Risk Factors for Persistent Atelectasis and Development of a Predictive Nomogram After Video-Assisted Thoracoscopic Lobectomy in Non-Small Cell Lung Cancer Patients: A Retrospective Cohort Study

Background: On most postoperative imaging examinations, atelectasis after lobectomy is a transient finding. We focused on the more clinically consequential scenario: persistent collapse of at least one segment of the lung beyond the first postoperative week. These cases may delay mobilisation, impair secretion clearance, and increase the need for additional airway care after video-assisted thoracoscopic (VATS) lobectomy for non-small cell lung cancer (NSCLC). This study investigated perioperative predictors of this specific endpoint and used the identified factors to develop a nomogram.Methods: We reviewed 1268 consecutive NSCLC patients treated with VATS lobectomy from January 2020 to December 2024. Persistent atelectasis meant radiographic collapse of at least one pulmonary segment for 7 or more postoperative days. Before modelling, 40 candidate predictors were coded. We then reviewed these predictors according to clinical domain, avoided retaining overlapping variables that represented the same construct, and performed 1000 bootstrap resamples for internal validation. Age, smoking exposure, forced expiratory volume in 1 second (FEV1), and operative time were additionally examined as continuous variables, and restricted cubic splines were used to assess possible nonlinear associations.Results: Persistent atelectasis was recorded in 368 patients (29.0%). Twenty non-redundant variables were included in the adjusted analysis, corresponding to 18.4 events per variable. The final score retained age ≥65 years, smoking history ≥20 pack-years, chronic obstructive pulmonary disease (COPD), FEV1 <60% predicted, upper lobe resection, and operative time ≥180 minutes (all p < 0.001). The apparent area under the receiver operating characteristic curve (AUC) was 0.842 (95% confidence interval (CI) 0.818–0.865); the bootstrap-corrected AUC was 0.837 (95% CI 0.813–0.861). Calibration was acceptable (Hosmer-Lemeshow chi-square = 5.41, p = 0.714; slope = 0.92). Observed rates were 10.2%, 38.1%, and 79.0% in the low-, intermediate-, and high-risk groups (p for trend <0.001). Sensitivity analyses using continuous predictors produced similar discrimination, and spline analyses showed significant overall associations for age, smoking exposure, FEV1, and operative time.Conclusion: In this cohort, persistent atelectasis was associated with pulmonary reserve, chronic airway disease, lobe removal, and the length of the operation. Because operative time is included in the score, the complete nomogram is best applied during perioperative or early postoperative care rather than preoperatively. Therefore, further multicentre validation remains necessary.