James V. Felicetta, MD

This editorial could have quite a negative impact on my fellow endocrinologists. That’s because it could result in a significant reduction in the number of referrals to endocrinologists for patients with low thyroid-stimulating hormone (TSH) levels, at least that’s a best-case scenario. Despite the potentially negative impact on endocrine referrals, I feel compelled to suggest that you may be overreacting to low TSH levels in many of your patients.

You know what I mean—you see moderately suppressed TSH levels all of the time in day-to-day clinical practice. And of course, as a faithful provider, you harken back to your training days and remember the basics of the thyroid-pituitary axis. Theoretically at least, the pituitary gland secretes TSH in proportion to the need for the thyroid gland to put out more thyroid hormone. If the thyroid gland starts to fail for any reason (Hashimoto’s thyroiditis, an autoimmune disease, is by far the most common cause), the pituitary will detect that there is an insufficient amount of thyroid hormone floating around and will secrete more TSH to try to stimulate the thyroid to pump out more hormone.

Conversely, if the thyroid gland becomes overactive for any reason (Graves’ disease, another autoimmune phenomenon, is the most common cause here), then the secretion of TSH will be suppressed by the excess thyroid hormone. It is seemingly straightforward: An elevated TSH means hypothyroidism, and a suppressed TSH means an overactive thyroid gland.

Alas, dear reader, if only it were so simple! It turns out that a very large fraction of the low TSH levels seen in clinical practice are not related at all to an overactive thyroid gland, and no therapeutic intervention of any sort is indicated. In light of the intricate feedback loop, which controls the delicate balance between the secretion of thyroid hormone on the one hand and that of TSH on the other, how can that possibly be true?

The answer is that the feedback loop is nowhere near as simple and straightforward as you were taught as an eager student of human physiology. The thyrotroph cells in the pituitary, the ones that secrete TSH, do indeed respond rather exquisitely to the ambient levels of circulating thyroid hormones. But they are also very susceptible to a number of other circulating compounds that are quite capable of suppressing their output of TSH every bit as effectively as thyroid hormones.

The classic setting in which TSH levels are suppressed in the absence of true thyrotoxicosis is euthyroid sick syndrome (ESS). I always tell my trainees that the surest way to find patients with ESS is simply to ask for directions to the intensive care unit (ICU). Assuming that the patients in the ICU truly need to be there, every single one of them will display thyroid hormone changes consistent with ESS. It’s still not clear whether or not ESS is an adaptive or protective mechanism, but it occurs in virtually all patients who are sufficiently sick.

The first manifestation of ESS is low T3 syndrome, wherein the conversion of T4 to the more metabolically potent T3 is markedly reduced. Since T3 is by far the more physiologically active of these 2 thyroid hormones, the net effect of the block in conversion to T3 is a down-regulation of the thyroid axis. The dialing back of thyroid effect may well be a protective physiologic mechanism so the body can focus on defending against whatever severe physiologic insult set the whole process in motion in the first place. It may represent a turning down of the metabolic thermometer or burn rate.

If the underlying illness persists or worsens, usually the next manifestations of ESS are a suppression of TSH, and then a concomitant suppression of the production of T4 from the thyroid gland. So most truly ill individuals experience an uncoupling of the usual relationship between TSH levels and thyroid hormone levels. The suppression of the TSH levels in ESS is generally attributed to the circulating presence of abnormally high levels of cytokines associated with severe illness, including interleukins and a number of other potent mediators of inflammation.

It also turns out that less ill patients can also experience a suppression of TSH levels due to a number of circulating compounds, prominent among them are corticosteroids, catecholamines, and opioids. Thus, many patients who have chronically elevated levels of corticosteroids, catecholamines, or opioids will also have relatively suppressed levels of TSH without a hint that they are suffering from thyrotoxicosis.

Any endocrinologist who has been reading this is probably bored, but hopefully the rest of you have gained just a small bit of insight into the multiplicity of factors that can lead to low levels of TSH. In a perfectly healthy person with no reason to have elevated levels of corticosteroids or catecholamines, a low TSH level does indeed raise the concern for thyrotoxicosis, especially if the TSH is not measurable, as it usually is with true thyrotoxicosis. But in patients who are ill, all bets are off. A low TSH level is very probably not an indicator of excess circulating thyroid hormone.

It is hoped my fellow endocrinologists will now receive fewer consults for low TSH levels, and they can concentrate on something more important, such as trying to tame those pesky glucose levels in our ever-increasing glut of diabetic patients.

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

This editorial could have quite a negative impact on my fellow endocrinologists. That’s because it could result in a significant reduction in the number of referrals to endocrinologists for patients with low thyroid-stimulating hormone (TSH) levels, at least that’s a best-case scenario. Despite the potentially negative impact on endocrine referrals, I feel compelled to suggest that you may be overreacting to low TSH levels in many of your patients.

You know what I mean—you see moderately suppressed TSH levels all of the time in day-to-day clinical practice. And of course, as a faithful provider, you harken back to your training days and remember the basics of the thyroid-pituitary axis. Theoretically at least, the pituitary gland secretes TSH in proportion to the need for the thyroid gland to put out more thyroid hormone. If the thyroid gland starts to fail for any reason (Hashimoto’s thyroiditis, an autoimmune disease, is by far the most common cause), the pituitary will detect that there is an insufficient amount of thyroid hormone floating around and will secrete more TSH to try to stimulate the thyroid to pump out more hormone.

Conversely, if the thyroid gland becomes overactive for any reason (Graves’ disease, another autoimmune phenomenon, is the most common cause here), then the secretion of TSH will be suppressed by the excess thyroid hormone. It is seemingly straightforward: An elevated TSH means hypothyroidism, and a suppressed TSH means an overactive thyroid gland.

Alas, dear reader, if only it were so simple! It turns out that a very large fraction of the low TSH levels seen in clinical practice are not related at all to an overactive thyroid gland, and no therapeutic intervention of any sort is indicated. In light of the intricate feedback loop, which controls the delicate balance between the secretion of thyroid hormone on the one hand and that of TSH on the other, how can that possibly be true?

The answer is that the feedback loop is nowhere near as simple and straightforward as you were taught as an eager student of human physiology. The thyrotroph cells in the pituitary, the ones that secrete TSH, do indeed respond rather exquisitely to the ambient levels of circulating thyroid hormones. But they are also very susceptible to a number of other circulating compounds that are quite capable of suppressing their output of TSH every bit as effectively as thyroid hormones.

The classic setting in which TSH levels are suppressed in the absence of true thyrotoxicosis is euthyroid sick syndrome (ESS). I always tell my trainees that the surest way to find patients with ESS is simply to ask for directions to the intensive care unit (ICU). Assuming that the patients in the ICU truly need to be there, every single one of them will display thyroid hormone changes consistent with ESS. It’s still not clear whether or not ESS is an adaptive or protective mechanism, but it occurs in virtually all patients who are sufficiently sick.

The first manifestation of ESS is low T3 syndrome, wherein the conversion of T4 to the more metabolically potent T3 is markedly reduced. Since T3 is by far the more physiologically active of these 2 thyroid hormones, the net effect of the block in conversion to T3 is a down-regulation of the thyroid axis. The dialing back of thyroid effect may well be a protective physiologic mechanism so the body can focus on defending against whatever severe physiologic insult set the whole process in motion in the first place. It may represent a turning down of the metabolic thermometer or burn rate.

If the underlying illness persists or worsens, usually the next manifestations of ESS are a suppression of TSH, and then a concomitant suppression of the production of T4 from the thyroid gland. So most truly ill individuals experience an uncoupling of the usual relationship between TSH levels and thyroid hormone levels. The suppression of the TSH levels in ESS is generally attributed to the circulating presence of abnormally high levels of cytokines associated with severe illness, including interleukins and a number of other potent mediators of inflammation.

It also turns out that less ill patients can also experience a suppression of TSH levels due to a number of circulating compounds, prominent among them are corticosteroids, catecholamines, and opioids. Thus, many patients who have chronically elevated levels of corticosteroids, catecholamines, or opioids will also have relatively suppressed levels of TSH without a hint that they are suffering from thyrotoxicosis.

Any endocrinologist who has been reading this is probably bored, but hopefully the rest of you have gained just a small bit of insight into the multiplicity of factors that can lead to low levels of TSH. In a perfectly healthy person with no reason to have elevated levels of corticosteroids or catecholamines, a low TSH level does indeed raise the concern for thyrotoxicosis, especially if the TSH is not measurable, as it usually is with true thyrotoxicosis. But in patients who are ill, all bets are off. A low TSH level is very probably not an indicator of excess circulating thyroid hormone.

It is hoped my fellow endocrinologists will now receive fewer consults for low TSH levels, and they can concentrate on something more important, such as trying to tame those pesky glucose levels in our ever-increasing glut of diabetic patients.

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

This editorial could have quite a negative impact on my fellow endocrinologists. That’s because it could result in a significant reduction in the number of referrals to endocrinologists for patients with low thyroid-stimulating hormone (TSH) levels, at least that’s a best-case scenario. Despite the potentially negative impact on endocrine referrals, I feel compelled to suggest that you may be overreacting to low TSH levels in many of your patients.

You know what I mean—you see moderately suppressed TSH levels all of the time in day-to-day clinical practice. And of course, as a faithful provider, you harken back to your training days and remember the basics of the thyroid-pituitary axis. Theoretically at least, the pituitary gland secretes TSH in proportion to the need for the thyroid gland to put out more thyroid hormone. If the thyroid gland starts to fail for any reason (Hashimoto’s thyroiditis, an autoimmune disease, is by far the most common cause), the pituitary will detect that there is an insufficient amount of thyroid hormone floating around and will secrete more TSH to try to stimulate the thyroid to pump out more hormone.

Conversely, if the thyroid gland becomes overactive for any reason (Graves’ disease, another autoimmune phenomenon, is the most common cause here), then the secretion of TSH will be suppressed by the excess thyroid hormone. It is seemingly straightforward: An elevated TSH means hypothyroidism, and a suppressed TSH means an overactive thyroid gland.

Alas, dear reader, if only it were so simple! It turns out that a very large fraction of the low TSH levels seen in clinical practice are not related at all to an overactive thyroid gland, and no therapeutic intervention of any sort is indicated. In light of the intricate feedback loop, which controls the delicate balance between the secretion of thyroid hormone on the one hand and that of TSH on the other, how can that possibly be true?

The answer is that the feedback loop is nowhere near as simple and straightforward as you were taught as an eager student of human physiology. The thyrotroph cells in the pituitary, the ones that secrete TSH, do indeed respond rather exquisitely to the ambient levels of circulating thyroid hormones. But they are also very susceptible to a number of other circulating compounds that are quite capable of suppressing their output of TSH every bit as effectively as thyroid hormones.

The classic setting in which TSH levels are suppressed in the absence of true thyrotoxicosis is euthyroid sick syndrome (ESS). I always tell my trainees that the surest way to find patients with ESS is simply to ask for directions to the intensive care unit (ICU). Assuming that the patients in the ICU truly need to be there, every single one of them will display thyroid hormone changes consistent with ESS. It’s still not clear whether or not ESS is an adaptive or protective mechanism, but it occurs in virtually all patients who are sufficiently sick.

The first manifestation of ESS is low T3 syndrome, wherein the conversion of T4 to the more metabolically potent T3 is markedly reduced. Since T3 is by far the more physiologically active of these 2 thyroid hormones, the net effect of the block in conversion to T3 is a down-regulation of the thyroid axis. The dialing back of thyroid effect may well be a protective physiologic mechanism so the body can focus on defending against whatever severe physiologic insult set the whole process in motion in the first place. It may represent a turning down of the metabolic thermometer or burn rate.

If the underlying illness persists or worsens, usually the next manifestations of ESS are a suppression of TSH, and then a concomitant suppression of the production of T4 from the thyroid gland. So most truly ill individuals experience an uncoupling of the usual relationship between TSH levels and thyroid hormone levels. The suppression of the TSH levels in ESS is generally attributed to the circulating presence of abnormally high levels of cytokines associated with severe illness, including interleukins and a number of other potent mediators of inflammation.

It also turns out that less ill patients can also experience a suppression of TSH levels due to a number of circulating compounds, prominent among them are corticosteroids, catecholamines, and opioids. Thus, many patients who have chronically elevated levels of corticosteroids, catecholamines, or opioids will also have relatively suppressed levels of TSH without a hint that they are suffering from thyrotoxicosis.

Any endocrinologist who has been reading this is probably bored, but hopefully the rest of you have gained just a small bit of insight into the multiplicity of factors that can lead to low levels of TSH. In a perfectly healthy person with no reason to have elevated levels of corticosteroids or catecholamines, a low TSH level does indeed raise the concern for thyrotoxicosis, especially if the TSH is not measurable, as it usually is with true thyrotoxicosis. But in patients who are ill, all bets are off. A low TSH level is very probably not an indicator of excess circulating thyroid hormone.

It is hoped my fellow endocrinologists will now receive fewer consults for low TSH levels, and they can concentrate on something more important, such as trying to tame those pesky glucose levels in our ever-increasing glut of diabetic patients.

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

At first it seems to be a fairly straightforward proposition. The older gentleman you are seeing in the clinic reports that he has been running rather low on energy in recent weeks, and he also mentions that there’s not much lead in his pencil these days. As a conscientious clinician, you immediately entertain the possibility that hypogonadism might explain some of his symptoms. You dutifully order up a total testosterone level, and then a free testosterone level when the total comes back low, recognizing that binding protein abnormalities might produce a low total even when the clinically relevant free level is still normal. Both levels do come back well below the age-adjusted lower limits of normal, which gives you some transient level of satisfaction that you have identified a very significant factor contributing to your patient’s difficulties.

You have confirmed a deficiency of a major hormone, and it seems logical that you would want to restore the hormone level to normal in this particular patient. But before you reach for your prescription pad (or your mouse), a fundamental question hangs uneasily in the air. Are you going to be doing more harm than good by prescribing testosterone replacement therapy (TRT) to this rather trusting older fellow? In light of recent studies, might you actually increase this patent’s chances of a heart attack or a stroke? That would not be a nice thing to do to this pleasant older gentleman. (As a newly minted senior citizen, I pray mightily that my own caregivers adhere rigorously to Hippocrates’ hoary admonition to, above all, do no harm.)

I’m not going to be able to resolve this clinical conundrum definitively in this editorial. (Please don’t stop reading just yet!) But maybe a review of the pros and cons for testosterone replacement therapy can help you faithful readers gain just a little bit better sense of the operative risk/benefit considerations at play here.

Let’s look first at the case for prescribing TRT when the laboratory test values show definitive evidence of low testosterone levels. I don’t want to delve into the distracting issue of which form of testosterone replacement to consider, which pits injections vs gels vs patches vs pills (don’t use the potentially hepatotoxic methyltestosterone pills passed out like candy by some urologists). Apart from the possibly relevant issue of peaks and troughs seen with injection therapy, the same risk/benefit considerations pretty much apply to all forms of TRT.

The benefits of TRT clearly include an increase in lean muscle mass, an increase in red blood cell concentration due to the hematopoetic effects of the male hormone, and a reduction in both the total amount and the percentage of body fat. A number of studies have shown that testosterone enhances insulin sensitivity—surely a good thing given the massive number of older patients with either prediabetes or full-blown type 2 diabetes. Some men also report a significant increase in their hard-to-define-but-still important sense of manliness, and sometimes a major improvement in their ability to perform in the sack. The latter effects, though, are often very modest and of considerably less potency (sorry, pun intended) than seen with sildenafil or one of the other PDE-5 inhibitors. In spite of all these seemingly positive effects, the clear majority of men report that they really don’t feel much different after starting on TRT, and many discontinue it on their own after relatively short periods, especially those enduring intramuscular injections every 2 weeks.

So the benefits derived from TRT are not really very impressive in many patients. What about the downside of giving testosterone? Surely there can’t be any problems associated with simply replacing an important hormone that has fallen to low levels? After all, we don’t hesitate to give thyroid hormone to hypothyroid patients, to give growth hormone to children with low levels of this critical hormone, or to give insulin to diabetic patients whose pancreases are not producing enough of that life-saving hormone.

For a very long time the risk/benefit arguments over whether or not to give TRT were almost entirely theoretical. Those in favor cited the several aforementioned benefits, and those in opposition decried replacement therapy as a perverse form of tinkering with nature by trying to alter the natural decline in the levels of certain hormones that were part and parcel of the natural aging process.

Then along came 3 rather worrisome studies in fairly rapid-fire succession, which seamed collectively to deliver a true body blow to TRT. However, a closer examination of these studies reveals that each is so severely flawed that no meaningful conclusions can be derived from any of them.

The Testosterone in Older Men with Sarcopenia (TOM) trial was a randomized trial of TRT vs placebo in older men (mean age 74 years) with mobility limitations (sarcopenia, after all, means decreased muscle bulk) and a high prevalence of chronic disease.¹ The trial was stopped early because of a much higher occurrence of self-reported cardiovascular-related adverse events. However, these adverse events were extremely disparate and were all self-reported; none had been prespecified outcomes. Any objective observer would have to conclude that the study was poorly designed and that no meaningful conclusions can be drawn from its premature termination.

The second trial that seemed to cast doubt on the safety of TRT suffered from an even worse design. It was a retrospective cohort study of 8,709 veterans aged 60 to 64 years with low testosterone levels who were undergoing coronary artery angiography. The authors reported in the Journal of the American Medical Association that those receiving testosterone therapy had a higher risk of experiencing a composite outcome of all-cause mortality, myocardial infarction (MI), or cerebrovascular accident than did those who had not received testosterone therapy (hazard ratio [HR] = 1.29; 95% confidence interval [CI]: 1.04-1.58).² Right off the bat, you should be very wary of any HR emanating from a retrospective study that shows a small increase in risk of 29%; it’s only when a HR is 2.0 or more that it’s likely you’re looking at a real phenomenon. But to add insult to injury, the percentage of actual adverse outcomes was actually SMALLER in those taking testosterone than in those who did not get any! The authors had used such an incredibly tortured series of risk adjustments for a variety of comorbidities that they actually managed to stand the raw numbers on their head.

The third study, which had seamed at first blush to demonstrate cardiovascular toxicity of TRT, was a much larger retrospective cohort study of 55,793 men who had received replacement testosterone.³ The authors reported an increase in the relative risk of MI in the first 3 months after starting testosterone compared with the risk of MI in the same men in the prior year (relative risk [RR] = 1.36). However, the much more important absolute risk increase was very, very low, with only an additional 1.25 cases of MI seen over 1,000 patient-years. Apart from the fact that a RR of 1.36 is most unimpressive in a retrospective study, the simple fact that the men were older by a few months after TRT is probably more than adequate to explain this tiny increase in apparent risk.

The FDA has monitored these studies closely and has chosen not to make a determination that there is an increased risk of cardiovascular events associated with TRT. That is not at all the same as saying that it has been proven to be completely free of cardiovascular risk; rather it is a common-sense acknowledgment that there is not any convincing evidence to date of such a risk.

Thus, the conscientious clinician is left to conclude that TRT is a reasonable option in symptomatic patients who have been shown to have low levels of free testosterone. It has not been conclusively demonstrated that TRT will have significant beneficial effects, but neither has it been proven to have any true cardiovascular toxicity. It is a therapy worth trying in those symptomatic patients who understand that they will be receiving therapy of uncertain benefit, if any, and with the possibility of uncertain risk, if any.

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

At first it seems to be a fairly straightforward proposition. The older gentleman you are seeing in the clinic reports that he has been running rather low on energy in recent weeks, and he also mentions that there’s not much lead in his pencil these days. As a conscientious clinician, you immediately entertain the possibility that hypogonadism might explain some of his symptoms. You dutifully order up a total testosterone level, and then a free testosterone level when the total comes back low, recognizing that binding protein abnormalities might produce a low total even when the clinically relevant free level is still normal. Both levels do come back well below the age-adjusted lower limits of normal, which gives you some transient level of satisfaction that you have identified a very significant factor contributing to your patient’s difficulties.

You have confirmed a deficiency of a major hormone, and it seems logical that you would want to restore the hormone level to normal in this particular patient. But before you reach for your prescription pad (or your mouse), a fundamental question hangs uneasily in the air. Are you going to be doing more harm than good by prescribing testosterone replacement therapy (TRT) to this rather trusting older fellow? In light of recent studies, might you actually increase this patent’s chances of a heart attack or a stroke? That would not be a nice thing to do to this pleasant older gentleman. (As a newly minted senior citizen, I pray mightily that my own caregivers adhere rigorously to Hippocrates’ hoary admonition to, above all, do no harm.)

I’m not going to be able to resolve this clinical conundrum definitively in this editorial. (Please don’t stop reading just yet!) But maybe a review of the pros and cons for testosterone replacement therapy can help you faithful readers gain just a little bit better sense of the operative risk/benefit considerations at play here.

Let’s look first at the case for prescribing TRT when the laboratory test values show definitive evidence of low testosterone levels. I don’t want to delve into the distracting issue of which form of testosterone replacement to consider, which pits injections vs gels vs patches vs pills (don’t use the potentially hepatotoxic methyltestosterone pills passed out like candy by some urologists). Apart from the possibly relevant issue of peaks and troughs seen with injection therapy, the same risk/benefit considerations pretty much apply to all forms of TRT.

The benefits of TRT clearly include an increase in lean muscle mass, an increase in red blood cell concentration due to the hematopoetic effects of the male hormone, and a reduction in both the total amount and the percentage of body fat. A number of studies have shown that testosterone enhances insulin sensitivity—surely a good thing given the massive number of older patients with either prediabetes or full-blown type 2 diabetes. Some men also report a significant increase in their hard-to-define-but-still important sense of manliness, and sometimes a major improvement in their ability to perform in the sack. The latter effects, though, are often very modest and of considerably less potency (sorry, pun intended) than seen with sildenafil or one of the other PDE-5 inhibitors. In spite of all these seemingly positive effects, the clear majority of men report that they really don’t feel much different after starting on TRT, and many discontinue it on their own after relatively short periods, especially those enduring intramuscular injections every 2 weeks.

So the benefits derived from TRT are not really very impressive in many patients. What about the downside of giving testosterone? Surely there can’t be any problems associated with simply replacing an important hormone that has fallen to low levels? After all, we don’t hesitate to give thyroid hormone to hypothyroid patients, to give growth hormone to children with low levels of this critical hormone, or to give insulin to diabetic patients whose pancreases are not producing enough of that life-saving hormone.

For a very long time the risk/benefit arguments over whether or not to give TRT were almost entirely theoretical. Those in favor cited the several aforementioned benefits, and those in opposition decried replacement therapy as a perverse form of tinkering with nature by trying to alter the natural decline in the levels of certain hormones that were part and parcel of the natural aging process.

Then along came 3 rather worrisome studies in fairly rapid-fire succession, which seamed collectively to deliver a true body blow to TRT. However, a closer examination of these studies reveals that each is so severely flawed that no meaningful conclusions can be derived from any of them.

The Testosterone in Older Men with Sarcopenia (TOM) trial was a randomized trial of TRT vs placebo in older men (mean age 74 years) with mobility limitations (sarcopenia, after all, means decreased muscle bulk) and a high prevalence of chronic disease.¹ The trial was stopped early because of a much higher occurrence of self-reported cardiovascular-related adverse events. However, these adverse events were extremely disparate and were all self-reported; none had been prespecified outcomes. Any objective observer would have to conclude that the study was poorly designed and that no meaningful conclusions can be drawn from its premature termination.

The second trial that seemed to cast doubt on the safety of TRT suffered from an even worse design. It was a retrospective cohort study of 8,709 veterans aged 60 to 64 years with low testosterone levels who were undergoing coronary artery angiography. The authors reported in the Journal of the American Medical Association that those receiving testosterone therapy had a higher risk of experiencing a composite outcome of all-cause mortality, myocardial infarction (MI), or cerebrovascular accident than did those who had not received testosterone therapy (hazard ratio [HR] = 1.29; 95% confidence interval [CI]: 1.04-1.58).² Right off the bat, you should be very wary of any HR emanating from a retrospective study that shows a small increase in risk of 29%; it’s only when a HR is 2.0 or more that it’s likely you’re looking at a real phenomenon. But to add insult to injury, the percentage of actual adverse outcomes was actually SMALLER in those taking testosterone than in those who did not get any! The authors had used such an incredibly tortured series of risk adjustments for a variety of comorbidities that they actually managed to stand the raw numbers on their head.

The third study, which had seamed at first blush to demonstrate cardiovascular toxicity of TRT, was a much larger retrospective cohort study of 55,793 men who had received replacement testosterone.³ The authors reported an increase in the relative risk of MI in the first 3 months after starting testosterone compared with the risk of MI in the same men in the prior year (relative risk [RR] = 1.36). However, the much more important absolute risk increase was very, very low, with only an additional 1.25 cases of MI seen over 1,000 patient-years. Apart from the fact that a RR of 1.36 is most unimpressive in a retrospective study, the simple fact that the men were older by a few months after TRT is probably more than adequate to explain this tiny increase in apparent risk.

The FDA has monitored these studies closely and has chosen not to make a determination that there is an increased risk of cardiovascular events associated with TRT. That is not at all the same as saying that it has been proven to be completely free of cardiovascular risk; rather it is a common-sense acknowledgment that there is not any convincing evidence to date of such a risk.

Thus, the conscientious clinician is left to conclude that TRT is a reasonable option in symptomatic patients who have been shown to have low levels of free testosterone. It has not been conclusively demonstrated that TRT will have significant beneficial effects, but neither has it been proven to have any true cardiovascular toxicity. It is a therapy worth trying in those symptomatic patients who understand that they will be receiving therapy of uncertain benefit, if any, and with the possibility of uncertain risk, if any.

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

At first it seems to be a fairly straightforward proposition. The older gentleman you are seeing in the clinic reports that he has been running rather low on energy in recent weeks, and he also mentions that there’s not much lead in his pencil these days. As a conscientious clinician, you immediately entertain the possibility that hypogonadism might explain some of his symptoms. You dutifully order up a total testosterone level, and then a free testosterone level when the total comes back low, recognizing that binding protein abnormalities might produce a low total even when the clinically relevant free level is still normal. Both levels do come back well below the age-adjusted lower limits of normal, which gives you some transient level of satisfaction that you have identified a very significant factor contributing to your patient’s difficulties.

You have confirmed a deficiency of a major hormone, and it seems logical that you would want to restore the hormone level to normal in this particular patient. But before you reach for your prescription pad (or your mouse), a fundamental question hangs uneasily in the air. Are you going to be doing more harm than good by prescribing testosterone replacement therapy (TRT) to this rather trusting older fellow? In light of recent studies, might you actually increase this patent’s chances of a heart attack or a stroke? That would not be a nice thing to do to this pleasant older gentleman. (As a newly minted senior citizen, I pray mightily that my own caregivers adhere rigorously to Hippocrates’ hoary admonition to, above all, do no harm.)

I’m not going to be able to resolve this clinical conundrum definitively in this editorial. (Please don’t stop reading just yet!) But maybe a review of the pros and cons for testosterone replacement therapy can help you faithful readers gain just a little bit better sense of the operative risk/benefit considerations at play here.

Let’s look first at the case for prescribing TRT when the laboratory test values show definitive evidence of low testosterone levels. I don’t want to delve into the distracting issue of which form of testosterone replacement to consider, which pits injections vs gels vs patches vs pills (don’t use the potentially hepatotoxic methyltestosterone pills passed out like candy by some urologists). Apart from the possibly relevant issue of peaks and troughs seen with injection therapy, the same risk/benefit considerations pretty much apply to all forms of TRT.

The benefits of TRT clearly include an increase in lean muscle mass, an increase in red blood cell concentration due to the hematopoetic effects of the male hormone, and a reduction in both the total amount and the percentage of body fat. A number of studies have shown that testosterone enhances insulin sensitivity—surely a good thing given the massive number of older patients with either prediabetes or full-blown type 2 diabetes. Some men also report a significant increase in their hard-to-define-but-still important sense of manliness, and sometimes a major improvement in their ability to perform in the sack. The latter effects, though, are often very modest and of considerably less potency (sorry, pun intended) than seen with sildenafil or one of the other PDE-5 inhibitors. In spite of all these seemingly positive effects, the clear majority of men report that they really don’t feel much different after starting on TRT, and many discontinue it on their own after relatively short periods, especially those enduring intramuscular injections every 2 weeks.

So the benefits derived from TRT are not really very impressive in many patients. What about the downside of giving testosterone? Surely there can’t be any problems associated with simply replacing an important hormone that has fallen to low levels? After all, we don’t hesitate to give thyroid hormone to hypothyroid patients, to give growth hormone to children with low levels of this critical hormone, or to give insulin to diabetic patients whose pancreases are not producing enough of that life-saving hormone.

For a very long time the risk/benefit arguments over whether or not to give TRT were almost entirely theoretical. Those in favor cited the several aforementioned benefits, and those in opposition decried replacement therapy as a perverse form of tinkering with nature by trying to alter the natural decline in the levels of certain hormones that were part and parcel of the natural aging process.

Then along came 3 rather worrisome studies in fairly rapid-fire succession, which seamed collectively to deliver a true body blow to TRT. However, a closer examination of these studies reveals that each is so severely flawed that no meaningful conclusions can be derived from any of them.

The Testosterone in Older Men with Sarcopenia (TOM) trial was a randomized trial of TRT vs placebo in older men (mean age 74 years) with mobility limitations (sarcopenia, after all, means decreased muscle bulk) and a high prevalence of chronic disease.¹ The trial was stopped early because of a much higher occurrence of self-reported cardiovascular-related adverse events. However, these adverse events were extremely disparate and were all self-reported; none had been prespecified outcomes. Any objective observer would have to conclude that the study was poorly designed and that no meaningful conclusions can be drawn from its premature termination.

The second trial that seemed to cast doubt on the safety of TRT suffered from an even worse design. It was a retrospective cohort study of 8,709 veterans aged 60 to 64 years with low testosterone levels who were undergoing coronary artery angiography. The authors reported in the Journal of the American Medical Association that those receiving testosterone therapy had a higher risk of experiencing a composite outcome of all-cause mortality, myocardial infarction (MI), or cerebrovascular accident than did those who had not received testosterone therapy (hazard ratio [HR] = 1.29; 95% confidence interval [CI]: 1.04-1.58).² Right off the bat, you should be very wary of any HR emanating from a retrospective study that shows a small increase in risk of 29%; it’s only when a HR is 2.0 or more that it’s likely you’re looking at a real phenomenon. But to add insult to injury, the percentage of actual adverse outcomes was actually SMALLER in those taking testosterone than in those who did not get any! The authors had used such an incredibly tortured series of risk adjustments for a variety of comorbidities that they actually managed to stand the raw numbers on their head.

The third study, which had seamed at first blush to demonstrate cardiovascular toxicity of TRT, was a much larger retrospective cohort study of 55,793 men who had received replacement testosterone.³ The authors reported an increase in the relative risk of MI in the first 3 months after starting testosterone compared with the risk of MI in the same men in the prior year (relative risk [RR] = 1.36). However, the much more important absolute risk increase was very, very low, with only an additional 1.25 cases of MI seen over 1,000 patient-years. Apart from the fact that a RR of 1.36 is most unimpressive in a retrospective study, the simple fact that the men were older by a few months after TRT is probably more than adequate to explain this tiny increase in apparent risk.

The FDA has monitored these studies closely and has chosen not to make a determination that there is an increased risk of cardiovascular events associated with TRT. That is not at all the same as saying that it has been proven to be completely free of cardiovascular risk; rather it is a common-sense acknowledgment that there is not any convincing evidence to date of such a risk.

Thus, the conscientious clinician is left to conclude that TRT is a reasonable option in symptomatic patients who have been shown to have low levels of free testosterone. It has not been conclusively demonstrated that TRT will have significant beneficial effects, but neither has it been proven to have any true cardiovascular toxicity. It is a therapy worth trying in those symptomatic patients who understand that they will be receiving therapy of uncertain benefit, if any, and with the possibility of uncertain risk, if any.

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Faithful readers may recall from previous editorials that I’m not particularly happy with the new hypertension guidelines issued recently by the JNC 8 authors. I am especially concerned that the new recommendations of a blood pressure goal of < 150/90 mm Hg for people aged > 60 years, like myself, could lead to a real deterioration in blood pressure control. We know that adherence to the previous goal of < 140/90 mm Hg for this age group has hardly been optimal, so why in the world would we want to relax our targets even further? I have confronted several of the JNC 8 writers with my concerns, and they have reluctantly acknowledged that I am hardly alone in my worries.

But one thing I never saw coming was that the new guidelines would confound one of the important clinical trials I’ve been participating in over the past 4 years. I’m referring to the National Institutes of Health-funded Systolic Blood Pressure Intervention Trial (SPRINT), which was designed to compare 2 systolic blood pressure goals, the traditional 140 mm Hg goal and a more aggressive 120 mm Hg goal.

One thing that is particularly confounding in the context of the new guidelines for those aged > 60 years is that we SPRINT investigators were instructed specifically to recruit as many patients as possible aged > 75 years, so that we could get a clear sense of what the systolic goal should be in this particularly high-risk population. The study architects didn’t even consider testing a goal of 150 mm Hg systolic. In a similar vein, we also worked very hard to over-recruit 2 other groups of high-risk patients, those who had already had a cardiovascular event and those with mild renal insufficiency.

The new guidelines wound up impacting my conduct of the SPRINT trial. An intellectually curious trial subject in his late 70s took a keen interest in the question: What is the optimal systolic blood pressure goal? As it turns out, he was among those who had been randomized to the more aggressive systolic goal of 120 mm Hg. At his most recent visit, he caught me off guard by asking why we were testing a blood pressure goal of 120 mm Hg in someone of his age. He had read that people aged > 60 years needed a blood pressure goal of only 150 mm Hg,according to the latest expert recommendations.

Initially I was flummoxed by his question. Perhaps I should have anticipated that some of our subjects might have questions, but I have to admit that the thought had not occurred to me. I was pleased to see that he was not at all agitated at the apparent disconnect. He was merely curious as to how there could be such a discrepancy between guidelines intended for the general public and the study goal of 120 mm Hg. This proved to be an important teachable moment. After gathering my wits, I was able to explain the difference between guidelines and hypotheses that are carefully tested in clinical trials. I was especially eager to let him know that the true science of a clinical trial trumps the value of clinical guidelines, which are based on the best clinical judgments and guesstimates of leaders in the field.

The key to understanding the role of clinical guidelines is to recognize that they simply represent the most informed opinions available, given the sum total of clinical information that is available at that time. Clinical guidelines are based on evidence as much as is humanly possible, but there are often gaps in what we have learned from published clinical trials. Such trials are inherently limited with respect to the insights they can provide, because funding limitations invariably dictate that hard choices must be made in terms of the hypotheses that can be tested and the populations that can be studied. So the total amount of available data from clinical trials is almost invariably insufficient to answer a significant number of clinical questions definitively.

And that’s why a well-designed clinical trial trumps whatever expert guidelines may seem pertinent to the clinical question at stake. Yes, the JNC 8 authors may have determined (albeit with a significant contrarian minority report) that their best reading of the available literature was that there was no definitive evidence supporting a blood pressure goal of < 150/90 mm Hg in those aged > 60 years. But it must be recognized that the absence of such definitive evidence to date does not at all mean that a lower goal might one day be shown to be superior to the JNC 8 recommendations. And that’s where the SPRINT trial comes in: It’s specifically designed to test the hypothesis that a lower systolic goal of 120 mm Hg might be superior in terms of clinical outcomes to the higher goal of 140 mm Hg. Well-designed clinical trials are the mechanism through which meaningful clinical data are accrued; those data can then inform clinical guidelines.

I am happy to report that my alert SPRINT subject grasped the point rather quickly. As a retired engineer, he understood the importance of obtaining definitive data rather than relying forevermore upon the best guesses of well-meaning experts in the field. Clinical guidelines are useful as far as they go, but they are heavily dependent upon the generation of clinically valid data from randomized clinical trials. My SPRINT subject left the clinic with a renewed commitment to getting his systolic blood pressure down to the assigned goal of 120 mm Hg. All of us should follow his example and try mightily to keep in mind the distinction between clinical guidelines and actual data generated from randomized clinical trials. 

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Faithful readers may recall from previous editorials that I’m not particularly happy with the new hypertension guidelines issued recently by the JNC 8 authors. I am especially concerned that the new recommendations of a blood pressure goal of < 150/90 mm Hg for people aged > 60 years, like myself, could lead to a real deterioration in blood pressure control. We know that adherence to the previous goal of < 140/90 mm Hg for this age group has hardly been optimal, so why in the world would we want to relax our targets even further? I have confronted several of the JNC 8 writers with my concerns, and they have reluctantly acknowledged that I am hardly alone in my worries.

But one thing I never saw coming was that the new guidelines would confound one of the important clinical trials I’ve been participating in over the past 4 years. I’m referring to the National Institutes of Health-funded Systolic Blood Pressure Intervention Trial (SPRINT), which was designed to compare 2 systolic blood pressure goals, the traditional 140 mm Hg goal and a more aggressive 120 mm Hg goal.

One thing that is particularly confounding in the context of the new guidelines for those aged > 60 years is that we SPRINT investigators were instructed specifically to recruit as many patients as possible aged > 75 years, so that we could get a clear sense of what the systolic goal should be in this particularly high-risk population. The study architects didn’t even consider testing a goal of 150 mm Hg systolic. In a similar vein, we also worked very hard to over-recruit 2 other groups of high-risk patients, those who had already had a cardiovascular event and those with mild renal insufficiency.

The new guidelines wound up impacting my conduct of the SPRINT trial. An intellectually curious trial subject in his late 70s took a keen interest in the question: What is the optimal systolic blood pressure goal? As it turns out, he was among those who had been randomized to the more aggressive systolic goal of 120 mm Hg. At his most recent visit, he caught me off guard by asking why we were testing a blood pressure goal of 120 mm Hg in someone of his age. He had read that people aged > 60 years needed a blood pressure goal of only 150 mm Hg,according to the latest expert recommendations.

Initially I was flummoxed by his question. Perhaps I should have anticipated that some of our subjects might have questions, but I have to admit that the thought had not occurred to me. I was pleased to see that he was not at all agitated at the apparent disconnect. He was merely curious as to how there could be such a discrepancy between guidelines intended for the general public and the study goal of 120 mm Hg. This proved to be an important teachable moment. After gathering my wits, I was able to explain the difference between guidelines and hypotheses that are carefully tested in clinical trials. I was especially eager to let him know that the true science of a clinical trial trumps the value of clinical guidelines, which are based on the best clinical judgments and guesstimates of leaders in the field.

The key to understanding the role of clinical guidelines is to recognize that they simply represent the most informed opinions available, given the sum total of clinical information that is available at that time. Clinical guidelines are based on evidence as much as is humanly possible, but there are often gaps in what we have learned from published clinical trials. Such trials are inherently limited with respect to the insights they can provide, because funding limitations invariably dictate that hard choices must be made in terms of the hypotheses that can be tested and the populations that can be studied. So the total amount of available data from clinical trials is almost invariably insufficient to answer a significant number of clinical questions definitively.

And that’s why a well-designed clinical trial trumps whatever expert guidelines may seem pertinent to the clinical question at stake. Yes, the JNC 8 authors may have determined (albeit with a significant contrarian minority report) that their best reading of the available literature was that there was no definitive evidence supporting a blood pressure goal of < 150/90 mm Hg in those aged > 60 years. But it must be recognized that the absence of such definitive evidence to date does not at all mean that a lower goal might one day be shown to be superior to the JNC 8 recommendations. And that’s where the SPRINT trial comes in: It’s specifically designed to test the hypothesis that a lower systolic goal of 120 mm Hg might be superior in terms of clinical outcomes to the higher goal of 140 mm Hg. Well-designed clinical trials are the mechanism through which meaningful clinical data are accrued; those data can then inform clinical guidelines.

I am happy to report that my alert SPRINT subject grasped the point rather quickly. As a retired engineer, he understood the importance of obtaining definitive data rather than relying forevermore upon the best guesses of well-meaning experts in the field. Clinical guidelines are useful as far as they go, but they are heavily dependent upon the generation of clinically valid data from randomized clinical trials. My SPRINT subject left the clinic with a renewed commitment to getting his systolic blood pressure down to the assigned goal of 120 mm Hg. All of us should follow his example and try mightily to keep in mind the distinction between clinical guidelines and actual data generated from randomized clinical trials. 

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Faithful readers may recall from previous editorials that I’m not particularly happy with the new hypertension guidelines issued recently by the JNC 8 authors. I am especially concerned that the new recommendations of a blood pressure goal of < 150/90 mm Hg for people aged > 60 years, like myself, could lead to a real deterioration in blood pressure control. We know that adherence to the previous goal of < 140/90 mm Hg for this age group has hardly been optimal, so why in the world would we want to relax our targets even further? I have confronted several of the JNC 8 writers with my concerns, and they have reluctantly acknowledged that I am hardly alone in my worries.

But one thing I never saw coming was that the new guidelines would confound one of the important clinical trials I’ve been participating in over the past 4 years. I’m referring to the National Institutes of Health-funded Systolic Blood Pressure Intervention Trial (SPRINT), which was designed to compare 2 systolic blood pressure goals, the traditional 140 mm Hg goal and a more aggressive 120 mm Hg goal.

One thing that is particularly confounding in the context of the new guidelines for those aged > 60 years is that we SPRINT investigators were instructed specifically to recruit as many patients as possible aged > 75 years, so that we could get a clear sense of what the systolic goal should be in this particularly high-risk population. The study architects didn’t even consider testing a goal of 150 mm Hg systolic. In a similar vein, we also worked very hard to over-recruit 2 other groups of high-risk patients, those who had already had a cardiovascular event and those with mild renal insufficiency.

The new guidelines wound up impacting my conduct of the SPRINT trial. An intellectually curious trial subject in his late 70s took a keen interest in the question: What is the optimal systolic blood pressure goal? As it turns out, he was among those who had been randomized to the more aggressive systolic goal of 120 mm Hg. At his most recent visit, he caught me off guard by asking why we were testing a blood pressure goal of 120 mm Hg in someone of his age. He had read that people aged > 60 years needed a blood pressure goal of only 150 mm Hg,according to the latest expert recommendations.

Initially I was flummoxed by his question. Perhaps I should have anticipated that some of our subjects might have questions, but I have to admit that the thought had not occurred to me. I was pleased to see that he was not at all agitated at the apparent disconnect. He was merely curious as to how there could be such a discrepancy between guidelines intended for the general public and the study goal of 120 mm Hg. This proved to be an important teachable moment. After gathering my wits, I was able to explain the difference between guidelines and hypotheses that are carefully tested in clinical trials. I was especially eager to let him know that the true science of a clinical trial trumps the value of clinical guidelines, which are based on the best clinical judgments and guesstimates of leaders in the field.

The key to understanding the role of clinical guidelines is to recognize that they simply represent the most informed opinions available, given the sum total of clinical information that is available at that time. Clinical guidelines are based on evidence as much as is humanly possible, but there are often gaps in what we have learned from published clinical trials. Such trials are inherently limited with respect to the insights they can provide, because funding limitations invariably dictate that hard choices must be made in terms of the hypotheses that can be tested and the populations that can be studied. So the total amount of available data from clinical trials is almost invariably insufficient to answer a significant number of clinical questions definitively.

And that’s why a well-designed clinical trial trumps whatever expert guidelines may seem pertinent to the clinical question at stake. Yes, the JNC 8 authors may have determined (albeit with a significant contrarian minority report) that their best reading of the available literature was that there was no definitive evidence supporting a blood pressure goal of < 150/90 mm Hg in those aged > 60 years. But it must be recognized that the absence of such definitive evidence to date does not at all mean that a lower goal might one day be shown to be superior to the JNC 8 recommendations. And that’s where the SPRINT trial comes in: It’s specifically designed to test the hypothesis that a lower systolic goal of 120 mm Hg might be superior in terms of clinical outcomes to the higher goal of 140 mm Hg. Well-designed clinical trials are the mechanism through which meaningful clinical data are accrued; those data can then inform clinical guidelines.

I am happy to report that my alert SPRINT subject grasped the point rather quickly. As a retired engineer, he understood the importance of obtaining definitive data rather than relying forevermore upon the best guesses of well-meaning experts in the field. Clinical guidelines are useful as far as they go, but they are heavily dependent upon the generation of clinically valid data from randomized clinical trials. My SPRINT subject left the clinic with a renewed commitment to getting his systolic blood pressure down to the assigned goal of 120 mm Hg. All of us should follow his example and try mightily to keep in mind the distinction between clinical guidelines and actual data generated from randomized clinical trials. 

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review the complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Most of you are no doubt familiar with the opening of Charles Dickens’ classic novel A Tale of Two Cities: “It was the best of times, it was the worst of times.” In many ways Dickens has presaged my current thinking about the new lipid guidelines that were recently issued. On the one hand, they are truly the very best guidelines that could possibly be produced at the present time; on the other hand, they may also be the worst set of guidelines that could possibly be promulgated on the practice community at the present moment.

Let’s first take a good look at what the new guidelines actually recommend. First, it’s important to understand that the parentage of the new guidelines has changed in a very important way from that of earlier recommendations. The previous lipid guidelines—the National Cholesterol Education Panel (NCEP) recommendations—were issued in 2001 and were sponsored and endorsed by the National Institutes of Health (NIH) as the federal government’s best effort at lipid recommendations for the general practice community. An update in 2004 further advised that a low-density lipoprotein cholesterol (LDL-C) goal of < 70 mg/dL is appropriate for many patients with preexisting vascular disease.

New guidelines were clearly overdue. Indeed, an expert panel had already been convened and was hard at work. However, a year or so ago the NIH made a critical strategic decision that it no longer wanted to be in the guideline business. The NIH rather abruptly decided that there would be no further iterations of the NCEP guidelines.

Fortunately the NIH did not simply drop the ball. Rather, it decided to pass the baton (mixed metaphor—sorry!) to a joint task force of the American Heart Association (AHA) and the American College of Cardiology (ACC). After all, who better to ponder lipid goals than the vascular experts who populate these 2 august societies? These 2 groups took a good look at the work that had been done by the expert panel, and decided that they would bless the new recommendations.

The fruit of the years of labor were finally presented at the annual meeting of the AHA held in Dallas in 2013. I didn’t make it to all of the heart sessions during that meeting. (I was distracted to a considerable extent by the 50th anniversary commemoration of President John F. Kennedy’s assassination going on just a few blocks away from the convention center at the Texas School Book Depository site.) But I can tell you I was definitely at the AHA session where the guidelines were formally presented, and it was indeed a lively and controversial session.

By far the biggest change in the new guidelines, representing both its greatest strength and its greatest weakness, is the new emphasis on overall cardiovascular risk assessment rather than on the attainment of a certain defined LDL-C goal. Indeed, a feature of the new guidelines, which many find disconcerting, is that there is no longer any mention whatsoever of LDL-C goals or targets!

The guidelines are also heavily statin-centric; other classes of lipid-lowering agents, such as fibrates or niacin, receive short shrift indeed. The recommendations are that statins should be prescribed routinely for each of the following “statin benefit groups”:

1. Patients who have clinical atherosclerotic cardiovascular disease and thus fall into the “secondary prevention” category.

2. Those with LDL-C levels of  ≥ 190 mg/dL and who have no secondary cause, such as certain medications or diseases such as hypothyroidism or nephrotic syndrome.

3. Patients with diabetes without established cardiovascular disease aged 40 to 75 years with LDL-C levels between 70 mg/dL and 189 mg/dL.

4. Patients without diabetes with established atherosclerotic cardiovascular disease aged 40 to 75 years with LDL-C levels from 70 mg/dL to 189 mg/dL and a calculated cardiovascular risk of at least 7.5% over the next 10 years.

The fourth category is potentially the most confusing for conscientious providers. The risk calculator that determines whether or not someone has a risk of > 7.5% over the next year is not the traditional Framingham risk calculator, with which many providers are familiar. Rather, it is a brand-new, improved risk calculator devised by the panel. The calculator can be found on both the AHA and ACC websites and in iOS and Android apps (See App Corner, p.38).

To make things even more confusing, once it has been determined that a statin is indicated, the dosing of the statin, either low, moderate, or high intensity, must be selected on the basis of determined risk level. Fortunately, the panel has given us a nice table defining which statins qualify for inclusion in each of these 3 intensity categories. As a general rule, the low-intensity statins should almost always be avoided. But the determination of whether moderate or high-intensity statins are indicated gets somewhat murky. The first two aforementioned classes both deserve high-intensity statins. However, patients with diabetes who haven’t had an event could go with either moderate- or high-intensity statins, and those patients without diabetes or LDL-C levels ≥ 190, but with a 10-year risk of at least 7.5%, can also receive either moderate- or high-intensity statins.

So there it is, and it all does make a certain amount of sense. You first determine the patient’s risk category, which determines whether or not statins are indicated. If they are, you then decide what level of potency your prescribed statin should possess. There is no need to go checking LDL-C levels later, because your therapy is not targeted at any particular LDL-C level. You might want to check occasionally, though, just as a way of assessing patient compliance.

So what should we make of all of this? From a purely scientific point of view, it seems abundantly clear that these are the most scientifically valid set of guidelines that have ever been produced, generated by genuine experts who bent over backward to examine every possible relevant study. The new risk calculator is clearly a broader-based tool than the Framingham calculator, which was based on now-dated data from a very narrow heavily-white population basis. Although the new risk calculator has been criticized by some as a very imperfect tool that overestimates risk in some subpopulations, I firmly believe it is considerably less imperfect than the Framingham tool. It is, quite frankly, the best risk calculator anyone could come up with at this time, and the cutoff for treatment at a risk of 7.5% or higher over 10 years seems eminently reasonable to me.

So what’s the problem with the new guidelines? I think you astute readers already know what the problem is: These guidelines simply represent way too radical a change for the huge bulk of busy, harried providers out there. The average primary care provider is currently struggling to complete a multifaceted patient encounter in 15 minutes or less and then document it in excruciating detail. He or she is going to be extremely hard-pressed to master and implement the new guidelines. The guidelines are indeed the most scientifically accurate and thorough guidelines that could be humanly produced, but they represent such a radical change from previous guidelines that a huge number of providers are going to be playing catch-up for a long time. I hope that their learning curve can be a very rapid one, but I worry that these scientifically pristine guidelines will be slow to find their way into general practice. 

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Most of you are no doubt familiar with the opening of Charles Dickens’ classic novel A Tale of Two Cities: “It was the best of times, it was the worst of times.” In many ways Dickens has presaged my current thinking about the new lipid guidelines that were recently issued. On the one hand, they are truly the very best guidelines that could possibly be produced at the present time; on the other hand, they may also be the worst set of guidelines that could possibly be promulgated on the practice community at the present moment.

Let’s first take a good look at what the new guidelines actually recommend. First, it’s important to understand that the parentage of the new guidelines has changed in a very important way from that of earlier recommendations. The previous lipid guidelines—the National Cholesterol Education Panel (NCEP) recommendations—were issued in 2001 and were sponsored and endorsed by the National Institutes of Health (NIH) as the federal government’s best effort at lipid recommendations for the general practice community. An update in 2004 further advised that a low-density lipoprotein cholesterol (LDL-C) goal of < 70 mg/dL is appropriate for many patients with preexisting vascular disease.

New guidelines were clearly overdue. Indeed, an expert panel had already been convened and was hard at work. However, a year or so ago the NIH made a critical strategic decision that it no longer wanted to be in the guideline business. The NIH rather abruptly decided that there would be no further iterations of the NCEP guidelines.

Fortunately the NIH did not simply drop the ball. Rather, it decided to pass the baton (mixed metaphor—sorry!) to a joint task force of the American Heart Association (AHA) and the American College of Cardiology (ACC). After all, who better to ponder lipid goals than the vascular experts who populate these 2 august societies? These 2 groups took a good look at the work that had been done by the expert panel, and decided that they would bless the new recommendations.

The fruit of the years of labor were finally presented at the annual meeting of the AHA held in Dallas in 2013. I didn’t make it to all of the heart sessions during that meeting. (I was distracted to a considerable extent by the 50th anniversary commemoration of President John F. Kennedy’s assassination going on just a few blocks away from the convention center at the Texas School Book Depository site.) But I can tell you I was definitely at the AHA session where the guidelines were formally presented, and it was indeed a lively and controversial session.

By far the biggest change in the new guidelines, representing both its greatest strength and its greatest weakness, is the new emphasis on overall cardiovascular risk assessment rather than on the attainment of a certain defined LDL-C goal. Indeed, a feature of the new guidelines, which many find disconcerting, is that there is no longer any mention whatsoever of LDL-C goals or targets!

The guidelines are also heavily statin-centric; other classes of lipid-lowering agents, such as fibrates or niacin, receive short shrift indeed. The recommendations are that statins should be prescribed routinely for each of the following “statin benefit groups”:

1. Patients who have clinical atherosclerotic cardiovascular disease and thus fall into the “secondary prevention” category.

2. Those with LDL-C levels of  ≥ 190 mg/dL and who have no secondary cause, such as certain medications or diseases such as hypothyroidism or nephrotic syndrome.

3. Patients with diabetes without established cardiovascular disease aged 40 to 75 years with LDL-C levels between 70 mg/dL and 189 mg/dL.

4. Patients without diabetes with established atherosclerotic cardiovascular disease aged 40 to 75 years with LDL-C levels from 70 mg/dL to 189 mg/dL and a calculated cardiovascular risk of at least 7.5% over the next 10 years.

The fourth category is potentially the most confusing for conscientious providers. The risk calculator that determines whether or not someone has a risk of > 7.5% over the next year is not the traditional Framingham risk calculator, with which many providers are familiar. Rather, it is a brand-new, improved risk calculator devised by the panel. The calculator can be found on both the AHA and ACC websites and in iOS and Android apps (See App Corner, p.38).

To make things even more confusing, once it has been determined that a statin is indicated, the dosing of the statin, either low, moderate, or high intensity, must be selected on the basis of determined risk level. Fortunately, the panel has given us a nice table defining which statins qualify for inclusion in each of these 3 intensity categories. As a general rule, the low-intensity statins should almost always be avoided. But the determination of whether moderate or high-intensity statins are indicated gets somewhat murky. The first two aforementioned classes both deserve high-intensity statins. However, patients with diabetes who haven’t had an event could go with either moderate- or high-intensity statins, and those patients without diabetes or LDL-C levels ≥ 190, but with a 10-year risk of at least 7.5%, can also receive either moderate- or high-intensity statins.

So there it is, and it all does make a certain amount of sense. You first determine the patient’s risk category, which determines whether or not statins are indicated. If they are, you then decide what level of potency your prescribed statin should possess. There is no need to go checking LDL-C levels later, because your therapy is not targeted at any particular LDL-C level. You might want to check occasionally, though, just as a way of assessing patient compliance.

So what should we make of all of this? From a purely scientific point of view, it seems abundantly clear that these are the most scientifically valid set of guidelines that have ever been produced, generated by genuine experts who bent over backward to examine every possible relevant study. The new risk calculator is clearly a broader-based tool than the Framingham calculator, which was based on now-dated data from a very narrow heavily-white population basis. Although the new risk calculator has been criticized by some as a very imperfect tool that overestimates risk in some subpopulations, I firmly believe it is considerably less imperfect than the Framingham tool. It is, quite frankly, the best risk calculator anyone could come up with at this time, and the cutoff for treatment at a risk of 7.5% or higher over 10 years seems eminently reasonable to me.

So what’s the problem with the new guidelines? I think you astute readers already know what the problem is: These guidelines simply represent way too radical a change for the huge bulk of busy, harried providers out there. The average primary care provider is currently struggling to complete a multifaceted patient encounter in 15 minutes or less and then document it in excruciating detail. He or she is going to be extremely hard-pressed to master and implement the new guidelines. The guidelines are indeed the most scientifically accurate and thorough guidelines that could be humanly produced, but they represent such a radical change from previous guidelines that a huge number of providers are going to be playing catch-up for a long time. I hope that their learning curve can be a very rapid one, but I worry that these scientifically pristine guidelines will be slow to find their way into general practice. 

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.

Most of you are no doubt familiar with the opening of Charles Dickens’ classic novel A Tale of Two Cities: “It was the best of times, it was the worst of times.” In many ways Dickens has presaged my current thinking about the new lipid guidelines that were recently issued. On the one hand, they are truly the very best guidelines that could possibly be produced at the present time; on the other hand, they may also be the worst set of guidelines that could possibly be promulgated on the practice community at the present moment.

Let’s first take a good look at what the new guidelines actually recommend. First, it’s important to understand that the parentage of the new guidelines has changed in a very important way from that of earlier recommendations. The previous lipid guidelines—the National Cholesterol Education Panel (NCEP) recommendations—were issued in 2001 and were sponsored and endorsed by the National Institutes of Health (NIH) as the federal government’s best effort at lipid recommendations for the general practice community. An update in 2004 further advised that a low-density lipoprotein cholesterol (LDL-C) goal of < 70 mg/dL is appropriate for many patients with preexisting vascular disease.

New guidelines were clearly overdue. Indeed, an expert panel had already been convened and was hard at work. However, a year or so ago the NIH made a critical strategic decision that it no longer wanted to be in the guideline business. The NIH rather abruptly decided that there would be no further iterations of the NCEP guidelines.

Fortunately the NIH did not simply drop the ball. Rather, it decided to pass the baton (mixed metaphor—sorry!) to a joint task force of the American Heart Association (AHA) and the American College of Cardiology (ACC). After all, who better to ponder lipid goals than the vascular experts who populate these 2 august societies? These 2 groups took a good look at the work that had been done by the expert panel, and decided that they would bless the new recommendations.

The fruit of the years of labor were finally presented at the annual meeting of the AHA held in Dallas in 2013. I didn’t make it to all of the heart sessions during that meeting. (I was distracted to a considerable extent by the 50th anniversary commemoration of President John F. Kennedy’s assassination going on just a few blocks away from the convention center at the Texas School Book Depository site.) But I can tell you I was definitely at the AHA session where the guidelines were formally presented, and it was indeed a lively and controversial session.

By far the biggest change in the new guidelines, representing both its greatest strength and its greatest weakness, is the new emphasis on overall cardiovascular risk assessment rather than on the attainment of a certain defined LDL-C goal. Indeed, a feature of the new guidelines, which many find disconcerting, is that there is no longer any mention whatsoever of LDL-C goals or targets!

The guidelines are also heavily statin-centric; other classes of lipid-lowering agents, such as fibrates or niacin, receive short shrift indeed. The recommendations are that statins should be prescribed routinely for each of the following “statin benefit groups”:

1. Patients who have clinical atherosclerotic cardiovascular disease and thus fall into the “secondary prevention” category.

2. Those with LDL-C levels of  ≥ 190 mg/dL and who have no secondary cause, such as certain medications or diseases such as hypothyroidism or nephrotic syndrome.

3. Patients with diabetes without established cardiovascular disease aged 40 to 75 years with LDL-C levels between 70 mg/dL and 189 mg/dL.

4. Patients without diabetes with established atherosclerotic cardiovascular disease aged 40 to 75 years with LDL-C levels from 70 mg/dL to 189 mg/dL and a calculated cardiovascular risk of at least 7.5% over the next 10 years.

The fourth category is potentially the most confusing for conscientious providers. The risk calculator that determines whether or not someone has a risk of > 7.5% over the next year is not the traditional Framingham risk calculator, with which many providers are familiar. Rather, it is a brand-new, improved risk calculator devised by the panel. The calculator can be found on both the AHA and ACC websites and in iOS and Android apps (See App Corner, p.38).

To make things even more confusing, once it has been determined that a statin is indicated, the dosing of the statin, either low, moderate, or high intensity, must be selected on the basis of determined risk level. Fortunately, the panel has given us a nice table defining which statins qualify for inclusion in each of these 3 intensity categories. As a general rule, the low-intensity statins should almost always be avoided. But the determination of whether moderate or high-intensity statins are indicated gets somewhat murky. The first two aforementioned classes both deserve high-intensity statins. However, patients with diabetes who haven’t had an event could go with either moderate- or high-intensity statins, and those patients without diabetes or LDL-C levels ≥ 190, but with a 10-year risk of at least 7.5%, can also receive either moderate- or high-intensity statins.

So there it is, and it all does make a certain amount of sense. You first determine the patient’s risk category, which determines whether or not statins are indicated. If they are, you then decide what level of potency your prescribed statin should possess. There is no need to go checking LDL-C levels later, because your therapy is not targeted at any particular LDL-C level. You might want to check occasionally, though, just as a way of assessing patient compliance.

So what should we make of all of this? From a purely scientific point of view, it seems abundantly clear that these are the most scientifically valid set of guidelines that have ever been produced, generated by genuine experts who bent over backward to examine every possible relevant study. The new risk calculator is clearly a broader-based tool than the Framingham calculator, which was based on now-dated data from a very narrow heavily-white population basis. Although the new risk calculator has been criticized by some as a very imperfect tool that overestimates risk in some subpopulations, I firmly believe it is considerably less imperfect than the Framingham tool. It is, quite frankly, the best risk calculator anyone could come up with at this time, and the cutoff for treatment at a risk of 7.5% or higher over 10 years seems eminently reasonable to me.

So what’s the problem with the new guidelines? I think you astute readers already know what the problem is: These guidelines simply represent way too radical a change for the huge bulk of busy, harried providers out there. The average primary care provider is currently struggling to complete a multifaceted patient encounter in 15 minutes or less and then document it in excruciating detail. He or she is going to be extremely hard-pressed to master and implement the new guidelines. The guidelines are indeed the most scientifically accurate and thorough guidelines that could be humanly produced, but they represent such a radical change from previous guidelines that a huge number of providers are going to be playing catch-up for a long time. I hope that their learning curve can be a very rapid one, but I worry that these scientifically pristine guidelines will be slow to find their way into general practice. 

Author disclosures
The author reports no actual or potential conflicts of interest with regard to this article.

Disclaimer
The opinions expressed herein are those of the author and do not necessarily reflect those of Federal Practitioner, Frontline Medical Communications Inc., the U.S. Government, or any of its agencies. This article may discuss unlabeled or investigational use of certain drugs. Please review complete prescribing information for specific drugs or drug combinations—including indications, contraindications, warnings, and adverse effects—before administering pharmacologic therapy to patients.