A comparative study between ultrasound evaluation, liver stiffness measurement and biopsy for staging of hepatic fibrosis in patients with chronic liver disease

Background: Percutaneous liver biopsy (LB) is the gold standard test for staging of hepatic fibrosis. Liver stiffness measurement (LSM) is used as non-invasive method for fibrosis, while Ultrasound (US) is principally used for detection of liver complications. Aim: This prospective study blindly investigates the accuracy of ultra- sonographic parameters in predicting the fibrosis staging compared to LB and LSM, in chronic liver disease. Methods: 61 patients (41/20 M/F, 19 HCV, 19 HBV, 23 NASH) who performed a valid LB (specimen length > 200mm) for staging of hepatic damage were consequently enrolled. Within 3 months they have been analyzed for LSM and US by two experienced blind gastroenterologists. Biopsy specimens were scored following Metavir in viral patients and Brunt in NASH. At least ten successful LSM (success-rate > 60% and IQR < 30%) were obtained. Nineteen US and Doppler parameters (liver surfaces, boundaries and echogenicity, 6 hepatic diameters, portal and median sovrahepatic venous flow and diameter, splenic diameter) were evaluated and summarized in a US classification ranging from F0 (absence of fibrosis) to F4 (signs of cirrhosis). Results: Patients’ distribution was Metavir/Brunt F0=5, F1=18, F2=16, F3=3, F4=19. LSM was significantly correlated with fibrosis stage (Spearman’s rho=0.6245, p = 0.0019). US classification was signifi- cantly correlated with histological fibrosis stage (Spearman’s rho=0.7642, p < 0.0001). The agreement between the two observers for US signs of fi- brosis was good (unweighed-kappa=0.4913, p < 0.0001). All cases staged as F3 and F4 at US corresponded to cirrhosis at biopsy. The 95% of F0 and F1 at US corresponded to F1 or F2 at histological analysis. The detection at US of irregular liver surface was always indicative of cirrhosis (SP 86%, SN 63%). A discordance between US/histological/LSM investigations was registered in patients with moderate fibrosis (F2), specially in presence of steatosis. In NASH, epatomegaly with steatosis alone at US was related to presence fibrosis (p < 0.05). Conclusions: Our preliminary data suggest that an US pointed at specific parameters is accurate in predicting staging of fibrosis in liver disease and it might be used in association with other non-invasive tests to ameliorate the diagnosis of fibrosis.