Purpose or Case Report: Portal hypertension is a manifestation of Fontan-related venous congestion and hepatic fibrosis. We studied the prevalence of radiologic evidence of portal hypertension (RP) and its association with Fontan hemodynamics and adverse outcomes. We also tested the validity of the VAST score, used in a prior study, in our population.
Methods & Materials: Fontan patients who underwent cross-sectional abdominal imaging between 2012 and 2019 were retrospectively reviewed. RP was defined as the presence of ≥2 of the following: varices (V), ascites (A), or splenomegaly (S). The VAST score was calculated as 1 point for each of V, A, S, or T (thrombocytopenia). High Fontan pressure was defined as Fontan pressure >13 mm Hg on catheterization. Fontan failure (FF) was defined as diuretic escalation, ventricular assist device placement, heart transplant, or death. Associations were assessed using univariate and multivariate logistic regression.
Results: We included 124 patients with mean age and time since Fontan of 24±9 and 17±7 years, respectively. The overall distribution of V, A, S, and T was 18 (14%), 26 (21%), 36 (30%, n=119), and 36 (29%), respectively. 25 (20%) patients met the definition of RP, and 35 (28%) patients had a VAST score ≥2. FF was noted in 27 (22%) patients at the time of abdominal imaging or during follow-up.

On univariate analysis, patients with RP were more likely to have high Fontan pressure (79% vs 30%, p<0.0001) measured at a median interval of 1.0 (IQR 0.2-2.8) year from abdominal imaging. They were also more likely to experience FF (OR 4.1, 95% CI 1.6-10.6, p=0.003). FF was associated with VAST score ≥2 (OR 3.9, 95% CI 1.6-9.6, p=0.002) but not with thrombocytopenia alone (OR 1.3, 95% CI 0.5-3.2, p=0.58). Patients with FF were older (28±13 vs 23±8 years, p=0.02) and were more likely to have high Fontan pressure (67% vs 30%, p=0.002) and moderate or severe ventricular dysfunction (VD) on imaging (28% vs 7%, p=0.004).

On multivariate analysis, RP was independently associated with FF when adjusted for VD and age (OR 4.1, 95% CI 1.5-11.6, p=0.007), but not when adjusted for high Fontan pressure (OR 1.4, 95% CI 0.4-5.4 p=0.602). Similarly, VAST score ≥2 remained significant after adjusting for VD and age (OR 3.6, 95% CI 1.4-9.6, p=0.01) but not for high Fontan pressure (OR 1.7, 95% CI 0.5-5.8, p=0.36).
Conclusions: Radiologic evidence of portal hypertension may herald higher Fontan pressures and increased risk of FF. Thrombocytopenia did not independently predict adverse outcomes in our study.