Objectives: Enhanced Recovery After Surgery aims to accelerate recovery, with length of stay as a key metric. This study assessed temporal trends in short-term outcomes within a maturing programme and identified factors associated with increased hospital stay. Methods: Data were prospectively collected for consecutive patients undergoing lung cancer resection following a 14-step protocol between 2013 and 2023. Primary outcome was length of stay. Secondary outcomes included 30-day mortality, morbidity, re-admission, and reoperation rates. Predictors of length of stay were analysed using linear regression. Results: We included 2192 patients; procedures included lobectomy (61%), wedge resection (23%), segmentectomy (10%), pneumonectomy (3.5%), and bi-lobectomy (2.7%). Video-assisted thoracoscopic surgery was used in 80% of cases. Median length of stay decreased from 5 to 4 days (P < .001), while protocol adherence increased from 10/14 to 12/14 (P = .01). In-hospital mortality (2.9% to 1.0%, P < .001) and major morbidity (12.2% to 5.6%, P < .001) both declined. In multivariable linear regression, factors associated with longer stay included age (β = 0.17, CI 0.13 to 0.20, P < .001), higher American Society of Anesthesiologists score (β = 1.12, CI 1.04 to 2.2, P = .02), open surgery (β = 1.0, CI 0.17 to 2.2, P = .043), thoracoscopic-to-open converted surgery (β = 1.49, CI 0.96 to 1.9, P = .03), and intensive care (β = 3.4, CI 2.5 to 4.3, P < .001). Protective factors were early mobilization (β = −0.90, CI −1.9 to 0.33, P = .005) and opioid avoidance (β = −0.72, CI −2.4 to 0.99, P = .038). Conclusions: Sustained use of an Enhanced Recovery After Surgery programme was associated with shorter hospitalization and reduced morbidity. Factors associated with length of stay can identify patients at risk of delayed recovery and prioritize elements for optimization within recovery pathways.

Background/Objectives: Total cell-free DNA (cfDNA) is not tumour-specific and may be affected by peri-operative injury and inflammation. We evaluated peri-operative plasma cfDNA dynamics and explored associations with recurrence and mortality in resectable non-small cell lung cancer (NSCLC). Methods: This prospective pilot cohort included 35 patients undergoing lobectomy with systematic nodal dissection and 30 healthy controls. Samples were collected pre-operatively, from the tumour-draining pulmonary vein before major manipulation, peripherally after resection, and on postoperative day 14. Analyses included group comparisons, receiver operating characteristic analysis, logistic regression and Cox regression. Results: Pre-operative cfDNA was higher in patients than controls (mean 1.96 versus 1.08 ng/µL; p = 0.002). Discrimination was modest (AUC 0.67, 95% CI 0.54–0.80), with sensitivity 49% and specificity 93% at an internally selected cut-off. cfDNA differed across timepoints (overall p = 0.008), with higher intra-operative and post-resection values than pre-operative values. Among 34 patients with complete three-year outcomes, 12 recurred and 13 died. Pre-operative cfDNA was associated with recurrence in multivariable logistic regression (adjusted OR 2.47, 95% CI 1.07–5.67; p = 0.034). Day-14 cfDNA was higher among patients who died (5.35 versus 2.17 ng/µL; p = 0.008). Conclusions: Total cfDNA showed modest diagnostic discrimination and exploratory outcome associations. Small sample size, potential model overfitting and incomplete analytical documentation limit interpretation. These hypothesis-generating findings require independent validation with standardised methods, peri-operative covariates and tumour-informed circulating tumour DNA assays.