High incidence of insulin resistance and dysglycemia amongst nondiabetic cardiac surgical patients.

Abstract

Undiagnosed glycometabolic dysfunction is prominent amongst nondiabetic cardiac surgical patients, whereas perioperative dysglycemia is associated with adverse outcomes. This study assessed whether the preoperative level of insulin resistance predicts the degree of perioperative dysglycemia in nondiabetic, normoglycemic cardiac surgical patients. Twenty-two nondiabetic patients awaiting cardiac operations were assessed for metabolic parameters and whole-body insulin resistance (mean glucose infusion [GINF] rate) using the hyperinsulinemic-euglycemic clamp. Intraoperative and postoperative glucose levels and treatment requirements were analyzed. Linear regression analysis was used to find predictors of baseline, peak intraoperative, and mean postoperative fasting blood glucose (FBG). The mean GINF recorded in nondiabetic, normoglycemic patients was 3.5 ± 1.4 mg/kg/min. The mean peak intraoperative and mean postoperative FBG concentrations were 154.9 ± 34.2 mg/dL (range, 108.1 to 227.0 mg/dL) and 120.7 ± 16.2 mg/dL (range, 100.9 to 154.9 mg/dL), respectively. The GINF correlated inversely with mean peak intraoperative (r = -0.7, p = 0.02) and mean postoperative FBG (r = -0.8, p = 0.01). The GINF did not correlate with preoperative FBG levels (r = 0.3, p = 0.4). Preoperative FBG did not correlate with peak intraoperative (r = 0.4, p = 0.5) or mean postoperative FBG (r = 0.5, p = 0.3). Nondiabetic, normoglycemic cardiac surgical patients are highly insulin resistant using the hyperinsulinemic-euglycemic clamp. Preoperative insulin resistance, not FBG, is significantly associated with the development of perioperative dysglycemia. Insulin resistance screening may be useful to identify insulin resistance preoperatively and predict the degree of perioperative dysglycemia in cardiac surgical patients but should be performed with a more appropriate and reproducible test.