Reference
The Friedewald equation
The formula behind almost every LDL cholesterol result printed since 1972 — what each term is, where the divisors come from, and exactly where it stops working.
The formula
LDL cholesterol is not measured in a standard lipid panel. It is calculated from the three measurements that are made — total cholesterol, HDL cholesterol and triglycerides — by subtracting everything the equation believes is not LDL:
mg/dL LDL-C = TC - HDL-C - TG / 5 mmol/L LDL-C = TC - HDL-C - TG / 2.2
Read it as three steps. Total cholesterol counts the cholesterol in every lipoprotein particle. HDL cholesterol counts the cholesterol in the protective particles. What remains is cholesterol in the atherogenic particles — LDL, VLDL and their remnants. The division estimates the VLDL share and removes it, leaving LDL cholesterol.
Where the divisors come from
In fasting plasma, a VLDL particle carries roughly five times as much triglyceride as cholesterol by mass. That single observation is the divisor of 5: dividing the triglyceride concentration by 5 converts a triglyceride measurement into an estimate of VLDL cholesterol. In SI units the same ratio becomes a divisor of about 2.18, and laboratory practice rounds it to 2.2.
Because 2.2 is a rounded figure rather than an exact conversion, the two published forms are not mathematically equivalent. A panel worked in mg/dL with 5 and the same panel worked in mmol/L with 2.2 disagree by roughly 0.01 mmol/L on a typical result — about 1 mg/dL. Any calculator that computes in one unit and converts the answer for display will hide that small but real difference, which is why this site computes in whichever unit you type and says which one it used.
Worked example
TC 200 mg/dL, HDL-C 50 mg/dL, TG 150 mg/dL
LDL-C = 200 - 50 - 150 / 5
= 200 - 50 - 30
= 120 mg/dLThe same panel in SI units — 5.17, 1.29 and 1.69 mmol/L — gives 5.17 − 1.29 − 1.69/2.2 = 3.11 mmol/L. Converting the 120 mg/dL result across gives 3.10 mmol/L.
Validity
Where the equation stops working
Every one of these is a reason to distrust a calculated LDL cholesterol. The first is a hard limit published in the original paper; the rest are the conditions the divisor was never calibrated for.
Triglycerides 400 mg/dL (4.5 mmol/L) or more
The published limit. Above it, VLDL particles carry proportionally less cholesterol than the divisor of 5 assumes, so the equation underestimates LDL cholesterol. The error grows with triglyceride level and can reach tens of mg/dL.
Triglycerides below about 50 mg/dL (0.6 mmol/L)
The VLDL estimate collapses towards zero and becomes unstable, which can leave the calculated LDL cholesterol a few mg/dL too high.
A non-fasting sample
Triglycerides rise after eating, and every extra triglyceride molecule is read as VLDL cholesterol, so the LDL result is pushed down. The divisor assumes fasting plasma.
Dysbetalipoproteinaemia (type III) or chylomicronaemia
These conditions put triglyceride in particles the fixed divisor was never calibrated for — remnant particles in type III disease, chylomicrons after a fatty meal — so the VLDL estimate is wrong in both directions.
LDL cholesterol below about 70 mg/dL (1.8 mmol/L)
Not a validity limit but an accuracy one: when LDL is small, the fixed VLDL estimate carries a larger relative error. This is the specific gap the Martin–Hopkins adjustable factor was designed to close.
Alternatives
What to use instead
Three options exist when the Friedewald equation is not appropriate. None of them change what you should do about a result that looks wrong — that is a conversation for a clinician.
Martin–Hopkins (adjustable 180-cell) method
Replaces the fixed factor of 5 with an adjustable factor drawn from a 180-cell table indexed by triglyceride and non-HDL cholesterol levels. It stays accurate at triglyceride levels where the Friedewald equation is no longer valid.
Useful when: Non-fasting samples and triglycerides between 400 and 800 mg/dL (4.5–9 mmol/L).
Sampson (NIH) equation
A regression equation published by the NIH that is less biased than Friedewald at low LDL cholesterol and high triglyceride levels, and which has been adopted by some large laboratories.
Useful when: Low LDL cholesterol and/or raised triglycerides.
Direct measurement (homogeneous assay)
The laboratory measures LDL cholesterol directly instead of estimating it. Also the only way to report LDL cholesterol when triglycerides are above 400 mg/dL.
Useful when: Triglycerides above 400 mg/dL, or when an exact value is needed for treatment decisions.
Differences
Why your calculator and your lab report disagree
Small disagreements are normal and rarely mean anyone made a mistake. Four things explain almost all of them:
- Rounding order. A laboratory rounds the three inputs to whole mg/dL or two decimals of mmol/L before calculating, then rounds the answer. Doing the arithmetic on unrounded values moves a result by a milligram or two.
- A different equation. Laboratories that use Martin–Hopkins or Sampson report values that differ from Friedewald, most visibly at high triglycerides and at low LDL cholesterol.
- A direct assay. Some methods measure LDL cholesterol physically, so their result has no reason to match a calculated one — particularly above 400 mg/dL triglycerides.
- The unit path. As described above, the mg/dL and mmol/L forms differ by about 0.01 mmol/L on a typical panel. Compare like with like: a mg/dL result against a mg/dL result.
If the difference is more than a few mg/dL, check that all three inputs were transcribed correctly and that the samples were both fasting. Then ask the laboratory which method it used.
The rest of the panel
Four measurements, four pages: the composite that stays valid when triglycerides are high, the conversion factors, the full category chart, and the equation itself.
- Non-HDL cholesterol calculatorTotal cholesterol minus HDL cholesterol — the part of the panel that carries the atherogenic particles, and the measurement that stays valid when triglycerides are high.
- mg/dL to mmol/L converterConvert total cholesterol, LDL-C, HDL-C and triglycerides between mg/dL and mmol/L, using the separate factors each analyte actually needs.
- LDL cholesterol chartEvery LDL-C category from optimal to very high, in mg/dL and mmol/L side by side, with the treatment goals that sit underneath them.
References
- 1.Friedewald WT, Levy RI, Fredrickson DS. Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge. Clin Chem. 1972;18(6):499-502. PubMed (opens in a new tab)
- 2.Martin SS, Blaha MJ, Elshazly MB, et al. Comparison of a novel method vs the Friedewald equation for estimating low-density lipoprotein cholesterol levels from the standard lipid profile. JAMA. 2013;310(19):2061-2068. PubMed (opens in a new tab)
- 3.Sampson M, Ling C, Sun Q, et al. A new equation for calculation of low-density lipoprotein cholesterol in patients with normolipidemia and/or hypertriglyceridemia. JAMA Cardiol. 2020;5(5):540-548. PubMed (opens in a new tab)
- 4.National Cholesterol Education Program (NCEP) Expert Panel. Third Report of the NCEP Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults (Adult Treatment Panel III), final report. Circulation. 2002;106(25):3143-3421. PubMed (opens in a new tab)
Content last reviewed 16 September 2026.Every cut-point and formula on this page is taken from the sources above.
Run your panel through it.
The calculator on the home page applies exactly this equation, in the unit you choose, and tells you when the answer should not be trusted.