Association of atrial septal fenestration with outcomes after atrioventricular septal defect repair

Objective: During repair of atrioventricular septal defect (AVSD), surgeons might leave an atrial level shunt when concerned about postoperative physiology, or as part of routine practice. However, the association of fenestration with outcomes is unclear. We sought to determine factors associated with mortality after biventricular repair of AVSD. Methods: We included 581 patients enrolled from 32 Congenital Heart Surgeons' Society institutions from January 1, 2012, to June 1, 2020 in the Congenital Heart Surgeons' Society AVSD cohort. Parametric multiphase hazard analysis was used to identify factors associated with mortality. A random effect model was used to account for possible intersite variability in mortality. Results: An atrial fenestration was placed during repair in 133/581 (23%) patients. Overall 5-year survival after repair was 91%. Patients who had fenestration had an 83% 5-year survival versus 93% for those not fenestrated (P < .001). Variables associated with mortality in multivariable hazard analysis included institutional diagnosis of ventricular unbalance (hazard ratio [HR], 2.7 [95% confidence interval (CI): 1.5-4.9]; P = .003), preoperative mechanical ventilation (HR, 4.1 [95% CI, 1.3-13.1]; P = .02), atrial fenestration (HR, 2.8 [95% CI, 1.5-4.9]; P < .001), and reoperation for ventricular septal defect (HR, 4.0 [95% CI, 1.3-13.1]; P = .002). There was no difference in measures of ventricular unbalance for comparisons of fenestrated with nonfenestrated patients. No significant interinstitution variability in mortality was observed on the basis of the random effect model (P = .7). Conclusions: An atrial communication at biventricular repair of AVSD is associated with significantly reduced long-term survival after adjusting for other known associated factors, including unbalance. These findings might challenge the routine practice of fenestration.