BACKGROUND: The authors applied an optimization model (direct search) to find the optimal combination of bupivacaine dose, fentanyl dose, clonidine dose, and infusion rate for continuous postoperative epidural analgesia. METHODS: One hundred ninety patients undergoing 48-h thoracic epidural analgesia after major abdominal surgery were studied. Combinations of the variables of bupivacaine dose, fentanyl dose, clonidine dose, and infusion rate were investigated to optimize the analgesic effect (monitored by verbal descriptor pain score) under restrictions dictated by the incidence and severity of side effects. Six combinations were empirically chosen and investigated. Then a stepwise optimization model was applied to determine subsequent combinations until no decrease in the pain score after three consecutive steps was obtained. RESULTS: Twenty combinations were analyzed. The optimization procedure led to a reduction in the incidence of side effects and in the mean pain scores. The three best combinations of bupivacaine dose (mg/h), fentanyl dose (microg/h), clonidine dose (microg/h), and infusion rate (ml/h) were: 9-21-5-7, 8-30-0-9, and 13-25-0-9, respectively. CONCLUSIONS: Given the variables investigated, the aforementioned combinations may be the optimal ones to provide postoperative analgesia after major abdominal surgery. Using the direct search method, the enormous number of possible combinations of a therapeutic strategy can be reduced to a small number of potentially useful ones. This is accomplished using a scientific rather than an arbitrary procedure.
BACKGROUND: The authors applied an optimization model (direct search) to find the optimal combination of bupivacaine dose, fentanyl dose, clonidine dose, and infusion rate for continuous postoperative epidural analgesia. METHODS: One hundred ninety patients undergoing 48-h thoracic epidural analgesia after major abdominal surgery were studied. Combinations of the variables of bupivacaine dose, fentanyl dose, clonidine dose, and infusion rate were investigated to optimize the analgesic effect (monitored by verbal descriptor pain score) under restrictions dictated by the incidence and severity of side effects. Six combinations were empirically chosen and investigated. Then a stepwise optimization model was applied to determine subsequent combinations until no decrease in the pain score after three consecutive steps was obtained. RESULTS: Twenty combinations were analyzed. The optimization procedure led to a reduction in the incidence of side effects and in the mean pain scores. The three best combinations of bupivacaine dose (mg/h), fentanyl dose (microg/h), clonidine dose (microg/h), and infusion rate (ml/h) were: 9-21-5-7, 8-30-0-9, and 13-25-0-9, respectively. CONCLUSIONS: Given the variables investigated, the aforementioned combinations may be the optimal ones to provide postoperative analgesia after major abdominal surgery. Using the direct search method, the enormous number of possible combinations of a therapeutic strategy can be reduced to a small number of potentially useful ones. This is accomplished using a scientific rather than an arbitrary procedure.