LDL Cholesterol: The Evidence Behind the Calculator

A 56-year-old man from Mangaluru comes for review six months after a coronary stent. He takes his statin every night. His lipid profile reads: total cholesterol 150 mg/dL, HDL 34 mg/dL, triglycerides 250 mg/dL. The report, calculated by the laboratory with the Friedewald equation, gives an LDL of 66 mg/dL. The intern writes “LDL below 70, at target” and moves on. This is an illustrative case, not a real patient.

By the end of this article you will know why that 66 may not be true, and what the same profile gives with the other two equations. You will also know when no calculated LDL should be trusted at all.

1. Three Equations From One Lipid Profile

“Wait, the LDL is calculated? I thought the lab measured it.” Most labs do not. They measure total cholesterol, HDL and triglycerides, and work the LDL out. All three equations in the tool subtract an estimate of VLDL cholesterol from the non-HDL cholesterol. They differ only in how that estimate is made. Here they are, with where each was derived:

Equation	LDL cholesterol, mg/dL	Derived from	Used here
Friedewald, 1972	TC − HDL − TG/5	448 patients with familial hyperlipoproteinaemia or their relatives	Below 400 mg/dL triglycerides, for comparison
Martin-Hopkins, 2013	TC − HDL − TG/factor, the factor (3.1 to 9.5) read from a 180-cell table of triglyceride and non-HDL strata	1,350,908 lipid profiles, cholesterol measured after ultracentrifugation	Below 400 mg/dL, leading
Sampson (NIH), 2020	TC/0.948 − HDL/0.971 − (TG/8.56 + TG × non-HDL/2140 − TG²/16100) − 9.44	8656 patients, 18,715 beta-quantification results	Up to 800 mg/dL; leading from 400

Look at the third column. Friedewald was worked out in 1972 from 448 patients with familial hyperlipoproteinaemia or their relatives. Martin-Hopkins and Sampson came from many thousands of measurements against a reference method. That difference in where they came from is most of the story.

2. Why Friedewald Reads Low

Friedewald estimates the VLDL cholesterol as TG/5. The factor of 5 fixes the ratio of triglyceride to VLDL cholesterol. In the Martin-Hopkins data the median ratio was 5.2, and it climbed as triglycerides rose, so at high triglycerides TG/5 overstates the VLDL and the LDL comes out low.

“How much does it matter?” Most when the LDL itself is low, which is exactly the patient on treatment. The same error in the VLDL is a larger share of a small LDL. Martin and colleagues asked a sharp question: when an equation says the LDL is below 70, how often is it really below 70?

Triglycerides	Estimate below 70 confirmed below 70: Martin-Hopkins	Friedewald
100 to 149 mg/dL	94.3%	79.9%
150 to 199 mg/dL	92.4%	61.3%
200 to 399 mg/dL	84.0%	40.3%

Read the last row slowly. With triglycerides of 200 to 399 mg/dL, a Friedewald LDL below 70 was confirmed below 70 only 40.3% of the time. Our patient’s triglycerides are 250.

3. Our Patient, Three Ways

His non-HDL cholesterol is 150 − 34 = 116 mg/dL. Now the three equations, worked from the formulas in the table above:

Equation	Working	LDL, mg/dL
Friedewald	150 − 34 − 250/5	66
Martin-Hopkins	150 − 34 − 250/7.6 (the factor for triglycerides 248 to 292 and non-HDL 100 to 129)	83
Sampson (NIH)	The formula in the table, with TG 250 and non-HDL 116	75

So the “at target” LDL of 66 is probably an LDL in the 80s. Below 400 mg/dL of triglycerides, the tool leads with Martin-Hopkins. The National Lipid Association recommends it across the LDL range up to 399 mg/dL, and the 2026 ACC/AHA guideline prefers Martin-Hopkins or Sampson over Friedewald and over direct assays.

4. Does It Hold in Indian Patients?

A fair question, since none of these equations was derived in India. One Indian study has looked:

5. Validation in Indian Patients

In 3028 Indian patients undergoing coronary revascularisation, all three equations read below the direct LDL. Martin-Hopkins fell short by the least: a mean of 5.2 mg/dL, against 7.2 for Sampson and 10.5 for Friedewald. Friedewald placed 24.6 per cent of those with a direct LDL above 70 mg/dL below 70. The reference in such studies is a direct assay, not ultracentrifugation, so part of each difference belongs to the assay.

Notice the last sentence. In this study, the comparison was against a direct assay, so it cannot settle which equation is truest. But the direction is the same as in the American data: Friedewald reads lowest, and it moves people below 70 who are not there.

5. When No Calculated LDL Should Be Trusted

Some profiles defeat every equation. Above 800 mg/dL of triglycerides, none of the three is valid, and the non-HDL cholesterol is the number to use. From 400 to 800 mg/dL, the tool uses Sampson, the only one of the three reported accurate to 800. It marks the result as a wider estimate, because the National Lipid Association recommends no estimating equation at 400 or more. In type III dysbetalipoproteinaemia, Martin-Hopkins and Friedewald are furthest from the measured value; it needs an apolipoprotein B to recognise. And a direct assay is not the answer at high triglycerides either, because its accuracy also falls as they rise.

“Then what is safe?” The non-HDL cholesterol. It is total minus HDL, needs no VLDL estimate, and is reliable fasting or not. The Lipid Association of India treats it as a co-primary target beside LDL.

6. The Common Mistake

The common mistake is to read the LDL on the report as a measurement. It is easy to see why: it sits in the same column as the measured values, with the same units. But it is an estimate, and the Friedewald estimate is lowest exactly where treatment decisions are made: a treated patient, a low LDL, raised triglycerides. Look at the triglycerides before you believe the LDL.

Back to Our Patient

His Friedewald LDL of 66 becomes 83 by Martin-Hopkins and 75 by Sampson. He is probably not at a target of 70. The decision about stepping up his lipid-lowering therapy should rest on the Martin-Hopkins value, not on the 66. His non-HDL cholesterol of 116 is a second check that does not depend on any estimate.

Try This

A woman’s fasting profile shows triglycerides of 620 mg/dL. The laboratory prints a Friedewald LDL. What do you do with it? Illustrative case.

Answer. Set it aside. Friedewald is not used at 400 or more. Sampson is the only one of the three reported accurate up to 800, and even then the result is a wider estimate. Use her non-HDL cholesterol, and look for a cause of the high triglycerides.

Take-Home Points

• The LDL on most reports is calculated, not measured.
• Friedewald reads low when the triglycerides are high, and the error matters most when the LDL is low.
• Below 400 mg/dL of triglycerides, lead with Martin-Hopkins; from 400 to 800, Sampson gives only a wider estimate.
• Above 800 mg/dL, trust no calculated LDL: use the non-HDL cholesterol.
• Non-HDL cholesterol needs no estimate and is reliable fasting or not.

For what the statins themselves do, and the questions patients ask about them, see Cholesterol-Lowering Statin Medications: Do You Know the Real Story?; for the coronary side, Acute Coronary Syndrome: The Evidence Behind the Pathway.

Declaration: the author built and maintains the tool. It is free to use. The figures in this article are those the tool carried in its Evidence tab on 5 November 2026, each from the source the tool names. The illustrative cases follow the tool’s formulas; check the tool’s result for a real patient.

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.
2. Martin SS, Blaha MJ, Elshazly MB, Toth PP, Kwiterovich PO, Blumenthal RS, 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):2061-8.
3. Sampson M, Ling C, Sun Q, Harb R, Ashmaig M, Warnick R, et al. A new equation for calculation of low-density lipoprotein cholesterol in patients with normolipidemia and/or hypertriglyceridemia. JAMA Cardiol. 2020;5(5):540-8.
4. Martin SS, Giugliano RP, Murphy SA, Wasserman SM, Stein EA, Ceška R, et al. Comparison of low-density lipoprotein cholesterol assessment by Martin/Hopkins estimation, Friedewald estimation, and preparative ultracentrifugation: insights from the FOURIER trial. JAMA Cardiol. 2018;3(8):749-53.
5. Ginsberg HN, Rosenson RS, Hovingh GK, Letierce A, Samuel R, Poulouin Y, et al. LDL-C calculated by Friedewald, Martin-Hopkins, or NIH equation 2 versus beta-quantification: pooled alirocumab trials. J Lipid Res. 2022;63(1):100148.
6. Wilson PWF, Jacobson TA, Martin SS, Jackson EJ, Le NA, Davidson MH, et al. Lipid measurements in the management of cardiovascular diseases: practical recommendations a scientific statement from the national lipid association writing group. J Clin Lipidol. 2021;15(5):629-48.
7. Nordestgaard BG, Langsted A, Mora S, Kolovou G, Baum H, Bruckert E, et al. Fasting is not routinely required for determination of a lipid profile: clinical and laboratory implications including flagging at desirable concentration cut-points. A joint consensus statement from the European Atherosclerosis Society and European Federation of Clinical Chemistry and Laboratory Medicine. Eur Heart J. 2016;37(25):1944-58.
8. Blumenthal RS, Morris PB, Gaudino M, Johnson HM, Anderson TS, Bittner VA, et al. 2026 ACC/AHA/AACVPR/ABC/ACPM/ADA/AGS/APhA/ASPC/NLA/PCNA guideline on the management of dyslipidemia: a report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation. 2026;153(17):e1154-e1276.
9. Puri R, Mehta V, Bansal M, Shetty S, Yusuf J, Agarwala R, et al. Navigating cardiovascular risk and lipid management in Indian patients: key messages from the Lipid Association of India 2024 Consensus Statement IV. J Assoc Physicians India. 2024;72(8):80-2.
10. Bansal M, Kaushal P, Kasliwal RR, Chandra P, Kapoor R, Chouhan N, et al. Different methods of low-density lipoprotein cholesterol estimation and the impact on lipid-lowering therapy in patients with coronary artery disease. J Assoc Physicians India. 2026;74(2):57-61.