Predictors of clinically relevant postoperative morbidity following curative intent surgical resection for non-small cell lung cancer.

Postoperative morbidity remains an important challenge following curative-intent surgical treatment for non-small cell lung cancer (NSCLC), despite advances in minimally invasive thoracic surgery and perioperative management. Identification of factors associated with clinically relevant postoperative complications may improve perioperative risk stratification and patient selection. The aim of the present study was to evaluate predictors of clinically relevant postoperative morbidity following pulmonary resection for NSCLC in a single tertiary oncologic center. A retrospective observational cohort study was conducted that included 163 consecutive patients who underwent curative-intent anatomical pulmonary resection for NSCLC. Clinically relevant postoperative morbidity was defined as Clavien-Dindo grade ≥ II. Demographic, clinical, functional, and perioperative variables were analyzed, including age, Charlson Comorbidity Index (CCI), pulmonary function (FEV1), surgical approach, pathological stage, and type of resection. Comparative analysis between patients with and without clinically relevant morbidity was performed. Univariate and multivariate logistic regression analyses were subsequently performed to identify factors associated with postoperative morbidity. Model performance was evaluated using receiver operating characteristic (ROC) curve analysis. Clinically relevant postoperative morbidity occurred in 21.5% of patients. Patients who developed postoperative complications tended to be older and presented lower preoperative FEV1 values compared with patients without clinically relevant morbidity. Increased CCI demonstrated an association with increased postoperative morbidity risk (adjusted OR = 2.04, 95% CI, 0.54-7.70), while lower FEV1 values were associated with increased postoperative risk (adjusted OR = 0.98, 95% CI, 0.95-1.01). Increasing age also demonstrated a modest association with postoperative complications (adjusted OR = 1.02, 95% CI, 0.97-1.07). The final multivariate model demonstrated fair discriminatory performance, with an area under the ROC curve (AUC) of 0.645. Clinically relevant postoperative morbidity following NSCLC surgery appears to be influenced by multiple patient-related and procedure-related factors, particularly comorbidity burden, pulmonary reserve, and age. Although the predictive performance of the present exploratory model was modest, careful preoperative evaluation and individualized perioperative risk stratification remain essential in thoracic oncologic surgery.