On the Failure of Insulin to Affect Hyperglycemia During Cardiac Surgery

Abstract

In Response: We greatly appreciate Dr. Schricker’s comments concerning our study (1) and the interesting topic of control of intraoperative hyperglycemia during cardiac surgery with cardiopulmonary bypass (CPB). His questions centered on our methodology and the observations that our protocol did not affect intraoperative glucose, although postoperative glucose was lower in the group that received insulin. As Dr. Schricker surmised, the duration of insulin administration was limited to time on CPB. Insulin was administered centrally through tubing that was not preconditioned to decrease binding. The patients were comparable with regards to preoperative medications, intraoperative fluid replacement, blood product transfusions, amount of cell saver blood replaced, and catecholamine administration. Five patients in the control group and three patients in the insulin group received catecholamine infusions during the time of separation from CPB, for an average time duration of 30 and 45 min, respectively. The anesthetic agents were comparable between groups. The initial heparin dose that was administered was indeed weight-based, but the two groups received similar doses overall (32,439 U and 33,071 U on average for the control and insulin groups, respectively, P = 0.869). This likely reflects the subsequent doses administered in response to activated clotting times, as it is well known that the response to heparin dosing is highly variable. While the group receiving insulin did indeed have a greater body mass index (BMI), upon multivariate analysis this factor did not affect glucose levels. Our study is in agreement with previously published observations by Chaney et al. that an intraoperative insulin infusion leads to lower postoperative glucose levels but unchanged intraoperative levels (2). As Dr. Schricker observes, Chaney et al. continued their insulin protocol through the period of sternal closure (2) and our protocol ceased insulin at the time of separation from CPB. It should be noted that the primary stimulus for hyperglycemia during cardiac surgery is CPB (3). Hence, once this profound stimulus ceases, it is not surprising that normal glucose control returns quickly, and the exogenous insulin administered will have an additive effect. This would be expected to lead to lower postoperative glucose levels. We strongly agree with Dr. Schricker’s concerns against allowing patients to remain hyperglycemic, especially given the recent exciting findings of Van den Berghe et al. (4). It must be noted that CPB is a potent stimulus for hyperglycemia (3), as well as an activator of the inflammatory response (5). While the hyperglycemia associated with CPB has been recognized for many years, the impact of this transient alteration in glucose control in the face of the activation of the systemic inflammatory response is poorly understood. We previously found that diabetic patients undergoing CPB have improved neutrophil phagocytosis and improved intraoperative glycemic control with an aggressive insulin infusion (6). However, the issue of hyperglycemia associated with CPB and nondiabetic patients is incompletely understood. Indeed, we were prompted to study this issue in light of the lack of consensus as to the importance of hyperglycemia during CPB in this patient population. Athos J. Rassias, MD Mark P. Yeager, MD