Yoshino · The New England journal of medicine 2020 · Non-randomized matched controlled study · n=22

Effects of Diet versus Gastric Bypass on Metabolic Function in Diabetes.

Cited 242 times in the scientific literature.

Level 3 - non-randomized controlled study

Non-randomized controlled comparative clinical study

PubMed 32813948 · doi:10.1056/NEJMoa2003697 · record verified 2026-08-27

What was done

Researchers compared the metabolic effects of matched (~18%) weight loss achieved through either Roux-en-Y gastric bypass surgery or diet alone in 22 patients with obesity and type 2 diabetes. The primary outcome was the change in hepatic insulin sensitivity measured by stages 1 and 2 of a 3-stage hyperinsulinemic-euglycemic pancreatic clamp. Secondary outcomes included muscle insulin sensitivity (insulin-stimulated glucose disposal), beta-cell function (insulin secretion relative to insulin sensitivity), and 24-hour plasma glucose and insulin area-under-the-curve profiles.

What was found

Both groups showed nearly identical improvements across all primary and secondary metabolic parameters: - Suppression of endogenous glucose production increased similarly during clamp stage 1 (diet: +7.04 μmol/kg fat-free mass/min, 95% CI 4.74 to 9.33; surgery: +7.02, 95% CI 3.21 to 10.84) and stage 2 (diet: +5.39, 95% CI 2.44 to 8.34; surgery: +5.37, 95% CI 2.41 to 8.33), with no significant differences between groups. - Insulin-stimulated glucose disposal increased from 30.5 ± 15.9 to 61.6 ± 13.0 μmol/kg fat-free mass/min with diet and from 29.4 ± 12.6 to 54.5 ± 10.4 with surgery (no significant difference). - Beta-cell function increased by 1.83 units (95% CI 1.22 to 2.44) with diet versus 1.11 units (95% CI 0.08 to 2.15) with surgery (no significant difference). - 24-hour plasma glucose and insulin AUC reductions were equivalent in both groups, and no major complications occurred.

Why it matters

This study demonstrates that the therapeutic metabolic benefits of Roux-en-Y gastric bypass on insulin sensitivity and beta-cell function are driven by the magnitude of weight loss itself, rather than weight-independent endocrine mechanisms.

Limits

The sample size was very small (n=22 total). The study design was not randomized, introducing potential selection bias. Findings reflect physiology at a single matched weight-loss milestone (~18%) and do not address long-term weight maintenance, feasibility, or extended clinical outcomes.

Cited by