الفهرس 
INTRODUCTIONOnly 14 pages are availabe for public view
 |  | from | 221 |  |  |
| | | from | 221 | | |
Abstract
Biliary atresia is an idiopathic inflammatory process involving the bile ducts resulting in obstruction of biliary tract, chronic cholestasis, progressive fibrosis and eventually biliary cirrhosis. The basic etiology of BA is still not clear.
The differential diagnosis of BA includes other disorders of NC and the most important objective in such cases is to distinguish atretic cause from non-atretic causes. The diagnosis of BA, particularly distinguishing it from other causes of liver injury in the neonatal period, is challenging as there is a high degree of overlap in clinical, biochemical, imaging, and histological characteristics of BA and other causes of NC. There is an increased need of early and correct diagnosis of BA because timely surgical portoenterostomy is necessary for improved biliary drainage. Surgery therefore in BA has to be performed at an earliest opportunity preferably within first 8 weeks of life without loss of time. What investigations can distinguish BA from other non-BA conditions has been a matter of debate.
Therefore, this study aimed to assess hepatic immunohistochemical expression of HES1 and SOX17 in BA cases and some other neonatal cholestatic disorders due to causes other than BA, in order to investigate their suggestive pathogenic and diagnostic roles in BA.
This retrospective study included 61 infants with neonatal cholestasis (32 with BA and 29 with non-BA cholestasis) in whom liver biopsy was indicated for etiological diagnosis. For control purpose, ten liver specimens were obtained from cases without
cholestasis. The studied cases were divided into BA, non-BA and control groups.
BA and non- BA cases were age matched; the median age of BA cases was 70 days and in non- BA cases was 60 days. BA group showed predominance of female gender (59.4%), while male gender was predominant in non- BA group (62.1%).
The current study showed that, out of all clinical and radiological findings; clay stool, hepatomegaly and non–contractile gallbladder emerged as indicators of BA and could achieve significant statistical difference between BA and non- BA groups with P- value <0.05.
As regard liver function tests and CBC parameters; liver transaminases (AST and ALT), GGT and platelets were significantly higher in BA cases than that in non- BA cases with (P <0.05). GGT was the most accurate test (70%), achieving 91% sensitivity and 70% specificity in discrimination between BA and non-BA groups.
Many of histopathological findings showed significant statistical difference between BA and non-BA groups. Portal changes such as; portal tract edema, intraductal and intracanalicular bile plugs, generalized bile ductular proliferation and marked bridging fibrosis between portal tracts were more prominent in BA cases than that in non- BA cases (P<0.05). In contrast, hepatocellular changes such as; hepatocellular swelling and extramedullary hematopoiesis were more prominent in non- BA cases than that in BA cases with P-value <0.05.
Lee and Looi histological scoring system could achieve significant statistical difference between BA and non-BA groups with 94% sensitivity, 80% specificity and a cut off value of (7)
Our study showed that HES1 was expressed in 84.4% of BA cases compared to 37.9% in non- BA cases and 60% in control cases. These results could achieve highly significant statistical difference between BA and non- BA groups with P-value <0.0001 and a clinical performance of 81% sensitivity, 70% specificity, 68% PPV, 82% NPV and 75% as diagnostic accuracy in discrimination between BA and non- BA cases. Furthermore, more than half of positive BA cases for HES1 (51.9%) showed high expression of HES1 compared to only 18.2% in non- BA cases. In contrast, there was no significant statistical difference between BA and control groups as well as non- BA and control groups regarding HES1 expression (P>0.05).
The current study showed that SOX17 immunoreactivity was positive in all of the studied cases. A bout 90% of BA cases showed low expression of SOX17, while more than half of non- BA cases (55.2%) showed high SOX17 expression and this achieved highly significant statistical difference between BA and non- BA groups with P-value <0.0001. There was no significant statistical difference between BA and control groups as well as non- BA and control groups regarding SOX17 (P>0.05).
By studying the relation between both of the immunohistochemical stains (HES1 and SOX17), we found that there was trend to negative correlation between HES1 score and SOX17 score but not reach the level of statistical significance (P=0.063)
The present study showed significant association between positive HES1 expression and many of the studied variables that were associated to BA such as; clay stool (P<0.05), elevated GGT level (P<0.05) and many histopathological findings such as; intraductal and intracanalicular bile plugs, portal tract edema, marked bridging
fibrosis of portal tracts, mild periductular neutrophils and mild, generalized bile ductular proliferation (P< 0.05). HES1 also showed significant positive correlation to Lee and Looi histological score.
Regarding SOX17, low expression of SOX17 showed significant association with clay stool (P<0.05), non-contractile GB (P<0.05) and high level of each of AST, GGT and platelets (P<0.05) as well as many histopathological findings such as; intraductal and intracanalicular bile plugs, necrosis of few hepatocytes and mild infiltration of portal tracts by inflammatory cells, (P< 0.05), most of these variables were associated to BA. Morever, there was negative significant correlation between SOX17 score and Lee and Looi score.