Diagnostic evaluation of infantile cholestasis.

OBJECTIVE: To evaluate diagnostic accuracy of some important clinical manifestations and different investigations in infantile cholestasis. MATERIAL AND METHOD: Infants diagnosed with prolong conjugated hyperbilirubinemia and admitted to Chiang Mai University Hospital between Jan 1999 and Feb 2003. Demographic and clinical data were recorded Routine biochemical tests, and serology for TORCHS infections were carried out. An abdominal ultrasonography, DISIDA scan and percutaneous/open liver biopsy were performed. Hyperechoic band at the level of portal bifurcation, named triangular cord (TC) sign was blindly assessed on ultrasonography by the same radiologist. The patients were diagnosed as BA if either operative findings of atretic common bile duct/ gallbladder or evidence of bile duct obstruction demonstrated by intraoperative cholangiography was noted RESULTS: Sixty-one patients were diagnosed as BA (n = 31) and NH (n = 30) with an average age at diagnosis of 88.6 and 63.1 days respectively. Concerning clinical presentations, only the presence of acholic stool was significantly different between BA and NH (p = 0.006). The GGT level of greater than 500 IU/L was significantly found in BA (p < 0.001). The acholic stool and GGT level more than 500 IU/L were highly specific for BA at 100 and 96.6% respectively. In addition, the sensitivity and specificity of US-TC and DISIDA scan were 87.4, 100 and 89. 7, 92.0% respectively. The accuracy for diagnosis of BA were highest by DISIDA scan (96.3) followed by US-TC (86.9), GGT level of > 500 IU/L(81.0) and acholic stool (80.3) in order CONCLUSION: There was no single laboratory investigation that could precisely make a definite diagnosis of BA. The acholic stool and GGT level of higher than 500 IU/L were highly specific for BA. The TC in ultrasound is noninvasive and easily available tests when combined with acholic stool and the GGT level is suggested plan of management.