TyG index calculator

Why an index built from triglycerides

The standard way to estimate insulin resistance from a blood sample is HOMA-IR, and it needs a fasting insulin. Many general practices will not order one, many insurers will not pay for one, and insulin assays are poorly standardised between laboratories anyway.

The TyG index sidesteps all of that. It uses fasting triglycerides and fasting glucose — two values that appear on essentially every routine lipid panel already in your file. Simental-Mendía and colleagues published it in Metabolic Syndrome and Related Disorders in 2008, validated against the hyperinsulinaemic-euglycaemic clamp, which is the reference standard for measuring insulin sensitivity.

The biology behind it is not arbitrary. Insulin normally suppresses the release of fatty acids from fat tissue and the production of triglyceride-rich particles by the liver. When that suppression weakens, fasting triglycerides drift up. Raised triglycerides and raised fasting glucose together are a fingerprint of the same problem.

The formula

TyG is the natural logarithm of fasting triglycerides in mg/dL multiplied by fasting glucose in mg/dL, with the result divided by two. The calculator accepts either unit for both values and converts before computing — triglycerides at 88.57 mg/dL per mmol/L, glucose at 18.0156. The conversion has to happen first, because a logarithm of a product shifts by a constant when the units change, which would move every published cut-off without any visible sign that it had.

Because it is a logarithm, the scale is compressed: the whole meaningful range sits between roughly four and six, and a difference of a tenth is not trivial. That also means the index is insensitive to small measurement noise, which is part of why it reproduces well.

What the thresholds are worth

Less than HbA1c thresholds are, and the page should say so plainly. Rather than pick one number, the middle band above spans the published ones. Its lower edge is the cut-point from the original 2008 paper, which validated the index against HOMA-IR in a Mexican population; its upper edge is the highest well-sourced value, from an Indian cohort scored against metabolic syndrome. The one clamp-validated figure — from the same Mexican group in 2010, measured against the euglycaemic-hyperinsulinaemic clamp, which is the reference standard for insulin sensitivity — sits between them. No professional body has adopted any of the three.

One practical warning, because it silently doubles the answer on other sites: the index is written two ways in the literature. Taking the logarithm of the product and then halving it — the convention this calculator uses, and the one the cut-offs above belong to — gives values in the fours and fives. Halving inside the logarithm instead gives the same index shifted to roughly twice that. Both conventions are correct and their published thresholds differ accordingly, so a result from one compared against a threshold from the other is simply wrong. If a calculator gives you a number near nine and holds it against a threshold near four and a half, close the tab.

What TyG is genuinely good at is ranking. Within one population, a higher TyG reliably marks lower insulin sensitivity, and large cohort studies have linked higher values to later type 2 diabetes and to cardiovascular events. Treating your own value as a position on a scale, tracked over time, is sound. Treating a single reading as a pass or fail is not.

When it does not apply

Triglycerides are the volatile half of the formula. They rise sharply after a meal and after alcohol, so the sample must be genuinely fasting — most laboratories ask for at least eight hours, and alcohol the night before will distort it. Familial hypertriglyceridaemia, untreated hypothyroidism, nephrotic syndrome and some medications raise triglycerides for reasons that have nothing to do with insulin, and TyG cannot tell those apart from insulin resistance.

It is also not a diagnostic test and does not appear in any diagnostic guideline. If fasting insulin is available to you, HOMA-IR is the more direct measure; TyG is the one you can compute from bloods you already have.