Reading a lipid panel without a doctor in the room
1. Pick the units your report uses
mg/dL in the United States, India and much of Asia; mmol/L in the UK, Europe, Canada and Australia. Getting this wrong is the single easiest way to alarm yourself unnecessarily — a total cholesterol of 5.2 mmol/L is 200 mg/dL, and reading one as the other is the difference between fine and frightening.
2. Enter total cholesterol and HDL
These two alone give you the ratio, which is what this tool is named for. Both should come from the same blood draw — mixing values from panels taken months apart produces a ratio that describes nobody.
3. Add triglycerides if your report has them
With triglycerides the tool can also estimate LDL and calculate non-HDL cholesterol, which most guidelines now regard as more useful than LDL alone. Nearly every standard panel reports triglycerides, so it is usually just a matter of finding the line.
4. Choose your sex
The HDL thresholds are genuinely sex-specific — the low-HDL cut-off is 40 mg/dL for men and 50 for women, because women typically run higher and the same absolute figure carries different meaning. A tool that applies one number to everyone is misreading half its visitors.
5. Read the ratio alongside the individual values
The ratio is a useful summary and a poor substitute for the components. It is possible to reach an excellent ratio with an unremarkable HDL and a low total, or with a high HDL masking an LDL that still needs attention. Both are shown, deliberately.
6. Compare the two LDL estimates
When your triglycerides are ordinary the two methods agree closely and you can stop reading. When they diverge, that gap is the information — and at high triglycerides it can be the difference between two LDL categories, which is exactly when it matters most.
What the ratio is, and what it is not
The total cholesterol to HDL ratio is total cholesterol divided by HDL cholesterol. A total of 200 mg/dL with an HDL of 50 gives a ratio of 4.0.
Its appeal is that it captures a relationship rather than a level. Cholesterol is carried around the body by lipoproteins, and the ones doing the carrying behave very differently. Low-density lipoprotein deposits cholesterol into artery walls; high-density lipoprotein carries it back to the liver. Two people with identical total cholesterol can be in very different positions depending on how it is distributed, and the ratio is a compact way of expressing that.
Commonly cited interpretation puts a ratio below 3.5 as favourable, around 4.5 as average, and above 5 as unfavourable, with lower generally better. Those figures are worth treating as landmarks rather than thresholds, because no major guideline now treats the ratio as a treatment target in its own right.
What has replaced it in guideline thinking is non-HDL cholesterol — total cholesterol minus HDL. That single subtraction captures every cholesterol-carrying particle that can contribute to atherosclerosis: LDL, VLDL, IDL and lipoprotein(a). It does not require fasting, it does not require an estimate, and it predicts risk at least as well as LDL. If you take one number from this page other than the ratio, take that one.
Why there are two LDL estimates, and when they part company
Most laboratories do not measure LDL directly. Direct measurement requires ultracentrifugation, which is slow and expensive, so LDL on a routine report is almost always calculated.
The standard method is the Friedewald equation, published by Friedewald, Levy and Fredrickson in Clinical Chemistry in June 1972 and still in daily use more than fifty years later: LDL = total cholesterol − HDL − (triglycerides ÷ 5) in mg/dL. The division by five is an estimate of VLDL cholesterol, and it is a fixed assumption.
That assumption is where the trouble lives. The ratio of triglycerides to VLDL cholesterol is not constant — it changes as triglycerides rise and as LDL falls. Friedewald becomes progressively less reliable above roughly 200 mg/dL of triglycerides and is generally regarded as invalid above 400, which is why the previous version of this page printed a dash there and told you to ask for a direct measurement. Honest, and not much help.
The Sampson equation, published by Sampson and colleagues in JAMA Cardiology in May 2020, was developed at the National Institutes of Health specifically to address this. It is a closed-form equation validated against direct measurement up to 800 mg/dL of triglycerides, and it performs notably better than Friedewald at low LDL levels — increasingly relevant as more people are treated to lower targets.
A third method is worth knowing about because your own laboratory may be using it. Martin and colleagues published an alternative in JAMA on 20 November 2013 which replaces the fixed divisor of five with a value looked up from a table of 180 cells, chosen according to your own triglyceride and non-HDL levels. Many laboratories have adopted it, so if a reported LDL differs from what this page calculates, a different estimation method is the most likely reason rather than an error by either party.
The tool shows both. At ordinary triglyceride levels they agree within about one to three mg/dL and the distinction is academic. As triglycerides climb they separate: on a panel of total 200, HDL 60, triglycerides 400, Friedewald returns an LDL of 60 and Sampson returns 77. Those two figures fall in different LDL categories. That is not a rounding difference, and hiding it behind a single number would be the wrong call.
Reading each value on its own
The bands below are the ones in general clinical use, set out in the Third Report of the National Cholesterol Education Program Expert Panel — ATP III — published in JAMA on 16 May 2001 and still the reference most laboratory reports are printed against.
Total cholesterol — desirable below 200 mg/dL (5.2 mmol/L), borderline 200 to 239, high at 240 and above. On its own it is the least informative number on the panel, because it does not distinguish the particles doing harm from the ones doing good.
LDL cholesterol — optimal below 100 mg/dL (2.6 mmol/L), near-optimal 100 to 129, borderline high 130 to 159, high 160 to 189, very high at 190 and above. The near-optimal band was missing from the previous version’s table, so a very common result fell between two printed rows.
HDL cholesterol — low below 40 mg/dL for men and below 50 for women; 60 mg/dL (1.6 mmol/L) and above is generally regarded as protective. Worth knowing that very high HDL, above roughly 90, is no longer assumed to be straightforwardly beneficial; recent evidence suggests the relationship is U-shaped rather than linear.
Triglycerides — normal below 150 mg/dL (1.7 mmol/L), borderline 150 to 199, high 200 to 499, very high at 500 and above, where the risk of pancreatitis becomes a concern in its own right.
Non-HDL cholesterol — the usual target is 30 mg/dL above whatever the LDL goal is, so below 130 for someone whose LDL goal is 100.
Why your target is not on this page
This is the part most cholesterol calculators get wrong by omission. There is no single LDL target that applies to everyone, and any page that gives you one is not telling you the truth.
Modern guidelines set targets by overall cardiovascular risk, not by the lipid panel alone. Someone with no risk factors, normal blood pressure and no family history is treated very differently from someone who has already had a heart attack, or who has diabetes, chronic kidney disease or familial hypercholesterolaemia. For the highest-risk groups, targets below 70 mg/dL and in some cases below 55 are used. For someone at low risk the same LDL figure may need no intervention at all.
Risk is estimated with tools that take in age, sex, smoking status, blood pressure, diabetes, and often ethnicity and family history — the pooled cohort equations in the United States, QRISK3 in the UK, SCORE2 in Europe. None of that can be derived from four numbers on a lab report, which is why this page shows you where your values sit and stops there.
Getting the blood test right in the first place
Fasting used to be standard and is now often not required. Total cholesterol, HDL and non-HDL are all reliable non-fasting. Triglycerides are the value that fasting affects most, so if your triglycerides are being watched, or an LDL calculation matters, a 9 to 12 hour fast is still commonly requested. Follow whatever your own clinic asks for rather than this page.
Several things move a panel temporarily and are worth avoiding beforehand: acute illness or infection, which can lower cholesterol for weeks; recent surgery; alcohol in the preceding day or two, which raises triglycerides substantially; and pregnancy, during which lipid levels rise markedly and a routine panel is not interpretable against ordinary thresholds. Standing versus lying down during the draw can change values by a few per cent, and a tourniquet left on too long can too.
What this page is not
It is arithmetic on numbers you have already been given, and nothing on it is a diagnosis, a treatment recommendation or medical advice. It cannot tell you whether you need a statin, whether your result is good for someone in your circumstances, or what your target should be. Those questions require your full risk profile and belong with a doctor.
If your results are outside the ranges shown, the useful next step is a conversation with whoever ordered the test, not a decision made here. If your triglycerides are above 500, that warrants prompt medical attention because of pancreatitis risk. And a very high LDL — particularly with a family history of early heart disease — can indicate familial hypercholesterolaemia, an inherited condition that is common, underdiagnosed and very treatable once identified.
Nothing you enter leaves your browser
Your lipid values are used locally to compute the ratio and the estimates. Nothing is transmitted, stored or logged. There is no account and no upload, and the page keeps working with the connection off once loaded — which for medical results is the only sensible arrangement.
Questions people ask about cholesterol numbers
What is a good cholesterol ratio?
Below 3.5 is generally regarded as favourable, around 4.5 as average, above 5 as unfavourable. Treat those as landmarks: no major guideline uses the ratio as a treatment target, and non-HDL cholesterol is the more useful summary figure.
Why do my two LDL figures differ?
Because they use different assumptions about VLDL. Friedewald divides triglycerides by a fixed five; Sampson does not. They agree closely at normal triglycerides and diverge as triglycerides rise or LDL falls, which is precisely when the difference matters.
Which LDL estimate should I trust?
At ordinary triglycerides, either. Above roughly 200 mg/dL of triglycerides, Sampson is the better-validated of the two. Above 400, Friedewald should not be used at all — and if the number is going to change your treatment, a directly measured LDL is worth asking for.
Is high HDL always good?
Up to a point. Above roughly 60 mg/dL is generally regarded as protective, but very high levels are no longer assumed to be better still — the relationship appears U-shaped rather than linear.
Do I need to fast before a cholesterol test?
Often not. Total, HDL and non-HDL are reliable non-fasting. Triglycerides are the value most affected, so fasting is still commonly requested when they matter. Follow your own clinic’s instruction.
What is non-HDL cholesterol and why does it matter?
Total cholesterol minus HDL. It captures every atherogenic particle in one number, needs no estimate and no fasting, and predicts risk at least as well as LDL. Guidelines have moved toward it for exactly those reasons.
My total cholesterol is high but my ratio is good. Which counts?
Neither on its own. That combination usually means a high HDL is lifting the total, which is generally reassuring — but your LDL and non-HDL still need reading, and your overall risk still governs what anyone would do about it.
Can I lower cholesterol without medication?
Diet, physical activity, weight loss, stopping smoking and reducing alcohol all move lipid values, sometimes substantially, and are worth doing regardless. Whether that is sufficient depends entirely on your starting point and your overall risk, and it is a decision to make with a doctor rather than instead of one.
Do you store my results?
No. Nothing is saved or transmitted.
Sources behind these numbers
- Friedewald WT, Levy RI, Fredrickson DS, “Estimation of the concentration of low-density lipoprotein cholesterol in plasma, without use of the preparative ultracentrifuge”, Clinical Chemistry, June 1972, 18(6):499–502 (PMID 4337382).
- Sampson M et al, “A New Equation for Calculation of Low-Density Lipoprotein Cholesterol in Patients With Normolipidemia and/or Hypertriglyceridemia”, JAMA Cardiology, May 2020, 5(5):540–548 (PMID 32101259) — validated to 800 mg/dL of triglycerides.
- Expert Panel on Detection, Evaluation, and Treatment of High Blood Cholesterol in Adults, “Executive Summary of the Third Report of the National Cholesterol Education Program (NCEP) Expert Panel (Adult Treatment Panel III)”, JAMA, 16 May 2001, 285(19):2486–2497 (PMID 11368702) — the reference bands used on this page.
- Martin SS et al, “Comparison of a novel method vs the Friedewald equation for estimating low-density lipoprotein cholesterol levels from the standard lipid profile”, JAMA, 20 November 2013, 310(19):2061–2068 (PMID 24240933) — the adjustable-divisor method many laboratories now use.
- NHLBI, National Institutes of Health — blood cholesterol, for the reference ranges used here.
- American Heart Association — about cholesterol.
- American College of Cardiology, for the risk-based framing of targets.