Prevalence and Risk Factors of Significant Fibrosis in Patients With Nonalcoholic Fatty Liver Without Steatohepatitis

BACKGROUND & AIMS: In patients with nonalcoholic fatty liver disease (NAFLD), nonalcoholic steatohepatitis (NASH) is a risk factor for the development of fibrosis. However, fibrosis has been observed in livers of patients without NASH. We aimed to estimate the prevalence of fibrosis in patients without NASH and risk factors for fibrosis. METHODS: We analyzed data from 1738 subjects (44.9% with severe obesity) in a cross-sectional liver biopsy cohort enrolled at referral centers in Italy and Finland. Biopsy specimens were analyzed histologically by a blinded pathologist at each center, and a diagnosis of NASH was made based on steatosis (≥5% of hepatocytes), hepatocellular ballooning, and lobular inflammation. We also collected data on demographic features, metabolic comorbidities, and genetic factors, and performed logistic regression analyses. Findings were validated using data from 118 consecutive patients with NAFLD who underwent sequential liver biopsies at tertiary referral centers in Italy. RESULTS: In the cross-sectional cohort, 132 of 389 patients (33.9%) with significant fibrosis had no NASH and 39 patients (10.0%) had no inflammation. The dissociation between NASH and fibrosis was significantly greater in patients with severe obesity (P < .005). Steatosis, ballooning, and lobular inflammation each were associated independently with significant fibrosis (P < .001); age, adiposity, fasting hyperglycemia, and the PNPLA3 I148M variant also were associated with fibrosis. In patients without, but not in those with, NASH, significant fibrosis was associated with steatosis grade and the PNPLA3 I148M variant. In patients without NASH, age, fasting hyperglycemia, ballooning, and inflammation were associated with fibrosis. In the validation cohort, 16 of 47 patients (34.0%) with clinically significant fibrosis did not have NASH at baseline. In patients with fibrosis without baseline NASH, worsening of fibrosis (based on later biopsies) was associated with fasting hyperglycemia and more severe steatosis (P = .016). CONCLUSIONS: In an analysis of biopsy specimens collected from patients with NAFLD at a single time point, one third of patients with significant fibrosis did not have NASH. We validated this finding in a separate cohort. In patients without NASH, fasting hyperglycemia, severe steatosis, mild inflammation or ballooning, and the PNPLA3 I148M variant identified those at risk of significant fibrosis.