Isolated (IHV) drainage can prevent uniform delivery of hepatic venous effluent to the pulmonary circulation after Fontan completion, predisposing to pulmonary arteriovenous malformations (PAVMs), hypoxemia, and thrombosis. Comparative data across operative strategies remain limited. Accordingly, Fontan configurations that differ in how hepatic venous return is incorporated and distributed within the cavopulmonary pathway may plausibly influence oxygenation and PAVM risk, although direct flow quantification is required for definitive mechanistic confirmation. We conducted a multicenter retrospective cohort study. Among 988 Fontan procedures performed between 2014 and 2025, we identified 99 patients with IHV drainage who underwent Fontan completion between 2014 and 2020 and had ≥ 5 years of follow-up. Patients received one of three techniques: intra-extracardiac Fontan (IECF, n = 42), lateral tunnel (LTT, n = 17), or extracardiac Fontan with hepatic-vein graft incorporation (ECF-HV, n = 40). Prespecified 5-year primary outcomes were oxygen saturation, PAVM, and thrombosis. Secondary outcomes included mean pulmonary artery pressure (PAP), arrhythmia, protein-losing enteropathy (PLE), pleural effusion, NYHA class, and mortality. Group comparisons used nonparametric tests with Dunn post-hoc testing and exact tests; sensitivity analyses excluded fenestrated patients. Baseline demographics, pulmonary artery indices, and pre-Fontan hemodynamics were similar across groups. At 5 years, oxygen saturation was higher with IECF (mean [SD] 92 [4]%) than with LTT (85 [6]%; p = 0.002) and ECF-HV (89 [5]%; p = 0.001). Mean PAP was lower with IECF (12 [2] mmHg) than with LTT (14 [3] mmHg; p = 0.001) and ECF-HV (13 [2] mmHg; p = 0.003). PAVM incidence was 4.8% with IECF versus 29.4% with LTT and 7.5% with ECF-HV (overall p = 0.017). Thrombosis occurred in 2.4% with IECF versus 29.4% with LTT and 5.0% with ECF-HV (overall p = 0.002). NYHA class distribution favored IECF but did not reach statistical significance (overall p = 0.12). Kaplan–Meier analysis showed no statistically significant difference in 5-year survival (log-rank p = 0.11). Findings were consistent after excluding fenestrated patients. In patients with IHV drainage undergoing Fontan completion, IECF was associated with higher 5-year oxygen saturation, lower PAP, and lower rates of PAVMs and thrombosis compared with LTT and ECF-HV, without a statistically significant difference in survival by log-rank testing. These findings may reflect differences in hepatic venous incorporation and pathway geometry; however, direct flow-based assessment (e.g., 4D-flow MRI, MRI-based energy-loss analysis, angiographic flow mapping, or computational flow modeling) was not performed, and mechanistic interpretation should therefore remain speculative.

Outcomes of Fontan Modifications for Isolated Hepatic Vein Drainage: 5 Years Comparative Analysis

期刊论文
发表日期：

2026年5月9日

查看全文

Onur Benli,
Mustafa Kemal Avşar

摘要

Isolated (IHV) drainage can prevent uniform delivery of hepatic venous effluent to the pulmonary circulation after Fontan completion, predisposing to pulmonary arteriovenous malformations (PAVMs), hypoxemia, and thrombosis. Comparative data across operative strategies remain limited. Accordingly, Fontan configurations that differ in how hepatic venous return is incorporated and distributed within the cavopulmonary pathway may plausibly influence oxygenation and PAVM risk, although direct flow quantification is required for definitive mechanistic confirmation. We conducted a multicenter retrospective cohort study. Among 988 Fontan procedures performed between 2014 and 2025, we identified 99 patients with IHV drainage who underwent Fontan completion between 2014 and 2020 and had ≥ 5 years of follow-up. Patients received one of three techniques: intra-extracardiac Fontan (IECF, n = 42), lateral tunnel (LTT, n = 17), or extracardiac Fontan with hepatic-vein graft incorporation (ECF-HV, n = 40). Prespecified 5-year primary outcomes were oxygen saturation, PAVM, and thrombosis. Secondary outcomes included mean pulmonary artery pressure (PAP), arrhythmia, protein-losing enteropathy (PLE), pleural effusion, NYHA class, and mortality. Group comparisons used nonparametric tests with Dunn post-hoc testing and exact tests; sensitivity analyses excluded fenestrated patients. Baseline demographics, pulmonary artery indices, and pre-Fontan hemodynamics were similar across groups. At 5 years, oxygen saturation was higher with IECF (mean [SD] 92 [4]%) than with LTT (85 [6]%; p = 0.002) and ECF-HV (89 [5]%; p = 0.001). Mean PAP was lower with IECF (12 [2] mmHg) than with LTT (14 [3] mmHg; p = 0.001) and ECF-HV (13 [2] mmHg; p = 0.003). PAVM incidence was 4.8% with IECF versus 29.4% with LTT and 7.5% with ECF-HV (overall p = 0.017). Thrombosis occurred in 2.4% with IECF versus 29.4% with LTT and 5.0% with ECF-HV (overall p = 0.002). NYHA class distribution favored IECF but did not reach statistical significance (overall p = 0.12). Kaplan–Meier analysis showed no statistically significant difference in 5-year survival (log-rank p = 0.11). Findings were consistent after excluding fenestrated patients. In patients with IHV drainage undergoing Fontan completion, IECF was associated with higher 5-year oxygen saturation, lower PAP, and lower rates of PAVMs and thrombosis compared with LTT and ECF-HV, without a statistically significant difference in survival by log-rank testing. These findings may reflect differences in hepatic venous incorporation and pathway geometry; however, direct flow-based assessment (e.g., 4D-flow MRI, MRI-based energy-loss analysis, angiographic flow mapping, or computational flow modeling) was not performed, and mechanistic interpretation should therefore remain speculative.