Association of Baseline Triglyceride-Glucose Index with Poor Glycemic Control and Diabetes Remission After Metabolic Surgery.

Abstract

The utility of insulin resistance (IR) as a predictor of diabetes remission after metabolic surgery is not well-defined. We assessed the association of baseline surrogate IR indices including triglyceride-glucose (TyG) index and homeostatic model assessment for IR (HOMA-IR) with glycemic control and diabetes remission after metabolic surgery. Patients with type 2 diabetes scheduled for metabolic surgery were recruited at a single-center (n = 149; age: 44 ± 10years, 47.7% men, body mass index: 41.5 ± 7.5kg/m2), and followed-up for 12months postoperatively. The relationships between the IR indices and poor glycemic control (HbA1c ≥ 7%) at baseline or complete diabetes remission (HbA1c < 6% without glucose-lowering medications at 12months) post-surgery were examined. Elevated TyG index was associated with poor glycemic control cross-sectionally. Compared with non-remitters, lower baseline TyG index levels were observed in individuals with complete diabetes remission after surgery (P = 0.012); whereas HOMA-IR was not significantly different. Consistently, the proportion of diabetes non-remitters (compared to remitters) increased with increasing TyG tertiles from 1 to 3 (P = 0.015). Both TyG index (relative risk = 0.62, 95% CI = 0.42-0.91, P = 0.014) and TyG tertile 1 (relative risk = 1.99, 95% CI = 1.25-3.24, P = 0.003) independently predicted diabetes remission. The TyG index identified diabetes remission with an area under the curve of 0.68. The optimal TyG threshold was 9.41, yielding a sensitivity of 69.6%, specificity of 60.9%, positive predictive value of 64.0%, and negative predictive value of 66.7%. TyG index, previously suggested to predominantly reflect muscle IR, outperforms HOMA-IR as an IR indicator associated with glycemic control and diabetes remission after metabolic surgery.