Healthcare & Clinical Calculations
Triglyceride-glucose index: two formulas and TyG-BMI
Compare two published triglyceride-glucose formulas and calculate a clearly labeled TyG-BMI research example.
The parentheses change the result
The triglyceride-glucose (TyG) index is studied as a surrogate marker in metabolic research. Publications use two different expressions: ln(TG × glucose / 2), with division inside the natural logarithm, and ln(TG × glucose) / 2, with division outside it. Here TG and fasting glucose are entered as numerical concentrations in mg/dL. They are different numerical scales, so a threshold or TyG-BMI value from one expression must not be applied to the other.
For example, a 2025 cohort analysis uses the inside-division expression, while a 2026 cohort uses the outside-division expression and cites the original authors’ 2020 clarification. Both study the association of an index with health outcomes; neither formula by itself establishes an individual diagnosis.
Version A: division inside the logarithm
For TG 165 mg/dL and fasting glucose 108 mg/dL, ln(165 × 108 / 2) rounds to 9.095. This worksheet requires both values to be positive numbers.
=IF(COUNT(B2:B3)<>2,"Enter triglycerides and glucose",IF(MIN(B2:B3)<=0,"Inputs must be positive",ROUND(LN(B2*B3/2),3)))Version B: division outside the logarithm
The same inputs give ln(165 × 108) / 2 ≈ 4.894. Parentheses and the units used by a source matter when reproducing its results.
=IF(COUNT(B2:B3)<>2,"Enter triglycerides and glucose",IF(MIN(B2:B3)<=0,"Inputs must be positive",ROUND(LN(B2*B3)/2,3)))TyG-BMI example using version A
Some studies multiply the inside-division TyG expression by BMI in kg/m². With BMI 29.4, the example gives approximately 267.4 using the unrounded version A value. TyG-BMI based on the other expression would use a different numerical scale. BMI is not needed to calculate either TyG expression on its own.
=IF(NOT(ISNUMBER(E2)),E2,IF(COUNT(B4)<>1,"Enter BMI",IF(B4<=0,"BMI must be positive",ROUND(LN(B2*B3/2)*B4,1))))Interpretation and limitations
Observed associations and proposed cutoffs depend on the formula, population and study design. The index is not a direct measurement of insulin resistance and does not diagnose MASLD, diabetes, cardiovascular disease or an individual’s future risk. Use established clinical assessment and applicable guidance rather than a universal cutoff.
Try the example
Use the original values or change them to test your understanding.