Diagnostic accuracy of two-dimensional shear wave elastography in detecting hepatic fibrosis in children with autoimmune hepatitis, biliary atresia and other chronic liver diseases

Background Although fibrosis is the main determinant of liver stiffness, other disease-related factors usually disregarded in studies on liver elastography, such as inflammation and cholestasis, may influence liver stiffness. Objective To evaluate the accuracy of two-dimensional (2-D) shear wave elastography in assessing liver fibrosis in children with chronic liver disease by controlling for the confounding role of several disease- and patient-related factors. Materials and methods Three disease groups were studied: 1) various chronic liver diseases, 2) autoimmune hepatitis and 3) biliary atresia. The METAVIR (meta-analysis of histological data in viral hepatitis) score was used for fibrosis staging and grading of necroinflammatory activity. Multiple linear regression was used to evaluate the relationship between liver stiffness measurements and disease-related factors. The diagnostic accuracy of elastography for predicting fibrosis stages was assessed by calculating the area under the receiver operating characteristic curves. Results The various chronic liver diseases group (n=32; 7.1 +/- 4.9 [mean +/- standard deviation] years) showed liver stiffness of 8.9 +/- 5.0 (mean +/- standard deviation) kPa, the autoimmune hepatitis group (n=33; 8.1 +/- 4.4 years) of 7.1 +/- 2.7 kPa, and the biliary atresia group (n=19; 0.2 +/- 0.1 years) of 19.7 +/- 15.2 kPa. Liver stiffness measurements differed across METAVIR fibrosis categories in all disease groups. The highest values were found in biliary atresia, at fibrosis stages >= F2 (F2: 12.4 +/- 1.6 kPa, F3: 17.8 +/- 2 kPa, F4: 41.5 +/- 12.4 kPa). Liver stiffness was strongly associated only with fibrosis (P<0.0001) in various chronic liver diseases, but with necroinflammatory activity (P<0.0001) and fibrosis (P=0.002) in autoimmune hepatitis, and with age (P<0.0001), fibrosis (P<0.0001) and cholestasis (P=0.009) in biliary atresia. Optimal cutoffs for detecting advanced fibrosis (>= F3) were 16 kPa (area under curve: 0.98; sensitivity: 87.5%; specificity: 96.7%) in biliary atresia and 8.7 kPa (area under curve: 0.98; sensitivity: 93.8%; specificity: 96.1%) in other chronic liver diseases. Conclusion Two-dimensional shear wave elastography is reliable in assessing liver fibrosis in children with chronic liver diseases.

Automated summary

Two-dimensional shear wave elastography is reliable in assessing liver fibrosis in children with chronic liver diseases, controlling for the influence of various disease and patient-related factors.