Highlight
1. Serum bile acid (sBA) levels measured within one year after successful Kasai procedure (KP) strongly predict the development of portal hypertension (PH) over the following five years.
2. sBA thresholds at 6 and 11 months post-KP provide 100% sensitivity in predicting high-risk esophageal varices (HRV).
3. sBA measurement offers a non-invasive biomarker to monitor residual cholestasis and stratify risk for complications after KP in biliary atresia patients.
4. Early identification of at-risk patients may allow more timely clinical interventions to prevent severe portal hypertension outcomes.
Study Background
Biliary atresia is a progressive fibro-obliterative cholangiopathy of infancy that disrupts bile flow, leading to cholestasis, liver fibrosis, cirrhosis, and eventual liver failure. The Kasai portoenterostomy procedure, typically performed in early infancy, aims to restore bile drainage to delay or prevent liver transplantation. However, even among children with successful KP—defined by normalized serum bilirubin levels—many still develop significant complications such as portal hypertension (PH) and high-risk esophageal varices (HRV), which pose risks of bleeding and morbidity.
Detecting PH and HRV early is clinically important to optimize surveillance and management strategies. Current approaches rely on invasive procedures like endoscopy or indirect assessments that may be limited in sensitivity or specificity. Serum bile acids (sBA) reflect hepatobiliary function and residual cholestasis, making them a promising non-invasive biomarker. This study evaluated the predictive value of early sBA levels in children with successful KP for the subsequent development of PH and HRV within five years post-KP.
Study Design
This retrospective monocentric observational study included 60 children diagnosed with biliary atresia who underwent successful KP, defined as achieving a total serum bilirubin ≤25 μmol/L within six months after KP. sBA levels were measured during the first postoperative year, specifically at median timepoints of 6 months (range 4.5–9 months) and 11 months (range 9–12 months) after KP.
The investigators compiled longitudinal clinical, biological, and digestive endoscopic data up to 5 years after KP to determine the incidence of portal hypertension and development of high-risk esophageal varices. The predictive accuracy of sBA measurements for these endpoints was analyzed using receiver operating characteristic (ROC) curves to calculate area under the curve (AUC), sensitivities, and threshold values.
Key Findings
sBA levels measured at a median of 6 months and again at 11 months post-KP robustly predicted portal hypertension development at 3 and 5 years, with AUCs ranging between 0.89 and 0.93 (p<0.0003), indicating excellent discriminatory power. Notably, sBA concentrations also predicted the occurrence of high-risk esophageal varices within 5 years, with AUCs between 0.74 and 0.75 (p<0.04).
Specific sBA thresholds were identified: a cutoff of 56 μmol/L at 6 months and 30 μmol/L at 11 months post-KP predicted HRV development with a sensitivity of 100%. This implies that all patients who developed high-risk varices had sBA levels above these thresholds at these time points, yielding high negative predictive value.
The study reinforces the role of persistent residual cholestasis, as reflected by elevated serum bile acids despite successful bilirubin normalization, as a key driver of portal hypertension pathogenesis and variceal risk in biliary atresia survivors.
Expert Commentary
This study addresses an important clinical gap in monitoring children after KP. While total serum bilirubin serves as an initial marker of KP success, the persistence of subtle cholestasis captured by sBA levels predicts subsequent complications more precisely. Measurement of serum bile acids is feasible, minimally invasive, and cost-effective, making it an attractive surveillance biomarker.
Limitations include the retrospective monocentric nature potentially limiting generalizability. Additionally, while sBA predicts development of PH and HRV, this study does not directly assess impact on clinical outcomes such as variceal bleeding or survival. Further prospective multicenter validation and integration with other noninvasive markers like liver stiffness measurement could refine risk stratification.
Mechanistically, elevated bile acids may worsen hepatic inflammation and fibrosis, perpetuating portal hypertension progression. Early identification of patients at risk through sBA could inform clinical decisions such as timing of endoscopic surveillance, prophylactic therapies, or early referral for transplantation evaluation.
Conclusion
Serum bile acid levels measured within the first postoperative year in children with biliary atresia who undergo successful Kasai portoenterostomy reliably predict the development of portal hypertension and high-risk esophageal varices within five years. These findings advocate for incorporating sBA measurement into routine post-KP surveillance to facilitate early intervention and improve long-term outcomes.
Further multicenter prospective studies are warranted to confirm these thresholds, evaluate combined biomarker models, and determine associations with clinical events to optimize personalized management pathways for this vulnerable population.
Funding and ClinicalTrials.gov
The original study did not report specific funding sources or clinical trial registration numbers. Future studies could benefit from multicenter collaboration and prospective design.
References
- Grimaud E, Gardin A, Ackermann O, et al. Serum bile acid levels predict the development of portal hypertension and high-risk esophageal varices following successful Kasai in biliary atresia. Hepatology. 2025;84(1):117-128. doi:10.1002/hep.31557
- Moyer KA, Mack C, Sokol RJ. Biliary atresia: Clinical outcomes and management strategies. Semin Liver Dis. 2018;38(3):255-264. doi:10.1055/s-0038-1666840
- Fagiuoli S, Alaggio R, Maggiore G. Non-invasive assessment of portal hypertension: State of the art. Clin Liver Dis (Hoboken). 2022;19(1):18-25. doi:10.1002/cld.1131
