Commentary

With me is Dr. Rudolf de Boer, from University Medical Center, Groningen, The Netherlands, who has done some of the basic science work; Dr. Salvatore Di Somma, who is an emergency physician at San Andrea Hospital at the University Sapienza in Rome; and Dr. Alan Maisel, a world-renowned biomarker researcher from the University of California San Diego at the Veterans Administration. I am Frank Peacock, a Professor in Emergency Medicine and the Director of Research at the Baylor College of Medicine.

Welcome, gentlemen. I’d like to open this up with Dr. Boer, starting with some discussion on what he knows about this marker.

Dr. de Boer: Yes, thank you, Dr. Peacock, for your introduction. I will briefly introduce galectin-3 to you, although I’m sure most of you have become familiar with it to some extent. It was actually not so long ago that galectin-3 was scientifically discovered in the 1980s. When it was discovered, its function was not fully recognized. It was only 8 years ago that galectin-3 was associated with heart failure development and cardiac remodeling.¹

Galectin-3 is a member of the galectin family, which is a large family comprising about 20 galectins. Galectin-3 is unique because it’s a so-called chimera-type galectin, meaning that upon lectin binding, it forms multidimers, shaping a mesh, which is thought to add to the stiffness of the interstitial matrix, and in the case of heart failure, it adds stiffness to the extracellular myocardial matrix.² Galectin-3 can be activated by all kinds of lectins, and in the normal heart, these ligands are believed to be major cellular proteins such as laminin and collagens. In damaged organs in general, and in heart failure in particular, galectin-3 is predominantly released by macrophages and the activating lectins are clearly more abundant, and this is generally thought to be the mechanism of galectin-3 activation in heart failure.

As I mentioned, a first seminal paper¹ on the potential role of galectin-3 in heart failure was published in 2004. In a rat model with transition from compensated to decompensated heart failure, galectin-3 was the most strongly differentially expressed gene in this transitional phase. With further experimentation, the investigators showed that adding galectin-3 into the pericardial sac rendered superphysiological levels of galectin-3, and with this perturbation, they were able to provoke adverse cardiac remodeling with collagen deposition and left ventricular dysfunction. So, not only does galectin-3 seem to be associated with heart failure development, but may also be a part of the pathophysiology of cardiac remodeling and heart failure. This, of course, is very interesting.

Furthermore, galectin-3 seems to be a factor that is very dominant in the matrix: it does not predominantly act on cardiomyocytes, but on fibroblasts and the matrix compartment of the heart. It is strictly colocalized with sites of fibrosis where injury occurs, and it was shown that the primary source of galectin-3 in the heart includes all kinds of inflammatory cells, mainly macrophages, at the site of injury, for example, post-myocardial infarction or in hypertensive models. However, some other cells also produce galectin-3 including cardiac fibroblasts and mast cells. Mast cells are clearly less abundant, and fibroblasts are very abundant but produce a rather low amount of galectin-3. So overall, the primary source of galectin-3 is believed to be macrophages. As it is currently understood, when galectin-3 is produced, it turns cardiac fibroblasts into matrix-producing myofibroblasts, which then adds to the remodeling, specifically extracellular matrix production and fibrogenesis, not so much left ventricular hypertrophy, and this may contribute to the ultimate culmination into overt heart failure.

This is a general scheme of how we think of galectin-3 right now. As said before, interestingly, galectin-3 is associated with the disease process, but when activated, it has also been shown to add to the pathophysiology. Besides the naturally occurring binding lectins, investigators have developed neutralizing lectins that may bind to galectin-3 and render it inactive. This approach may actually downregulate galectin-3 activity with the aim to attenuate the galectin-3—driven progress of cardiac remodeling and to halt heart failure development. Recently, these neutralizing lectins have been tested in several animal models of hypertension, and it was shown that galectin-3 can be inhibited; this is associated with the attenuation of end-organ damage.³,⁴

Besides cardiac fibrosis, it has been observed that galectin-3 is also activated in liver fibrosis,⁵ kidney fibrosis,⁶ and lung fibrosis.⁷ Experimental studies using pharmacological compounds specifically targeting galectin-3 in non-cardiac disease have also been quite successful in attenuating tissue fibrosis.⁷ I think this adds an extra level of interest to this protein—that is, it’s not only a marker of disease severity, but also some kind of culprit biomarker, suggesting that it is part of the pathophysiology of heart failure, and can be a target of specific therapy.

Dr. Peacock: Perfect, Dr. Boer, that’s a really nice review of what we know up to this point about galectin-3. I think the next question is—is it ready for the clinical world? And if so, how would we be using it? Dr. Di Somma, what do you think about this and the use of galectin-3 in the emergency department?

Dr. Di Somma: That’s a very important question, because for patients presenting to an emergency department for shortness of breath due to acute heart failure, the physician has 3 main important jobs to do. First, a quick diagnosis is needed to start immediately with treatment because from papers in the literature,⁸ it is very clear that how fast you are in making the diagnosis and starting the treatment is going to have an important impact on both the patient’s outcome and length of stay. So, the first job is to give an appropriate diagnosis, but of no less importance is to immediately perform a risk specification, which is very important in terms of determining how aggressive your treatment must be, as well as the disposition based on this certification analysis of your patients with acute heart failure.

It seems possible now to discriminate between patients arriving in the Emergency Department with signs and symptoms of acute heart failure with galectin-3 levels of <17.8 pg/mL compared to patients with a galectin-3 level greater than this cutoff. Patients with a galectin-3 level of <17.8 pg/mL have a better prognosis after 30 days on follow-up examination, which, from the emergency physician’s point of view, is of great importance in immediately understanding the severity of the patient’s condition, because the disposition for these patients should be different.

Galectin-3 levels could be influenced by age and the kidney function. So if the patient is very old (> 75 years) and his estimated glomerular filtration rate is below 30 mL/min, the galectin-3 levels could be higher than expected. This could be related to the ongoing fibrotic process in the kidney that galectin-3 is mirroring. Moreover, in patients presenting with concomitant oncologic disease, galectin levels could be also higher.

For acute heart failure treatment, we now have many options, but in majority of the cases, the therapy is based on the use of intravenous diuretics. The current guidelines recommend high doses of intravenous diuretics, but also consider the possibility that, if immediately after a patient presents to the emergency department, you are able to make a correct diagnosis and determine the treatment and the following disposition based on biomarkers, the dosage of diuretic should also be based on appropriate patient risk stratification. A galectin-3 level of >17.8 pg/mL for instance indicates that a patient is at high risk for hospitalization, so you must be more cautious with this patient, possibly with a more aggressive intravenous diuretic treatment. Furthermore, the emergency department physician could also decide on the basis of a galectin-3 level of >17.8 pg/mL whether or not to discharge these patients only after infusion or to admit this patient to the hospital. On the contrary, since the job of the emergency department physician is mostly to rule out less severe cases, a galectin-3 level of <17.8 pg/mL (indicating less heart fibrosis) would probably indicate that it is safe to discharge a patient after just a small increase in the diuretic treatment dosage.

The use of galectin-3 will also have an important impact on the problem of overcrowding in the emergency department. In the recertification of patients coming to the emergency department with acute heart failure, there is an important possibility that adding the opportunity to measure galectin-3 can positively impact treatment from the emergency department physician.

Dr. Peacock: I think you brought up some extremely important points. You have to make a diagnosis first. Galectin-3 does not diagnose heart failure, but as you said, once you’ve established the probability that the patient has heart failure, galectin-3 is an outstanding prognostic agent. As Dr. de Boer said, it reflects generalized fibrosis, and there is a lack of specificity in that situation. However, once you’ve decided that the diagnosis is heart failure, then it’s an excellent prognostic, and if you define the cutoff point as 17.8 pg/mL, it identifies patients who would be at lower or higher risk for adverse outcomes subsequent to discharge.

This is really important for emergency medicine because we really don’t have anything that tells us what the future will look like in heart failure, except for B-type natriuretic peptide (BNP), which is really poor at predicting the next 2 weeks, and troponin, which is excellent, but is only elevated in a very small minority of cases using the contemporary assays available in the United States at this time.

Dr. Maisel: You can also make a case that once a diagnosis of heart failure is made—it’s sort of a moderate or mild-to-moderate heart failure, the BNP level is just a little above the grey zone, maybe 500 or 600 pg/mL, and the patient responds well to the diuretic—this isn’t like a troponin-positive acute coronary syndrome patient where, even if it’s just slightly positive, you’re still going to admit them. But, if you could show a propensity to have fibrosis in the short term and long term in this case of heart failure, and especially, if the patient is new to the system, then you may want to admit the patient to make sure this person gets on excellent therapy—and I’ll be talking about the role of mineralocorticoid receptor antagonists in a few minutes.

I think it’s significant to say, “Maybe this guy should not go home.” His level is above 17.8 pg/mL, and he has mild heart failure. He’s prone to just go downhill with each recurrent heart failure admission, and we need to take a closer look. If, in fact, you don’t admit him, then this is a person I would bring back within probably the first week to keep a close eye on.

Dr. Peacock: Dr. Maisel, what are you going to do when Dr. Di Somma sees a patient in the emergency department, their galectin-3 level is 27 pg/mL, and he sends the patient upstairs? Now, you’re at the receiving end.

With me is Dr. Rudolf de Boer, from University Medical Center, Groningen, The Netherlands, who has done some of the basic science work; Dr. Salvatore Di Somma, who is an emergency physician at San Andrea Hospital at the University Sapienza in Rome; and Dr. Alan Maisel, a world-renowned biomarker researcher from the University of California San Diego at the Veterans Administration. I am Frank Peacock, a Professor in Emergency Medicine and the Director of Research at the Baylor College of Medicine.

Welcome, gentlemen. I’d like to open this up with Dr. Boer, starting with some discussion on what he knows about this marker.

Dr. de Boer: Yes, thank you, Dr. Peacock, for your introduction. I will briefly introduce galectin-3 to you, although I’m sure most of you have become familiar with it to some extent. It was actually not so long ago that galectin-3 was scientifically discovered in the 1980s. When it was discovered, its function was not fully recognized. It was only 8 years ago that galectin-3 was associated with heart failure development and cardiac remodeling.¹

Galectin-3 is a member of the galectin family, which is a large family comprising about 20 galectins. Galectin-3 is unique because it’s a so-called chimera-type galectin, meaning that upon lectin binding, it forms multidimers, shaping a mesh, which is thought to add to the stiffness of the interstitial matrix, and in the case of heart failure, it adds stiffness to the extracellular myocardial matrix.² Galectin-3 can be activated by all kinds of lectins, and in the normal heart, these ligands are believed to be major cellular proteins such as laminin and collagens. In damaged organs in general, and in heart failure in particular, galectin-3 is predominantly released by macrophages and the activating lectins are clearly more abundant, and this is generally thought to be the mechanism of galectin-3 activation in heart failure.

As I mentioned, a first seminal paper¹ on the potential role of galectin-3 in heart failure was published in 2004. In a rat model with transition from compensated to decompensated heart failure, galectin-3 was the most strongly differentially expressed gene in this transitional phase. With further experimentation, the investigators showed that adding galectin-3 into the pericardial sac rendered superphysiological levels of galectin-3, and with this perturbation, they were able to provoke adverse cardiac remodeling with collagen deposition and left ventricular dysfunction. So, not only does galectin-3 seem to be associated with heart failure development, but may also be a part of the pathophysiology of cardiac remodeling and heart failure. This, of course, is very interesting.

Furthermore, galectin-3 seems to be a factor that is very dominant in the matrix: it does not predominantly act on cardiomyocytes, but on fibroblasts and the matrix compartment of the heart. It is strictly colocalized with sites of fibrosis where injury occurs, and it was shown that the primary source of galectin-3 in the heart includes all kinds of inflammatory cells, mainly macrophages, at the site of injury, for example, post-myocardial infarction or in hypertensive models. However, some other cells also produce galectin-3 including cardiac fibroblasts and mast cells. Mast cells are clearly less abundant, and fibroblasts are very abundant but produce a rather low amount of galectin-3. So overall, the primary source of galectin-3 is believed to be macrophages. As it is currently understood, when galectin-3 is produced, it turns cardiac fibroblasts into matrix-producing myofibroblasts, which then adds to the remodeling, specifically extracellular matrix production and fibrogenesis, not so much left ventricular hypertrophy, and this may contribute to the ultimate culmination into overt heart failure.

This is a general scheme of how we think of galectin-3 right now. As said before, interestingly, galectin-3 is associated with the disease process, but when activated, it has also been shown to add to the pathophysiology. Besides the naturally occurring binding lectins, investigators have developed neutralizing lectins that may bind to galectin-3 and render it inactive. This approach may actually downregulate galectin-3 activity with the aim to attenuate the galectin-3—driven progress of cardiac remodeling and to halt heart failure development. Recently, these neutralizing lectins have been tested in several animal models of hypertension, and it was shown that galectin-3 can be inhibited; this is associated with the attenuation of end-organ damage.³,⁴

Besides cardiac fibrosis, it has been observed that galectin-3 is also activated in liver fibrosis,⁵ kidney fibrosis,⁶ and lung fibrosis.⁷ Experimental studies using pharmacological compounds specifically targeting galectin-3 in non-cardiac disease have also been quite successful in attenuating tissue fibrosis.⁷ I think this adds an extra level of interest to this protein—that is, it’s not only a marker of disease severity, but also some kind of culprit biomarker, suggesting that it is part of the pathophysiology of heart failure, and can be a target of specific therapy.

Dr. Peacock: Perfect, Dr. Boer, that’s a really nice review of what we know up to this point about galectin-3. I think the next question is—is it ready for the clinical world? And if so, how would we be using it? Dr. Di Somma, what do you think about this and the use of galectin-3 in the emergency department?

Dr. Di Somma: That’s a very important question, because for patients presenting to an emergency department for shortness of breath due to acute heart failure, the physician has 3 main important jobs to do. First, a quick diagnosis is needed to start immediately with treatment because from papers in the literature,⁸ it is very clear that how fast you are in making the diagnosis and starting the treatment is going to have an important impact on both the patient’s outcome and length of stay. So, the first job is to give an appropriate diagnosis, but of no less importance is to immediately perform a risk specification, which is very important in terms of determining how aggressive your treatment must be, as well as the disposition based on this certification analysis of your patients with acute heart failure.

It seems possible now to discriminate between patients arriving in the Emergency Department with signs and symptoms of acute heart failure with galectin-3 levels of <17.8 pg/mL compared to patients with a galectin-3 level greater than this cutoff. Patients with a galectin-3 level of <17.8 pg/mL have a better prognosis after 30 days on follow-up examination, which, from the emergency physician’s point of view, is of great importance in immediately understanding the severity of the patient’s condition, because the disposition for these patients should be different.

Galectin-3 levels could be influenced by age and the kidney function. So if the patient is very old (> 75 years) and his estimated glomerular filtration rate is below 30 mL/min, the galectin-3 levels could be higher than expected. This could be related to the ongoing fibrotic process in the kidney that galectin-3 is mirroring. Moreover, in patients presenting with concomitant oncologic disease, galectin levels could be also higher.

For acute heart failure treatment, we now have many options, but in majority of the cases, the therapy is based on the use of intravenous diuretics. The current guidelines recommend high doses of intravenous diuretics, but also consider the possibility that, if immediately after a patient presents to the emergency department, you are able to make a correct diagnosis and determine the treatment and the following disposition based on biomarkers, the dosage of diuretic should also be based on appropriate patient risk stratification. A galectin-3 level of >17.8 pg/mL for instance indicates that a patient is at high risk for hospitalization, so you must be more cautious with this patient, possibly with a more aggressive intravenous diuretic treatment. Furthermore, the emergency department physician could also decide on the basis of a galectin-3 level of >17.8 pg/mL whether or not to discharge these patients only after infusion or to admit this patient to the hospital. On the contrary, since the job of the emergency department physician is mostly to rule out less severe cases, a galectin-3 level of <17.8 pg/mL (indicating less heart fibrosis) would probably indicate that it is safe to discharge a patient after just a small increase in the diuretic treatment dosage.

The use of galectin-3 will also have an important impact on the problem of overcrowding in the emergency department. In the recertification of patients coming to the emergency department with acute heart failure, there is an important possibility that adding the opportunity to measure galectin-3 can positively impact treatment from the emergency department physician.

Dr. Peacock: I think you brought up some extremely important points. You have to make a diagnosis first. Galectin-3 does not diagnose heart failure, but as you said, once you’ve established the probability that the patient has heart failure, galectin-3 is an outstanding prognostic agent. As Dr. de Boer said, it reflects generalized fibrosis, and there is a lack of specificity in that situation. However, once you’ve decided that the diagnosis is heart failure, then it’s an excellent prognostic, and if you define the cutoff point as 17.8 pg/mL, it identifies patients who would be at lower or higher risk for adverse outcomes subsequent to discharge.

This is really important for emergency medicine because we really don’t have anything that tells us what the future will look like in heart failure, except for B-type natriuretic peptide (BNP), which is really poor at predicting the next 2 weeks, and troponin, which is excellent, but is only elevated in a very small minority of cases using the contemporary assays available in the United States at this time.

Dr. Maisel: You can also make a case that once a diagnosis of heart failure is made—it’s sort of a moderate or mild-to-moderate heart failure, the BNP level is just a little above the grey zone, maybe 500 or 600 pg/mL, and the patient responds well to the diuretic—this isn’t like a troponin-positive acute coronary syndrome patient where, even if it’s just slightly positive, you’re still going to admit them. But, if you could show a propensity to have fibrosis in the short term and long term in this case of heart failure, and especially, if the patient is new to the system, then you may want to admit the patient to make sure this person gets on excellent therapy—and I’ll be talking about the role of mineralocorticoid receptor antagonists in a few minutes.

I think it’s significant to say, “Maybe this guy should not go home.” His level is above 17.8 pg/mL, and he has mild heart failure. He’s prone to just go downhill with each recurrent heart failure admission, and we need to take a closer look. If, in fact, you don’t admit him, then this is a person I would bring back within probably the first week to keep a close eye on.

Dr. Peacock: Dr. Maisel, what are you going to do when Dr. Di Somma sees a patient in the emergency department, their galectin-3 level is 27 pg/mL, and he sends the patient upstairs? Now, you’re at the receiving end.

With me is Dr. Rudolf de Boer, from University Medical Center, Groningen, The Netherlands, who has done some of the basic science work; Dr. Salvatore Di Somma, who is an emergency physician at San Andrea Hospital at the University Sapienza in Rome; and Dr. Alan Maisel, a world-renowned biomarker researcher from the University of California San Diego at the Veterans Administration. I am Frank Peacock, a Professor in Emergency Medicine and the Director of Research at the Baylor College of Medicine.

Welcome, gentlemen. I’d like to open this up with Dr. Boer, starting with some discussion on what he knows about this marker.

Dr. de Boer: Yes, thank you, Dr. Peacock, for your introduction. I will briefly introduce galectin-3 to you, although I’m sure most of you have become familiar with it to some extent. It was actually not so long ago that galectin-3 was scientifically discovered in the 1980s. When it was discovered, its function was not fully recognized. It was only 8 years ago that galectin-3 was associated with heart failure development and cardiac remodeling.¹

Galectin-3 is a member of the galectin family, which is a large family comprising about 20 galectins. Galectin-3 is unique because it’s a so-called chimera-type galectin, meaning that upon lectin binding, it forms multidimers, shaping a mesh, which is thought to add to the stiffness of the interstitial matrix, and in the case of heart failure, it adds stiffness to the extracellular myocardial matrix.² Galectin-3 can be activated by all kinds of lectins, and in the normal heart, these ligands are believed to be major cellular proteins such as laminin and collagens. In damaged organs in general, and in heart failure in particular, galectin-3 is predominantly released by macrophages and the activating lectins are clearly more abundant, and this is generally thought to be the mechanism of galectin-3 activation in heart failure.

As I mentioned, a first seminal paper¹ on the potential role of galectin-3 in heart failure was published in 2004. In a rat model with transition from compensated to decompensated heart failure, galectin-3 was the most strongly differentially expressed gene in this transitional phase. With further experimentation, the investigators showed that adding galectin-3 into the pericardial sac rendered superphysiological levels of galectin-3, and with this perturbation, they were able to provoke adverse cardiac remodeling with collagen deposition and left ventricular dysfunction. So, not only does galectin-3 seem to be associated with heart failure development, but may also be a part of the pathophysiology of cardiac remodeling and heart failure. This, of course, is very interesting.

Furthermore, galectin-3 seems to be a factor that is very dominant in the matrix: it does not predominantly act on cardiomyocytes, but on fibroblasts and the matrix compartment of the heart. It is strictly colocalized with sites of fibrosis where injury occurs, and it was shown that the primary source of galectin-3 in the heart includes all kinds of inflammatory cells, mainly macrophages, at the site of injury, for example, post-myocardial infarction or in hypertensive models. However, some other cells also produce galectin-3 including cardiac fibroblasts and mast cells. Mast cells are clearly less abundant, and fibroblasts are very abundant but produce a rather low amount of galectin-3. So overall, the primary source of galectin-3 is believed to be macrophages. As it is currently understood, when galectin-3 is produced, it turns cardiac fibroblasts into matrix-producing myofibroblasts, which then adds to the remodeling, specifically extracellular matrix production and fibrogenesis, not so much left ventricular hypertrophy, and this may contribute to the ultimate culmination into overt heart failure.

This is a general scheme of how we think of galectin-3 right now. As said before, interestingly, galectin-3 is associated with the disease process, but when activated, it has also been shown to add to the pathophysiology. Besides the naturally occurring binding lectins, investigators have developed neutralizing lectins that may bind to galectin-3 and render it inactive. This approach may actually downregulate galectin-3 activity with the aim to attenuate the galectin-3—driven progress of cardiac remodeling and to halt heart failure development. Recently, these neutralizing lectins have been tested in several animal models of hypertension, and it was shown that galectin-3 can be inhibited; this is associated with the attenuation of end-organ damage.³,⁴

Besides cardiac fibrosis, it has been observed that galectin-3 is also activated in liver fibrosis,⁵ kidney fibrosis,⁶ and lung fibrosis.⁷ Experimental studies using pharmacological compounds specifically targeting galectin-3 in non-cardiac disease have also been quite successful in attenuating tissue fibrosis.⁷ I think this adds an extra level of interest to this protein—that is, it’s not only a marker of disease severity, but also some kind of culprit biomarker, suggesting that it is part of the pathophysiology of heart failure, and can be a target of specific therapy.

Dr. Peacock: Perfect, Dr. Boer, that’s a really nice review of what we know up to this point about galectin-3. I think the next question is—is it ready for the clinical world? And if so, how would we be using it? Dr. Di Somma, what do you think about this and the use of galectin-3 in the emergency department?

Dr. Di Somma: That’s a very important question, because for patients presenting to an emergency department for shortness of breath due to acute heart failure, the physician has 3 main important jobs to do. First, a quick diagnosis is needed to start immediately with treatment because from papers in the literature,⁸ it is very clear that how fast you are in making the diagnosis and starting the treatment is going to have an important impact on both the patient’s outcome and length of stay. So, the first job is to give an appropriate diagnosis, but of no less importance is to immediately perform a risk specification, which is very important in terms of determining how aggressive your treatment must be, as well as the disposition based on this certification analysis of your patients with acute heart failure.

It seems possible now to discriminate between patients arriving in the Emergency Department with signs and symptoms of acute heart failure with galectin-3 levels of <17.8 pg/mL compared to patients with a galectin-3 level greater than this cutoff. Patients with a galectin-3 level of <17.8 pg/mL have a better prognosis after 30 days on follow-up examination, which, from the emergency physician’s point of view, is of great importance in immediately understanding the severity of the patient’s condition, because the disposition for these patients should be different.

Galectin-3 levels could be influenced by age and the kidney function. So if the patient is very old (> 75 years) and his estimated glomerular filtration rate is below 30 mL/min, the galectin-3 levels could be higher than expected. This could be related to the ongoing fibrotic process in the kidney that galectin-3 is mirroring. Moreover, in patients presenting with concomitant oncologic disease, galectin levels could be also higher.

For acute heart failure treatment, we now have many options, but in majority of the cases, the therapy is based on the use of intravenous diuretics. The current guidelines recommend high doses of intravenous diuretics, but also consider the possibility that, if immediately after a patient presents to the emergency department, you are able to make a correct diagnosis and determine the treatment and the following disposition based on biomarkers, the dosage of diuretic should also be based on appropriate patient risk stratification. A galectin-3 level of >17.8 pg/mL for instance indicates that a patient is at high risk for hospitalization, so you must be more cautious with this patient, possibly with a more aggressive intravenous diuretic treatment. Furthermore, the emergency department physician could also decide on the basis of a galectin-3 level of >17.8 pg/mL whether or not to discharge these patients only after infusion or to admit this patient to the hospital. On the contrary, since the job of the emergency department physician is mostly to rule out less severe cases, a galectin-3 level of <17.8 pg/mL (indicating less heart fibrosis) would probably indicate that it is safe to discharge a patient after just a small increase in the diuretic treatment dosage.

The use of galectin-3 will also have an important impact on the problem of overcrowding in the emergency department. In the recertification of patients coming to the emergency department with acute heart failure, there is an important possibility that adding the opportunity to measure galectin-3 can positively impact treatment from the emergency department physician.

Dr. Peacock: I think you brought up some extremely important points. You have to make a diagnosis first. Galectin-3 does not diagnose heart failure, but as you said, once you’ve established the probability that the patient has heart failure, galectin-3 is an outstanding prognostic agent. As Dr. de Boer said, it reflects generalized fibrosis, and there is a lack of specificity in that situation. However, once you’ve decided that the diagnosis is heart failure, then it’s an excellent prognostic, and if you define the cutoff point as 17.8 pg/mL, it identifies patients who would be at lower or higher risk for adverse outcomes subsequent to discharge.

This is really important for emergency medicine because we really don’t have anything that tells us what the future will look like in heart failure, except for B-type natriuretic peptide (BNP), which is really poor at predicting the next 2 weeks, and troponin, which is excellent, but is only elevated in a very small minority of cases using the contemporary assays available in the United States at this time.

Dr. Maisel: You can also make a case that once a diagnosis of heart failure is made—it’s sort of a moderate or mild-to-moderate heart failure, the BNP level is just a little above the grey zone, maybe 500 or 600 pg/mL, and the patient responds well to the diuretic—this isn’t like a troponin-positive acute coronary syndrome patient where, even if it’s just slightly positive, you’re still going to admit them. But, if you could show a propensity to have fibrosis in the short term and long term in this case of heart failure, and especially, if the patient is new to the system, then you may want to admit the patient to make sure this person gets on excellent therapy—and I’ll be talking about the role of mineralocorticoid receptor antagonists in a few minutes.

I think it’s significant to say, “Maybe this guy should not go home.” His level is above 17.8 pg/mL, and he has mild heart failure. He’s prone to just go downhill with each recurrent heart failure admission, and we need to take a closer look. If, in fact, you don’t admit him, then this is a person I would bring back within probably the first week to keep a close eye on.

Dr. Peacock: Dr. Maisel, what are you going to do when Dr. Di Somma sees a patient in the emergency department, their galectin-3 level is 27 pg/mL, and he sends the patient upstairs? Now, you’re at the receiving end.

That’s one of the reasons we’re having this conversation. One of the more controversial issues in medicine has been that of hypertension in pregnant women. The guidelines have not changed much in the last 50 years, although some effort to update these guidelines has been made recently. Today, we discuss the following: (1) The definition of elevated BP in pregnancy: is it too complicated? (2) Indications for therapy; should you wait until the BP increases to levels of 160/105 to 110 mm Hg? Is there any danger in allowing the systolic BP to remain elevated for a few months in a pregnant woman? (3) The indications for interruption of pregnancy; should we begin treatment earlier and continue to use the drugs that have been suggested over the years? During pregnancy, hypertension is typically classified as chronic hypertension (pre-existing or onset prior to 20 weeks gestation), gestational hypertension, preeclampsia, or preeclampsia superimposed on chronic hypertension.

I’m Dr. Marvin Moser, Clinical Professor of Medicine at Yale. We have with us Dr. Phyllis August, Professor of Research and Medicine and OB/GYN at Cornell Weill Medical College; Vesna Garovic, a Professor of Medicine at the Mayo Clinic who has done a great deal of work on hypertension in pregnancy; and Dr. Carl Rose, Associate Professor of Maternal-Fetal Medicine at the Mayo Clinic, to discuss this controversial subject.

Dr. August, let’s start with you. What about the definition of hypertension? We have a very complicated list of definitions of hypertension in pregnancy; do you want to review some of those? Should we simplify the criteria for clinicians?

Dr. August: I think we need to distinguish between the diagnostic category and the definition of hypertension. The definition of hypertension for the general, non-pregnant population is not very different from that for pregnant women. Obstetricians speak about mild hypertension in pregnancy, which they define as a systolic pressure of 140 to 150 mm Hg and diastolic pressure of 90 to 109 mm Hg, and then severe hypertension. There are 2 categories for hypertension: severe hypertension is a pressure of 160/110 mm Hg and above, and mild to moderate hypertension is a pressure below that level. The diagnosis of hypertension is made at a pressure of 140/90 mm Hg and above, so really, it’s not very different from Stages 1 and 2 of essential hypertension outlined in the last Joint National Committee report.¹

Preeclampsia, preeclampsia superimposed on chronic hypertension, chronic hypertension alone, and gestational hypertension are the diagnostic categories within which you can have either mild to moderate or severe hypertension.

Dr. Moser: Do we need all these designations? I know that the pathophysiology of preeclampsia is different from that of essential hypertension, but are the definitions helpful for the clinician? What’s the difference, for example, if you have someone with chronic hypertension who becomes pregnant? What if you have someone with pre-eclampsia superimposed on chronic hypertension, and you have these 4 categories. Are they helpful? Does this definition affect therapy?

Dr. August: They are helpful because what we really need to identify is the condition of the woman who, because of the pregnancy, is either developing new hypertension, or that of a woman who had hypertension before, but because of the pregnancy, her hypertension is worsening. That situation can change quickly, leading to serious maternal vmorbidity and fetal morbidity. This also helps to identify women who really need to be watched closely and addressed in a different way from someone who, at the beginning of pregnancy, had a pressure of 150/100 mm Hg and has the same pressure even at mid-pregnancy. I think the diagnostic categories are useful, and the fact that they’ve been used for many decades supports their effectiveness. They’re not perfect, but they are useful.

Dr. Moser: Dr. Rose, do you want to add to that?

Dr. Rose: As an obstetrician, I would suggest that these categories allow us to communicate with one another using common terminology that we collectively understand. It also influences the obstetric management of these patients.

Dr. Garovic: It is likely that young women who don’t see a physician and present with a BP of 150/100 mm Hg when they are pregnant, have had hypertension for a longer period, which has been asymptomatic and subclinical and that it has been brought to the attention of the clinician only due to the pregnancy.

Having said that, I’m not sure that it helps that much with respect to treatment because target levels at which BP treatment is started are decreasing, especially in the tertiary centers. My approach is to treat a pregnant woman with a documented systolic BP of 150 mm Hg on at least 2 occasions, whether symptomatic or asymptomatic and at risk for preeclampsia or not.

Dr. Moser: In the last American Society of Hypertension recommendations,² it was clearly stated that the level of the BP should guide treatment and the decision of when to interrupt or deliver the pregnancy. So, if it is the level of BP that the clinician is following, why do we need sub-designations? In other words, if the woman has chronic hypertension to start with, you’re obviously going to try and keep the systolic BP below 140 mm Hg. If she develops hypertension in the first trimester or second trimester, which is a rare occurrence, you’re going to reduce the BP. If she develops hypertension in the third trimester with or without proteinuria, for example, you’re going to treat it. So, why don’t we simplify this and treat the BP without 4 or 5 definitions? Of course, if the BP rises and there is proteinuria in the third trimester, other options will also need to be considered. Also, proteinuria in the second trimester is an indication of preeclampsia.

Dr. August: In my experience, if you have a patient who has a BP reading of 120/80 mm Hg at the start of pregnancy, and then at 20 weeks, the BP is 150/100 mm Hg, even with no proteinuria, you have to be very worried about the patient because she’s developing either superimposed preeclampsia or gestational hypertension, until proven otherwise. That’s different from a BP of 150/100 mm Hg that was stable and started out that way at 8 weeks of pregnancy, and then at 20 weeks, it still is 150/100 mm Hg. There has been no change, and no evidence for an evolving process that is potentially dangerous to the mother and fetus, so I would disagree that it’s just a number.

Dr. Moser: How would you treat the patient differently, Dr. August?

Dr. August: If a woman had a clear, well-documented pressure of 110 to 120/70 to 80 mm Hg early in pregnancy and then in the second trimester and developed mild to moderate or Stage 1 hypertension, I would make sure to check all her laboratory data and rule out preeclampsia, but I would be very worried that she was developing a pregnancy-related disorder like pre-eclampsia or gestational hypertension, and I would examine her very frequently.

Dr. Moser: But would you treat her BP?

Dr. August: I would probably treat her BP if it was 150/95 to 99 mm Hg. Even though the guidelines say 160 mm Hg, 150 mm Hg is close to 160 mm Hg and since the patient’s pressure is currently at 150 mm Hg, it may increase to 160 mm Hg in an hour from now when the patient is a little more stressed and uncomfortable; therefore, 150 mm Hg is my cutoff.

Dr. Moser: Consider this: the normal systolic pressure of a pregnant woman is 110 mm Hg, 120 mm Hg, or even lower sometimes, and even in the second and third trimester, it shouldn’t be more than 120 mm Hg, and 130 mm Hg at the most. Now, if a woman’s pressure exceeds the 140/90 mm Hg cutoff of hypertension, why do obstetricians wait until the level increases to 150 mm Hg, or as you said, 160 mm Hg?

Dr. August: They wait because they think lowering the BP too much might compromise the placental perfusion.

Dr. Moser: Dr. Garovic or Dr. Rose, can you comment on the concept that an elevated systolic pressure for a couple of months in pregnancy isn’t going to do any harm?

Dr. Rose: I agree with Dr. August that the pathophysiology of the processes is different. In other words, I do not believe that we should simply treat hypertension and categorize all the accompanying processes as a similar prenatal complication. If a mother develops new-onset hypertension late in pregnancy, certainly this represents a different clinical scenario from one who develops hypertension at 18 weeks of gestation, although implications for the pregnancy may ultimately be similar.

Dr. August: But I would argue that a woman who develops hypertension at 18 weeks and didn’t have it at 12 weeks is at a more serious risk for preterm birth than the woman who develops hypertension in the third trimester.

Dr. Moser: Let me argue on the other side again. Such patients are at risk, but what are you going to do about the risk? You’re going to check some enzymes and chemistries, etc, but what is the only thing that you might do to reduce the risk of progression of hypertension or a small placenta, which may be a result of infarcts from elevated BP? I’m trying to determine whether just lowering the BP, even though the pathophysiology is different, might prevent progression of hypertension. I know it may not prevent the progression of preeclampsia, but early treatment might prevent or delay interruption of pregnancy, based upon the criteria of an elevated BP.

Dr. Garovic: I just want to clarify the point that I was discussing earlier. A BP of 150/100 mm Hg at 21 weeks’ gestation may represent a significantly elevated BP, as the patient’s BP, at that time of pregnancy, should be even lower than her pre-pregnancy levels. I would not treat somebody with a BP of 150/100 mm Hg based on a single office reading for the same reason that we may not treat hypertension based on a single office reading in the general population. However, if I have enough evidence that a patient has sustained hypertension, especially in the second trimester when the BP should be lower than non-pregnant levels, I will treat that patient. Treatment is given after a 6-hour BP reading or nurse-monitored BP readings over a 30-minute to 1-hour period to generate enough evidence to show that BP is certainly steadily elevated.

I would like to discuss a recent patient: she was a 36-year-old woman with a BP of 150/100 mm Hg. It was her first pregnancy, and she was approximately 12 weeks pregnant. She was referred to us from one of the regional clinics close to Rochester. She underwent a retroperitoneal ultrasound, which showed bilateral renal artery stenosis/fibromuscular secondary to fibromuscular dysplasia. So, she had a BP of 150/100 mm Hg, but by the time she saw us, her systolic pressure was 160 to 170 mm Hg and diastolic pressure was in the 100s range. Dr. Rose and I aggressively treated her with 300 mg labetalol twice a day and nifedipine once a day; her BP effectively was reduced to 120 to 130 mm Hg systolic pressure and 70 to 80 mm Hg diastolic pressure, and she delivered a full-term baby boy at 38 gestational weeks.

Dr. August: Let me ask you a question: did she really have renovascular hypertension? Was she successfully revascularized and did her blood pressure normalize after revascularization?

Dr. Garovic: At 6 months post-partum, she underwent a renal angiogram with bilateral angioplasties with a subsequent normalization of her BP. She remains normotensive, when off BP medications.

Dr. Moser: Do you believe that if we treat elevated BP early, we might delay or prevent iatrogenic preterm deliveries?

Dr. August: Yes, I think your point is that if somebody in the third trimester has preeclampsia or gestational hypertension and severe hypertension, and you lower their BP and, if all other parameters are normal, the baby is fine, and the mother’s platelet count is normal, you can prevent her from having a preterm birth. You can extend the pregnancy by preventing the BP from reaching levels that are too dangerous and thus require delivery.

That’s one of the reasons we’re having this conversation. One of the more controversial issues in medicine has been that of hypertension in pregnant women. The guidelines have not changed much in the last 50 years, although some effort to update these guidelines has been made recently. Today, we discuss the following: (1) The definition of elevated BP in pregnancy: is it too complicated? (2) Indications for therapy; should you wait until the BP increases to levels of 160/105 to 110 mm Hg? Is there any danger in allowing the systolic BP to remain elevated for a few months in a pregnant woman? (3) The indications for interruption of pregnancy; should we begin treatment earlier and continue to use the drugs that have been suggested over the years? During pregnancy, hypertension is typically classified as chronic hypertension (pre-existing or onset prior to 20 weeks gestation), gestational hypertension, preeclampsia, or preeclampsia superimposed on chronic hypertension.

I’m Dr. Marvin Moser, Clinical Professor of Medicine at Yale. We have with us Dr. Phyllis August, Professor of Research and Medicine and OB/GYN at Cornell Weill Medical College; Vesna Garovic, a Professor of Medicine at the Mayo Clinic who has done a great deal of work on hypertension in pregnancy; and Dr. Carl Rose, Associate Professor of Maternal-Fetal Medicine at the Mayo Clinic, to discuss this controversial subject.

Dr. August, let’s start with you. What about the definition of hypertension? We have a very complicated list of definitions of hypertension in pregnancy; do you want to review some of those? Should we simplify the criteria for clinicians?

Dr. August: I think we need to distinguish between the diagnostic category and the definition of hypertension. The definition of hypertension for the general, non-pregnant population is not very different from that for pregnant women. Obstetricians speak about mild hypertension in pregnancy, which they define as a systolic pressure of 140 to 150 mm Hg and diastolic pressure of 90 to 109 mm Hg, and then severe hypertension. There are 2 categories for hypertension: severe hypertension is a pressure of 160/110 mm Hg and above, and mild to moderate hypertension is a pressure below that level. The diagnosis of hypertension is made at a pressure of 140/90 mm Hg and above, so really, it’s not very different from Stages 1 and 2 of essential hypertension outlined in the last Joint National Committee report.¹

Preeclampsia, preeclampsia superimposed on chronic hypertension, chronic hypertension alone, and gestational hypertension are the diagnostic categories within which you can have either mild to moderate or severe hypertension.

Dr. Moser: Do we need all these designations? I know that the pathophysiology of preeclampsia is different from that of essential hypertension, but are the definitions helpful for the clinician? What’s the difference, for example, if you have someone with chronic hypertension who becomes pregnant? What if you have someone with pre-eclampsia superimposed on chronic hypertension, and you have these 4 categories. Are they helpful? Does this definition affect therapy?

Dr. August: They are helpful because what we really need to identify is the condition of the woman who, because of the pregnancy, is either developing new hypertension, or that of a woman who had hypertension before, but because of the pregnancy, her hypertension is worsening. That situation can change quickly, leading to serious maternal vmorbidity and fetal morbidity. This also helps to identify women who really need to be watched closely and addressed in a different way from someone who, at the beginning of pregnancy, had a pressure of 150/100 mm Hg and has the same pressure even at mid-pregnancy. I think the diagnostic categories are useful, and the fact that they’ve been used for many decades supports their effectiveness. They’re not perfect, but they are useful.

Dr. Moser: Dr. Rose, do you want to add to that?

Dr. Rose: As an obstetrician, I would suggest that these categories allow us to communicate with one another using common terminology that we collectively understand. It also influences the obstetric management of these patients.

Dr. Garovic: It is likely that young women who don’t see a physician and present with a BP of 150/100 mm Hg when they are pregnant, have had hypertension for a longer period, which has been asymptomatic and subclinical and that it has been brought to the attention of the clinician only due to the pregnancy.

Having said that, I’m not sure that it helps that much with respect to treatment because target levels at which BP treatment is started are decreasing, especially in the tertiary centers. My approach is to treat a pregnant woman with a documented systolic BP of 150 mm Hg on at least 2 occasions, whether symptomatic or asymptomatic and at risk for preeclampsia or not.

Dr. Moser: In the last American Society of Hypertension recommendations,² it was clearly stated that the level of the BP should guide treatment and the decision of when to interrupt or deliver the pregnancy. So, if it is the level of BP that the clinician is following, why do we need sub-designations? In other words, if the woman has chronic hypertension to start with, you’re obviously going to try and keep the systolic BP below 140 mm Hg. If she develops hypertension in the first trimester or second trimester, which is a rare occurrence, you’re going to reduce the BP. If she develops hypertension in the third trimester with or without proteinuria, for example, you’re going to treat it. So, why don’t we simplify this and treat the BP without 4 or 5 definitions? Of course, if the BP rises and there is proteinuria in the third trimester, other options will also need to be considered. Also, proteinuria in the second trimester is an indication of preeclampsia.

Dr. August: In my experience, if you have a patient who has a BP reading of 120/80 mm Hg at the start of pregnancy, and then at 20 weeks, the BP is 150/100 mm Hg, even with no proteinuria, you have to be very worried about the patient because she’s developing either superimposed preeclampsia or gestational hypertension, until proven otherwise. That’s different from a BP of 150/100 mm Hg that was stable and started out that way at 8 weeks of pregnancy, and then at 20 weeks, it still is 150/100 mm Hg. There has been no change, and no evidence for an evolving process that is potentially dangerous to the mother and fetus, so I would disagree that it’s just a number.

Dr. Moser: How would you treat the patient differently, Dr. August?

Dr. August: If a woman had a clear, well-documented pressure of 110 to 120/70 to 80 mm Hg early in pregnancy and then in the second trimester and developed mild to moderate or Stage 1 hypertension, I would make sure to check all her laboratory data and rule out preeclampsia, but I would be very worried that she was developing a pregnancy-related disorder like pre-eclampsia or gestational hypertension, and I would examine her very frequently.

Dr. Moser: But would you treat her BP?

Dr. August: I would probably treat her BP if it was 150/95 to 99 mm Hg. Even though the guidelines say 160 mm Hg, 150 mm Hg is close to 160 mm Hg and since the patient’s pressure is currently at 150 mm Hg, it may increase to 160 mm Hg in an hour from now when the patient is a little more stressed and uncomfortable; therefore, 150 mm Hg is my cutoff.

Dr. Moser: Consider this: the normal systolic pressure of a pregnant woman is 110 mm Hg, 120 mm Hg, or even lower sometimes, and even in the second and third trimester, it shouldn’t be more than 120 mm Hg, and 130 mm Hg at the most. Now, if a woman’s pressure exceeds the 140/90 mm Hg cutoff of hypertension, why do obstetricians wait until the level increases to 150 mm Hg, or as you said, 160 mm Hg?

Dr. August: They wait because they think lowering the BP too much might compromise the placental perfusion.

Dr. Moser: Dr. Garovic or Dr. Rose, can you comment on the concept that an elevated systolic pressure for a couple of months in pregnancy isn’t going to do any harm?

Dr. Rose: I agree with Dr. August that the pathophysiology of the processes is different. In other words, I do not believe that we should simply treat hypertension and categorize all the accompanying processes as a similar prenatal complication. If a mother develops new-onset hypertension late in pregnancy, certainly this represents a different clinical scenario from one who develops hypertension at 18 weeks of gestation, although implications for the pregnancy may ultimately be similar.

Dr. August: But I would argue that a woman who develops hypertension at 18 weeks and didn’t have it at 12 weeks is at a more serious risk for preterm birth than the woman who develops hypertension in the third trimester.

Dr. Moser: Let me argue on the other side again. Such patients are at risk, but what are you going to do about the risk? You’re going to check some enzymes and chemistries, etc, but what is the only thing that you might do to reduce the risk of progression of hypertension or a small placenta, which may be a result of infarcts from elevated BP? I’m trying to determine whether just lowering the BP, even though the pathophysiology is different, might prevent progression of hypertension. I know it may not prevent the progression of preeclampsia, but early treatment might prevent or delay interruption of pregnancy, based upon the criteria of an elevated BP.

Dr. Garovic: I just want to clarify the point that I was discussing earlier. A BP of 150/100 mm Hg at 21 weeks’ gestation may represent a significantly elevated BP, as the patient’s BP, at that time of pregnancy, should be even lower than her pre-pregnancy levels. I would not treat somebody with a BP of 150/100 mm Hg based on a single office reading for the same reason that we may not treat hypertension based on a single office reading in the general population. However, if I have enough evidence that a patient has sustained hypertension, especially in the second trimester when the BP should be lower than non-pregnant levels, I will treat that patient. Treatment is given after a 6-hour BP reading or nurse-monitored BP readings over a 30-minute to 1-hour period to generate enough evidence to show that BP is certainly steadily elevated.

I would like to discuss a recent patient: she was a 36-year-old woman with a BP of 150/100 mm Hg. It was her first pregnancy, and she was approximately 12 weeks pregnant. She was referred to us from one of the regional clinics close to Rochester. She underwent a retroperitoneal ultrasound, which showed bilateral renal artery stenosis/fibromuscular secondary to fibromuscular dysplasia. So, she had a BP of 150/100 mm Hg, but by the time she saw us, her systolic pressure was 160 to 170 mm Hg and diastolic pressure was in the 100s range. Dr. Rose and I aggressively treated her with 300 mg labetalol twice a day and nifedipine once a day; her BP effectively was reduced to 120 to 130 mm Hg systolic pressure and 70 to 80 mm Hg diastolic pressure, and she delivered a full-term baby boy at 38 gestational weeks.

Dr. August: Let me ask you a question: did she really have renovascular hypertension? Was she successfully revascularized and did her blood pressure normalize after revascularization?

Dr. Garovic: At 6 months post-partum, she underwent a renal angiogram with bilateral angioplasties with a subsequent normalization of her BP. She remains normotensive, when off BP medications.

Dr. Moser: Do you believe that if we treat elevated BP early, we might delay or prevent iatrogenic preterm deliveries?

Dr. August: Yes, I think your point is that if somebody in the third trimester has preeclampsia or gestational hypertension and severe hypertension, and you lower their BP and, if all other parameters are normal, the baby is fine, and the mother’s platelet count is normal, you can prevent her from having a preterm birth. You can extend the pregnancy by preventing the BP from reaching levels that are too dangerous and thus require delivery.

That’s one of the reasons we’re having this conversation. One of the more controversial issues in medicine has been that of hypertension in pregnant women. The guidelines have not changed much in the last 50 years, although some effort to update these guidelines has been made recently. Today, we discuss the following: (1) The definition of elevated BP in pregnancy: is it too complicated? (2) Indications for therapy; should you wait until the BP increases to levels of 160/105 to 110 mm Hg? Is there any danger in allowing the systolic BP to remain elevated for a few months in a pregnant woman? (3) The indications for interruption of pregnancy; should we begin treatment earlier and continue to use the drugs that have been suggested over the years? During pregnancy, hypertension is typically classified as chronic hypertension (pre-existing or onset prior to 20 weeks gestation), gestational hypertension, preeclampsia, or preeclampsia superimposed on chronic hypertension.

I’m Dr. Marvin Moser, Clinical Professor of Medicine at Yale. We have with us Dr. Phyllis August, Professor of Research and Medicine and OB/GYN at Cornell Weill Medical College; Vesna Garovic, a Professor of Medicine at the Mayo Clinic who has done a great deal of work on hypertension in pregnancy; and Dr. Carl Rose, Associate Professor of Maternal-Fetal Medicine at the Mayo Clinic, to discuss this controversial subject.

Dr. August, let’s start with you. What about the definition of hypertension? We have a very complicated list of definitions of hypertension in pregnancy; do you want to review some of those? Should we simplify the criteria for clinicians?

Dr. August: I think we need to distinguish between the diagnostic category and the definition of hypertension. The definition of hypertension for the general, non-pregnant population is not very different from that for pregnant women. Obstetricians speak about mild hypertension in pregnancy, which they define as a systolic pressure of 140 to 150 mm Hg and diastolic pressure of 90 to 109 mm Hg, and then severe hypertension. There are 2 categories for hypertension: severe hypertension is a pressure of 160/110 mm Hg and above, and mild to moderate hypertension is a pressure below that level. The diagnosis of hypertension is made at a pressure of 140/90 mm Hg and above, so really, it’s not very different from Stages 1 and 2 of essential hypertension outlined in the last Joint National Committee report.¹

Preeclampsia, preeclampsia superimposed on chronic hypertension, chronic hypertension alone, and gestational hypertension are the diagnostic categories within which you can have either mild to moderate or severe hypertension.

Dr. Moser: Do we need all these designations? I know that the pathophysiology of preeclampsia is different from that of essential hypertension, but are the definitions helpful for the clinician? What’s the difference, for example, if you have someone with chronic hypertension who becomes pregnant? What if you have someone with pre-eclampsia superimposed on chronic hypertension, and you have these 4 categories. Are they helpful? Does this definition affect therapy?

Dr. August: They are helpful because what we really need to identify is the condition of the woman who, because of the pregnancy, is either developing new hypertension, or that of a woman who had hypertension before, but because of the pregnancy, her hypertension is worsening. That situation can change quickly, leading to serious maternal vmorbidity and fetal morbidity. This also helps to identify women who really need to be watched closely and addressed in a different way from someone who, at the beginning of pregnancy, had a pressure of 150/100 mm Hg and has the same pressure even at mid-pregnancy. I think the diagnostic categories are useful, and the fact that they’ve been used for many decades supports their effectiveness. They’re not perfect, but they are useful.

Dr. Moser: Dr. Rose, do you want to add to that?

Dr. Rose: As an obstetrician, I would suggest that these categories allow us to communicate with one another using common terminology that we collectively understand. It also influences the obstetric management of these patients.

Dr. Garovic: It is likely that young women who don’t see a physician and present with a BP of 150/100 mm Hg when they are pregnant, have had hypertension for a longer period, which has been asymptomatic and subclinical and that it has been brought to the attention of the clinician only due to the pregnancy.

Having said that, I’m not sure that it helps that much with respect to treatment because target levels at which BP treatment is started are decreasing, especially in the tertiary centers. My approach is to treat a pregnant woman with a documented systolic BP of 150 mm Hg on at least 2 occasions, whether symptomatic or asymptomatic and at risk for preeclampsia or not.

Dr. Moser: In the last American Society of Hypertension recommendations,² it was clearly stated that the level of the BP should guide treatment and the decision of when to interrupt or deliver the pregnancy. So, if it is the level of BP that the clinician is following, why do we need sub-designations? In other words, if the woman has chronic hypertension to start with, you’re obviously going to try and keep the systolic BP below 140 mm Hg. If she develops hypertension in the first trimester or second trimester, which is a rare occurrence, you’re going to reduce the BP. If she develops hypertension in the third trimester with or without proteinuria, for example, you’re going to treat it. So, why don’t we simplify this and treat the BP without 4 or 5 definitions? Of course, if the BP rises and there is proteinuria in the third trimester, other options will also need to be considered. Also, proteinuria in the second trimester is an indication of preeclampsia.

Dr. August: In my experience, if you have a patient who has a BP reading of 120/80 mm Hg at the start of pregnancy, and then at 20 weeks, the BP is 150/100 mm Hg, even with no proteinuria, you have to be very worried about the patient because she’s developing either superimposed preeclampsia or gestational hypertension, until proven otherwise. That’s different from a BP of 150/100 mm Hg that was stable and started out that way at 8 weeks of pregnancy, and then at 20 weeks, it still is 150/100 mm Hg. There has been no change, and no evidence for an evolving process that is potentially dangerous to the mother and fetus, so I would disagree that it’s just a number.

Dr. Moser: How would you treat the patient differently, Dr. August?

Dr. August: If a woman had a clear, well-documented pressure of 110 to 120/70 to 80 mm Hg early in pregnancy and then in the second trimester and developed mild to moderate or Stage 1 hypertension, I would make sure to check all her laboratory data and rule out preeclampsia, but I would be very worried that she was developing a pregnancy-related disorder like pre-eclampsia or gestational hypertension, and I would examine her very frequently.

Dr. Moser: But would you treat her BP?

Dr. August: I would probably treat her BP if it was 150/95 to 99 mm Hg. Even though the guidelines say 160 mm Hg, 150 mm Hg is close to 160 mm Hg and since the patient’s pressure is currently at 150 mm Hg, it may increase to 160 mm Hg in an hour from now when the patient is a little more stressed and uncomfortable; therefore, 150 mm Hg is my cutoff.

Dr. Moser: Consider this: the normal systolic pressure of a pregnant woman is 110 mm Hg, 120 mm Hg, or even lower sometimes, and even in the second and third trimester, it shouldn’t be more than 120 mm Hg, and 130 mm Hg at the most. Now, if a woman’s pressure exceeds the 140/90 mm Hg cutoff of hypertension, why do obstetricians wait until the level increases to 150 mm Hg, or as you said, 160 mm Hg?

Dr. August: They wait because they think lowering the BP too much might compromise the placental perfusion.

Dr. Moser: Dr. Garovic or Dr. Rose, can you comment on the concept that an elevated systolic pressure for a couple of months in pregnancy isn’t going to do any harm?

Dr. Rose: I agree with Dr. August that the pathophysiology of the processes is different. In other words, I do not believe that we should simply treat hypertension and categorize all the accompanying processes as a similar prenatal complication. If a mother develops new-onset hypertension late in pregnancy, certainly this represents a different clinical scenario from one who develops hypertension at 18 weeks of gestation, although implications for the pregnancy may ultimately be similar.

Dr. August: But I would argue that a woman who develops hypertension at 18 weeks and didn’t have it at 12 weeks is at a more serious risk for preterm birth than the woman who develops hypertension in the third trimester.

Dr. Moser: Let me argue on the other side again. Such patients are at risk, but what are you going to do about the risk? You’re going to check some enzymes and chemistries, etc, but what is the only thing that you might do to reduce the risk of progression of hypertension or a small placenta, which may be a result of infarcts from elevated BP? I’m trying to determine whether just lowering the BP, even though the pathophysiology is different, might prevent progression of hypertension. I know it may not prevent the progression of preeclampsia, but early treatment might prevent or delay interruption of pregnancy, based upon the criteria of an elevated BP.

Dr. Garovic: I just want to clarify the point that I was discussing earlier. A BP of 150/100 mm Hg at 21 weeks’ gestation may represent a significantly elevated BP, as the patient’s BP, at that time of pregnancy, should be even lower than her pre-pregnancy levels. I would not treat somebody with a BP of 150/100 mm Hg based on a single office reading for the same reason that we may not treat hypertension based on a single office reading in the general population. However, if I have enough evidence that a patient has sustained hypertension, especially in the second trimester when the BP should be lower than non-pregnant levels, I will treat that patient. Treatment is given after a 6-hour BP reading or nurse-monitored BP readings over a 30-minute to 1-hour period to generate enough evidence to show that BP is certainly steadily elevated.

I would like to discuss a recent patient: she was a 36-year-old woman with a BP of 150/100 mm Hg. It was her first pregnancy, and she was approximately 12 weeks pregnant. She was referred to us from one of the regional clinics close to Rochester. She underwent a retroperitoneal ultrasound, which showed bilateral renal artery stenosis/fibromuscular secondary to fibromuscular dysplasia. So, she had a BP of 150/100 mm Hg, but by the time she saw us, her systolic pressure was 160 to 170 mm Hg and diastolic pressure was in the 100s range. Dr. Rose and I aggressively treated her with 300 mg labetalol twice a day and nifedipine once a day; her BP effectively was reduced to 120 to 130 mm Hg systolic pressure and 70 to 80 mm Hg diastolic pressure, and she delivered a full-term baby boy at 38 gestational weeks.

Dr. August: Let me ask you a question: did she really have renovascular hypertension? Was she successfully revascularized and did her blood pressure normalize after revascularization?

Dr. Garovic: At 6 months post-partum, she underwent a renal angiogram with bilateral angioplasties with a subsequent normalization of her BP. She remains normotensive, when off BP medications.

Dr. Moser: Do you believe that if we treat elevated BP early, we might delay or prevent iatrogenic preterm deliveries?

Dr. August: Yes, I think your point is that if somebody in the third trimester has preeclampsia or gestational hypertension and severe hypertension, and you lower their BP and, if all other parameters are normal, the baby is fine, and the mother’s platelet count is normal, you can prevent her from having a preterm birth. You can extend the pregnancy by preventing the BP from reaching levels that are too dangerous and thus require delivery.

In 2011, the National Lipid Association (NLA) released clinical guidance and expert panel recommendations regarding the screening and treatment of patients with FH.¹ Two US advocacy groups have emerged that promote awareness and screening for this condition, and several new LDL-lowering drugs are in different phases of development. These agents have and will be evaluated in the FH population.

Recent observational data remind us that medication is the current cornerstone therapy for FH. Lipid-lowering therapies reduce the risk of CVD in these patients as compared with those who are not treated. There remain, however, many patients who cannot tolerate therapy or who, despite maximal pharmacologic treatment, do not achieve the recommended LDL targets. For these patients, LDL apheresis is an option available at approximately 35 centers throughout the United States. Finally, with new national cholesterol guidelines underway, there is renewed interest and awareness regarding the identification and treatment of patients with lipid disorders.

I’m Dr. James Underberg, Clinical Assistant Professor of Medicine at the NYU School of Medicine, NYU Center for Cardiovascular Disease. I am also Director of the lipid clinic at Bellevue Hospital in New York. I’m joined today by 3 internationally recognized experts in the field of hypercholesterolemia: Dr. Eliot Brinton, President of the Utah Lipid Center and Director of Atherometabolic Research at the Utah Foundation for Biomedical Research; Dr. Patrick Moriarty of the University of Kansas Medical Center; and Dr. Mary McGowan, Chief Medical Officer of the FH Foundation.

Dr. Brinton, can you briefly review the epidemiology, associated cardiovascular risk, and different diagnostic criteria for FH?

DR. BRINTON: This is among the most commonly occurring, clinically significant, monogenic metabolic disorders. It is codominant, so the clinical picture differs among homozygotes, heterozygotes, and the unaffected.

Heterozygous FH occurs in approximately 1 of 300 to 1 of 500 persons in the general population, although certain populations have a much higher prevalence. In groups such as the French Canadians and Dutch Afrikaners, 1 in 100 individuals are heterozygotes due to a founder effect. FH homozygotes are less common—about 1 in a million—and it is estimated that up to 10 000 homozygous FH patients are present worldwide.

One of the biggest clinical problems that we face with FH is detecting it or finding affected patients in the general population. The disease is treatable, but it is important to start treatment as early as possible. It has been very hard to identify such patients because they tend to be scattered throughout the population. Many of them may have never had even a single lipid panel, much less been referred to a lipidologist for proper care of their cholesterol disorder.

Another enormous challenge that we face in dealing with FH is that the patients have very, very high LDL-C levels—usually in the range of 250 to 500 mg/dL—in heterozygotes and usually above 500 mg/dL in homozygotes. These are untreated levels, and of course, once treatment is started, the levels fall. The underlying cause of FH in most cases is a defect in the LDL receptor.

The action of the LDL receptor in the liver is the major mechanism responsible for clearing LDL particles from the bloodstream, and if one has a defective receptor such that it is either never synthesized or has little or no capacity to bind to the LDL particle, then the LDL levels greatly increase in the blood. Of course, if someone is heterozygous, he/she will have one functioning receptor and one missing or malfunctioning receptor. By definition, homozygotes have defects in both copies of the LDL receptor gene; therefore, they have very little, if any, LDL receptor activity. This is classic FH, but there are other causes of severe hypercholesterolemia too. Although they are genetically and causally distinct, the increase in LDL-C levels may be similar to that in classic FH. When 2 distinct mutations cause the same phenotype, they are said to be phenocopies of each other.

For example, approximately 10% of patients with FH have a defect in apolipoprotein B (apoB), which is the major ligand for the LDL receptor, rather than a defect in the LDL receptor itself. A much smaller number of patients appear to have a defect in proprotein convertase subtilisin/kexin type 9 (PCSK9), which is a factor that helps process the LDL receptor. Because this factor functions to reduce the activity of LDL receptors, it is a rare gain-of-function mutation that reduces LDL receptor activity, thereby causing an increase in LDL-C levels.

Interestingly, even though a defect in the LDL receptor is by far the most common cause of FH, these defects are very heterogeneous in nature; thousands of individual mutations of the LDL receptor have been discovered in genotyping studies of FH patients. However, genetically isolated populations often have a founder effect with more uniform mutations as well as a higher prevalence of FH. Generally, the most striking aspect of FH is that even though the phenotype tends to be quite similar from genotype to genotype, there is incredible genetic heterogeneity.

FH was one of the very first monogenic disorders with important clinical sequelae ever discovered. These patients tend to have premature CVD, which is CVD occurring in men younger than 45 years and women younger than 55 years of age. CVD is quite uncommon among people in these age groups in the general population. The risk ratio of CVD for people with to those without heterozygous FH is very high at these younger ages, and we generally see a 5- to 20-fold increase, even though, in absolute terms, there are few events. The risk ratios and absolute rates of CVD are much higher in FH homozygotes, especially in those younger than 25 years of age.

Past middle age, CVD becomes quite common in the general population, although obviously, FH patients are also at progressively increased risk for developing CVD at older ages. In homozygous FH patients, CVD can occur before the age of 10 years. It’s really very troubling to see a young child have a heart attack, for example. Thankfully, with improved diagnosis and treatment, we’re now seeing homozygous FH patients getting well into their teens and even sometimes into adulthood before they have their first cardiovascular event.

DR. UNDERBERG: So, how would you diagnose the condition?

DR. BRINTON: That’s a great question. The diagnosis usually should be made clinically, in my opinion. In this modern era, it is tempting to jump right ahead to genetic testing, and indeed, genetic testing can help make or confirm a diagnosis of FH; however, in a significant percentage of patients who truly have FH, the genetic abnormality cannot be determined. More importantly, treatment must be guided by the lipid levels, not the genotype; therefore, lipid testing is much more important in patient management.

Screening can start with determining just the total cholesterol level in a non-fasting specimen. If it’s above 200 mg/dL, then FH may be present, and a fasting lipid panel with an LDL-C level is needed. In children or adolescents, if the LDL-C level is above 160 mg/dL or the non-high density lipoprotein (HDL), which is total cholesterol minus HDL cholesterol (HDL-C), is above 190 mg/dL, then FH is strongly suspected.

In adults over the age of 20 years, the cutoffs are higher because the LDL-C level increases with age, and FH is suspected if the LDL-C level is above 190 mg/dL or if the non-HDL-C level is above 220 mg/dL. FH is very likely at higher levels: for individuals aged below 20 years, this level is an LDL-C of above 190 mg/dL; for those aged 20 to 30 years, it’s an LDL-C level of above 220 mg/dL; and for those aged 30 years or older, the LDL-C threshold is 250 mg/dL.

DR. UNDERBERG: Are those numbers are from the Make Early Diagnosis to Prevent Early Deaths (MEDPED) criteria?²

DR. BRINTON: Yes, and they are also from the NLA Expert Panel statement,³ which, as you mentioned, came out in the middle of 2011.

DR. UNDERBERG: That’s US-based. Are there any other criteria that are used globally?

DR. BRINTON: These are non-US criteria, but they’re fairly similar, so I think we should focus on the US-based guidelines. Screening and diagnostic cutoffs are, of course, always a tradeoff between sensitivity and specificity. Lower cutoffs will have greater sensitivity but less specificity. Thus, you’re going to have a larger number of non-FH patients that you think might have FH. If you set the criteria or the cutoffs higher, then you’re much more likely to have a true FH case, but you will miss some patients with FH due to variability in LDL-C levels, even among patients with an identical FH mutation. These LDL-C variations can be due to variability in other genetic factors and environmental influences.

DR. UNDERBERG: Dr. McGowan, we’ve heard a little bit about diagnostic criteria. How are we doing right now with regard to diagnosis both here and globally? What options do we have to improve the diagnosis-making process with respect to screening, awareness, and other factors?

DR. MCGOWAN: Unfortunately, we’re doing fairly poorly, both in the United States and abroad. Roughly 20% of people with FH have received a diagnosis of FH. That doesn’t mean that people who have FH and have not yet been diagnosed are not being treated with lipid lowering agents: they may be. However, they are often not receiving an adequate dose of medication, and the failure to diagnose means a missed opportunity to educate patients and screen relatives; remember, this is an autosomal codominant disorder. Roughly half of the first-degree relatives of a person living with FH will also have FH.

General practitioners and even cardiologists often don’t think of FH when they have a young person with a coronary event admitted to their critical care unit. This is very unfortunate. It may occur in part because coronary disease is such a common disorder in the United States. When physicians miss the diagnosis of FH, they miss the opportunity to educate patients about cascade screening. This is the process I just referred to. Cascade screening involves screening all first-degree relatives of an index case and repeating the exercise in the first-degree relatives of the newly identified patients. As Dr. Brinton pointed out, FH is treatable, and the sooner we make the diagnosis, the more likely we are to prevent a cardiac event.

There are certainly some physical findings that we look for in patients with FH, but these are often quite subtle and are frequently missed. We can look for xanthomas in the Achilles tendon or the extensor tendons of the hands. Corneal arcus is not pathognomonic for FH, but when seen in the correct setting, it may help you make a diagnosis.

These physical findings are actually used in some of the FH-screening tools. Dr. Brinton mentioned the MEDPED, which was developed in Utah. Other screening tools include the Simon Broome criteria⁴ and the Dutch Lipid Clinic Network criteria.⁵ Both the Simon Broome and the Dutch Lipid Clinic Network criteria evaluate lipid levels in the context of physical findings and determine the probability that a person has FH.

DR. UNDERBERG: Does making the diagnosis of FH alter the way you would treat someone, especially with respect to how aggressively you would treat someone?

DR. MCGOWAN: This is a very important question. When we think about patients with FH, one of the things we know is that they’ve had elevated lipids since birth. In fact, if a person inherits the FH from his/her mother, he/she has been exposed to the mother’s very elevated lipids in utero and may have a greater cardiac risk than somebody who inherited the FH from his/her father. Having elevated lipids from birth is very different from gaining weight at the age of 40 years and developing hyperlipidemia in mid-life. The NLA has specifically pointed out that we should not be using the Framingham Risk Score in patients with FH because, in fact, they should all be considered very high risk. Anyone with FH should be treated very aggressively.

When we compare children with FH to their unaffected siblings, we see a significant difference in terms of carotid intima-media thickness by approximately 12 years of age.⁶ The vascular trajectory is very different in people with FH as compared to people who develop hyperlipidemia later in life. We should not be satisfied with achieving an LDL-C level of 190, 160, or even 130 mg/dL. We should try to reduce the LDL level as far as possible. Sometimes, this is very difficult because patients with FH don’t respond to lipid-lowering agents as well as patients with polygenic hyperlipidemia.

For example, homozygotes achieve about a 25% reduction in the LDL level at maximum doses of statins in combination with ezetimibe.⁷ Clearly, homozygous FH patients are at very high risk, and unfortunately, they don’t get the 50% to 60% reduction that we see when we treat other people with high doses of statins.⁸ Heterozygotes have a better response, but it’s still not the same response that you see in people with polygenic hyperlipidemia. Therefore, the answer to your question is yes, we should treat FH patients differently.

DR. BRINTON: I generally agree with what Dr. McGowan has said, but I am slightly more agnostic. In my view, there isn’t much clinical value in making a genetic diagnosis of FH, and here’s why: For everyone with suspected FH, we need to obtain the best-possible family history. If the person has a family history of CVD, especially before middle age and especially in more than 1 close relative, 2 things should occur: first, aggressive LDL lowering in the patient, and second, aggressive screening in family members. In cases where little or no family history can be obtained—for example, if the patient was adopted—then we can either do genetic testing or simply assume the presence of FH and proceed to aggressive treatment and screening of any blood relatives of the patient. Therefore, a positive family history and a very high LDL-C level will, in fact, lead us to the aggressive treatment that Dr. McGowan mentioned and to family screening as well.

A lipidologist may want to make the diagnosis of FH in a formal genetic manner, and I can think of some scenarios in which this might be useful. I believe that some of the difficulty that we’ve had in the broader physician community is that FH can seem like an arcane entity with many complicated criteria. I fear that, often, primary care doctors may give up in despair and leave patients undiagnosed and untreated, or undertreated, and do not refer them. I think we need to keep our messages about FH very simple—not only for patients but also for general physicians.

Even though I’ve spent much of my career working in genetically oriented academic institutions, I feel that we and our patients are best served by focusing on the phenotype, or lipid levels, rather than on the genotype in cases of severe hypercholesterolemia. By definition, the levels in these cases are far above the LDL-C cutoffs, and we need to be prepared to give such patients high doses of more than one LDL-lowering medication, as Dr. McGowan has said. Given limited resources, I don’t think we should be worrying so much about the finer details of how to diagnose FH. Instead, we should put that effort toward aggressive screening and treatment.

DR. UNDERBERG: Dr. McGowan, there are some criteria elucidated by the NLA that might create subsets of FH patients that we should target for LDL-C lowering. Do you want to point some of these out?

In 2011, the National Lipid Association (NLA) released clinical guidance and expert panel recommendations regarding the screening and treatment of patients with FH.¹ Two US advocacy groups have emerged that promote awareness and screening for this condition, and several new LDL-lowering drugs are in different phases of development. These agents have and will be evaluated in the FH population.

Recent observational data remind us that medication is the current cornerstone therapy for FH. Lipid-lowering therapies reduce the risk of CVD in these patients as compared with those who are not treated. There remain, however, many patients who cannot tolerate therapy or who, despite maximal pharmacologic treatment, do not achieve the recommended LDL targets. For these patients, LDL apheresis is an option available at approximately 35 centers throughout the United States. Finally, with new national cholesterol guidelines underway, there is renewed interest and awareness regarding the identification and treatment of patients with lipid disorders.

I’m Dr. James Underberg, Clinical Assistant Professor of Medicine at the NYU School of Medicine, NYU Center for Cardiovascular Disease. I am also Director of the lipid clinic at Bellevue Hospital in New York. I’m joined today by 3 internationally recognized experts in the field of hypercholesterolemia: Dr. Eliot Brinton, President of the Utah Lipid Center and Director of Atherometabolic Research at the Utah Foundation for Biomedical Research; Dr. Patrick Moriarty of the University of Kansas Medical Center; and Dr. Mary McGowan, Chief Medical Officer of the FH Foundation.

Dr. Brinton, can you briefly review the epidemiology, associated cardiovascular risk, and different diagnostic criteria for FH?

DR. BRINTON: This is among the most commonly occurring, clinically significant, monogenic metabolic disorders. It is codominant, so the clinical picture differs among homozygotes, heterozygotes, and the unaffected.

Heterozygous FH occurs in approximately 1 of 300 to 1 of 500 persons in the general population, although certain populations have a much higher prevalence. In groups such as the French Canadians and Dutch Afrikaners, 1 in 100 individuals are heterozygotes due to a founder effect. FH homozygotes are less common—about 1 in a million—and it is estimated that up to 10 000 homozygous FH patients are present worldwide.

One of the biggest clinical problems that we face with FH is detecting it or finding affected patients in the general population. The disease is treatable, but it is important to start treatment as early as possible. It has been very hard to identify such patients because they tend to be scattered throughout the population. Many of them may have never had even a single lipid panel, much less been referred to a lipidologist for proper care of their cholesterol disorder.

Another enormous challenge that we face in dealing with FH is that the patients have very, very high LDL-C levels—usually in the range of 250 to 500 mg/dL—in heterozygotes and usually above 500 mg/dL in homozygotes. These are untreated levels, and of course, once treatment is started, the levels fall. The underlying cause of FH in most cases is a defect in the LDL receptor.

The action of the LDL receptor in the liver is the major mechanism responsible for clearing LDL particles from the bloodstream, and if one has a defective receptor such that it is either never synthesized or has little or no capacity to bind to the LDL particle, then the LDL levels greatly increase in the blood. Of course, if someone is heterozygous, he/she will have one functioning receptor and one missing or malfunctioning receptor. By definition, homozygotes have defects in both copies of the LDL receptor gene; therefore, they have very little, if any, LDL receptor activity. This is classic FH, but there are other causes of severe hypercholesterolemia too. Although they are genetically and causally distinct, the increase in LDL-C levels may be similar to that in classic FH. When 2 distinct mutations cause the same phenotype, they are said to be phenocopies of each other.

For example, approximately 10% of patients with FH have a defect in apolipoprotein B (apoB), which is the major ligand for the LDL receptor, rather than a defect in the LDL receptor itself. A much smaller number of patients appear to have a defect in proprotein convertase subtilisin/kexin type 9 (PCSK9), which is a factor that helps process the LDL receptor. Because this factor functions to reduce the activity of LDL receptors, it is a rare gain-of-function mutation that reduces LDL receptor activity, thereby causing an increase in LDL-C levels.

Interestingly, even though a defect in the LDL receptor is by far the most common cause of FH, these defects are very heterogeneous in nature; thousands of individual mutations of the LDL receptor have been discovered in genotyping studies of FH patients. However, genetically isolated populations often have a founder effect with more uniform mutations as well as a higher prevalence of FH. Generally, the most striking aspect of FH is that even though the phenotype tends to be quite similar from genotype to genotype, there is incredible genetic heterogeneity.

FH was one of the very first monogenic disorders with important clinical sequelae ever discovered. These patients tend to have premature CVD, which is CVD occurring in men younger than 45 years and women younger than 55 years of age. CVD is quite uncommon among people in these age groups in the general population. The risk ratio of CVD for people with to those without heterozygous FH is very high at these younger ages, and we generally see a 5- to 20-fold increase, even though, in absolute terms, there are few events. The risk ratios and absolute rates of CVD are much higher in FH homozygotes, especially in those younger than 25 years of age.

Past middle age, CVD becomes quite common in the general population, although obviously, FH patients are also at progressively increased risk for developing CVD at older ages. In homozygous FH patients, CVD can occur before the age of 10 years. It’s really very troubling to see a young child have a heart attack, for example. Thankfully, with improved diagnosis and treatment, we’re now seeing homozygous FH patients getting well into their teens and even sometimes into adulthood before they have their first cardiovascular event.

DR. UNDERBERG: So, how would you diagnose the condition?

DR. BRINTON: That’s a great question. The diagnosis usually should be made clinically, in my opinion. In this modern era, it is tempting to jump right ahead to genetic testing, and indeed, genetic testing can help make or confirm a diagnosis of FH; however, in a significant percentage of patients who truly have FH, the genetic abnormality cannot be determined. More importantly, treatment must be guided by the lipid levels, not the genotype; therefore, lipid testing is much more important in patient management.

Screening can start with determining just the total cholesterol level in a non-fasting specimen. If it’s above 200 mg/dL, then FH may be present, and a fasting lipid panel with an LDL-C level is needed. In children or adolescents, if the LDL-C level is above 160 mg/dL or the non-high density lipoprotein (HDL), which is total cholesterol minus HDL cholesterol (HDL-C), is above 190 mg/dL, then FH is strongly suspected.

In adults over the age of 20 years, the cutoffs are higher because the LDL-C level increases with age, and FH is suspected if the LDL-C level is above 190 mg/dL or if the non-HDL-C level is above 220 mg/dL. FH is very likely at higher levels: for individuals aged below 20 years, this level is an LDL-C of above 190 mg/dL; for those aged 20 to 30 years, it’s an LDL-C level of above 220 mg/dL; and for those aged 30 years or older, the LDL-C threshold is 250 mg/dL.

DR. UNDERBERG: Are those numbers are from the Make Early Diagnosis to Prevent Early Deaths (MEDPED) criteria?²

DR. BRINTON: Yes, and they are also from the NLA Expert Panel statement,³ which, as you mentioned, came out in the middle of 2011.

DR. UNDERBERG: That’s US-based. Are there any other criteria that are used globally?

DR. BRINTON: These are non-US criteria, but they’re fairly similar, so I think we should focus on the US-based guidelines. Screening and diagnostic cutoffs are, of course, always a tradeoff between sensitivity and specificity. Lower cutoffs will have greater sensitivity but less specificity. Thus, you’re going to have a larger number of non-FH patients that you think might have FH. If you set the criteria or the cutoffs higher, then you’re much more likely to have a true FH case, but you will miss some patients with FH due to variability in LDL-C levels, even among patients with an identical FH mutation. These LDL-C variations can be due to variability in other genetic factors and environmental influences.

DR. UNDERBERG: Dr. McGowan, we’ve heard a little bit about diagnostic criteria. How are we doing right now with regard to diagnosis both here and globally? What options do we have to improve the diagnosis-making process with respect to screening, awareness, and other factors?

DR. MCGOWAN: Unfortunately, we’re doing fairly poorly, both in the United States and abroad. Roughly 20% of people with FH have received a diagnosis of FH. That doesn’t mean that people who have FH and have not yet been diagnosed are not being treated with lipid lowering agents: they may be. However, they are often not receiving an adequate dose of medication, and the failure to diagnose means a missed opportunity to educate patients and screen relatives; remember, this is an autosomal codominant disorder. Roughly half of the first-degree relatives of a person living with FH will also have FH.

General practitioners and even cardiologists often don’t think of FH when they have a young person with a coronary event admitted to their critical care unit. This is very unfortunate. It may occur in part because coronary disease is such a common disorder in the United States. When physicians miss the diagnosis of FH, they miss the opportunity to educate patients about cascade screening. This is the process I just referred to. Cascade screening involves screening all first-degree relatives of an index case and repeating the exercise in the first-degree relatives of the newly identified patients. As Dr. Brinton pointed out, FH is treatable, and the sooner we make the diagnosis, the more likely we are to prevent a cardiac event.

There are certainly some physical findings that we look for in patients with FH, but these are often quite subtle and are frequently missed. We can look for xanthomas in the Achilles tendon or the extensor tendons of the hands. Corneal arcus is not pathognomonic for FH, but when seen in the correct setting, it may help you make a diagnosis.

These physical findings are actually used in some of the FH-screening tools. Dr. Brinton mentioned the MEDPED, which was developed in Utah. Other screening tools include the Simon Broome criteria⁴ and the Dutch Lipid Clinic Network criteria.⁵ Both the Simon Broome and the Dutch Lipid Clinic Network criteria evaluate lipid levels in the context of physical findings and determine the probability that a person has FH.

DR. UNDERBERG: Does making the diagnosis of FH alter the way you would treat someone, especially with respect to how aggressively you would treat someone?

DR. MCGOWAN: This is a very important question. When we think about patients with FH, one of the things we know is that they’ve had elevated lipids since birth. In fact, if a person inherits the FH from his/her mother, he/she has been exposed to the mother’s very elevated lipids in utero and may have a greater cardiac risk than somebody who inherited the FH from his/her father. Having elevated lipids from birth is very different from gaining weight at the age of 40 years and developing hyperlipidemia in mid-life. The NLA has specifically pointed out that we should not be using the Framingham Risk Score in patients with FH because, in fact, they should all be considered very high risk. Anyone with FH should be treated very aggressively.

When we compare children with FH to their unaffected siblings, we see a significant difference in terms of carotid intima-media thickness by approximately 12 years of age.⁶ The vascular trajectory is very different in people with FH as compared to people who develop hyperlipidemia later in life. We should not be satisfied with achieving an LDL-C level of 190, 160, or even 130 mg/dL. We should try to reduce the LDL level as far as possible. Sometimes, this is very difficult because patients with FH don’t respond to lipid-lowering agents as well as patients with polygenic hyperlipidemia.

For example, homozygotes achieve about a 25% reduction in the LDL level at maximum doses of statins in combination with ezetimibe.⁷ Clearly, homozygous FH patients are at very high risk, and unfortunately, they don’t get the 50% to 60% reduction that we see when we treat other people with high doses of statins.⁸ Heterozygotes have a better response, but it’s still not the same response that you see in people with polygenic hyperlipidemia. Therefore, the answer to your question is yes, we should treat FH patients differently.

DR. BRINTON: I generally agree with what Dr. McGowan has said, but I am slightly more agnostic. In my view, there isn’t much clinical value in making a genetic diagnosis of FH, and here’s why: For everyone with suspected FH, we need to obtain the best-possible family history. If the person has a family history of CVD, especially before middle age and especially in more than 1 close relative, 2 things should occur: first, aggressive LDL lowering in the patient, and second, aggressive screening in family members. In cases where little or no family history can be obtained—for example, if the patient was adopted—then we can either do genetic testing or simply assume the presence of FH and proceed to aggressive treatment and screening of any blood relatives of the patient. Therefore, a positive family history and a very high LDL-C level will, in fact, lead us to the aggressive treatment that Dr. McGowan mentioned and to family screening as well.

A lipidologist may want to make the diagnosis of FH in a formal genetic manner, and I can think of some scenarios in which this might be useful. I believe that some of the difficulty that we’ve had in the broader physician community is that FH can seem like an arcane entity with many complicated criteria. I fear that, often, primary care doctors may give up in despair and leave patients undiagnosed and untreated, or undertreated, and do not refer them. I think we need to keep our messages about FH very simple—not only for patients but also for general physicians.

Even though I’ve spent much of my career working in genetically oriented academic institutions, I feel that we and our patients are best served by focusing on the phenotype, or lipid levels, rather than on the genotype in cases of severe hypercholesterolemia. By definition, the levels in these cases are far above the LDL-C cutoffs, and we need to be prepared to give such patients high doses of more than one LDL-lowering medication, as Dr. McGowan has said. Given limited resources, I don’t think we should be worrying so much about the finer details of how to diagnose FH. Instead, we should put that effort toward aggressive screening and treatment.

DR. UNDERBERG: Dr. McGowan, there are some criteria elucidated by the NLA that might create subsets of FH patients that we should target for LDL-C lowering. Do you want to point some of these out?

In 2011, the National Lipid Association (NLA) released clinical guidance and expert panel recommendations regarding the screening and treatment of patients with FH.¹ Two US advocacy groups have emerged that promote awareness and screening for this condition, and several new LDL-lowering drugs are in different phases of development. These agents have and will be evaluated in the FH population.

Recent observational data remind us that medication is the current cornerstone therapy for FH. Lipid-lowering therapies reduce the risk of CVD in these patients as compared with those who are not treated. There remain, however, many patients who cannot tolerate therapy or who, despite maximal pharmacologic treatment, do not achieve the recommended LDL targets. For these patients, LDL apheresis is an option available at approximately 35 centers throughout the United States. Finally, with new national cholesterol guidelines underway, there is renewed interest and awareness regarding the identification and treatment of patients with lipid disorders.

I’m Dr. James Underberg, Clinical Assistant Professor of Medicine at the NYU School of Medicine, NYU Center for Cardiovascular Disease. I am also Director of the lipid clinic at Bellevue Hospital in New York. I’m joined today by 3 internationally recognized experts in the field of hypercholesterolemia: Dr. Eliot Brinton, President of the Utah Lipid Center and Director of Atherometabolic Research at the Utah Foundation for Biomedical Research; Dr. Patrick Moriarty of the University of Kansas Medical Center; and Dr. Mary McGowan, Chief Medical Officer of the FH Foundation.

Dr. Brinton, can you briefly review the epidemiology, associated cardiovascular risk, and different diagnostic criteria for FH?

DR. BRINTON: This is among the most commonly occurring, clinically significant, monogenic metabolic disorders. It is codominant, so the clinical picture differs among homozygotes, heterozygotes, and the unaffected.

Heterozygous FH occurs in approximately 1 of 300 to 1 of 500 persons in the general population, although certain populations have a much higher prevalence. In groups such as the French Canadians and Dutch Afrikaners, 1 in 100 individuals are heterozygotes due to a founder effect. FH homozygotes are less common—about 1 in a million—and it is estimated that up to 10 000 homozygous FH patients are present worldwide.

One of the biggest clinical problems that we face with FH is detecting it or finding affected patients in the general population. The disease is treatable, but it is important to start treatment as early as possible. It has been very hard to identify such patients because they tend to be scattered throughout the population. Many of them may have never had even a single lipid panel, much less been referred to a lipidologist for proper care of their cholesterol disorder.

Another enormous challenge that we face in dealing with FH is that the patients have very, very high LDL-C levels—usually in the range of 250 to 500 mg/dL—in heterozygotes and usually above 500 mg/dL in homozygotes. These are untreated levels, and of course, once treatment is started, the levels fall. The underlying cause of FH in most cases is a defect in the LDL receptor.

The action of the LDL receptor in the liver is the major mechanism responsible for clearing LDL particles from the bloodstream, and if one has a defective receptor such that it is either never synthesized or has little or no capacity to bind to the LDL particle, then the LDL levels greatly increase in the blood. Of course, if someone is heterozygous, he/she will have one functioning receptor and one missing or malfunctioning receptor. By definition, homozygotes have defects in both copies of the LDL receptor gene; therefore, they have very little, if any, LDL receptor activity. This is classic FH, but there are other causes of severe hypercholesterolemia too. Although they are genetically and causally distinct, the increase in LDL-C levels may be similar to that in classic FH. When 2 distinct mutations cause the same phenotype, they are said to be phenocopies of each other.

For example, approximately 10% of patients with FH have a defect in apolipoprotein B (apoB), which is the major ligand for the LDL receptor, rather than a defect in the LDL receptor itself. A much smaller number of patients appear to have a defect in proprotein convertase subtilisin/kexin type 9 (PCSK9), which is a factor that helps process the LDL receptor. Because this factor functions to reduce the activity of LDL receptors, it is a rare gain-of-function mutation that reduces LDL receptor activity, thereby causing an increase in LDL-C levels.

Interestingly, even though a defect in the LDL receptor is by far the most common cause of FH, these defects are very heterogeneous in nature; thousands of individual mutations of the LDL receptor have been discovered in genotyping studies of FH patients. However, genetically isolated populations often have a founder effect with more uniform mutations as well as a higher prevalence of FH. Generally, the most striking aspect of FH is that even though the phenotype tends to be quite similar from genotype to genotype, there is incredible genetic heterogeneity.

FH was one of the very first monogenic disorders with important clinical sequelae ever discovered. These patients tend to have premature CVD, which is CVD occurring in men younger than 45 years and women younger than 55 years of age. CVD is quite uncommon among people in these age groups in the general population. The risk ratio of CVD for people with to those without heterozygous FH is very high at these younger ages, and we generally see a 5- to 20-fold increase, even though, in absolute terms, there are few events. The risk ratios and absolute rates of CVD are much higher in FH homozygotes, especially in those younger than 25 years of age.

Past middle age, CVD becomes quite common in the general population, although obviously, FH patients are also at progressively increased risk for developing CVD at older ages. In homozygous FH patients, CVD can occur before the age of 10 years. It’s really very troubling to see a young child have a heart attack, for example. Thankfully, with improved diagnosis and treatment, we’re now seeing homozygous FH patients getting well into their teens and even sometimes into adulthood before they have their first cardiovascular event.

DR. UNDERBERG: So, how would you diagnose the condition?

DR. BRINTON: That’s a great question. The diagnosis usually should be made clinically, in my opinion. In this modern era, it is tempting to jump right ahead to genetic testing, and indeed, genetic testing can help make or confirm a diagnosis of FH; however, in a significant percentage of patients who truly have FH, the genetic abnormality cannot be determined. More importantly, treatment must be guided by the lipid levels, not the genotype; therefore, lipid testing is much more important in patient management.

Screening can start with determining just the total cholesterol level in a non-fasting specimen. If it’s above 200 mg/dL, then FH may be present, and a fasting lipid panel with an LDL-C level is needed. In children or adolescents, if the LDL-C level is above 160 mg/dL or the non-high density lipoprotein (HDL), which is total cholesterol minus HDL cholesterol (HDL-C), is above 190 mg/dL, then FH is strongly suspected.

In adults over the age of 20 years, the cutoffs are higher because the LDL-C level increases with age, and FH is suspected if the LDL-C level is above 190 mg/dL or if the non-HDL-C level is above 220 mg/dL. FH is very likely at higher levels: for individuals aged below 20 years, this level is an LDL-C of above 190 mg/dL; for those aged 20 to 30 years, it’s an LDL-C level of above 220 mg/dL; and for those aged 30 years or older, the LDL-C threshold is 250 mg/dL.

DR. UNDERBERG: Are those numbers are from the Make Early Diagnosis to Prevent Early Deaths (MEDPED) criteria?²

DR. BRINTON: Yes, and they are also from the NLA Expert Panel statement,³ which, as you mentioned, came out in the middle of 2011.

DR. UNDERBERG: That’s US-based. Are there any other criteria that are used globally?

DR. BRINTON: These are non-US criteria, but they’re fairly similar, so I think we should focus on the US-based guidelines. Screening and diagnostic cutoffs are, of course, always a tradeoff between sensitivity and specificity. Lower cutoffs will have greater sensitivity but less specificity. Thus, you’re going to have a larger number of non-FH patients that you think might have FH. If you set the criteria or the cutoffs higher, then you’re much more likely to have a true FH case, but you will miss some patients with FH due to variability in LDL-C levels, even among patients with an identical FH mutation. These LDL-C variations can be due to variability in other genetic factors and environmental influences.

DR. UNDERBERG: Dr. McGowan, we’ve heard a little bit about diagnostic criteria. How are we doing right now with regard to diagnosis both here and globally? What options do we have to improve the diagnosis-making process with respect to screening, awareness, and other factors?

DR. MCGOWAN: Unfortunately, we’re doing fairly poorly, both in the United States and abroad. Roughly 20% of people with FH have received a diagnosis of FH. That doesn’t mean that people who have FH and have not yet been diagnosed are not being treated with lipid lowering agents: they may be. However, they are often not receiving an adequate dose of medication, and the failure to diagnose means a missed opportunity to educate patients and screen relatives; remember, this is an autosomal codominant disorder. Roughly half of the first-degree relatives of a person living with FH will also have FH.

General practitioners and even cardiologists often don’t think of FH when they have a young person with a coronary event admitted to their critical care unit. This is very unfortunate. It may occur in part because coronary disease is such a common disorder in the United States. When physicians miss the diagnosis of FH, they miss the opportunity to educate patients about cascade screening. This is the process I just referred to. Cascade screening involves screening all first-degree relatives of an index case and repeating the exercise in the first-degree relatives of the newly identified patients. As Dr. Brinton pointed out, FH is treatable, and the sooner we make the diagnosis, the more likely we are to prevent a cardiac event.

There are certainly some physical findings that we look for in patients with FH, but these are often quite subtle and are frequently missed. We can look for xanthomas in the Achilles tendon or the extensor tendons of the hands. Corneal arcus is not pathognomonic for FH, but when seen in the correct setting, it may help you make a diagnosis.

These physical findings are actually used in some of the FH-screening tools. Dr. Brinton mentioned the MEDPED, which was developed in Utah. Other screening tools include the Simon Broome criteria⁴ and the Dutch Lipid Clinic Network criteria.⁵ Both the Simon Broome and the Dutch Lipid Clinic Network criteria evaluate lipid levels in the context of physical findings and determine the probability that a person has FH.

DR. UNDERBERG: Does making the diagnosis of FH alter the way you would treat someone, especially with respect to how aggressively you would treat someone?

DR. MCGOWAN: This is a very important question. When we think about patients with FH, one of the things we know is that they’ve had elevated lipids since birth. In fact, if a person inherits the FH from his/her mother, he/she has been exposed to the mother’s very elevated lipids in utero and may have a greater cardiac risk than somebody who inherited the FH from his/her father. Having elevated lipids from birth is very different from gaining weight at the age of 40 years and developing hyperlipidemia in mid-life. The NLA has specifically pointed out that we should not be using the Framingham Risk Score in patients with FH because, in fact, they should all be considered very high risk. Anyone with FH should be treated very aggressively.

When we compare children with FH to their unaffected siblings, we see a significant difference in terms of carotid intima-media thickness by approximately 12 years of age.⁶ The vascular trajectory is very different in people with FH as compared to people who develop hyperlipidemia later in life. We should not be satisfied with achieving an LDL-C level of 190, 160, or even 130 mg/dL. We should try to reduce the LDL level as far as possible. Sometimes, this is very difficult because patients with FH don’t respond to lipid-lowering agents as well as patients with polygenic hyperlipidemia.

For example, homozygotes achieve about a 25% reduction in the LDL level at maximum doses of statins in combination with ezetimibe.⁷ Clearly, homozygous FH patients are at very high risk, and unfortunately, they don’t get the 50% to 60% reduction that we see when we treat other people with high doses of statins.⁸ Heterozygotes have a better response, but it’s still not the same response that you see in people with polygenic hyperlipidemia. Therefore, the answer to your question is yes, we should treat FH patients differently.

DR. BRINTON: I generally agree with what Dr. McGowan has said, but I am slightly more agnostic. In my view, there isn’t much clinical value in making a genetic diagnosis of FH, and here’s why: For everyone with suspected FH, we need to obtain the best-possible family history. If the person has a family history of CVD, especially before middle age and especially in more than 1 close relative, 2 things should occur: first, aggressive LDL lowering in the patient, and second, aggressive screening in family members. In cases where little or no family history can be obtained—for example, if the patient was adopted—then we can either do genetic testing or simply assume the presence of FH and proceed to aggressive treatment and screening of any blood relatives of the patient. Therefore, a positive family history and a very high LDL-C level will, in fact, lead us to the aggressive treatment that Dr. McGowan mentioned and to family screening as well.

A lipidologist may want to make the diagnosis of FH in a formal genetic manner, and I can think of some scenarios in which this might be useful. I believe that some of the difficulty that we’ve had in the broader physician community is that FH can seem like an arcane entity with many complicated criteria. I fear that, often, primary care doctors may give up in despair and leave patients undiagnosed and untreated, or undertreated, and do not refer them. I think we need to keep our messages about FH very simple—not only for patients but also for general physicians.

Even though I’ve spent much of my career working in genetically oriented academic institutions, I feel that we and our patients are best served by focusing on the phenotype, or lipid levels, rather than on the genotype in cases of severe hypercholesterolemia. By definition, the levels in these cases are far above the LDL-C cutoffs, and we need to be prepared to give such patients high doses of more than one LDL-lowering medication, as Dr. McGowan has said. Given limited resources, I don’t think we should be worrying so much about the finer details of how to diagnose FH. Instead, we should put that effort toward aggressive screening and treatment.

DR. UNDERBERG: Dr. McGowan, there are some criteria elucidated by the NLA that might create subsets of FH patients that we should target for LDL-C lowering. Do you want to point some of these out?