The combined effect of alcohol and obesity on risk of liver disease: a systematic review and meta-analysis

Abstract

Background: liver disease caused by alcohol and obesity is preventable by risk factor modification. Evidence from individual studies on risk in patients who have both risk factors is inconsistent. We performed a systematic review and meta-analysis to quantify the risk of liver disease associated with combinations of BMI and alcohol consumption.

Methods: the full study protocol is registered with PROSPERO (CRD42016046508). Inclusion criteria were: English language publications in peer-reviewed journals; adults from general population cohorts without pre-existing liver disease; BMI and alcohol quantifiably measured; follow-up at least 10 years duration. Outcomes were incident morbidity/mortality from cirrhosis and/or hepatocellular carcinoma. A one-stage meta-analysis was performed on original count data, using a Poisson regression log-linear model. The log-linear model has the property that joint effects are multiplicative. Reference categories were normal weight (BMI<25) and alcohol consumption above zero but within UK limits (<112 g/wk).
Results: the search identified 2589 studies. 50 underwent full text review. Eight cohorts (Two from the USA, six from Europe) were included in the meta-analysis, totalling 1,029,962 participants. There was no interaction between alcohol and obesity in the model. Risks of liver disease in those with both increased BMI and alcohol consumption were significantly increased and were multiplicative, as per the properties of the log-linear model (figure 1). Compared to normal weight participants drinking <112 g/wk (UK recommended limit), the relative risk of liver disease in those who were overweight and drinking above limits was 3.60 (95%CI 3.22-4.02) and the relative risk in those who were obese and drinking above limits was 5.84 (95%CI 5.09–6.70).

Conclusion: overweight and obese patients drinking >112 g/wk are at significantly increased risk of liver disease. This risk should inform lifestyle advice given to patients and risk stratification by healthcare professionals. Alcohol thresholds imposed in NAFLD criteria and pathways may mean that patients with both increased BMI and alcohol risk factors, at risk of Both Alcohol and Fatty Liver Disease (BAFLD), are not receiving appropriate risk assessment or clinical care. The current guidelines for alcohol consumption may not be appropriate for obese patients

More information

Identifiers

ORCID for Julie Parkes: orcid.org/0000-0002-6490-395X

ORCID for Dankmar Bohning: orcid.org/0000-0003-0638-7106

ORCID for Paul Roderick: orcid.org/0000-0001-9475-6850

ORCID for Ryan Buchanan: orcid.org/0000-0003-0850-5575

Catalogue record

Export record

Altmetrics