Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2017 Feb;65(2):491-500.
doi: 10.1002/hep.28899. Epub 2016 Dec 24.

Metabolic Profiling of Fatty Liver in Young and Middle-Aged Adults: Cross-sectional and Prospective Analyses of the Young Finns Study

Affiliations
Free PMC article

Metabolic Profiling of Fatty Liver in Young and Middle-Aged Adults: Cross-sectional and Prospective Analyses of the Young Finns Study

Jari E Kaikkonen et al. Hepatology. .
Free PMC article

Abstract

Nonalcoholic fatty liver is associated with obesity-related metabolic disturbances, but little is known about the metabolic perturbations preceding fatty liver disease. We performed comprehensive metabolic profiling to assess how circulating metabolites, such as lipoprotein lipids, fatty acids, amino acids, and glycolysis-related metabolites, reflect the presence of and future risk for fatty liver in young adults. Sixty-eight lipids and metabolites were quantified by nuclear magnetic resonance metabolomics in the population-based Young Finns Study from serum collected in 2001 (n = 1,575), 2007 (n = 1,509), and 2011 (n = 2,002). Fatty liver was diagnosed by ultrasound in 2011 when participants were aged 34-49 years (19% prevalence). Cross-sectional associations as well as 4-year and 10-year risks for fatty liver were assessed by logistic regression. Metabolites across multiple pathways were strongly associated with the presence of fatty liver (P < 0.0007 for 60 measures in age-adjusted and sex-adjusted cross-sectional analyses). The strongest direct associations were observed for extremely large very-low-density lipoprotein triglycerides (odds ratio [OR] = 4.86 per 1 standard deviation, 95% confidence interval 3.48-6.78), other very-low-density lipoprotein measures, and branched-chain amino acids (e.g., leucine OR = 2.94, 2.51-3.44). Strong inverse associations were observed for high-density lipoprotein measures, e.g., high-density lipoprotein size (OR = 0.36, 0.30-0.42) and several fatty acids including omega-6 (OR = 0.37, 0.32-0.42). The metabolic associations were attenuated but remained significant after adjusting for waist, physical activity, alcohol consumption, and smoking (P < 0.0007). Similar aberrations in the metabolic profile were observed already 10 years before fatty liver diagnosis.

Conclusion: Circulating lipids, fatty acids, and amino acids reflect fatty liver independently of routine metabolic risk factors; these metabolic aberrations appear to precede the development of fatty liver in young adults. (Hepatology 2017;65:491-500).

Figures

Figure 1
Figure 1
Cross‐sectional associations of metabolic measures with presence of fatty liver (n = 1,939‐2,002 of whom 339‐372 had diagnosed fatty liver in 2011). ORs and their 95% confidence intervals are per 1 standard deviation increment in the metabolic measures and shown with adjustment for sex and age (blue) and additionally for waist, alcohol intake, leisure‐time physical activity and smoking (red). P values listed in exponential format denote metabolite associations that were statistically significant associations when accounting for Bonferroni correction (P < 0.0007). Abbreviations: Apo, apolipoprotein; CI, confidence interval; DHA, docosahexaenoic acid; FA, fatty acid; HOMA‐IR, homeostatic model assessment of insulin resistance; IDL, intermediate‐density lipoprotein; LDL, low‐density lipoprotein; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; SFA, saturated fatty acid.
Figure 2
Figure 2
Prospective associations of metabolic measures with 10‐year risk for fatty liver (n = 1,516‐1,575 with metabolite data at the 2001 baseline, of whom 263‐275 had fatty liver diagnosed in 2011). ORs (95% confidence intervals) are per 1 standard deviation increment in the baseline metabolic measures and shown with adjustment for sex and age (blue) and additionally for baseline waist, alcohol intake, leisure‐time physical activity, and smoking (red). P values listed in exponential format denote metabolite associations that were statistically significant associations when accounting for Bonferroni correction (P < 0.0007). Individuals with suspected fatty liver in 2001 were excluded from analyses (alanine aminotransferase >30 U/L). Abbreviations: Apo, apolipoprotein; CI, confidence interval; DHA, docosahexaenoic acid; FA, fatty acid; HOMA‐IR, homeostatic model assessment of insulin resistance; IDL, intermediate‐density lipoprotein; LDL, low‐density lipoprotein; MUFA, monounsaturated fatty acid; PUFA, polyunsaturated fatty acid; SFA, saturated fatty acid.

Similar articles

See all similar articles

Cited by 18 articles

See all "Cited by" articles

References

    1. Cohen JC, Horton JD, Hobbs HH. Human fatty liver disease: old questions and new insights. Science 2011;332:1519‐1523. - PMC - PubMed
    1. Kwak MS, Kim D, Chung GE, Kim W, Kim YJ, Yoon JH. Role of physical activity in nonalcoholic fatty liver disease in terms of visceral obesity and insulin resistance. Liver Int 2014;35:944‐952. - PubMed
    1. Yasutake K, Kohjima M, Kotoh K, Nakashima M, Nakamuta M, Enjoji M. Dietary habits and behaviors associated with nonalcoholic fatty liver disease. World J Gastroenterol 2014;20:1756‐1767. - PMC - PubMed
    1. Lu H, Liu H, Hu F, Zou L, Luo S, Sun L. Independent association between nonalcoholic fatty liver disease and cardiovascular disease: a systematic review and meta‐analysis. Int J Endocrinol 2013;2013:124958. - PMC - PubMed
    1. Targher G, Day CP, Bonora E. Risk of cardiovascular disease in patients with nonalcoholic fatty liver disease. N Engl J Med 2010;363:1341‐1350. - PubMed

Publication types

MeSH terms

Feedback