Purpose: We delineate predictive factors of pulmonary morbidity in patients who receive combination chemotherapy with bleomycin and undergo surgical resection of residual disease, and establish updated guidelines for perioperative management. Materials and Methods: A total of 77 patients with high volume stage II to IV nonseminomatous germ cell tumors underwent 97 major surgical procedures a mean of 6.4 months following high dose combination chemotherapy, including bleomycin (mean 437.5 units per 8.2 courses), between 1988 and 1995 at the University of Texas M. D. Anderson Cancer Center. The importance of preoperative pulmonary status, anesthesia time, fraction of inspired oxygen, fluid balance, bleomycin dose, number of acute toxicity episodes, oxygen saturation problems and pulmonary symptoms was examined. Cases were divided into groups according to whether there were postoperative oxygen saturation problems (19) or not (58). Results: There were no significant differences in age, weight, bleomycin dose, number of acute toxicity episodes, cardiac ejection fraction or preoperative pulmonary symptoms between the 2 groups. Restrictive spirometry patterns were seen in 26 of 74 patients (35%), only 9 of whom had postoperative oxygen saturation problems. Mean induction fractional inspired oxygen was 87% (median 100%) for an average of 56 minutes. Intraoperative fractional inspired oxygen averaged 40% for a mean duration of 8.1 hours. Postoperative oxygen saturation problems, consisting of prolonged intubation, pulmonary edema, dyspnea, tachypnea or desaturation requiring diuresis, occurred in 19 patients (25%). Surgery/anesthesia time, amount of blood transfused, estimated blood loss, fluid balance, type of fluid given (all p <0.0001) and preoperative forced vital capacity (p = 0.012) were significant predictors of postoperative oxygen saturation problems on univariate analysis. On multivariate analysis only the amount of blood transfused, preoperative forced vital capacity and surgical time in descending order remained significant. Maintained intraoperative fractional inspired oxygen was not significant on either analysis. There were no deaths. Conclusions: Perioperative oxygen restriction in patients treated with bleomycin is not necessary. Intravenous fluid management, including transfusion, appears to be the most significant factor affecting postoperative pulmonary morbidity and overall clinical outcome. In addition, post-chemotherapy forced vital capacity and operative time are significant predictive factors of procedure related pulmonary morbidity.
