Non-HDL cholesterol = total cholesterol − HDL cholesterol.
Non-HDL cholesterol is read against LDL cholesterol plus a margin. The ATP III report set non-HDL goals 30 mg/dL above the matching LDL goals, because it treated VLDL cholesterol of 30 mg/dL or less as normal. So where a clinician has an LDL goal in mind, the non-HDL goal sits 30 mg/dL higher.
The 2018 AHA/ACC cholesterol guideline lists non-HDL cholesterol of 190 to 219 mg/dL among its risk-enhancing factors. These are features that can tip a borderline or intermediate 10-year risk estimate toward a closer conversation about prevention. The guideline uses LDL cholesterol as its main number for planning, and non-HDL cholesterol is read next to it as an extra signal.
A result in or above that band does not mean heart disease is present, and a result below it does not rule risk out. Age, blood pressure, smoking, diabetes and family history all sit alongside non-HDL cholesterol in a full risk estimate, which a clinician puts together.
Non-HDL cholesterol rises with anything that raises LDL or the triglyceride-rich particles, VLDL and their remnants. Diet pattern, excess body weight, physical inactivity and poorly controlled diabetes tend to push it up, while weight loss and regular activity tend to bring it down over weeks to months.
The ATP III report lists conditions that can raise blood lipids, including an underactive thyroid, some kidney conditions and obstructive liver disease. It also names medicines that can shift lipids, such as progestins, anabolic steroids and corticosteroids. Statins and other lipid medicines lower non-HDL cholesterol, so a result taken on treatment reads differently from one taken off it.
Because HDL is subtracted, a fall in HDL nudges non-HDL cholesterol up by the same amount, even if total cholesterol has not changed. The reverse also holds. A rise in HDL with total cholesterol unchanged lowers the result. That is worth keeping in mind when comparing two panels, because the change may come from either input rather than from the particles non-HDL is meant to track.
Non-HDL cholesterol changes little after an ordinary meal, and the 2018 guideline accepts a nonfasting panel for a first look at risk. That makes it one of the more stable values on a lipid panel from one draw to the next.
A result well above earlier panels is still worth repeating under similar conditions before reading much into it, ideally away from a recent illness, a big change in alcohol or a crash diet. After starting or changing a statin or lifestyle plan, the 2018 guideline suggests a repeat lipid panel 4 to 12 weeks later, then every 3 to 12 months as needed.
A non-HDL cholesterol in the risk-enhancing band or higher, especially with other risk factors, is worth discussing with a clinician. Very high values, or a large jump between panels with no obvious reason, are worth raising soon rather than at the next routine check.
The subtraction itself is exact, so the limits come from the inputs. Both values need to come from the same blood draw and be in the same units. Mixing a total cholesterol from one lab with an HDL from another can shift the result by more than real biology would.
Non-HDL cholesterol counts cholesterol, not particles. Two people with the same non-HDL value can carry different numbers of LDL and VLDL particles, which is what apolipoprotein B measures more directly. It also includes the cholesterol in lipoprotein(a) without separating it out, and lipoprotein(a) needs its own test.
The bands quoted here come from adult guidelines. They are not designed for children, for pregnancy, or for rare inherited lipid disorders, where a specialist reads the panel differently. They also come from population studies, so they describe average risk across many people rather than the exact risk for any one person.
Take a total cholesterol of 230 mg/dL and an HDL of 40 mg/dL from the same panel. The formula is total cholesterol minus HDL, so 230 minus 40 gives a non-HDL cholesterol of 190 mg/dL. The calculator shows it as a whole number in mg/dL. In mmol/L the subtraction works the same way, as long as both inputs use that unit.
That result sits at the bottom of the 190 to 219 mg/dL band the 2018 guideline lists as a risk-enhancing factor. On its own it is a reason to talk with a clinician, who reads it with LDL, triglycerides, blood pressure, smoking status and the rest of a 10-year risk estimate.
Using the ATP III margin, a clinician who had an LDL goal of 130 mg/dL in mind for this person would look for a non-HDL value of 160 mg/dL or lower. This result is 30 mg/dL above that. If the same panel showed an LDL near 150, most of the non-HDL value would be LDL, and if it showed an LDL near 120, more would sit in triglyceride-rich particles.
LDL cholesterol is one part of non-HDL cholesterol. Non-HDL also includes the cholesterol in VLDL and their remnants, so it captures more of the cholesterol in particles linked with plaque, and it needs no estimate of VLDL. That matters most when triglycerides are raised, which is when calculated LDL is least reliable.
A 2011 meta-analysis by Sniderman and colleagues ranked apolipoprotein B as the strongest of the three markers for cardiovascular risk, with non-HDL cholesterol next and LDL cholesterol behind both. A 2012 meta-analysis of people on statins by Boekholdt and colleagues found on-treatment non-HDL cholesterol was more strongly linked with later events than LDL cholesterol.
The total cholesterol to HDL ratio uses the same two inputs but divides instead of subtracts. Apolipoprotein B needs a separate test. In practice a clinician often reads LDL and non-HDL cholesterol together and adds apolipoprotein B when the two point in different directions.
The Heart health panel measures the values this calculator uses. Related: cholesterol non hdl, total cholesterol, hdl cholesterol.
A non-HDL cholesterol calculator subtracts HDL cholesterol from total cholesterol. The answer is the cholesterol carried inside every atherogenic particle in your blood, including LDL, VLDL, remnant lipoproteins and Lp(a). You need only two numbers from a standard lipid panel, no fasting sample is required, and a clinician reads the figure alongside LDL and ApoB.
The calculator on this page does that arithmetic and keeps the units consistent. What follows explains the inputs, what the output actually describes, why the number can shift between tests, and which markers make it more informative.
The calculator takes two values from a routine lipid panel and performs one subtraction. Total cholesterol is the cholesterol contained in every lipoprotein circulating in your blood. HDL cholesterol is the portion carried in high-density lipoproteins, the particles that are not thought to drive plaque formation. Remove the HDL fraction from the total and what remains is the cholesterol sitting inside the particles that can deposit in an artery wall.
A worked example in conventional units. If total cholesterol is 210 mg/dL and HDL cholesterol is 55 mg/dL, non-HDL cholesterol is 155 mg/dL. In SI units, a total cholesterol of 5.4 mmol/L and an HDL of 1.4 mmol/L gives a non-HDL cholesterol of 4.0 mmol/L. Both inputs have to be in the same units before you subtract, which is the single most common source of a nonsensical answer. For cholesterol, mg/dL divided by roughly 38.67 converts to mmol/L.
Two features make this calculation unusual among lipid metrics. First, both inputs are measured directly by the laboratory rather than estimated, so the output is exact arithmetic rather than a model. Second, because neither input depends on a triglyceride term, the result does not require a fasting sample. Calculated LDL cholesterol, by contrast, is commonly derived using the Friedewald equation, which uses triglycerides and therefore becomes unreliable after a meal or when triglycerides are high. Non-HDL cholesterol sidesteps that problem entirely, which is one reason it appears on panels drawn at any time of day.
The calculator cannot be more accurate than the numbers you feed it. If the lipid panel itself was drawn during an acute illness or processed with an unusual assay, the subtraction will faithfully reproduce that distortion.
Non-HDL cholesterol is a single number representing the cholesterol content of all apolipoprotein B containing lipoproteins. Every atherogenic particle carries exactly one apoB molecule on its surface, and that family includes LDL, VLDL, intermediate-density lipoproteins, chylomicron remnants and Lp(a). LDL cholesterol accounts for only part of that group. The remainder, mostly the triglyceride-rich particles and their remnants, is invisible on an LDL-only reading but is captured the moment you subtract HDL from the total.
This matters because, outside rare inherited lipid disorders, the total number of apoB particles is what determines cardiovascular risk 6. Research suggests that non-HDL cholesterol performs at least as well as LDL cholesterol as a risk marker across a broad range of populations, and that its advantage widens in people with raised triglycerides or metabolic disease 1. The metric is not new either. Non-HDL cholesterol has appeared in cholesterol-lowering guidance in various forms for decades, usually as a secondary target alongside LDL cholesterol 2.
It is also where so-called residual risk often hides. Higher non-HDL cholesterol is associated with cardiovascular events that occur despite an LDL cholesterol value that looks acceptable on paper 3. The prognostic signal is not limited to coronary disease either. Studies suggest that non-HDL cholesterol carries prognostic information for adverse outcomes after acute ischemic stroke 4.
Your result is a measure of how much atherogenic cholesterol is in circulation at the moment of the draw, nothing more. It is not a diagnosis, and no single lipid value establishes whether someone has cardiovascular disease.
The categories below are the ones most commonly printed on laboratory reports in conventional units, with approximate SI equivalents. They describe population-level bands, not personal targets.
Below about 130 mg/dL (roughly 3.4 mmol/L). Generally reported as desirable for adults at average risk. The useful next step is to keep an eye on the trend across future panels rather than treat one in-range result as permanent.
About 130 to 159 mg/dL (roughly 3.4 to 4.1 mmol/L). Above the desirable band. This may reflect dietary pattern, body composition, physical activity, alcohol intake, thyroid function or genetics, and it is worth reviewing with a clinician alongside triglycerides, blood pressure and glucose markers.
About 160 to 189 mg/dL (roughly 4.1 to 4.9 mmol/L). Commonly labelled borderline high. At this level the rest of the risk picture matters more than the number alone, including family history, smoking status and whether ApoB or Lp(a) have ever been measured.
About 190 mg/dL and above (roughly 4.9 mmol/L and above). Reported as high, and above roughly 220 mg/dL as very high. A result in this territory is suboptimal and is worth discussing with a clinician promptly so that the cause and the overall risk profile can be assessed properly. It should never be ignored, and it should never be self-managed.
Two caveats sit on top of all of these bands. Personal targets are set by a clinician for each individual, and for someone with established cardiovascular disease, diabetes or a strong family history, the number a clinician aims for may be considerably lower than the general desirable band. And a result that sits inside the desirable band does not rule out risk, particularly when triglycerides are elevated or Lp(a) has never been checked.
The subtraction is exact, so any error comes from the inputs or from the timing of the draw.
Biological variability. Lipid values fluctuate from week to week in the same person. One panel is a snapshot, and a single point should not be over-interpreted against a threshold.
Acute illness or injury. Infection, surgery, recent heart attack or stroke and other acute stress can temporarily depress cholesterol values, making a panel drawn in that window unrepresentative of your usual state.
Very high triglycerides. When triglycerides are substantially elevated, calculated LDL cholesterol loses reliability, which is part of why non-HDL cholesterol is preferred in that situation. Studies suggest that triglyceride concentration influences how readily non-HDL cholesterol goals are reached in lipid-lowering trials 5.
Assay and laboratory differences. HDL cholesterol in particular can be measured by different direct methods, and results are not always perfectly interchangeable between laboratories. Comparing serial results from the same lab is more informative than comparing across providers.
Unit and transcription errors. Mixing mg/dL and mmol/L, or reading HDL where the report shows LDL, produces a number that looks plausible and is wrong.
Recent change in circumstances. Weight change, a shift in alcohol intake, pregnancy, untreated thyroid disease and new medications can all move lipids. Context belongs in the interpretation.
These three markers answer slightly different questions, and they are complementary rather than interchangeable.
| Marker | What it reflects | How it is obtained |
|---|---|---|
| LDL cholesterol | Cholesterol carried in LDL particles only | Usually calculated from the panel, sometimes measured directly |
| Non-HDL cholesterol | Cholesterol carried in all apoB-containing particles | Simple subtraction of two directly measured values, no extra test, no fasting |
| ApoB | The number of atherogenic particles, since each carries one apoB | Measured directly, usually ordered as a separate test |
The conceptual distinction is cholesterol mass versus particle count. Two people can share an identical non-HDL cholesterol value while carrying very different numbers of particles, because particles vary in how much cholesterol they hold. Research suggests that apolipoprotein B is physiologically the more direct marker of atherogenic particle burden than either LDL cholesterol or non-HDL cholesterol 6. In routine practice the three are often read together, and reviews of contemporary lipid management describe how each is used in cardiovascular care 7.
The practical appeal of non-HDL cholesterol is that it costs nothing extra. If you already have a standard lipid panel, you already have it.
The marker to read alongside the calculator output is the laboratory-reported Non-HDL Cholesterol value itself, which gives you the reference range your lab applies and lets you track the trend over repeat draws rather than relying on one calculation.
Around it, a few others change the interpretation. Triglycerides indicate how much of the non-HDL fraction sits in remnant particles. ApoB adds the particle count. Lp(a) is largely genetically determined and is typically measured at least once in a lifetime, since it is included in non-HDL cholesterol but is not separable from it. Markers of inflammation and glucose regulation, such as hs-CRP and HbA1c, describe the metabolic environment those particles are circulating in. Blood pressure, family history, smoking status and age carry at least as much weight as any lipid value when a clinician estimates overall risk.
Non-HDL cholesterol appears on any panel that reports both total cholesterol and HDL cholesterol, so most standard lipid panels contain it either explicitly or implicitly. The Heart Health panel measures the lipid fractions together with related cardiovascular markers, so the subtraction is already done and placed in context rather than left to a calculator.
FitScript is a labs-first membership built around 150+ biomarkers tested per year, with repeat testing so that a value can be read as a trend rather than a one-off. Atlas, FitScript's AI health advisor, interprets how your lipid markers sit relative to the rest of your biology, and any treatment decision is made by a licensed provider who evaluates whether it is appropriate for you.
Bring the result to a clinician if it sits above the desirable band, if it has risen meaningfully from a previous panel, if your triglycerides are also elevated, or if cardiovascular disease runs in your family. A result in the high or very high categories is worth discussing promptly, as is any lipid panel drawn alongside new symptoms such as chest discomfort, breathlessness on exertion or leg pain when walking, which deserve same-day medical attention rather than a calculator.
What a clinician does with the number depends on the whole picture. Management may involve nutrition, activity, alcohol, sleep, body composition, investigating thyroid or kidney contributors, and in some cases medication, which only a licensed provider can evaluate and prescribe. Discuss your results and any treatment with a clinician or licensed provider before changing anything. This article is educational and is not medical advice.
If you are mapping out the next thing to read, does coffee raise cholesterol covers one of the more common everyday questions about lipid panels, and how do cholesterol lowering medications work explains the mechanisms behind the treatment options a clinician may raise.
No. LDL is one part of it; non-HDL also includes the cholesterol in VLDL and IDL particles.
It comes from two standard lipid values and covers all the particles considered atherogenic, so it adds to what LDL shows.
The 2018 US guideline lists 190 to 219 mg/dL as a risk-enhancing factor. A clinician reads it with your other risk factors.
Non-HDL cholesterol equals total cholesterol minus HDL cholesterol. Both values must be in the same units. For example, a total cholesterol of 210 mg/dL with an HDL of 55 mg/dL gives 155 mg/dL, and a total of 5.4 mmol/L with an HDL of 1.4 mmol/L gives 4.0 mmol/L.
Not for this particular number. Total cholesterol and HDL cholesterol are both measured directly and neither is meaningfully distorted by a recent meal, so non-HDL cholesterol can be calculated from a non-fasting panel. Calculated LDL cholesterol is more sensitive to recent eating because it uses a triglyceride term. Follow whatever instructions your clinician or lab gives you.
Laboratories commonly report values below roughly 130 mg/dL, or about 3.4 mmol/L, as desirable for adults at average risk, with higher bands labelled above desirable, borderline high and high. Those are population categories, not personal goals. A clinician sets an individual target based on your full risk profile, and for some people that target is considerably lower.
LDL cholesterol counts only the cholesterol inside LDL particles. Non-HDL cholesterol counts the cholesterol inside every atherogenic particle, including VLDL, intermediate-density lipoproteins, remnant particles and Lp(a) as well as LDL. That makes it a broader measure, and the gap between the two widens when triglycerides are elevated.
They answer different questions. Non-HDL cholesterol measures cholesterol mass in atherogenic particles, while ApoB counts the particles themselves, since each carries one apoB molecule. Research suggests ApoB is the more direct marker of particle burden, but it requires a separate test, whereas non-HDL cholesterol comes free with any standard lipid panel. Many clinicians review both.
Yes, and that pattern is one of the reasons the metric is useful. It usually reflects cholesterol carried in triglyceride-rich particles and their remnants, which an LDL reading does not capture. If your LDL is in range but non-HDL is not, that discrepancy is worth raising with a clinician alongside your triglyceride result.
Testing frequency depends on your age, risk profile and whether anything is being actively managed, so it is a decision for your clinician. What helps in every case is consistency. Repeat testing at the same laboratory makes the trend readable, and a trend is far more informative than one isolated value.
No. A calculator performs arithmetic on two lab values and nothing else. It cannot diagnose a condition, assess your arteries, or account for blood pressure, family history, smoking, glucose control or inflammation. Interpreting what a result means for you is a clinical judgement, and it should be made by a clinician reviewing your full picture.