A new method to predict postoperative lung function: Quantitative breath sound measurements

Background: Currently, predicted postoperative (PPO) lung function (forced expiratory volume in 1 second [PPO-FEV₁] and diffusion capacity of the lung for carbon monoxide [PPO-Dlco]) estimated from spirometry and regional perfusion is used to select patients for lung resection. Vibration response imaging (VRI) analyzes lung sounds and quantifies regional acoustic energy. Single-center studies suggest that this noninvasive, radiation-free method of quantifying lung function is comparable to the reference standard. Methods: A prospective, multiinstitutional United States study comparing VRI with perfusion in patient assessment for lung resection enrolled 163 patients, with 135 currently available for analysis. PPO values were calculated by subtracting the fraction of segments to be resected in a lung (113 lobectomies, 20 pneumonectomies) multiplied by the percentage of acoustic energy (VRI) or perfusion of that lung. We compared the two methods with each other, with actual postoperative pulmonary function tests, and the rate of complications as predicted by PPO values above or below 40%. Results: Good agreement was found between calculated estimations of postoperative lung function using VRI and perfusion measurements (PPO-FEV₁%: r = 0.95; -8% to 11.5%; PPO-Dlco: r = 0.97; -6.6% to 9.5%), although larger discrepancies were noted between the actual VRI and perfusion measurements (-17 to 24). The VRI and perfusion methods provided excellent agreement in categorization of patients into low or elevated risk based on PPO values of above or below 40% (95% for PPO-FEV₁%; 94% for PPO-Dlco) and similar correlations with actual postoperative values (r = 0.74 and r = 0.67 for FEV₁; r = 0.72 and r = 0.67 for Dlco). Conclusions: VRI may offer a simple, noninvasive, and radiation-free alternative to lung scintigraphy for predicting postoperative lung function in patients with lung malignancies.