Predictive Value of Preoperative Chest Computed Tomography-Measured Diaphragm Thickness Combined with Pulmonary Function for Postoperative Pulmonary Complications after Anatomic Lung Resection in Elderly Patients with Clinical Stage I Non-Small Cell Lung Cancer.

BackgroundPostoperative pulmonary complications (PPCs) remain a major cause of adverse recovery after lung resection in elderly patients with early-stage non-small cell lung cancer (NSCLC). This study evaluated the predictive value of preoperative chest computed tomography (CT)-measured diaphragm thickness combined with pulmonary function for PPCs after anatomic lung resection.MethodsIn this single-center retrospective study, 248 patients aged 65 years or older with clinical stage I NSCLC who underwent lobectomy or segmentectomy between January 1, 2020 and December 31, 2025 were included. Preoperative CT-measured diaphragm thickness, pulmonary function indices, perioperative variables, and PPCs occurring within 30 days after surgery or during hospitalization were analyzed. Multivariable logistic regression and receiver operating characteristic analyses were performed.ResultsPPCs occurred in 54 of 248 patients (21.8%). Multivariable analysis showed that chronic obstructive pulmonary disease (odds ratio [OR], 2.29; 95% CI, 1.12-4.69; P = .023) and longer operative time (OR per 10-min increase, 1.13; 95% CI, 1.04-1.23; P = .004) were associated with increased PPC risk, whereas greater mean bilateral diaphragm thickness (OR per 0.1-mm increase, 0.88; 95% CI, 0.81-0.96; P = .004) and higher FEV1%pred (OR, 0.96; 95% CI, 0.93-0.99; P = .009) were protective. The area under the curve was 0.791 for mean bilateral diaphragm thickness, 0.752 for FEV1%pred, and 0.842 for the combined model; the bootstrap-corrected AUC was 0.829.ConclusionPreoperative CT-measured mean bilateral diaphragm thickness and FEV1%pred independently predicted PPCs after anatomic lung resection in elderly patients with clinical stage I NSCLC. Their combination provided improved predictive performance for preoperative risk stratification.