MATERIALS AND METHODS: A prospective and observational study was conducted on consecutive infants with NC referred to the University of Malaya Medical Centre, Malaysia, between November 1996 and May 2004.
RESULTS: The 3 most common causes of cholestasis among the 146 infants with NC studied were idiopathic neonatal hepatitis (n = 63, 43%), BA (n = 35, 24%) and congenital cytomegalovirus hepatitis (n = 13, 9%). Common clinical features at presentation were jaundice (100%), hepatomegaly (95%), splenomegaly (52%) and pale stools (47%). Three clinical features noted to be sensitive for BA were the presence of acholic or variably acholic stools on admission, a liver which was firm/hard in consistency and a palpable liver of ≥4 cm (sensitivity of 77%, 80% and 94%, respectively), but the corresponding specificity was poor (51%, 65% and 39%, respectively). The stools of 2 children with BA were pigmented initially but became acholic subsequently.
CONCLUSIONS: We did not find any single clinical feature with sufficient sensitivity and specificity to differentiate BA from other causes of NC. Repeated inspection of stools colour is necessary as occasionally, patients with BA may have initial pigmented stools. Biochemical assessment and imaging studies are important in the assessment of any infant with NC.
METHODS: This was a descriptive, retrospective analysis, conducted at the Department of Paediatrics, University Malaya Medical Centre, Malaysia. All children who had esophagogastroduodenoscopy (EGD) and colonoscopy from January 2008 to June 2011 were included. An endoscopy was considered appropriate when its indication complied with the NASPGHAN and ASGE guideline. All endoscopic findings were classified as either positive (presence of any endoscopic or histologic abnormality) or negative (no or minor abnormality, normal histology); effecting a positive contributive (a change in therapeutic decisions or prognostic consequences) or non-contributive yield (no therapeutic or prognostic consequences).
RESULTS: Overall, 76% of the 345 procedures (231 EGD alone, 26 colonoscopy alone, 44 combined EGD and colonoscopy) performed in 301 children (median age 7.0 years, range 3 months to 18 years) had a positive endoscopic finding. Based on the NASPGHAN and ASGE guideline, 99.7% of the procedures performed were considered as appropriate. The only inappropriate procedure (0.3%) was in a child who had EGD for assessment of the healing of gastric ulcer following therapy in the absence of any symptoms. The overall positive contributive yield for a change in diagnosis and/or management was 44%. The presence of a positive endoscopic finding was more likely to effect a change in the therapeutic plan than an alteration of the initial diagnosis. A total of 20 (5.8%) adverse events were noted, most were minor and none was fatal.
CONCLUSION: The NASPGHAN and ASGE guideline is more likely to predict a positive endoscopic finding but is less sensitive to effect a change in the initial clinical diagnosis or the subsequent therapeutic plan.