Journal of Diabetes Investigation (Dec 2024)

Metabolic dysfunction‐associated steatotic liver disease: A superior predictor for incident type 2 diabetes over traditional criteria – NAGALA study

  • Kimiko Sakai,
  • Takuro Okamura,
  • Ema Toyokuni,
  • Hiroshi Okada,
  • Akihiro Obora,
  • Takao Kojima,
  • Masahide Hamaguchi,
  • Michiaki Fukui

DOI
https://doi.org/10.1111/jdi.14315
Journal volume & issue
Vol. 15, no. 12
pp. 1788 – 1796

Abstract

Read online

ABSTRACT Aims/Introduction The 2023 Delphi consensus recommended the use of new term, metabolic dysfunction‐associated steatotic liver disease (MASLD), aiming conceptual shift from the conventional non‐alcoholic fatty liver disease (NAFLD). The association between NAFLD and type 2 diabetes mellitus (T2DM) development is well known. This study aimed to examine the correlation between MASLD and T2DM development, comparing their utility as predictors. Materials and Methods This retrospective cohort study obtained data from a medical health checkup program conducted at Asahi University Hospital, Japan, between 2004 and 2021. Logistic regression analysis was used to assess the association between MASLD and incident T2DM over 5 years. To compare the predictive utility of NAFLD and MASLD, receiver operating characteristic curves were drawn, followed by area under the curve (AUC) comparisons. Results In total, 15,039 participants (59.6% males; median [interquartile range {IQR}] age, 44 [38, 50] years) were included. Out of 2,682 participants meeting the criteria for MASLD, 234 individuals (8.7%) developed T2DM. Multivariate analysis revealed a significantly elevated risk of T2DM in MASLD compared with the reference healthy group (without steatotic liver disease or cardiometabolic risk), presenting an OR of 127.00 (95% CI 40.40–399.00, P < 0.001). The concordance rate of diagnosis between NAFLD and MASLD was 98.7%. The AUC values were 0.799 for NAFLD and 0.807 for MASLD, respectively. Comparative analysis of the AUC showed a statistical difference between NAFLD and MASLD (P < 0.001). Conclusions MASLD was shown to be a significant risk factor for incident T2DM, exhibiting a potentially higher predictive capacity than conventional NAFLD.

Keywords