HDL Cholesterol Unit Converter

Convert HDL cholesterol between mg/dL, mmol/L, mg/L and g/L using exactly the same factor as LDL and total cholesterol, because it is exactly the same molecule. Then read the answer against a threshold rather than a target: HDL-C is an input to risk estimation and is not something any guideline asks you to treat.

Last updated 18 Sep 2026

HDL cholesterol converter

mg/dL ⇄ mmol/L

HDL cholesterol

mg/dL × 0.0258632 = mmol/L. The same factor converts total, LDL and non-HDL cholesterol, because all four are cholesterol. It does not convert a triglyceride.

Read against the thresholds forBoth bounds are published, and neither is a treatment target. The lower bound is the sex-specific figure that counts a low HDL-C as a risk marker — it is an ATP III major risk factor at under 40 mg/dL and a metabolic-syndrome criterion at under 40 in men and under 50 in women. The upper bound is the concentration above which two Copenhagen cohorts found all-cause mortality rising again. The bar therefore reads high at both ends, which is the honest shape for this analyte and not the shape most readers expect.

1.24mmol/LExample

HDL cholesterol 48 mg/dL in a man

Formula, and why the factor is the same as LDL’s

mmol/L = mg/dL × 0.0258632
mg/dL = mmol/L × 38.665
because 1 mg/dL is 0.01 g/L, and 0.01 ÷ 386.65 g/mol = 2.58632 × 10⁻⁵ mol/L

0.0258632: the molar mass of cholesterol, 386.65 g/mol for C₂₇H₄₆O, divided into a milligram per decilitre. That is the whole derivation, and nothing about high-density lipoprotein enters into it
"HDL cholesterol": names the particle the cholesterol was riding on, not a different molecule. The laboratory measures cholesterol mass in the fraction that stays in solution after the apolipoprotein-B-containing particles are precipitated or blocked. The measurand is cholesterol, so the factor is cholesterol’s
same as LDL-C and total: total cholesterol, LDL-C, HDL-C and non-HDL cholesterol all share 0.0258632. There is no separate HDL conversion to remember, and a table that prints one is printing the same number twice
not triglycerides: a triglyceride on the same report needs 0.011294, from an assumed mass of 885.45 g/mol. Using the cholesterol factor on a triglyceride overstates it more than twofold
the mental check: mmol/L is roughly the mg/dL figure divided by 40. The pairs worth holding are 40 mg/dL ≈ 1.0, 50 ≈ 1.3, 60 ≈ 1.55 and 116 ≈ 3.0 mmol/L

Worked example

HDL cholesterol 48 mg/dL in a man
48 × 0.0258632 = 1.24 mmol/L
48 × 10 = 480 mg/L, and 48 × 0.01 = 0.480 g/L
Above the 40 mg/dL threshold that makes a low HDL-C a major risk factor, and a long way below the concentration at which mortality turns back up
No action follows from that. HDL-C is an input to the risk equation, not a target, so the next number to read is the LDL-C or the non-HDL cholesterol on the same report
For the measurements that do carry goals, see the LDL cholesterol converter and the non-HDL cholesterol calculator

One factor, four cholesterol measurements

The first four rows share a factor because they share a measurand. Readers reasonably expect an "HDL conversion factor" to be its own number, and it is not: HDL, LDL and VLDL are lipoprotein particles, and the thing being quantified in all three cases is the cholesterol they carry. The only row that needs a different factor is the last one, and getting those two confused is the one conversion error on a lipid profile with clinical consequences — the cholesterol factor applied to a triglyceride inflates it by more than twofold.
Measurement	What the laboratory measures	mg/dL to mmol/L factor
Total cholesterol	All cholesterol in the sample	0.0258632 — see the total cholesterol converter
LDL cholesterol	Cholesterol in the LDL fraction, usually calculated rather than measured	0.0258632 — see the LDL cholesterol converter
HDL cholesterol	Cholesterol in the fraction left after apoB-containing particles are removed	0.0258632 — this page
Non-HDL cholesterol	Total minus HDL, so cholesterol on every atherogenic particle	0.0258632 — see the non-HDL cholesterol calculator
Triglycerides	Triglyceride mass, a different molecule entirely	0.011294 — see the triglyceride converter

Where a low HDL-C actually appears in guidance

The last row is the sentence most often quoted from this subject and it should be read as history. It belongs to a 2001 framework that decided LDL goals by counting risk factors, and that framework no longer exists: the 2018 guideline replaced the count with a quantitative risk estimate plus a list of risk-enhancing factors, and HDL-C appears nowhere on that list. The 2026 guideline sets numerical goals for LDL-C and non-HDL cholesterol and none for HDL-C. What survives is the first three rows: a low HDL-C still counts against a patient, and HDL-C is still a required input to the risk equation. Subtracting a risk factor for a high one is not current practice and has not been for some years.
Framework	How HDL-C is used	Threshold
NCEP ATP III, 2001, major risk factors	Counted as one of five major risk factors modifying the LDL goal	"Low HDL cholesterol (<40 mg/dL)"
Metabolic syndrome criteria, carried into the 2018 AHA/ACC guideline	One of the five criteria, three of which define the syndrome	"reduced HDL-C, <40 mg/dL in men; <50 in women"
AHA PREVENT risk equations, 2023	A required input variable, alongside total cholesterol	Continuous — no cut-point
2018 AHA/ACC risk-enhancing factors	Not listed, in any form	—
2026 ACC/AHA dyslipidaemia guideline treatment goals	No HDL-C goal. The goals are "LDL-C <55 mg/dL (1.4 mmol/L) and non–HDL-C <85 mg/dL (2.2 mmol/L)" in secondary prevention	—
NCEP ATP III, 2001, negative risk factor	Historical. "HDL cholesterol >60 mg/dL counts as a ‘negative’ risk factor; its presence removes one risk factor from the total count"	>60 mg/dL

Every attempt to treat HDL-C, and what happened

This table is the reason the page does not tell you to raise your HDL-C. Two drugs raised it substantially and produced no benefit or net harm. The third is printed here as an exception rather than left out, because leaving it out would make the argument tidier than the evidence: anacetrapib doubled HDL-C and did reduce major coronary events by nine per cent. It also cut non-HDL cholesterol by eighteen per cent, which is a mechanism already known to work, so the trial cannot separate the two, and it was abandoned rather than licensed. The defensible reading is narrow and worth stating as such: no trial has shown that raising HDL-C, as distinct from lowering atherogenic lipoproteins, reduces events, which is why HDL-C is a risk variable and not a target.
Trial	What it did to HDL-C	What it did to outcomes
ILLUMINATE, torcetrapib plus atorvastatin, 15,067 patients	"an increase of 72.1% in high-density lipoprotein cholesterol", with LDL-C down 24.9%	"terminated prematurely because of an increased risk of death and cardiac events": cardiovascular events HR 1.25 (1.09–1.44), death from any cause HR 1.58 (1.14–2.19)
AIM-HIGH, niacin added to statin	"significantly increased the median HDL cholesterol level from 35 mg per deciliter (0.91 mmol per liter) to 42 mg per deciliter (1.08 mmol per liter)"	Primary endpoint HR 1.02 (0.87–1.21), P=0.79. "no incremental clinical benefit … despite significant improvements in HDL cholesterol and triglyceride levels"
REVEAL, anacetrapib — the exception	"increased HDL cholesterol levels by 104% (43 mg/dL [1.12 mmol/L])"	Major coronary events DID fall: rate ratio 0.91 (0.85–0.97), P=0.004. But it also "reduced the mean level of non-HDL cholesterol by 18% (17 mg/dL [0.44 mmol/L])", major atherosclerotic events missed significance at 0.93 (0.86–1.00), and the drug was never marketed

The top of the scale, where the association reverses

From the Copenhagen City Heart Study and the Copenhagen General Population Study, 52,268 men and 64,240 women, reporting "U-shaped" associations in both sexes. Two things follow for this page. The upper bound of each reference group above is taken from this study, so the bar under the converted number does not run off the top into an implied "better". And a very high HDL-C is a finding rather than a reassurance: the practical differential is alcohol intake, oestrogen or other drug effects, and rarely a genetic cause such as cholesteryl ester transfer protein deficiency. These are observational hazard ratios and do not establish that the HDL-C is what harms anyone; they establish that reading the top of this scale as protective is not supported.
HDL cholesterol	Men, all-cause mortality	Women, all-cause mortality
Lowest-risk concentration	1.9 mmol/L (about 73 mg/dL) — the reference	2.4 mmol/L (about 93 mg/dL) — the reference
2.5–2.99 mmol/L (97–115 mg/dL)	HR 1.36 (1.09–1.70)	—
3.0–3.49 mmol/L (116–134 mg/dL)	HR 2.06 (1.44–2.95) at ≥3.0	HR 1.10 (0.83–1.46)
3.5 mmol/L and above (≥135 mg/dL)	—	HR 1.68 (1.09–2.58)

The same molecule as LDL cholesterol, and the opposite kind of number

The conversion is the least interesting thing on this page, and it is worth being explicit about why. A milligram per decilitre of cholesterol is 0.01 g/L, and 0.01 divided by cholesterol’s molar mass of 386.65 g/mol gives 2.58632 × 10⁻⁵ mol/L, so the factor from mg/dL to mmol/L is 0.0258632 and the reciprocal is 38.665. That factor is identical for total cholesterol, for LDL cholesterol, for non-HDL cholesterol and for the number on this page, because all four are measurements of cholesterol. "HDL cholesterol" names the particle the cholesterol was carried on — a small, dense, apolipoprotein-A-1-containing lipoprotein — and not a chemically distinct substance. Readers arrive expecting a separate HDL factor and there is not one. The only factor on a lipid profile that differs is the triglyceride one, 0.011294, and confusing the two is the single conversion error here with consequences.

What makes HDL-C worth a page of its own is that it is the one lipid measurement on the report with no target attached, and the page is written so the bar under the converted figure does not imply otherwise. A low HDL-C is a real risk marker. ATP III counted under 40 mg/dL as one of five major risk factors; the metabolic syndrome criteria carried into current guidance use under 40 mg/dL in men and under 50 in women; and the AHA’s PREVENT equations require HDL-C as an input variable alongside total cholesterol. None of that makes it a thing to treat. The 2026 ACC/AHA dyslipidaemia guideline sets numerical goals for LDL-C and for non-HDL cholesterol and sets none for HDL-C, and the 2018 guideline’s list of risk-enhancing factors does not mention HDL-C at all.

The reason for that absence is a run of trials. Torcetrapib, a cholesteryl ester transfer protein inhibitor, raised HDL-C by 72 per cent in 15,067 patients and ILLUMINATE was stopped early for an increased risk of death and cardiac events, with hazard ratios of 1.25 for cardiovascular events and 1.58 for death from any cause. Niacin raised median HDL-C from 35 to 42 mg/dL on top of a statin in AIM-HIGH and returned a primary-endpoint hazard ratio of 1.02, the authors concluding that there was no incremental clinical benefit despite significant improvements in HDL cholesterol. The honest exception is REVEAL: anacetrapib roughly doubled HDL-C and did reduce major coronary events, with a rate ratio of 0.91. It also lowered non-HDL cholesterol by 18 per cent, which is a mechanism already known to reduce events, the secondary atherosclerotic endpoint missed significance, and the drug was never brought to market. So the claim this page will defend is the narrow one: no trial has shown that raising HDL-C, as distinct from lowering atherogenic lipoproteins, reduces events.

There is a second reason not to read the top of the scale as good news, and it is why both reference groups here have an upper bound. In the Copenhagen City Heart Study and the Copenhagen General Population Study, 52,268 men and 64,240 women, all-cause mortality described a U: lowest at about 1.9 mmol/L in men and 2.4 mmol/L in women, and rising again above those, to a hazard ratio of 2.06 in men at or above 3.0 mmol/L and 1.68 in women at or above 3.5 mmol/L. Those are observational figures and they do not show that a high HDL-C harms anybody. What they do show is that a report of 120 mg/dL is a finding to explain — alcohol, oestrogen or another drug effect, occasionally a cholesteryl ester transfer protein deficiency — and not a result to congratulate a patient on. The one sentence this page exists to displace is "higher is better".

One piece of ATP III does still get quoted and should be dated when it is. The 2001 at-a-glance states that HDL cholesterol above 60 mg/dL counts as a negative risk factor and that its presence removes one risk factor from the total count. That rule belonged to a framework which decided LDL goals by counting risk factors, and that framework was replaced in 2018 by quantitative risk estimation plus a defined list of risk-enhancing factors, on which HDL-C does not appear. Subtracting a risk factor for a high HDL-C is not current practice.

Practically, if you are holding a lipid profile: convert the numbers here, then spend your attention on the ones with goals. The LDL cholesterol converter and the non-HDL cholesterol calculator carry the treatment targets, the apoB to apoA-1 ratio calculator and the lipoprotein(a) converter cover the risk-enhancing measurements the 2026 guideline names, and the total cholesterol to HDL ratio calculator, LDL to HDL ratio calculator and triglyceride to HDL ratio calculator use the HDL-C from this page as a denominator. If the LDL-C on your report was calculated rather than measured, the Friedewald LDL calculator shows what that calculation depends on.

Frequently asked questions

Is the HDL cholesterol conversion factor different from the LDL one?

No. Both are 0.0258632 from mg/dL to mmol/L, and so are total cholesterol and non-HDL cholesterol. The factor comes from cholesterol’s molar mass of 386.65 g/mol, and all four measurements quantify cholesterol; HDL and LDL name the lipoprotein particles carrying it, not different molecules. The only lipid on the same report with a different factor is the triglyceride, which needs 0.011294.

What is a normal HDL cholesterol?

There is no reference interval in the usual sense, because the published figures are risk thresholds rather than the bounds of a healthy population. The sex-specific low thresholds are the ones in routine use: under 40 mg/dL (about 1.0 mmol/L) counts as a major risk factor in ATP III and as a metabolic-syndrome criterion in men, and under 50 mg/dL (about 1.3 mmol/L) is the metabolic-syndrome criterion in women. There is no published upper bound of normality, but there is a concentration above which mortality rises again — see the next answer.

Is a very high HDL cholesterol good?

The evidence does not support reading it that way. In two Copenhagen cohorts totalling 52,268 men and 64,240 women, all-cause mortality was lowest at about 1.9 mmol/L in men and 2.4 mmol/L in women and rose again above that, reaching a hazard ratio of 2.06 in men at or above 3.0 mmol/L (116 mg/dL) and 1.68 in women at or above 3.5 mmol/L (135 mg/dL). Those are observational associations rather than proof of harm, but a markedly high HDL-C is a result to explain — alcohol intake, oestrogen or another drug effect, rarely a cholesteryl ester transfer protein deficiency — not a result to celebrate.

Should I try to raise my HDL cholesterol?

No guideline asks you to, and the trial record explains why. Torcetrapib raised HDL-C by 72 per cent and ILLUMINATE was stopped early for excess death and cardiac events. Niacin raised median HDL-C from 35 to 42 mg/dL in AIM-HIGH with a primary-endpoint hazard ratio of 1.02 and no incremental benefit. Anacetrapib in REVEAL did reduce major coronary events, by nine per cent, but it also lowered non-HDL cholesterol by eighteen per cent and was never marketed, so it cannot show that the HDL-C rise was what helped. The measurements with treatment goals are LDL-C and non-HDL cholesterol.

Does HDL cholesterol above 60 mg/dL still remove a risk factor?

Not in current guidance. The sentence comes from the 2001 NCEP ATP III at-a-glance, which states that HDL cholesterol above 60 mg/dL counts as a negative risk factor and removes one risk factor from the total count. It belonged to a scheme that set LDL goals by counting risk factors; that scheme was replaced in 2018 by quantitative risk estimation with a defined list of risk-enhancing factors, and HDL-C does not appear on that list. The 2026 ACC/AHA dyslipidaemia guideline sets goals for LDL-C and non-HDL cholesterol and none for HDL-C.

If HDL-C is not a target, why does my laboratory report it?

Because it is still needed for risk estimation and for the derived numbers. The AHA PREVENT equations require HDL cholesterol as an input alongside total cholesterol, systolic pressure, BMI, eGFR and the rest. Non-HDL cholesterol, which does carry a treatment goal, is total cholesterol minus HDL-C and cannot be calculated without it. And the Friedewald and Sampson estimates of LDL-C both subtract HDL-C. It is a load-bearing measurement that is not itself a target.

Related calculators

References

National Cholesterol Education Program. ATP III At-A-Glance: Quick Desk Reference. NIH/NHLBI, 2001. HDL classification "<40 Low", ">60 High"; major risk factors include "Low HDL cholesterol (<40 mg/dL)"; "HDL cholesterol >60 mg/dL counts as a ‘negative’ risk factor; its presence removes one risk factor from the total count."
Grundy SM, Stone NJ, Bailey AL, et al. 2018 AHA/ACC/multisociety guideline on the management of blood cholesterol. Circulation. 2019;139(25):e1082–e1143. Metabolic syndrome criterion "reduced HDL-C, <40 mg/dL in men; <50 in women"; HDL-C does not appear in the risk-enhancing factors table.
2026 ACC/AHA/AACVPR/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of dyslipidemia. Circulation. 2026. "LDL-C and non–high-density lipoprotein cholesterol (HDL-C) treatment goals are back to guide LLT"; secondary prevention "a goal of LDL-C <55 mg/dL (1.4 mmol/L) and non–HDL-C <85 mg/dL (2.2 mmol/L) is recommended". No HDL-C goal is set.
Khan SS, Matsushita K, Sang Y, et al. Development and validation of the American Heart Association’s PREVENT equations. Circulation. 2024;149(6):430–449. Required inputs: "sex, age, total cholesterol, HDL cholesterol, systolic blood pressure, BMI, eGFR, diabetes, current smoking status, antihypertensive use, and lipid-lowering therapy use".
Barter P, Gotto AM, LaRosa JC, et al. Effects of torcetrapib in patients at high risk for coronary events (ILLUMINATE). N Engl J Med. 2007;357:2109–2122. "an increase of 72.1% in high-density lipoprotein cholesterol and a decrease of 24.9% in low-density lipoprotein cholesterol"; cardiovascular events HR 1.25 (1.09–1.44); death from any cause HR 1.58 (1.14–2.19); "terminated prematurely because of an increased risk of death and cardiac events".
AIM-HIGH Investigators. Niacin in patients with low HDL cholesterol levels receiving intensive statin therapy. N Engl J Med. 2011;365:2255–2267. HDL-C "from 35 mg per deciliter (0.91 mmol per liter) to 42 mg per deciliter (1.08 mmol per liter)"; primary endpoint HR 1.02 (0.87–1.21), P=0.79; "no incremental clinical benefit from the addition of niacin to statin therapy".
HPS3/TIMI55–REVEAL Collaborative Group. Effects of anacetrapib in patients with atherosclerotic vascular disease. N Engl J Med. 2017;377:1217–1227. HDL-C "increased … by 104% (43 mg/dL [1.12 mmol/L])"; non-HDL-C reduced "by 18% (17 mg/dL [0.44 mmol/L])"; major coronary events rate ratio 0.91 (0.85–0.97), P=0.004; major atherosclerotic events 0.93 (0.86–1.00), P=0.052.
Madsen CM, Varbo A, Nordestgaard BG. Extreme high high-density lipoprotein cholesterol is paradoxically associated with high mortality in men and women: two prospective cohort studies. Eur Heart J. 2017;38(32):2478–2486. 52,268 men and 64,240 women; men HR 2.06 (1.44–2.95) at HDL-C ≥3.0 mmol/L against a reference of 1.9 mmol/L; women HR 1.68 (1.09–2.58) at ≥3.5 mmol/L against 2.4 mmol/L; "U-shaped" associations in both sexes.

Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.