Nonalcoholic Fatty Liver Disease Is Associated With Decreased Lung Function

Chang-Hoon Lee; Seung Ho Choi; Goh Eun Chung; Boram Park; Min-Sun Kwak


Liver International. 2018;38(11):2091-2100. 

In This Article


We evaluated the association between NAFLD and lung function in healthcare examinees. The major findings are as follows. Firstly, NAFLD was associated with decreased lung function at baseline. Secondly, NAFLD itself was not associated with increased lung function decline rate in longitudinal analysis. Thirdly, the intermediate to high probability for hepatic fibrosis as evaluated by noninvasive methods was associated with an increased FVC decline rate. Figure 3 shows a schematic summary of these results.

Figure 3.

Schematic summary of study results. Non-alcoholic fatty liver disease is associated with decreased lung function, but not with increased lung function decline rate. However, NAFLD with fibrosis is associated with rapid decline of FVC

Several previous cross-sectional studies showed an association between NAFLD and reduced lung function. One study showed an inverse association between NAFLD and decreased FVC and FEV1 in men.[11] We recently reported that NAFLD (as defined by a high ALT level) was significantly associated with reduced lung function, especially in males, on the basis of nationwide survey data.[14] This study, which is the largest study ever adjusting for potential metabolic parameters based on detailed questionnaire and laboratory findings, showed consistent results presenting an inverse association between NAFLD and lung function (FEV1 and FVC). In this study, NAFLD was associated with decreased lung function not only in males but also in females. The inconsistency with our previous results,[14] which showed reduced lung function in males only, is probably due to different definitions of NAFLD. NAFLD was defined as elevated ALT in our previous study; however, it was defined by ultrasonography in this study. In addition, study size might also have influenced this difference. This study is the largest-ever study of its kind, including 11 892 subjects; thus, the previously insignificant trend between NAFLD and lung function in females was noted as significant in this study. Systemic inflammation[27,28] and insulin-resistance-related muscle dysfunction[29,30] could be plausible mechanisms explaining this link. Interestingly, the association between NAFLD and lower lung function remains little changed in our study, even after adjustment for different metabolic covariates, which suggests that NAFLD per se is independently associated with reduced lung function.

Although our study and previous studies show that NAFLD is strongly associated with reduced lung function, it is undetermined whether NAFLD affects lung function or vice versa in the cross-sectional design. In addition, lung function decline is a particularly important health outcome in the general population[15,16] and in people with chronic respiratory diseases.[17,18] To our knowledge, this is the first study investigating the impact of NAFLD on lung function decline. In longitudinal analysis of the PS-matched cohort, there was no significant difference in lung function decline rate between people with and without NAFLD in either sex. Interestingly, however, the FVC decline was accelerated in NAFLD patients with fibrosis evaluated by noninvasive method. In fact, it has been recognized that NASH and fibrosis is more strongly associated with extrahepatic complications and poor outcome than simple steatosis is. Simple steatosis shows a benign long-term outcome without excess mortality compared with the general population, while patients with NASH or fibrosis have increased long-term mortality with a hazard ratio of 1.004–1.05 and a standardized mortality ratio of 1.34–2.6.[31] Extrahepatic complications of NAFLD such as decreased kidney function[32] and incident diabetes[33] become more common as the severity of fibrosis increases. Consistent with these previous studies showing the association between hepatic fibrosis and extrahepatic complications, this study showed that hepatic fibrosis was associated with accelerated lung function decline. There are several plausible explanations for this relationship. Firstly, chronic inflammation could be a common element of hepatic fibrosis and lung function decline. Inflammatory cytokines and chemokines[34] including tumour necrosis factor-alpha[35,36] and transforming growth factor beta[36,37] have been reported to be associated with both liver fibrosis and lung function decline. Adiponectin, an anti-inflammatory cytokine, is also associated with liver fibrosis[38] and decreased lung function.[28] However, we could not analyse the mediation effect of inflammatory cytokines in our study. Secondly, the ageing process, which is accompanied by the accrual of inflammation according to the inflamm-ageing theory, could be a common mechanism of hepatic fibrosis and lung function decline. Hepatic fibrosis in NAFLD is associated with ageing[39] via increased reactive oxygen species formation, DNA damage, and activation of the NF–κB pathway.[39,40] Lung function decline is also an ageing process secondary to changes including loss of parenchymal tissue and gas exchange surface, reduction in intercostal muscle mass and strength and decrease in chest wall compliance.[41] The third plausible explanation is the gut-lung microbiota axis. Gut dysbiosis can induce colonic inflammation and bacterial translocation, which accelerate the progression of NAFLD.[42] Lung dysbiosis is associated with lung function.[43] Recently, an interchange between gut and lung microbiota has been noted, suggesting that changes in one compartment could affect the microbial composition and function in the other compartment.[44]

This study has several strengths. Firstly, this is the first longitudinal study evaluating changes in lung function according to the presence of NAFLD. We included subjects who were followed up for more than 3 years and thus underwent spirometry more than three times in addition to the baseline evaluation. Thus, we could evaluate the change in lung function precisely. Secondly, this is the largest study ever to investigate the association of NAFLD and lung function. Our cross-sectional analysis validated the previously suggested inverse association between NAFLD and lung function. Thirdly, we adjusted for a comprehensive set of potential metabolic parameters, including anthropometric measurements, lipid panel metrics and past medical history including history of diabetes and hypertension. In addition, a detailed questionnaire reviewed underlying chronic liver disease and lung disease, which could have affected the results. Smoking status, which is one of most important factors affecting lung function, was also accurately adjusted for. Fourthly, we used appropriate statistical analysis including both covariate-adjusted regression and PS-matched analyses.

We should acknowledge several limitations. Firstly, we diagnosed fatty liver disease on the basis of ultrasonography, as invasive histological confirmation via liver biopsy would have been unethical in the context of routine screening and health checkups. Although ultrasonography has inter– and intra-observer variability, it has the strengths of noninvasiveness, satisfactory sensitivity and specificity, low cost, and repeatability.[45,46] To reduce inter– and intra-observer variation, we ensured that all ultrasonographic examinations were performed by board-certified radiologists using standard methods. Secondly, hepatic fibrosis was evaluated by noninvasive methods, not by histological examination, although NFS and FIB–4 have been well validated through many previous studies.[21,47] Thirdly, we could not clarify the mechanisms of this association between NAFLD and decreased lung function. We could only provide some hypotheses.

In conclusion, NAFLD was associated with decreased lung function but not with lung function decline rates in a propensity-matched cohort. Hepatic fibrosis, as evaluated by noninvasive methods, was significantly associated with increased lung function decline in the longitudinal analysis. This study is the first large-scale longitudinal study evaluating the association between NAFLD and lung function.