Update of the US National Cholesterol Education Program (NCEP) Adult Treatment Panel (ATP) III Guidelines: An Expert Interview With Christie M Ballantyne, MD

Introduction

Editor's Note: 
Dr Ballantyne is Director of the Center for Cardiovascular Disease Prevention at Baylor College of Medicine and the Methodist DeBakey Heart Center, Houston, Texas. Dr Ballantyne also serves as Professor of Medicine and Associate Chief and Clinical Director of the Section of Atherosclerosis and Lipoprotein Research in the Department of Medicine at Baylor College of Medicine; Director of the Atherosclerosis and Lipid Laboratory at Baylor; and Co-director of the Lipid Metabolism and Atherosclerosis Clinic at The Methodist Hospital.

Dr Ballantyne's clinical research is focused on the prevention of atherosclerotic vascular disease. He has more than 150 published works in the areas of atherosclerosis, lipids, and inflammation. He is an Associate Editor for the American Heart Association journal Circulation and for www.lipidsonline.org.

In July 2004, the National Cholesterol Education Program (NCEP) coordinating committee issued an update to the third Adult Treatment Panel (ATP III) guidelines on the detection, evaluation, and treatment of high blood cholesterol in adults.^[1]The update was based on evidence from 5 clinical trials published since the ATP III guidelines^[2] were issued in 2001: the Heart Protection Study (HPS); ^[3] the PROspective Pravastatin in elderly individuals at risk of vascular disease (PROSPER);^[4] the Antihypertensive and Lipid-Lowering Treatment to Prevent Heart Attack Trial (ALLHAT-LLT);^[5] the Anglo-Scandinavian Cardiac Outcomes Trial—Lipid Lowering Arm (ASCOT-LLA);^[6] and the Pravastatin or Atorvastatin Evaluation and Infection Therapy – Thrombolysis in Myocardial Infarction 22 (PROVE-IT TIMI 22).^[7]

On the basis of these 5 trials, the main recommended modifications to ATP III were:

The new recommendations and therapeutic options are endorsed by the National Heart, Lung, and Blood Institute, the American Heart Association, and the American College of Cardiology.

In this interview with Medscape, Dr. Ballantyne, who is not directly associated with the NCEP expert panel that devised the ATP III guidelines or the update, discusses the recommendations and therapeutic options proposed in it.

Medscape: Do you think there was a need for new cholesterol guidelines at this stage?

Dr. Ballantyne: I think there is still a lot of confusion about the latest NCEP recommendations, but I believe that for the most part they were on target. They are not new guidelines; this is basically an update of ATP III and they are based on key new evidence. As more scientific data appear, it is completely appropriate to provide updated evidence with regard to how physicians can incorporate that information into clinical practice. I think that the NCEP panel has done that. The document is very carefully worded.

Medscape: The new document refers to "therapeutic options" rather than "recommendations." Could that be confusing for physicians?

Dr. Ballantyne: It depends on whether guidelines are supposed to be cookbook medicine or whether they are supposed to help physicians to take care of individual patients, in which case most of this has to be informed by clinical judgment. Although the guidelines are quite complicated, the recommendations are based on clinical trial evidence. It is intended that they do avoid becoming quickly out of date and so they leave room for clinical judgment .

Medscape: Which of the 5 trials had most influence in generating the need for an update of ATP III?

Dr. Ballantyne: I think that the one that was most straightforward is the Heart Protection Study (HPS). This was an enormous study, in over 20,000 individuals, that very clearly showed that high-risk patients benefit from LDL-cholesterol-lowering drug therapy regardless of their LDL-cholesterol level at baseline. This is a very important group of people who have a very high event rate. The ATP III recommendations were to initiate drug therapy where LDL-cholesterol was > 130 mg/dL. There was a "gray zone" of 100-129 mg/dL, and for these patients, ATP III did not mention using drugs. Drug therapy was not even a consideration for people with LDL-cholesterol < 100 mg/dL.

From an analysis we did of data from the third National Health and Nutrition Examination Survey (NHANES III, 1998 to 1994),^[8] we estimated that in the United States alone there are 28 million people who are coronary heart disease (CHD) or CHD risk equivalents, 15 million with LDL-cholesterol > 130 mg/dL, 8 million LDL-cholesterol 100-129 mg/dL, in the gray zone, and 5 million with LDL-cholesterol < 100 mg/dL. So of the 28 million people at high risk in the United States, ATP III gave clearcut guidelines for 15 million but none for the other 13 million. It would have been irresponsible not to issue guidelines for these individuals based on evidence from trials that these people benefit from drug therapy. The ATP III guidelines stated that drug therapy was optional in a group of 8 million high-risk patients with LDL-cholesterol > 100 mg/dL and < 130 mg.dL.

Medscape: So physicians might previously have been offering these patients lifestyle changes only?

Dr. Ballantyne: Exactly, and another 5 million people who also had LDL-cholesterol < 100 mg/dL were not even considered for drug therapy. So this is a fairly large public health issue, and it was completely appropriate to get that information out.

Medscape: Why was it done now?

Dr. Ballantyne: I think some people would question why it was not done even sooner, because these are very important issues for both patients and clinicians. When you have evidence that is strong and where there is a scientific consensus, you would like to have your guidelines to conform with the evidence. Clearly, the guidelines were not conforming to the evidence, so it was necessary to put out an update for this.

Medscape: What are the key points in the updated guidelines?

Dr Ballantyne: The first one is the greater focus on global risk. The guidelines now say that for people at very high risk, treat everyone with an LDL-cholesterol > 100 mg/dL and also consider treating those with LDL-cholesterol < 100 mg/dL. That is an option based on very strong evidence, and it has tremendous benefits for patients. I do not think anyone can look at a 20,000-person study like HPS and say that the data do not make sense.

The second key point is similar, but for the moderately high-risk patient, for whom the option is to treating at LDL-cholesterol between 100 and 129 mg/dL to achieve a target of < 100 mg/dL. The data that we have from the Anglo-Scandinavian Cardiac Outcomes Trial (ASCOT) are also very straightforward, showing that there is benefit in treating these patients. This is another very large population, and I do not believe that there is any doubt that these people benefit from therapy.

An important point is that therapy should be sufficient to reduce LDL-cholesterol levels by 30% to 40%, because that is what all 5 studies achieved. The updated guidelines estimate the daily doses of currently available statins that can achieve this: atorvastatin 10 mg, lovastatin 4 mg, pravastatin 40 mg, simvastatin 20-40 mg, fluvastatin 40-80 mg, and rosuvastatin 5-10 mg. Another study that was not mentioned in the update was the Lescol Intervention Prevention Study (LIPS).^[9] It was another study that used 80 mg of fluvastatin (Lescol).

Medscape: What is your opinion about the evidence or lack of it for a lower limit for LDL-cholesterol lowering?

Dr. Ballantyne: The updated guidelines have created a new category of very high-risk people with an optional target of > 70 mg/dL. This is probably controversial. The ATP III guidelines recommended a target of < 100 mg/dL, but ATP II said ≤ 100 mg/dL,^[10] a subtle distinction, but implying that 100 mg/dL is good enough. Saying less than 100 mg/L, you are really leaving it open to clinical judgment. Some people may have been saying that we should be getting LDL-cholesterol down to < 90 mg/dL, maybe < 70 mg/dL even before this update came out. The update was based on the PROVE-IT TIMI-22 study, which suggested that in very high-risk patients, a goal of < 70 mg/dL could be appropriate. The update does point out that this is only for very high-risk patients and not for everyone, because there are some large trials in progress that will provide evidence for people who have more stable CHD. The Treating to New Targets (TNT) study comparing atorvastatin 10 mg vs 80 mg has been completed, and the results are expected in early 2005.^[11] The results of the Incremental Decrease in Events through Aggressive Lipid Lowering (IDEAL) study, comparing atorvastatin 80 mg versus simvastatin as used in the Scandinavian Simvastatin Survival Study (4S),^[12] ie, 20 mg titrated to 40 mg, to get cholesterol to < 200 mg/dL,^[13] and the results of the Study of the Effectiveness of Additional Reductions of Cholesterol and Homocysteine (SEARCH), which is comparing simvastatin 80 mg vs 20 mg, are expected later the same year. Until we have those results, you would not be recommending a lower target in other people at high risk..

The guidelines point out that it is quite difficult to get LDL-cholesterol down to < 70 mg/dL. One issue is that to achieve this frequently requires a reduction of ≥ 50%, and to do this there are currently only 3 daily dose options: atorvastatin 80 mg, rosuvastatin 20-40 mg, or the new combination of ezetimibe/simvastatin starting with 10/20 mg or higher. All these will achieve a > 50% reduction in LDL-cholesterol as a single pill, otherwise it is necessary to go with multiple drugs. I think that is why there was a lot of caution with the wording of the update, recognizing that it is harder to achieve these levels, and why the major focus was to ensure that people are on a good dose of statin to achieve at least a 30% to 40% reduction.

Whether there is a lower limit for healthy LDL-cholesterol levels is an interesting question. HPS included a large number of people with LDL-cholesterol < 100 mg/dL at baseline and treated with 40 mg of simvastatin. The biggest safety concern had been based on epidemiologic studies, mostly in Asia, that very low levels of LDL-cholesterol would lead to an increased risk for hemorrhagic stroke. It is very important to note that in HPS, where there were many individuals who achieved very low levels of LDL-cholesterol, no increased risk for hemorrhagic stroke was observed; rather, there was a reduction in ischemic stroke overall. So it does not appear that studies of pharmacologic therapy supported what occurred in epidemiologic studies, and you have to wonder why, in an epidemiologic study, did someone have cholesterol that low? It may be that they had chronic liver disease or another condition that was responsible.

There is the need for data from the trials that are comparing doses given to get LDL-cholesterol reductions of 30% to 40% compared with higher doses that achieve greater reductions, because there is the issue of how much more benefit they confer along with more side effects. That is an area where we clearly need to be able to calculate benefits and risks. We do not have many data on the comparison of the highest and lowest doses, nor do we have the evidence about event reduction compared with LDL-cholesterol reduction with the ezetimibe/simvastatin/ezetimibe combination. This is an area where there can be a very legitimate difference of opinion as to what we should be doing for patients and in terms of just how low to go.

Medscape: There has been some controversy about safety of statins at high dosages recently. Do you foresee any revisions would be needed?

Dr. Ballantyne: Where that controversy comes up is with regard to the highest dose. In the PROVE-IT study, there was a significantly greater event reduction and there also were significantly greater adverse effects in the course of liver function tests. If we look at the trade-off, however, it seems quite worth it.

The Aggrastat to Zocor (TIMI 21) (A to Z) study, which compared an intensive vs a less intensive simvastatin regimen in patients with acute coronary syndrome, did not achieve statistical significance in terms of clinical benefit except for perhaps the most important endpoint, CHD, and the intensive regimen was associated with increased side effects.^[14] So we do need some more evidence before we can say that everyone should get a 50% reduction in LDL-cholesterol. There could be differences between the top dose of one statin or another in terms of side-effect profile, or it may be other drugs people are taking or other conditions. So this was probably one reason why the latest update was not ATP IV, because there was not enough evidence for changing all the targets.

Medscape: What about the importance of raising levels of HDL-cholesterol?

Dr Ballantyne: Probably the most controversial part of the guidelines was the footnote, "If a high risk person has high triglycerides or low HDL-cholesterol, combining a fibrate or nicotinic acid with an LDL-lowering drug is a therapeutic option . . ." I do this routinely, it is logical, but it is an area where we do not have all the evidence. We are doing a study based on this concept, ELIMIT, in which we are comparing statin alone vs statin plus niacin plus ezetimibe, with the MRI of peripheral arterial disease as the primary endpoint. There have not been a lot of data on outcomes with statin/fibrate or statin/niacin combinations compared with statin alone. But please note that the update says to consider this, not to routinely add fibrates or niacin to a statin.

Medscape: Some of the evidence cited in the ATP III update has been challenged by the Center for Science in the Public Interest (CSPI).^[15]

Dr Ballantyne: There was a mistake in that the NCEP panel should have listed all of the potential conflicts of interest; I think everyone agrees with that. But that is very different from saying that the work is wrong. They said a lot of things were taken out of context, such as that ASCOT-LLA did not show that there was benefit in people with diabetes. ASCOT-LLA was halted early because of a significant reduction in cardiovascular events compared with placebo. Analysis of subgroups failed to show significance among the diabetic population, but this was probably due to the sample size, shortened follow-up period, and higher drop in statin use among the diabetic patients on placebo. HPS showed benefit of cholesterol-lowering therapy in the almost 6000 patients with diabetes.^[16] HPS included many elderly patients who clearly also had benefit. The recently published Collaborative Atorvastatin Diabetes Study (CARDS), however, involved 2800 patients with type 2 diabetes and was halted 2 years early in June 2003 because patients on atorvastatin had significant reductions in events compared with those receiving placebo regardless of their baseline LDL-cholesterol.^[17] So I think these critics are taking a piece out of studies and are ignoring the totality of the data.

A valid argument can be made that not all diabetics are CHD risk equivalents, but the vast majority of them are and thus are at very high risk. It may have been an oversimplification of the guidelines to say that all diabetics are CHD risk equivalent, but it may have been considered that the guidelines were already too complicated to include another treatment algorithm for the ~10% of diabetic patients who are not CHD risk equivalent.

Medscape: The last question is about the complexity of the guidelines. They have been called complicated, and maybe they are compared with other guidelines in cardiology.

Dr Ballantyne: There are many nuances in the guidelines. I have tried to get a summarized version produced that can be put to a primary care audience, but it is very difficult to rewrite them and keep the versions consistent. Unfortunately, there are controversies, there is some complexity, and the guidelines probably reflect that. I hope that they can be simpler in the future.

Meantime, the key take-home message is that when you have identified someone who is at high risk or at moderately high risk that you decide to treat, give the drug doses that will achieve a 30% to 40% reduction in LDL-cholesterol. If you use these, you will get that reduction. Lifestyle modification should be implemented in all high-risk patients. Another issue, which is very straightforward, is that in high-risk patients, drug therapy should be initiated if LDL-cholesterol is < 100 mg/dL and considered if it is > 100 mg/dL. That is a very key point, broadening the net for high-risk patients. The other key point is that for moderately high-risk people with LDL-cholesterol 100-129 mg/dL, drug therapy is considered optional. Another focus is to get people on the right therapy. I think we have a lot of consensus on that.

The optional goal of < 70 mg/dL in very high-risk patients is one of great interest to people who are cardiologists or who take care of a lot of these very high-risk patients, it is also where probably the most controversy has arisen. I personally believe that the evidence from ongoing clinical trials will support this recommendation.