A pilot study estimating liver fibrosis with ultrasound shear-wave elastography: does the cause of liver disease or location of measurement affect performance?

Abstract

The purpose of this study was to evaluate the diagnostic accuracy of real-time shear-wave elastography for assessment of liver fibrosis in an unselected patient population, comparing shear-wave elastography measurements obtained at and remote from the site of random liver biopsy. In a prospective study of 50 patients (21 with and 29 without hepatitis C) referred for clinically indicated random liver biopsy for diffuse liver disease, shear-wave elastography measurements were taken from four locations before biopsy: one at the left lobe, two at the right lobe, and one at the biopsy location. The mean, minimum, maximum, and SD of shear-wave elastography were compared with pathologic grading. Steatosis and serum markers were analyzed using multiple logistic regression. Optimized shear-wave elastography thresholds were calculated using AUC analysis. The AUC (95% CI) at the biopsy site, ipsilateral lobe, and contralateral lobe were 0.82 (0.63-1.0), 0.84 (0.67-1.0), and 0.59 (0.19-0.99) in hepatitis C patients; 0.89 (0.75-1.0), 0.88 (0.73-1.0), and 0.93 (0.80-1.0) in nonhepatitis C patients; and 0.85 (0.74-0.96), 0.89 (0.79-0.99), and 0.80 (0.67-0.93) in all patients, respectively. Optimized biopsy site shear-wave elastography values for detecting Metavir score F2 or greater were 1.87 m/s (75% sensitivity and specificity), 2.00 m/s (80% sensitivity and specificity), and 1.89 m/s (76% sensitivity and specificity) in hepatitis C, nonhepatitis C, and all patients, respectively. Steatosis and serum markers were not significant. Real-time shear-wave elastography accurately predicted significant fibrosis (stage ≥ 2) in an unselected patient population with diffuse disease, including patients with and without hepatitis C. Shear-wave elastography best predicts pathologic grading when taken at the biopsy site or ipsilateral lobe in hepatitis C patients. Percentage steatosis was not predictive of shear-wave elastography results.