Limited recovery of β-cell function after gastric bypass despite clinical diabetes remission

Abstract

The mechanisms responsible for the remarkable remission of type 2 diabetes after Roux-en-Y gastric bypass (RYGBP) are still puzzling. To elucidate the role of the gut, we compared β-cell function assessed during an oral glucose tolerance test (OGTT) and an isoglycemic intravenous glucose clamp (iso-IVGC) in: 1) 16 severely obese patients with type 2 diabetes, up to 3 years post-RYGBP; 2) 11 severely obese normal glucose-tolerant control subjects; and 3) 7 lean control subjects. Diabetes remission was observed after RYGBP. β-Cell function during the OGTT, significantly blunted prior to RYGBP, normalized to levels of both control groups after RYGBP. In contrast, during the iso-IVGC, β-cell function improved minimally and remained significantly impaired compared with lean control subjects up to 3 years post-RYGBP. Presurgery, β-cell function, weight loss, and glucagon-like peptide 1 response were all predictors of postsurgery β-cell function, although weight loss appeared to be the strongest predictor. These data show that β-cell dysfunction persists after RYGBP, even in patients in clinical diabetes remission. This impairment can be rescued by oral glucose stimulation, suggesting that RYGBP leads to an important gastrointestinal effect, critical for improved β-cell function after surgery.

Trial registration: ClinicalTrials.gov NCT00571220.

Figures

Figure 1

Effect of RYGBP on ISR and β-cell function. A: ISR during the OGTT or iso-IVGC over the entire experiment (180 min) was not different between groups. In all groups and conditions, ISR AUC0–180 min was significantly greater during the OGTT vs. iso-IVGC (P < 0.05). B: ISX was significantly impaired during the OGTT in the OB-DM group presurgery and significantly increased immediately after surgery; this increase was maintained up to 3 years postsurgery. However, no increase in ISX was observed during the iso-IVGC. In all groups and conditions, ISX was significantly greater during the OGTT vs. iso-IVGC (P < 0.05). C: ISR during the first 60 min of the OGTT was significantly increased immediately after surgery and this increase was maintained up to 3 years postsurgery. No increase in ISR during the iso-IVGC was observed. In all groups and conditions, ISR AUC0–60 min was significantly greater during the OGTT vs. iso-IVGC (P < 0.05). D: In the OB-DM group presurgery, BCGS during the OGTT and iso-IVGC was significantly lower than both control groups. After surgery, O-BCGS normalized to the levels of both control groups by 1 month and was further increased at 1 to 2 years. In contrast, IV-BCGS remained significantly lower compared with both control groups up to 3 years. BCGS was significantly greater during the OGTT vs. iso-IVGC in the OB-NGT group and OB-DM group at all conditions postsurgery (P < 0.05), but not in the LEAN or OB-DM group prior to surgery. E: In the OB-DM group presurgery, DI during the OGTT and iso-IVGC was significantly impaired compared with both control groups. After surgery, O-DI normalized to the levels of the OB-NGT group by 1 month and LEAN by 1 year. In contrast, IV-DI remained significantly lower compared with the LEAN control subjects up to 3 years postsurgery. DI was significantly greater during the OGTT vs. iso-IVGC in the OB-NGT group and in the OB-DM group at all conditions postsurgery (P < 0.05), but not in the LEAN or OB-DM group prior to surgery. Data are mean ± SEM. *P < 0.05 vs. LEAN; †P < 0.05 vs. OB-NGT; ‡P < 0.05 vs. OB-DM0; §P < 0.05 vs. OB-DM1M.

Figure 2

ISR during the OGTT and iso-IVGC and the incretin effect of ISR. A–G: ISR AUC (pmol/kg/min) was significantly greater during the OGTT compared with iso-IVGC under all groups and conditions (P < 0.05). H: The incretin effect of ISR was blunted in subjects with diabetes prior to surgery (D0), but normalized from 1 month onwards after surgery. Data are mean ± SEM. *P < 0.05 vs. LEAN; †P < 0.05 vs. OB-NGT; ‡P < 0.05 vs. OB-DM0. D0, obese group with diabetes prior to RYGBP surgery; D1M, D1Y, D2Y, D3Y, obese group with diabetes at 1 month, 1 year, 2 years, and 3 years after RYGBP surgery, respectively; L, lean; OB, obese NGT control subjects.

Figure 3

Effect of RYGBP on the DI during the OGTT and iso-IVGC. A: OGTT. O-DI, significantly impaired in OB-DM prior to surgery (P < 0.05 vs. both LEAN and OB-NGT groups), improved rapidly and significantly, as illustrated by a shift upwards and to the right. O-DI in the OB-DM group normalized to levels of OB-NGT control subjects by 1 month and was not significantly different from the LEAN from 1 year onwards (P < 0.05). B: Iso-IVGC. IV-DI was significantly impaired in OB-DM prior to surgery (P < 0.05 vs. both LEAN and OB-NGT groups). Contrary to the O-DI, IV-DI improved minimally, albeit significantly (P < 0.05), with a shift to the right and minimal shift upward, after RYGBP. However, IV-DI remained significantly lower than the OB-NGT control subjects at 1 month and lower than the LEAN control subjects at all time points postsurgery (P < 0.05). Mean ± SEM for all groups except LEAN. LEAN presented as each individual subject.