A scoring system for biliary atresia: is this the right one?

A prospective multicentre study evaluating the performance of the modified simple biliary atresia scoring system in predicting biliary atresia

PURPOSE

Early diagnosis of biliary atresia (BA) is critical for best outcomes, but is challenged by overlapping clinical manifestations with other causes of obstructive jaundice in neonates. We evaluate the performance of the modified Simple BA Scoring System (SBASS) in diagnosing BA.

METHODS

We performed a prospective, cross-sectional study on infants with cholestatic jaundice (June 2021-December 2022). Modified SBASS scoring was applied and compared to the eventual diagnosis (as per intraoperative cholangiogram (IOC) and liver histopathology). The score (0-6), consists of gall bladder length < 1.6 cm (+ 1), presence of triangular cord sign (+ 1), conjugated bilirubin:total bilirubin ratio > 0.7(+ 2), gamma-glutamyl transferase (GGT) ≥ 200 U/L (+ 2).

RESULTS

73 were included: Fifty-two (71%) had BA. In the non-BA group, 6 (28%) had percutaneous cholangiography (PTC) while 15 (72%) had intraoperative cholangiogram (IOC). At a cut-off of 3, the modified SBASS showed sensitivity of 96.2%, specificity of 61.9% and overall accuracy of 86.3% in diagnosing BA. Area under receiver operating characteristic curve was 0.901. GGT had the highest sensitivity (94.2%), while triangular cord sign showed the highest specificity at 95.2%.

CONCLUSION

The SBASS provides a bedside, non-invasive scoring system for exclusion of BA in infantile cholestatic jaundice and reduces the likelihood of negative surgical explorations.

Development and Validation of a Machine Learning-Based Prediction Model for Detection of Biliary Atresia

Background and Aims

Biliary atresia is a rare and devastating bile duct disease that occurs during the neonatal period. Timely identification and prompt surgical intervention is critical for improving the outcome. The aim of the study was to develop a new machine learning-based prediction model for the detection of biliary atresia.

Methods

Neonates aged <100 days with cholestasis at least once were retrospectively screened in 2 tertiary referral hospitals between 2015 and 2020. Simple demographic data, routine laboratory indices, and imaging findings of ultrasonography and hepatobiliary scintigraphy were used as features in the multivariate analysis. The extreme gradient boosting (XGBoost) framework was used to develop prediction models according to the diagnostic steps.

Results

Among 1605 enrolled neonates with all-cause cholestasis, 145 (9%) were included as having biliary atresia. Direct bilirubin, gamma-glutamyl transpeptidase, abdominal sonography, and hepatobiliary scan were the most impactful features in prediction models. The Step II XGBoost model, consisting of nonimaging inputs, showed excellent discriminatory performance (area under the curve = 0.97). The Step III and IV XGBoost models showed near-perfect performances (area under the curve = 0.998 and 0.999, respectively). In external validation (n = 912 with 118 [12.9%] biliary atresia), XGBoost-based prediction models consistently showed acceptable performances. Utilizing shapley additive explanation values also provided visualized insight and explanation of the contribution of features in detecting biliary atresia. The models were integrated into a web-based diagnostic tool for case-level application.

Conclusion

We introduced a new machine learning-based prediction model for detecting biliary atresia in the largest cohorts of neonatal cholestasis.

Development and validation of a novel nomogram and risk score for biliary atresia in patients with cholestasis

BACKGROUND

Timely discriminating biliary atresia (BA) from other causes of cholestasis is important but challenging.

AIMS

To develop a useful diagnostic nomogram and a simplified scoring system to diagnosing BA.

STUDY DESIGN

All medical records of the patients who were consecutively admitted to our institution with cholestasis from March 2016 to December 2020 were retrospectively searched. The patients were allocated to the derivation cohort (n = 343) and the validation cohort (n = 246). Multivariable logistic regression models were used to construct the nomogram. The nomogram was validated in both cohorts. The simplified risk score was derived from the nomogram.

RESULTS

The nomogram was constructed based on presence of clay stool, gallbladder length, gallbladder emptying index, shear wave elastography value, and gamma-glutamyl transferase level. This model showed good calibration and discrimination ability, with the C-index of 0.968 (95% CI: 0.951-0.984). The discriminating ability is most prominent in the 61-90 days group, with AUC of 0.982 (95% CI: 0.955-1.000). The simplified risk score identified most patients with very high or low risk of BA, and was capable of exempting 64.3% non-BA patients from intraoperative cholangiogram procedure.

CONCLUSIONS

This novel diagnostic nomogram had good discrimination and calibration abilities. The simplified scoring system showed significant clinical utility.

Knowledge, Attitudes, and Behaviors of Pediatric Primary Care Providers on Management of Cholestasis

Biliary atresia is a common cause of cholestasis in infants and is a time-sensitive diagnosis. A survey was distributed to pediatric primary care providers in order to assess variations in diagnosis and management of cholestasis. Participants were identified from physician parent groups on social media and regional pediatric residency programs. Information on knowledge and interpretation of screening tests, past experience/behavior, confidence, and comfort level managing cholestasis, as well as demographic information was collected. Out of 116 eligible respondents, 94.8% were confident in diagnosing hyperbilirubinemia but only 10.3% knew the biochemical definition of direct hyperbilirubinemia. Of the 56% of providers who had some knowledge of the guidelines, 18.5% stated the guidelines changed the way they evaluate cholestasis. These results demonstrate a gap in knowledge of diagnosing and evaluating cholestasis, which could provide the framework for standardized screening, leading to earlier identification of biliary atresia.

Risk Estimation for Biliary Atresia in Patients with Neonatal Cholestasis: Development and Validation of a Risk Score

Purpose To develop and validate a scoring system based on clinical and imaging features to predict the risk for biliary atresia in patients with neonatal cholestasis. Materials and Methods Patients with neonatal cholestasis who underwent both ultrasonography (US) and hepatobiliary scintigraphy (n = 480) were retrospectively identified from two tertiary referral hospitals from January 2000 to February 2017. Patients from one hospital were classified as the derivation cohort (n = 371), and those from the other hospital were classified as the validation cohort (n = 109). Clinical and imaging features associated with biliary atresia were assessed. Histopathologic or intraoperative cholangiographic findings served as the reference standard for biliary atresia. A prediction model was developed by using logistic regression and was then transformed into a scoring system. The scoring system was internally and externally validated. Results Among the 371 patients in the derivation cohort, 97 (26.15%) had biliary atresia. A scoring system was constructed with the following variables: full-term birth, presence of the triangular cord sign at US, abnormal gallbladder morphology at US, and failure of radioisotope excretion to the small bowel at hepatobiliary scintigraphy. The maximum possible score with this system is 7 points. This system enabled differentiation of biliary atresia in the derivation cohort (C statistic, 0.981; 95% confidence interval [CI]: 0.970, 0.992) and the validation cohort (C statistic, 0.995; 95% CI: 0.987, 1.000). The risk score also showed good calibration in both the derivation and the validation cohorts (P = .328 and P = .281, respectively). Conclusion A simple scoring system combining clinical and imaging features can help accurately estimate the risk for biliary atresia in patients with neonatal cholestasis.

Incremental Value of MR Cholangiopancreatography in Diagnosis of Biliary Atresia

Purpose

To evaluate the incremental value of a combination of magnetic resonance cholangiopancreatography (MRCP) and ultrasonography (US), compared to US alone, for diagnosing biliary atresia (BA) in neonates and young infants with cholestasis.

Materials and Methods

The institutional review board approved this retrospective study. The US and MRCP studies were both performed on 64 neonates and young infants with BA (n = 41) or without BA (non-BA) (n = 23). Two observers reviewed independently the US alone set and the combined US and MRCP set, and graded them using a five-point scale. Diagnostic performance was compared using pairwise comparison of the receiver operating characteristics (ROC) curve. The sensitivity, specificity, accuracy, positive predictive value (PPV), and negative predictive value were assessed.

Results

The diagnostic performance (the area under the ROC curve [A_z]) for diagnosing BA improved significantly after additional review of MRCP images; A_z improved from 0.688 to 0.901 (P = .015) for observer 1 and from 0.676 to 0.901 (P = .011) for observer 2. The accuracy of MRCP combined with US (observer 1, 95% [61/64]; observer 2 92% [59/64]) and PPV (observer 1, 95% [40/42]; observer 2 91% [40/44]) were significantly higher than those of US alone for both observers (accuracy: observer 1, 73% [47/64], P = 0.003; observer 2, 72% [46/64], P = 0.004; PPV: observer 1, 76% [35/46], P = 0.016; observer 2, 76% [34/45], P = 0.013). Interobserver agreement of confidence levels was good for US alone (ĸ = 0.658, P < .001) and was excellent for the combined set of US and MRCP (ĸ = 0.929, P < .001).

Conclusion

Better diagnostic performance was achieved with the combination of US and MRCP than with US alone for the evaluation of BA in neonates and young infants with cholestasis.