular disease.³⁵

Despite its name, the syndrome does not involve the presence of classic ovarian cysts. In fact, the cysts associated with polycystic ovary syndrome are dense accumulations of antral follicles arranged peripherally in the ovarian cortex; they should not be removed surgically as they represent the ovarian reserve.

Although ovaries that appear polycystic on transvaginal ultrasonography are often associated with the syndrome, they are not invariably present and are not absolutely required for the diagnosis of polycystic ovary syndrome based on the most commonly used criteria.³⁵ Several diagnostic criteria have been proposed for polycystic ovary syndrome and its phenotypes. The 2003 revised Rotterdam criteria require 2 out of the following 3 features:

• Oligo-ovulation or anovulation
• Evidence of hyperandrogenism, whether clinical (eg, acne or hirsutism) or based on laboratory testing
• Polycystic-appearing ovaries on ultrasonography.

There is no single test that can diagnose the disease. Although polycystic ovary syndrome is often characterized by elevated LH levels, LH–FSH ratios, and fasting insulin levels, these are not diagnostic criteria. The diagnosis hinges on excluding other causes of anovulation such as thyroid disease, hyperprolactinemia, 21-hydroxylase deficiency, androgen-producing neoplasms, and Cushing syndrome. In addition to checking serum testosterone levels, irregular menstrual cycles and infertility should be assessed at minimum with measurement of TSH, prolactin, and day-3 FSH. Obese women should be screened for metabolic syndrome, which should include an assessment of impaired glucose tolerance with a 2-hour oral glucose tolerance test.³⁶

Women with polycystic ovary syndrome are known to have insulin resistance, which is difficult to assess and is independent of their body mass index.³⁷ They often report a family history of diabetes or a personal history of gestational diabetes or giving birth to infants who are large for gestational age. Although most women diagnosed with insulin resistance and anovulatory infertility will not yet have a diagnosis of diabetes, women with polycystic ovary syndrome are 3 to 7 times more likely to develop type 2 diabetes later in life³⁷ and are at increased risk of lipid abnormalities, cardiovascular disease, and stroke. Therefore, interventions to address the compounding influences of polycystic ovary syndrome and obesity can improve fertility outcomes and help prevent long-term sequelae that accompany the syndrome.

Treatment for women with polycystic ovary syndrome attempting conception includes lifestyle modifications, medications for ovulation induction, and possible use of insulin sensitizers. Metformin alone is not effective as a single agent for achieving pregnancy.³⁸ Diet, weight loss, and exercise can have dramatic effects on ovulation and pregnancy and should be highly encouraged.

Ovulation induction is often required in anovulatory women, either in combination with lifestyle modifications or used subsequently if modifications are not successful. Letrozole is advised as the initial agent in women with obesity and anovulatory infertility rather than clomiphene citrate; a side-by-side comparison demonstrated increased rates of ovulation and live birth with letrozole.³⁹

Once-daily letrozole 2.5 mg or clomiphene 50 mg can be prescribed for 5 days, from cycle days 3 through 7 to cycle days 5 through 9. If this initial dosing fails to result in ovulation, the dose can be increased. Known adverse effects are hot flashes, headaches, ovarian cysts, and increased risk of multiple gestation.

Metformin should be considered as an adjunct to fertility treatments in women with polycystic ovary syndrome, especially those with obesity or impaired glucose tolerance, or if there is no response to standard ovulation induction.

Ovarian hyperstimulation syndrome (cystic enlargement of the ovaries with potentially dangerous fluid and electrolyte imbalances) can occur in women with polycystic ovary syndrome; however, it rarely occurs with oral medications.

For millions of couples, a primary care physician may be the first point of contact for fertility concerns. Statistics from the US Centers for Disease Control and Prevention indicate that 12% of women ages 15 to 44 received fertility services from 2006 to 2010.¹ Despite seeking services, most couples requested only advice or testing rather than treatments such as ovulation-inducing medications, surgery, or, rarely, assisted reproductive technologies including in vitro fertilization. Based on these data, primary care physicians are in a unique position to offer guidance and provide fertility services in most circumstances without the need for referral.

This article reviews the answers to questions patients frequently ask, and outlines a practical framework for the evaluation and management of the infertile couple.

MANY PATIENTS SEEK INFORMATION

At least 1 million medical visits per year are for women seeking help in becoming pregnant, with the number increasing over the last several decades.¹ Reasons for the increase include delayed childbearing and the effects of aging on the female reproductive system (“female reproductive aging”), as well as the availability of increasingly effective treatments for infertility.

While the prevalence of infertility in US couples is widely quoted as 10% to 15%,² there is no estimate for the number of fertility-related questions patients routinely pose to care providers. These questions often relate to coital timing, use of lubricants, positioning, and the use of fertility trackers and ovulation predictors.

A 2017 study of women with 12 months of infertility found that only 8% sought subspecialist care vs care from a general physician or provider, indicating that generalists are most often the first point of contact.³ The majority (92%) of women responding to a survey regarding fertility-awareness education indicated a preference for immediate counseling from their general practitioner.⁴

Although some healthcare providers may consider infertility simply a quality-of-life issue, the World Health Organization classifies it as a disease, and as such it warrants identification, assessment, and intervention.⁵ Further, patients with infertility are known to experience considerable psychological distress related to their condition. In a comparison study, women with infertility experienced levels of psychological distress similar to the level in patients with cancer and patients with chronic medical illness.⁶

In the current era, general practitioners and women’s health specialists may also now address patients’ questions about reproductive aging and egg-freezing, which is now an established technology.⁷

FAILURE TO CONCEIVE AFTER 1 YEAR

flyckt_infertility_t1.jpg

The American Society of Reproductive Medicine (ASRM) defines infertility as failure to conceive after 1 year of appropriately timed unprotected intercourse⁸; 85% of couples will have achieved a pregnancy within this time period.⁹ In practice, some women are evaluated sooner if they are of advanced maternal age (> age 35) or report a history of chemotherapy or radiation exposure, anovulation, or risk factors for obstructed fallopian tubes (ie, endometriosis, fibroids, or pelvic inflammatory disease). Common causes of infertility are listed in Table 1.

As women approach age 40, the potential for fertility decreases rapidly and significantly. Women in their later 30s have only half the fertility of women in their early 20s.¹⁰ Misperceptions of aging and female fertility have been fueled by widely publicized celebrity births from women in their 40s and even 50s, without disclosing the use of frozen or donor eggs. This unfortunate fact affects women actively trying to conceive as well as women who wish to delay childbearing due to lack of a partner or for personal or professional reasons. Primary care physicians should be able to provide counseling relevant to female reproductive aging and make suitable and timely referrals for fertility preservation if indicated.

AN EMOTIONAL ISSUE

In approaching the couple with infertility, it is important to proceed with great sensitivity for the socioemotional context of this diagnosis. For both the male and female partner, infertility can be highly stigmatizing, and can be viewed as a personal or relationship failure.

Couples should be encouraged to ask embarrassing or uncomfortable questions. Although this may not be feasible in many circumstances, interviews should ideally be conducted with both partners individually as well as together, to allow sensitive issues to be shared. In some cases, a partner may be unaware of a history of a sexually transmitted infection, a prior abortion, the use of testosterone supplements or medications to enhance male sexual performance, or a vasectomy or tubal ligation during a previous relationship.

It is not unusual that the anxiety of infertility can cause decreased libido and sexual and erectile dysfunction. These issues can further complicate the problem of conceiving, and couples counseling is not uncommonly required.11 Patients are often reassured to know that they are not alone in their diagnosis.

 

 

LOOK FOR CLUES

Before embarking on a series of tests, the primary care physician can carefully evaluate for clues that may guide the diagnostic evaluation. The approach can be individualized based on the patient’s age, duration of subfertility (ie, how long they have been trying to become pregnant), and risk factors. But as a general rule, regardless of age, couples who have been trying to conceive for more than 1 year should be encouraged to pursue additional testing.

Because each month presents a new cycle of hope (often followed by intense disappointment), the prevailing sentiment to “just give it a little more time” must be countered by education and counseling. The primary care physician must increase awareness that lack of pregnancy in the stated time periods is a compelling reason for evaluation.

History-taking in the infertile couple should include a complete gynecologic and menstrual history. A history of sexually transmitted diseases that can cause tubal disease, such as gonorrhea and Chlamydia, is significant. Both partners should be assessed for a history of prior conceptions, past medical or surgical problems, medications, and exposures to environmental toxins including alcohol, tobacco, and drugs.

A detailed physical examination can provide clues to the cause of subfertility, especially if signs of obesity, androgen excess, or insulin resistance are present.

QUESTIONS OFTEN ASKED BY COUPLES TRYING TO CONCEIVE

Clinicians are frequently asked questions related to sexual practices and lifestyle in relation to fertility and should be comfortable responding to questions in these areas.

Does frequent ejaculation ‘use up’ my sperm?

Men should be reassured that frequent ejaculations do not decrease sperm counts; even daily ejaculation does not deplete the concentration of sperm. Male partners can be reassured that “saving up” is not an effective strategy; in fact, abstinence periods of greater than 5 days can adversely affect semen parameters.¹²

How often should we have sex?

Infrequent intercourse (< 1 time per week) reduces the monthly chance of conceiving.¹³ There does not seem to be a significant improvement in fecundity with daily intercourse vs intercourse on alternate days. Strict schedules surrounding intercourse may increase stress, and reassurance should be offered that intercourse need not be regimented. Every 1 to 2 days should suffice.

Are any sexual positions better for conception?

There is no evidence that particular coital positioning or remaining supine after intercourse improves fertility. Sperm can be found within the endocervix within seconds of ejaculation, irrespective of sexual position.

What is the window of fertility?

There is good evidence that the fertile window lasts approximately 6 days and closes after ovulation.^13,14 Women with regular cycles can determine their typical day of ovulation based on menstrual tracking. Intercourse should begin about 6 days before ovulation and should continue every 1 to 2 days for 1 week to fully capture this window.

Should we change our lifestyle?

Couples seeking pregnancy should be advised to limit alcohol and caffeine use, completely abstain from cigarette smoking or illicit drug use, and maintain a healthy body mass index.

Very few data exist to support particular diets or supplements to promote fertility, including antioxidants and herbal remedies. Folic acid supplementation is recommended in all women attempting to conceive to reduce the incidence of birth defects.

Do lubricants reduce fertility?

Although there seem to be no differences in fecundity rates in couples using commercial lubricants, most water-based lubricants are best avoided in couples with infertility, as adverse effects on sperm have been demonstrated in vitro.¹⁵ If lubrication is needed, couples may try mineral oil, canola oil, or hydroxyetyl­cellulose-based lubricants (eg, Pre-seed).

Do fertility trackers work?

Many couples with primary infertility perceive that coital timing is critical and worry that their infertility is due to poorly timed intercourse; in fact, this is seldom the case.

Despite widespread marketing of urinary luteinizing hormone (LH) detection kits and electronic trackers and monitors, there is no clear evidence that these methods improve monthly rates of conception.

Women with a regular menstrual cycle should be encouraged to take notice when their cervical mucus appears clear and slippery (a sign of ovulation). Not all women are able to detect these fluctuations; however, for those who can, observing cervical mucus changes appears to be equivalent or superior to predictor kits in predicting conception.¹⁶

A PRACTICAL FRAMEWORK FOR EVALUATING THE INFERTILE COUPLE

To assess for the common factors identified in Table 1, the essential investigation of the infertile couple includes:

• Semen analysis
• Confirmation of ovulation
• Hysterosalpingography.

Consideration can also be given to ovarian reserve testing in women at risk of diminished ovarian reserve. The above investigation can be performed simultaneously to allow for prompt identification of any issues. Further, infertility is often a combination of problems (eg, anovulation in the woman together with a problem in the man), so an incomplete evaluation may overlook a coexisting diagnosis and lead to delays in treatment and pregnancy.

flyckt_infertility_t2.jpg

Although abnormal results from this screening will likely prompt referral to a fertility specialist, most patients seeking management from their primary care physician simply want assessment and education.

Tests that are no longer typically used in clinical practice are outlined in Table 2.

 

 

OVARIAN RESERVE TESTING AND FEMALE REPRODUCTIVE AGING

Ovarian reserve refers to the number of fertilizable oocytes that remain in the ovary. This reserve changes over time, and changes occur rapidly as women approach and enter their 30s. Though not the case in men, the age of the female partner is an independent risk factor for infertility. This discrepancy is due to loss of ovarian reserve, chromosome abnormalities in embryos, and the development of medical conditions with age that affect fertility.

Testing for ovarian reserve does not necessarily predict an overall inability to achieve a live birth,¹⁷ but it can predict response to exogenous gonadotropins and, to some degree, the chance for successful pregnancy with assisted reproductive technology.¹⁸

The ASRM states that testing for diminished ovarian reserve may provide useful information in women who have had a previous poor response to gonadotropins and in women planning assisted reproductive technology.¹⁹ The ASRM also indicates that the following are risk factors for diminished ovarian reserve, and clinicians may target the assessment accordingly¹⁹:

• Age 35 or older
• History of exposure to chemotherapy or pelvic radiation
• Family history of early menopause (age < 40)
• History of ovarian surgery
• Unexplained or idiopathic fertility.

Although several tests of ovarian reserve exist, either an antimullerian hormone (AMH) test or a combined cycle day-3 follicle-stimulating hormone (FSH) and estradiol level are the 2 tests commonly used in clinical practice. Antral follicle counts are an ultrasonographic measure used by infertility specialists but rarely by primary care physicians. Assays such as inhibin are rarely ordered and have limited clinical utility.

The AMH test

Many reproductive endocrinologists rely on the AMH level as a single test of ovarian reserve as it is easy to obtain, has a relatively low cost, and offers stable results. AMH is produced by the granulosa cells of the ovarian antral follicles and is readily detected in serum samples.

Conveniently for the clinician, levels of this hormone remain stable throughout the menstrual cycle and therefore can be tested on any day and at any time of day. Lower serum AMH levels (< 1 ng/mL) have been shown to correspond to diminished ovarian stimulation with gonadotropins as well as decreased embryo quality and poor pregnancy outcomes with assisted reproductive technology.¹⁹

Nevertheless, despite overall stability, AMH levels can be falsely lowered in women using exogenous hormones or with a diagnosis of hypogonadotropic hypogonadism. Levels may be higher than expected in women with polycystic ovary syndrome due to higher numbers of antral and preantral follicles in the polycystic ovary.

The day-3 follicle-stimulating hormone test

FSH and 17-beta estradiol testing can be ordered in combination to assess function of the hypothalamic-pituitary-ovarian axis on day 3 of the menstrual cycle. There is some flexibility, however, and testing obtained on cycle day 2, 3, or 4 yields equivalent results.

Although there are no strict cutoffs, FSH levels that appear elevated (> 10–20 IU/L) are associated with lower chances of conceiving with in vitro fertilization in multiple studies.²⁰

The test is limited by levels that may fluctuate cycle to cycle, and reassuring test results do not necessarily indicate that a woman will achieve a pregnancy. Although a serum estradiol value alone is not a useful test, it can be used in combination with day-3 FSH to screen for diminished ovarian reserve.

As premature recruitment of a follicle can cause an early follicular rise in estradiol, FSH may be falsely suppressed on day 3. For example, a “normal” day-3 FSH combined with an elevated day-3 17-beta estradiol level of 60 to 80 pg/mL is associated with a poor response to medical treatments for infertility.

Female reproductive aging

Aging of the female reproductive system is a central threat to fertility, and prompt assessment and referral are warranted for women age 35 or older who have been trying to conceive for more than 6 months. The ASRM recommends that women over age 40 be evaluated immediately.²¹

A prevailing misconception is that regular menstrual cycles correspond with normal fertility. In reality, women lose their ability to achieve a healthy live birth in the 5 to 10 years preceding menopause. Although all women who do not desire pregnancy should still use appropriate contraception to avoid unintended pregnancy, women who do desire pregnancy should be aware of these physiologic changes.

Classic age-related changes in ovarian reserve are accompanied by a steep rise in aneuploidy and miscarriage risk.²² This is particularly relevant as women increasingly delay childbearing in modern society. Loss of fertility begins at 32 and abruptly accelerates at age 37²¹; this fact is poorly communicated to and understood by patients. In a 2018 study of highly educated women, most respondents failed to identify that 45-year-old women can only rarely achieve a successful pregnancy.²³

In recent decades, the percentage of women who delay childbearing until after age 35 has steadily increased. There is a widespread misconception that fertility treatments and assisted reproductive technology can compensate for female reproductive aging. Primary care physicians can play a central role in reminding couples that age remains the single greatest predictor of natural fertility and the chance of success with assisted reproduction.

Further, for women who desire future fertility and are without a partner, primary care physicians can counsel them regarding the availability of donor insemination or egg freezing. Studies confirm that women want clinicians to initiate information on reproductive health, and 80% of women undergoing elective egg-freezing for fertility preservation wished that they had done so at an earlier age.^24,25

 

 

FEMALE PERITONEAL AND STRUCTURAL CAUSES

Women with endometriosis, fibroids, or a history of tubal disease have impaired fecundity. Pelvic imaging is an essential component of their evaluation. Although hysterosalpingography is the mainstay of tubal assessment, in select cases ultrasonography or hysteroscopy may be indicated.

Tubal disease and hysterosalpingography

Tubal disease remains one of the most common causes of infertility in the US females. In most cases, tubal damage is secondary to pelvic inflammatory disease from infection with gonorrhea or Chlamydia, or both.

Rates of confirmed tubal-factor infertility have been shown to increase with both the severity of the infection and the number of past infections.²⁶ In a landmark study, 1 episode of pelvic inflammatory disease was associated with a 12% risk of tubal-factor infertility, whereas 3 infections carried a risk as high as 54%. Pelvic inflammatory disease is also known to increase the risk of ectopic pregnancy.

To assess tubal patency, hysterosalpingography, a radiographic procedure, is typically performed using fluoroscopy and injected contrast material. Some centers may offer sonohysterography as a radiation-free alternative, depending on sonographic skill and experience. Both tests are best scheduled in the window between the end of menstrual bleeding and ovulation. In practice, patients with regular cycles can typically schedule hysterosalpingography between cycle days 5 and 12.

In patients with known hydrosalpinx (a distended fallopian tube due to blockage) or a history of pelvic infection, doxycycline should be given before the procedure.²⁷ Patients with demonstrated hydrosalpinx on hysterosalpingography should receive doxycycline 100 mg twice daily for 5 days to prevent posthysterosalpingography pelvic inflammatory disease.²⁷ Patients with active pelvic or cervical infection should not undergo hysterosalpingography .

Women with confirmed hydrosalpinx or tubal obstruction can be referred for laparoscopy. Gynecologic surgeons will plan their approach based on whether the obstruction is proximal (near the uterus) or distal (near the ovary) as well as whether hydrosalpinx, abnormal tubal architecture, salpingitis isthmica nodosa, or peritubal adhesions are noted. Tubal surgery can be effective in mild cases of tubal disease; however, as in vitro fertilization is becoming more effective, patients with moderate or severe tubal disease are increasingly being referred directly for assisted reproductive technology. Before undergoing assisted reproductive technology, hydrosalpinx will need to be addressed, as it can decrease clinical pregnancy rates with in vitro fertilization.

Endometriosis

Endometriosis is found in 21% to 47% of women with subfertility²⁸ and commonly causes pain, ovarian cysts, and tubal disease. There is often a delay of 7 to 8 years for diagnosis due to the misapprehension that severe dysmenorrhea is normal. Women with an affected first-degree family member are at substantially increased risk.

Although endometriosis is commonly thought to result from reflux of endometrial tissue into the peritoneal cavity with menses, there are multiple proposed mechanisms for the disease.²⁹ The pathogenesis of endometriosis is enigmatic, and there are likely as yet undetermined immunologic and genetic predispositions that confer increased risk.

Common symptoms of endometriosis are dysmenorrhea, dyspareunia, and pelvic pain, and these are sometimes accompanied by bowel and bladder symptoms. Pelvic examination classically demonstrates an immobile uterus and uterosacral nodularity; palpation of these nodules can elicit pain. On laparoscopy, endometriosis can range from minimal to severe; however, stage of endometriosis correlates poorly with reported symptoms.³⁰

Consideration of surgery is based on clinical history, results of the pelvic examination, and possible findings on ultrasonography or hysterosalpingography. Although positive findings on imaging can support a plan for intervention, endometriosis is largely a peritoneal disease, and evidence of tubal damage or ovarian cysts is rarely evident on ultrasonography. In women with menstrual complaints (eg, dysmenorrhea, heavy menstrual bleeding, abnormal uterine bleeding) and a history of infertility, ultrasonography may be useful in determining the presence of uterine pathology such as ovarian cyst or endometrioma, large hydrosalpinx, polyp, or substantial fibroid burden—any of which may have a significant impact on female fertility.

In the absence of a reliable blood test or imaging study, the gold standard for the diagnosis of endometriosis continues to be laparoscopic surgery. Hormonal treatments for endometriosis symptoms are not effective in improving infertility and will preclude pregnancy. Laparoscopic surgery is more successful in improving pregnancy rates in women with advanced disease: pregnancy rates after surgery can be as high as 60% in women with ovarian endometriomas but are significantly lower in women with removal of minimal to mild disease.^30,31 Women over age 35 or who present with low ovarian reserve and whose male partner has semen abnormalities should consider moving directly to assisted reproductive technology rather than pursuing endometriosis surgery.

 

 

MALE FACTOR INFERTILITY

Although male partners are often highly engaged in and supportive of the fertility evaluation, some are reluctant to undergo testing, and some wish to undergo semen analysis only after female factors have been ruled out. Our practice is to evaluate male factors immediately, due to the high contribution of male factors (up to 40% of cases) either alone or in combination with female factors.³²

Men at particularly increased risk of semen abnormalities include those with a history of chemotherapy or radiation or exposure to toxins (eg, environmental exposures, alcohol, tobacco, illicit substances) and prescribed medications.

At a minimum, for the male partner, a reproductive history should be taken and a semen analysis ordered. Men should be directly queried about testosterone use, as this often-used anabolic steroid hormone can severely impair sperm production.

flyckt_infertility_t3.jpg

Normal semen parameters as designated by the World Health Organization³³ are listed in Table 3. Home collection can be offered at some centers to allay any uneasiness associated with the procedure. Although frequent ejaculation does not appear to affect sperm counts, the ASRM recommends performing formal semen analysis after a window of abstinence of 2 to 5 days.³⁴ The test should be repeated if the result is abnormal, as transient influences such as recent illness may manifest in the sperm parameters for up to 3 months after recovery; this extended effect is related to the duration of normal germ cell maturation. Although there are some differences in sperm parameters of older men, reproductive success does not seem significantly diminished.

Men who have low sperm counts, motility, or morphology scores based on World Health Organization criteria should not be deemed “infertile,” as there is significant variation from one analysis to the next, and normal fertility has been reported in men with notably low sperm counts. Particular caution should be exercised in interpreting low morphology scores in men with normal counts and motility, as this parameter appears to have the least prognostic value in this context. Men with abnormal semen analyses should be referred to a specialist for further urologic evaluation and treatment.

Treatments for male factor infertility include surgery, steroid hormones, and possibly intrauterine insemination or assisted reproductive technology. In even the most challenging cases, male infertility is now largely treatable with intracytoplasmic sperm injection with assisted reproductive technology. While most advances in in vitro fertilization have been evolutionary, intracytoplasmic sperm injection was revolutionary. This breakthrough technology allows a single sperm to be injected directly into the oocyte. Sperm for this procedure can be obtained either from the ejaculate or from microsurgical testicular sperm extraction.

ANOVULATION

flyckt_infertility_t4.jpg

Anovulation manifests with oligo- or amenorrhea and may explain up to 40% of female infertility.² There are myriad causes of anovulation (Table 4); however, polycystic ovary syndrome is the most common.

A thorough menstrual history can be informative, as most females of reproductive age have a fairly predictable 25-to-35-day monthly menstrual cycle. Women presenting with menstrual charting with this pattern do not require laboratory confirmation of ovulation. Basal body temperatures are rarely used currently, as they are time-consuming, can induce stress, and are confirmatory rather than predictive of ovulation. Endometrial biopsy for endometrial “dating” is no longer performed in infertile women.

If laboratory confirmation is desired, LH kit testing with a commercially available test or a luteal phase serum progesterone obtained 7 days after suspected ovulation can be obtained. A serum progesterone level higher than 3 ng/mL is indicative of ovulation.¹⁹ Due to the notable fluctuations in ovulatory-appearing progesterone levels over several hours, caution must be taken in interpreting a lower-normal level as indicative of a luteal phase insufficiency.

Polycystic ovary syndrome

Polycystic ovary syndrome is important to understand because it is a metabolic condition that predisposes patients to a variety of health risks. Along with gynecologic consequences such as infertility, abnormal uterine bleeding, and endometrial pathology, it is often accompanied by alterations in glucose and lipid metabolism, obesity, hypertension, and cardiovascular disease.³⁵

Despite its name, the syndrome does not involve the presence of classic ovarian cysts. In fact, the cysts associated with polycystic ovary syndrome are dense accumulations of antral follicles arranged peripherally in the ovarian cortex; they should not be removed surgically as they represent the ovarian reserve.

Although ovaries that appear polycystic on transvaginal ultrasonography are often associated with the syndrome, they are not invariably present and are not absolutely required for the diagnosis of polycystic ovary syndrome based on the most commonly used criteria.³⁵ Several diagnostic criteria have been proposed for polycystic ovary syndrome and its phenotypes. The 2003 revised Rotterdam criteria require 2 out of the following 3 features:

• Oligo-ovulation or anovulation
• Evidence of hyperandrogenism, whether clinical (eg, acne or hirsutism) or based on laboratory testing
• Polycystic-appearing ovaries on ultrasonography.

There is no single test that can diagnose the disease. Although polycystic ovary syndrome is often characterized by elevated LH levels, LH–FSH ratios, and fasting insulin levels, these are not diagnostic criteria. The diagnosis hinges on excluding other causes of anovulation such as thyroid disease, hyperprolactinemia, 21-hydroxylase deficiency, androgen-producing neoplasms, and Cushing syndrome. In addition to checking serum testosterone levels, irregular menstrual cycles and infertility should be assessed at minimum with measurement of TSH, prolactin, and day-3 FSH. Obese women should be screened for metabolic syndrome, which should include an assessment of impaired glucose tolerance with a 2-hour oral glucose tolerance test.³⁶

Women with polycystic ovary syndrome are known to have insulin resistance, which is difficult to assess and is independent of their body mass index.³⁷ They often report a family history of diabetes or a personal history of gestational diabetes or giving birth to infants who are large for gestational age. Although most women diagnosed with insulin resistance and anovulatory infertility will not yet have a diagnosis of diabetes, women with polycystic ovary syndrome are 3 to 7 times more likely to develop type 2 diabetes later in life³⁷ and are at increased risk of lipid abnormalities, cardiovascular disease, and stroke. Therefore, interventions to address the compounding influences of polycystic ovary syndrome and obesity can improve fertility outcomes and help prevent long-term sequelae that accompany the syndrome.

Treatment for women with polycystic ovary syndrome attempting conception includes lifestyle modifications, medications for ovulation induction, and possible use of insulin sensitizers. Metformin alone is not effective as a single agent for achieving pregnancy.³⁸ Diet, weight loss, and exercise can have dramatic effects on ovulation and pregnancy and should be highly encouraged.

Ovulation induction is often required in anovulatory women, either in combination with lifestyle modifications or used subsequently if modifications are not successful. Letrozole is advised as the initial agent in women with obesity and anovulatory infertility rather than clomiphene citrate; a side-by-side comparison demonstrated increased rates of ovulation and live birth with letrozole.³⁹

Once-daily letrozole 2.5 mg or clomiphene 50 mg can be prescribed for 5 days, from cycle days 3 through 7 to cycle days 5 through 9. If this initial dosing fails to result in ovulation, the dose can be increased. Known adverse effects are hot flashes, headaches, ovarian cysts, and increased risk of multiple gestation.

Metformin should be considered as an adjunct to fertility treatments in women with polycystic ovary syndrome, especially those with obesity or impaired glucose tolerance, or if there is no response to standard ovulation induction.

Ovarian hyperstimulation syndrome (cystic enlargement of the ovaries with potentially dangerous fluid and electrolyte imbalances) can occur in women with polycystic ovary syndrome; however, it rarely occurs with oral medications.