الفهرس 
CholestasisOnly 14 pages are availabe for public view
 |  | from | 217 |  |  |
| | | from | 217 | | |
Abstract
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. It is of paramount importance to differentiate BA from other causes of NC early as the success of Kasai portoenterostomy is directly related to the age of the child. However, making a definite diagnosis before subjecting a child to laparotomy and peroperative cholangiogram is not an easy task. No single test is 100% accurate. This prospective study included 78 patients recruited over a period of two years (2009-2011), from the Pediatric Hepatology department, National Liver Institute, Minofiya University, and department of Pediatrics, faculty of medicine, Minofiya University. The studied infants were 53 infants with cholestasis and 25 non-hepatic infants. After confirming the final diagnosis, they were divided into three groups. The first group (Gr.I) including patients with BA (n = 27). Patients of Gr.I who underwent Kasai operation were 24 cases. They underwent follow up after Kasai operation (clinically, US color Doppler and laboratory). The second group (Gr.II) including patients with cholestasis due to other causes than BA (n = 26). The third group (Gr.II) was non-hepatic infants and neonates (n = 25). Our results showed that: 1- Clinically: Clay stool was found in 92.6% of BA and in a considerable percent (42%) of non-BA. Clay stool had a high sensitivity (92.5%) and NPV (88.2%), but low specificity 166 (57.69%), and PPV (69.4%) in discriminating between BA and non-BA group (P <0.01). 2- Laboratory: a- GGT level were significantly higher in BA patients (416.1 U/L ± 214.5) than those in non-BA group (247.5 U/L ± 288.14) and at a cut-off value of >286 U/L, it had 74% sensitivity, 73 % specificity, 74% PPV, 73% NPV and 73.6% accuracy in discriminating between BA and non-BA group (P<0.05). b- Prothrombin concentration at a cut-off value of >91% had 77.8% sensitivity, 61.5 % specificity, 68.75% PPV, 76.2% NPV and 71.6% accuracy in discriminating between BA and non-BA group (P<0.05). c- Platelets count at a cut-off value of >349 x10³/μL has a sensitivity of 70.4%, 61.5 % specificity, 65.5% PPV, 66.6% NPV and 66% accuracy in discriminating between BA and non-BA group (P <0.05). 3- Histopathology: a- Ductular proliferation in liver biopsy had the best performance among liver biopsy criteria in diagnosis of patients in BA, where it had 100% sensitivity and 100% NPV, 86.95% specificity, 90% PPV, and 94% accuracy. The presence of bile plugs had 96.29% sensitivity, 65.2% specificity, 76.47% PPV, 93.75% NPV and 82% accuracy and the presence of intracellular cholestasis had 100% sensitivity and 100% NPV, but low specificity (34.7%) and PPV (64.2%) with 70% accuracy in favor of BA cases (P <0.01 for all) b- The majority of BA group showed either fibrous expansion of most portal tracts (44.4%) or focal porto-portal bridging (40.7%), while the majority of non-BA group showed absent or portal expantion of some portal tracts (52.2%) (P <0.01). c- Multinucleated giant hepatocytes was predominant feature in Gr.II (81.8%) more than that in BA group (22.2%) (P<0.05). The presence of such feature in liver biopsy had 77.7% sensitivity, 81.8% specificity, 84% PPV and 75% NPV in discriminating between BA and non-BA group in favor of non-BA group (P <0.01). 4- Ultrasonography and Doppler: a- The mean GB length in BA (1.7 cm ± 0.8) was significantly shorter than those in non-BA (2.4 cm ± 0.6) (P <0.05). GB length at a cutoff value of <20.5 mm had 75% sensitivity, 72.2% specificity, 72.2% PPV and 77.2% NPV in discriminating between BA and non-BA group (P<0.001). b-The presence of non-contractile, rudimentary or nonvisualized GB had high sensitivity (92.5%) and NPV (87.75%), but lower specificity (53.8%) and PPV (67.75%) and 73.6% accuracy in discriminating between BA and non- BA group in favor of BA (P <0.01). c- Positive TC-sign had low sensitivity and NPP (59.2%, 67.5% respectively), but high specificity and PPV (88.46%, 84.2% respectively) in diagnosing BA. 168 d-HAD was significantly higher in BA (2.5 ± 0.55 mm) than that in both non-BA (1.87 ± 0.63) and non-hepatic control group (1.6 ± 0.47) groups (P<0.05 for both), while there was no significant statistical difference between non-BA and non-hepatic control group (P>0.05 for both). HAD at a cutoff value of >2.05 mm had 77.8% sensitivity, 73.1% specificity, 77.7% PPV, 76.9% NPV and 77.34% accuracy in diagnosis of BA pateints (P <0.01). f-Hepatic subcapsular flow was found in 96.3% of BA, while it was found only in one case of non-BA and not found in non-hepatic control group (P<0.01). Positive hepatic subcapsular flow had 96.29% sensitivity, 96.15% specificity, 96.29% PPV 96.15% NPV and 96.2% accuracy in discriminating BA from non-BA patients (P <0.001). g- HARI was (0.72 ± 0.07) in BA, (0.71 ± 0.12) in non-BA group and there was no statistical significant difference between both (P >0.05), while both groups were significantly higher than non-hepatic control group (0.66 ± 0.09) (P <0.05). h- HAD/PVD ratio was significantly higher in BA (0.495 ± 0.13) than that in non-BA group (0.389 ± 0.11) and nonhepatic control group (0.34 ± 0.08) (P <0.005), but not significantly different between both of non-BA group and non-hepatic control group (P >0.05). HAD/PVD ratio at a cutoff value of >0.45 has 63% sensitivity, 76.9% specificity, 73.9% PPV, 66.6% NPV and 69.8% accuracy in favor of BA diagnosis (P <0.01). j- All BA cases that underwent DTA test (n = 17) had no bile-stained secretions. In addition (3/8) of non-BA group had no bile-stained secretions in DTA test and (5/8) had bilestained aspirate. Absence of bile in DTA, for cases that underwent this test, had high sensitivity (100%), NPP (100%) and PPV (85%), but low specificity (62.5%), in favor of BA diagnosis (P <0.01). When combining different parameters together, we found that the presence of clay stool, non-contractile GB and hepatic subcapsular flow had a sensitivity of 92.6% and 100% specificity, 100% PPV, 92.6% NPV and 96.2% accuracy. Which may help early and non-invasive diagnosis of BA especially with the drawbacks that may be faced depending on invasive tequniqeues (liver biopsy, hepatobiliry scentigraphy and DTA test). TC-sign, hepatic subcapsular flow, mean GB length, non-contractile GB, DTA test , AST, ALT, GGT, Prothrombin %, Platelets, HAD, HAD/PVD ratio, bile plugs, absent multi-nucleated giant hepatocytes and grade of fibrosis with their significant cut-off values and significances had the best stepwise multivariate logistic regression model of factors predicting BA where mean GB length, non-contractile GB, DTA test, bile plugs, absent multi-nucleated giant hepatocytes and the highest grade of fibrosis are considered independent predictors for the patients of Gr.I while other variables are considered dependant predictors. 170 After Kasai operation, on follow up we found that: 1- Cholangitis was the most frequent post-operative complication (50%) of type-II BA post-Kasai patients. 2- The mean age at Kasai operation in BA patients with failed Kasai did not differ (80.17 ± 15.13 days) from those with successful Kasai (76.42.17 ± 12 days), yet it was lower in those with successful Kasai. 3- BA Patients with failed Kasai were mainly type-III BA (76.5%) and had more frequent post-operative complications during follow up period. 4- Follow up of HARI in post-Kasai patients by color Doppler, did not differ significantly between BA patients who had failed vs. succeed Kasai. However, the mean of HARI in patients who had failed Kasai then died by liver failure was increased ≥ 0.80 which was higher than that in patients with successful Kasai ≥0.72. In conclusions, clinical, laboratory, ultrasonography, color Doppler US and histopathological findings could significantly discriminate BA from other causes of NC, with especial emphasis to hepatic subcapsular flow that has the best accuracy. In addition color Doppler US may be of benefit in follow up of BA patients after Kasai operation in particular HARI.