"My LDL is low, so I'm probably fine." It's one of the most common assumptions people make after a cholesterol test, and for decades, most cardiovascular guidance reinforced it by focusing almost entirely on LDL. Framingham Heart Study data tells a more complete story, and it's one that HDL cholesterol deserves far more credit for.
The Study That Tested This Directly
Researchers analyzing the Framingham Offspring Study, following 2,439 men and 2,812 women over 20 years, set out to answer a specific question: does the relationship between HDL and heart disease risk depend on what your LDL or total cholesterol happens to be? If HDL only mattered when LDL was already high, that would suggest HDL isn't truly independent, just a modifier of LDL's effect.
The answer was clear. Risk of coronary heart disease rose two to threefold from the lowest to the highest total cholesterol to HDL ratio group, and this pattern held true regardless of whether LDL or total cholesterol was low, moderate, or high. In the researchers' own words, the ratio predicted heart disease risk equally well at low and high cholesterol values. Even within the group with the lowest LDL cholesterol, people with low HDL still carried meaningfully elevated risk.
What This Actually Means
This finding directly challenges the "low LDL means I'm safe" assumption. A person with LDL cholesterol of 100 mg/dL and HDL of 25 mg/dL does not carry the same risk as someone with LDL of 100 mg/dL and HDL of 85 mg/dL, even though their LDL numbers are identical. HDL is doing real, independent work here, not simply riding along with whatever LDL happens to be doing.
| LDL Level | HDL Level | General Pattern |
|---|---|---|
| Low (100 mg/dL) | High (85 mg/dL) | Lower risk |
| Low (100 mg/dL) | Low (25 mg/dL) | Meaningfully elevated risk despite low LDL |
| High (220 mg/dL) | High (85 mg/dL) | Risk reduced compared to low HDL at the same LDL |
| High (220 mg/dL) | Low (25 mg/dL) | Highest risk category |
This lines up with something we've mentioned in our guide to total cholesterol and expanded on in our article on the TC/HDL ratio: a single number in isolation, whether it's LDL or total cholesterol, tells an incomplete story without knowing where HDL stands.
The Classic Framingham Risk Chart
This chart comes from a classic presentation of Framingham Heart Study data published by Dr. William Castelli in 1988, and it shows the same pattern described above in a single picture. Along the bottom, LDL cholesterol climbs from 100 to 220 mg/dL. Running the other direction, HDL cholesterol climbs from 25 to 85 mg/dL. The height of each bar is the four year risk of developing coronary heart disease for someone with that specific combination.
Look at the two back corners of the chart. Someone with an LDL of 220 mg/dL and an HDL of only 25 mg/dL sits under the tallest, reddest bar, the highest risk group by far. Follow that same LDL of 220 mg/dL over to a high HDL of 85 mg/dL instead, and the bar drops noticeably. The same shift shows up even at the lowest LDL level of 100 mg/dL, where low HDL still pushes the bar higher than high HDL does.
That is the same independent effect described throughout this article, illustrated with actual Framingham risk figures instead of general categories. HDL is not simply riding along with LDL, it is changing the height of the bar on its own, at every LDL level on the chart.
An Older, Complementary Finding
This wasn't an isolated result. An earlier Framingham analysis following 2,748 participants for 12 years found that people in the lowest HDL group had roughly four times the risk of dying from coronary heart disease compared to those in the highest HDL group, an association that held up after adjusting for total cholesterol, blood pressure, weight, and smoking status. In other words, accounting for all of those other factors didn't make HDL's effect disappear. It remained a distinct, standalone signal.
An Important Distinction: Independent Risk Marker vs. Treatment Target
Here's where this needs to be handled carefully, because it's a genuinely easy point to misread. Finding that low HDL independently predicts risk is not the same as proving that artificially raising HDL with medication will lower that risk. These are two different scientific questions, observation versus intervention, and they don't always have the same answer.
This distinction became very real in clinical trials. In the AIM-HIGH trial, over 3,400 patients with existing cardiovascular disease and low HDL were given niacin, a medication known to raise HDL substantially, on top of statin therapy. HDL levels rose as expected. But the trial was stopped early because niacin produced no reduction in heart attacks, strokes, or cardiovascular death compared to placebo. A separate large trial testing a different HDL-raising approach reached a similar disappointing conclusion.
This doesn't erase the Framingham findings, low HDL remains a genuine, independent warning sign. It does mean the relationship is more complex than "just raise the number." Part of the explanation may lie in HDL's function rather than its quantity, a distinction we explore in our article on HDL cholesterol efflux. Simply pushing the number higher with certain medications doesn't necessarily restore the biological work that healthy HDL does.
Why Research Is Shifting Toward HDL Function, Not Just HDL Numbers
Given that niacin and similar quantity-focused drugs kept failing, researchers have spent the last decade pivoting toward a different question: instead of simply raising HDL cholesterol, can a treatment improve how well HDL actually functions, particularly its ability to pull cholesterol out of artery walls?
This led to trials of infused apolipoprotein A-I, HDL's main functional protein, designed to directly boost cholesterol efflux capacity rather than the HDL number itself. The largest of these, the AEGIS-II trial, tested an infused form of apoA-I called CSL112 in over 18,000 patients following a heart attack. The overall trial did not meet its primary goal of reducing cardiovascular events at 90 days. However, a closer look at the data offered a genuinely interesting clue: among patients who still had elevated LDL cholesterol at baseline, the treatment was associated with a meaningful reduction in cardiovascular death and repeat heart attacks that persisted out to a full year. Among patients whose LDL was already well controlled, no benefit appeared.
Researchers involved described this as an important signal rather than a dead end, since it suggests that improving HDL function may still matter, particularly in people who haven't yet gotten their other risk factors under control. This is exactly why continued investment in developing substances that genuinely improve HDL's function, not just its concentration, remains a meaningful research priority. Getting this right could eventually offer real protection for the many people whose low or poorly functioning HDL currently goes unaddressed simply because past attempts targeted the wrong variable.
Frequently Asked Questions
Q: If drugs that raise HDL don't reduce heart attacks, why does low HDL still matter?
Because low HDL is a genuine warning sign of underlying risk, even if the specific drugs tested so far haven't successfully reversed that risk by raising the number artificially. It's similar to how a fever can be a real, meaningful sign of infection even if simply lowering the temperature with medication doesn't cure the underlying illness.
Q: Does this mean low HDL can't be improved at all?
Not at all. Lifestyle approaches, including regular exercise, weight management, and not smoking, have consistently been associated with healthier HDL levels and function. The disappointing trial results are specific to certain medications, not to lifestyle change.
Q: Should I ask my doctor to focus more on my HDL number specifically?
It's worth discussing your full lipid panel, including HDL, as part of your overall cardiovascular risk conversation, rather than fixating on LDL alone. Your doctor can help interpret your HDL level in the context of your total risk profile.
Key Takeaway
Framingham Heart Study data shows that low HDL cholesterol predicts heart disease risk independently of LDL or total cholesterol, holding true even among people with low LDL who might otherwise assume they're in the clear. At the same time, clinical trials testing HDL-raising medications haven't shown the same benefit from pushing the number up artificially, a reminder that being a strong risk marker and being an effective treatment target are two different things. The practical takeaway isn't to obsess over a single HDL number, but to make sure it isn't overlooked entirely just because LDL looks fine.
References
- Nam BH, Kannel WB, D'Agostino RB. Search for an Optimal Atherogenic Lipid Risk Profile: From the Framingham Study. American Journal of Cardiology. 2006;97(3):372-375. doi.org/10.1016/j.amjcard.2005.08.055
- Wilson PWF, Abbott RD, Castelli WP. High Density Lipoprotein Cholesterol and Mortality: The Framingham Heart Study. Arteriosclerosis. 1988;8(6):737-741. doi.org/10.1161/01.ATV.8.6.737
- The AIM-HIGH Investigators. Niacin in Patients with Low HDL Cholesterol Levels Receiving Intensive Statin Therapy. New England Journal of Medicine. 2011;365(24):2255-2267. doi.org/10.1056/NEJMoa1107579
- Gibson CM, Duffy D, Korjian S, et al. Apolipoprotein A-I Infusions and Cardiovascular Outcomes in Acute Myocardial Infarction According to Baseline LDL Cholesterol Levels: The AEGIS-II Trial. Journal of the American College of Cardiology. 2024. doi.org/10.1016/j.jacc.2024.03.396
- Castelli WP. Cholesterol and Lipids in the Risk of Coronary Artery Disease: The Framingham Heart Study. Canadian Journal of Cardiology. 1988;4(Suppl A):5A-10A. PubMed
Comments
Post a Comment