Insulin-like growth factor 1 predicts post-load hypoglycemia following bariatric surgery: a prospective cohort study

Abstract

Postprandial hypoglycemia is a complication following gastric bypass surgery, which frequently remains undetected. Severe hypoglycemic episodes, however, put patients at risk, e.g., for syncope. A major cause of hypoglycemia following gastric bypass is hyperinsulinemic nesidioblastosis. Since pancreatic islets in nesidioblastosis overexpress insulin-like growth factor 1 (IGF-1) receptor α and administration of recombinant IGF-1 provokes hypoglycemia, our main objective was to investigate the occurrence of post-load hypoglycemia one year after bariatric surgery and its relation to pre- and post-operative IGF-1 serum concentrations. We evaluated metabolic parameters including 2 h 75 g oral glucose tolerance test (OGTT) and measured IGF-1 serum concentration in thirty-six non-diabetic patients (29 f/7 m), aged 41.3±2.0 y with a median (IQR) BMI of 30.9 kg/m2 (27.5-34.3 kg/m2), who underwent elective bariatric surgery (predominantly gastric bypass, 83%) at our hospital. Post-load hypoglycemia as defined by a 2 h glucose concentration <60 mg/dl was detected in 50% of patients. Serum insulin and C-peptide concentration during the OGTT and HOMA-IR (homeostatic model assessment-insulin resistance) were similar in hypoglycemic and euglycemic patients. Strikingly, pre- and post-operative serum IGF-1 concentrations were significantly higher in hypoglycemic patients (p = 0.012 and p = 0.007 respectively). IGF-1 serum concentration before surgery negatively correlated with 2 h glucose concentration during the OGTT (rho = -0.58, p = 0.0003). Finally, IGF-1 serum concentrations before and after surgery significantly predicted post-load hypoglycemia with odds ratios of 1.28 (95%CI:1.03-1.55, p = 0.029) and 1.18 (95%CI:1.03-1.33, p = 0.015), respectively, for each 10 ng/ml increment. IGF-1 serum concentration could be a valuable biomarker to identify patients at risk for hypoglycemia following bariatric surgery independently of a diagnostic OGTT. Thus, IGF-1 testing could help to prevent a significant complication of gastric bypass surgery.

Figure 1. Glucose, insulin and C-peptide concentration during pre-operative and post-operative OGTT.

(A) Glucose concentrations during the course of the pre-operative 2 h OGTT (n = 35, full lines) compared to the post-operative OGTT (n = 35, dashed lines). (B) Histogram of post-operative 2 h glucose concentrations during the OGTT (n = 35). Post-load hypoglycemia was defined by a 2 h glucose concentration<60 mg/dl. (C) Insulin and (D) C-peptide concentrations during the course of the pre-operative 2 h OGTT (n = 35, full lines) compared to the post-operative OGTT (n = 35, dashed lines).

Figure 2. Post-operative OGTT glucose, insulin and C-peptide concentration in patients with euglycemia and hypoglycemia.

(A) Glucose, (B) insulin and (C) C-peptide concentrations during the course of the post-operative 2 h OGTT in patients with euglycemia (n = 17, full lines) and patients with post-load hypoglycemia (n = 18, dashed lines). Differences between pre and post-operative values were calculated with paired t-test. Differences between patients with euglycemia and hypoglycemia were calculated by unpaired student's t-test.*p<0.05.

Figure 3. IGF-1 in patients with euglycemia and post-load hypoglycemia at 2 h during the post-operative OGTT.

(A) Box-plots of pre- and post-operative serum IGF-1 concentrations in patients with euglycemia (n = 17, black bars) and post-load hypoglycemia (n = 18, white bars). Outliers are represented by dots. (B) Pre-operative serum IGF-1 concentration plotted against 2 h glucose concentration during the post-operative OGTT (n = 35). (C) Post-operative serum IGF-1 concentration plotted against 2 h glucose concentration during the post-operative OGTT (n = 35). (D) ROC-AUC curve for detecting post-load hypoglycemia according to pre-operative IGF-1 concentration (n = 35). Differences between IGF-1 concentrations between the two groups were calculated by unpaired student's t-test. The association between pre- and post-operative serum IGF-1 concentration and post-operative 2 h glucose concentration during the OGTT was analyzed using Spearman's rank correlation,*p<0.05.