Epidural Anesthesia and Analgesia Decrease the Postoperative Incidence of Insulin Resistance in Preoperative Insulin-Resistant Subjects Only Francesco Donatelli, MD* Angelo Vavassori, MD* Simona Bonfanti, MD* Piervirgilio Parrella, SD* Luca Lorini, MD* Roberto Fumagalli, MD† Franco Carli, MD, MPhil‡ BACKGROUND: Insulin resistance (IR) is a feature of the endocrine stress response to surgery. It is not known whether a preoperative state of IR would affect the postoperative endocrine response. We sought to characterize the preoperative state of IR in a group of patients undergoing elective hip and knee arthroplasty, and to determine to what extent perioperative epidural analgesia modifies the postopera- tive state of IR in those who are and are not insulin-resistant before surgery. METHODS: Sixty patients undergoing either hip or knee arthroplasty were screened by using the homeostatic model assessment (HOMA) in two populations: insulin- resistant patients and noninsulin-resistant patients, whereas HOMA is fasting insulin (U/mL)  fasting glucose (mmol/L)/22.5. The patients belonging to each population were then randomly assigned to receive either intraoperative epidural blockade followed by postoperative epidural analgesia (epidural group) or general anesthesia followed by patient-controlled analgesia (control group). Analgesia was assessed with visual analog scale up to 48 h after surgery and HOMA was repeated at the end of surgery and 48 h after surgery to determine the postoperative state of IR. RESULTS: Epidural anesthesia and analgesia significantly influenced the postopera- tive HOMA score (smaller proportion of IR) in the postoperative period only in those patients who were insulin-resistant before surgery (P  0.01). In contrast, noninsulin-resistant patients had a similar postoperative proportion of IR between the epidural and control groups (P  0.05). At rest and during movement, visual analog scale scores were not different between groups at the end of surgery and in the first and second days after surgery. CONCLUSIONS: Epidural anesthesia and analgesia compared to general anesthesia followed by patient-controlled analgesia decreased the incidence of IR soon after surgery and 48 h after surgery only in patients who were insulin-resistant before surgery. (Anesth Analg 2007;104:1587–93) Insulin resistance (IR) is a state in which normogly- cemia is maintained with an elevated insulin concen- tration (1–3). There is a threefold increase in risk for coronary artery disease and stroke in subjects with IR when compared with individuals who have normal glucose tolerance (4 –7). A state of IR has been con- firmed in several different types of stress, including accidental trauma, sepsis, and burning injury (8 –10). However, it is only several years later that studies on IR after surgery have been performed (11,12). These studies found that IR is a feature of the catabolic reactions after major surgery because the postopera- tive increase in the circulating levels of catabolic hormones increases the production of glucose and weakens its peripheral use (13–15). The greatest de- gree of IR has been demonstrated, although not con- sistently, on the second postoperative day after upper abdominal surgery, and normalization seemed to oc- cur after approximately 3 wk (11). It is not clear which factors predict the degree of change in IR after surgery. A previous study (12) involving patients undergoing elective abdominal sur- gery reported that factors such as age, body mass index, and duration of surgery had no influence on the relative increase in IR after surgery. Therefore, the change in postoperative IR is a metabolic variable influenced by the degree of the surgical trauma itself, rather than by predisposing or associated factors. In contrast, other studies found that predisposing and associated factors such as a preoperative IR state and type of anesthesia influenced glucose metabolism after surgery (16 –18). From the *Department of Cardiovascular Medicine, Ospedali Riuniti di Bergamo, Largo Barozzi n. 3, Bergamo, Italy; †Department of Anesthesia and Intensive Care, Universita ` degli Studi Milano Bicocca Via Cadore 48, Monza, Italy; and ‡Department of Anesthe- sia, McGill University Health Centre, Montreal, Quebec, Canada. Accepted for publication February 14, 2007. Address correspondence and reprint requests to Franco Carli, Department of Anesthesia, McGill University Health Centre, Cedar Avenue Room #D10.144, Montreal, Quebec, Canada 1650. Address e-mail to franco.carli@mcgill.ca. Copyright © 2007 International Anesthesia Research Society DOI: 10.1213/01.ane.0000261506.48816.5c Vol. 104, No. 6, June 2007 1587