Objective: We aimed to evaluate the risk of incident type 2 diabetes (T2D) in metabolically healthy, obese (MHO) individuals and the role of fatty liver disease (FLD) in this potentially causal relationship.

Methods: We enrolled 514,866 participants from the Korean National Health Insurance Service-National Health Screening Cohort in 2002-2015. Subjects were categorized into metabolically healthy, non-obese (MHNO), metabolically unhealthy, non-obese (MUNO), MHO, and metabolically unhealthy, obese (MUO) groups. Well-validated FLD indices were used in this study; Fatty Liver Index (FLI), Hepatic Steatosis Index (HSI), and Visceral Adiposity Index (VAI).

Results: Over 5.6 years, 67,807 (27.8%) of 244,301 subjects free of T2D and liver disease besides FLD developed T2D. The risk correlated with obesity (HR, 1.35; 95% CI 1.33-1.37) and the number of metabolic risk factors; HRs for 1, 2, 3, and 4 risk factors were 1.43 (95% CI, 1.39-1.47), 1.92 (95% CI, 1.87-1.97), 2.51 (95% CI, 2.44-2.58), and 3.14 (95% CI, 3.02-3.26), respectively. Compared with the MHNO group, the risk of T2D was significantly higher in MUO (multivariate HR, 1.95; 95% CI, 1.91-2.00), MUNO (HR, 1.63; 95% CI, 1.60-1.67), and MHO (HR, 1.28; 95% CI, 1.25-1.30) individuals. Higher FLD scores among MHO individuals also exhibited a strong correlation with the incident T2D, compared with both the MHNO group and the MHO counterpart with lower FLD scores (P < 0.05 for all comparisons). Survival trees analysis based on log-rank test revealed that the best cut-off points to predict incident T2D were 25.1, 33.4, and 1.7 for FLI, HSI, and VAI, respectively.

Conclusion: The number of metabolic risk factors, obesity, and FLD are significant predictors of incident T2D. Higher T2D risk in MHO individuals, with an excess risk related to higher FLD scores, necessitates more intensive monitoring and interventions in these individuals.


Y. Kang: None. Y. Cho: None. J. Yoo: None. J. Lee: None. J. Choi: None. J. Park: None. W. Lee: None. C. Jung: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.