Genetic and lifestyle risk factors for advanced liver disease among men and women

Abstract

Background and Aim Liver disease is traditionally categorized as alcoholic and non‐alcoholic. We studied various risk factors predictive of advanced non‐viral liver disease in general population and analyzed the interaction between these factors and alcohol consumption. Methods Persons without underlying liver disease who participated in the Health2000 or FINRISK studies 1992–2012 comprised a cohort of 41 260 individuals. Pattern of alcohol consumption and metabolic, lifestyle‐related, and anthropometric parameters were analyzed with Cox regression analysis using severe liver disease hospitalization, cancer, or death as end‐point. Viral liver diseases were excluded. Results A total of 355 liver events occurred during the mean 12.4‐year follow‐up (511 789 person‐years). In the multivariate model, age (hazard ratio [HR] 1.03, P = 0.0083 for men; HR 1.04, P = 0.0198 for women), waist‐to‐hip ratio (WHR) (HR 1.52, P = 0.0006 for men; HR 1.58, P = 0.0167 for women), patatin‐like phospholipase‐containing domain 3 mutations (HR 1.9, P = 0.024 for men; HR 2.7, P = 0.0109 for women), and weekly binge drinking (HR 2.4, P = 0.0024 for men; HR 7.4, P < 0.0001 for women) predicted development of severe liver disease. Among men, diabetes (HR 2.7, P = 0.0002), average alcohol consumption (HR for 10 g/day 1.1, P = 0.0022), non‐married status (HR 1.9, P = 0.0397 for single; HR 2.4, P = 0.0002 for widowed/separated), and serum high‐density lipoprotein (HR 2.2, P = 0.0022) and non‐high‐density lipoprotein cholesterol (HR 1.2, P = 0.0237) were additional risk factors. Alcohol intake increased the risk especially among persons with high WHR (P for interaction 0.009). Conclusions Age, patatin‐like phospholipase‐containing domain 3 haplotype, and WHR increase the risk for development of severe liver disease. We found strong synergism between alcohol and central obesity. Binge drinking is an additional risk factor.

Interaction Between Alcohol Use and Metabolic Risk Factors for Liver Disease: A Critical Review of Epidemiological Studies

“…The majority of these studies come from Japan, and ethnicity may influence alcohol metabolism and therefore modify alcoholic liver toxicity (Kourkoumpetis and Sood, ). In fact, the findings in the Japanese studies are in sharp contrast with those in some similar Western studies (Alatalo et al, ; Lau et al, ; Ruhl and Everhart, ; Suomela et al, ) and in contrast with longitudinal general population‐based studies reporting increased risk for severe liver disease at alcohol intake even lower than 2 drinks per day (Åberg et al, ; Askgaard et al, ; Hagström et al, , ; Rehm et al, ; Sahlman et al, ; Simpson et al, ). Simple liver steatosis or elevated liver enzymes may be poor outcome measures in population studies due to unclear prognostic relevance.…”

“…Measures that better reflect abdominal, also called visceral or central, adiposity, such as waist circumference and WHR, seem superior to BMI in predicting incident liver disease (Åberg et al, ; Andreasson et al, ; Ioannou et al, ; Pang et al, ). The supraadditive interaction effect also seems to become stronger when obesity is measured by the WHR (Sahlman et al, ) or waist circumference (Åberg et al, ).…”

“…Extending these findings, among obese men in the general population with a waist–hip ratio (WHR) in the highest tertile, 1 daily alcohol drink yielded a similar relative risk for incident liver disease as 4 daily drinks in nonobese men (Fig. ; Sahlman et al, ). This means that, in the presence of marked abdominal obesity, hepatic toxicity of alcohol increases by a number of 4.…”

“…Interaction between average alcohol intake (grams of EtOH per day) and waist–hip ratio for the risk of incident clinical liver disease among men in the Finnish general population. (Reprinted with permission from Sahlman et al, ).…”

Insulin Resistance and Genetic Risk Predict Liver‐Related Outcomes and Death in Nonalcoholic Fatty Liver Disease

“…To validate this model built on cross‐sectional observations, we evaluated whether the PRS predicts incident liver‐related outcomes in the FINRISK 1992‐2012 and Health 2000 surveys, Finnish cross‐sectional population surveys . Subjects were linked with the National Registries for Hospitalizations, Death, and Cancer, to determine new diagnoses of cirrhosis, liver cancer, or liver‐related mortalities . Individuals with NAFLD at baseline, defined as a Fatty Liver Index score of 60 or higher, were included.…”