Evaluation of sound speed for detection of liver fibrosis: prospective comparison with transient dynamic elastography and histology

Abstract

Objective: The degree of liver fibrosis determines the prognosis and treatment of patients with chronic viral hepatitis. Transient elastography (TE) has been accepted as a noninvasive method for assessment of liver fibrosis. Sound velocity (SV) changes are also dependent on elastic properties of tissue. The aim of this pilot study was to evaluate whether SV estimation of liver tissue allows the determination of fibrosis stages in patients with chronic viral hepatitis.

Methods: Prospectively, 50 healthy volunteers and 149 patients received stiffness (TE, 50-Hz vibrator, 5-MHz array) and SV (conventional ultrasound, C5-2-MHz transducer) measurements. Eighty-four patients received representative liver biopsies. The estimated SV and stiffness were compared using liver biopsy as a reference (METAVIR fibrosis stage [F] scoring system [Hepatology 1996; 24:289-293]). Descriptive statistics, analysis of variance, receiver operating characteristic curve analysis, and box plot analysis as well as intra-operator and interoperator reproducibility analyses were performed.

Results: The SV ranged from 1540 to 1650 m/s. The mean SV ± SD was significantly different between healthy volunteers (1559 ± 11 m/s) and patients with F0-F3 (1575 ± 21 mm/s) and F4 (1594 ± 18 m/s) disease (P < .001). For detection of liver cirrhosis, the area under the receiver operating characteristic curve for SV was 0.80 (95% confidence interval, 0.69-0.89). With a cutoff value of 1589 m/s, the sensitivity, specificity, and positive and negative predictive values of SV for detection of liver cirrhosis were 82%, 76%, 70%, and 86%, respectively. Sound velocity measurements were reproducible (15%) and had acceptable operator independence (19%).

Conclusions: The SV of liver tissue depends on the fibrosis stage. An SV of 1589 m/s or higher detects cirrhosis with high sensitivity. Therefore, SV measurement appears to be a promising new method for noninvasive quantification of liver fibrosis.