Biomarker
HDL Cholesterol (HDL-C)
Cholesterol carried by high-density lipoprotein particles, historically called protective, but pharmaceutical trials raising it have repeatedly failed to reduce cardiovascular events.
Summary Does raising HDL cholesterol actually protect against heart disease? Show / hide ↓
HDL-C is the amount of cholesterol carried by high-density lipoprotein particles, often called “good cholesterol.” These particles help transport cholesterol back to the liver, but the amount of HDL-C alone may not explain heart protection. A 2024 genetic study found only a modest link with lower coronary artery disease risk, about 16% lower odds, and this link mostly disappeared after considering other blood-fat markers called ApoA1 and ApoB. Several large trials of drugs that substantially raised HDL-C failed to reduce heart attacks or deaths, while a 2022 study linked very high HDL-C with higher death risk in some groups. A standard lipid panel measures HDL-C directly, but raising it by itself has not been shown to improve health outcomes.
What this means for you: Low HDL-C can signal higher overall cardiovascular risk, but raising HDL-C alone is not an established way to prevent heart disease. Focus on overall risk rather than buying something specifically to increase HDL-C.
conflicting evidenceHDL cholesterol (HDL-C) measures cholesterol carried by high-density lipoprotein particles, long taught as "good cholesterol" because HDL particles participate in reverse cholesterol transport, moving cholesterol away from artery walls back to the liver. The traditional claim: higher HDL-C protects against cardiovascular disease. That claim has weakened substantially over the past decade as genetic and drug-trial evidence failed to confirm HDL-C itself as a causal protective factor.
A 2024 Mendelian randomization meta-analysis of coronary artery disease found a modest inverse association between HDL lipoproteins and CAD (pooled odds ratio 0.84, 95% CI 0.72-0.98), but this protective signal largely disappeared when analyzed together with ApoA1 and ApoB, suggesting HDL-C's apparent protection may be substantially confounded by its correlation with other lipid measures rather than reflecting an independent causal effect [1]. Multiple large pharmaceutical trials of drugs designed specifically to raise HDL-C, including the CETP inhibitors torcetrapib, dalcetrapib, and evacetrapib, failed to reduce cardiovascular events despite substantially raising HDL-C, some of the strongest available evidence that raising HDL-C by pharmacological means does not translate into cardiovascular benefit [4]. Separately, a National Heart, Lung, and Blood Institute-highlighted 2022 study found that very high HDL-C levels were associated with increased, not decreased, mortality risk in some populations, prompting the observation that the relationship may be U-shaped rather than simply linear.
How to measure it: a standard lipid panel measures HDL-C directly via enzymatic assay, requiring no calculation, unlike LDL-C [3]. It is typically reported alongside total cholesterol, LDL-C, and triglycerides in one blood draw.
How to intervene on it: aerobic exercise, weight loss, smoking cessation, and moderate alcohol consumption modestly raise HDL-C in observational and small trial data, though none of these have been shown to improve cardiovascular outcomes specifically through the HDL-C-raising mechanism as opposed to their other effects [3]. Niacin raises HDL-C substantially but failed to reduce cardiovascular events in the AIM-HIGH and HPS2-THRIVE trials when added to statin therapy, reinforcing that the biomarker moving does not guarantee outcome benefit [5].
Critics within cardiology now largely treat isolated HDL-C as a weak, potentially non-causal risk marker, useful mainly as one input into overall risk calculators rather than as an independent treatment target, a position reinforced by the ApoB-focused reassessment of lipid risk markers [6]. HDL functionality, meaning how well HDL particles actually perform cholesterol efflux, may matter more than the raw cholesterol content, but functional HDL testing is not yet standardized or widely available clinically [2].
The plain takeaway: HDL-C is still used in risk calculators because low HDL-C correlates with elevated overall risk, but drug trials specifically designed to raise HDL-C have repeatedly failed to reduce heart attacks or deaths, so raising HDL-C by itself is not an established way to lower cardiovascular risk.
References
Every numbered citation in this entry links here. Each reference links out to the primary source.
-
[1]
HDL-CAD pooled OR 0.84 (95% CI 0.72-0.98), but protective effect diminished when adjusted for ApoA1/ApoB (both p>0.05).
-
[2]
Study challenges 'good cholesterol's' role in universally predicting heart disease risk Tier 3
NIH-funded study finding HDL-C's relationship with cardiovascular risk varies by race and may not be uniformly protective, including elevated risk signals at very high levels in some groups.
-
[3]
HDL cholesterol, cardiovascular disease, and mortality Tier 3
Review summarizing the weakening evidence base for HDL-C as an independent causal protective factor, including CETP inhibitor trial failures.
-
[4]
Effects of torcetrapib in patients at high risk for coronary events Tier 1
RCT halted early: torcetrapib substantially raised HDL-C but increased cardiovascular events and mortality, a landmark negative result for the HDL-raising hypothesis.
-
[5]
RCT: niacin raised HDL-C substantially but showed no incremental cardiovascular benefit when added to statin therapy; trial stopped early for futility.
-
[6]
Apolipoprotein B in cardiovascular risk assessment Tier 3
Contextualizes why guidelines increasingly favor ApoB or non-HDL-C over isolated HDL-C for risk assessment.
Further reading
Curated external sources for a deeper dive. External links open in a new tab.
Follow HDL Cholesterol (HDL-C) through the chain: the mechanism that moves it, the molecule that targets it, the products that dose it, and the trials that tested it.
See HDL Cholesterol (HDL-C) on the Longevity Map →