The interaction of nonalcoholic fatty liver disease and smoking on mortality among adults in the United States.

BACKGROUND & AIMS: Nonalcoholic fatty liver disease (NAFLD) has become the most common liver disease in Western countries. Smoking and diabetes mellitus (DM) have been shown to increase mortality; however, whether NAFLD adds to the detrimental effect of smoking in DM and non-DM patients is unknown. We evaluated the possible interactive effect of NAFLD and smoking on mortality risk in a US population-based sample.

METHODS

Cross-sectional data from 11 205 participants in the third National Health and Nutrition Examination Survey were analysed. NAFLD was defined as ultrasonographic hepatic steatosis without evidence of other liver diseases. Proportional hazards regression modelling was used to test for the multiplicative interaction of NAFLD and smoking on overall mortality, controlling for DM.

RESULTS

36.5% of the participants had NAFLD of whom 21.1% were current smokers, while among non-NAFLD subjects, 26.2% reported current smoking. Smoking was associated with a hazard ratio (HR) of 2.23 (95% confidence interval (CI): 1.87-2.65) among non-NAFLD subjects, and 2.31 (95% CI: 1.33-2.92, P < 0.01) among NAFLD patients. In contrast, the HR for NAFLD was 1.01 (95% CI: 0.78-1.31, P = 0.96) among smokers and 0.98 (95% CI: 0.87-1.10, P = 0.73) among non-smokers. There was no evidence of interaction between NAFLD and smoking (HR = 1.01, 95% CI: 0.74-1.38, P = 0.94) in the combined model.

CONCLUSION

We found that smoking increased mortality by two-fold among the US population. Although the magnitude of the increase in mortality did not differ from that in non-NAFLD subjects, smoking represents a modifiable determinant of long-term outcomes in NAFLD patients.