The Long-term Effect of Fenestration Status on Clinical Outcomes: A Fontan Outcome Registry Using CMR Examination (FORCE) Study (RF_FR_390)

Background: A fenestration has shown early postoperative benefits in patients who undergo the Fontan procedure. However, the long-term effects of an open fenestration are unclear, leading to wide practice style variations of fenestration creation and closure. This study aimed to evaluate the long-term outcomes of the fenestration using data from the multi-institutional Fontan Outcome Registry Using CMR Examination (FORCE).

Methods: We included 2,714 patients and categorized them by fenestration status reported at the first post-Fontan CMR (open, spontaneous closure, surgical or device closure, and non-fenestrated Fontan). Echocardiograms within two years were cross-referenced to ascertain the fenestration status. Patients who had undergone a fenestration creation between Fontan and CMR, usually to address a complication, were excluded. We defined a composite outcome as the earliest occurrence of death, listing or receiving a heart transplant, plastic bronchitis, or protein-losing enteropathy from time of CMR. We examined CMR function measurements across fenestration status categories, and studied the association of fenestration with the composite outcome using multivariable (MV) Cox proportional hazard models. Model 1 included clinical variables only while Model 2 included clinical and CMR variables. A nested analysis was performed to account for patients nested within unique centers.

Results: The cohort had a median age at CMR of 14.3 years [IQR 10.3, 18.7] (41% female). Patients with open fenestration were younger, less likely to have a systemic left ventricle, and had higher indexed ventricular mass (Table 1). Non-fenestrated Fontan patients were more likely to have a systemic left ventricle, lower indexed single ventricle end-diastolic volume (SVEDV), higher ejection fraction, and lower indexed ventricular mass. The median follow-up time to outcome was 3.75 years [IQR 1.09, 7.25]. Dominant ventricular morphology, age at Fontan, and number of cardiac surgeries were associated with the outcome in the MV model (Table 2). Compared to a non-fenestrated Fontan, an open fenestration was associated with an increased risk of adverse outcomes in Model 1 adjusting for clinical covariates (HR 1.65 [1.06, 2.58], p = 0.03). Device or surgical and spontaneous fenestration closure were not associated with an increased risk of the outcome. When CMR parameters were included in Model 2, every 10ml/m2 increase in SVEDV was associated with a 5% increase in the hazard of clinical outcome, while no further association remained between fenestration status and outcome.

Conclusion: The current practice of fenestration closure in selected patients does not worsen clinical outcomes. Open fenestration at post-Fontan CMR is not an independent risk factor of adverse outcomes after accounting for EDV, while EDV remained a significant predictor. The findings likely reflect valid clinical decision-making in keeping the fenestration open in higher-risk patients with worse cardiac function and more dilated ventricles by CMR.

Kaplan-Meier curve of freedom from a composite clinical outcome of death, listing or receiving heart transplant, plastic bronchitis, and protein losing enteropathy, from the time of first post-Fontan CMR. Patients were stratified by fenestration status (open fenestration or baffle leak, spontaneous closure, device or surgical closure, non-fenestrated Fontan).