Predictors of operative mortality and cardiopulmonary morbidity in the National Emphysema Treatment Trial

We sought to identify predictors of operative mortality, pulmonary morbidity, and cardiovascular morbidity after lung volume reduction surgery. Univariate and multivariate logistic regression analyses were performed. Candidate predictors included demographic characteristics, physical condition characteristics, pulmonary function measures, measures of the distribution of emphysema as determined by radiologists and by means of computerized analysis of chest computed tomographic scans, and measures of exercise capacity, dyspnea, and quality of life. End points analyzed were operative mortality (death within 90 days of the operation), major pulmonary morbidities (tracheostomy, failure to wean, reintubation, pneumonia, and ventilator for ≥3 days), and cardiovascular morbidities (infarction, pulmonary embolus, or arrhythmia requiring treatment). Five hundred eleven patients in the non–high-risk group of the National Emphysema Treatment Trial underwent lung volume reduction. The incidence of operative mortality was 5.5%, major pulmonary morbidity occurred in 29.8% of patients, and cardiovascular morbidity occurred in 20.0% of patients. Predictors for these end points are as follows: Relative odds P value Operative mortality Non–upper-lobe predominance (radiologist) 2.99 .009 Pulmonary morbidity Age in years 1.05 .02 FEV 1 % predicted 0.97 .05 D lco % predicted 0.97 .01 Cardiovascular morbidity Age in years 1.07 .004 Oral steroid use 1.72 .04 Non–upper-lobe predominance (QIA α measure) 2.67