Tomoya Fukuoka, Hiroyo Ishii, Emi Ito, Yuki Matsuoka, Yoshikazu Miura, Takeshi Kimura, Kazuhiko Bessho, Takehisa Ueno, Yasuji Kitabatake
PV-PSV proved useful for predicting HRV, suggesting that it may serve as a noninvasive test in children with BA.
OBJECTIVES: Portal vein (PV) flow velocity is considered an indicator of portal hypertension. However, few studies have evaluated the predictive performance of PV flow velocity for high-risk varices (HRV) in children with biliary atresia (BA). In this study, we evaluated the predictive performance of PV peak systolic velocity (PSV) for HRV in children with BA.
METHODS: We retrospectively reviewed consecutive children with BA, aged 18 years or younger, who underwent initial upper endoscopy and abdominal ultrasonography within 30 days of endoscopy at Osaka University Hospital between 2010 and 2024. HRV was defined as Grade 3 esophagogastric varices (EGV), or Grade 2 EGV with red color signs and/or gastric varices along the cardia. Clinical data, including PV-PSV and previously reported prediction indices for varices, were compared between children with and without HRV.
RESULTS: Seventy children were included, 23 of whom had HRV. Variables that differed significantly between children with and without HRV were albumin, total bilirubin, clinical prediction rule, King's variceal prediction score, risk score, and PV-PSV. On multivariate analysis, PV-PSV was the only independent predictor of HRV. PV-PSV yielded the highest area under the receiver operating characteristic curve for predicting HRV (0.871), with 82.6% sensitivity and 78.0% specificity at a cutoff of 16.7 cm/s.
CONCLUSIONS: PV-PSV proved useful for predicting HRV, suggesting that it may serve as a noninvasive test in children with BA.