The effect of nuts on markers of glycemic control: a systematic review and meta-analysis of randomized controlled trials.

Background: Observational evidence suggests higher nut consumption is associated with better glycemic control; however, it is unclear if this association is causal.

Objectives: We aimed to conduct a systematic review and meta-analysis of randomized controlled trials to examine the effect of tree nuts and peanuts on markers of glycemic control in adults.

Methods: A systematic review and meta-analysis of randomized controlled trials was conducted. A total of 1063 potentially eligible articles were screened in duplicate. From these articles, 40 were eligible for inclusion and data from these articles were extracted in duplicate. The weighted mean difference (WMD) between the nut intervention and control arms was determined for fasting glucose, fasting insulin, glycated hemoglobin (HbA1c), and homeostasis model assessment of insulin resistance (HOMA-IR) using the DerSimonian and Laird random-effects method. For outcomes where a limited number of studies were published, a qualitative synthesis was presented.

Results: A total of 40 randomized controlled trials including 2832 unique participants, with a median duration of 3 mo (range: 1-12 mo), were included. Overall consumption of tree nuts or peanuts had a favorable effect on HOMA-IR (WMD: -0.23; 95% CI: -0.40, -0.06; I2 = 51.7%) and fasting insulin (WMD: -0.40 μIU/mL; 95% CI: -0.73, -0.07 μIU/mL; I2 = 49.4%). There was no significant effect of nut consumption on fasting blood glucose (WMD: -0.52 mg/dL; 95% CI: -1.43, 0.38 mg/dL; I2 = 53.4%) or HbA1c (WMD: 0.02%; 95% CI: -0.01%, 0.04%; I2 = 51.0%).

Conclusions: Consumption of peanuts or tree nuts significantly decreased HOMA-IR and fasting insulin; there was no effect of nut consumption on HbA1c or fasting glucose. The results suggest that nut consumption may improve insulin sensitivity. In the future, well-designed clinical trials are required to elucidate the mechanisms that account for these observed effects.