In order to determine the incidence of postoperative pulmonary complications (POPC) and the value of preoperative spirometry to predict pulmonary complications after upper abdominal surgery, 24 women and 36 men (total 60 patients) were studied prospectively (mean age 48.3 years). On the day before the operation and for 15 days after the operation, each patients's respiratory status was assessed by clinical examination, chest radiography, spirometry and blood gas analysis, and patients were monitored for pulmonary complications by a chest physician and a surgeon independently. In this study, postoperative pulmonary complications developed in 21 (35%) patients (pneumonia in 10 patients, bronchitis in nine patients, atelectasis in one patient, pulmonary embolism in one patient). Of 31 patients with abnormal preoperative spirometry, 14 (45.2%) patients showed complications, whereas among 29 patients with normal preoperative spirometry, 7 (24.1%) patients showed complications (P<0.05). The incidence of POPC was higher in patients with advanced age, smoking, preoperative abnormal findings obtained from physical examination of the chest, higher ASA class and longer duration of operation. The sensitivity (0.76) and specificity (0.79) of abnormal preoperative findings obtained from physical examination to predict POPC were higher than abnormal preoperative spirometry (0.67 and 0.56 retrospectively). There was no significant difference between patients with and without pulmonary complications in regard to weight, serum albumin, type of incision, incidence of abnormal preoperative blood gases and duration of postoperative hospital stay. We conclude that POPC is still a serious cause of postoperative morbidity.