O-35 EPIDEMIOLOGY, CLINICAL AND TISSUE CHARACTERISTICS OF A LARGE COHORT OF NAFLD/NASH FROM SOUTH AMERICA

Introduction and Objectives

Some of the highest rates of non-alcoholic fatty liver disease (NAFLD) in the world are present in the South American continent. Indeed, recent reports suggest that NAFLD is becoming a common cause of hepatocellular carcinoma in the continent. Nonetheless, little is known about the epidemiology and tissue finings of NAFLD in the region. We provide an extensive assessment of the inter-relation of NAFLD with metabolic variables as well as medication intake and biopsy findings in South America.

Materials and Methods

A retrospective chart review of patients with NAFLD from 5 countries in Latin America (Argentina, Brazil, Peru, Ecuador and Colombia) via the South American Liver Research Network (SALRN). Diagnosis of NAFLD was obtained via imaging reports and biopsies. Logistic regression models were used to examine associations between clinical and tissue characteristics with individual patient features. Each center was responsible for its own ethics approval.

Results

2722 patients from five different centers (and five different countries) were included in the analysis with proportions being the following: Argentina 556 (20%), Brazil 596 (22%), Colombia 1490 (55%), Ecuador 50 (2%) and Peru 30 (1%). The median age was 53 years (IQR 21-41) and median BMI 29 kg/m² (IQR 26-36). 63% were female. Biopsy reports were available for 35% (n=947) with 25% (n=232) of those showing significant fibrosis, 27% (n=254) severe steatosis, and 65% (n=616) inflammation. Only 17% of subjects had diabetes mellitus, 34% dyslipidemia, and 31% Hypertension. Median ALT for the entire cohort was 38 IU (IQR 25-65) and AST 28 IU (IQR 21-41). Of 1407 subjects with medication information, 29% were on lipid lowering agents, 12% on aspirin, 28% on metformin and 5% on vitamin E. Independent predictors of significant fibrosis (≥ F2) on biopsy were: Diabetes mellitus (OR =2.97, 95% CI, 2.12 – 4.15, p < 0.0001), hypertension (OR =1.59, 95% CI, 1.17 – 2.17, p = 0.003), and metformin (OR =2.71, 95% CI, 1.82 – 4.02, p < 0.0001). There was no statistically significant association between F≥ 2 fibrosis and obesity or overweight. Diabetes and Hypertension were both independently associated with severe steatosis (OR =1.93, p = 0.0001 and OR =2.13, p < 0.0001, respectively).

Conclusions

This study provides critical information defining the epidemiology of NAFLD in South America, showing important correlations between hypertension and diabetes mellitus with clinically significant biopsy findings.