Subchronic sleep restriction causes tissue-specific insulin resistance.
Level 2 - randomized trial
Randomized crossover trial in humans
PubMed 25658017 · doi:10.1210/jc.2014-3911
What was done
Fourteen subjects participated in a randomized crossover trial comparing 5 nights of sleep restriction (4 hours in bed per night) against 5 nights of normal sleep (8 hours in bed per night), each preceded by 2 acclimatization nights and separated by a washout period. Insulin sensitivity was evaluated using a hyperinsulinemic-euglycemic clamp and stable isotope tracers for hepatic glucose production. Secondary measures included 24-hour urinary stress hormones (cortisol, metanephrine, normetanephrine), fasting nonesterified fatty acids (NEFA), beta-hydroxybutyrate, and indirect calorimetry for resting energy expenditure and respiratory quotient.
What was found
Sleep restriction decreased whole-body insulin sensitivity by 25% (P = .008) and peripheral insulin sensitivity by 29% (P = .003) compared to normal sleep. Hepatic insulin sensitivity (endogenous glucose production) did not change significantly, though percent gluconeogenesis increased (P = .03). Urinary cortisol increased by 21% (P = .04), metanephrine by 8% (P = .014), and normetanephrine by 18% (P = .002). Fasting NEFA rose by 62% and beta-hydroxybutyrate rose by 55%. Resting energy expenditure was unchanged (P = 0.98), but respiratory quotient decreased from 0.81 ± 0.02 to 0.75 ± 0.02 (P = .045).
Why it matters
This study shows that short-term sleep restriction induces insulin resistance primarily in peripheral tissues rather than the liver, a process accompanied by elevated stress hormones and increased lipolysis.
Limits
The study had a very small sample size of 14 participants. The abstract does not specify participant demographics such as age, sex, or baseline health status. The five-night protocol reflects acute sleep restriction and does not capture long-term metabolic adaptations.