Abstract
Objectives
To investigate the prognostic significance of aspartate aminotransferase to platelet ratio index (APRI) in relation to histopathological features across the clinical course of biliary atresia (BA).
Methods
In this observational cohort study, we enrolled 135 BA patients with available APRI values at Kasai portoenterostomy (KPE, n = 116) or at post‐KPE follow‐up (n = 70; serum samples, n = 139; liver biopsies, n = 139). APRI was matched to manual scorings of liver histology, ductular reaction (DR) analysed using neural network model and fibrosis quantified from Sirius Red‐stained sections.
Results
APRI was elevated at both KPE and post‐KPE follow‐up (0.92 vs. 1.2, p = 0.5), and associated with biochemical markers of cholestasis, and decreased liver function. At KPE, APRI was higher in patients failing to resolve cholestasis postoperatively (0.80 vs. 0.96, p = 0.02) and both at KPE (0.77 vs. 0.94, p = 0.06) and post‐KPE (0.7 vs. 2.1, p < 0.001) in those with subsequent need for liver transplant (LT). Across the disease course, APRI moderately predicted LT need (at KPE, HR = 1.4, p = 0.01; post‐KPE, HR = 1.2, p < 0.001). APRI correlated with DR (at KPE, R = 0.37, p < 0.001; post‐KPE, R = 0.35, p < 0.001), composed of biliary epithelium (R = 0.34, p < 0.001; R = 0.25, p < 0.01) and parenchymal intermediate hepatocytes (R = 0.33, p < 0.001; R = 0.43, p < 0.001) during the entire clinical course. APRI showed no correlation with quantified liver fibrosis at KPE (R = 0.16, p = 0.13), but correlated with quantified fibrosis (R = 0.22, p = 0.02) and Metavir staging (R = 0.46, p < 0.001) after KPE. No association was observed with portal inflammatory cell infiltration or histological cholestasis.
Conclusions
APRI reflects biliary and portal injury rather than generalised liver fibrosis and associates with poor prognosis in BA.
Keywords: cholestasis, ductular reaction, fibrosis, inflammation, prognosis
What is Known
Biliary atresia is the leading cause for liver transplantation in children due to rapidly progressing cholestasis and fibrosis
Aspartate aminotransferase to platelet ratio index (APRI) is a widely used non‐invasive biomarker for liver fibrosis in various liver diseases
Elevated APRI reflects severe disease stage, risk of complications, and poor prognosis
What is New
-
•
APRI reflects the degree of histological ductular reaction in biliary atresia at and after Kasai portoenterostomy (KPE)
-
•
APRI was not associated with liver fibrosis at KPE, or with portal inflammation at any point of the disease course
-
•
APRI associated with poor short‐ and long‐term prognosis in biliary atresia
1. INTRODUCTION
Biliary atresia (BA) is a progressive cholestatic disease of infancy and the leading cause of liver transplantation (LT) in children. 1 Fibroinflammatory obliteration of bile ducts leads to histological and biochemical cholestasis, ductular reaction (DR), portal inflammation, and biliary cirrhosis with a rapidly progressing course. 1 The injury results in liver failure within 2 years of age, if cholestasis is not operatively resolved. 2 The first‐line surgical treatment, Kasai portoenterostomy (KPE), aims to restore bile flow by excising the extrahepatic bile duct remnants and reconstruction with a Roux‐en‐Y jejunal loop. 1 , 2 Subsequently, 60%–75% of patients require LT by adulthood due to progressive liver injury. 1
In various adult and paediatric liver diseases, aspartate aminotransferase (AST) to platelet ratio index (APRI) is clinically used as a non‐invasive marker of liver fibrosis and a predictor of chronic liver disease complications. 3 , 4 , 5 , 6 Although a recent meta‐analysis pooling small BA cohorts linked APRI with significant portal fibrosis at KPE, APRI's link to liver fibrosis and other histological markers of BA liver injury remains unclear. 6 , 7 Still, it has proven clinical value in predicting biochemical response to KPE, development of portal hypertension, and long‐term prognosis in BA. 5 , 8
The purpose of this study was to examine APRI's relationship with histopathological findings of liver injury along with its clinical utility within our large cohort from two tertiary‐centres. Thus, we calculated APRI both at KPE and post‐KPE follow‐up in relation to outcome and associated it with liver histology, assessed both with conventional pathology and a novel quantitative deep learning model.
2. METHODS
2.1. Ethics statement
The study protocol was reviewed and approved by the National Health Service Research Ethics Committee in the UK (12/WA/0282 and 18/SC/0058) and Research Ethics Committee (HUS/7296/2024) and the review board of the Helsinki University Hospital (HUS/564/2024). The study design conformed to the Declaration of Helsinki. Informed consent for the use of samples in the research purposes was obtained from the patients' legal guardians.
2.2. Patients and specimens
Children with BA who underwent KPE in Finland during 2000–2020 or at King's College Hospital, London, UK, during 2005–2013 were enrolled (Table 1). Study population covered all patients (n = 135) with available APRI measurement determined at KPE (patients, n = 116; serum samples, n = 116) or at post‐KPE follow‐up (patients, n = 70; serum samples, n = 139). All post‐KPE follow‐up patients had APRI determined also at KPE to assess longitudinal changes (Table 1).
Table 1.
Clinical characteristics and biochemistry.
| All patients | At follow‐up | |
|---|---|---|
| Patients, n | 135 | 70 |
| Liver biopsies, a n | 259 | 139 |
| Place of care, n (%) | ||
| Helsinki, Finland | 64 (47%) | 62 (89%) |
| London, UK | 63 (47%) | 0 |
| Other | 8 (6%) | 8 (11%) |
| Male, n (%) | 60 (44%) | 32 (46%) |
| Type of BA, b n (%) | ||
| I | 1 (1%) | 1 (2%) |
| II | 6 (4%) | 6 (9%) |
| III | 127 (95%) | 62 (89%) |
| Cystic disease, c n (%) | 15 (11%) | 8 (11%) |
| Splenic malformation, n (%) | 17 (13%) | 9 (13%) |
| Isolated disease, n (%) | 104 (77%) | 49 (70%) |
| Age at KPE, d | 62 (41–81) | 64 (40–84) |
| Clearance of jaundice, d n (%) | 81 (60%) | 50 (71%) |
| Outcome, n (%) | ||
| NL | 53 (40%) | 35 (50%) |
| LT | 72 (53%) | 30 (42%) |
| Death with NL | 6 (4%) | 1 (1%) |
| Death after LT | 4 (3%) | 4 (6%) |
| NL survival, y | 7.07 (1.59–12.5) | 9.8 (4.72–14.6) |
| 2‐year, n (%, 95% CI) | 83 (61%, 53%–70%) | 49 (70%, 58%–80%) |
| 5‐year, n (%, 95% CI) | 69 (51%, 42%–60%) | 46 (66%, 53%–77%) |
| Age at LT, e y | 1.96 (0.91–7.47) | 6.53 (1.29–9.82) |
| Liver biochemistry at biopsy | ||
| Bilirubin, µmol/L | 99 (12–161) | 13 (7–45) |
| Conjugated bilirubin, µmol/L | 57 (5–120) | 7 (3–23) |
| AST, U/L | 127 (68–209) | 86 (51–140) |
| ALT, U/L | 81 (40–129) | 56 (30–106) |
| GGT, U/L | 175 (58–491) | 62 (28–156) |
| ALP, U/L | 440 (288–637) | 377 (256–559) |
| Prealbumin, mg/L | 116 (88–162) | 140 (104–169) |
| Bile acids, µmol/L | 99 (28–195) | 49 (20.8–168) |
| APRI | 1.07 (0.56–2.08) | 1.37 (0.58–2.22) |
| Liver histopathology | ||
| DR% | 6 (3–13) | 5 (2–16) |
| BE% | 4 (2–7) | 3 (2–5) |
| PIH% | 1 (0–5) | 1 (0–145 |
| Proportional Sirius Red fibrosis, f % | 15 (9–26) | 21 (14–32) |
| Metavir stage, g 0–4 | 3 (2–4) | 3 (2–4) |
| Fibroscan, kPA h | 18.7 (10.7–31.2) | 16.2 (10.5–30.8) |
Note: Data are presented as median with interquartile range or frequencies.
Abbreviations: ALP, alkaline phosphatase; ALT, alanine transaminase; APRI, aspartate transaminase to platelet ratio index; AST, aspartate aminotransferase; BA, biliary atresia; BE, biliary epithelium; DR, ductular reaction; GGT, gamma‐glutamyl transferase; kPa, kilopascals; KPE, Kasai portoenterostomy; LT, liver transplantation; NL, native liver; PIH, parenchymal intermediate hepatocytes.
Fifty‐five patients had biopsy specimens at both KPE and follow‐up, including 34 with repeated follow‐up biopsy specimens comprising 121 punch biopsies, 107 wedge biopsies, and the remainder specimens of unknown type.
Information available for 134 patients.
Information available for 133 patients.
Decrease of serum bilirubin concentration < 20 μmol/L after KPE.
Including 27 patients those who underwent transplantation shortly after Kasai (< 12 months).
Proportional Sirius Red fibrosis: information available for 122 patients.
Metavir fibrosis: information available for 132 patients.
Fibroscan information available for 39 patients combining collectively 63 samples.
Among included patients, all KPE patients (biopsies, n = 116) and follow‐up patients (patients, n = 70; biopsies, n = 139) had digitised image of liver biopsy obtained at same time point as APRI measurement. KPE liver biopsies were obtained in both study centres and final BA diagnosis was confirmed with histopathological assessment of liver parenchyma and bile duct remnants. Post‐KPE biopsies comprised protocol liver biopsies (n = 107) and biopsies obtained at LT (n = 32) of only patients operated in Finland. Following national BA programme, core needle biopsies are obtained at KPE, 1 year after KPE and in every 5 years thereafter. 9 Follow‐up biopsies are performed under general anaesthesia for upper gastrointestinal endoscopy surveillance for varices in Finland. 10 , 11 Routine biopsies at LT from explanted livers were also included in the study (Table 1).
Baseline clinical data was assessed, and routine serum liver biochemistry and liver stiffness measurements (LSM) were retrospectively collected (Table 1). Clearance of jaundice (COJ) was defined as a decrease in serum total bilirubin to <20 μmol/L following KPE.
2.2.1. Liver staining and immunohistochemistry
Cytokeratin‐7 (CK7) immunohistochemistry and haematoxylin‐eosin staining was performed fully automated in Helsinki University Hospital routine laboratory. Collagen fibres were stained with Sirius Red colour staining. Laboratory methods are provided in Supplementary Methods.
2.2.2. Liver histopathology
Liver histopathology was evaluated with both conventional scorings and quantitative neural network model for digitised images of histological slides. The CharBADR model, a neural network for cytokeratin‐7 (CK7) stained biopsy specimens of BA livers, quantified DR, including the biliary epithelium (BE) and parenchymal intermediate hepatocytes (PIH). 12 DR and BE were expressed as a proportional area from the entire liver tissue (DR% and BE%, respectively), and PIH from parenchymal area (PIH%).
Liver fibrosis was scored using Metavir staging and quantified with CaseViewer HistoQuant software (3DHISTECH, Budapest, Hungary). In the quantification, colour intensity and shade of liver tissue image was measured and Sirius Red positive area was proportioned to the entire liver tissue area avoiding liver capsule. 13
Additionally, BE, PIH, portal inflammatory cell infiltration, as well as intraductal, intracellular, and intracanalicular cholestasis were scored manually by two pathologists. Further information about scorings is depicted in Supplementary Methods.
2.3. Statistical analysis
RStudio (R version 4.3.1) was used for the data analysis. Continuous variables are expressed as medians and interquartile ranges (IQRs) and compared using Mann‐Whitney U, Kruskal–Wallis, Spearman rank correlations, and temporal changes of APRI assessed with Wilcoxon signed rank tests. Survival analyses were performed using univariate and multivariable Cox proportional hazards regression, with LT or death during follow‐up as the outcome event. For longitudinal analyses, a time‐dependent Cox proportional hazards model was fitted in the follow‐up cohort, modelling APRI as a time‐varying covariate after Box‐Cox transformation. Cox proportional hazard ratios (HR) with 95% confidence intervals (CI) were calculated using either APRI as a continuous variable or dichotomous APRI cutoff and confirmed graphically using Kaplan–Meier survival curves. Optimal cut‐offs of APRI were chosen using receiver operating characteristic (ROC) analysis based on the maximum sum of specificity and sensitivity for the event and internally confirmed with bootstrapping method. Statistical significance was set at p < 0.05. Patients who were lost to follow‐up or had missing data for a given analysis were excluded from the respective analysis.
3. RESULTS
3.1. Patient characteristics
Baseline patient characteristics at the time of liver biopsy are detailed in Table 1. Median age at KPE was 62 (41–81) days in the entire study cohort. Following KPE, 81 (60%) achieved COJ. One hundred twenty‐seven (95%) patients had type III BA, and 17 (13%) had Biliary Atresia Splenic Malformation syndrome (BASM). Native‐liver survival (NLS) rates among all patients at 2 and 5 years were 61% and 51%, respectively. Follow‐up biopsies were collected at a median age of 4.1 years (range, 0.4–20.6 years).
3.2. APRI in BA
When comparing KPE and the first follow‐up biopsy of individual patients, APRI was similarly elevated at KPE and follow‐up (0.92 (0.54–1.54) vs. 1.17 (0.55–2.08), p = 0.50). BA patients undergoing KPE over 60 days of age presented with notably higher APRI (0.58 vs. 1.14, p < 0.001) at KPE compared to infants operated earlier. In addition, KPE age positively correlated with APRI (Figure 1, R = 0.45, p < 0.001). No significant difference in APRI was detected in patients with isolated BA compared to patients having associated anomalies (0.90 vs. 0.69, p = 0.22), or between different BA types (1 vs. 2 vs. 3, p = 0.48). BASM had no statistically significant effect on APRI (0.63 vs. 0.90, p = 0.08). Age at sampling (R = −0.04, p = 0.61) was not associated with APRI during post‐KPE follow‐up.
Figure 1.
Association of APRI at KPE with clinical outcomes. Violin plots of APRI and (A) Age at KPE, (B) COJ and (C) outcome at 5 years. Violin plots display individual data points. Linear regression plots (D) stratified by LT status, showing APRI trajectories from KPE through follow‐up including a longitudinal analysis of the correlation between APRI and Metavir. Illustration of Spearman rank correlation (E) between APRI and liver stiffness during post‐KPE follow‐up. APRI, aspartate aminotransferase to platelet ratio index; COJ, clearance of jaundice; kPa, kilopascals; KPE, Kasai portoenterostomy; LSM, liver stiffness measurement; LT, liver transplant; NL, native liver; NLS, native liver survival.
3.3. Association with other non‐invasive markers of liver injury
APRI associated with biochemical cholestasis, hepatocyte injury, and decreased hepatic synthetic function both at KPE and during postoperative follow‐up. APRI measured at KPE, positively correlated with total (R = 0.48, p < 0.001) and conjugated bilirubin (R = 0.56, p < 0.001), AST (R = 0.86, p < 0.001) ALT (R = 0.69, p < 0.001), ALP (R = 0.39, p < 0.01), serum total bile acids (R = 0.48, p < 0.01) and negatively with thromboplastin time (R = −0.34, p < 0.05), and factor V (R = −0.60, p < 0.001).
In line with the associations observed at KPE, serum APRI measured at follow‐up positively correlated with total (R = 0.59, p < 0.001) and conjugated bilirubin (R = 0.62, p < 0.001), AST (R = 0.74, p < 0.001), ALT (R = 0.61, p < 0.001), GGT (R = 0.37, p < 0.001), ALP (R = 0.26, p < 0.01) and bile acids (R = 0.68, p < 0.001) and negatively with prealbumin (R−0.58, p < 0,001), thromboplastin time (R = −0.28, p < 0.001), and factor V (−0.40, p < 0.001). Additionally, APRI closely correlated with LSM at post‐KPE follow‐up (R = 0.72, p < 0.001) (Figure 1).
3.4. Relation to histopathology at KPE
A positive correlation of APRI and the neural network‐quantified DR% was observed at KPE (Figure 2). This association extended to discriminated DR components BE% and PIH%, both of which independently demonstrated similar associations with APRI (Figure 2). Both relationships were similar for manual scorings (BE scoring, R = 0.34, p < 0.001; PIH scoring, R = 0.33, p < 0.001). Inflammatory cell infiltration in portal areas or different histological cholestasis types did not correlate APRI at KPE (Figure 2). No significant correlations with Sirius Red‐quantified liver fibrosis or with Metavir staging were found at KPE (Figure 2).
Figure 2.
Association of APRI with liver histopathology at KPE. Spearman rank correlations between APRI and (A) DR%, (B) proportional Sirius Red ‐ liver fibrosis, (C) BE%, (D) PIH% at KPE. Violin plots of APRI and manual scores of (E) intracellular cholestasis, (F) intracanalicular cholestasis, (G) intraductal cholestasis, (H) portal inflammation, and (I) Metavir fibrosis stage at KPE. Violin plots display individual data points. APRI, aspartate aminotransferase to platelet ratio index; BE, biliary epithelium; DR, ductular reaction; PIH, parenchymal intermediate hepatocytes.
3.5. Relation to outcome at KPE
Patients who failed to normalise their bilirubin levels postoperatively had higher serum APRI at KPE compared to patients achieving COJ (0.80 vs. 0.96) (Figure 1). During a follow‐up period of 5 years, patients with an event presented with higher APRI at KPE compared to patients with their NL (Figure 1).
To assess clinical prognostic utility, APRI at KPE was stratified based on COJ and NLS using area under the receiver operating characteristic (AUROC) analysis. The AUROC value for APRI in predicting outcome were identical for COJ (0.62) and for 5‐year NLS (0.61) (Supporting Information: Figure S1). In regression survival models, APRI predicted NL survival in the entire KPE cohort and patients with associated anomalies (Supporting Information: Table S1).
3.6. Relation to histopathology during post‐KPE follow‐up
During the post‐KPE follow‐up, the relationship between APRI and DR% grew stronger compared to at KPE (Figure 3). However, while the correlation between APRI and PIH% accentuated, the correlation with BE% grew weaker during postoperative follow‐up. Manual scorings of PIH (R = 0.43, p < 0.001) and BE (R = 0.25, p = 0.004) showed comparable correlations. Post‐KPE APRI positively correlated with histological cholestasis scores, Sirius Red‐quantified liver fibrosis, and Metavir fibrosis stage, but not with portal inflammation score (Figure 3 and Supporting Information: Table S2).
Figure 3.
Association of APRI with liver histopathology at post‐KPE follow‐up. Spearman rank correlations between APRI and (A) DR%, (B) proportional Sirius Red ‐ liver fibrosis, (C) BE%, (D) PIH% in post‐KPE liver biopsies. Violin plots of APRI and manual scores of (E) intracellular cholestasis, (F) intracanalicular cholestasis, (G) intraductal cholestasis, (H) portal inflammation, (I) Metavir fibrosis stage at post‐KPE follow‐up. Violin plots display individual data points. APRI, aspartate aminotransferase to platelet ratio index; BE, biliary epithelium; DR, ductular reaction; PIH, parenchymal intermediate hepatocytes.
3.7. Relation to outcome at post‐KPE follow‐up
Patients failing to resolve cholestasis after KPE had significantly higher APRI at follow‐up when compared to patients with COJ (2.58 vs. 1.22, p < 0.001). In patients with APRI measured both at KPE and post‐KPE follow‐up, APRI continued to rise after KPE in patients who subsequently underwent LT (median: 1.09 vs. 2.07; p = 0.017, IQR: 1.50 vs. 1.55), while values remained largely unchanged in NL survivors (median: 0.85 vs. 0.58; p = 0.055, IQR: 0.83 vs. 0.96) (Figure 1). In patients with multiple follow‐up biopsies (n = 55), APRI correlated strongly with Metavir stage (R = 0.45, p < 0.001) and moderately with Sirius Red (R = 0.32, p < 0.001). Among native liver survivors with serial APRI measurements (n = 30), APRI levels decreased in over half (n = 18, 60%).
When including all patients sampled at follow‐up, higher serum APRI was observed in patients who subsequently underwent LT compared to NL patients (median: 2.07 vs. 0.70, p < 0.001).
During post‐KPE follow‐up, AUROC value for 5‐year NL survival was 0.75, while the optimal cutoff was more than two‐fold higher to that observed at KPE (Supplementary Figure S1). Following KPE, APRI was predictive for NL survival in all patients, patients without associated anomalies, patients operated <60 days of age, and patients with COJ in Cox regression analyses (Supporting Information: Table S1).
4. DISCUSSION
These findings suggest that APRI reflects primarily histological bile duct and portal injury but also associates with histological liver fibrosis after KPE. In addition, APRI linked with poor prognosis both in treatment‐naïve and post‐KPE BA.
APRI was first introduced as a non‐invasive alternative to liver biopsy in predicting advanced viral hepatitis in adults. 14 Since that, its use has been expanded to assess fibrosis and subsequent liver complications in paediatric hepatopathies, including BA, due to its simplicity and accessibility from routine blood tests. 14 , 15 Subsequent research has demonstrated a significant correlation between APRI values and histopathological fibrosis severity in biopsy‐confirmed BA cohorts. 16 , 17 As a result, APRI has been proposed for long‐term monitoring after KPE and to support the timing of LT. 5 , 18 However, discrepancies in cut‐off thresholds, generally ranging between 0.8 and 2.0, have limited APRI's reliability as a standalone predictor. 6 Most clinical protocols recommend using APRI in combination with other serum biomarkers and imaging modalities to improve diagnostic accuracy of disease progression in BA. 8 , 15 , 16 , 18 In a smaller study comparing different biomarkers, APRI was identified as the most predictive non‐invasive marker at presentation, with an AUROC cut‐off of 0.91, which is similar to our current findings. 19 Nonetheless, meta‐analyses and multicentre studies continue to report moderate overall accuracy, with variability across populations and disease stages. 6 , 20 All in all, higher APRI values link with poor outcomes following KPE. 21
Beyond BA, APRI has been examined in other paediatric and adult cholestatic hepatobiliary pathologies. 22 , 23 , 24 , 25 Where applicable, APRI showed moderate correlation with fibrosis, yet it consistently underperformed when compared to imaging or composite indices. 3 , 26 , 27 Consequently, it has not demonstrated sufficient diagnostic accuracy to replace biopsy or reliably stratify BA histology at the time of diagnosis. 6
DR is recognised as a histopathological response to cholestatic liver injury, characterised by cholangiocyte proliferation and activation of hepatic progenitor cells, promoting inflammation and fibrogenesis. 28 The extent of DR closely associates with the degree of disease progression and transplant‐free survival in BA, demonstrating its predictive value not only at KPE but also throughout the postoperative period. 29 , 30 Notably, DR continues to drive fibrogenesis even after successful KPE despite normalisation of bilirubin. 31 , 32 The prognostic ability of DR at the time of KPE is dominated by the expansion of BE secondary to cholangiocyte proliferation, while PIH becomes the main DR element observed among long‐term native liver survivors, most likely resembling transdifferentiating hepatocytes or hepatic progenitor cell activation. 13
Our results suggest an interesting relationship between APRI and DR, supported by evolution of correlations at different disease stages: initially with BE at KPE, and subsequently with PIH during follow‐up. These findings align with the relationships between DR components and outcomes, showing a shift from BE at KPE to PIH at follow‐up. 28 As liver disease progresses, hepatocyte injury becomes more extensive, leading to persistently elevated AST levels. 11 With advancing liver fibrosis and portal hypertension, splenic sequestration lowers platelets while decreased thrombopoietin production further contributes to thrombocytopenia 32 , 33 This relationship is reflected by the continuous increase in serum APRI in patients of our cohort who subsequently underwent LT. In the current study, among 12 patients with rising APRI values following KPE, 9 (75%) demonstrated both progression of Metavir fibrosis and development of portal hypertension, whereas 3 (25%) showed fibrosis progression without clinical signs of portal hypertension. On the other hand, decreasing APRI trend was observed in patients who survived with their native livers in line with milder disease progression. LSM is well associated with advanced stages of hepatic fibrosis and cirrhosis in adults and children. 3 , 16 , 17 Our results also showed that APRI was closely correlated with LSM leading us to assume that they may reflect similar histopathological processes in post‐KPE liver injury, which requires further research. Overall, our findings suggest that APRI reflects primarily biliary and portal injury with increasing correlation to liver fibrosis after KPE and could serve as an additional non‐invasive marker for progression of liver dysfunction in BA especially following KPE.
Limitations of this study include a retrospective study design and the use of archived liver specimens. Although liver histopathology was confirmed with human scorings, image processing neural network might be unexpectedly predisposed to technical inaccuracies when analysing liver sections collected over several decades. Quantification of entire Sirius Red‐positive area does not distinguish between portal, periportal, and centrilobular fibrosis while conventional scorings are exposed to human error and leave room for interpretation between scores. As liver specimens and laboratory tests were obtained for clinical purposes, complete follow‐up data were not available for all patients at each stage. Post‐KPE samples were available from only one centre, although recent outcome data appear comparable between the two study centres. 34 , 35
5. CONCLUSIONS
The purpose of this study was to examine APRI's relationship with histopathological features of liver injury and its clinical utility in a large cohort of BA patients. Our findings indicate that APRI is associated with histological biliary and portal injury, as evidenced by its correlation with DR and its components at the time of KPE and throughout postoperative follow‐up. Additionally, APRI's correlation with liver fibrosis only became evident during follow‐up. Taken together, these results suggest that APRI functions primarily as a marker of active biliary and portal injury. Further research is needed to explore the role of histological liver fibrosis in the disease course of BA and effectiveness of other complementary markers potentially enhancing the accuracy of non‐invasive predictors.
CONFLICT OF INTEREST STATEMENT
Nelli Sjöblom reports consulting income from Aiforia Technologies unrelated to the submitted study. All authors declare no conflict of interest.
Supporting information
Supplementary Figure 1. Prognostic utility of APRI for KPE outcomes. AUROC curve for (A) Clearance of jaundice and (B) native liver survival. Kaplan‐Meier survival curve. using optimal cut‐off for patients at (C) KPE and (D) follow‐up AUROC, area under the receiver. operating characteristic; APRI, aspartate aminotransferase to platelet ratio index; NLS, native liver. survival; KPE, Kasai portoenterostomy.
Supporting Material.
Supplementary Table 1. Cox survival analysis for APRI.
Supplementary Table 2. Spearman correlation analysis of follow‐up samples for APRI.
ACKNOWLEDGMENTS
The study was supported by Sigrid Juselius Foundation, Finnish Paediatric Foundation, Helsinki University Hospital Fund, The Finnish Association for Transplantation Surgery, and Mary and Georg C. Ehrnrooth Foundation. The funding sources were not involved in conduct of the research and publication process. Open access publishing facilitated by Helsingin yliopisto, as part of the Wiley ‐ FinELib agreement.
REFERENCES
- 1. Tam PKH, Wells RG, Tang CSM, et al. Biliary atresia. Nat Rev Dis Primers. 2024;10(1):47. 10.1038/s41572-024-00533-x [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Bezerra JA, Wells RG, Mack CL, et al. Biliary atresia: clinical and research challenges for the twenty‐first century. Hepatology. 2018;68(3):1163‐1173. 10.1002/hep.29905 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Liguori A, Zoncapè M, Casazza G, Easterbrook P, Tsochatzis EA. Staging liver fibrosis and cirrhosis using non‐invasive tests in people with chronic hepatitis B to inform WHO 2024 guidelines: a systematic review and meta‐analysis. Lancet Gastroenterol Hepatol. 2025;10(4):332‐349. 10.1016/S2468-1253(24)00437-0 [DOI] [PubMed] [Google Scholar]
- 4. Mansoor S, Yerian L, Kohli R, et al. The evaluation of hepatic fibrosis scores in children with nonalcoholic fatty liver disease. Dig Dis Sci. 2015;60(5):1440‐1447. 10.1007/s10620-014-3494-7 [DOI] [PubMed] [Google Scholar]
- 5. Muntean A, Kronfli R, Makin E, Davenport M. The AST‐to‐platelet ratio index (APRi) at kasai portoenterostomy: standing the test of time. J Pediatr Surg. 2023;58(12):2347‐2351. 10.1016/j.jpedsurg.2023.06.012 [DOI] [PubMed] [Google Scholar]
- 6. Brits E, Brown S, Botes L, Sempa JB, Pienaar M. Aspartate aminotransferase‐to‐platelet ratio index (APRi) as biomarker for liver damage in biliary atresia (BA): a meta‐analysis. J Pediatr Surg. 2025;60(4):162234. 10.1016/j.jpedsurg.2025.162234 [DOI] [PubMed] [Google Scholar]
- 7. Lind RC, Verkade HJ, Porte RJ, Hulscher JBF. Aspartate transaminase‐to‐platelet ratio index is not correlated with severity of fibrosis or survival in children with biliary atresia. J Pediatr Gastroenterol Nutr. 2012;54(5):698‐699. 10.1097/MPG.0b013e318244d19d [DOI] [PubMed] [Google Scholar]
- 8. Hukkinen M, Lohi J, Heikkilä P, et al. Noninvasive evaluation of liver fibrosis and portal hypertension after successful portoenterostomy for biliary atresia. Hepatology Communications. 2019;3(3):382‐391. 10.1002/hep4.1306 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Pakarinen MP, Hukkinen M, Nyholm I, Lohi J, Jahnukainen T. Decrease of post‐portoenterostomy liver fibrosis in a nationally centralized cohort of biliary atresia patients. J Pediatr Surg. 2025;60(11):162505. 10.1016/j.jpedsurg.2025.162505 [DOI] [PubMed] [Google Scholar]
- 10. Lampela H, Kosola S, Koivusalo A, et al. Endoscopic surveillance and primary prophylaxis sclerotherapy of esophageal varices in biliary atresia. J Pediatr Gastroenterol Nutr. 2012;55(5):574‐579. 10.1097/MPG.0b013e31825f53e5 [DOI] [PubMed] [Google Scholar]
- 11. Hukkinen M, Pihlajoki M, Pakarinen MP. Predicting native liver injury and survival in biliary atresia. Semin Pediatr Surg. 2020;29(4):150943. 10.1016/j.sempedsurg.2020.150943 [DOI] [PubMed] [Google Scholar]
- 12. Nyholm I, Hukkinen M, Lohi J, et al. Accumulation of altered serum bile acids predicts liver injury after portoenterostomy in biliary atresia. J Hepatol. 2025;83(2):440‐452. 10.1016/j.jhep.2025.01.025 [DOI] [PubMed] [Google Scholar]
- 13. Nyholm I, Hukkinen M, Pihlajoki M, et al. Serum FGF19 predicts outcomes of Kasai portoenterostomy in biliary atresia. Hepatology. 2023;77(4):1263‐1273. 10.1097/HEP.0000000000000048 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Wai CT, Greenson JK, Fontana RJ, et al. A simple noninvasive index can predict both significant fibrosis and cirrhosis in patients with chronic hepatitis C. Hepatology. 2003;38(2):518‐526. 10.1053/jhep.2003.50346 [DOI] [PubMed] [Google Scholar]
- 15. Kim SY, Seok JY, Han SJ, Koh H. Assessment of liver fibrosis and cirrhosis by aspartate aminotransferase‐to‐platelet ratio index in children with biliary atresia. J Pediatr Gastroenterol Nutr. 2010;51(2):198‐202. 10.1097/MPG.0b013e3181da1d98 [DOI] [PubMed] [Google Scholar]
- 16. de Lédinghen V, Le Bail B, Rebouissoux L, et al. Liver stiffness measurement in children using FibroScan: feasibility study and comparison with fibrotest, aspartate transaminase to platelets ratio index, and liver biopsy. J Pediatr Gastroenterol Nutr. 2007;45(4):443‐450. 10.1097/MPG.0b013e31812e56ff [DOI] [PubMed] [Google Scholar]
- 17. Voutilainen S, Kivisaari R, Lohi J, Jalanko H, Pakarinen MP. A prospective comparison of noninvasive methods in the assessment of liver fibrosis and esophageal varices in pediatric chronic liver diseases. J Clin Gastroenterol. 2016;50(8):658‐663. 10.1097/MCG.0000000000000532 [DOI] [PubMed] [Google Scholar]
- 18. Grieve A, Makin E, Davenport M. Aspartate aminotransferase‐to‐platelet ratio index (APRi) in infants with biliary atresia: prognostic value at presentation. J Pediatr Surg. 2013;48(4):789‐795. 10.1016/j.jpedsurg.2012.10.010 [DOI] [PubMed] [Google Scholar]
- 19. Takahashi Y, Matsuura T, Maeda S, et al. Factors predicting the need for liver transplantation in biliary atresia patients after 18 years of age. Pediatr Surg Int. 2024;40(1):218. 10.1007/s00383-024-05805-x [DOI] [PubMed] [Google Scholar]
- 20. Hukkinen M, Fischler B, Liliemark U, et al. Multicenter study on early predictors of biliary atresia outcomes. J Pediatr Surg. 2025;60(9):162404. 10.1016/j.jpedsurg.2025.162404 [DOI] [PubMed] [Google Scholar]
- 21. Suominen JS, Lampela H, Heikkilä P, Lohi J, Jalanko H, Pakarinen MP. APRi predicts native liver survival by reflecting portal fibrogenesis and hepatic neovascularization at the time of portoenterostomy in biliary atresia. J Pediatr Surg. 2015;50(9):1528‐1531. 10.1016/j.jpedsurg.2014.11.046 [DOI] [PubMed] [Google Scholar]
- 22. Shiau H, Guffey D, Loomes KM, et al. Biopsy validated study of biomarkers for liver fibrosis and transplant prediction in inherited cholestasis. Hepatology Communications. 2020;4(10):1516‐1526. 10.1002/hep4.1569 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Song C, Lewis S, Kamath A, et al. Primary sclerosing cholangitis: diagnostic performance of MRI compared to blood tests and clinical scoring systems for the evaluation of histopathological severity of disease. Abdominal Radiology. 2020;45(2):354‐364. 10.1007/s00261-019-02366-9 [DOI] [PubMed] [Google Scholar]
- 24. Lewindon PJ, Puertolas‐Lopez MV, Ramm LE, et al. Accuracy of transient elastography data combined with APRI in detection and staging of liver disease in pediatric patients with cystic fibrosis. Clin Gastroenterol Hepatol. 2019;17(12):2561‐2569. 10.1016/j.cgh.2019.03.015 [DOI] [PubMed] [Google Scholar]
- 25. Lee J, Vali Y, Boursier J, et al. Prognostic accuracy of FIB‐4, NAFLD fibrosis score and APRI for NAFLD‐related events: A systematic review. Liver Int. 2021;41(2):261‐270. 10.1111/liv.14669 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26. Stasi C, Leoncini L, Biagini MR, et al. Assessment of liver fibrosis in primary biliary cholangitis: comparison between indirect serum markers and fibrosis morphometry. Digest Liver Dis. 2016;48(3):298‐301. 10.1016/j.dld.2015.10.024 [DOI] [PubMed] [Google Scholar]
- 27. Halász T, Horváth G, Kiss A, et al. Evaluation of histological and non‐invasive methods for the detection of liver fibrosis: the values of histological and digital morphometric analysis, liver stiffness measurement and APRI score. Pathol Oncol Res. 2016;22(1):1‐6. 10.1007/s12253-015-9964-1 [DOI] [PubMed] [Google Scholar]
- 28. Sato K, Marzioni M, Meng F, Francis H, Glaser S, Alpini G. Ductular reaction in liver diseases: pathological mechanisms and translational significances. Hepatology. 2019;69(1):420‐430. 10.1002/hep.30150 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Xiao MH, Wu S, Liang P, et al. Mucosal‐associated invariant T cells promote ductular reaction through amphiregulin in biliary atresia. EBioMedicine. 2024;103:105138. 10.1016/j.ebiom.2024.105138 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Mavila N, Siraganahalli Eshwaraiah M, Kennedy J. Ductular reactions in liver injury, regeneration, and disease progression‐an overview. Cells. 2024;13(7):579. 10.3390/cells13070579 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Lampela H, Kosola S, Heikkilä P, Lohi J, Jalanko H, Pakarinen MP. Native liver histology after successful portoenterostomy in biliary atresia. J Clin Gastroenterol. 2014;48(8):721‐728. 10.1097/MCG.0000000000000013 [DOI] [PubMed] [Google Scholar]
- 32. Hukkinen M, Kerola A, Lohi J, Jahnukainen T, Heikkilä P, Pakarinen MP. Very low bilirubin after portoenterostomy improves survival of the native liver in patients with biliary atresia by deferring liver fibrogenesis. Surgery. 2019;165(4):843‐850. 10.1016/j.surg.2018.10.032 [DOI] [PubMed] [Google Scholar]
- 33. Mitchell O, Feldman DM, Diakow M, Sigal SH. The pathophysiology of thrombocytopenia in chronic liver disease. Hepatic Med: Evid Res. 2016;8:39‐50. 10.2147/HMER.S74612 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Davenport M, Makin E, Ong EG, Sharif K, Dawrant M, Alizai N. The outcome of a centralization program in biliary atresia: twenty years and beyond. Ann Surg. 2025;281(4):608‐614. 10.1097/SLA.0000000000006273 [DOI] [PubMed] [Google Scholar]
- 35. Hukkinen M, Kerola A, Lohi J, et al. Treatment policy and liver histopathology predict biliary atresia outcomes: results after national centralization and protocol biopsies. J Am Coll Surg. 2018;226(1):46‐57.e1. 10.1016/j.jamcollsurg.2017.09.009 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Supplementary Figure 1. Prognostic utility of APRI for KPE outcomes. AUROC curve for (A) Clearance of jaundice and (B) native liver survival. Kaplan‐Meier survival curve. using optimal cut‐off for patients at (C) KPE and (D) follow‐up AUROC, area under the receiver. operating characteristic; APRI, aspartate aminotransferase to platelet ratio index; NLS, native liver. survival; KPE, Kasai portoenterostomy.
Supporting Material.
Supplementary Table 1. Cox survival analysis for APRI.
Supplementary Table 2. Spearman correlation analysis of follow‐up samples for APRI.