Cardiac reoperations in patients with prior thoracic radiation: A single-center analysis of outcomes and key lessons

OBJECTIVES: Thoracic radiation therapy (TRT) increases the risk of primary and reoperative cardiac surgery (RCS). However, the influence of radiation extent on RCS is unknown. Thus, we examine the outcomes of RCS in patients with prior TRT, comparing mantle versus nonmantle radiation. METHODS: The cohort comprised all adult patients with prior TRT who underwent RCS at a single center from March 2011 to March 2025. The cohort was matched (1:3) using baseline characteristics and operative components to patients who underwent RCS without prior TRT. Subanalysis of patients with prior TRT compared those with mantle versus nonmantle radiation. RESULTS: Seven hundred eight patients were included, comprising 177 in the TRT cohort (65 mantle, 112 nonmantle) and 531 in the matched no-TRT cohort. Compared with matched controls, TRT patients had higher rates of postoperative renal failure requiring dialysis (16.9% vs 8.3%; risk ratio, 2.04 [95% CI, 1.33-3.15]; P = .001) and hospital stay (28.2 vs 19.8 days; mean difference, 8.4 days [95% CI, 3.5-13.3 days]; P = .01). Operative mortality was higher in the TRT cohort (11% vs 5.5%; risk ratio, 2.28 [95% CI, 1.34-3.86]; P = .01). In subanalysis, mantle radiation patients had longer hospital stay compared with those receiving nonmantle radiation (35.3 ± 35.4 vs 24.3 ± 23.9 days; P = .03), higher rate of mechanical circulatory support (13.8% vs 4.4%; P = .026), delayed chest closure (12.3% vs 2.6%; P = .011), and operative mortality (15.4% vs 8%; P = .06). Mantle radiation patients had worse 1- and 5-year survival (60% vs 86% and 36% vs 61%, respectively; P = .001). CONCLUSIONS: Prior TRT is associated with increased postoperative complications and operative mortality after RCS. Notably, mantle radiation appears to confer a greater risk than nonmantle radiation, with worse survival and a trend toward higher operative mortality, although larger studies are needed to confirm this observation.