HDL is commonly called the good cholesterol and LDL the bad cholesterol. That shorthand is useful but incomplete. HDL carries cholesterol away from arteries, while LDL delivers it to tissues and can deposit it in artery walls when there is too much. The clinical picture depends on particle number, size, and context, not just the label.
What LDL and HDL actually do
Cholesterol is a waxy lipid that cells need for membrane structure and hormone synthesis. It cannot dissolve in blood, so it travels inside lipoproteins. These are particles with a lipid core and a protein shell. The shell proteins determine where the particle goes and what receptors recognize it.
LDL stands for low-density lipoprotein. It is the end product of a delivery chain. The liver secretes very-low-density lipoprotein (VLDL), which offloads triglycerides to muscle and fat. As VLDL shrinks, it becomes intermediate-density lipoprotein, then LDL. LDL's job is to deliver cholesterol to peripheral tissues. Cells pull LDL in through the LDL receptor. When receptors are scarce or LDL particles are excessive, LDL lingers in the bloodstream and can cross into the arterial intima.
HDL stands for high-density lipoprotein. It is assembled in the liver and intestine as lipid-poor particles. HDL can accept cholesterol from macrophages in artery walls through a transporter called ABCA1. This process is called reverse cholesterol transport. HDL then hands cholesterol to the liver, either directly or via other lipoproteins. In plain terms, LDL is a delivery truck and HDL is a pickup service.
The mechanistic caveat matters. HDL's cholesterol-carrying capacity is not the same as its function. Some HDL particles become dysfunctional in chronic inflammation and stop promoting efflux. That is one reason raising HDL cholesterol has not reliably lowered heart attack risk in trials.
Why the "good" label oversimplifies biology
The good-versus-bad framing comes from epidemiology. High LDL cholesterol associates with higher atherosclerotic cardiovascular disease (ASCVD) risk. High HDL cholesterol associates with lower risk in observational studies. But association is not the same as causation.
Mendelian randomization studies test causality using genetic variants. Variants that raise LDL cholesterol from birth increase ASCVD risk. Variants that raise HDL cholesterol do not consistently lower risk. This is a key distinction. LDL appears to be a causal driver; HDL is more often a marker of metabolic health.
HDL also behaves differently across ranges. Very high HDL cholesterol, above roughly 80 to 100 mg/dL, has been linked in some cohorts to increased risk. The mechanism is debated. Possibilities include dysfunctional large HDL particles, impaired cholesterol efflux capacity, or confounding by other conditions. The takeaway is not that high HDL is dangerous. It is that HDL is not a simple protective currency.
LDL is also heterogeneous. Small, dense LDL particles may be more atherogenic because they penetrate the artery wall more readily and bind less tightly to liver receptors. Particle number, often estimated by apolipoprotein B (apoB), can matter more than LDL cholesterol mass. Two people can share an LDL cholesterol of 130 mg/dL but have very different particle counts.
The numbers that shape risk
US lipid panels report cholesterol in milligrams per deciliter (mg/dL). LDL cholesterol is usually calculated using the Friedewald equation when triglycerides are below 400 mg/dL. Above that threshold, direct measurement or a different equation is needed.
For adults without established ASCVD, an LDL cholesterol below 100 mg/dL is often described as optimal. A value of 100 to 129 mg/dL is near-optimal. Values of 130 to 159 mg/dL are borderline high. Values of 160 to 189 mg/dL are high. Values of 190 mg/dL and above are very high and warrant evaluation for familial hypercholesterolemia.
HDL cholesterol thresholds are less tidy. In men, below 40 mg/dL is considered low. In women, below 50 mg/dL is considered low. Higher values are generally favorable, but there is no proven target to raise HDL for prevention.
Triglycerides add context. A triglyceride level below 150 mg/dL is normal. Values of 150 to 499 mg/dL are elevated. Values of 500 mg/dL and above raise pancreatitis risk and require separate attention. High triglycerides often accompany low HDL and small LDL particles, a pattern called atherogenic dyslipidemia.
Non-HDL cholesterol is LDL cholesterol plus all other apoB-containing lipoproteins. It captures VLDL remnants that LDL cholesterol alone misses. A non-HDL cholesterol below 130 mg/dL is a common secondary target for people at moderate risk.
A comparison of the two lipoproteins
| Feature | LDL | HDL |
|---|---|---|
| Primary role | Delivers cholesterol to tissues | Removes cholesterol from tissues |
| Direction of transport | Liver to periphery | Periphery to liver |
| Main structural protein | Apolipoprotein B-100 | Apolipoprotein A-I |
| Standard range framing | Lower is generally better | Higher is generally better, with caveats |
| Common threshold | Below 100 mg/dL optimal in average-risk adults | Below 40 mg/dL (men) or 50 mg/dL (women) is low |
| Causal evidence for ASCVD | Strong | Weak for raising HDL as a strategy |
| Therapeutic implication | Lowering reduces events | Raising has not reliably reduced events |
The table shows why the labels persist and where they break down. LDL is the more actionable target in most clinical settings. HDL is a useful risk marker but a poor therapeutic target on its own.
Edge cases where the labels fail
Familial hypercholesterolemia is an inherited condition affecting LDL receptor function. Untreated LDL cholesterol can reach 190 to 400 mg/dL or higher from childhood. The cumulative exposure to LDL drives early atherosclerosis. In this setting, LDL is not just a marker; it is the direct cause of disease.
Low HDL is often a signal rather than a standalone problem. It clusters with insulin resistance, central obesity, high triglycerides, and physical inactivity. Treating the low HDL number in isolation misses the underlying metabolic issue. The better approach is to address the drivers.
Some people have genetic HDL deficiency with very low HDL cholesterol but no clear increase in ASCVD. Examples include Tangier disease and apoA-I deficiency. These rare conditions show that HDL cholesterol level alone does not determine risk.
Chronic kidney disease, diabetes, and inflammatory states can produce dysfunctional HDL. The particle is present, but its efflux capacity is impaired. Standard HDL cholesterol tests do not capture this. This is a professional caveat worth stating plainly: a normal HDL number does not guarantee normal HDL function.
Common misconceptions and what replaces them
Misconception one: LDL is always bad. LDL is essential for cell function. The problem is excess and retention in the artery wall, not the molecule itself. Very low LDL achieved with medication has a strong safety record in large trials, but the framing should be about risk reduction, not moral categories.
Misconception two: raising HDL prevents heart attacks. Trials of niacin and CETP inhibitors raised HDL cholesterol but did not consistently reduce cardiovascular events. Some trials showed harm. The lesson is that HDL cholesterol is not a validated treatment target for most people.
Misconception three: a normal total cholesterol means normal risk. Total cholesterol includes HDL, so a high HDL can mask a high LDL. A total cholesterol of 200 mg/dL with an HDL of 70 mg/dL implies an LDL near 110 mg/dL, but the same total with an HDL of 35 mg/dL implies a much higher LDL. Always look at the components.
Misconception four: diet alone explains cholesterol. Saturated fat intake raises LDL in many people, but genetics, thyroid status, medications, and weight change also matter. A person with a strong family history may need medication regardless of diet.
Misconception five: the reference range on a lab report means "healthy for you." Reference ranges are population-based. A person with diabetes or prior heart attack needs a lower LDL than an average healthy adult.
Practical actions and when to act
The first step is to get a full lipid panel, not just total cholesterol. Ask for LDL cholesterol, HDL cholesterol, triglycerides, and non-HDL cholesterol. If triglycerides are above 400 mg/dL, ask how LDL was determined.
The second step is to estimate your own risk. For adults aged 40 to 75 without ASCVD, a 10-year risk estimate guides whether statin therapy is indicated. Risk calculators use age, blood pressure, cholesterol values, diabetes status, and smoking.
The third step is to reconcile thresholds explicitly. A primary prevention target of LDL below 100 mg/dL may be appropriate for average risk. For someone with diabetes and multiple risk factors, a target below 70 mg/dL is often used. For established ASCVD, a target below 55 mg/dL is recommended by some guidelines. These are not contradictory; they reflect different absolute risk levels.
The fourth step is to address lifestyle drivers. Reducing saturated fat, increasing soluble fiber, and losing excess weight can lower LDL. Aerobic exercise and weight loss can raise HDL modestly and lower triglycerides. The effect sizes vary by person.
The fifth step is to know the red flags that warrant prompt medical evaluation. LDL cholesterol of 190 mg/dL or higher at any age. Chest pain, shortness of breath, or symptoms of a heart attack or stroke. Suspected familial hypercholesterolemia, such as early heart disease in a close relative. Triglycerides of 500 mg/dL or higher. New low HDL with unexplained weight loss or systemic symptoms.
Limitations and what the evidence cannot settle
The evidence strongly supports lowering LDL to reduce ASCVD risk in high-risk people. The evidence does not support raising HDL as a standalone prevention strategy. That asymmetry is the core answer to the keyword question.
Several gaps remain. HDL function tests are not widely available in routine care. ApoB testing is increasingly used but not universal. Lipoprotein(a) is an independent risk factor that standard panels miss. The optimal LDL target for very old adults is uncertain. The long-term effects of very low LDL achieved without medication are less well studied than medication-induced lowering.
This article is general educational information, not a diagnosis or treatment plan. Cholesterol management depends on your full medical history, medications, and family history. Discuss your lipid panel and risk estimate with a qualified local clinician before starting or stopping any therapy.