Abstract
RATIONALE: Previous investigations have identified several potential predictors of outcomes from lung volume reduction surgery (LVRS). A concern regarding these studies has been their small sample size, which may limit generalizability. We therefore sought to examine radiographic and physiologic predictors of surgical outcomes in a large, multicenter clinical investigation, the National Emphysema Treatment Trial. OBJECTIVES: To identify objective radiographic and physiological indices of lung disease that have prognostic value in subjects with chronic obstructive pulmonary disease being evaluated for LVRS. METHODS: A subset of the subjects undergoing LVRS in the National Emphysema Treatment Trial underwent preoperative high-resolution computed tomographic (CT) scanning of the chest and measures of static lung recoil at total lung capacity (SRtlc) and inspiratory resistance (Ri). The relationship between CT measures of emphysema, the ratio of upper to lower zone emphysema, CT measures of airway disease, SRtlc, Ri, the ratio of residual volume to total lung capacity (RV/TLC), and both 6-month postoperative changes in FEV(1) and maximal exercise capacity were assessed. MEASUREMENTS AND MAIN RESULTS: Physiological measures of lung elastic recoil and inspiratory resistance were not correlated with improvement in either the FEV(1) (R = -0.03, P = 0.78 and R = -0.17, P = 0.16, respectively) or maximal exercise capacity (R = -0.02, P = 0.83 and R = 0.08, P = 0.53, respectively). The RV/TLC ratio and CT measures of emphysema and its upper to lower zone ratio were only weakly predictive of postoperative changes in both the FEV(1) (R = 0.11, P = 0.01; R = 0.2, P < 0.0001; and R = 0.23, P < 0.0001, respectively) and maximal exercise capacity (R = 0.17, P = 0.0001; R = 0.15, P = 0.002; and R = 0.15, P = 0.002, respectively). CT assessments of airway disease were not predictive of change in FEV(1) or exercise capacity in this cohort. CONCLUSIONS: The RV/TLC ratio and CT measures of emphysema and its distribution are weak but statistically significant predictors of outcome after LVRS.