Incidence and risk factors for Non-Alcoholic Fatty Liver Disease in an urban, adult Sri Lankan population – a community cohort follow-up study

Abstract

INTRODUCTION: We previously reported a community prevalence of 33% for NAFLD in an urban, adult Sri Lankan population. We also found a significant association between patatin-like phospholipase domain containing 3 (PNPLA3) gene rs738409 polymorphism, and susceptibility to NAFLD in the same population, after testing 10 selected single nucleotide polymorphisms (SNPs) in a case control study. AIMS & METHODS: The aim of this study was to assess the incidence and risk factors for NAFLD in this population after seven years of follow-up. The study population consisted of 42-71-year-old adults, originally selected by age stratified random sampling from electoral lists from Ragama, Sri Lanka. The target population was screened initially in 2007 and subsequently invited back for re-evaluation in 2014. On both occasions they were assessed using a structured interview, clinical and anthropometric measurements, liver ultrasound, and biochemical and serological tests. NAFLD was diagnosed on established ultrasound criteria for fatty liver (two out of three criteria: increased echogenecity of the liver compared to kidney and spleen, obliteration of the vascular architecture of the liver and deep attenuation of the ultrasonic signal), safe alcohol consumption (Asian standards: 514 units/week for men, 57 units/week for females) and absence of hepatitis B and C markers. Non-NAFLD controls were defined as subjects who did not have any of the ultrasound criteria for NAFLD. We also performed an updated case-control study to investigate associations of selected genetic variants with incident NAFLD [SNPs: PNPLA3 (rs738409), LYPLAL1 (rs12137855), GCKR (rs780094), PPP1R3B (rs4240624) and NCAN (rs2228603), APOC3 (rs2854117 and rs2854116), ADIPOR2 (rs767870) and STAT3 (rs6503695 and rs9891119)]. RESULTS: Of the 2985 original study participants, 2155 (72.2%) (1244 women and 911 men; mean age 59.2 years [SD, 7.7]) participated in the follow-up assessment. 1322 [mean age 58.9 years (SD, 7.6), 483 (53.0%) men and 839 (67.4%) women] had NAFLD. Out of 795 [466 (58.6%) women] participants who did not have NAFLD in the original study, 365 [226 (61.9%) women, mean age 58.6 years (SD, 7.9)] had developed NAFLD after 7 years, giving an annual incidence rate 6.6%. On multivariate analysis, increased waist circumference [OR 1.96(1.30 – 2.97), p=0.001], BMI4 23 kg/m2 [OR 2.93(1.99 – 4.30), p50.001] and raised plasma triglycerides (TG) [OR 1.49(1.03 – 2.13), p=0.03] were independently predictive of incident NAFLD in this cohort, while raised BP and reduced HDL, were not. In the updated association study involving 1310 cases and 427 controls, we found borderline association with NAFLD at two of the 10 candidate loci: rs4240624 at PPP1R3B and rs738409 at PNPLA3 (one-tailed P=0.044 and 0.033, respectively). CONCLUSION: In this community cohort follow-up study in an urban, adult population in Sri Lanka, the annual incidence of NAFLD was 6.6%. Incident NAFLD was associated with features of the metabolic syndrome, and showed tendency of association at PNPLA3 and PPP1R3B gene polymorphisms. Disclosure of Interest: None declared