Background: Postoperative morbidity and mortality in destroyed lung are relatively high. We tried to identify the prognostic factors affecting postoperative morbidity and mortality in destroyed lung through a retrospective study. Material and method: The retrospective study was undertaken in 112 patients who had undergone pneumonectomy or pleuropneumonectomy for destroyed lung at Severance Hospital from 1970 to 2000. We analyzed the correlation between postoperative morbidity and mortality and etiology, duration of disease, preoperative FEV1, presence or absence of peroperative empyema, operation timing, the side of operation, duration of operation, and operation type. Result: There were 55 men and 57 women, aged 20 to 81 years (mean 44 years). Etiologic diseases were tuberculosis in 86 patients(76.8%) including tuberculos empyema in 20 and tuberculous bronchiectasis in 4, pyogenic empyema in 12(10,7%), bronchiectasis in 12(10.7%), and lung abscess in 2(1.8%). Postoperative morbidity were 25%(n=28) and postoperative mortality was 6%(n=7). The presence of preoperative empyema(p=0.016), pleuropneumonectomy(p=0.037) and preoperative FEV1 of less than 1.75 L(P=0.048) significantly increased the postoperative morbidity, If operation time was less than 300min, postoperative morbidity(p=0.002) and mortality(p=0.03) were significantly low. Conclusion: Postoperative morbidity and mortality in destroyed lung were acceptable. Postoperative morbidity and mortality were significantly low when operation time was less than 300 min. Preoperative existence of empyema, pleuropneumonectomy and preoperative FEV1 of less than 1.75 L significantly increased postoperative morbidity.