A Nomogram Predicting the Prognosis of Children With Biliary Atresia After Hepatoportoenterostomy

Abstract

Background: Although Kasai portoenterostomy (KPE) is performed timely for most children with biliary atresia (BA), the native liver survival (NLS) is still poor due to the progressive liver fibrosis. Many children have to receive liver transplantation (LT) within 2 years after KPE. Early prediction of the prognosis permits the implementation of prophylactic treatments for BA children. However, studies about the prediction are limited. Objective: The purpose of this study is to establish a nomogram to predict the prognosis of BA children within 2 years after KPE. Methods: The follow-up data of 151 BA children were retrospectively reviewed, and were randomly divided into a training cohort for constructing a nomogram (n = 103) and a validation cohort (n = 48). In the training cohort, patients were divided into Group A and Group B according to whether death or LT were observed within 2 years post-KPE. Multivariate Cox regression based on the baseline characteristics, liver function indicators and LSM (liver stiffness measurement) values at KPE and 3 months after KPE was utilized for the establishment of the nomogram in predicting the prognosis of BA within 2 years after KPE. The discrimination and calibration of the nomogram were internally and externally validated. Results: Fifty-six BA children were included in Group A and 47 were included in group B. Age at KPE, METAVIR score F4, LSM at 3 months, first onset of cholangitis within 3 months, and jaundice clearance time were the independent predictors for the prognosis of BA children within 2 years after KPE (all P < 0.05). The developed nomogram based on these independent predictors showed good discrimination and calibration by the internal and external validation. Its performance was better than each predictor in predicting the prognosis (all P < 0.05). Conclusions: The established nomogram based on the indicators from the first 3 months after KPE may be useful for predicting the prognosis of BA children within 2 years post-KPE and helpful for the consideration of LT.

Keywords: LSM; biliary atresia; kasai portoenterostomy; liver transplantation; nomogram; prognosis.

Figure 1

SWE measurements in BA children. The SWE showed a stiffness color map (top) and a grayscale image (bottom). The LSM value in the ROI of (A) was 10.2 kPa (METAVIR score F2), and (B) was 28.2 kPa (METAVIR score F4). SWE, shear wave elatography; LSM, liver stiffness measurement; ROI, region of interest.

Figure 2

Kaplan-Meier analysis for the prognosis of BA children within 2 years after KPE. The NLS probabilities in the training cohort and validation cohort were 45.6 and 45.8%, respectively. BA, biliary atresia; KPE, Kasai portoenterostomy; NLS, native liver survival.

Figure 3

Trends of LSM, ALT, and TBil within 3 months of follow-up post-KPE. The general trends of LSM (A) and ALT (B) were rising first and then decreasing. TBil (C) showed a significant decreasing trend after KPE. However, the 75th percentiles of TBil at 2 and 3 months after KPE increased. LSM, liver stiffness measurement; ALT, aspartate aminotransferase; TBil, total billrubin.

Figure 4

Forest plot of Cox regression analysis for predictors of the prognosis of BA children within 2 year post-KPE. Age at KPE, METAVIR score F4, LSM at 3 months, first onset of cholangitis within 3 months, and jaundice clearance time were the independent predictors. BA, biliary atresia; KPE, Kasai portoenterostomy; LSM, liver stiffness measurement; AST, aspartate aminotransferase; ALT, alanine amino-transferase; TBil, total billrubin; DBil, direct billrubin.

Figure 5

Nomogram for predicting the probability of NLS in BA children within 2 years post-KPE. The top row showed the point assignment for each variable. Rows 2–6 indicated the variables included in the nomogram. The bottom row showed the probability of NLS. BA, biliary atresia; KPE, Kasai portoenterostomy; NLS, native liver survival; LSM, liver stiffness measurement.

Figure 6

Internal and external validation of the nomogram predicting the prognosis of BA within 2 years post-KPE. The ROC curves were plotted for the discrimination of the nomogram. The AUCs in the training cohort and validation cohort were 0.930 (A) and 0.878 (B), respectively, indicting good discrimination. The calibration curves were plotted for evaluating the calibration of nomogram-predicted prognosis and actual prognosis estimated using KM analysis. The calibration plots in the training and validation datasets were shown in (C,D). Both indicated that the nomogram-predicted progonisis compared very well with the actual prognosis. KPE, Kasai portoenterostomy; ROC, receiver operating characteristic curve; AUC, area under the curve; KM, Kaplan-Meier.

Figure 7

ROC curves of age of KPE, jaundice clearance time, LSM at 3 months, METAVIR score, and first onset of cholangitis within 3 months in predicting the prognosis of BA children within 2 years post-KPE. The AUCs of them [age of KPE: 0.680 (95% CI: 0.581–0.769), jaundice clearance time: 0.800 (95% CI: 0.710–0.872), LSM at 3 months: 0.837 (95% CI: 0.752–0.903), METAVIR score: 0.652 (95% CI: 0.551–0.743), and first onset of cholangitis within 3 months: 0.642, (95% CI: 0.541–0.734)] were lower than the nomogram (all P < 0.05). ROC, receiver operating characteristic curve; AUC, area under the curve; KPE, Kasai portoenterostomy; LSM, liver stiffness measurement; CI, confidence interval.