METHODS To evaluate the efficacy of postthoracotomy analgesia with intermittent epidural fentanyl. 50 patients were allocated randomly into 2 groups. The first group received intermittent epidural fentanyl and the second group received intermittent intravenous analgesia using pethidine. The variables studied were: pain score; total amount of additional intravenous opioid analgesia, and ventilatory function parameters [forced vital capacity (FVC), forced expiratory volume in the first second (FEV1) and FEV1/FVC ratio]: RESULTS In the first postoperative day, pain scores were higher in the epidural group (P = 0.034), but there was no significant difference between mean pain scores in the second and third days (P = 0.61, P = 0.15, respectively). On all three days, significantly more additional analgesics were required in the epidural group. A difference was found between both groups in the post- to pre-operative FEV1, FVC and FEV1/FVC ratios, with the better preservation of the ventilatory function in the epidural group (P = 0.001, 0.013, <0.0001, respectively). CONCLUSIONS The analgesic effect of intermittent epidural fentanyl is not adequate and postoperative pain relief has not any significant advantage over the more easily-applied intravenous analgesia. However, better preservation of ventilatory function makes epidural fentanyl a useful adjunct analgesia in reduction of post-thoracotomy pulmonary complications.