BOB Ped Surg 2023 - Maria Soledad Jara Valdivia, CIPESUR - Presentation

Why This Test Matters

Biliary atresia exists as a diagnostic emergency because it leads to liver cirrhosis without early treatment 0:00. The problem is that liver biopsy — the traditional diagnostic tool — often cannot reliably separate BA from other causes of infant cholestasis 0:20. Portal fibrosis and bile duct proliferation are considered the histologic hallmarks of BA 0:40, but familial intrahepatic cholestasis, neonatal hepatitis, and Alagille syndrome can all produce nearly identical patterns 0:50. This overlap forces clinicians into a difficult position: operate on infants who may not have BA, or delay surgery in those who do.

The Core Problem

The diagnosis of BA can be challenging as its histopathologic features overlap with those of other pediatric cholestatic liver diseases 0:20. A pathologist examining a biopsy from a jaundiced infant sees portal fibrosis and proliferating bile ducts 0:40 — findings that could represent BA requiring urgent Kasai portoenterostomy, or neonatal hepatitis that will resolve with supportive care, or a genetic cholestasis that needs entirely different management. Clinical context helps, but the tissue itself often remains ambiguous.

How CD56 Immunostaining Works

During cirrhosis, biliary epithelial structures express immunohistochemical markers of immaturity such as CD56 1:10. CD56 is a neural cell adhesion molecule that appears on immature or regenerating bile ductules. The hypothesis is that the aggressive fibrosis and duct injury in BA drives more extensive ductular immaturity than other cholestatic conditions, making CD56 expression more prominent and widespread 1:10.

Jara Valdivia and colleagues tested this by staining liver biopsies from 89 infants: 30 with confirmed BA, 28 with non-cirrhotic liver disease (tumors, choledochal cysts, focal nodular hyperplasia), and 31 with cirrhosis from causes other than BA (autoimmune hepatitis, neonatal hepatitis, familial intrahepatic cholestasis) 3:05. Two pathologists, blinded to the clinical diagnosis, scored each biopsy as CD56-positive if more than 5% of ductular epithelial cells stained positive 3:20.

The results were striking. CD56 positivity had a sensitivity of 74% for BA versus 16% and 9% in the other groups 3:20, with specificity of 87.7% 3:35. The positive predictive value was 76% 3:45 and the negative predictive value was 80% 3:55, with an area under the curve of 0.81 4:05. When the investigators examined different staining thresholds within the BA group, they found that the 6-25% labeling range had higher sensitivity while maintaining adequate specificity 4:10.

What This Means in Practice

The test performs best as a rule-out tool. An NPV of 86.2% shows that if CD56 is negative, biliary atresia is unlikely 4:55. This matters because a negative result can spare an infant from unnecessary surgery 4:55. The specificity of 87.7% and likelihood ratio positive of 6.04 4:45 mean that a positive result substantially increases the probability of BA 4:45, though not enough to confirm the diagnosis on its own — you still need the full clinical picture.

The sensitivity of 74% 3:20 means the test will miss about one in cases of BA if used alone. This is the critical limitation 3:20. CD56 staining cannot replace clinical judgment, imaging, or intraoperative cholangiography 5:05. It adds information; it does not provide certainty.

Where Practice Remains Uncertain

This study compared BA against a mixed group of other cholestatic and non-cholestatic conditions 3:05. The real clinical question — can CD56 distinguish BA from neonatal hepatitis in an ambiguous case — requires head-to-head comparison in the subset of biopsies where the pathologist is genuinely uncertain. The study included only 31 patients with non-BA cirrhosis 3:05, and we do not know how many of those were the true diagnostic dilemmas.

The 5% threshold for calling a biopsy positive was chosen by the investigators 3:20, and while their subanalysis suggested the 6-25% range performed well 4:10, this cutoff needs validation in other centers. Immunostaining is not perfectly reproducible, and what one pathologist scores as positive another might call negative.

When to Use This Test

CD56 IHC stain is helpful as a complementary test in liver biopsies to rule out biliary atresia as an etiology for neonates presenting with cholestasis 5:05. Order it when the biopsy shows portal fibrosis and duct proliferation 0:40 but the clinical picture is ambiguous — normal gallbladder on ultrasound, equivocal HIDA scan, or features that could fit either BA or another cholestatic process. A negative result argues against BA and supports continued medical management 4:55. A positive result raises the probability of BA enough to warrant more aggressive investigation or earlier surgical exploration 4:45.

Do not use CD56 staining as a standalone test 5:05. Do not delay surgery in a clinically obvious case of BA to wait for immunostaining results. And do not assume a positive result confirms the diagnosis — it shifts probabilities, nothing more 4:45.