National analysis of short-term outcomes after pulmonary resections on cardiopulmonary bypass

Background Pulmonary resections using cardiopulmonary bypass (CPB) are infrequently performed. Their short-term outcomes are not well described. We queried the National Inpatient Sample over a 10-year period (2001 to 2011) to more clearly delineate the short-term outcomes of patients undergoing pulmonary resections on CPB. Methods We identified all patients 18 years and older who underwent pulmonary lobectomy (LB) or pneumonectomy (PN) on CPB; lung transplantations were excluded. We then grouped these patients based on the setting in which bypass was used: LB/PN with planned CPB (group 1), LB/PN with concomitant on-pump cardiac procedure (group 2), or LB/PN requiring CPB secondary to injury (group 3). Demographic data and inhospital outcomes were obtained for each patient. Results In all, 843 patients underwent LB or PN on CPB during the study period. Lobectomies were the most commonly performed procedure overall. Inhospital mortality for groups 1, 2, and 3 were 22% (n = 58), 16% (n = 61), and 57% (n = 115), respectively. Complications were prevalent across all groups. Routine discharge was achieved by fewer than half of all patients: 48% of group 1 (n = 128); 34% of group 2 (n = 129); and 18% of group 3 (n = 36). Pneumonectomy (odds ratio 2.74, 95% confidence interval: 1.00 to 7.53, p = 0.049) as well as using CPB either as part of a combined cardiac surgery (odds ratio 1.48, 95% confidence interval: 0.39 to 5.59, p = 0.002) or because of injury (odds ratio 6.52, 95% confidence interval: 2.13 to 19.99, p = 0.002) were found to be significant multivariate predictors of short-term mortality. Conclusions Pulmonary resections on CPB carry considerable short-term mortality and morbidity, but some risk can be partially mitigated when bypass is planned preoperatively.