Dear Editor
We read with interest the meta-analysis on global prevalence of non-alcoholic fatty liver disease (NAFLD) by Younossi et al. (1) published ahead of print in Hepatology. We commend the authors for undertaking such a task given the fractured state of the literature. However, their selection of studies for the meta-analysis seems to be incorrect. Specifically, Fig. 3e, which depicts NAFLD prevalence in North America based on non-invasive imaging, includes data from 13 published articles. Nine of these studies are from the same cohort (NHANES III), however, only one of these should have been included (2).
Furthermore, the study by Church et al. (3) contained participants selected by nine categories of body mass index (BMI) and fitness with equal number of participants per category, and cannot be used to calculate prevalence of NAFLD, which is strongly correlated with BMI. In fact, any samples where the sampling frame was based on risk factors correlated with the outcome (NAFLD), such as BMI, diabetes mellitus, hypertension, dyslipidemia, or other components of the metabolic syndrome, should not be combined with population-based samples in a meta-analysis because clinical heterogeneity is too large to derive meaningful pooled estimates (2).
With regard to Fig. 3b, the study by Kim et al. (4) was conducted among patients who received a liver transplant and thus cannot be used to estimate prevalence in the general population. The study by Hong et al. (5) had a 16.3% prevalence of NAFLD per Supplementary Table A (1), which is correct, but in Fig. 3b the prevalence was incorrectly calculated at 24.77%. Finally, we identified at least 10 more studies from Asia (Japan and China in particular) that were not included in the meta-analysis by Younossi et al. (1).
We have not checked other results presented; however, based on the above errors, we have concerns about the validity and meaningfulness of the study as a whole.
Acknowledgments
Funding
The work was financially supported in part by a grant from the National Institute on Alcohol Abuse and Alcoholism at the National Institutes of Health (grant number R21AA023521-01A1 to MR). The sponsor of the study (NIAAA) had no role in the writing of this letter.
Footnotes
Conflict of Interest: None declared.
References
- 1.Younossi ZM, Koenig AB, Abdelatif D, Fazel Y, Henry L, Wymer M. Global Epidemiology of Non-Alcoholic Fatty Liver Disease-Meta-Analytic Assessment of Prevalence, Incidence and Outcomes. Hepatology. 2015 doi: 10.1002/hep.28431. [DOI] [PubMed] [Google Scholar]
- 2.Higgins JPT, Green S, editors. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011] The Cochrane Collaboration; 2011. Available from www.cochrane-handbook.org. [Google Scholar]
- 3.Church TS, Kuk JL, Ross R, Priest EL, Biltoft E, Blair SN. Association of cardiorespiratory fitness, body mass index, and waist circumference to nonalcoholic fatty liver disease. Gastroenterology. 2006;130:2023–2030. doi: 10.1053/j.gastro.2006.03.019. [DOI] [PubMed] [Google Scholar]
- 4.Kim H, Lee K, Lee KW, Yi NJ, Lee HW, Hong G, Choi Y, et al. Histologically proven non-alcoholic fatty liver disease and clinically related factors in recipients after liver transplantation. Clin Transplant. 2014;28:521–529. doi: 10.1111/ctr.12343. [DOI] [PubMed] [Google Scholar]
- 5.Hong Z, Yanfang J, Shumei H, Jie S, Qing G, Xiangwei F, Bing D, et al. Relationship between serum aminotransferase levels and metabolic disorders in northern China. Turk J Gastroenterol. 2012;23:699–707. doi: 10.4318/tjg.2012.0589. [DOI] [PubMed] [Google Scholar]