Digital quantification is more precise than traditional semiquantitation of hepatic steatosis: correlation with fibrosis in 220 treatment-naïve patients with chronic hepatitis C.

BACKGROUND: Steatosis, as associated with chronic hepatitis C (CHC) and non-alcoholic fatty liver disease (NAFLD), has been considered a risk factor for development of fibrosis. AIMS: Our aims were to determine if correlations existed between the degree of steatosis and fibrosis in treatment-naïve CHC patients, and to compare the accuracy of digital image analysis with semiquantification (manual assessment) to quantify hepatic steatosis.
METHODS: We studied 220 treatment-naïve, liver biopsy-proven CHC patients, including a serial biopsy sub-cohort of 37 patients with a mean interval of 3.82 years. Steatosis and fibrosis % were evaluated using digital quantification of steatosis (DQS) and fibrosis contrasted with manual assessment.
RESULTS: Most patients had <6% steatosis measured manually and digitally. Overall, manual assessment of steatosis was 3.78 times greater than DQS. Increasing steatosis % was associated with advancing fibrosis stage, both manually and digitally. Intraobserver reliability for DQS showed higher intraclass correlation reproducibility (r = 0.98, P < 0.001) than the manual method (r = 0.81, P < 0.01). Interobserver concordance for DQS had an average measure intraclass correlation of r = 0.99. Cirrhotics were more likely than non-cirrhotics to have grade 2 steatosis.
CONCLUSIONS: Increased steatosis was associated with increased fibrosis. DQS was consistently more precise and reproducible than manual assessment of steatosis in grades 1 (1 to <6%) and 2 (6 to <34%), and may prove to be especially preferable in clinical trials of pharmacotherapeutic agents.