CME Credit

The A. Webb Roberts Center for Continuing Medical Education (CME) of Baylor Health Care System, Dallas, Texas is accredited by the Accreditation Council for Continuing Medical Education to provide continuing medical education for physicians.

The A. Webb Roberts Center for Continuing Medical Education of Baylor Health Care System designates this educational activity for a maximum of 1 AMA PRA Category 1 Credit . Physicians should only claim credit commensurate with the extent of their participation in the activity.

CME Provider Privacy Policy and Contact Information

The A. Webb Roberts Center for Continuing Medical Education of Baylor Health Care System (214-820-2317) observes the privacy and confidentially of CME information and the personal information of CME participants.

Third parties receive only aggregated data about CME activities that are relevant to their interests and/or the activities they support.

CME Instructions

Physicians may receive CME credit for this activity. After reading this report, go online to http://www.ajconline.org , find the link for the CME register on the right-hand bottom side of the screen, complete a post-test with a minimum score of 80%, complete an evaluation, and print a certificate.

Computer Requirements

Windows 2000, Pentium 3 or greater, 512 RAM, 80-gigabyte storage

Combination of Media: Print and Internet.

Estimated Time to Complete: 1 hour.

Release Date: January, 2011.

Termination Date: January, 2012.

Cardiovascular risk factors: are they all “created” equal?

Dr. Viggiani: I’d like to begin our discussion by reviewing the established risk factors for CHD. Data from the Framingham Heart Study and other studies have identified elevated cholesterol, high blood pressure, smoking, diabetes mellitus, overweight, and physical inactivity as risk factors for CVD events. The effects of such things as age, gender, and triglyceride and HDL levels on CVD events have also been evaluated in the Framingham study. Dr. Roberts, do you think all of these risk factors are equally contributing to the genesis of CVD events?

Dr. Roberts: The Framingham studies have tended to add up the CVD risk factors, and the more factors a patient has, the greater their risk. I think that has diffused the real issue. I think that cholesterol is not “a” risk factor. Rather, I think that cholesterol is “the” cause of atherosclerosis, and other factors such as systemic hypertension, elevated glucose, diabetes mellitus, inactivity, aging, and overweight worsen the risk of an elevated LDL cholesterol or non-HDL cholesterol. There are 4 supporting pieces of evidence that cholesterol is the cause of atherosclerosis: (1) atherosclerosis can be produced in herbivores by giving cholesterol or saturated fat; (2) cholesterol is present in atherosclerotic plaques; (3) epidemiologic studies (Seven Country Study, for example) comparing people with high levels of cholesterol to those with low levels clearly show that populations with high cholesterol levels have a much higher frequency of developing atherosclerotic events and a much higher frequency of dying from those events; the quantity of plaque in the arteries in people with elevated levels of cholesterol is much greater than in those with lower levels of cholesterol; (4) treatment studies, particularly the statin studies, show that when cholesterol levels are lowered, atherosclerotic events decrease, the chance of dying from these events decrease, and the quantity of plaque decreases. Atherosclerosis cannot be produced in nonhuman herbivores by making them hypertensive, or by blowing cigarette smoke in their faces over the course of their lifetime, or by raising their blood glucose levels. The only way to produce atherosclerosis experimentally in herbivores (for example, rabbits, monkeys) is by giving cholesterol (egg yolks) or saturated fat.

Dr. Viggiani: Does everyone agree with this concept?

Dr. Gotto: I don’t entirely agree. I would agree that there likely is a substrate of cholesterol and LDL that’s necessary to produce atherosclerotic plaque, but that may be proved or disproved if we get LDL levels down low enough. We haven’t gotten LDLs down low enough at this point to be able to say how low can you go and totally eliminate atherosclerosis, which conforms to Dr. Roberts’ premise. We know there’s a relation between atherosclerosis, coronary risk, and cholesterol and LDL levels; it is a log-linear relation. That means the curve slopes off as LDL levels decrease. At some point on the curve we’ll get to a point where a further decrease of LDL would not give further benefit, or at least it would take so long to get the benefit that you won’t see it in the lifetime of a clinical trial. I agree that a certain amount of cholesterol and LDL is necessary for atherosclerosis but I’m not sure that’s equivalent to saying that cholesterol is “the” cause of atherosclerosis. Other risk factors occur as well. You see diabetics, hypertensives, and patients who smoke developing atherosclerosis at much lower levels of LDL than people who don’t have the other risk factors. I think it is misleading to say cholesterol and LDL are “the” cause of atherosclerosis.

Dr. LaRosa: I think it’s probably true that if the LDL level is lowered down to the level of herbivores, or of very primitive populations, that atherosclerosis will not occur. Of course, hypertension is also uncommon in these populations who do not gain excess weight or eat a lot of simple carbohydrates. From a practical point of view, I tend to agree with Bill Roberts, that you have to have substrate. It probably doesn’t take very much LDL above the levels seen in herbivores, which is around 50 to 60 mg/dl. If you get much above that, you start to develop enough substrate to develop atherosclerosis, and then all these other risk factors come into play. If you get the LDL down around 50 mg/dl, you probably don’t have much atherogenesis occurring. Or if you do have it, it’s going on at a very low level. In the TNT [Treating to New Targets] study, we actually looked at quintiles of LDL levels; the lowest quintile was <64 mg/dl. Even at those levels, if you look at HDL or you look at blood pressure, there is a difference in risk between those who have low HDLs and high HDLs. That is, people with lower HDLs still have a higher risk. People with higher blood pressures still have a higher risk. From a practical point of view, it doesn’t really matter because we live in a society where most people’s LDLs are not 50 or 60 mg/dl. In Western societies, these other risk factors aren’t really as important.

Dr. Guyton: Clinicians are most familiar with the NCEP [National Cholesterol Education Program] paradigm, which views cholesterol, specifically LDL and non-HDL cholesterol, as factors to treat. There are several major risk factors that help clinicians make a decision on whether to lower LDL cholesterol: hypertension, diabetes mellitus, smoking, family history of atherosclerotic events, low HDL, and then age and gender. A couple of recent studies have focused attention on cholesterol. One is the recent data on PCSK9 [proprotein convertase subtilisin/kexin type 9], a serine protease that inactivates LDL receptors in the liver. Data from the Atherosclerosis Risk in Communities (ARIC) study showed that small decreases in LDL cholesterol due to PCSK9 loss of function mutations were associated with markedly lower CHD incidence. For example, a mutation giving a 15% decrease of LDL cholesterol levels was associated with a 44% lower CHD incidence.

This was attributed to the fact that genetic decrease of LDL is present from birth, unlike lowering of LDL by medication over only 4 or 5 years in clinical trials. The other point that I want to make helps to explain why treatment of LDL in particular and possibly treatment of HDL are so important in patients who have risk based on other factors, whether it’s a history of smoking or hypertension or diabetes. Improving cholesterol levels is more of an opportunity than lowering blood pressure levels or treating diabetes. The blood glucose level cannot be lowered to 1/2 the average level in the population. Blood pressure cannot be lowered to a level that’s 1/2 the average level in the population. The LDL cholesterol level can be lowered to a level that’s 1/2 the average level in the American population. That fact makes cholesterol an opportunity factor, particularly in patients with diabetes, where the risk of atherosclerosis is so high.

In diabetes, the risk of atherosclerosis is augmented tremendously, probably by glycosylation reactions or other factors distinct from the specific changes in the lipids. We know there’s a diabetic dyslipidemia but that doesn’t even come close to explaining why diabetics have so much risk. Nevertheless, although glucose would be suspected to be the mediator of that augmented risk in diabetes, the opportunity to prevent coronary events is still with LDL cholesterol, by making LDL cholesterol levels much lower than the average population level. We need to think about opportunity as well as risk.

Dr. Roberts: Do you think if you could get the entire population to have serum LDL levels <50 mg/dl that you would not have to worry about atherosclerosis, even in people who smoke cigarettes or had hypertension?

Dr. Guyton: Probably not.

Dr. Roberts: So would you agree that systemic hypertension without an LDL >50 mg/dl does not cause plaques?

Dr. Guyton: Yes, I agree with that.

Dr. Roberts: Do you think cigarette smoking in people with LDLs <50 mg/dl causes plaques?

Dr. Guyton: No. If the LDL is <50 mg/dl, atherosclerosis rarely occurs.

Dr. Roberts: OK. You also agree that lipids cause atherosclerosis?

Dr. Guyton: I agree.

Dr. Roberts: If one goes down the list of risk factors and asks the question, “Do you have to have this factor to have atherosclerosis?,” you eliminate every 1 of them except lipids.

Dr. LaRosa: I’m inclined to agree with you, but I don’t think we have absolute evidence for that. I’ve always believed that what you’re saying makes a lot of sense. You don’t see atherosclerosis in populations where LDLs are very, very low, perhaps because the people don’t live long enough or because they have so many other health problems. And they rarely become obese, so there are other things in their favor. But I’m inclined to agree with you, although it doesn’t mean that we’re going to live in a society like that any time soon. We live in a society where there’s sufficient substrate for the rest of these things to come into play and accelerate the process.

Dr. Guyton: The educational message for physicians and the public is that by the time we develop arterial plaques, the treatment strategy usually has to involve more than just pushing the LDL down. Even primary prevention is actually treatment of atherosclerosis by the time we intervene. In primary prevention we’re preventing heart attacks, but we’re also treating atherosclerosis, and atherosclerosis that often is fairly developed by that time. Cholesterol is already in the arterial wall.

Dr. Roberts: It is possible to prevent plaques if we get the LDL low enough. Wouldn’t you agree?

Dr. Guyton: Yes. We are close to seeing that. Results from The Measuring Effects on Intima Media Thickness: an Evaluation Of Rosuvastatin trial, which compared rosuvastatin 40 mg to placebo, showed that active treatment completely stabilized the carotid IMT. Half the people in that trial were still growing plaques and 1/2 of them were regressing plaques. Therefore, we cannot be certain of avoiding new plaque, even with the strongest, most efficacious statin treatment available.

Dr. Gotto: The Acute Stroke Accurate Prediction (ASAP) carotid trial, published in 2001, compared high-dose atorvastatin [80 mg] to simvastatin [40 mg]. There was still progression of carotid plaques. There was some regression, but that occurred only with the high-dose arm. The study implied, however, that aggressive enough treatment might totally prevent the further progression of plaque.

Dr. Guyton: That trial involved only patients with familial hypercholesterolemia, so they began with quite high LDL levels.

Dr. Roberts: Eighty milligrams of atorvastatin doesn’t lower many people’s LDLs to <50 mg/dl.

Dr. Gotto: It depends on what the baseline is. Jerome Cohen reported that the prevalence of abnormal LDL decreased from 43.5% in NHANES II to 36.3% in NHANES 1999 to 2006. In the Action to Control Cardiovascular Risk in Diabetes trial, the mean LDL was 105 mg/100 ml and in the JUPITER [Justification for Use of Statins in Prevention: An Intervention Trial Evaluating Rosuvastatin] trial the median LDL was 108 mg/dl at baseline. These data show that LDL levels are becoming lower.

Dr. LaRosa: The stunning thing about JUPITER is that the mean LDL level was lowered to an average of 55 mg/dl. Admittedly, that means 50% of patients were above and 50% were below that level.

Dr. Gotto: The event rate turned out to be 2 times as low as we were expecting; that’s why the study was stopped at 1.9 years.

Dr. LaRosa: To have a mortality benefit that was measurable in 2 years says something. It certainly broadly hints that when the LDL level gets that low, atherogenesis is not occurring. Admittedly, that’s only indirect evidence. Nevertheless, these results are astounding.

Dr. Gotto: The JUPITER results show what can be done with primary prevention in people who we previously thought were at so-called low risk.

Dr. Roberts: The JUPITER trial in part was based on your TexCAPS [Texas Coronary Atherosclerosis Prevention Study] trial?

Dr. Gotto: Yes, in part it was because we found when we did a subgroup post hoc analysis with Paul Ridker that if an LDL or a CRP was above the median or the mean, 1 benefited from treatment with lovastatin, and if the 2 were below the mean, there was no benefit. But, again, that was a subgroup post hoc analysis.

Dr. Roberts: Was the high-sensitivity [hs] CRP in your trial considered normal if it was <2 or <1 mg/L?

Dr. Gotto: It may not have been exactly 2, because we took above and below the mean or median. We analyzed it both ways.

Dr. Roberts: John, isn’t an hs-CRP <1 mg/L generally considered normal?

Dr. Guyton: An AHA/CDC [American Heart Association/Centers for Disease Control ad Prevention] consensus conference in 2003 divided it more or less into population tertiles: 0 to 1 mg/L, 1 to 3 mg/L, and >3 mg/L, which correspond to low, medium, and high risk, respectively.

Dr. Gotto: The hs-CRP varies with populations. It differs in different parts of the world. There is no universal normal level. We selected 2 mg/L as a cutoff for AFCAPS [Air Force Coronary Atherosclerosis Prevention Study]/TexCAPS, but the PROVE-IT [Pravastatin or Atorvastatin Evaluation and Infection Therapy] post hoc subgroup analysis showed that people with the lowest rate of recurrent events were those who had an LDL <70 mg/dl and a CRP <1 mg/L.

Dr. Viggiani: Dr. Gotto, increased CRP has emerged as a potential new indicator of cardiovascular risk. How should the CRP fit into clinical practice? Should we use it as a screening tool?

Dr. Gotto: CRP is made in the liver. It is stimulated by interleukin-6 and it’s a general measurement of inflammation. Some believe that it’s actually involved in the atherosclerotic process itself. No receptors have been identified for CRP. Others think it’s just a secondary indication of inflammation. CRP may also be higher in patients with higher levels of LDL and higher levels of oxidized LDL. A number of studies carried out by Ridker and colleagues suggested that hs-CRP is a strong and independent predictor of coronary risk, and that LDL and CRP together are the strongest predictors. LDL and CRP are lowered by statins, and they appear to be lowered independently of each other. Individuals who have the most LDL decrease with a statin also have the most CRP decrease, but for a given individual they’re not directly linked.

The current recommendation of the AHA/CDC is to use hs-CRP in patients with intermediate risk—that is a 10-year risk for a coronary event of >10% but <20%—to make a decision about whether to commit the patient to long-term statin therapy. Based on the JUPITER study I recommend that patients undergoing a cardiovascular risk assessment—including measurements of lipids, blood pressure, and questioning about smoking and diet—should have an hs-CRP measured.

Dr. LaRosa: I agree. I think that JUPITER would have been a stronger study if we’d had a control group with normal CRPs, or what we consider to be normal. I think we’re going to get into the same problem with CRP as we did with LDL, when we talk about normal and abnormal, as if there were some point below which everything is OK and you didn’t have to worry about it. There is very good circumstantial evidence that that those patients with high CRPs are a group that’s at higher risk of CVD events than were anticipated. It would be nicer if we had been able to look at how a group of people with CRPs <1 mg/L performed versus a group of people with CRPs >2 mg/L.

Dr. Guyton: Would you say that AFCAPS/TexCAPS data are actually a better test of that hypothesis?

Dr. LaRosa: Yes, except that was not the primary hypothesis of those studies. I think there must be other things that make patients more or less susceptible to coronary events. At a given cholesterol level, we know that some people are more likely to have an event than other people. And maybe CRP is a measurement of some sort of vascular susceptibility to atherogenesis.

When we do clinical trials we’re generally looking at 5-year exposures. Thus far, in the major observational, epidemiologic studies, we’ve been able to gather enough data to look at 10-year exposures. What we really want to do is look at lifetime exposures. A 25-year-old man with familial hypercholesterolemia may be at fairly low risk of developing an event in the next 10 years, but his risk is extraordinarily high for developing an event over his lifetime. As yet, Framingham has not been able to accumulate enough data to give us lifetime risk numbers. I expect that data relatively soon.

Dr. Guyton: A number of people are looking at lifetime risk in trying to develop that concept, including Donald Lloyd-Jones, Peter Wilson, and others.

Dr. LaRosa: That’s an important distinction for practitioners. We’re almost surely underestimating lifetime risk in clinical trials and in observation studies.

Dr. Guyton: I don’t think it’s appropriate to measure CRP across the board. I’m not ready to take all the men >50 years old and all the women >60 years old and measure their CRP and put them on a statin if their CRP is high. CRP is an extraordinarily strong risk factor and it’s comparable to, and probably greater than, the total cholesterol:HDL ratio, which is extraordinarily strong. We might even say that as a single blood test applied to a population, CRP beats them all. It is a little more powerful than LDL or HDL alone, and lipid parameters are only comparable when you start combining them in ratios.

One problem of applying CRP measurements to an individual is that you have the variability of CRP during an acute-phase response. If someone has an infection, the CRP increases. Even in the absence of a classic acute-phase response, CRP is quite variable, so it’s hard to know whether you’re properly classifying people on the basis of a single determination of CRP. If the CRP is substantially >3 mg/L, and we retest and confirm that it was not simply an acute-phase response, we’ve got a patient who’s got a problem. That’s the way I’ve been using it in our lipid clinic. CRP also correlates very highly with obesity, and weight loss lowers CRP. I use it sometimes as a marker of how well patients are reducing what I’m beginning to call their “active fat deposits,” as we begin to see CRP levels decrease before the patient gets down to normal weight. One woman in my practice, who had a gastric bypass operation and subsequently lost 80 lb, decreased her CRP from approximately 20 to 0.7 mg/L. We had multiple measurements before and after to show that this was really a true decrease in the CRP.

Dr. Roberts: In the JUPITER trial, CRP levels in the rosuvastatin arm went from a median of 4.2 mg/L at baseline down to 2.2 at 24 months. Do you think that played a major role in the enormous decrease in events in that trial, or do you think you can reasonably attribute the enormous decrease in events to the lowering of the LDL alone?

Dr. Guyton: I think we can definitively say that lowering LDL works, but whether lowering LDL explains all the decreases is difficult to say. I think it’s even more difficult to say that truly lowering the CRP was the factor that really linked with the reduction of events.

Dr. LaRosa: I think CRP, whatever its long-term future, may be helpful in the short term by demonstrating the value of so-called primary prevention or what others call “earlier treatment.” There is a lot of debate about the cost of treating large numbers of people, and because of this some rationalize that we should treat only people who already have established disease. That’s kind of closing the barn door after the horse has left. But then the question becomes should one start intervening in a population of people before they’ve had clinical evidence of coronary disease? Should we put everybody >18 year of age on statins? We’ve all seen a dramatic decrease in coronary bypass surgery and angioplasty over the previous 5 years or so. There’s every reason to believe that this trend is related to the fact that statins are being more widely used. If we now start to treat people who have increased CRPs, rather than because they have coronary disease, and find other ways of selecting people who are at higher risk but who don’t yet have clinical disease, we may see that trend accelerate.

Other emerging assessment tools

Dr. Viggiani: What other assessment tools are emerging that may help further identify patients at higher risk for cardiovascular events?

Dr. Guyton: We are beginning to use increasingly the imaging tests that quantify coronary calcium and carotid IMT. There is considerable debate about the economic value of these procedures, and insurance companies usually don’t pay for them. And yet, coronary calcium quantification has been shown to add appreciably to the Framingham risk calculation. I find coronary calcium scanning most useful in patients who might have high triglycerides, fairly low LDL cholesterol, a low HDL, and I’m not sure what the atherogenic environment is in that patient. Calcium in coronary arteries correlates quantitatively very well with the total burden of coronary atherosclerosis. It’s not a test for diagnosing chest pain and yet it’s highly predictive of future coronary death. Coronary calcium scoring is not the same thing as computed tomographic angiography. Many people confuse these 2 tests. Computed tomographic angiography is basically a heart catheterization without the catheter using intravenous contrast material and using the computer to visualize it in the coronary arteries by computed tomography. Coronary calcium scoring does not use any contrast material, but looks for calcium deposits that occur in the atherosclerotic lesions.

Dr. LaRosa: An important issue about these specific tests is their cost, and whether or not they are economically practical to implement. That’s why I think CRP is attractive because it’s inexpensive and we can get some of the same information that these imaging tests provide.

Treatment options for dyslipidemia

Dr. Viggiani: Let’s turn to specific treatment options for patients with dyslipidemia. Many clinicians initially start with so-called therapeutic lifestyle changes. What can be said about these interventions?

Dr. Guyton: Unfortunately, true evidence-based medicine may never be able to tell us what the effect of diet is because you’re not going to have a large, randomized, double-blind, controlled trial of a specific dietary intervention. Nevertheless, dietary changes can be extremely important. We need to look at data from the point of view of the patient and try to exclude all potential biases when interpreting the data.

Dr. Roberts: Do you think exercise prevents atherosclerotic plaques?

Dr. Guyton: That’s a tough question, but I think it probably does. Exercise modulates the atherosclerotic process in a favorable manner. There are better cardiovascular outcomes in the physically fit than in the unfit.

Dr. Gotto: Physically fit people have less obesity, a lower degree of insulin resistance, and lower levels of CRP than unfit persons. Exercise with weight control decreases blood pressure, increases HDL, and decreases triglycerides.

Dr. Roberts: And arteries enlarge with exercise such that it takes more plaque to narrow them.

Dr. Gotto: A decrease in risk of a coronary event parallels LDL decrease. Diet, bile acid resins, ileal bypass operations, and statins decrease approximately along the same straight line. Niacin and fibrates are off the line, a fact suggesting that the 2 lower risk by another mechanism. We don’t yet have data for ezetimibe. The Cholesterol Treatment Trialists have proposed that a decrease of LDL by 1 mmol/L gives a 20% decrease in events. The JUPITER study is consistent with these results, that is, about a 50% decrease in events for every 50 mg/dl decrease of LDL.

Dr. LaRosa: That percentage concerns relative risk decrease, not absolute risk decrease. Relative risk involves relatively fewer events prevented.

Dr. Gotto: The JUPITER data are only 2-year data. The event decrease might have been 75% if the data went out to 5 years. I realize this is a lot of extrapolation, but I’m trying to fit the data from JUPITER to the other curves that I mentioned earlier. The JUPITER data are consistent with the results of the other trials.

Dr. Roberts: The percent decrease in LDL corresponds with the percent decrease in events during that period?

Dr. Gotto: Yes.

Dr. Guyton: There are exceptions to that rule; for example, when the heart is already damaged. Statins do not really lower events very much, if at all, in patients with advanced heart failure. In the Simvastatin and Ezetimibe in Aortic Stenosis (SEAS) study, which was a study of patients with aortic stenosis, the ezetimibe/simvastatin combination lowered ischemic cardiovascular events by about 22%, so this would be an exception to what Bill was saying.

Dr. LaRosa: The amazing thing is after all of these data, the Ezetimibe and Simvastatin in Hypercholesterolemia Enhances Atherosclerosis Regression (ENHANCE) trial came out and immediately raised the question among the media of whether the LDL hypothesis was still a hypothesis. That trial evaluated carotid IMT. SEAS is the closest thing we have to evaluate the effect of ezetimibe on events, but unfortunately, that trial concerns patients with aortic stenosis and is not comparable to any other population reported in trials.

Dr. Guyton: We also have the subanalysis of the SANDS [Stop Atherosclerosis in Native Diabetics Study] study, which showed a regression of carotid IMT in patients treated to aggressive targets, and the regression was similar whether the targets were achieved with simvastatin alone or with ezetimibe/simvastatin combination.

Dr. Gotto: These ENHANCE patients had a normal IMT at baseline. The control group was on simvastatin.

Dr. Guyton: The ENHANCE study had been modeled after the ASAP, a study we discussed earlier. In the ASAP study the mean carotid IMT was approximately 0.92 mm. In the ENHANCE study it was 0.69 mm; that’s almost a normal thickness. Patients in ENHANCE and ASAP were selected in the same manner, by the same group, in the same country, namely Holland, and all patients had familial hypercholesterolemia. It was a surprise to see that patients in ENHANCE had an essentially normal baseline IMT, whereas those patients in the earlier ASAP study had an abnormal thickness. This finding perhaps speaks to the effectiveness of statin treatment that those people were receiving during the 6-year interval between the 2 trials; 80% of the people in ENHANCE had previously been on statins.

Dr. Roberts: But there was a significant difference in the treatment groups. ENHANCE was evaluating the effectiveness of ezetimibe as add-on therapy to simvastatin 80 mg, and I don’t think the trial went on long enough to show a difference between groups.

Dr. Gotto: It’s hard to improve on a normal artery, especially if the normal artery is being treated with simvastatin 80 mg.

Dr. LaRosa: Is there anybody who thinks that ENHANCE was an adequate test of ezetimibe?

Dr. Guyton: I think it was a reasonable attempt to test the hypothesis, but it turned out to have a null result, neither good nor bad, with a p value of 0.29, almost a 1/3 chance of getting that particular result. Let me bring up an old study: the University of California, San Francisco a Slone Center Office-Based Research Network study in people with familial hypercholesterolemia. That study demonstrated by quantitative angiography a regression of coronary atherosclerosis if patients were intensively treated. The intensive treatment of the cholesterol lowered the mean LDL to 172 mg/dl, and the atherosclerosis actually lessened. Those patients had not been previously treated with statins; they didn’t have statins for 2/3 of the study.

Dr. Roberts: In the ENHANCE study, the 2 treatment arms had baseline LDLs of about 318 mg/dl. The combination treatment arm lowered the mean LDL to 141 mg/dl, whereas the simvastatin only arm achieved a mean LDL of 193 mg/dl.

Dr. Guyton: Let me suggest, Bill, that it may not be the LDL in the plasma that’s driving the growth of plaques. Rather, it may be how much cholesterol the LDL has delivered to the lesion, and it’s that delivery that’s driving the growth of the plaque. Lesion progression or regression generally in these studies with anatomic end points demonstrates very tiny differences. In contrast, the vulnerability to plaque rupture is substantially changed, possibly by affecting the most sensitive part of the plaque—the shoulder region of the plaque—where the plaque becomes weak and ruptures. To see a difference in carotid IMT, the cholesterol content of the lesion probably needs to change substantially and 2 years is not enough time to change what the effect of treatment over a previous 6-year or 10-year period had been in those patients.

Dr. Roberts: But as Tony commented, one has to start with some abnormal IMT to decrease that thickness. You only go 1 way if you start with an essentially normal wall thickness. I understand that 3 measurements were done in the ENHANCE study: common carotid, the so-called bulb, and the proximal internal carotid. The common carotid artery infrequently gets much plaque in it. I see a lot of carotid endarterectomy specimens and most of the plaque is in the internal carotid artery. I would like to know if there was any difference in the 2 groups in the internal carotid artery without incorporating the other 2 measurements.

Dr. LaRosa: Let me pose this question. How should practicing clinicians interpret the data from ENHANCE and SEAS? If a clinician has a patient on ezetimibe/simvastatin, should they leave them on it?

Dr. Roberts: Absolutely.

Dr. Gotto: I feel the same way.

Dr. Guyton: Yes.

Dr. LaRosa: Yes. These data are inconclusive and don’t shake my faith in the fact that the LDL lowering is what is important.

Dr. Gotto: There are 2 ongoing studies looking at the combination of ezetimibe/simvastatin that should give us additional information. The Improved Reduction of Outcomes: Vytorin Efficacy International Trial study is an outcomes study that may have up to 18,000 participants. The SHARP [Study of Heart and Renal Protection] study, which includes patients with renal disease, has a placebo comparison.

Dr. Guyton: Yet ezetimibe does not have clinical outcome data at this point.

Dr. Roberts: Correct. Atorvastatin didn’t either for 5 years, and yet it became the best-selling statin within a few months.

Dr. Gotto: None of the statins had clinical outcome data when they were approved.

Dr. LaRosa: Do the ENHANCE results change the way you would use or think about carotid IMT as a screening test or as a way of following people with atherosclerosis?

Dr. Guyton: I’m doing carotid IMT measurements in a few patients who can afford it and in whom questions have arisen. I had a patient with moderately high cholesterol, but she had a very strong suggestion of tendon xanthomas. Her carotid IMT turned out to be 1.0 mm at 45 years of age, which is quite thick. I’m doing it much more as a baseline test. I use the test for patients with very low HDL cholesterol, which is not always atherogenic. These patients are usually young and do not have coronary calcium. Atherosclerotic progression can be determined with repeat testing 3 to 5 years after the baseline test. In an individual patient, there is little value in repeating carotid IMT before 3 years, because it is not a very precise technique.

Dr. LaRosa: Carotid IMT may correlate as a predictor of events, but the correlation between changes in thickness and changes in events is not well demonstrated.

Dr. Guyton: That’s true at the individual level, although there are pretty good data at the population level.

Dr. LaRosa: I’m specifically referring to clinical trial data. We had the opportunity to follow carotid IMT in subgroups of patients in several large trials, but unfortunately did not do it.

Dr. Guyton: The carotid IMT measurements have to be done in a very specialized laboratory. It takes 30 to 45 minutes minimum, so it’s not a quick screening procedure. The technician and the person reading the test have to be carefully trained.

Side effects of treatment

Dr. Viggiani: What about side effects of treatments? What are the most important side effects that we need to think about and how can we avoid or minimize them?

Dr. Gotto: The most significant side effects of statins are those related to muscle. Fortunately, the myopathy and rhabdomyolysis associated with increased creatine kinase levels >10 times the upper limit of normal are rare—about 1 to 0.1 in 1,000—although with cerivastatin, now off the market, there was much more muscle toxicity. The myopathy was discovered quite early on with lovastatin—in 2 patients who had heart transplants and were on lovastatin 80 mg plus high doses of cyclosporine. They were retested after their dialysis and after they had cleared the rhabdomyolysis. They were able to tolerate lovastatin 20 mg plus cyclosporine without difficulty. Because myopathy is a well-recognized side effect of statins, patients have to be warned about the risk of muscle weakness, pain, or tenderness. There is not much benefit to doing a baseline screening creatine kinase, but there are certain subgroups of patients in whom there’s increased susceptibility to this side effect: patients on multiple drugs, elderly patients, and underweight patients. It’s common to see patients on statins complaining of musculoskeletal symptoms and they are convinced that the statin is causing them. Data from the placebo-controlled studies, however, show no difference in this type of complaint between the placebo and the statin groups. There have been a couple of almost anecdotal studies reporting some mitochondrial changes, but the studies are not convincing.

Dr. Guyton: An observational study suggested that 5% to 19% of people taking the higher doses of statins will stop the statins because of myalgia or some other musculoskeletal complaint. We don’t know what the outcome would be if those patients were rechallenged with a placebo or the statin. We don’t know whether any significant percentage of those people actually have a biological effect on the muscles that could be revealed by a placebo-controlled trial. Tony is exactly right. There’s essentially a <1% difference in the percentage of people from the big randomized studies who complain of myalgia depending on whether they’re taking a statin or a placebo.

Dr. Roberts: This is 1 of the potential advantages of using the combination of simvastatin and ezetimibe, or essentially any statin and ezetimibe, because ezetimibe is a statin-sparing drug. One can use lower doses of the statin in combination with ezetimibe and get the same lowering effect on LDL as that achieved by tripling the statin dose alone. Nanette Wenger told me that before starting a patient on a statin, she questions the patient about potential myalgia symptoms: do you have any problems in your muscles, any pain or soreness, and so on? She writes their responses down and then starts them on the statin. Then, if they come back in with complaints of muscle soreness or pain, she reads them what she wrote down before putting them on the statin.

Dr. Guyton: Before putting somebody on a high dose of simvastatin, I get a thyroid-stimulating hormone level to be sure thyroid disease is not present because this is a risk factor for myopathy with statin treatment.

Dr. Roberts: What would you do with a 35-year-old woman with a total cholesterol of 350 mg/dl, for example? Would you routinely do a thyroid hormone study in that person?

Dr. Guyton: I’ve seen patients who were functioning fairly well and had a high level of cholesterol, and it was completely due to rather severe thyroid disease. At the start of our clinical trials we always get a thyroid-stimulating hormone level. We had 2 patients in 1 trial who each presented with thyroid disease, and yet they walked in and were enthusiastic volunteers for the clinical trial.

Overall, statins are very safe. The Statin Safety Conference, sponsored by the National Lipid Association, suggested that an increase of the hepatic transaminases is essentially the only hepatic side effect of statins. When you’re using the highest dose of a statin, usually an 80-mg dose, approximately 2% of patients have hepatic transaminase increases. A panel of 3 liver experts, 2 of whom had specifically studied the relation between statins and hepatic safety, gave their opinion that regulatory agencies should remove the requirement for hepatic transaminase monitoring, because it really didn’t make any difference clinically. Fatal or permanent liver damage occurs in statin-treated patients at rates essentially equal to rates in the general population. These experts, therefore, considered the monitoring of transaminases unnecessary. Unfortunately, the Food and Drug Administration (FDA) and the European Regulatory Agency have not looked again at that question.

Dr. LaRosa: I would like to comment on the issue of cancer and statin use. This issue has come up in some clinical trials. In the CARE [Cholesterol and Recurrent Events] and PROSPER [Pravastatin in Elderly Individuals at Risk of Vascular Disease] trials there were several cases of breast cancer. There have been 2 meta-analyses that have not demonstrated a cancer-related effect, although there was 1 meta-analysis from the Veterans Administration that indicated that there was a correlation between achieved LDL and an increased risk of cancer, although not related to the type of statin. The general issue for me is this: the widespread use of statins has resulted in a decrease in coronary disease and that almost surely at some point will extend the length of life. How are we going to separate the potential effect of statins to extend life span such that patients develop cancer rather than die from complications of atherosclerosis?

Dr. Guyton: The trials you mentioned are not the largest statin trials. Trials with 17,000 or 20,000 patients have not shown an increase in cancer.

Dr. LaRosa: I agree. I believe that it is a nonissue.

Dr. Guyton: There is a very extensive summary of statin safety in the National Lipid Association’s Statin Safety Taskforce Report. That report was published in 2006 in The American Journal of Cardiology . The following year a safety supplement on the nonstatin lipid-modifying drugs was published. Those reports provide extensive references that answer a lot of questions on the safety of lipid-altering agents.

Dr. LaRosa: The concern about statin side effects is only justified by the fact that so many people take them. These are among the safest drugs that have ever been approved by the FDA. They are much safer than aspirin, which of course never had to go through the FDA approval process.

Dr. Gotto: I remember when statins were first approved. If a patient said they had >2 alcoholic drinks daily, a statin would not be prescribed. The patients were sent to an ophthalmologist to get a slit-lamp examination. Early on there were just a whole battery of things the patient had to go through. Now, they are prescribed much more readily.

Intensive versus moderate lowering of LDL: how low should LDL go?

Dr. Viggiani: There are several studies that examined the benefits of intensive versus moderate lipid lowering. Dr. LaRosa, how low is low enough?

Dr. LaRosa: Neither TNT nor Incremental Decrease in End Points Through Aggressive Lipid Lowering study nor PROVE-IT answered the question of which strategy is better. One thing that JUPITER demonstrated is that getting LDL lower than in any of those previous studies gave patients additional benefits. Is there a point at which there’s no more benefit to lowering LDL, and is there a point at which lowering LDL becomes dangerous? There are some persons with LDLs in the 20s and 30s and they seem to do fine.

Dr. Gotto: A daughter of 1 of the subjects in the Dallas Heart Study who was a compound heterozygote for 2 PCSK9 mutations had an LDL of 14 mg/dl. This patient was studied extensively and there were no clinically significant abnormalities found. Patients with PCSK9 mutations are different from those with abetalipoproteinemia, who have a very severe deficiency of LDL; some of that difference may be related to lower levels of fat-soluble vitamins during early development.

Dr. Roberts: At birth, the serum LDL cholesterol is about 40 mg/dl.

Dr. LaRosa: I think that none of these studies have defined the level of LDL below which there’s no additional decrease in risk. The sample size required to try to answer that question would have to be enormous. From JUPITER, there was benefit with LDLs into the mid-50s. How will we study patients with levels going from 55 to 35 mg/dl? The absolute event rate is likely to be so low that even if we could do it we’d have to have a million people in the study, and they would have to be followed for a very long time. It will be a difficult question to answer.

Treating low HDL and increased triglycerides

Dr. Viggiani: What are the current treatment options for patients with low HDL and elevated triglycerides?

Dr. LaRosa: Perhaps the first question to ask is, should patients be treated with drugs to increase HDL in the first place? I think there is very little trial evidence to justify that.

Dr. Gotto: Actually, I think there is good evidence for benefit of using statins in patients with low HDL. Statins work very well in these patients. We did a primary prevention study based on low HDL, AFCAPS/TexCAPS. Patients with similar levels of LDL—not high enough to treat—had increased risk associated with low HDL abolished by statins, just as the increased risk associated with a high CRP is abolished by statins. I think for low HDL or high triglycerides, one should start with nonpharmacologic therapy: lifestyle modification, diet, and exercise; and if that doesn’t work, then the main classes of drugs to consider are the fibrates and nicotinic acid for isolated hypertriglyceridemia and the ω-3 fatty acids.

Dr. Viggiani: Data from Investigation of Lipid Level Management Using Coronary Ultrasound to Assess Reduction of Atherosclerosis by CETP Inhibition and HDL Elevation and Investigation of lipid level management to understand its impact in atherosclerotic events (ILLUMINATE) indicated that use of the cholesterol ester transport protein [CETP] inhibitor, torcetrapib, in combination with atorvastatin, did not reverse the atherogenic process and, in fact, increased the risk of morbidity and mortality. What should we know about the CETP inhibitors?

Dr. Gotto: We know that torcetrapib had off-target effects: it increased blood pressure, increased aldosterone secretion, caused electrolyte changes consistent with aldosterone, and it was associated with increasing cardiovascular events and death, although it increased HDL and lowered LDL. At 12 months in the ILLUMINATE trial, HDL was increased by >72% and LDL decreased nearly 25% compared to baseline. There’s 1 ongoing trial, and another 1 is planned, with different agents in this class that don’t seem to cause the aldosterone secretion or blood pressure increase. Whether or not CETP inhibition will be beneficial is debatable. One can argue that having a higher level of HDL will have anti-inflammatory effects and other beneficial effects, even if it doesn’t promote reverse cholesterol transport. There’s a debate about what proportion of cholesterol is carried from peripheral tissues back to the liver by HDL and how much goes into the LDL pathway by CETP. For a patient with a very effective LDL receptor mechanism and LDL clearance mechanism, one might not want to block that transfer into the LDL pathway. So, there are hypothetical arguments on both sides. Until we get the results of an outcomes trial, we’re not going to know whether or not these agents are effective.

Dr. Guyton: We should remember not to equate HDL increase with CETP inhibition because we actually have a very good HDL-increasing drug in niacin. Unfortunately, niacin has 2 issues. First, it’s difficult for many people to take, although with good patient counseling I believe 80% to 90% of people can take it. Second, niacin until recently has not had a large pharmaceutical company put up the funds for a large clinical trial. The Coronary Drug Project was 1 of the first National Institutes of Health (NIH)-sponsored multicenter, randomized trials. It was conducted in men who had had a previous myocardial infarction. Niacin prevented 27% of recurrent nonfatal heart attacks in that study. It also decreased the combined end point of cardiovascular death and myocardial infarction, although this result was not statistically significant. It also decreased cerebrovascular events by 26%. The investigators were looking for a decrease in total mortality in this study. Because they didn’t find it, the results were considered disappointing and very few patients continued niacin after the end of the study. Nine years after the trial ended, Canner and colleagues tracked down the study participants. They found that there was an 11% relative risk decrease in total mortality in the niacin group [52.0% vs 58.2%, p = 0.0004] and an absolute mortality risk decrease that is 1 of the largest absolute risk decreases in any trial. It was not, however, the primary end point of the main study.

Almost entirely forgotten are the 555 patients reported by Carlson and Rosenhamer in the Stockholm Ischemic Heart Disease trial. Survivors of myocardial infarction were randomly assigned to open-label combined treatment with niacin and clofibrate or to no lipid treatment. Clofibrate is a drug we don’t use any more because 1 study showed it to increase total mortality when used alone. The combination of niacin with clofibrate, however, decreased CHD mortality by 36% and total mortality by 26%.

Beyond that there are 3 anatomic end point studies that actually showed CVD event decreases. Each of these small trials had approximately 150 to 200 participants, and each used niacin in combination with other drugs: simvastatin, gemfibrozil, and/or colestipol. Despite their small size, each of these studies showed a decreased risk of CVD events.

I believe that after statins, niacin has the most consistent and best record of improved outcomes of the lipid-modifying drugs. Although more large randomized trials have been performed with fibrates such as gemfibrozil and fenofibrate, niacin beats fibrates for consistency of benefit. Currently, 2 large randomized trials are in process with niacin: the Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglycerides and Impact on Global Health Outcomes study and the Heart Protection Study 2 Treatment of HDL to Reduce the Incidence of Vascular Events study.

Dr. LaRosa: Niacin has annoying side effects for most people.

Dr. Guyton: The number of serious side effects is small.

Dr. LaRosa: Yes, but unlike statins, some people have died from niacin-induced liver failure.

Dr. Guyton: There is 1 report that I’m aware of, but that patient had a previous liver transplant.

Dr. LaRosa: Niacin also has a measurable impact on glucose tolerance. I agree that niacin in doses effective enough to have an impact on lipids has a good track record in terms of efficacy, but it’s a miserable drug to take.

Dr. Guyton: I disagree. When patients see what happens to their lipids on niacin therapy, they usually are happy to tolerate the niacin-induced flushing. Moreover, flushing is subject to tachyphylaxis; with regular dosing, flushing events become milder and rarer over a period of weeks. What is needed is good instructions: precede the niacin dosing by a 325-mg aspirin tablet, and take the niacin in the middle of a meal or with a light snack at bedtime. All the hepatic toxicity with niacin has occurred essentially in people taking slow-release, over-the-counter niacin tablets, given 2 times/day (morning and evening). There is 1 report of serious liver toxicity with regular niacin, and no report of severe liver toxicity with the prescription extended-release niacin.

Dr. Gotto: In a study looking at niacin in combination with a prostacyclin inhibitor, there was increased patient acceptability, but there still was a significant degree of discontinuation, primarily due to flushing. The main reason patients don’t take niacin is the flushing. For clinicians who prescribe it a lot, they may get a much higher rate of adherence, but even in the trials where they tried to improve patient tolerance and acceptability, there are still pretty high discontinuation rates.

Dr. Roberts: Does anybody use bile acid resins anymore?

Dr. LaRosa: Almost never.

Dr. Gotto: Colesevelam works pretty well in patients who need additional decreases or who are statin intolerant. I have patients who have been on these agents since the 1970s.

Dr. Guyton: Bile acid sequestrants in my view have a very strong role in patients with familial hypercholesterolemia. Many of these patients do not get to goal with a statin alone, or with a statin and ezetimibe, or even with a statin, ezetimibe, and niacin.