Abstract
The global spread of coronavirus disease 2019 (COVID-19) has had a profound public health impact, particularly on perioperative management, rendering the optimization of timing for post-infection thoracic oncologic surgery a pressing clinical concern. This multicenter retrospective cohort study included adult patients who underwent elective video-assisted thoracic oncologic surgery in February 2023 with confirmed COVID-19 infection ≥ 4 weeks prior. A matched historical control cohort from February 2019 was used for comparison. Propensity score matching (PSM) and inverse probability of treatment weighting (IPTW) were applied to adjust for confounders. Subgroup analyses were conducted based on clinical variables, and logistic regression was used to assess the association between infection-to-surgery interval and PPCs. A total of 846 patients were included. After PSM and IPTW, the incidence of PPCs remained comparable between the COVID-19 and no-COVID-19 groups (PSM: 26.1% vs. 31.8%, p = 0.784; IPTW: 28.0% vs. 29.7%, p = 0.615). No significant differences in PPC rates were observed across infection-to-surgery intervals (4–6, 6–8, and 8–12 weeks; p = 0.953). Prior COVID-19 infection was associated with higher postoperative WBC counts and lower lymphocyte levels, but not with increased PPCs risk. Smoking history was an independent predictor of PPCs (OR: 2.503, p = 0.005), while infection timing was not. Thoracic oncologic surgery may be considered ≥ 4 weeks after COVID-19 recovery in carefully selected patients. Further prospective studies are needed to assess safety in earlier postoperative intervals and among patients recovering from severe infection. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1038/s41598-026-39978-3.