US Lipid Guidelines Revisited: Any Regrets?

Tricia Ward


July 14, 2016

Editor's Note: It's nearing 3 years since the release of the controversial American College of Cardiology (ACC)/American Heart Association (AHA) guidelines for the treatment of blood cholesterol to reduce atherosclerotic cardiovascular disease (ASCVD) risk in adults. Much has been published, both pro and con, regarding the decision to eliminate specific targets for low-density lipoprotein cholesterol (LDL-C) targets. Medscape checked in with guideline writing committee member Donald M. Lloyd-Jones, MD, ScM. Dr Lloyd-Jones is professor and chair of the Department of Preventive Medicine at Northwestern University Feinberg School of Medicine, and is professor of medicine (cardiology) at Northwestern Medicine, both in Chicago, Illinois.

Dr Lloyd-Jones | Medscape: Was the lack of randomized controlled data the reason for eliminating specific LDL-C targets from the 2013 ACC/AHA cholesterol guidelines?[1]

Dr Lloyd-Jones: Yes, that was the major reason. No studies have used either different intensities of statins or different combinations of medications to try to get to a specific LDL-C level; therefore, it's very difficult to say what the appropriate goal ought to be. In the guidelines we took pains to say that, in general, we think lower LDL-C is better, but it matters how you get there.

We were thinking not only about the benefits of lowering LDL-C, but also about the potential risks from adverse medication reactions and other side effects. When you take the net benefit into account, that's where you come to the conclusion that lower LDL-C generally is better, but it matters how you get there, and in whom, which is a good way to sum up the guidelines. | Medscape: A meta-analysis from Israel by Leibowitz,[2] published in JAMA Internal Medicine, showed support for a goal of < 100 mg/dL (2.6 mmol/L) but not < 70 mg/dL (1.8 mmol/L) in patients with established ischemic heart disease. The accompanying editorial[3] notes that the study "raises questions about the practice of statin dosing by intensity. What do you think of that paper?

Dr Lloyd-Jones: It's fair to say that those data would be consistent with that hypothesis, but they certainly can't say anything definitive about whether that's correct or not. This is a very nicely done study, but as the authors are careful to point out, these are observational data. These were not patients who were randomized to a strategy of lowering their LDL-C to < 70 mg/dL versus < 100 mg/dL or < 130 mg/dL (3.4 mmol/L). There are many, many variables that we just can't account for that could explain the findings.

For example, we don't have any information about the baseline LDL-C levels of these patients. We know where they ended up (< 70 mg/dL, < 100 mg/dL, or < 130 mg/dL), but we don't know where they started. For any given dose of statin medication, the major determinant of where you end up is your LDL-C level when you start. That's an important driver of your risk. Without that information, it's difficult to understand the implications of these data for selecting a target.

The other important feature of where you end up is the statin dose. Many things go into that decision: patient preference, provider preference, what's generic, what's available, etc. There are a lot of moving parts to this analysis that are not well accounted for. It's a reach to say, as the authors did—and certainly the editorialists did—that these data suggest that we should use a target of < 100 mg/dL and not aim lower. I saw nothing in the paper that supports the editorialists' conclusion that we should now move away from evidence-based statin intensity dosing to absolute LDL-C targets. | Medscape: In preparation for this interview, you sent a paper by Bangalore et al[4] which pooled patients from three large randomized trials and split them by into groups by attained LDL-C levels (≤ 70 vs > 70 mg/dL) and percent LDL-C reduction (≥ 50% vs < 50%). They concluded that percent LDL-C reduction provides incremental prognostic value over statin dose and attained LDL-C levels. Do you believe that these data are stronger than the Israeli paper?

Dr Lloyd-Jones: The Bangalore paper is a much more controlled experiment than the Leibowitz analysis. They looked at where patients ended up, but they also considered where they started and the intensity of the statin therapy. They assessed which of those variables actually contribute to understanding who's at higher and lower risk. The strong suggestion from those data is that the percent lowering of LDL-C and the dose of statin are the two most important pieces of the equation, whether or not you get to an LDL-C of < 70 mg/dL. Their other observation is that the actual achieved LDL-C levels don't really tell you much about residual risk.

We need to put both of these studies in the mix as we consider the next iteration of the guidelines. They are important but certainly not the final word on the potential LDL-C goals, whether that's percent lowering or an actual LDL-C level. The JAMA Internal Medicine editorialists, in particular, went far beyond the data in their conclusion that we should shoot for < 100 mg/dL and that such a target is clearly better than percent LDL-C reduction. There's nothing in the Israeli paper about percent LDL-C reduction because we don't know the baseline values. Their conclusion is well beyond what that paper can address, and the Bangalore paper suggests quite the opposite.

...If people really take the time to read what we said, we came to fairly conservative conclusions as a result of very strong data. I think that they are going to stand the test of time. | Medscape: In terms of percent LDL-C reduction, the guidelines support aiming for a 50% reduction. Is that within about 4-6 weeks of treatment?

Dr Lloyd-Jones: For these higher-risk patients, who are secondary prevention patients because they already have atherosclerotic cardiovascular disease, we would typically start at high-intensity statin therapy and then remeasure at about 4-6 weeks to see what kind of LDL-C response we got.

We had a table in the guidelines of the doses of which drugs would be considered low, moderate, or high intensity. It's really a function of dose and strength of the statin, and not patient factors. As detailed in the recent paper from Israel, low intensity would be simvastatin < 20 mg/day or equivalent. Intermediate would be up to 40 mg of simvastatin/day or equivalent in other statins. And high would be simvastatin 80 mg/day or equivalent or greater potency and intensity. As you probably know, simvastatin at 80 mg, and even at 40 mg, is restricted by FDA black box warnings because of the concerns about drug interactions. I think we're likely to see use of that particular medication falling. | Medscape: Another distinction between the studies that our readers will point out is that the Bangalore study is funded by industry, and one of the authors is a Pfizer employee, whereas the Leibowitz Israeli study has no funding from industry. Can you comment on that?

Dr Lloyd-Jones: The first author, Bangalore, is not an industry employee. I'm not here to defend them, and we always need to be aware of potential conflicts of interest. But that being said, the design of the Bangalore study, whereby patients were randomized to statin or not, or higher- versus lower-intensity statin, takes into account a lot more of the potential reasons that one might achieve a lower or higher LDL-C level. The Leibowitz study did a nice job with their propensity and sensitivity analyses of trying to account for things we know about, but it doesn't account for things that we don't know about.

An example of how observational data can't necessarily control for everything is that if you look at the people in the Leibowitz analysis, who got to an LDL-C < 70 mg/dL, they were much more likely to have diabetes than the groups that got to higher LDL-C levels. At baseline they're already at higher risk for cardiovascular disease. At the other end of the spectrum, the people who got to an LDL-C between 100 and 130 mg/dL had far more comorbidities. They're sicker and they're clearly more likely to die of something else than to have a cardiovascular event and get benefit from a statin.

When you have imbalances like that, with higher risk in the LDL-C < 70 mg/dL group and a competing non-CV risk of dying in the higher LDL-C groups, you're going to get a lot of noise around that signal—and it's difficult to filter out all of that noise.

We can't say that this is the final word, by any stretch, on what appropriate LDL-C goals might be. It's not that simple. | Medscape: There are guidelines from the National Lipid Association,[5] the European Society of Cardiology,[6] and the Canadian Cardiovascular Society[7] that continue to include LDL-C targets. Do you think that is confusing for clinicians?

Dr Lloyd-Jones: Yes. It is a little confusing. The National Lipid Association recommendations are not a guideline, and they appropriately note that their recommendations are not based on a formal systematic review of the evidence. We don't always have direct data to answer the questions we'd like to answer, and that's where it's helpful to have expert consensus recommendations.

In regard to the targets, I think that it's increasingly difficult to support them or to have a one-size-fits-all number. It's pretty clear that that's not the way the biology works. If we look at some of the targets, they are < 100 mg/dL according to the TP-III[8] and National Lipid Association; in Canada, < 80 mg/dL [2 mmol/L]; in Europe, < 70 mg/dL. That should tell you that nobody really knows and that these targets are not evidence-based.

I still tell my patients that, in general, I would like their LDL-C to be as low as possible, but we're going to be thoughtful about how we get there because we want to balance the potential benefits against the potential harms. | Medscape: Since the original 2013 guidelines, IMPROVE-IT[9] was published and PCSK9 inhibitors were approved. Where do they fit in?

Dr Lloyd-Jones: That was the main driver for the 2016 ACC "Expert Consensus Decision Pathway on the Role of Non-statin Therapies for LDL-Cholesterol Lowering in the Management of Atherosclerotic Cardiovascular Disease Risk," which was just published in the Journal of the American College of Cardiology.[10]

This document builds on the 2013 cholesterol guidelines. It's a consensus statement and as such is not a formal systematic review of the evidence. We wanted to provide explicit guidance for practitioners considering when and whether to use other medications once they have a patient on a statin, or in patients who, for whatever reason, can't take a statin. We discuss when you might consider ezetimibe or PCSK9 inhibitors, or perhaps bile acid sequestrants, and that niacin should not be considered for most patients. It addresses a number of different types of patient groups and how to weigh the potential benefits and harms of all of the adjunctive medications. This is in the context of patients who either are already on the appropriate dose of statin or can't take a statin. | Medscape: For the final few questions, I'll ask you to tackle the skeptics. First are those who believe in statins for secondary prevention but are concerned that the drugs are overused in primary prevention and given a higher precedence over diet and lifestyle.

Dr Lloyd-Jones: We certainly see the largest absolute benefit for statins in terms of reducing risk in people who need secondary prevention. I don't think anybody would argue with that (and that's the population in the Leibowitz study). I don't think that the Israeli authors or anybody else is saying that secondary prevention is an inappropriate venue for using statins.

There's been a lot of debate about the relative benefits of using them in primary prevention. I think it is indisputable that statins reduce the risk not only for heart attacks and strokes but also for total mortality in primary-prevention patients. The question is, which primary prevention patients? They need to be at high enough risk so that the benefits of taking a statin outweigh any potential adverse effects.

In the guidelines, we took great pains and wrote at great length about the appropriate levels for which they should be considered. That level of 7.5%, or even 5%, ASCVD risk has been validated[11,12] over and over again in a number of different settings as not only being cost-effective[13] (which it clearly is), but also as a reasonable point at which patients should be having conversations with their doctor. What the guidelines really do in relation to primary prevention is to ask: What's the likely benefit? What are potential harms? And what are the patient preferences? People who've actually read the guidelines would say, "You did a pretty good job of actually laying all that out so that patients and doctors can make better decisions." | Medscape: One of the problems is that people don't necessarily read all of the guidelines, and if they look at the algorithm in Figure 2 of the guidelines, they see that > 7.5% risk points to moderate- to high-intensity statin.

Dr Lloyd-Jones: For primary prevention, you can't get to a statin prescription without the patient-clinician discussion (per Figure 2 of the guidelines). Even those people who aren't taking the time to read the guideline shouldn't be mailing prescriptions without having a patient-clinician discussion. That's well outlined in the guidelines. | Medscape: The next critique relates to the side effects of statins. Could you address that?

Dr Lloyd-Jones: There are perhaps two major side effects that people worry about with statins. I don't think anyone seriously thinks that they cause cancers or any other major significant long-term side effects.

There's been debate about the importance of the increased diagnosis of diabetes in people who start statin therapy. That appears to be a real dose- and intensity-dependent phenomenon. But in terms of absolute numbers of increased rates of diabetes, it's quite small.

What has been nicely shown is that nobody with a normal blood sugar or without significant risk factors for diabetes becomes diabetic as a result of being started on a statin. The JUPITER investigators did a very nice analysis[14] where they showed that the date of diagnosis of diabetes in JUPITER was accelerated by 5.4 weeks for people who got diabetes as a result of being started on a statin vs the placebo group. Those people were going to get diabetes anyway, and you don't prevent diabetes by withholding a statin. In fact, you do those people a lot of good by giving them a statin because you're going to prevent subsequent heart attacks and strokes. You have to prevent the diabetes with lifestyle; that we know.

I think the diabetes issue is largely a nonissue that gets far too much attention. We need to be preventing diabetes regardless, and we need to put our patients with diabetes on statins to prevent the devastating consequences. This is a small amount of change in blood glucose levels in a small number of patients, pushing them over a diagnostic threshold that they were going to cross anyway 5.4 weeks later.

The other major issue that gets a lot of airtime is myalgias, or muscle aches, with statins. It is certainly fair to say that about 20% of people who take a statin will have muscle aches, but the very well-designed studies in all-comers who repeatedly have myalgias on statins show that fewer than half of them can tell when they're getting a statin and when they're getting a placebo. The true rate of myalgias in the population due to statins is much more like 5%, and not the 20% that gets bandied about in the literature. | Medscape: Is the 20% from studies like GAUSS-3?[15]

Dr Lloyd-Jones: GAUSS-3 included people who had had myalgias on multiple statins. There are some very nice data from Paul Thompson,[16] including some of his studies with CoQ10,[17] where the number truly affected is around 5% after multiple crossovers between a statin and placebo. About three quarters of the people couldn't tell when they were getting the medication and when they were getting the placebo in regard to the occurrence of muscle aches.

We have an effective class of medications, statins, that can prevent our leading cause of death and disability. Of course, any medication will have side effects, but we ignore the big signal here at our peril if all we're focusing on is the side effects. Unfortunately, a disservice has been done in putting so much emphasis on the potential side effects when these drugs are preventing heart attacks, strokes, and death that would otherwise be occurring in large numbers of people. | Medscape: That leads to the most extreme skeptics: the LDL-C–atherosclerosis hypothesis doubters. How do you respond to them?

Dr Lloyd-Jones: From my reading of the literature and decades and decades of very high-quality science, there is no doubt in my mind that cholesterol—and LDL-C in particular—are the central molecules in atherogenesis. No question about it. Certainly, other things contribute, but the central molecule in creating atherosclerotic plaques is LDL-C. Again, we ignore that science at our peril.

If these doubters have high cholesterol or other risk factors, and they're not at least considering taking a statin after all the trials we've seen (most recently even with HOPE-3[18]), I'm not sure that they're serving the interests of themselves and the public very well. | Medscape: It's been 2.5 years since the guidelines came out. If you were to go back in time, would you do anything differently regarding their rollout?

Dr Lloyd-Jones: The conclusions we came to are rock-solid foundational recommendations because they were based on randomized controlled clinical trial data. The framework in which we put them, particularly including the patient-clinician discussion (especially in primary prevention), was really revolutionary and an incredibly important step forward. And the evidence-based guidelines that have been published since then, from the Joint British Societies,[19] the USPSTF, and the VA,[20] have all taken remarkably similar approaches.

Unfortunately, we didn't get the opportunity to explain them in the amount of detail we would've liked before a number of people—some of whom perhaps even still have not read the guidelines—chimed in with their opinions. That's the nature of the beast. But if people really take the time to read what we said, we came to fairly conservative conclusions as a result of very strong data. I think that they are going to stand the test of time.

Disclosure: Donald M. Lloyd-Jones, MD, ScM, has disclosed no relevant financial relationships.

Follow Tricia Ward on Twitter: @_triciaward
For more Cardiology news and features, join us on Facebook and Twitter @theheartorg


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: