Chris’ Corner: What Is Going on With HDL?

We received a very thoughtful letter regarding my February column (see below), where I recounted the tumultuous year for lipid management, which was marked by four major failures for HDL treatments. It begs the simple question: what is going on with HDL? Is it, in fact, "good" cholesterol?

The whole idea of "good" and "bad" cholesterol came from an epidemiologic association of high LDL with higher risk of developing heart disease, while high HDL seemed to carry a lower risk. HDL appeared to exert a protective effect, so we all made a simple leap of faith—if we can increase HDL, we might be able to prevent heart disease. It has turned out not to be that simple.

As Dr. Roehm nicely reviews in his letter, the early data looked promising, but it's important to note these studies were carried out prior to the use of statins. More recently, trials evaluating niacin as add-on therapy with statins have presented one strike out after another (a bit like the 2012 Red Sox). A very reasonable explanation is that the modest effects of niacin can't further improve outcome in patients who are already treated well with statins, especially if their LDL is down to 62–63 mg/dL, as was the case in AIM-HIGH and HPS-2 THRIVE trials.

In the face of excellent current therapies, then, are HDL-raising therapies simply futile? A focus on HDL efflux as the key factor has also attracted attention. Dan Rader, MD, has shown that the function of HDL is a key factor, perhaps more important than the amount of HDL.

Should we throw in the towel and give up? Many of us still hold out hope for raising HDL. There are two approaches in advanced testing. (Disclosure: I am involved in trials/discussions in both these areas). The first is the use of potent cholesterylester transfer protein (CETP) inhibitors, anacetrapib and evacetrapib, that raise HDL to the order of 140% (taking HDL from about 40 to 100 mg/dL). That big of a pharmacologic difference could potentially be different than the 15–30% in the recent niacin and dalcetrapib trials. Even if the native HDL is not perfectly functioning, this should hopefully do something.

The second approach involves HDL infusions. This novel approach of giving a large dose of fully functioning HDL as a sort of drain cleaner could be a promising way to go. Phase II studies are ongoing and hopefully we will see results soon.

HDL has taken a few hits lately, but it's not down for the count quite yet. Now we're trying to really see if raising the so-called "good cholesterol" will actually be good.

Letter to the Editor

Dr. Cannon,

I had the pleasure of reading your February CardioSource WorldNews column. In the column, you touched on the topics of whether niacin is beneficial and the status of HDL as a risk factor.

Niacin clearly does not seem to be of benefit when used with statins as evidenced by AIM-HIGH (Atherothrombosis Intervention in Metabolic Syndrome with Low HDL/High Triglyceride and Impact on Global Health Outcomes) and the recently halted HPS2-THRIVE trial (Heart Protection Study 2-Treatment of HDL to Reduce the Incidence of Vascular Events).

The older literature, as you suggest, before the statin era, shows a benefit of niacin in post-MI patients.

In the Coronary Drug Project,1 the post-MI patients randomized to niacin versus placebo had a 27% reduction in nonfatal MI (8.9% vs. 12.2%; p = 0.002) and a 21% decrease in stroke and TIA (7.7% vs. 9.7%; p = 0.02). Additional follow-up showed a statistically significant reduction in total mortality for the niacin group.2

Similarly, in a randomized open label trial conducted in post-MI patients in Sweden,3 niacin combined with clofibrate compared to placebo resulted in a 27% reduction in total mortality (21.9% vs. 29.7%; p < 0.05) and a 36% reduction in CV mortality (16.8% vs. 26.4%: p < 0.01).

One reasonable interpretation of the data is that niacin was effective in secondary prevention in the pre-statin era, but not with concomitant statin therapy. Of course, an alternative approach is to lump the Coronary Drug Project and the Swedish trial in a meta-analysis with the much larger AIM-HIGH and HPS2-THRIVE, which will result in the conclusion that niacin is of no benefit.

I personally think the data are best explained by the modern-era, intensive statin treatment group constituting a different study population, and that the niacin trial information from the earlier times is valid.

I suspect the strength of HDL as a risk factor will be strongest in populations where statins are not in use and less predictive in populations where statins are being used.

Interestingly, what is thought to be an independent risk factor waxes and wanes. For many years triglycerides were thought to not be an independent risk factor and totally explained by HDL.

Your column was thought-provoking. Keep up the good work.


Eric Roehm, MD


1. Coronary Drug Project Research Group. JAMA. 1975;231:360-81.
2. Canner PL, Berge KG, Wenger NK, et al. J Am Coll Cardiol. 1986;8:1245-55.
3. Carlson LA, Rosenhamer G. Acta Med Scand. 1988;223:405-18.

Clinical Topics: Diabetes and Cardiometabolic Disease, Dyslipidemia, Prevention, Lipid Metabolism, Nonstatins

Keywords: Heart Diseases, Follow-Up Studies, Secondary Prevention, Risk Factors, Niacin, Lipoproteins, HDL, Triglycerides

< Back to Listings