Perioperative and Mid-Term Outcomes of Primary Versus Redo Pericardiectomy for Constrictive Pericarditis: A...

Background: Direct comparative evidence between redo and primary pericardiectomy for constrictive pericarditis (CP) is limited to small, uncontrolled series. We compared perioperative and mid-term outcomes between the two cohorts and explored factors associated with all-cause death. Methods: In this single-center retrospective cohort study, 137 consecutive patients with CP undergoing radical pericardiectomy between April 2018 and December 2023 were allocated to a primary pericardiectomy cohort (PPC, n = 86) or a redo pericardiectomy cohort (RPC, n = 51). The primary endpoint was all-cause death; secondary endpoints included low cardiac output syndrome (LCOS), unplanned reoperation, acute kidney injury (AKI), atrial fibrillation, and postoperative New York Heart Association class. Survival was analyzed with Kaplan-Meier curves and log-rank tests, prognostic factors with Firth-penalized Cox regression (with baseline adjustment and multiple imputation), and the origin of the survival difference was explored via a sensitivity analysis excluding operative deaths. Results: The RPC had more preoperative atrial fibrillation (64.7% vs. 36.0%, p = 0.002) and a larger left atrial diameter (54.8 vs. 47.0 mm, p < 0.001) than the PPC. Operative mortality was higher in the RPC (15.7% vs. 2.3%; relative risk 6.745, p = 0.006), as was LCOS (13.7% vs. 3.5%, p = 0.026). Seventeen patients died (6 PPC, 11 RPC) over a median follow-up of 3.6 years; cumulative overall survival was lower in the RPC (log-rank p = 0.013; 3-year survival 78.4% vs. 94.2%). Postoperative LCOS was the strongest independent predictor of death (hazard ratio [HR] 22.526, p < 0.001); redo pericardiectomy (HR 3.031; 95% CI 1.034-9.679; p = 0.043) and AKI stage (HR 1.848; 95% CI 1.107-3.088; p = 0.020) were independently associated with death; postoperative albumin showed a borderline protective association (HR 0.913, p = 0.048). Among perioperative survivors, no statistically significant difference in long-term survival was detected (HR 1.275; 95% CI 0.279-5.819; p = 0.753). Conclusions: Redo pericardiectomy was associated with substantially higher perioperative mortality; among perioperative survivors, long-term survival did not differ detectably between cohorts, although estimates were imprecise. Because the excess risk was accompanied by potentially modifiable perioperative events, intensified risk assessment and meticulous perioperative management warrant evaluation as approaches to improving outcomes in this high-risk population.