Objective: To perform blood gas analysis of the respiratory response to transdermal fentanyl in dogs which have experienced an open-chest surgical procedure. Design: Prospective trial. Setting: Veterinary Teaching Hospital Surgical Research and Student Laboratory. Intervention: Dogs were purchased for a surgical laboratory. Students performed a cranial abdominal exploratory and diaphragmatic hernia repair. Sixteen dogs were divided into 2 groups. Dogs received transdermal fentanyl (group F), using an average dose of 4.8 mg/kg/hr, applied to the caudal-lateral abdomen 22 hours before surgery, or intravenous buprenorphine (group B; 0.02 mg/kg) given 1 hour prior to anesthetic induction and every 6 hours postoperatively. All dogs received intravenous acepromazine (0.05 mg/kg) preoperatively and every 6 hours postoperatively. Dogs were instrumented with carotid artery catheters. Measurements and main results: Arterial blood gas values were analyzed every 2 hours postoperatively. Plasma fentanyl levels were analyzed every 4 hours postoperatively. The mean carbon dioxide tension (PCO 2 ) did not exceed 45 mmHg in either group. The range in mean PCO 2 levels was 32.9 (+ 3.4)À38.1 (+ 3.9) in group B and 34.7 (+ 3.25)À43.6 (+ 5.5) in group F. At 2 time points, the mean PCO 2 was significantly lower in group B compared with normal levels in group F. Hypoxemia occurred in both the groups. The range in mean oxygen tension (PO 2 ) was 76.5 (+ 18.3)À91.1 (+ 16.3) in group B and 76.0 (+ 10.8)À96.6 (+ 7.6) in group F. There was no significant difference in PO 2 between groups. Levels of fentanyl considered to be analgesic were maintained for the postoperative period. Conclusions: The use of a relatively high dose of transdermal fentanyl did not induce postoperative hypoventilation as evidenced by serial arterial blood gas analysis in this model. (J Vet Emerg Crit Care 2002; 12(2): 81±87)