Determination of preoperative pulmonary function is crucial in avoiding complications from pulmonary resection. Many have employed static pulmonary function testing in an attempt to decrease morbidity and mortality from lung resections. The purpose of the present study was to correlate preoperative static pulmonary function, one-second forced expiratory volume (FEV1), and exercise O2 consumption (MVO2) with postoperative morbidity and mortality. Fifty consecutive patients underwent preoperative FEV1 and MVO2 determinations. A criterion for surgical resection included an FEV1 greater than 1.7 liters for pneumonectomy, greater than 1.2 liters for lobectomy, and greater than 0.9 liters for wedge resection. The surgeon was blinded as to the results of MVO2 studies. Mean age was 63.8 years (range, 47 to 76 years). There were 10 pneumonectomies, 28 lobectomies, and 12 wedge resections. Among the 50 surgical candidates selected solely on the standard FEV1 values, mortality was 4% (2/50) and morbidity, 12% (6/50). Stratification on the basis of exercise performance showed a 29% mortality (2/7) and a 43% morbidity (3/7) in patients with an MVO2 less than 10 ml/kg/min. Patients with an MVO2 less than 20 but greater than 10 ml/kg/min had a 10.7% morbidity (3/28), and there were no deaths. No patients with an MVO2 greater than 20 ml/kg/min sustained any morbidity or died (p less than 0.001). We conclude that exercise is an important criterion in the preoperative evaluation of patients for pulmonary surgery. An MVO2 less than 10 ml/kg/min is associated with significant morbidity and mortality.