Savvy Psychopharmacology

Ms. W, age 27, presents with a chief concern of “depression.” She describes a history of several hypo­manic episodes as well as the current depres­sive episode, prompting a bipolar II disorder diagnosis. She is naïve to all psychotropics. You plan to initiate a mood-stabilizing agent. What would you include in your initial workup before starting treatment and how would you monitor her as she continues treatment?

Mood stabilizers are employed to treat bipolar spectrum disorders (bipolar I, bipolar II, and cyclothymic disorder) and schizoaffective disorder, bipolar type. Some evidence suggests that mood stabiliz­ers also can be used for treatment-resistant depressive disorders and borderline per­sonality disorder.¹ Mood stabilizers include lithium, valproate, carbamazepine, oxcar­bazepine, and lamotrigine.^2-5

This review focuses on applications and monitoring of mood stabilizers for bipolar I and II disorders. We also will briefly review atypical antipsychotics because they also are used to treat bipo­lar spectrum disorders (see the September 2013 issue of Current Psychiatry at CurrentPsychiatry.com for a more detailed article on monitoring of antipsychotics).⁶

There are several well-researched guidelines used to guide clinical prac­tice.^2-5 Many guidelines recommend base­line and routine monitoring parameters based on the characteristics of the agent used. However, the International Society for Bipolar Disorders (ISBD) guidelines highlight the importance of monitoring medical comorbidities, which are common among patients with bipolar disorder and can affect pharmacotherapy and clinical outcomes. These recommendations are similar to metabolic monitoring guidelines for antipsychotics.⁵

Reviews of therapeutic monitoring show that only one-third to one-half of patien

taking a mood stabilizer are appropriately monitored. Poor adher­ence to guideline recommendations often is observed because of patients’ lack of insight or medication adherence and because psychiatric care generally is segre­gated from other medical care.7-9

Baseline testing
The ISBD guidelines recommend an initial workup for all patients that includes:
  • waist circumference or body mass index (BMI), or both
  • blood pressure
  • complete blood count (CBC)
  • electrolytes
  • blood urea nitrogen (BUN) and creatinine
  • liver function tests (LFTs)
  • fasting glucose
  • fasting lipid profile.

In addition, medical history, cigarette smoking status, alcohol intake, and family history of cardiovascular disease, cerebro­vascular disease, hypertension, dyslip­idemia, and diabetes mellitus should be documented. Rule out pregnancy in women of childbearing potential.² The Figure describes monitoring parameters based on selected agent.

Agent-specific monitoring
Lithium. Patients beginning lithium ther­apy should undergo thyroid function testing and, for patients age >40, ECG mon­itoring. Educate patients about potential side effects of lithium, signs and symptoms of lithium toxicity, and the importance of avoiding dehydration. Adding or changing certain medications could elevate the serum lithium level (eg, diuretics, angiotensin-converting enzyme [ACE]-inhibitors, nonsteroidal anti-inflammatory drugs [NSAIDs], COX-2 inhibitors).

Lithium can cause weight gain and adverse effects in several organ systems, including:
  • gastrointestinal (GI) (nausea, vomit­ing, abdominal pain, loss of appetite, diarrhea)
  • renal (nephrogenic diabetes insipidus, tubulointerstitial renal disease)
  • neurologic (tremors, cognitive dulling, raised intracranial pressure)
  • endocrine (thyroid and parathyroid dysfunction)
  • cardiac (benign electrocardiographic changes, conduction abnormalities)
  • dermatologic (acne, psoriasis, hair loss)
  • hematologic (benign leukocytosis).

Lithium has a narrow therapeutic index (0.5 to 1.2 mEq/L), which means that small changes in the serum level can result in therapeutic inefficacy or toxic­ity. Lithium toxicity can cause irreversible organ damage or death. Serum lithium levels, symptomatic response, emergence and evolution of adverse drug reactions (ADRs), and the recognition of patient risk factors for toxicity can help guide dosing. From a safety monitoring viewpoint, lith­ium toxicity, renal and endocrine adverse effects, and potential drug interactions are foremost concerns.

Lithium usually is started at a low, divided dosages to minimize side effects, and titrated according to response. Check lithium levels before and after each dose increase. Serum levels reach steady state 5 days after dosage adjustment, but might need to be checked sooner if a rapid increase is necessary, such as when treat­ing acute mania, or if you suspect toxicity.

If the patient has renal insufficiency, it may take longer for the lithium to reach steady state; therefore, delaying a blood level beyond 5 days may be necessary to gauge a true steady state. Also, anytime a medication that interferes with lithium renal elimination, such as diuretics, ACE inhibitors, NSAIDs, COX-2 inhibitors, is added or the dosage is changed, a new lithium level will need to be obtained to reassess the level in 5 days, assuming adequate renal function. In general, renal function and thyroid function should be evaluated once or twice during the first 6 months of lithium treatment.

Subsequently, renal and thyroid func­tion can be checked every 6 months to 1 year in stable patients or when clinically indicated. Check a patient’s weight after 6 months of therapy, then at least annually.²

Valproic acid (VPA) and its derivatives. The most important initial monitoring for VPA therapy includes LFTs and CBC. Before initiating VPA treatment, take a medical history, with special attention to hepatic, hematologic, and bleeding abnor­malities. Therapeutic blood monitoring can be conducted once steady state is achieved and as clinically necessary thereafter.

VPA can be administered at an ini­tial starting dosage of 20 to 30 mg/kg/d in inpatients. In outpatients it is given in low, divided doses or as once-daily dosing using an extended-release formulation to minimize GI and neurologic toxicity and titrated every few days. Target serum level is 50 to 125 μg/mL.

Side effects of VPA include GI distress (eg, anorexia, nausea, dyspepsia, vomiting, diarrhea), hematologic effects (reversible leukopenia, thrombocytopenia), hair loss, weight gain, tremor, hepatic effects (benign LFT elevations, hepatotoxicity), osteoporo­sis, and sedation. Patients with prior or cur­rent hepatic disease may be at greater risk for hepatotoxicity. There is an association between VPA and polycystic ovarian syn­drome. Rare, idiosyncratic, but potentially fatal adverse events with valproate include irreversible hepatic failure, hemorrhagic pancreatitis, and agranulocytosis.

Older monitoring standards indicated taking LFTs and CBC every 6 months and serum VPA level as clinically indicated. According to ISBD guidelines, weight, CBC, LFTs, and menstrual history should be monitored every 3 months for the first year and then annually; blood pressure, bone status (densitometry), fasting glu­cose, and fasting lipids should be moni­tored only in patients with related risk factors. Routine ammonia levels are not recommended but might be indicated if a patient has sudden mental status changes or change in condition.²

Carbamazepine and oxcarbazepine. The most important initial monitoring for car­bamazepine therapy includes LFTs, renal function, electrolytes, and CBC. Before treatment, take a medical history, with special emphasis on history of blood dys­crasias or liver disease. After initiating car­bamazepine, CBC, LFTs, electrolytes, and renal function should be done monthly for 3 months, then repeated annually.

Carbamazepine is a substrate and an inducer of the cytochrome P450 (CYP) system, so it can reduce levels of many other drugs including other antiepileptics, warfarin, and oral contraceptives. Serum level of carbamazepine can be measured at trough after 5 days, with a target level of 4 to 12 μg/mL. Two levels should be drawn, 4 weeks apart, to establish thera­peutic dosage secondary to autoinduction of the CYP450 system.²

As many as one-half of patients experi­ence side effects with carbamazepine. The most common side effects include fatigue, nausea, and neurologic symptoms (dip­lopia, blurred vision, and ataxia). Less frequent side effects include skin rashes, leukopenia, liver enzyme elevations, thrombocytopenia, hyponatremia, and hypo-osmolality. Rare, potentially fatal side effects include agranulocytosis, aplas­tic anemia, thrombocytopenia, hepatic failure, and exfoliative dermatitis (eg, Stevens-Johnson syndrome).

Patients of Asian descent who are taking carbamazepine should undergo genetic testing for the HLA-B*1502 enzyme because persons with this allele are at higher risk of developing Stevens-Johnson syndrome. Also, patients should be edu­cated about the signs and symptoms of these rare adverse reactions so that medi­cal treatment is not delayed should these adverse events present.

Lamotrigine does not require further lab­oratory monitoring beyond the initial rec­ommended workup. The most important variables to consider are interactions with other medications (especially other antiep­ileptics, such as VPA and carbamazepine) and observing for rash. Titration takes several weeks to minimize risk of develop­ing a rash.² Similar to carbamazepine, the patient should be educated on the signs and symptoms of exfoliative dermatitis (eg, Stevens-Johnson syndrome) so that medical treatment is sought out should this reaction occur.

Atypical antipsychotics. Baseline workup includes the general monitoring param­eters described above. Atypical anti­psychotics have a lower incidence of extrapyramidal side effects than typical antipsychotics, but are associated with an increased risk of metabolic complications. Other major ADRs to consider are cardiac effects and hyperprolactinemia; clinicians should therefore inquire about a personal or family history of cardiac problems, including congenital long QT syndrome. Patients should be screened for any medi­cations that can prolong the QTc interval or interact with the metabolism of medica­tions known to cause QTc prolongation.

Measure weight monthly for the first 3 months, then every 3 months to monitor for metabolic side effects during ongoing treatment. Obtain blood pressure and fast­ing glucose every 3 months for the first year, then annually. Repeat a fasting lipid profile 3 months after initiating treatment, then annually. Cardiac effects and prolac­tin levels can be monitored as needed if clinically indicated.²

CASE CONTINUED
You discuss with Ms. W choices of a mood sta­bilizing agent to treat her bipolar II disorder; she agrees to start lithium. Before initiating treatment, you obtain her weight (and calcu­late her BMI), blood pressure, CBC, electrolyte levels, BUN and creatinine levels, liver func­tion tests, fasting glucose, fasting lipid profile, and thyroid panel. You also review her medi­cal history, lifestyle factors (cigarette smok­ing status, alcohol intake), and family history. A urine pregnancy screen is negative. The pharmacist assists in screening for potential drug-drug interactions, including over-the-counter medications that Ms. W occasionally takes as needed. She is counseled on the use of NSAIDS because these drugs can increase the lithium level.

Ms. W tolerates and responds well to lith­ium. No further dosing recommendations are made, based on clinical response. You measure her weight at 6 months, then annu­ally. Renal function and thyroid function are monitored at 3 and 6 months after lithium is initiated, and then annually. One year after starting lithium, she continues to tolerate the medication and has minimal metabolic side effects.

Related Resources
• McInnis MG. Lithium for bipolar disorder: A re-emerging treatment for mood instability. Current Psychiatry. 2014; 13(6):38-44.
• Stahl SM. Stahl’s illustrated mood stabilizers. New York, NY: Cambridge University Press; 2009.

Drug Brand Names
Carbamazepine • Tegretol      Valproic acid • Depacon, Depakote
Lamotrigine • Lamictal           Warfarin • Coumadin
Lithium • Lithobid, Eskalith
Oxcarbazepine • Trileptal

Disclosure
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Ms. W, age 27, presents with a chief concern of “depression.” She describes a history of several hypo­manic episodes as well as the current depres­sive episode, prompting a bipolar II disorder diagnosis. She is naïve to all psychotropics. You plan to initiate a mood-stabilizing agent. What would you include in your initial workup before starting treatment and how would you monitor her as she continues treatment?

Mood stabilizers are employed to treat bipolar spectrum disorders (bipolar I, bipolar II, and cyclothymic disorder) and schizoaffective disorder, bipolar type. Some evidence suggests that mood stabiliz­ers also can be used for treatment-resistant depressive disorders and borderline per­sonality disorder.¹ Mood stabilizers include lithium, valproate, carbamazepine, oxcar­bazepine, and lamotrigine.^2-5

This review focuses on applications and monitoring of mood stabilizers for bipolar I and II disorders. We also will briefly review atypical antipsychotics because they also are used to treat bipo­lar spectrum disorders (see the September 2013 issue of Current Psychiatry at CurrentPsychiatry.com for a more detailed article on monitoring of antipsychotics).⁶

There are several well-researched guidelines used to guide clinical prac­tice.^2-5 Many guidelines recommend base­line and routine monitoring parameters based on the characteristics of the agent used. However, the International Society for Bipolar Disorders (ISBD) guidelines highlight the importance of monitoring medical comorbidities, which are common among patients with bipolar disorder and can affect pharmacotherapy and clinical outcomes. These recommendations are similar to metabolic monitoring guidelines for antipsychotics.⁵

Reviews of therapeutic monitoring show that only one-third to one-half of patien

taking a mood stabilizer are appropriately monitored. Poor adher­ence to guideline recommendations often is observed because of patients’ lack of insight or medication adherence and because psychiatric care generally is segre­gated from other medical care.7-9

Baseline testing
The ISBD guidelines recommend an initial workup for all patients that includes:
  • waist circumference or body mass index (BMI), or both
  • blood pressure
  • complete blood count (CBC)
  • electrolytes
  • blood urea nitrogen (BUN) and creatinine
  • liver function tests (LFTs)
  • fasting glucose
  • fasting lipid profile.

In addition, medical history, cigarette smoking status, alcohol intake, and family history of cardiovascular disease, cerebro­vascular disease, hypertension, dyslip­idemia, and diabetes mellitus should be documented. Rule out pregnancy in women of childbearing potential.² The Figure describes monitoring parameters based on selected agent.

Agent-specific monitoring
Lithium. Patients beginning lithium ther­apy should undergo thyroid function testing and, for patients age >40, ECG mon­itoring. Educate patients about potential side effects of lithium, signs and symptoms of lithium toxicity, and the importance of avoiding dehydration. Adding or changing certain medications could elevate the serum lithium level (eg, diuretics, angiotensin-converting enzyme [ACE]-inhibitors, nonsteroidal anti-inflammatory drugs [NSAIDs], COX-2 inhibitors).

Lithium can cause weight gain and adverse effects in several organ systems, including:
  • gastrointestinal (GI) (nausea, vomit­ing, abdominal pain, loss of appetite, diarrhea)
  • renal (nephrogenic diabetes insipidus, tubulointerstitial renal disease)
  • neurologic (tremors, cognitive dulling, raised intracranial pressure)
  • endocrine (thyroid and parathyroid dysfunction)
  • cardiac (benign electrocardiographic changes, conduction abnormalities)
  • dermatologic (acne, psoriasis, hair loss)
  • hematologic (benign leukocytosis).

Lithium has a narrow therapeutic index (0.5 to 1.2 mEq/L), which means that small changes in the serum level can result in therapeutic inefficacy or toxic­ity. Lithium toxicity can cause irreversible organ damage or death. Serum lithium levels, symptomatic response, emergence and evolution of adverse drug reactions (ADRs), and the recognition of patient risk factors for toxicity can help guide dosing. From a safety monitoring viewpoint, lith­ium toxicity, renal and endocrine adverse effects, and potential drug interactions are foremost concerns.

Lithium usually is started at a low, divided dosages to minimize side effects, and titrated according to response. Check lithium levels before and after each dose increase. Serum levels reach steady state 5 days after dosage adjustment, but might need to be checked sooner if a rapid increase is necessary, such as when treat­ing acute mania, or if you suspect toxicity.

If the patient has renal insufficiency, it may take longer for the lithium to reach steady state; therefore, delaying a blood level beyond 5 days may be necessary to gauge a true steady state. Also, anytime a medication that interferes with lithium renal elimination, such as diuretics, ACE inhibitors, NSAIDs, COX-2 inhibitors, is added or the dosage is changed, a new lithium level will need to be obtained to reassess the level in 5 days, assuming adequate renal function. In general, renal function and thyroid function should be evaluated once or twice during the first 6 months of lithium treatment.

Subsequently, renal and thyroid func­tion can be checked every 6 months to 1 year in stable patients or when clinically indicated. Check a patient’s weight after 6 months of therapy, then at least annually.²

Valproic acid (VPA) and its derivatives. The most important initial monitoring for VPA therapy includes LFTs and CBC. Before initiating VPA treatment, take a medical history, with special attention to hepatic, hematologic, and bleeding abnor­malities. Therapeutic blood monitoring can be conducted once steady state is achieved and as clinically necessary thereafter.

VPA can be administered at an ini­tial starting dosage of 20 to 30 mg/kg/d in inpatients. In outpatients it is given in low, divided doses or as once-daily dosing using an extended-release formulation to minimize GI and neurologic toxicity and titrated every few days. Target serum level is 50 to 125 μg/mL.

Side effects of VPA include GI distress (eg, anorexia, nausea, dyspepsia, vomiting, diarrhea), hematologic effects (reversible leukopenia, thrombocytopenia), hair loss, weight gain, tremor, hepatic effects (benign LFT elevations, hepatotoxicity), osteoporo­sis, and sedation. Patients with prior or cur­rent hepatic disease may be at greater risk for hepatotoxicity. There is an association between VPA and polycystic ovarian syn­drome. Rare, idiosyncratic, but potentially fatal adverse events with valproate include irreversible hepatic failure, hemorrhagic pancreatitis, and agranulocytosis.

Older monitoring standards indicated taking LFTs and CBC every 6 months and serum VPA level as clinically indicated. According to ISBD guidelines, weight, CBC, LFTs, and menstrual history should be monitored every 3 months for the first year and then annually; blood pressure, bone status (densitometry), fasting glu­cose, and fasting lipids should be moni­tored only in patients with related risk factors. Routine ammonia levels are not recommended but might be indicated if a patient has sudden mental status changes or change in condition.²

Carbamazepine and oxcarbazepine. The most important initial monitoring for car­bamazepine therapy includes LFTs, renal function, electrolytes, and CBC. Before treatment, take a medical history, with special emphasis on history of blood dys­crasias or liver disease. After initiating car­bamazepine, CBC, LFTs, electrolytes, and renal function should be done monthly for 3 months, then repeated annually.

Carbamazepine is a substrate and an inducer of the cytochrome P450 (CYP) system, so it can reduce levels of many other drugs including other antiepileptics, warfarin, and oral contraceptives. Serum level of carbamazepine can be measured at trough after 5 days, with a target level of 4 to 12 μg/mL. Two levels should be drawn, 4 weeks apart, to establish thera­peutic dosage secondary to autoinduction of the CYP450 system.²

As many as one-half of patients experi­ence side effects with carbamazepine. The most common side effects include fatigue, nausea, and neurologic symptoms (dip­lopia, blurred vision, and ataxia). Less frequent side effects include skin rashes, leukopenia, liver enzyme elevations, thrombocytopenia, hyponatremia, and hypo-osmolality. Rare, potentially fatal side effects include agranulocytosis, aplas­tic anemia, thrombocytopenia, hepatic failure, and exfoliative dermatitis (eg, Stevens-Johnson syndrome).

Patients of Asian descent who are taking carbamazepine should undergo genetic testing for the HLA-B*1502 enzyme because persons with this allele are at higher risk of developing Stevens-Johnson syndrome. Also, patients should be edu­cated about the signs and symptoms of these rare adverse reactions so that medi­cal treatment is not delayed should these adverse events present.

Lamotrigine does not require further lab­oratory monitoring beyond the initial rec­ommended workup. The most important variables to consider are interactions with other medications (especially other antiep­ileptics, such as VPA and carbamazepine) and observing for rash. Titration takes several weeks to minimize risk of develop­ing a rash.² Similar to carbamazepine, the patient should be educated on the signs and symptoms of exfoliative dermatitis (eg, Stevens-Johnson syndrome) so that medical treatment is sought out should this reaction occur.

Atypical antipsychotics. Baseline workup includes the general monitoring param­eters described above. Atypical anti­psychotics have a lower incidence of extrapyramidal side effects than typical antipsychotics, but are associated with an increased risk of metabolic complications. Other major ADRs to consider are cardiac effects and hyperprolactinemia; clinicians should therefore inquire about a personal or family history of cardiac problems, including congenital long QT syndrome. Patients should be screened for any medi­cations that can prolong the QTc interval or interact with the metabolism of medica­tions known to cause QTc prolongation.

Measure weight monthly for the first 3 months, then every 3 months to monitor for metabolic side effects during ongoing treatment. Obtain blood pressure and fast­ing glucose every 3 months for the first year, then annually. Repeat a fasting lipid profile 3 months after initiating treatment, then annually. Cardiac effects and prolac­tin levels can be monitored as needed if clinically indicated.²

CASE CONTINUED
You discuss with Ms. W choices of a mood sta­bilizing agent to treat her bipolar II disorder; she agrees to start lithium. Before initiating treatment, you obtain her weight (and calcu­late her BMI), blood pressure, CBC, electrolyte levels, BUN and creatinine levels, liver func­tion tests, fasting glucose, fasting lipid profile, and thyroid panel. You also review her medi­cal history, lifestyle factors (cigarette smok­ing status, alcohol intake), and family history. A urine pregnancy screen is negative. The pharmacist assists in screening for potential drug-drug interactions, including over-the-counter medications that Ms. W occasionally takes as needed. She is counseled on the use of NSAIDS because these drugs can increase the lithium level.

Ms. W tolerates and responds well to lith­ium. No further dosing recommendations are made, based on clinical response. You measure her weight at 6 months, then annu­ally. Renal function and thyroid function are monitored at 3 and 6 months after lithium is initiated, and then annually. One year after starting lithium, she continues to tolerate the medication and has minimal metabolic side effects.

Related Resources
• McInnis MG. Lithium for bipolar disorder: A re-emerging treatment for mood instability. Current Psychiatry. 2014; 13(6):38-44.
• Stahl SM. Stahl’s illustrated mood stabilizers. New York, NY: Cambridge University Press; 2009.

Drug Brand Names
Carbamazepine • Tegretol      Valproic acid • Depacon, Depakote
Lamotrigine • Lamictal           Warfarin • Coumadin
Lithium • Lithobid, Eskalith
Oxcarbazepine • Trileptal

Disclosure
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Ms. W, age 27, presents with a chief concern of “depression.” She describes a history of several hypo­manic episodes as well as the current depres­sive episode, prompting a bipolar II disorder diagnosis. She is naïve to all psychotropics. You plan to initiate a mood-stabilizing agent. What would you include in your initial workup before starting treatment and how would you monitor her as she continues treatment?

Mood stabilizers are employed to treat bipolar spectrum disorders (bipolar I, bipolar II, and cyclothymic disorder) and schizoaffective disorder, bipolar type. Some evidence suggests that mood stabiliz­ers also can be used for treatment-resistant depressive disorders and borderline per­sonality disorder.¹ Mood stabilizers include lithium, valproate, carbamazepine, oxcar­bazepine, and lamotrigine.^2-5

This review focuses on applications and monitoring of mood stabilizers for bipolar I and II disorders. We also will briefly review atypical antipsychotics because they also are used to treat bipo­lar spectrum disorders (see the September 2013 issue of Current Psychiatry at CurrentPsychiatry.com for a more detailed article on monitoring of antipsychotics).⁶

There are several well-researched guidelines used to guide clinical prac­tice.^2-5 Many guidelines recommend base­line and routine monitoring parameters based on the characteristics of the agent used. However, the International Society for Bipolar Disorders (ISBD) guidelines highlight the importance of monitoring medical comorbidities, which are common among patients with bipolar disorder and can affect pharmacotherapy and clinical outcomes. These recommendations are similar to metabolic monitoring guidelines for antipsychotics.⁵

Reviews of therapeutic monitoring show that only one-third to one-half of patien

taking a mood stabilizer are appropriately monitored. Poor adher­ence to guideline recommendations often is observed because of patients’ lack of insight or medication adherence and because psychiatric care generally is segre­gated from other medical care.7-9

Baseline testing
The ISBD guidelines recommend an initial workup for all patients that includes:
  • waist circumference or body mass index (BMI), or both
  • blood pressure
  • complete blood count (CBC)
  • electrolytes
  • blood urea nitrogen (BUN) and creatinine
  • liver function tests (LFTs)
  • fasting glucose
  • fasting lipid profile.

In addition, medical history, cigarette smoking status, alcohol intake, and family history of cardiovascular disease, cerebro­vascular disease, hypertension, dyslip­idemia, and diabetes mellitus should be documented. Rule out pregnancy in women of childbearing potential.² The Figure describes monitoring parameters based on selected agent.

Agent-specific monitoring
Lithium. Patients beginning lithium ther­apy should undergo thyroid function testing and, for patients age >40, ECG mon­itoring. Educate patients about potential side effects of lithium, signs and symptoms of lithium toxicity, and the importance of avoiding dehydration. Adding or changing certain medications could elevate the serum lithium level (eg, diuretics, angiotensin-converting enzyme [ACE]-inhibitors, nonsteroidal anti-inflammatory drugs [NSAIDs], COX-2 inhibitors).

Lithium can cause weight gain and adverse effects in several organ systems, including:
  • gastrointestinal (GI) (nausea, vomit­ing, abdominal pain, loss of appetite, diarrhea)
  • renal (nephrogenic diabetes insipidus, tubulointerstitial renal disease)
  • neurologic (tremors, cognitive dulling, raised intracranial pressure)
  • endocrine (thyroid and parathyroid dysfunction)
  • cardiac (benign electrocardiographic changes, conduction abnormalities)
  • dermatologic (acne, psoriasis, hair loss)
  • hematologic (benign leukocytosis).

Lithium has a narrow therapeutic index (0.5 to 1.2 mEq/L), which means that small changes in the serum level can result in therapeutic inefficacy or toxic­ity. Lithium toxicity can cause irreversible organ damage or death. Serum lithium levels, symptomatic response, emergence and evolution of adverse drug reactions (ADRs), and the recognition of patient risk factors for toxicity can help guide dosing. From a safety monitoring viewpoint, lith­ium toxicity, renal and endocrine adverse effects, and potential drug interactions are foremost concerns.

Lithium usually is started at a low, divided dosages to minimize side effects, and titrated according to response. Check lithium levels before and after each dose increase. Serum levels reach steady state 5 days after dosage adjustment, but might need to be checked sooner if a rapid increase is necessary, such as when treat­ing acute mania, or if you suspect toxicity.

If the patient has renal insufficiency, it may take longer for the lithium to reach steady state; therefore, delaying a blood level beyond 5 days may be necessary to gauge a true steady state. Also, anytime a medication that interferes with lithium renal elimination, such as diuretics, ACE inhibitors, NSAIDs, COX-2 inhibitors, is added or the dosage is changed, a new lithium level will need to be obtained to reassess the level in 5 days, assuming adequate renal function. In general, renal function and thyroid function should be evaluated once or twice during the first 6 months of lithium treatment.

Subsequently, renal and thyroid func­tion can be checked every 6 months to 1 year in stable patients or when clinically indicated. Check a patient’s weight after 6 months of therapy, then at least annually.²

Valproic acid (VPA) and its derivatives. The most important initial monitoring for VPA therapy includes LFTs and CBC. Before initiating VPA treatment, take a medical history, with special attention to hepatic, hematologic, and bleeding abnor­malities. Therapeutic blood monitoring can be conducted once steady state is achieved and as clinically necessary thereafter.

VPA can be administered at an ini­tial starting dosage of 20 to 30 mg/kg/d in inpatients. In outpatients it is given in low, divided doses or as once-daily dosing using an extended-release formulation to minimize GI and neurologic toxicity and titrated every few days. Target serum level is 50 to 125 μg/mL.

Side effects of VPA include GI distress (eg, anorexia, nausea, dyspepsia, vomiting, diarrhea), hematologic effects (reversible leukopenia, thrombocytopenia), hair loss, weight gain, tremor, hepatic effects (benign LFT elevations, hepatotoxicity), osteoporo­sis, and sedation. Patients with prior or cur­rent hepatic disease may be at greater risk for hepatotoxicity. There is an association between VPA and polycystic ovarian syn­drome. Rare, idiosyncratic, but potentially fatal adverse events with valproate include irreversible hepatic failure, hemorrhagic pancreatitis, and agranulocytosis.

Older monitoring standards indicated taking LFTs and CBC every 6 months and serum VPA level as clinically indicated. According to ISBD guidelines, weight, CBC, LFTs, and menstrual history should be monitored every 3 months for the first year and then annually; blood pressure, bone status (densitometry), fasting glu­cose, and fasting lipids should be moni­tored only in patients with related risk factors. Routine ammonia levels are not recommended but might be indicated if a patient has sudden mental status changes or change in condition.²

Carbamazepine and oxcarbazepine. The most important initial monitoring for car­bamazepine therapy includes LFTs, renal function, electrolytes, and CBC. Before treatment, take a medical history, with special emphasis on history of blood dys­crasias or liver disease. After initiating car­bamazepine, CBC, LFTs, electrolytes, and renal function should be done monthly for 3 months, then repeated annually.

Carbamazepine is a substrate and an inducer of the cytochrome P450 (CYP) system, so it can reduce levels of many other drugs including other antiepileptics, warfarin, and oral contraceptives. Serum level of carbamazepine can be measured at trough after 5 days, with a target level of 4 to 12 μg/mL. Two levels should be drawn, 4 weeks apart, to establish thera­peutic dosage secondary to autoinduction of the CYP450 system.²

As many as one-half of patients experi­ence side effects with carbamazepine. The most common side effects include fatigue, nausea, and neurologic symptoms (dip­lopia, blurred vision, and ataxia). Less frequent side effects include skin rashes, leukopenia, liver enzyme elevations, thrombocytopenia, hyponatremia, and hypo-osmolality. Rare, potentially fatal side effects include agranulocytosis, aplas­tic anemia, thrombocytopenia, hepatic failure, and exfoliative dermatitis (eg, Stevens-Johnson syndrome).

Patients of Asian descent who are taking carbamazepine should undergo genetic testing for the HLA-B*1502 enzyme because persons with this allele are at higher risk of developing Stevens-Johnson syndrome. Also, patients should be edu­cated about the signs and symptoms of these rare adverse reactions so that medi­cal treatment is not delayed should these adverse events present.

Lamotrigine does not require further lab­oratory monitoring beyond the initial rec­ommended workup. The most important variables to consider are interactions with other medications (especially other antiep­ileptics, such as VPA and carbamazepine) and observing for rash. Titration takes several weeks to minimize risk of develop­ing a rash.² Similar to carbamazepine, the patient should be educated on the signs and symptoms of exfoliative dermatitis (eg, Stevens-Johnson syndrome) so that medical treatment is sought out should this reaction occur.

Atypical antipsychotics. Baseline workup includes the general monitoring param­eters described above. Atypical anti­psychotics have a lower incidence of extrapyramidal side effects than typical antipsychotics, but are associated with an increased risk of metabolic complications. Other major ADRs to consider are cardiac effects and hyperprolactinemia; clinicians should therefore inquire about a personal or family history of cardiac problems, including congenital long QT syndrome. Patients should be screened for any medi­cations that can prolong the QTc interval or interact with the metabolism of medica­tions known to cause QTc prolongation.

Measure weight monthly for the first 3 months, then every 3 months to monitor for metabolic side effects during ongoing treatment. Obtain blood pressure and fast­ing glucose every 3 months for the first year, then annually. Repeat a fasting lipid profile 3 months after initiating treatment, then annually. Cardiac effects and prolac­tin levels can be monitored as needed if clinically indicated.²

CASE CONTINUED
You discuss with Ms. W choices of a mood sta­bilizing agent to treat her bipolar II disorder; she agrees to start lithium. Before initiating treatment, you obtain her weight (and calcu­late her BMI), blood pressure, CBC, electrolyte levels, BUN and creatinine levels, liver func­tion tests, fasting glucose, fasting lipid profile, and thyroid panel. You also review her medi­cal history, lifestyle factors (cigarette smok­ing status, alcohol intake), and family history. A urine pregnancy screen is negative. The pharmacist assists in screening for potential drug-drug interactions, including over-the-counter medications that Ms. W occasionally takes as needed. She is counseled on the use of NSAIDS because these drugs can increase the lithium level.

Ms. W tolerates and responds well to lith­ium. No further dosing recommendations are made, based on clinical response. You measure her weight at 6 months, then annu­ally. Renal function and thyroid function are monitored at 3 and 6 months after lithium is initiated, and then annually. One year after starting lithium, she continues to tolerate the medication and has minimal metabolic side effects.

Related Resources
• McInnis MG. Lithium for bipolar disorder: A re-emerging treatment for mood instability. Current Psychiatry. 2014; 13(6):38-44.
• Stahl SM. Stahl’s illustrated mood stabilizers. New York, NY: Cambridge University Press; 2009.

Drug Brand Names
Carbamazepine • Tegretol      Valproic acid • Depacon, Depakote
Lamotrigine • Lamictal           Warfarin • Coumadin
Lithium • Lithobid, Eskalith
Oxcarbazepine • Trileptal

Disclosure
The authors report no financial relationships with any company whose products are mentioned in this article or with manufacturers of competing products.

Mr. D, age 30, has a 12-year history of schizophrenia and is experiencing worsening auditory hallucinations despite reported medication adherence. He has been taking clozapine, maintenance dos­ages 500 to 700 mg/d, for 4 years and smokes 2 packs of cigarettes a day. When Mr. D is admitted to a nonsmoking inpatient psychi­atric facility, he receives nicotine transdermal patches, 21 mg/d, for nicotine withdrawal. Mr. D’s most recent outpatient clozapine dosage, 700 mg/d, is resumed. All laboratory tests, including complete blood count with differen­tial, are within normal limits at admission.

Five days later Mr. D is tachycardic with a heart rate of 109 beats per minute. When assessing Mr. D, we notice he has alogia and that, when he does speak, his speech is slowed with a 4 to 5 second delay in response. He also appears sedated. We observe occasional mild jerking of his shoul­der and lower legs.

Mr. D reports that his auditory hallucina­tions have lessened since his admission, but complains of difficulty remembering infor­mation and feeling tired during the day. The treatment team suspects clozapine toxicity; his trough clozapine level is 1,350 ng/mL (therapeutic range, 350 to 1,000 ng/mL).

It is well documented that cigarette smoke can induce cytochrome P450 (CYP) isoenzymes, specifically CYP1A1, CYP1A2, and CYP2E1. Because clozapine is primarily metabolized by CYP1A2 (approximately 70%), smoking can induce clozapine metabolism and abruptly stopping smoking can increase clozapine levels.¹ The polycyclic aromatic hydrocarbons, not the nicotine, found in cigarettes are thought to be responsible for CYP1A2 induction; therefore, use of a nicotine replacement product did not prevent the increase in Mr. D’s clozapine levels.
 

Examining the evidence

Meyer¹ evaluated clozapine levels before and after implementation of a hospital-wide smoking ban (N = 11). Clozapine dosages were not adjusted at the time of the smoking ban, which resulted in a mean 72% increase in clozapine levels after a min­imum of 2 weeks as nonsmokers. Even after eliminating 2 outliers, the mean increase in clozapine levels was 36.1%. Murayama-Sung et al² reported a statistically signifi­cant increase in the level of clozapine (46%, P = .004) and the level of norclozapine (23%, P = .02) after a hospital-wide smoking ban was instituted (N = 14). However, the pre-change and post-change in the ratio of clo­zapine to norclozapine level was not found to be statistically significant. Haslemo et al³ found that smoking as few as 7 to 12 cigarettes a day was sufficient for maximum induction of CYP1A2. Because Mr. D was smoking 2 packs of cigarettes a day (40 cigarettes) with an clozapine dosage 700 mg/d as an outpatient, he likely expe­rienced significant induction of clozapine metabolism through CYP1A2, which was no longer present when he stopped smoking.

Therapeutic clozapine concentrations are typically above 350 and 420 ng/mL.⁴ Concentrations >700 ng/mL are associ­ated with increased adverse effects, but generally are not associated with a higher response; levels >900 ng/mL have been associated with toxicity.⁴ Clozapine-treated patients on a stable dosage who smoke can experience clozapine-related adverse effects after admission to a smoke-free facility secondary to an increase in the clozapine concentration (Table 1).⁴

Faber and Fuhr⁶ recommended reduc­ing the dosage of a CYP1A2 substrate medication, such as clozapine, olanzap­ine, or theophylline, by 10% each day until the dosage has been reduced by 40% in patients who stop smoking. Lowe and Ackman⁵ proposed reducing the clozapine dosage by 30% to 40% to achieve a pre-cessation serum concentration at 1 week. For Mr. D, this would mean decreasing the clozapine dosage to 425 to 500 mg/d.

Assuming that Mr. D’s clozapine dosage is decreased during his hospitalization and that he resumes smoking after discharge, it is likely the dosage will need to be increased. It may take several weeks to see maximal induction, because new CYP enzymes need to be synthesized when the patient resumes smoking.⁷ One recommendation is to increase the clozapine dosage by a factor of 1.5 over 2 to 4 weeks, with close monitoring of the clozapine concentration and adverse effects because this increase is approximate.⁷ Depending on when Mr. D’s follow-up appointment is scheduled, the practitioner may need to plan a dosage adjustment to prevent a decrease in his clozapine level caused by smoking to prevent a worsening of symptoms and rehospitalization.

This case emphasizes the importance of asking clozapine-treated patients about their smoking history when they are admit­ted to a smoke-free facility. For several reasons, >60% of patients with schizophre­nia smoke cigarettes⁸ (Table 2).^9-14 Patients who smoke and are on a stable dosage of clozapine might require a dos­age reduction when they are admitted to a smoke-free facility to avoid adverse effects. If the dosage is not adjusted, a patient may experience clozapine-induced adverse effects, such as tachycardia, sedation, and seizures. It is likely that patients such as Mr. D will experience fluctuation in the clo­zapine level and possibly changes in effi­cacy and tolerability transitioning between inpatient and outpatient settings if the dos­age is not adjusted.

Drug Brand Names
Clozapine • Clozaril             Theophylline • Theo-Dur
Olanzapine • Zyprexa

Disclosure
The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Mr. D, age 30, has a 12-year history of schizophrenia and is experiencing worsening auditory hallucinations despite reported medication adherence. He has been taking clozapine, maintenance dos­ages 500 to 700 mg/d, for 4 years and smokes 2 packs of cigarettes a day. When Mr. D is admitted to a nonsmoking inpatient psychi­atric facility, he receives nicotine transdermal patches, 21 mg/d, for nicotine withdrawal. Mr. D’s most recent outpatient clozapine dosage, 700 mg/d, is resumed. All laboratory tests, including complete blood count with differen­tial, are within normal limits at admission.

Five days later Mr. D is tachycardic with a heart rate of 109 beats per minute. When assessing Mr. D, we notice he has alogia and that, when he does speak, his speech is slowed with a 4 to 5 second delay in response. He also appears sedated. We observe occasional mild jerking of his shoul­der and lower legs.

Mr. D reports that his auditory hallucina­tions have lessened since his admission, but complains of difficulty remembering infor­mation and feeling tired during the day. The treatment team suspects clozapine toxicity; his trough clozapine level is 1,350 ng/mL (therapeutic range, 350 to 1,000 ng/mL).

It is well documented that cigarette smoke can induce cytochrome P450 (CYP) isoenzymes, specifically CYP1A1, CYP1A2, and CYP2E1. Because clozapine is primarily metabolized by CYP1A2 (approximately 70%), smoking can induce clozapine metabolism and abruptly stopping smoking can increase clozapine levels.¹ The polycyclic aromatic hydrocarbons, not the nicotine, found in cigarettes are thought to be responsible for CYP1A2 induction; therefore, use of a nicotine replacement product did not prevent the increase in Mr. D’s clozapine levels.
 

Examining the evidence

Meyer¹ evaluated clozapine levels before and after implementation of a hospital-wide smoking ban (N = 11). Clozapine dosages were not adjusted at the time of the smoking ban, which resulted in a mean 72% increase in clozapine levels after a min­imum of 2 weeks as nonsmokers. Even after eliminating 2 outliers, the mean increase in clozapine levels was 36.1%. Murayama-Sung et al² reported a statistically signifi­cant increase in the level of clozapine (46%, P = .004) and the level of norclozapine (23%, P = .02) after a hospital-wide smoking ban was instituted (N = 14). However, the pre-change and post-change in the ratio of clo­zapine to norclozapine level was not found to be statistically significant. Haslemo et al³ found that smoking as few as 7 to 12 cigarettes a day was sufficient for maximum induction of CYP1A2. Because Mr. D was smoking 2 packs of cigarettes a day (40 cigarettes) with an clozapine dosage 700 mg/d as an outpatient, he likely expe­rienced significant induction of clozapine metabolism through CYP1A2, which was no longer present when he stopped smoking.

Therapeutic clozapine concentrations are typically above 350 and 420 ng/mL.⁴ Concentrations >700 ng/mL are associ­ated with increased adverse effects, but generally are not associated with a higher response; levels >900 ng/mL have been associated with toxicity.⁴ Clozapine-treated patients on a stable dosage who smoke can experience clozapine-related adverse effects after admission to a smoke-free facility secondary to an increase in the clozapine concentration (Table 1).⁴

Faber and Fuhr⁶ recommended reduc­ing the dosage of a CYP1A2 substrate medication, such as clozapine, olanzap­ine, or theophylline, by 10% each day until the dosage has been reduced by 40% in patients who stop smoking. Lowe and Ackman⁵ proposed reducing the clozapine dosage by 30% to 40% to achieve a pre-cessation serum concentration at 1 week. For Mr. D, this would mean decreasing the clozapine dosage to 425 to 500 mg/d.

Assuming that Mr. D’s clozapine dosage is decreased during his hospitalization and that he resumes smoking after discharge, it is likely the dosage will need to be increased. It may take several weeks to see maximal induction, because new CYP enzymes need to be synthesized when the patient resumes smoking.⁷ One recommendation is to increase the clozapine dosage by a factor of 1.5 over 2 to 4 weeks, with close monitoring of the clozapine concentration and adverse effects because this increase is approximate.⁷ Depending on when Mr. D’s follow-up appointment is scheduled, the practitioner may need to plan a dosage adjustment to prevent a decrease in his clozapine level caused by smoking to prevent a worsening of symptoms and rehospitalization.

This case emphasizes the importance of asking clozapine-treated patients about their smoking history when they are admit­ted to a smoke-free facility. For several reasons, >60% of patients with schizophre­nia smoke cigarettes⁸ (Table 2).^9-14 Patients who smoke and are on a stable dosage of clozapine might require a dos­age reduction when they are admitted to a smoke-free facility to avoid adverse effects. If the dosage is not adjusted, a patient may experience clozapine-induced adverse effects, such as tachycardia, sedation, and seizures. It is likely that patients such as Mr. D will experience fluctuation in the clo­zapine level and possibly changes in effi­cacy and tolerability transitioning between inpatient and outpatient settings if the dos­age is not adjusted.

Drug Brand Names
Clozapine • Clozaril             Theophylline • Theo-Dur
Olanzapine • Zyprexa

Disclosure
The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Mr. D, age 30, has a 12-year history of schizophrenia and is experiencing worsening auditory hallucinations despite reported medication adherence. He has been taking clozapine, maintenance dos­ages 500 to 700 mg/d, for 4 years and smokes 2 packs of cigarettes a day. When Mr. D is admitted to a nonsmoking inpatient psychi­atric facility, he receives nicotine transdermal patches, 21 mg/d, for nicotine withdrawal. Mr. D’s most recent outpatient clozapine dosage, 700 mg/d, is resumed. All laboratory tests, including complete blood count with differen­tial, are within normal limits at admission.

Five days later Mr. D is tachycardic with a heart rate of 109 beats per minute. When assessing Mr. D, we notice he has alogia and that, when he does speak, his speech is slowed with a 4 to 5 second delay in response. He also appears sedated. We observe occasional mild jerking of his shoul­der and lower legs.

Mr. D reports that his auditory hallucina­tions have lessened since his admission, but complains of difficulty remembering infor­mation and feeling tired during the day. The treatment team suspects clozapine toxicity; his trough clozapine level is 1,350 ng/mL (therapeutic range, 350 to 1,000 ng/mL).

It is well documented that cigarette smoke can induce cytochrome P450 (CYP) isoenzymes, specifically CYP1A1, CYP1A2, and CYP2E1. Because clozapine is primarily metabolized by CYP1A2 (approximately 70%), smoking can induce clozapine metabolism and abruptly stopping smoking can increase clozapine levels.¹ The polycyclic aromatic hydrocarbons, not the nicotine, found in cigarettes are thought to be responsible for CYP1A2 induction; therefore, use of a nicotine replacement product did not prevent the increase in Mr. D’s clozapine levels.
 

Examining the evidence

Meyer¹ evaluated clozapine levels before and after implementation of a hospital-wide smoking ban (N = 11). Clozapine dosages were not adjusted at the time of the smoking ban, which resulted in a mean 72% increase in clozapine levels after a min­imum of 2 weeks as nonsmokers. Even after eliminating 2 outliers, the mean increase in clozapine levels was 36.1%. Murayama-Sung et al² reported a statistically signifi­cant increase in the level of clozapine (46%, P = .004) and the level of norclozapine (23%, P = .02) after a hospital-wide smoking ban was instituted (N = 14). However, the pre-change and post-change in the ratio of clo­zapine to norclozapine level was not found to be statistically significant. Haslemo et al³ found that smoking as few as 7 to 12 cigarettes a day was sufficient for maximum induction of CYP1A2. Because Mr. D was smoking 2 packs of cigarettes a day (40 cigarettes) with an clozapine dosage 700 mg/d as an outpatient, he likely expe­rienced significant induction of clozapine metabolism through CYP1A2, which was no longer present when he stopped smoking.

Therapeutic clozapine concentrations are typically above 350 and 420 ng/mL.⁴ Concentrations >700 ng/mL are associ­ated with increased adverse effects, but generally are not associated with a higher response; levels >900 ng/mL have been associated with toxicity.⁴ Clozapine-treated patients on a stable dosage who smoke can experience clozapine-related adverse effects after admission to a smoke-free facility secondary to an increase in the clozapine concentration (Table 1).⁴

Faber and Fuhr⁶ recommended reduc­ing the dosage of a CYP1A2 substrate medication, such as clozapine, olanzap­ine, or theophylline, by 10% each day until the dosage has been reduced by 40% in patients who stop smoking. Lowe and Ackman⁵ proposed reducing the clozapine dosage by 30% to 40% to achieve a pre-cessation serum concentration at 1 week. For Mr. D, this would mean decreasing the clozapine dosage to 425 to 500 mg/d.

Assuming that Mr. D’s clozapine dosage is decreased during his hospitalization and that he resumes smoking after discharge, it is likely the dosage will need to be increased. It may take several weeks to see maximal induction, because new CYP enzymes need to be synthesized when the patient resumes smoking.⁷ One recommendation is to increase the clozapine dosage by a factor of 1.5 over 2 to 4 weeks, with close monitoring of the clozapine concentration and adverse effects because this increase is approximate.⁷ Depending on when Mr. D’s follow-up appointment is scheduled, the practitioner may need to plan a dosage adjustment to prevent a decrease in his clozapine level caused by smoking to prevent a worsening of symptoms and rehospitalization.

This case emphasizes the importance of asking clozapine-treated patients about their smoking history when they are admit­ted to a smoke-free facility. For several reasons, >60% of patients with schizophre­nia smoke cigarettes⁸ (Table 2).^9-14 Patients who smoke and are on a stable dosage of clozapine might require a dos­age reduction when they are admitted to a smoke-free facility to avoid adverse effects. If the dosage is not adjusted, a patient may experience clozapine-induced adverse effects, such as tachycardia, sedation, and seizures. It is likely that patients such as Mr. D will experience fluctuation in the clo­zapine level and possibly changes in effi­cacy and tolerability transitioning between inpatient and outpatient settings if the dos­age is not adjusted.

Drug Brand Names
Clozapine • Clozaril             Theophylline • Theo-Dur
Olanzapine • Zyprexa

Disclosure
The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Mr. M, age 28, was given a diagnosis of schizophrenia 6 years ago after experiencing a psychotic break involving auditory hallucinations and paranoia. Olanzapine, 10 mg/d, relieved his symptoms, but he stopped taking the drug after gaining 40 pounds and developing diabetes mellitus. He had 2 other hospital admissions for acute psychosis and has taken at least 1 other medication, the name of which he can’t recall. Recently, Mr. M was involuntarily admitted to the psychiatric ward of his local hospital. His psychiatrist started aripiprazole, 10 mg/d, which was titrated to 30 mg/d. After 2 weeks he reports only a slight decrease in hallucinations. His mother is growing concerned about the effectiveness of this medication and wants to know if it’s time to consider another drug.

Time to onset of action of antipsychotic agents has been debated since at least 1970.¹ Supporters of the delayed-onset hypothesis assert that antipsychotics take weeks or months to show significant improvement of symptoms because of the need for depolarization block for efficacy.² Trials of 4 to 6 weeks often are recommended for patients before failure is declared,^3,4 and trials of this length or longer have proved useful for first-episode patients.^5-7 Recent studies suggest, however, that response is cumulative for chronically ill Practice Points

• Chronically ill and first-episode patients may respond differently to antipsychotics.
• In chronically ill patients with schizophrenia, early non-response accurately predicts non-response at weeks 4 to 12 in 75% to 85% of patients. Early response accurately predicts sustained response at weeks 4 to 12 in approximately 50% to 70% of patients.
• In first-episode patients with schizophrenia, early non-response predicts non-response at weeks 12 to 16 in approximately 60% to 65% of patients. Early response predicts response at weeks 12 to 16 in approximately 60% to 75% of patients.triglyceride, and LDL levels, and a decrease in the HDL level.² These effects may be seen without an increase in BMI, and should be considered a direct effect of the antipsychotic.⁵ Although the mechanism by which dyslipidemia occurs is poorly understood, an increase in the blood glucose level is thought to be, in part, mediated by antagonism of M3 muscarinic receptors on pancreatic âpatients with most improvement occurring during weeks 1 and 2.^1,8

Two meta-analyses found the greatest rate of cumulative improvement in symptoms during the first 2 weeks.^1,8 These analyses included chronically ill patients with mean duration of illness of 15.5 and 10.4 years, respectively. Patients reported 21.9% and 20.5% reductions in symptoms from baseline at 2 weeks, with total responses between 30% at 4 weeks and 40% at 1 year, respectively. These meta-analyses indicate that most of the benefit from antipsychotics in this patient population occurs in the first 2 weeks, which supports the early-response hypothesis.

These observations led to questions about the predictive value of early response and minimum time to determine treatment failure. This article discusses the significance of early response and non-response to antipsychotics and their impact on treating patients with schizophrenia.

What are the predictive factors? How can they guide treatment?

Of the 8 studies in our literature review, only 2 reported early response rates >50%.^9,10 (see this article at CurrentPsychiatry.com for a Box describing the literature review.) Positive predictive value (PPV) ranged from 0.51 to 0.81, meaning that 51% to 81% of early responders continued to respond. Six of the 8 studies reported PPV of 50% to 70%. ^9,11-15 This appears to be true for chronic and first-episode patients, suggesting that 30% to 50% of early responders will fail to have a sustained response (Table 1,^9-16Table 2,^9-16 and Figure).

Compared with early response, early non-response is a more consistent predictor of final non-response. In every study of chronically ill patients, negative predictive value (NPV) was greater than PPV (Table 1).^9-16 NPVs in the literature suggest that 58% to 91% of early non-responders will continue to be non-responders. This seems to be true of chronically ill patients for whom NPVs consistently were between 75% and 85%. By comparison, in first-episode patients NPVs of 58% and 66% were calculated (Table 1^9-16 and Figure).^14,15

These observations suggest that reassessing drug therapy is indicated early in treatment for early non-responders, particularly in chronically ill patients. However, early non-response in a first-episode patient is not as strong a predictor of eventual treatment failure, supporting the idea that first-episode patients may experience a delayed response to therapy. Researchers studying onset of antipsychotic effect report that median time to response onset in first-episode patients may be ≥8 weeks.^6,8 In patients who do not achieve modest early response, assess dose, adherence, substance abuse, and psychosocial stressors.³ For patients without dose, adherence, substance use, or stress issues, switching drug therapy in chronically ill early non-responders is reasonable because the probability of a late response is small.

Individual patient characteristics determine how much these data aid clinical decision-making. If a patient has a good response to an antipsychotic in the first 2 weeks, continue the drug, but observe the patient closely because response may not be sustained. In first-episode patients who fail to respond within 2 weeks of starting an antipsychotic, it is reasonable to continue the drug for several weeks because these patients may be more likely to respond later in therapy.

Clinicians treating chronically ill patients who have failed several antipsychotics and demonstrate a poor response after 2 weeks of an appropriate antipsychotic dose are justified in changing medications because later significant response is unlikely. If a patient has a poor early response but has failed several other antipsychotics with few remaining alternatives, it is reasonable to continue the maximum tolerated dose of the current therapy because the patient may be a late responder. However, early non-response predicts future non-response in many patients.

Case continued

In the case described here, Mr. M is failing his current treatment regimen with a reasonable antipsychotic dose after 2 weeks. Because Mr. M has been on 2 antipsychotics and demonstrated a good response to olanzapine, changing medications should be considered.

Related Resource

• Correll CU, Hauser M. The year in psychosis and bipolar disorder: predicting response to schizophrenia treatment. www.medscape.com/viewarticle/735211_7.

Drug Brand Names

Aripiprazole • Abilify                Quetiapine • Seroquel

Haloperidol • Haldol                 Risperidone • Risperdal

Olanzapine • Zyprexa               Ziprasidone • Geodon

Paliperidone • Invega

Disclosures
Dr. Straley owns stock in Johnson & Johnson. Dr. Webster reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Mr. M, age 28, was given a diagnosis of schizophrenia 6 years ago after experiencing a psychotic break involving auditory hallucinations and paranoia. Olanzapine, 10 mg/d, relieved his symptoms, but he stopped taking the drug after gaining 40 pounds and developing diabetes mellitus. He had 2 other hospital admissions for acute psychosis and has taken at least 1 other medication, the name of which he can’t recall. Recently, Mr. M was involuntarily admitted to the psychiatric ward of his local hospital. His psychiatrist started aripiprazole, 10 mg/d, which was titrated to 30 mg/d. After 2 weeks he reports only a slight decrease in hallucinations. His mother is growing concerned about the effectiveness of this medication and wants to know if it’s time to consider another drug.

Time to onset of action of antipsychotic agents has been debated since at least 1970.¹ Supporters of the delayed-onset hypothesis assert that antipsychotics take weeks or months to show significant improvement of symptoms because of the need for depolarization block for efficacy.² Trials of 4 to 6 weeks often are recommended for patients before failure is declared,^3,4 and trials of this length or longer have proved useful for first-episode patients.^5-7 Recent studies suggest, however, that response is cumulative for chronically ill Practice Points

• Chronically ill and first-episode patients may respond differently to antipsychotics.
• In chronically ill patients with schizophrenia, early non-response accurately predicts non-response at weeks 4 to 12 in 75% to 85% of patients. Early response accurately predicts sustained response at weeks 4 to 12 in approximately 50% to 70% of patients.
• In first-episode patients with schizophrenia, early non-response predicts non-response at weeks 12 to 16 in approximately 60% to 65% of patients. Early response predicts response at weeks 12 to 16 in approximately 60% to 75% of patients.triglyceride, and LDL levels, and a decrease in the HDL level.² These effects may be seen without an increase in BMI, and should be considered a direct effect of the antipsychotic.⁵ Although the mechanism by which dyslipidemia occurs is poorly understood, an increase in the blood glucose level is thought to be, in part, mediated by antagonism of M3 muscarinic receptors on pancreatic âpatients with most improvement occurring during weeks 1 and 2.^1,8

Two meta-analyses found the greatest rate of cumulative improvement in symptoms during the first 2 weeks.^1,8 These analyses included chronically ill patients with mean duration of illness of 15.5 and 10.4 years, respectively. Patients reported 21.9% and 20.5% reductions in symptoms from baseline at 2 weeks, with total responses between 30% at 4 weeks and 40% at 1 year, respectively. These meta-analyses indicate that most of the benefit from antipsychotics in this patient population occurs in the first 2 weeks, which supports the early-response hypothesis.

These observations led to questions about the predictive value of early response and minimum time to determine treatment failure. This article discusses the significance of early response and non-response to antipsychotics and their impact on treating patients with schizophrenia.

What are the predictive factors? How can they guide treatment?

Of the 8 studies in our literature review, only 2 reported early response rates >50%.^9,10 (see this article at CurrentPsychiatry.com for a Box describing the literature review.) Positive predictive value (PPV) ranged from 0.51 to 0.81, meaning that 51% to 81% of early responders continued to respond. Six of the 8 studies reported PPV of 50% to 70%. ^9,11-15 This appears to be true for chronic and first-episode patients, suggesting that 30% to 50% of early responders will fail to have a sustained response (Table 1,^9-16Table 2,^9-16 and Figure).

Compared with early response, early non-response is a more consistent predictor of final non-response. In every study of chronically ill patients, negative predictive value (NPV) was greater than PPV (Table 1).^9-16 NPVs in the literature suggest that 58% to 91% of early non-responders will continue to be non-responders. This seems to be true of chronically ill patients for whom NPVs consistently were between 75% and 85%. By comparison, in first-episode patients NPVs of 58% and 66% were calculated (Table 1^9-16 and Figure).^14,15

These observations suggest that reassessing drug therapy is indicated early in treatment for early non-responders, particularly in chronically ill patients. However, early non-response in a first-episode patient is not as strong a predictor of eventual treatment failure, supporting the idea that first-episode patients may experience a delayed response to therapy. Researchers studying onset of antipsychotic effect report that median time to response onset in first-episode patients may be ≥8 weeks.^6,8 In patients who do not achieve modest early response, assess dose, adherence, substance abuse, and psychosocial stressors.³ For patients without dose, adherence, substance use, or stress issues, switching drug therapy in chronically ill early non-responders is reasonable because the probability of a late response is small.

Individual patient characteristics determine how much these data aid clinical decision-making. If a patient has a good response to an antipsychotic in the first 2 weeks, continue the drug, but observe the patient closely because response may not be sustained. In first-episode patients who fail to respond within 2 weeks of starting an antipsychotic, it is reasonable to continue the drug for several weeks because these patients may be more likely to respond later in therapy.

Clinicians treating chronically ill patients who have failed several antipsychotics and demonstrate a poor response after 2 weeks of an appropriate antipsychotic dose are justified in changing medications because later significant response is unlikely. If a patient has a poor early response but has failed several other antipsychotics with few remaining alternatives, it is reasonable to continue the maximum tolerated dose of the current therapy because the patient may be a late responder. However, early non-response predicts future non-response in many patients.

Case continued

In the case described here, Mr. M is failing his current treatment regimen with a reasonable antipsychotic dose after 2 weeks. Because Mr. M has been on 2 antipsychotics and demonstrated a good response to olanzapine, changing medications should be considered.

Related Resource

• Correll CU, Hauser M. The year in psychosis and bipolar disorder: predicting response to schizophrenia treatment. www.medscape.com/viewarticle/735211_7.

Drug Brand Names

Aripiprazole • Abilify                Quetiapine • Seroquel

Haloperidol • Haldol                 Risperidone • Risperdal

Olanzapine • Zyprexa               Ziprasidone • Geodon

Paliperidone • Invega

Disclosures
Dr. Straley owns stock in Johnson & Johnson. Dr. Webster reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Mr. M, age 28, was given a diagnosis of schizophrenia 6 years ago after experiencing a psychotic break involving auditory hallucinations and paranoia. Olanzapine, 10 mg/d, relieved his symptoms, but he stopped taking the drug after gaining 40 pounds and developing diabetes mellitus. He had 2 other hospital admissions for acute psychosis and has taken at least 1 other medication, the name of which he can’t recall. Recently, Mr. M was involuntarily admitted to the psychiatric ward of his local hospital. His psychiatrist started aripiprazole, 10 mg/d, which was titrated to 30 mg/d. After 2 weeks he reports only a slight decrease in hallucinations. His mother is growing concerned about the effectiveness of this medication and wants to know if it’s time to consider another drug.

Time to onset of action of antipsychotic agents has been debated since at least 1970.¹ Supporters of the delayed-onset hypothesis assert that antipsychotics take weeks or months to show significant improvement of symptoms because of the need for depolarization block for efficacy.² Trials of 4 to 6 weeks often are recommended for patients before failure is declared,^3,4 and trials of this length or longer have proved useful for first-episode patients.^5-7 Recent studies suggest, however, that response is cumulative for chronically ill Practice Points

• Chronically ill and first-episode patients may respond differently to antipsychotics.
• In chronically ill patients with schizophrenia, early non-response accurately predicts non-response at weeks 4 to 12 in 75% to 85% of patients. Early response accurately predicts sustained response at weeks 4 to 12 in approximately 50% to 70% of patients.
• In first-episode patients with schizophrenia, early non-response predicts non-response at weeks 12 to 16 in approximately 60% to 65% of patients. Early response predicts response at weeks 12 to 16 in approximately 60% to 75% of patients.triglyceride, and LDL levels, and a decrease in the HDL level.² These effects may be seen without an increase in BMI, and should be considered a direct effect of the antipsychotic.⁵ Although the mechanism by which dyslipidemia occurs is poorly understood, an increase in the blood glucose level is thought to be, in part, mediated by antagonism of M3 muscarinic receptors on pancreatic âpatients with most improvement occurring during weeks 1 and 2.^1,8

Two meta-analyses found the greatest rate of cumulative improvement in symptoms during the first 2 weeks.^1,8 These analyses included chronically ill patients with mean duration of illness of 15.5 and 10.4 years, respectively. Patients reported 21.9% and 20.5% reductions in symptoms from baseline at 2 weeks, with total responses between 30% at 4 weeks and 40% at 1 year, respectively. These meta-analyses indicate that most of the benefit from antipsychotics in this patient population occurs in the first 2 weeks, which supports the early-response hypothesis.

These observations led to questions about the predictive value of early response and minimum time to determine treatment failure. This article discusses the significance of early response and non-response to antipsychotics and their impact on treating patients with schizophrenia.

What are the predictive factors? How can they guide treatment?

Of the 8 studies in our literature review, only 2 reported early response rates >50%.^9,10 (see this article at CurrentPsychiatry.com for a Box describing the literature review.) Positive predictive value (PPV) ranged from 0.51 to 0.81, meaning that 51% to 81% of early responders continued to respond. Six of the 8 studies reported PPV of 50% to 70%. ^9,11-15 This appears to be true for chronic and first-episode patients, suggesting that 30% to 50% of early responders will fail to have a sustained response (Table 1,^9-16Table 2,^9-16 and Figure).

Compared with early response, early non-response is a more consistent predictor of final non-response. In every study of chronically ill patients, negative predictive value (NPV) was greater than PPV (Table 1).^9-16 NPVs in the literature suggest that 58% to 91% of early non-responders will continue to be non-responders. This seems to be true of chronically ill patients for whom NPVs consistently were between 75% and 85%. By comparison, in first-episode patients NPVs of 58% and 66% were calculated (Table 1^9-16 and Figure).^14,15

These observations suggest that reassessing drug therapy is indicated early in treatment for early non-responders, particularly in chronically ill patients. However, early non-response in a first-episode patient is not as strong a predictor of eventual treatment failure, supporting the idea that first-episode patients may experience a delayed response to therapy. Researchers studying onset of antipsychotic effect report that median time to response onset in first-episode patients may be ≥8 weeks.^6,8 In patients who do not achieve modest early response, assess dose, adherence, substance abuse, and psychosocial stressors.³ For patients without dose, adherence, substance use, or stress issues, switching drug therapy in chronically ill early non-responders is reasonable because the probability of a late response is small.

Individual patient characteristics determine how much these data aid clinical decision-making. If a patient has a good response to an antipsychotic in the first 2 weeks, continue the drug, but observe the patient closely because response may not be sustained. In first-episode patients who fail to respond within 2 weeks of starting an antipsychotic, it is reasonable to continue the drug for several weeks because these patients may be more likely to respond later in therapy.

Clinicians treating chronically ill patients who have failed several antipsychotics and demonstrate a poor response after 2 weeks of an appropriate antipsychotic dose are justified in changing medications because later significant response is unlikely. If a patient has a poor early response but has failed several other antipsychotics with few remaining alternatives, it is reasonable to continue the maximum tolerated dose of the current therapy because the patient may be a late responder. However, early non-response predicts future non-response in many patients.

Case continued

In the case described here, Mr. M is failing his current treatment regimen with a reasonable antipsychotic dose after 2 weeks. Because Mr. M has been on 2 antipsychotics and demonstrated a good response to olanzapine, changing medications should be considered.

Related Resource

• Correll CU, Hauser M. The year in psychosis and bipolar disorder: predicting response to schizophrenia treatment. www.medscape.com/viewarticle/735211_7.

Drug Brand Names

Aripiprazole • Abilify                Quetiapine • Seroquel

Haloperidol • Haldol                 Risperidone • Risperdal

Olanzapine • Zyprexa               Ziprasidone • Geodon

Paliperidone • Invega

Disclosures
Dr. Straley owns stock in Johnson & Johnson. Dr. Webster reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Mrs. D is a 28-year-old married woman who became depressed after her first pregnancy. The depression was treated successfully with paroxetine, 20 mg/d. Before beginning treatment, she reported low mood, spent most of the day in bed, was unable to care for herself, and confessed to thoughts of harming her child.

Mrs. D presents to your clinic asking whether she should continue her selective serotonin reuptake inhibitor (SSRI) because she and her husband are thinking about having a second child. Recently, she tells you, she saw a news article suggesting that anti­depressants show little benefit, and she is concerned that her baby might have a heart defect if she continues paroxetine.

Mrs. D wants to discontinue her medication, but her husband thought she should discuss doing so with you first. During this visit she takes a pregnancy test, which is positive. She wants to know what to do.

Evidence of risks and benefits

Clinicians should be aware of possible adverse effects of SSRI use and untreated depression (Table).^2-10 The available data precludes definitive associations between untreated depression and poor outcomes (Box). Studies of SSRI use during pregnancy have shown conflicting results for all potential outcomes. Absolute risk, with the exception of neonatal adaptation syndrome, is estimated to be small. Neonatal adaptation syndrome—which is characterized by jitteriness, poor muscle tone, weak cries, respiratory distress, hypoglycemia, low Apgar scores, and seizures—occurs in 15% to 30% of infants born to mothers taking SSRIs, but it is transient and resolves during the first weeks of life.

Treatment recommendations

Given the conflicting nature of the evidence, treatment plans should be individualized, weighing the risks and benefits of treatment and the patient’s beliefs and psychiatric history. Consider severity of symptoms and history, including effective therapy and history of relapse. For women with mild or moderate depression, cognitive-behavioral therapy might be an appropriate first-line therapy. However, non-pharmacotherapeutic interventions might not relieve severe depression or be available to all women. When discontinuing an SSRI before pregnancy, counsel the patient to not discontinue the medication abruptly and provide an appropriate taper schedule. See Related Resources for detailed recommendations from the American Psychiatric Association and the American College of Obstetricians and Gynecologists.

Reviewing the SSRI literature regarding pregnancy

Sertraline, paroxetine, citalopram, and fluoxetine are the most studied SSRIs during pregnancy; little information is available on escitalopram and fluvoxamine.¹¹ Prescribing preference generally is given to the medications with the most evidence; paroxetine may be an exception. In 2005, the FDA requested a change in paroxetine’s pregnancy category from C to D, indicating that adequate studies demonstrated a risk of congenital cardiac malformations.¹¹ Additional studies have been conducted, and the teratogenicity of paroxetine is debatable. A recent review reports 8 studies that suggest a malformation risk, compared with 15 studies that show no risk.¹²

The American Academy of Pediatrics considers SSRIs to be compatible with breast-feeding.¹³ The best-studied drugs include sertraline and paroxetine. Fluoxetine should be avoided when possible because a long elimination half-life can cause the drug to accumulate in the newborn, increasing the risk of irritability, hypertonia, sedation, and poor suckle.⁷

There is no best SSRI for all pregnant women. Risks and benefits, including previous treatment success and failure, should be taken into account before starting or switching therapy. Whenever possible, consider monotherapy to avoid compounding the risk of harm.

Related Resources

• Yonkers KA, Wisner KL, Stewart DE, et al. The management of depression during pregnancy: a report from the American Psychiatric Association and the American College of Obstetricians and Gynecologists. Gen Hosp Psychiatry. 2009;31:403-413.
• MGH Center for Women’s Mental Health. www.womensmentalhealth.org.

Drug Brand Names

Citalopram • Celexa      Escitalopram • Lexapro      Fluoxetine • Prozac

Fluvoxamine • Luvox     Paroxetine • Paxil             Sertraline • Zoloft

Disclosures
Dr. Leino reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products. Dr. Ellingrod receives grant support from the National Institute of Mental Health.

Mrs. D is a 28-year-old married woman who became depressed after her first pregnancy. The depression was treated successfully with paroxetine, 20 mg/d. Before beginning treatment, she reported low mood, spent most of the day in bed, was unable to care for herself, and confessed to thoughts of harming her child.

Mrs. D presents to your clinic asking whether she should continue her selective serotonin reuptake inhibitor (SSRI) because she and her husband are thinking about having a second child. Recently, she tells you, she saw a news article suggesting that anti­depressants show little benefit, and she is concerned that her baby might have a heart defect if she continues paroxetine.

Mrs. D wants to discontinue her medication, but her husband thought she should discuss doing so with you first. During this visit she takes a pregnancy test, which is positive. She wants to know what to do.

Evidence of risks and benefits

Clinicians should be aware of possible adverse effects of SSRI use and untreated depression (Table).^2-10 The available data precludes definitive associations between untreated depression and poor outcomes (Box). Studies of SSRI use during pregnancy have shown conflicting results for all potential outcomes. Absolute risk, with the exception of neonatal adaptation syndrome, is estimated to be small. Neonatal adaptation syndrome—which is characterized by jitteriness, poor muscle tone, weak cries, respiratory distress, hypoglycemia, low Apgar scores, and seizures—occurs in 15% to 30% of infants born to mothers taking SSRIs, but it is transient and resolves during the first weeks of life.

Treatment recommendations

Given the conflicting nature of the evidence, treatment plans should be individualized, weighing the risks and benefits of treatment and the patient’s beliefs and psychiatric history. Consider severity of symptoms and history, including effective therapy and history of relapse. For women with mild or moderate depression, cognitive-behavioral therapy might be an appropriate first-line therapy. However, non-pharmacotherapeutic interventions might not relieve severe depression or be available to all women. When discontinuing an SSRI before pregnancy, counsel the patient to not discontinue the medication abruptly and provide an appropriate taper schedule. See Related Resources for detailed recommendations from the American Psychiatric Association and the American College of Obstetricians and Gynecologists.

Reviewing the SSRI literature regarding pregnancy

Sertraline, paroxetine, citalopram, and fluoxetine are the most studied SSRIs during pregnancy; little information is available on escitalopram and fluvoxamine.¹¹ Prescribing preference generally is given to the medications with the most evidence; paroxetine may be an exception. In 2005, the FDA requested a change in paroxetine’s pregnancy category from C to D, indicating that adequate studies demonstrated a risk of congenital cardiac malformations.¹¹ Additional studies have been conducted, and the teratogenicity of paroxetine is debatable. A recent review reports 8 studies that suggest a malformation risk, compared with 15 studies that show no risk.¹²

The American Academy of Pediatrics considers SSRIs to be compatible with breast-feeding.¹³ The best-studied drugs include sertraline and paroxetine. Fluoxetine should be avoided when possible because a long elimination half-life can cause the drug to accumulate in the newborn, increasing the risk of irritability, hypertonia, sedation, and poor suckle.⁷

There is no best SSRI for all pregnant women. Risks and benefits, including previous treatment success and failure, should be taken into account before starting or switching therapy. Whenever possible, consider monotherapy to avoid compounding the risk of harm.

Related Resources

• Yonkers KA, Wisner KL, Stewart DE, et al. The management of depression during pregnancy: a report from the American Psychiatric Association and the American College of Obstetricians and Gynecologists. Gen Hosp Psychiatry. 2009;31:403-413.
• MGH Center for Women’s Mental Health. www.womensmentalhealth.org.

Drug Brand Names

Citalopram • Celexa      Escitalopram • Lexapro      Fluoxetine • Prozac

Fluvoxamine • Luvox     Paroxetine • Paxil             Sertraline • Zoloft

Disclosures
Dr. Leino reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products. Dr. Ellingrod receives grant support from the National Institute of Mental Health.

Mrs. D is a 28-year-old married woman who became depressed after her first pregnancy. The depression was treated successfully with paroxetine, 20 mg/d. Before beginning treatment, she reported low mood, spent most of the day in bed, was unable to care for herself, and confessed to thoughts of harming her child.

Mrs. D presents to your clinic asking whether she should continue her selective serotonin reuptake inhibitor (SSRI) because she and her husband are thinking about having a second child. Recently, she tells you, she saw a news article suggesting that anti­depressants show little benefit, and she is concerned that her baby might have a heart defect if she continues paroxetine.

Mrs. D wants to discontinue her medication, but her husband thought she should discuss doing so with you first. During this visit she takes a pregnancy test, which is positive. She wants to know what to do.

Evidence of risks and benefits

Clinicians should be aware of possible adverse effects of SSRI use and untreated depression (Table).^2-10 The available data precludes definitive associations between untreated depression and poor outcomes (Box). Studies of SSRI use during pregnancy have shown conflicting results for all potential outcomes. Absolute risk, with the exception of neonatal adaptation syndrome, is estimated to be small. Neonatal adaptation syndrome—which is characterized by jitteriness, poor muscle tone, weak cries, respiratory distress, hypoglycemia, low Apgar scores, and seizures—occurs in 15% to 30% of infants born to mothers taking SSRIs, but it is transient and resolves during the first weeks of life.

Treatment recommendations

Given the conflicting nature of the evidence, treatment plans should be individualized, weighing the risks and benefits of treatment and the patient’s beliefs and psychiatric history. Consider severity of symptoms and history, including effective therapy and history of relapse. For women with mild or moderate depression, cognitive-behavioral therapy might be an appropriate first-line therapy. However, non-pharmacotherapeutic interventions might not relieve severe depression or be available to all women. When discontinuing an SSRI before pregnancy, counsel the patient to not discontinue the medication abruptly and provide an appropriate taper schedule. See Related Resources for detailed recommendations from the American Psychiatric Association and the American College of Obstetricians and Gynecologists.

Reviewing the SSRI literature regarding pregnancy

Sertraline, paroxetine, citalopram, and fluoxetine are the most studied SSRIs during pregnancy; little information is available on escitalopram and fluvoxamine.¹¹ Prescribing preference generally is given to the medications with the most evidence; paroxetine may be an exception. In 2005, the FDA requested a change in paroxetine’s pregnancy category from C to D, indicating that adequate studies demonstrated a risk of congenital cardiac malformations.¹¹ Additional studies have been conducted, and the teratogenicity of paroxetine is debatable. A recent review reports 8 studies that suggest a malformation risk, compared with 15 studies that show no risk.¹²

The American Academy of Pediatrics considers SSRIs to be compatible with breast-feeding.¹³ The best-studied drugs include sertraline and paroxetine. Fluoxetine should be avoided when possible because a long elimination half-life can cause the drug to accumulate in the newborn, increasing the risk of irritability, hypertonia, sedation, and poor suckle.⁷

There is no best SSRI for all pregnant women. Risks and benefits, including previous treatment success and failure, should be taken into account before starting or switching therapy. Whenever possible, consider monotherapy to avoid compounding the risk of harm.

Related Resources

• Yonkers KA, Wisner KL, Stewart DE, et al. The management of depression during pregnancy: a report from the American Psychiatric Association and the American College of Obstetricians and Gynecologists. Gen Hosp Psychiatry. 2009;31:403-413.
• MGH Center for Women’s Mental Health. www.womensmentalhealth.org.

Drug Brand Names

Citalopram • Celexa      Escitalopram • Lexapro      Fluoxetine • Prozac

Fluvoxamine • Luvox     Paroxetine • Paxil             Sertraline • Zoloft

Disclosures
Dr. Leino reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products. Dr. Ellingrod receives grant support from the National Institute of Mental Health.

Mr. H, age 31, is admitted to an acute psychiatric unit with major depressive disorder, substance dependence, insomnia, and generalized anxiety. In the past, he was treated unsuccessfully with sertraline, fluoxetine, clonazepam, venlafaxine, and lithium. The treatment team starts Mr. H on quetiapine, titrated to 150 mg at bedtime, to address suspected bipolar II disorder.

At baseline, Mr. H is 68 inches tall and slightly overweight at 176 lbs (body mass index [BMI] 26.8 kg/m²). The laboratory reports his glycated hemoglobin (HbA_1c) at 5.4%; low-density lipoprotein (LDL), 60 mg/dL; total cholesterol, 122 mg/dL; triglycerides, 141 mg/dL; and high-density lipoprotein (HDL), 34 mg/dL.

Within 1 month, Mr. H experiences a 16% increase in body weight. HbA_1c increases to 5.6%; LDL, to 93 mg/dL. These metabolic changes are not addressed, and he continues quetiapine for another 5 months. At the end of 6 months, Mr. H weighs 223.8 lbs (BMI 34 kg/m²)—a 27% increase from baseline. HbA_1c is in the prediabetic range, at 5.9%, and LDL is 120 mg/dL.¹ The treatment team discusses the risks of further metabolic effects, cardiovascular disease, and diabetes with Mr. H. He agrees to a change in therapy.

An increase in weight is thought to be associated with the actions of antipsychotics on H1and 5-HT2c receptors.⁷ Clozapine and olanzapine pose the highest risk of weight gain. Quetiapine and risperidone are considered of intermediate risk; aripiprazole and ziprasidone present the lowest risk(Table 1).^5,7

Patients taking an atypical antipsychotic may experience an elevation of blood glucose, serum Practice Points

• All atypical antipsychotics carry a risk of metabolic disturbance; clozapine and olanzapine have the highest risk, followed by quetiapine and risperidone.
• Newer atypical antipsychotics may carry less of a risk of metabolic side effects, but long-term data are lacking.
• Obtain baseline and periodic monitoring of BMI, waist circumference, HbA1c, fasting plasma glucose, and fasting lipids.
• If you find an abnormality of any of these parameters, consider one or more of the following: switching to an agent that is less risky; decreasing the dose or discontinuing therapy; recommending diet and exercise; and referring the patient to a program or clinician with expertise in the management of weight, diabetes, or lipids.
• Use monotherapy when appropriate to decrease the risk of side effects.triglyceride, and LDL levels, and a decrease in the HDL level.² These effects may be seen without an increase in BMI, and should be considered a direct effect of the antipsychotic.⁵ Although the mechanism by which dyslipidemia occurs is poorly understood, an increase in the blood glucose level is thought to be, in part, mediated by antagonism of M3 muscarinic receptors on pancreatic â-cells.⁷ Clozapine and olanzapine pose the highest risk of dyslipidemia. Quetiapine and risperidone are considered of intermediate risk; the risk associated with quetiapine is closer to that of olanzpine.^8,9 Aripiprazole and ziprasidone present a lower risk of dyslipidemia and glucose elevations.⁵

Newer atypical antipsychotics, such as asenapine, iloperidone, paliperidone, and lurasidone, seem to have a lower metabolic risk profile, similar to those seen with aripiprazole and ziprasidone.⁵ Patients enrolled in initial clinical trials might not be antipsychotic naïve, however, and may have been taking a high metabolic risk antipsychotic. When these patients are switched to an antipsychotic that carries less of a metabolic risk, it might appear that they are experiencing a decrease in metabolic adverse events.

Metabolic data on newer atypical antipsychotics are limited; most have not been subject to long-term study. Routine monitoring of metabolic side effects is recommended for all atypical antipsychotics, regardless of risk profile.

Recommended monitoring

Because of the known metabolic side effects that occur in patients taking an atypical antipsychotic, baseline and periodic monitoring is recommended (Table 2).^2,10 BMI and waist circumference should be recorded at baseline and tracked throughout treatment. Ideally, obtain measurements monthly for the first 3 months of therapy, or after any medication adjustments, then at 6 months, and annually thereafter. Encourage patients to track their own weight.

HbA_1c and fasting plasma glucose levels should be measured at baseline and throughout the course of treatment. Obtain another set of measurements at 3 months, then annually thereafter, unless the patient develops type 2 diabetes mellitus.²

Obtaining a fasting lipid panel at baseline and periodically throughout the course of treatment is recommended. After baseline measurement, another panel should be taken at 3 months and annually thereafter. Guidelines of the American Diabetes Association recommend a fasting lipid panel every 5 years—however, good clinical practice dictates obtaining a lipid panel annually.

Managing metabolic side effects

Assess whether the patient can benefit from a lower dosage of current medication, switching to an antipsychotic with less of a risk of metabolic disturbance, or from discontinuation of therapy. In most cases, aim to use monotherapy because polypharmacy contributes to an increased risk of side effects.10

Weight management. Recommend nutrition counseling and physical activity for all patients who are overweight. Referral to a health care professional or to a program with expertise in weight management also might be beneficial.² Include family members and significant others in the patient’s education when possible.

Impaired fasting glucose. Encourage a low-carbohydrate, high-protein diet with high intake of vegetables. Patients should obtain at least 30 minutes of physical activity, five times a week. Referral to a diabetes self-management class also is appropriate. Consider referral to a primary care physician or a clinician with expertise in diabetes.²

Impaired fasting lipids. Encourage your patients to adhere to a heart-healthy diet that is low in saturated fats and to get adequate physical activity. Referral to a dietician and primary care provider for medical management of dyslipidemia might be appropriate.²

Related Resources

• American Diabetes Association. Guide to living with diabetes. www.diabetes.org/living-with-diabetes.
• MOVE! Weight Management Program for Veterans. www. move.va.gov.

Drug Brand Names

Aripiprazole • Abilify
Asenapine • Saphris
Clonazepam • Klonopin
Clozapine • Clozaril
Fluoxetine • Prozac
Iloperidone • Fanapt
Lithium • Eskalith, Lithobid
Lurasidone • Latuda
Olanzapine • Zyprexa
Paliperidone • Invega
Quetiapine • Seroquel
Risperidone • Risperdal
Sertraline • Zoloft
Venlafaxine • Effexor
Ziprasidone • Geodon

Disclosure

The authors report no financial relationships with any of the manufacturers mentioned in this article or with manufacturers of competing products.

Mr. H, age 31, is admitted to an acute psychiatric unit with major depressive disorder, substance dependence, insomnia, and generalized anxiety. In the past, he was treated unsuccessfully with sertraline, fluoxetine, clonazepam, venlafaxine, and lithium. The treatment team starts Mr. H on quetiapine, titrated to 150 mg at bedtime, to address suspected bipolar II disorder.

At baseline, Mr. H is 68 inches tall and slightly overweight at 176 lbs (body mass index [BMI] 26.8 kg/m²). The laboratory reports his glycated hemoglobin (HbA_1c) at 5.4%; low-density lipoprotein (LDL), 60 mg/dL; total cholesterol, 122 mg/dL; triglycerides, 141 mg/dL; and high-density lipoprotein (HDL), 34 mg/dL.

Within 1 month, Mr. H experiences a 16% increase in body weight. HbA_1c increases to 5.6%; LDL, to 93 mg/dL. These metabolic changes are not addressed, and he continues quetiapine for another 5 months. At the end of 6 months, Mr. H weighs 223.8 lbs (BMI 34 kg/m²)—a 27% increase from baseline. HbA_1c is in the prediabetic range, at 5.9%, and LDL is 120 mg/dL.¹ The treatment team discusses the risks of further metabolic effects, cardiovascular disease, and diabetes with Mr. H. He agrees to a change in therapy.

An increase in weight is thought to be associated with the actions of antipsychotics on H1and 5-HT2c receptors.⁷ Clozapine and olanzapine pose the highest risk of weight gain. Quetiapine and risperidone are considered of intermediate risk; aripiprazole and ziprasidone present the lowest risk(Table 1).^5,7

Patients taking an atypical antipsychotic may experience an elevation of blood glucose, serum Practice Points

• All atypical antipsychotics carry a risk of metabolic disturbance; clozapine and olanzapine have the highest risk, followed by quetiapine and risperidone.
• Newer atypical antipsychotics may carry less of a risk of metabolic side effects, but long-term data are lacking.
• Obtain baseline and periodic monitoring of BMI, waist circumference, HbA1c, fasting plasma glucose, and fasting lipids.
• If you find an abnormality of any of these parameters, consider one or more of the following: switching to an agent that is less risky; decreasing the dose or discontinuing therapy; recommending diet and exercise; and referring the patient to a program or clinician with expertise in the management of weight, diabetes, or lipids.
• Use monotherapy when appropriate to decrease the risk of side effects.triglyceride, and LDL levels, and a decrease in the HDL level.² These effects may be seen without an increase in BMI, and should be considered a direct effect of the antipsychotic.⁵ Although the mechanism by which dyslipidemia occurs is poorly understood, an increase in the blood glucose level is thought to be, in part, mediated by antagonism of M3 muscarinic receptors on pancreatic â-cells.⁷ Clozapine and olanzapine pose the highest risk of dyslipidemia. Quetiapine and risperidone are considered of intermediate risk; the risk associated with quetiapine is closer to that of olanzpine.^8,9 Aripiprazole and ziprasidone present a lower risk of dyslipidemia and glucose elevations.⁵

Newer atypical antipsychotics, such as asenapine, iloperidone, paliperidone, and lurasidone, seem to have a lower metabolic risk profile, similar to those seen with aripiprazole and ziprasidone.⁵ Patients enrolled in initial clinical trials might not be antipsychotic naïve, however, and may have been taking a high metabolic risk antipsychotic. When these patients are switched to an antipsychotic that carries less of a metabolic risk, it might appear that they are experiencing a decrease in metabolic adverse events.

Metabolic data on newer atypical antipsychotics are limited; most have not been subject to long-term study. Routine monitoring of metabolic side effects is recommended for all atypical antipsychotics, regardless of risk profile.

Recommended monitoring

Because of the known metabolic side effects that occur in patients taking an atypical antipsychotic, baseline and periodic monitoring is recommended (Table 2).^2,10 BMI and waist circumference should be recorded at baseline and tracked throughout treatment. Ideally, obtain measurements monthly for the first 3 months of therapy, or after any medication adjustments, then at 6 months, and annually thereafter. Encourage patients to track their own weight.

HbA_1c and fasting plasma glucose levels should be measured at baseline and throughout the course of treatment. Obtain another set of measurements at 3 months, then annually thereafter, unless the patient develops type 2 diabetes mellitus.²

Obtaining a fasting lipid panel at baseline and periodically throughout the course of treatment is recommended. After baseline measurement, another panel should be taken at 3 months and annually thereafter. Guidelines of the American Diabetes Association recommend a fasting lipid panel every 5 years—however, good clinical practice dictates obtaining a lipid panel annually.

Managing metabolic side effects

Assess whether the patient can benefit from a lower dosage of current medication, switching to an antipsychotic with less of a risk of metabolic disturbance, or from discontinuation of therapy. In most cases, aim to use monotherapy because polypharmacy contributes to an increased risk of side effects.10

Weight management. Recommend nutrition counseling and physical activity for all patients who are overweight. Referral to a health care professional or to a program with expertise in weight management also might be beneficial.² Include family members and significant others in the patient’s education when possible.

Impaired fasting glucose. Encourage a low-carbohydrate, high-protein diet with high intake of vegetables. Patients should obtain at least 30 minutes of physical activity, five times a week. Referral to a diabetes self-management class also is appropriate. Consider referral to a primary care physician or a clinician with expertise in diabetes.²

Impaired fasting lipids. Encourage your patients to adhere to a heart-healthy diet that is low in saturated fats and to get adequate physical activity. Referral to a dietician and primary care provider for medical management of dyslipidemia might be appropriate.²

Related Resources

• American Diabetes Association. Guide to living with diabetes. www.diabetes.org/living-with-diabetes.
• MOVE! Weight Management Program for Veterans. www. move.va.gov.

Drug Brand Names

Aripiprazole • Abilify
Asenapine • Saphris
Clonazepam • Klonopin
Clozapine • Clozaril
Fluoxetine • Prozac
Iloperidone • Fanapt
Lithium • Eskalith, Lithobid
Lurasidone • Latuda
Olanzapine • Zyprexa
Paliperidone • Invega
Quetiapine • Seroquel
Risperidone • Risperdal
Sertraline • Zoloft
Venlafaxine • Effexor
Ziprasidone • Geodon

Disclosure

The authors report no financial relationships with any of the manufacturers mentioned in this article or with manufacturers of competing products.

Mr. H, age 31, is admitted to an acute psychiatric unit with major depressive disorder, substance dependence, insomnia, and generalized anxiety. In the past, he was treated unsuccessfully with sertraline, fluoxetine, clonazepam, venlafaxine, and lithium. The treatment team starts Mr. H on quetiapine, titrated to 150 mg at bedtime, to address suspected bipolar II disorder.

At baseline, Mr. H is 68 inches tall and slightly overweight at 176 lbs (body mass index [BMI] 26.8 kg/m²). The laboratory reports his glycated hemoglobin (HbA_1c) at 5.4%; low-density lipoprotein (LDL), 60 mg/dL; total cholesterol, 122 mg/dL; triglycerides, 141 mg/dL; and high-density lipoprotein (HDL), 34 mg/dL.

Within 1 month, Mr. H experiences a 16% increase in body weight. HbA_1c increases to 5.6%; LDL, to 93 mg/dL. These metabolic changes are not addressed, and he continues quetiapine for another 5 months. At the end of 6 months, Mr. H weighs 223.8 lbs (BMI 34 kg/m²)—a 27% increase from baseline. HbA_1c is in the prediabetic range, at 5.9%, and LDL is 120 mg/dL.¹ The treatment team discusses the risks of further metabolic effects, cardiovascular disease, and diabetes with Mr. H. He agrees to a change in therapy.

An increase in weight is thought to be associated with the actions of antipsychotics on H1and 5-HT2c receptors.⁷ Clozapine and olanzapine pose the highest risk of weight gain. Quetiapine and risperidone are considered of intermediate risk; aripiprazole and ziprasidone present the lowest risk(Table 1).^5,7

Patients taking an atypical antipsychotic may experience an elevation of blood glucose, serum Practice Points

• All atypical antipsychotics carry a risk of metabolic disturbance; clozapine and olanzapine have the highest risk, followed by quetiapine and risperidone.
• Newer atypical antipsychotics may carry less of a risk of metabolic side effects, but long-term data are lacking.
• Obtain baseline and periodic monitoring of BMI, waist circumference, HbA1c, fasting plasma glucose, and fasting lipids.
• If you find an abnormality of any of these parameters, consider one or more of the following: switching to an agent that is less risky; decreasing the dose or discontinuing therapy; recommending diet and exercise; and referring the patient to a program or clinician with expertise in the management of weight, diabetes, or lipids.
• Use monotherapy when appropriate to decrease the risk of side effects.triglyceride, and LDL levels, and a decrease in the HDL level.² These effects may be seen without an increase in BMI, and should be considered a direct effect of the antipsychotic.⁵ Although the mechanism by which dyslipidemia occurs is poorly understood, an increase in the blood glucose level is thought to be, in part, mediated by antagonism of M3 muscarinic receptors on pancreatic â-cells.⁷ Clozapine and olanzapine pose the highest risk of dyslipidemia. Quetiapine and risperidone are considered of intermediate risk; the risk associated with quetiapine is closer to that of olanzpine.^8,9 Aripiprazole and ziprasidone present a lower risk of dyslipidemia and glucose elevations.⁵

Newer atypical antipsychotics, such as asenapine, iloperidone, paliperidone, and lurasidone, seem to have a lower metabolic risk profile, similar to those seen with aripiprazole and ziprasidone.⁵ Patients enrolled in initial clinical trials might not be antipsychotic naïve, however, and may have been taking a high metabolic risk antipsychotic. When these patients are switched to an antipsychotic that carries less of a metabolic risk, it might appear that they are experiencing a decrease in metabolic adverse events.

Metabolic data on newer atypical antipsychotics are limited; most have not been subject to long-term study. Routine monitoring of metabolic side effects is recommended for all atypical antipsychotics, regardless of risk profile.

Recommended monitoring

Because of the known metabolic side effects that occur in patients taking an atypical antipsychotic, baseline and periodic monitoring is recommended (Table 2).^2,10 BMI and waist circumference should be recorded at baseline and tracked throughout treatment. Ideally, obtain measurements monthly for the first 3 months of therapy, or after any medication adjustments, then at 6 months, and annually thereafter. Encourage patients to track their own weight.

HbA_1c and fasting plasma glucose levels should be measured at baseline and throughout the course of treatment. Obtain another set of measurements at 3 months, then annually thereafter, unless the patient develops type 2 diabetes mellitus.²

Obtaining a fasting lipid panel at baseline and periodically throughout the course of treatment is recommended. After baseline measurement, another panel should be taken at 3 months and annually thereafter. Guidelines of the American Diabetes Association recommend a fasting lipid panel every 5 years—however, good clinical practice dictates obtaining a lipid panel annually.

Managing metabolic side effects

Assess whether the patient can benefit from a lower dosage of current medication, switching to an antipsychotic with less of a risk of metabolic disturbance, or from discontinuation of therapy. In most cases, aim to use monotherapy because polypharmacy contributes to an increased risk of side effects.10

Weight management. Recommend nutrition counseling and physical activity for all patients who are overweight. Referral to a health care professional or to a program with expertise in weight management also might be beneficial.² Include family members and significant others in the patient’s education when possible.

Impaired fasting glucose. Encourage a low-carbohydrate, high-protein diet with high intake of vegetables. Patients should obtain at least 30 minutes of physical activity, five times a week. Referral to a diabetes self-management class also is appropriate. Consider referral to a primary care physician or a clinician with expertise in diabetes.²

Impaired fasting lipids. Encourage your patients to adhere to a heart-healthy diet that is low in saturated fats and to get adequate physical activity. Referral to a dietician and primary care provider for medical management of dyslipidemia might be appropriate.²

Related Resources

• American Diabetes Association. Guide to living with diabetes. www.diabetes.org/living-with-diabetes.
• MOVE! Weight Management Program for Veterans. www. move.va.gov.

Drug Brand Names

Aripiprazole • Abilify
Asenapine • Saphris
Clonazepam • Klonopin
Clozapine • Clozaril
Fluoxetine • Prozac
Iloperidone • Fanapt
Lithium • Eskalith, Lithobid
Lurasidone • Latuda
Olanzapine • Zyprexa
Paliperidone • Invega
Quetiapine • Seroquel
Risperidone • Risperdal
Sertraline • Zoloft
Venlafaxine • Effexor
Ziprasidone • Geodon

Disclosure

The authors report no financial relationships with any of the manufacturers mentioned in this article or with manufacturers of competing products.

Mr. H, age 33, was diagnosed with bipolar I disorder 9 years ago. For the past year, his mood symptoms have been well controlled with lithium 300 mg, 3 times a day, and olanzapine, 20 mg/d. He presents to the outpatient clinic for a routine visit complaining of insomnia, daytime sleepiness, and increased thirst. He also notes that his tremor has become more prominent over the last few weeks. Concerned about his symptoms, Mr. H’s clinician orders a comprehensive laboratory panel (Table).

Upon further questioning, Mr. H’s physician determines that his insomnia is caused by nocturnal urination, which is consistent with fluid and electrolyte imbalances seen in Mr. H’s laboratory panel. Mr. H is diagnosed with lithium-induced diabetes insipidus.

Although lithium’s exact mechanism of action is unknown, it is known that lithium can negatively affect the kidneys.^1,2 Typically, antidiuretic hormone (ADH) regulates water permeability in the collecting duct of the nephron, allowing water to be reabsorbed through simple diffusion in the kidney’s collecting duct (Figure).³ Chronic lithium use reduces or desensitizes the kidney’s ability to respond to ADH. Resistance to ADH occurs when lithium accumulates in the cells of the collecting duct and inhibits ADH’s ability to increase water permeability. This inhibition can cause some of Mr. H’s symptoms, such as polydipsia and polyuria, and is estimated to occur in approximately 40% of patients receiving long-term lithium therapy.^4,5

Diagnosis

Diagnosing lithium-induced nephrogenic diabetes insipidus (NDI) begins with a history of the patient’s symptoms and ordering lab tests.⁵ The next step involves a water restriction test, also known as a thirst test, to measure the patient’s ability to concentrate his or her urine. Baseline serum osmolality and electrolytes are compared with new values obtained after completing the water restriction test. Healthy people will have a 2-to-4-fold increase in urine osmolality compared with patients who have NDI. The last step includes administering desmopressin and differentiates between central diabetes insipidus and NDI.⁶

After desmopressin use, patients who have central diabetes insipidus will have a >50% increase in urine osmolality, whereas patients who have NDI will have <10% increase in urine osmolality. This distinction is important because patients with central diabetes insipidus might have more severe disease and might not benefit from measures commonly used for lithium-induced NDI.⁷

Prevention and management

Lithium-induced NDI is thought to be dose-dependent and may be prevented by using the lowest effective dose of lithium for an individual patient. It is important that patients taking lithium receive basic electrolyte, hematologic, liver function, renal function, and thyroid function tests at baseline and every 6 to 12 months after the lithium regimen is stable. Additionally, lithium levels should be monitored frequently. The frequency of these tests may range from twice weekly to every 3 to 4 months or longer, depending on the patient’s condition. This monitoring allows the prescriber to quickly identify emerging adverse effects.

Patients with impaired renal function and those with a urine output >3 liters a day are more susceptible to NDI and require monitoring every 3 months. Also, instruct patients to monitor their urine output and educate them about the dangers of fluid and electrolyte imbalances and the signs and symptoms of NDI, such as excessive thirst and urination.^1,2

When a patient experiences lithium-induced NDI, re-evaluate treatment and dosage, including simplifying the dosing regimen or switching to once-daily dosing, usually at bedtime. Once-daily dosing results in a lower overall lithium trough, which might allow the kidneys more “drug-free” time.4,5 Additionally, 12-hour lithium levels are approximately 20% higher with once-daily monitoring; continued monitoring is needed during this switch. Patients who have a moderate or severe form of lithium-induced NDI may need to discontinue lithium altogether. There are several options for treating lithium-induced NDI in patients who need to take lithium. Closely monitor kidney function and lithium routinely with these strategies.

Amiloride. This potassium-sparing diuretic minimizes accumulation of lithium by inhibiting collecting duct sodium channels. Studies have shown that amiloride can decrease mean urine volume, increase urine osmolality, and improve the kidneys’ ability to respond to exogenous arginine vasopressin.⁸

Thiazide diuretics produce mild sodium depletion, which decreases the distal tubule delivery of sodium, therefore increasing water reabsorption in the collecting duct. Hydrochlorothiazide has been shown to reduce urine output by >50% in patients with NDI on a sodium-restricted diet. Hydrochlorothiazide use requires careful monitoring of potassium and lithium levels. Use of a thiazide diuretic also might warrant decreasing the lithium dose by as much as 50% to prevent toxicity.^9,10

Low-sodium diet plus hydrochlorothiazide. This route provides another option to decrease urine output during lithium-induced NDI. A reduction in urine output has been shown to be directly proportional to a decrease in salt intake and excretion. Restricting sodium to <2.3 g/d is an appropriate goal for many patients to prevent reoccurring symptoms, which is more than the 3 g/d average that most Americans consume. Potassium and lithium levels must be monitored closely.⁹

Nonsteroidal anti-inflammatory drugs (NSAIDs). These drugs’ ability to inhibit prostaglandin synthesis prevents prostaglandins from antagonizing actions of ADH in the kidney. The result is increased urine concentration via the actions of ADH. Indomethacin has a greater effect than ibuprofen in increasing ADH’s actions on the kidney. Use of concomitant NSAIDs with lithium requires close monitoring of renal function tests.¹¹

Mr. H, age 33, was diagnosed with bipolar I disorder 9 years ago. For the past year, his mood symptoms have been well controlled with lithium 300 mg, 3 times a day, and olanzapine, 20 mg/d. He presents to the outpatient clinic for a routine visit complaining of insomnia, daytime sleepiness, and increased thirst. He also notes that his tremor has become more prominent over the last few weeks. Concerned about his symptoms, Mr. H’s clinician orders a comprehensive laboratory panel (Table).

Upon further questioning, Mr. H’s physician determines that his insomnia is caused by nocturnal urination, which is consistent with fluid and electrolyte imbalances seen in Mr. H’s laboratory panel. Mr. H is diagnosed with lithium-induced diabetes insipidus.

Although lithium’s exact mechanism of action is unknown, it is known that lithium can negatively affect the kidneys.^1,2 Typically, antidiuretic hormone (ADH) regulates water permeability in the collecting duct of the nephron, allowing water to be reabsorbed through simple diffusion in the kidney’s collecting duct (Figure).³ Chronic lithium use reduces or desensitizes the kidney’s ability to respond to ADH. Resistance to ADH occurs when lithium accumulates in the cells of the collecting duct and inhibits ADH’s ability to increase water permeability. This inhibition can cause some of Mr. H’s symptoms, such as polydipsia and polyuria, and is estimated to occur in approximately 40% of patients receiving long-term lithium therapy.^4,5

Diagnosis

Diagnosing lithium-induced nephrogenic diabetes insipidus (NDI) begins with a history of the patient’s symptoms and ordering lab tests.⁵ The next step involves a water restriction test, also known as a thirst test, to measure the patient’s ability to concentrate his or her urine. Baseline serum osmolality and electrolytes are compared with new values obtained after completing the water restriction test. Healthy people will have a 2-to-4-fold increase in urine osmolality compared with patients who have NDI. The last step includes administering desmopressin and differentiates between central diabetes insipidus and NDI.⁶

After desmopressin use, patients who have central diabetes insipidus will have a >50% increase in urine osmolality, whereas patients who have NDI will have <10% increase in urine osmolality. This distinction is important because patients with central diabetes insipidus might have more severe disease and might not benefit from measures commonly used for lithium-induced NDI.⁷

Prevention and management

Lithium-induced NDI is thought to be dose-dependent and may be prevented by using the lowest effective dose of lithium for an individual patient. It is important that patients taking lithium receive basic electrolyte, hematologic, liver function, renal function, and thyroid function tests at baseline and every 6 to 12 months after the lithium regimen is stable. Additionally, lithium levels should be monitored frequently. The frequency of these tests may range from twice weekly to every 3 to 4 months or longer, depending on the patient’s condition. This monitoring allows the prescriber to quickly identify emerging adverse effects.

Patients with impaired renal function and those with a urine output >3 liters a day are more susceptible to NDI and require monitoring every 3 months. Also, instruct patients to monitor their urine output and educate them about the dangers of fluid and electrolyte imbalances and the signs and symptoms of NDI, such as excessive thirst and urination.^1,2

When a patient experiences lithium-induced NDI, re-evaluate treatment and dosage, including simplifying the dosing regimen or switching to once-daily dosing, usually at bedtime. Once-daily dosing results in a lower overall lithium trough, which might allow the kidneys more “drug-free” time.4,5 Additionally, 12-hour lithium levels are approximately 20% higher with once-daily monitoring; continued monitoring is needed during this switch. Patients who have a moderate or severe form of lithium-induced NDI may need to discontinue lithium altogether. There are several options for treating lithium-induced NDI in patients who need to take lithium. Closely monitor kidney function and lithium routinely with these strategies.

Amiloride. This potassium-sparing diuretic minimizes accumulation of lithium by inhibiting collecting duct sodium channels. Studies have shown that amiloride can decrease mean urine volume, increase urine osmolality, and improve the kidneys’ ability to respond to exogenous arginine vasopressin.⁸

Thiazide diuretics produce mild sodium depletion, which decreases the distal tubule delivery of sodium, therefore increasing water reabsorption in the collecting duct. Hydrochlorothiazide has been shown to reduce urine output by >50% in patients with NDI on a sodium-restricted diet. Hydrochlorothiazide use requires careful monitoring of potassium and lithium levels. Use of a thiazide diuretic also might warrant decreasing the lithium dose by as much as 50% to prevent toxicity.^9,10

Low-sodium diet plus hydrochlorothiazide. This route provides another option to decrease urine output during lithium-induced NDI. A reduction in urine output has been shown to be directly proportional to a decrease in salt intake and excretion. Restricting sodium to <2.3 g/d is an appropriate goal for many patients to prevent reoccurring symptoms, which is more than the 3 g/d average that most Americans consume. Potassium and lithium levels must be monitored closely.⁹

Nonsteroidal anti-inflammatory drugs (NSAIDs). These drugs’ ability to inhibit prostaglandin synthesis prevents prostaglandins from antagonizing actions of ADH in the kidney. The result is increased urine concentration via the actions of ADH. Indomethacin has a greater effect than ibuprofen in increasing ADH’s actions on the kidney. Use of concomitant NSAIDs with lithium requires close monitoring of renal function tests.¹¹

Mr. H, age 33, was diagnosed with bipolar I disorder 9 years ago. For the past year, his mood symptoms have been well controlled with lithium 300 mg, 3 times a day, and olanzapine, 20 mg/d. He presents to the outpatient clinic for a routine visit complaining of insomnia, daytime sleepiness, and increased thirst. He also notes that his tremor has become more prominent over the last few weeks. Concerned about his symptoms, Mr. H’s clinician orders a comprehensive laboratory panel (Table).

Upon further questioning, Mr. H’s physician determines that his insomnia is caused by nocturnal urination, which is consistent with fluid and electrolyte imbalances seen in Mr. H’s laboratory panel. Mr. H is diagnosed with lithium-induced diabetes insipidus.

Although lithium’s exact mechanism of action is unknown, it is known that lithium can negatively affect the kidneys.^1,2 Typically, antidiuretic hormone (ADH) regulates water permeability in the collecting duct of the nephron, allowing water to be reabsorbed through simple diffusion in the kidney’s collecting duct (Figure).³ Chronic lithium use reduces or desensitizes the kidney’s ability to respond to ADH. Resistance to ADH occurs when lithium accumulates in the cells of the collecting duct and inhibits ADH’s ability to increase water permeability. This inhibition can cause some of Mr. H’s symptoms, such as polydipsia and polyuria, and is estimated to occur in approximately 40% of patients receiving long-term lithium therapy.^4,5

Diagnosis

Diagnosing lithium-induced nephrogenic diabetes insipidus (NDI) begins with a history of the patient’s symptoms and ordering lab tests.⁵ The next step involves a water restriction test, also known as a thirst test, to measure the patient’s ability to concentrate his or her urine. Baseline serum osmolality and electrolytes are compared with new values obtained after completing the water restriction test. Healthy people will have a 2-to-4-fold increase in urine osmolality compared with patients who have NDI. The last step includes administering desmopressin and differentiates between central diabetes insipidus and NDI.⁶

After desmopressin use, patients who have central diabetes insipidus will have a >50% increase in urine osmolality, whereas patients who have NDI will have <10% increase in urine osmolality. This distinction is important because patients with central diabetes insipidus might have more severe disease and might not benefit from measures commonly used for lithium-induced NDI.⁷

Prevention and management

Lithium-induced NDI is thought to be dose-dependent and may be prevented by using the lowest effective dose of lithium for an individual patient. It is important that patients taking lithium receive basic electrolyte, hematologic, liver function, renal function, and thyroid function tests at baseline and every 6 to 12 months after the lithium regimen is stable. Additionally, lithium levels should be monitored frequently. The frequency of these tests may range from twice weekly to every 3 to 4 months or longer, depending on the patient’s condition. This monitoring allows the prescriber to quickly identify emerging adverse effects.

Patients with impaired renal function and those with a urine output >3 liters a day are more susceptible to NDI and require monitoring every 3 months. Also, instruct patients to monitor their urine output and educate them about the dangers of fluid and electrolyte imbalances and the signs and symptoms of NDI, such as excessive thirst and urination.^1,2

When a patient experiences lithium-induced NDI, re-evaluate treatment and dosage, including simplifying the dosing regimen or switching to once-daily dosing, usually at bedtime. Once-daily dosing results in a lower overall lithium trough, which might allow the kidneys more “drug-free” time.4,5 Additionally, 12-hour lithium levels are approximately 20% higher with once-daily monitoring; continued monitoring is needed during this switch. Patients who have a moderate or severe form of lithium-induced NDI may need to discontinue lithium altogether. There are several options for treating lithium-induced NDI in patients who need to take lithium. Closely monitor kidney function and lithium routinely with these strategies.

Amiloride. This potassium-sparing diuretic minimizes accumulation of lithium by inhibiting collecting duct sodium channels. Studies have shown that amiloride can decrease mean urine volume, increase urine osmolality, and improve the kidneys’ ability to respond to exogenous arginine vasopressin.⁸

Thiazide diuretics produce mild sodium depletion, which decreases the distal tubule delivery of sodium, therefore increasing water reabsorption in the collecting duct. Hydrochlorothiazide has been shown to reduce urine output by >50% in patients with NDI on a sodium-restricted diet. Hydrochlorothiazide use requires careful monitoring of potassium and lithium levels. Use of a thiazide diuretic also might warrant decreasing the lithium dose by as much as 50% to prevent toxicity.^9,10

Low-sodium diet plus hydrochlorothiazide. This route provides another option to decrease urine output during lithium-induced NDI. A reduction in urine output has been shown to be directly proportional to a decrease in salt intake and excretion. Restricting sodium to <2.3 g/d is an appropriate goal for many patients to prevent reoccurring symptoms, which is more than the 3 g/d average that most Americans consume. Potassium and lithium levels must be monitored closely.⁹

Nonsteroidal anti-inflammatory drugs (NSAIDs). These drugs’ ability to inhibit prostaglandin synthesis prevents prostaglandins from antagonizing actions of ADH in the kidney. The result is increased urine concentration via the actions of ADH. Indomethacin has a greater effect than ibuprofen in increasing ADH’s actions on the kidney. Use of concomitant NSAIDs with lithium requires close monitoring of renal function tests.¹¹

Mrs. L, age 27, has a history of major depressive disorder with symptoms of anxiety. She was managed successfully for 2 years with bupropion XL, 300 mg/d, but was switched to venlafaxine, titrated to 225 mg/d, after she developed seizures secondary to a head injury sustained in a car accident. After the switch, Mrs. L’s mood deteriorated and she was hospitalized. Since then, she’s received several medication trials, including paroxetine, 30 mg/d, a selective serotonin reuptake inhibitor (SSRI), and the tricyclic antidepressant (TCA) nortriptyline, 75 mg/d, but she could not tolerate these medications because of severe xerostomia.

After taking sertraline, 150 mg/d, for 8 weeks, Mrs. L improves and has a Patient Health Questionnaire score of 6, indicating mild depression. Her initial complaints of diarrhea and nausea have resolved, but Mrs. L now reports that she and her husband are having marital difficulties because she cannot achieve orgasm during sexual intercourse. She did not have this problem when she was taking bupropion. Her husband occasionally takes the phosphodiesterase type 5 (PDE5) inhibitor sildenafil before intercourse, and Mrs. L asks you if this medication will help her achieve orgasm.

DSM-IV-TR defines sexual dysfunction as disturbances in sexual desire and/or in the sexual response cycle (excitement, plateau, orgasm, and resolution) that result in marked distress and interpersonal difficulty.¹ Sexual dysfunction can occur with the use of any antidepressant with serotonergic activity; it affects an estimated 50% to 70% of patients who take SSRIs.² Sexual dysfunction can occur with all SSRIs; however, higher rates of sexual dysfunction are found with citalopram, fluoxetine, paroxetine, and sertraline.³ Studies have suggested there may be a dose-side effect relationship with SSRI-induced sexual dysfunction.⁴

Several factors can increase a patient’s risk of sexual dysfunction and should be considered before prescribing an antidepressant or when a patient presents with new or worsening sexual dysfunction (Table 1).⁵ In general, nonserotonergic agents such as bupropion, mirtazapine, and nefazodone are associated with lower rates of sexual dysfunction. The pharmacology of these agents explains their decreased propensity to cause sexual dysfunction. These agents increase levels of dopamine in the mesolimbic dopaminergic system either by blocking reuptake (bupropion) or antagonizing the serotonin subtype-2 receptor and facilitating disinhibition of decreased dopamine downstream (nefazodone and mirtazapine).

Table 1

Risk factors for sexual dysfunction

One option for treating antidepressant-induced sexual dysfunction in women is PDE5 inhibitors, which are used to treat erectile dysfunction (ED). These medications ameliorate ED by inhibiting degradation of cyclic guanosine monophosphate by PDE5, which increases blood flow to the penis during sexual stimulation. Although these medications are not FDA-approved for treating sexual dysfunction in women, adjunctive PDE5 inhibitor treatment may be beneficial for sexual dysfunction in females because similar mediators, such as nitric oxide and cyclic guanosine monophosphate, involved in the nonadrenergic-noncholinergic signaling that controls sexual stimulation in men also are found in female genital tissue.⁶

When treating a woman with SSRI-induced sexual dysfunction, consider nonpharmacologic treatments both before and during pharmacotherapy (Table 2).^7,8 See Table 3 for a comparison of pharmacokinetics, side effects, and drug interactions of the 4 FDA-approved PDE5 inhibitors—avanafil, sildenafil, tadalafil, and vardenafil.

Table 2

Management strategies for SSRI-induced sexual dysfunction

Table 3

Phosphodiesterase type 5 inhibitors: A comparison

Limited evidence for sildenafil

Case reports, a few small open-label trials, and 1 prospective, randomized controlled trial (RCT) have evaluated sildenafil as an adjunctive treatment for serotonergic antidepressant-associated sexual dysfunction in women.^6,9 Nurnberg et al⁶ examined the efficacy of adjunctive sildenafil in women with SSRI-induced sexual dysfunction. This 8-week, placebo-controlled, double-blind, RCT used a flexible dose (50 or 100 mg), intention-to-treat design to assess the effect of sildenafil on 98 premenopausal women whose depression was in remission. Ten patients were taking the serotonin-norepinephrine inhibitor venlafaxine, 1 was taking the TCA clomipramine, and 87 were receiving an SSRI. Patients were instructed to take sildenafil or placebo 1 to 2 hours before sexual activity. The primary outcome was mean change from baseline on the Clinical Global Impression-Sexual Function (CGI-SF) scale.

Women taking sildenafil showed significant improvement compared with those taking placebo, with a treatment difference between groups of 0.8 (95% CI, 0.6 to 1.0; =.001). Additionally, 23% of sildenafil-treated patients reported no improvement with the intervention, compared with 73% of patients receiving placebo. Secondary outcomes using 3 validated scales that evaluated specific phases of sexual function found that patients’ orgasmic function significantly benefited from sildenafil treatment, while desire, arousal, and overall satisfaction were not significantly different.

Although these findings seem to support sildenafil for treating serotonergic antidepressant-associated sexual dysfunction in women, the study had a relatively small treatment effect in a well-defined patient population; therefore, replication in future trials and different patient populations is warranted. Overall, sildenafil was well tolerated, despite patient reports of headaches, flushing, visual disturbances, dyspepsia, nasal congestion, and palpitations. Finally, cost vs benefit should be considered; PDE5 inhibitors may not be covered by insurance or may require prior authorization.

CASE CONTINUED: Symptoms resolve

Bupropion is not an appropriate choice for Mrs. L because of her seizure risk. Mirtazapine is ruled out because in the past she experienced excessive somnolence that impaired her ability to function. You are not comfortable prescribing nefazodone because of its risk of hepatotoxicity or suggesting that Mrs. L take a “drug holiday” (stop taking any antidepressants for a short period) because of the risk of depressive relapse. You suggest that Mrs. L continue to take sertraline because sometimes antidepressant-induced sexual dysfunction resolves after ≥6 months of treatment with the same agent, but she is adamant that her relationship with her husband will deteriorate if she waits that long. She also declines cognitive-behavioral therapy because her job doesn’t allow the time or flexibility to commit to the sessions.

You prescribe sildenafil, 50 mg, and instruct Mrs. L to take 1 tablet 1 to 2 hours before sexual activity. This treatment improves her ability to achieve orgasm. She tolerates the drug well and after 8 weeks of treatment her CGI-SF score improves from 6 at baseline, indicating extreme dysfunction, to 2, indicating normal function. Ten months into her sertraline treatment, Mrs. L discovers she no longer requires sildenafil to achieve orgasm.

Related Resources

• Nurnberg HG. An evidence-based review updating the various treatment and management approaches to serotonin reuptake inhibitor-associated sexual dysfunction. Drugs Today (Barc). 2008;44(2):147-168.
• NIH Medline Plus. Sexual problems in women. www.nlm.nih.gov/medlineplus/sexualproblemsinwomen.html.
• Sturpe DA, Mertens MK, Scoville C. What are the treatment options for SSRI-related sexual dysfunction? J Fam Pract. 2002;51(8):681.

Drug Brand Names

• Amantadine • Symadine, Symmetrel
• Avanafil • Stendra
• Bupropion • Wellbutrin, Zyban
• Buspirone • BuSpar
• Citalopram • Celexa
• Clomipramine • Anafranil
• Fluoxetine • Prozac
• Granisetron • Kytril
• Mirtazapine • Remeron
• Nefazodone • Serzone
• Nitroglycerin • Nitrostat
• Nortriptyline • Pamelor
• Olanzapine • Zyprexa
• Paroxetine • Paxil
• Sertraline • Zoloft
• Sildenafil • Viagra
• Tadalafil • Cialis
• Vardenafil • Levitra
• Venlafaxine • Effexor

Disclosures

Dr. Burghardt receives grant or research support from the University of Michigan Depression Center.

Ms. Gardner reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Mrs. L, age 27, has a history of major depressive disorder with symptoms of anxiety. She was managed successfully for 2 years with bupropion XL, 300 mg/d, but was switched to venlafaxine, titrated to 225 mg/d, after she developed seizures secondary to a head injury sustained in a car accident. After the switch, Mrs. L’s mood deteriorated and she was hospitalized. Since then, she’s received several medication trials, including paroxetine, 30 mg/d, a selective serotonin reuptake inhibitor (SSRI), and the tricyclic antidepressant (TCA) nortriptyline, 75 mg/d, but she could not tolerate these medications because of severe xerostomia.

After taking sertraline, 150 mg/d, for 8 weeks, Mrs. L improves and has a Patient Health Questionnaire score of 6, indicating mild depression. Her initial complaints of diarrhea and nausea have resolved, but Mrs. L now reports that she and her husband are having marital difficulties because she cannot achieve orgasm during sexual intercourse. She did not have this problem when she was taking bupropion. Her husband occasionally takes the phosphodiesterase type 5 (PDE5) inhibitor sildenafil before intercourse, and Mrs. L asks you if this medication will help her achieve orgasm.

DSM-IV-TR defines sexual dysfunction as disturbances in sexual desire and/or in the sexual response cycle (excitement, plateau, orgasm, and resolution) that result in marked distress and interpersonal difficulty.¹ Sexual dysfunction can occur with the use of any antidepressant with serotonergic activity; it affects an estimated 50% to 70% of patients who take SSRIs.² Sexual dysfunction can occur with all SSRIs; however, higher rates of sexual dysfunction are found with citalopram, fluoxetine, paroxetine, and sertraline.³ Studies have suggested there may be a dose-side effect relationship with SSRI-induced sexual dysfunction.⁴

Several factors can increase a patient’s risk of sexual dysfunction and should be considered before prescribing an antidepressant or when a patient presents with new or worsening sexual dysfunction (Table 1).⁵ In general, nonserotonergic agents such as bupropion, mirtazapine, and nefazodone are associated with lower rates of sexual dysfunction. The pharmacology of these agents explains their decreased propensity to cause sexual dysfunction. These agents increase levels of dopamine in the mesolimbic dopaminergic system either by blocking reuptake (bupropion) or antagonizing the serotonin subtype-2 receptor and facilitating disinhibition of decreased dopamine downstream (nefazodone and mirtazapine).

Table 1

Risk factors for sexual dysfunction

One option for treating antidepressant-induced sexual dysfunction in women is PDE5 inhibitors, which are used to treat erectile dysfunction (ED). These medications ameliorate ED by inhibiting degradation of cyclic guanosine monophosphate by PDE5, which increases blood flow to the penis during sexual stimulation. Although these medications are not FDA-approved for treating sexual dysfunction in women, adjunctive PDE5 inhibitor treatment may be beneficial for sexual dysfunction in females because similar mediators, such as nitric oxide and cyclic guanosine monophosphate, involved in the nonadrenergic-noncholinergic signaling that controls sexual stimulation in men also are found in female genital tissue.⁶

When treating a woman with SSRI-induced sexual dysfunction, consider nonpharmacologic treatments both before and during pharmacotherapy (Table 2).^7,8 See Table 3 for a comparison of pharmacokinetics, side effects, and drug interactions of the 4 FDA-approved PDE5 inhibitors—avanafil, sildenafil, tadalafil, and vardenafil.

Table 2

Management strategies for SSRI-induced sexual dysfunction

Table 3

Phosphodiesterase type 5 inhibitors: A comparison

Limited evidence for sildenafil

Case reports, a few small open-label trials, and 1 prospective, randomized controlled trial (RCT) have evaluated sildenafil as an adjunctive treatment for serotonergic antidepressant-associated sexual dysfunction in women.^6,9 Nurnberg et al⁶ examined the efficacy of adjunctive sildenafil in women with SSRI-induced sexual dysfunction. This 8-week, placebo-controlled, double-blind, RCT used a flexible dose (50 or 100 mg), intention-to-treat design to assess the effect of sildenafil on 98 premenopausal women whose depression was in remission. Ten patients were taking the serotonin-norepinephrine inhibitor venlafaxine, 1 was taking the TCA clomipramine, and 87 were receiving an SSRI. Patients were instructed to take sildenafil or placebo 1 to 2 hours before sexual activity. The primary outcome was mean change from baseline on the Clinical Global Impression-Sexual Function (CGI-SF) scale.

Women taking sildenafil showed significant improvement compared with those taking placebo, with a treatment difference between groups of 0.8 (95% CI, 0.6 to 1.0; =.001). Additionally, 23% of sildenafil-treated patients reported no improvement with the intervention, compared with 73% of patients receiving placebo. Secondary outcomes using 3 validated scales that evaluated specific phases of sexual function found that patients’ orgasmic function significantly benefited from sildenafil treatment, while desire, arousal, and overall satisfaction were not significantly different.

Although these findings seem to support sildenafil for treating serotonergic antidepressant-associated sexual dysfunction in women, the study had a relatively small treatment effect in a well-defined patient population; therefore, replication in future trials and different patient populations is warranted. Overall, sildenafil was well tolerated, despite patient reports of headaches, flushing, visual disturbances, dyspepsia, nasal congestion, and palpitations. Finally, cost vs benefit should be considered; PDE5 inhibitors may not be covered by insurance or may require prior authorization.

CASE CONTINUED: Symptoms resolve

Bupropion is not an appropriate choice for Mrs. L because of her seizure risk. Mirtazapine is ruled out because in the past she experienced excessive somnolence that impaired her ability to function. You are not comfortable prescribing nefazodone because of its risk of hepatotoxicity or suggesting that Mrs. L take a “drug holiday” (stop taking any antidepressants for a short period) because of the risk of depressive relapse. You suggest that Mrs. L continue to take sertraline because sometimes antidepressant-induced sexual dysfunction resolves after ≥6 months of treatment with the same agent, but she is adamant that her relationship with her husband will deteriorate if she waits that long. She also declines cognitive-behavioral therapy because her job doesn’t allow the time or flexibility to commit to the sessions.

You prescribe sildenafil, 50 mg, and instruct Mrs. L to take 1 tablet 1 to 2 hours before sexual activity. This treatment improves her ability to achieve orgasm. She tolerates the drug well and after 8 weeks of treatment her CGI-SF score improves from 6 at baseline, indicating extreme dysfunction, to 2, indicating normal function. Ten months into her sertraline treatment, Mrs. L discovers she no longer requires sildenafil to achieve orgasm.

Related Resources

• Nurnberg HG. An evidence-based review updating the various treatment and management approaches to serotonin reuptake inhibitor-associated sexual dysfunction. Drugs Today (Barc). 2008;44(2):147-168.
• NIH Medline Plus. Sexual problems in women. www.nlm.nih.gov/medlineplus/sexualproblemsinwomen.html.
• Sturpe DA, Mertens MK, Scoville C. What are the treatment options for SSRI-related sexual dysfunction? J Fam Pract. 2002;51(8):681.

Drug Brand Names

• Amantadine • Symadine, Symmetrel
• Avanafil • Stendra
• Bupropion • Wellbutrin, Zyban
• Buspirone • BuSpar
• Citalopram • Celexa
• Clomipramine • Anafranil
• Fluoxetine • Prozac
• Granisetron • Kytril
• Mirtazapine • Remeron
• Nefazodone • Serzone
• Nitroglycerin • Nitrostat
• Nortriptyline • Pamelor
• Olanzapine • Zyprexa
• Paroxetine • Paxil
• Sertraline • Zoloft
• Sildenafil • Viagra
• Tadalafil • Cialis
• Vardenafil • Levitra
• Venlafaxine • Effexor

Disclosures

Dr. Burghardt receives grant or research support from the University of Michigan Depression Center.

Ms. Gardner reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Mrs. L, age 27, has a history of major depressive disorder with symptoms of anxiety. She was managed successfully for 2 years with bupropion XL, 300 mg/d, but was switched to venlafaxine, titrated to 225 mg/d, after she developed seizures secondary to a head injury sustained in a car accident. After the switch, Mrs. L’s mood deteriorated and she was hospitalized. Since then, she’s received several medication trials, including paroxetine, 30 mg/d, a selective serotonin reuptake inhibitor (SSRI), and the tricyclic antidepressant (TCA) nortriptyline, 75 mg/d, but she could not tolerate these medications because of severe xerostomia.

After taking sertraline, 150 mg/d, for 8 weeks, Mrs. L improves and has a Patient Health Questionnaire score of 6, indicating mild depression. Her initial complaints of diarrhea and nausea have resolved, but Mrs. L now reports that she and her husband are having marital difficulties because she cannot achieve orgasm during sexual intercourse. She did not have this problem when she was taking bupropion. Her husband occasionally takes the phosphodiesterase type 5 (PDE5) inhibitor sildenafil before intercourse, and Mrs. L asks you if this medication will help her achieve orgasm.

DSM-IV-TR defines sexual dysfunction as disturbances in sexual desire and/or in the sexual response cycle (excitement, plateau, orgasm, and resolution) that result in marked distress and interpersonal difficulty.¹ Sexual dysfunction can occur with the use of any antidepressant with serotonergic activity; it affects an estimated 50% to 70% of patients who take SSRIs.² Sexual dysfunction can occur with all SSRIs; however, higher rates of sexual dysfunction are found with citalopram, fluoxetine, paroxetine, and sertraline.³ Studies have suggested there may be a dose-side effect relationship with SSRI-induced sexual dysfunction.⁴

Several factors can increase a patient’s risk of sexual dysfunction and should be considered before prescribing an antidepressant or when a patient presents with new or worsening sexual dysfunction (Table 1).⁵ In general, nonserotonergic agents such as bupropion, mirtazapine, and nefazodone are associated with lower rates of sexual dysfunction. The pharmacology of these agents explains their decreased propensity to cause sexual dysfunction. These agents increase levels of dopamine in the mesolimbic dopaminergic system either by blocking reuptake (bupropion) or antagonizing the serotonin subtype-2 receptor and facilitating disinhibition of decreased dopamine downstream (nefazodone and mirtazapine).

Table 1

Risk factors for sexual dysfunction

One option for treating antidepressant-induced sexual dysfunction in women is PDE5 inhibitors, which are used to treat erectile dysfunction (ED). These medications ameliorate ED by inhibiting degradation of cyclic guanosine monophosphate by PDE5, which increases blood flow to the penis during sexual stimulation. Although these medications are not FDA-approved for treating sexual dysfunction in women, adjunctive PDE5 inhibitor treatment may be beneficial for sexual dysfunction in females because similar mediators, such as nitric oxide and cyclic guanosine monophosphate, involved in the nonadrenergic-noncholinergic signaling that controls sexual stimulation in men also are found in female genital tissue.⁶

When treating a woman with SSRI-induced sexual dysfunction, consider nonpharmacologic treatments both before and during pharmacotherapy (Table 2).^7,8 See Table 3 for a comparison of pharmacokinetics, side effects, and drug interactions of the 4 FDA-approved PDE5 inhibitors—avanafil, sildenafil, tadalafil, and vardenafil.

Table 2

Management strategies for SSRI-induced sexual dysfunction

Table 3

Phosphodiesterase type 5 inhibitors: A comparison

Limited evidence for sildenafil

Case reports, a few small open-label trials, and 1 prospective, randomized controlled trial (RCT) have evaluated sildenafil as an adjunctive treatment for serotonergic antidepressant-associated sexual dysfunction in women.^6,9 Nurnberg et al⁶ examined the efficacy of adjunctive sildenafil in women with SSRI-induced sexual dysfunction. This 8-week, placebo-controlled, double-blind, RCT used a flexible dose (50 or 100 mg), intention-to-treat design to assess the effect of sildenafil on 98 premenopausal women whose depression was in remission. Ten patients were taking the serotonin-norepinephrine inhibitor venlafaxine, 1 was taking the TCA clomipramine, and 87 were receiving an SSRI. Patients were instructed to take sildenafil or placebo 1 to 2 hours before sexual activity. The primary outcome was mean change from baseline on the Clinical Global Impression-Sexual Function (CGI-SF) scale.

Women taking sildenafil showed significant improvement compared with those taking placebo, with a treatment difference between groups of 0.8 (95% CI, 0.6 to 1.0; =.001). Additionally, 23% of sildenafil-treated patients reported no improvement with the intervention, compared with 73% of patients receiving placebo. Secondary outcomes using 3 validated scales that evaluated specific phases of sexual function found that patients’ orgasmic function significantly benefited from sildenafil treatment, while desire, arousal, and overall satisfaction were not significantly different.

Although these findings seem to support sildenafil for treating serotonergic antidepressant-associated sexual dysfunction in women, the study had a relatively small treatment effect in a well-defined patient population; therefore, replication in future trials and different patient populations is warranted. Overall, sildenafil was well tolerated, despite patient reports of headaches, flushing, visual disturbances, dyspepsia, nasal congestion, and palpitations. Finally, cost vs benefit should be considered; PDE5 inhibitors may not be covered by insurance or may require prior authorization.

CASE CONTINUED: Symptoms resolve

Bupropion is not an appropriate choice for Mrs. L because of her seizure risk. Mirtazapine is ruled out because in the past she experienced excessive somnolence that impaired her ability to function. You are not comfortable prescribing nefazodone because of its risk of hepatotoxicity or suggesting that Mrs. L take a “drug holiday” (stop taking any antidepressants for a short period) because of the risk of depressive relapse. You suggest that Mrs. L continue to take sertraline because sometimes antidepressant-induced sexual dysfunction resolves after ≥6 months of treatment with the same agent, but she is adamant that her relationship with her husband will deteriorate if she waits that long. She also declines cognitive-behavioral therapy because her job doesn’t allow the time or flexibility to commit to the sessions.

You prescribe sildenafil, 50 mg, and instruct Mrs. L to take 1 tablet 1 to 2 hours before sexual activity. This treatment improves her ability to achieve orgasm. She tolerates the drug well and after 8 weeks of treatment her CGI-SF score improves from 6 at baseline, indicating extreme dysfunction, to 2, indicating normal function. Ten months into her sertraline treatment, Mrs. L discovers she no longer requires sildenafil to achieve orgasm.

Related Resources

• Nurnberg HG. An evidence-based review updating the various treatment and management approaches to serotonin reuptake inhibitor-associated sexual dysfunction. Drugs Today (Barc). 2008;44(2):147-168.
• NIH Medline Plus. Sexual problems in women. www.nlm.nih.gov/medlineplus/sexualproblemsinwomen.html.
• Sturpe DA, Mertens MK, Scoville C. What are the treatment options for SSRI-related sexual dysfunction? J Fam Pract. 2002;51(8):681.

Drug Brand Names

• Amantadine • Symadine, Symmetrel
• Avanafil • Stendra
• Bupropion • Wellbutrin, Zyban
• Buspirone • BuSpar
• Citalopram • Celexa
• Clomipramine • Anafranil
• Fluoxetine • Prozac
• Granisetron • Kytril
• Mirtazapine • Remeron
• Nefazodone • Serzone
• Nitroglycerin • Nitrostat
• Nortriptyline • Pamelor
• Olanzapine • Zyprexa
• Paroxetine • Paxil
• Sertraline • Zoloft
• Sildenafil • Viagra
• Tadalafil • Cialis
• Vardenafil • Levitra
• Venlafaxine • Effexor

Disclosures

Dr. Burghardt receives grant or research support from the University of Michigan Depression Center.

Ms. Gardner reports no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.