Insulin resistance reduction, intermittent fasting, and human growth hormone: secondary analysis of a randomized trial

Intense intermittent fasting regimens safely reduce weight to a similar extent as caloric restriction. A previous trial reported low-frequency 26-week intermittent fasting reduced homeostatic model assessment of insulin resistance (HOMA-IR) without significant weight loss. During fasting, human growth hormone (HGH) increases substantially, but whether basal HGH modifies the effect of fasting on outcomes of repeated fasting is unknown. In a post hoc analysis of a randomized controlled trial (registration: clinicaltrials.gov, NCT02770313, May 12, 2016), subjects (N = 68) were adults ages 21-70 years with modest cholesterol elevation, ≥1 metabolic syndrome component, available HGH measurements, no chronic disease, and no statin or anti-diabetes medication. Randomization was 1:1 to intermittent fasting (24-hour, water-only, twice-per-week for 4 weeks, then once-per-week for 22 weeks) or 26-week ad libitum control. General linear modeling evaluated the interaction of trial arm with baseline HGH for HOMA-IR changes. Subjects with lower baseline HGH had 26-week HOMA-IR changes (p = 0.003) of -1.04 ± 0.99 for fasting versus 0.60 ± 1.04 for controls. Subjects with higher baseline HGH had HOMA-IR changes (p = 0.26) of -0.69 ± 0.75 (fasting) and -0.42 ± 0.92 (controls). The interaction of fasting with lower baseline HGH was significant (p-interaction=0.004). Results were similar for insulin and glucose. Weight loss at 26 weeks was not significantly different between fasting and controls (-1.74 ± 4.81 kg vs. 0.21 ± 3.50 kg, p = 0.08) and was not correlated with changes in HOMA-IR, insulin, glucose, and HGH. In conclusion, lower baseline HGH modified the effect of low-frequency water-only 24-hour fasting in profoundly reducing HOMA-IR over 26 weeks compared both to controls and to fasting subjects with higher baseline HGH.