PURPOSE: We evaluated the validity and reliability of magnetic resonance elastography (MRE) for staging hepatic fibrosis in patients with chronic hepatitis B. METHODS: The study included 73 patients with chronic hepatitis B and confirmed stages of pathological fibrosis. Two radiologists measured liver stiffness using MRE in all cases. We compared the area under the receiver operating characteristic (ROC) curve (Az) for distinguishing stages of fibrosis compared with MRE liver stiffness measurements and serum fibrosis markers. We used intraclass correlation coefficients to analyze interobserver agreement for measurements of liver stiffness and 2 one-sided t-tests to test the equivalence of the measurements by the 2 observers. RESULTS: ROC analyses revealed the significantly superior discrimination abilities of MRE for liver fibrosis staging (Az = 0.945 to 0.978 [Observer 1] and 0.936 to 0.967 [Observer 2]) to those of serum fibrosis markers (0.491 to 0.742) for both observers (P < 0.0004). The intraclass correlation coefficient between the 2 observers was excellent (ρ = 0.971), and the measurements of liver stiffness by the 2 observers were statistically equivalent within a 0.1-kPa difference (P = 0.0157) CONCLUSION: MRE is a valid and reliable technique for discriminating the stage of hepatic fibrosis in patients with chronic hepatitis B.
PURPOSE: We evaluated the validity and reliability of magnetic resonance elastography (MRE) for staging hepatic fibrosis in patients with chronic hepatitis B. METHODS: The study included 73 patients with chronic hepatitis B and confirmed stages of pathological fibrosis. Two radiologists measured liver stiffness using MRE in all cases. We compared the area under the receiver operating characteristic (ROC) curve (Az) for distinguishing stages of fibrosis compared with MRE liver stiffness measurements and serum fibrosis markers. We used intraclass correlation coefficients to analyze interobserver agreement for measurements of liver stiffness and 2 one-sided t-tests to test the equivalence of the measurements by the 2 observers. RESULTS: ROC analyses revealed the significantly superior discrimination abilities of MRE for liver fibrosis staging (Az = 0.945 to 0.978 [Observer 1] and 0.936 to 0.967 [Observer 2]) to those of serum fibrosis markers (0.491 to 0.742) for both observers (P < 0.0004). The intraclass correlation coefficient between the 2 observers was excellent (ρ = 0.971), and the measurements of liver stiffness by the 2 observers were statistically equivalent within a 0.1-kPa difference (P = 0.0157) CONCLUSION: MRE is a valid and reliable technique for discriminating the stage of hepatic fibrosis in patients with chronic hepatitis B.