Fasting Insulin

Fasting insulin measures how much hormone the pancreas has to release to keep glucose normal overnight. In early insulin resistance, glucose stays in range precisely because insulin has climbed to hold it there, so fasting glucose reads normal while fasting insulin is already elevated. That is the whole argument for measuring it. It is also the numerator of HOMA-IR, calculated as fasting insulin in microIU/mL times fasting glucose in mmol/L divided by 22.5.

The DECODE Insulin Study Group pooled eleven prospective European studies covering 6,156 men and 5,351 women aged 30 to 89, with 8.8 years of follow-up and 362 cardiovascular deaths in men and 70 in women. Comparing the highest quartile of fasting insulin with the lower quartiles, the hazard ratio for cardiovascular death was 1.54 (95% CI 1.16-2.03) in men and 2.66 (95% CI 1.45-4.90) in women after adjustment for age, smoking and other risk factors. Per interquartile range the figures were 1.13 (95% CI 1.05-1.22) in men and 1.25 (95% CI 1.08-1.45) in women [1]. Two-hour post-load insulin lost significance after full adjustment in both sexes, so the fasting value carried the signal [1].

Xiaohong Zhang and Chaoyang Wang meta-analysed 7 prospective studies covering 26,976 non-diabetic adults followed 5 to 19 years. Highest versus lowest fasting insulin gave a pooled relative risk of 1.13 (95% CI 1.00-1.27, P = 0.058) for all-cause mortality, while HOMA-IR gave 1.34 (95% CI 1.11-1.62, P = 0.002) and 2.11 (95% CI 1.01-4.41, P = 0.048) for cardiovascular mortality [2]. The composite index beat the raw hormone, which is a fair caution against reading fasting insulin alone.

Sara Faghihi-Kashani and Alireza Esteghamati followed 2,607 patients with type 2 diabetes for an average 7.2 years and recorded 299 hard coronary events. Each one-quartile increase in fasting insulin carried a hazard ratio of 1.18 (95% CI 1.06-1.32, P < 0.01), independent of gender, BMI, blood pressure, lipids, medication and HbA1c. Fasting glucose showed no association [3].

How it is measured: immunoassay on serum after 8 to 12 hours without food. Assays vary and are not standardised between labs, so track trends within one lab. Insulin is unstable at room temperature and haemolysis destroys it, so handling matters more than for most panels.

How to move it: in the 409-person PREMOTE trial, berberine alone lowered HbA1c by 0.99 percentage points (95% CI -1.16 to -0.83) over 12 weeks against 0.59 on placebo, P < 0.001 [4]. Weight loss and exercise remain the largest documented movers.