Keijzers · Diabetes care 2002 · randomized double-blind crossover trial · n=12

Caffeine can decrease insulin sensitivity in humans.

Cited 367 times in the scientific literature.

Level 2 - randomized trial

Individual randomized double-blind crossover trial

PubMed 11815511 · doi:10.2337/diacare.25.2.364 · record verified 2026-08-29

What was done

Researchers tested the acute effect of caffeine on insulin sensitivity using hyperinsulinemic-euglycemic glucose clamps. In a randomized, double-blind, crossover trial, 12 healthy volunteers received intravenous caffeine or placebo. Outcomes measured included whole-body glucose uptake corrected for insulin concentration, plasma catecholamines, free fatty acids, insulin, and hemodynamic parameters. A second experiment evaluated 10 healthy subjects using an identical protocol with the adenosine reuptake inhibitor dipyridamole.

What was found

Caffeine reduced insulin sensitivity by 15% compared with placebo (P < 0.05). Plasma free fatty acids increased and remained higher than during placebo (P < 0.05). Plasma epinephrine increased fivefold (P < 0.0005), plasma norepinephrine increased (P < 0.02), and blood pressure rose (P < 0.001). Dipyridamole did not alter insulin sensitivity, though it increased plasma norepinephrine (P < 0.01).

Why it matters

The study demonstrates that acute caffeine exposure impairs whole-body insulin sensitivity in healthy humans, an effect likely driven by catecholamine release rather than peripheral adenosine receptor antagonism.

Limits

The sample size was small (12 subjects in the caffeine arm, 10 in the dipyridamole arm). Caffeine was administered intravenously rather than consumed orally as a beverage. The study assessed only acute responses in healthy volunteers, leaving long-term metabolic effects and responses in individuals with diabetes unmeasured.

Cited by