Objectives: Liver fibrosis has been recognized as a potentially serious morbidity associated with Fontan’s circulation (the Fontan associated liver disease, FALD). The purpose of this study was to assess the prevalence of FALD with standard investigations and evaluate potential risk factors for more advanced changes.
Methods: Patients were screened for FALD by abdominal ultrasound and routine laboratory tests (interpretation based on age and gender adjusted reference values). A “liver disease score” including items from ultrasound and blood testing (5 each, maximum score 10 points) was calculated to compare the degree of liver involvement and evaluate risk factors for more advanced FALD.
Result: A total of 230 Fontan patients (male n = 136, female n = 94) seen for routine follow-up between May 2014 and June 2019 with abdominal ultrasound and routine blood testing were included. The median duration of follow-up since the Fontan surgery was 10 (IQR: 7–15) years. Sonographic signs of fibrotic changes included surface nodularity, heterogeneous parenchyma or hyperechoic lesions in 117 (50.9%) patients. Ascites (n = 24, 10.4%) and splenomegaly (n = 59, 25.7%) were suggestive of portal hypertension. Abnormal blood test results included thrombocytopenia (n = 47, 20.4%), elevated gamma-glutamyl-transpeptidase levels (n = 195, 84.8%), glutamic-oxaloacetic-transaminase levels above normal range (n = 23, 10.0%), reduced prothrombin activity excluding cases with oral anticoagulation (n = 33, 17.3%), and hypoalbuminemia (n = 10, 4.3%). The median liver disease score was 3 (IQR: 2–4). Scores of 5 or higher were found in 28 (12.2%) patients. Duration of follow-up after the Fontan surgery was longer in cases with elevated scores (14 [IQR: 10–20] years vs. 9 [IQR: 6–14] years, p = 0.001) and central venous pressure (CVP) was higher (13 [IQR: 11–15] vs. 11 [IQR: 9–12] mm Hg, p < 0.001) as well as end-diastolic pressure of the systemic ventricle (8 [IQR: 5–10] vs. 6 [IQR: 5–7] mm Hg, p = 0.001). A patent fenestration was not related to the severity of FALD (p = 0.421). Only longer follow-up duration (OR = 1.12 [1.05–1.20], p = 0.001) and higher CVP (OR = 1.53 [1.25–1.86], p < 0.001) were significant predictors of more advanced FALD in a multivariable binary logistic regression model.
Conclusion: Abdominal ultrasound and laboratory abnormalities suggestive of FALD are common during routine follow-up. More advanced findings are associated with longer follow-up and higher CVP. Strategies to prevent or attenuate FALD are warranted.
