Forrest plot assessing interaction between preselected baseline factors and the effect of combined epidural–general anesthesia versus general anesthesia alone on recurrence-free survival. The estimated overall hazard ratio was derived from a multivariable Cox proportional hazards model adjusted for age, sex, smoking history, American Society of Anesthesiologists (ASA) physical status, tumor-node-metastasis stage, and postoperative anticancer therapy. For the subgroup analyses, we assessed the treatment-by-covariate interaction on the primary outcome, adjusting for the same baseline variables.