Abstract
INTRODUCTION: According to the guidelines for preoperative assessment of lung resection candidates, patients with normal forced expiratory volume in 1 s (FEV(1)) and diffusing capacity of the lung for carbon monoxide (D (LCO)) are at low risk for post-operative pulmonary complications (PPC). However, PPC affect hospital length of stay and related healthcare costs. We aimed to assess risk of PPC for lung resection candidates with normal FEV(1) and D (LCO) (>80% predicted) and identify factors associated with PPC. METHODS: 398 patients were prospectively studied at two centres between 2017 and 2021. PPC were recorded from the first 30 post-operative days. Subgroups of patients with and without PPC were compared and factors with significant difference were analysed by uni- and multivariate logistic regression. RESULTS: 188 subjects had normal FEV(1) and D (LCO). Of these, 17 patients (9%) developed PPC. Patients with PPC had significantly lower pressure of end-tidal carbon dioxide (P (ETCO(2)) ) at rest (27.7 versus 29.9; p=0.033) and higher ventilatory efficiency (V'(E)/V'(CO(2)) ) slope (31.1 versus 28; p=0.016) compared to those without PPC. Multivariate models showed association between resting P (ETCO(2)) (OR 0.872; p=0.035) and V'(E)/V'(CO(2)) slope (OR 1.116; p=0.03) and PPC. In both models, thoracotomy was strongly associated with PPC (OR 6.419; p=0.005 and OR 5.884; p=0.007, respectively). Peak oxygen consumption failed to predict PPC (p=0.917). CONCLUSIONS: Resting P (ETCO(2)) adds incremental information for risk prediction of PPC in patients with normal FEV(1) and D (LCO). We propose resting P (ETCO(2)) be an additional parameter to FEV(1) and D (LCO) for preoperative risk stratification.