Carbon monoxide lung diffusion capacity improves risk stratification in patients without airflow limitation: evidence for systematic measurement before lung resection

Objective: In many centers, carbon monoxide Lung diffusion capacity (DLCO) is still not routinely measured in all patients but only inpatients with airflow limitation. The objective of the study was to assess the degree of correlation between forced expiratory volume in 1 s (FEV1) and DLCO, and verify whether a low predicted postoperative DLCO (ppoDLCO) could have a rote in predicting complications in patients without airflow limitation. Methods: We analyzed 872 patients submitted to lung resection between January 2000 and December 2004 in two units measuring systematically DLCO before operation. Correlation between FEV1 and DLCO was assessed in the entire dataset and in different subsets of patients. A number of variables were then tested for a possible association with postoperative cardiopulmonary complications in patients with FEV1 > 80% by univariate analysis. Variables with p < 0.10 at univariate analysis were used as independent variables in a stepwise logistic regression analysis (dependent variable: presence of cardiopulmonary morbidity), which was in turn validated by bootstrap analysis. Results: The correlation coefficients between FEV1 and DLCO in the entire dataset and in different subsets of lung resection candidates (stratified by age, gender, cause of operation, airflow limitation) were all below 0.5, showing a modest degree of correlation. Two hundred and nineteen of the 508 patients (43%) with FEV1 > 80% had DLCO < 80%. Moreover, in patients with FEV1 > 80%, logistic regression analysis showed that ppoDLCO < 40% was a significant and reliable predictor of postoperative complications (p = 0.004). Conclusion: The modest correlation between FEV1 and DLCO and the capacity of ppoDLCO to discriminate between patients with and without complications in subjects with a normal FEV1, warrants the routine measurement of DLCO in all candidates for lung resection, irrespective of their FEV1 value, in order to improve surgical risk stratification. (c) 2006 Elsevier B.V. All rights reserved.