Type 2 diabetes, sex and metabolic dysfunction-associated steatotic liver disease

Table 1 Epidemiological evidence that the association between steatotic liver disease and risk of dysglycemia and type 2 diabetes is modulated by sex

Method	Findings	Conclusion	Risk of bias	Ref.
This retrospective cohort study included 13596 men and 6037 women to examine the association between SLD, sex, and T2D in non-obese Japanese adults undergoing health check-ups. Baseline BMI was grouped into six categories, and SLD was diagnosed by abdominal ultrasonography	During the follow-up, 738 men and 138 women developed T2D. Men with a BMI of 20-22.9 kg/m2 and SLD had a higher risk, while women in this group did not. Among men, higher BMI without SLD did not raise the risk of diabetes, but women with a BMI of 23-27.4 kg/m2 had an increased risk regardless of SLD status	Sex influences the relationship between liver biomarkers and the development of T2D. SLD is associated with an increased risk of T2D in non-obese Asian males, but not in females	Low	Narisada et al[59]
This analysis used cross-sectional data from 1554 participants in the Maastricht study. T2D was determined by OGTT, and IHL content was measured via 3T Dixon MRI. Mediation analysis evaluated whether serum SHBG mediates the link between IHL content and T2D	IHL content was linked to T2D in both women (OR = 1.08, 95%CI: 1.04-1.14) and men (OR = 1.12, 95%CI: 1.08-1.17), and serum SHBG significantly mediated this association. Serum SHBG contributed more in women (OR = 1.04, 95%CI: 1.02-1.07; 50.9% mediated, 95%CI: 26.7-81.3) than men (OR = 1.02, 95%CI: 1.01-1.03; 17.2% mediated, 95%CI: 9.6-27.6). Repeating analyses with T2D proxies and covariate adjustment yielded similar results	Serum SHBG mediated the association between IHL content and T2D in both sexes, but its contribution was stronger in women than in men	Low	Simons et al[37]
Of 1309 screened individuals, 748 without diabetes were enrolled. Over 15 years, correlations between metabolic markers (including APN) and T2D onset were assessed in this group	In multivariate LRA, the independent predictors of incident T2D differed by sex: In men, only eGFR and SLD were significantly related to the onset of T2D. In women, APN was the only significant risk factor	Intervention strategies to prevent T2D vary between men and women	Low	Yoshimoto et al[60]
PPG was determined as the AUC for glucose during OGTT using the trapezoidal rule among 794 middle-aged Soweto cohort participants. PC analysis clustered sex hormones, liver enzymes, and cardiometabolic factors by sex. Multivariable linear regression assessed how much variance in PPG was explained by PCs and T2D PRSs, adjusting for relevant covariates in men and women	In men, the PCs’ cluster of sex hormones, liver enzymes, and cardiometabolic factors explained 10.6% of the variance in PPG, with PC1 (peripheral fat), PC2 (liver enzymes and steroid hormones), and PC3 (lipids and peripheral fat) contributing significantly to PPG. In women, PC factors of sex hormones, cardiometabolic factors, and liver enzymes explained a similar amount of the variance in PPG (10.8%), with PC1 (central fat) and PC2 (lipids and liver enzymes) contributing significantly to PPG	Variations in PPG responses to an OGTT among individuals may be influenced by factors such as body fat distribution, serum lipid profiles, liver enzyme levels, and steroid hormone concentrations, with differing effects observed between men and women	Low	Masango et al[61]
The PREVADIAB2 study assessed metabolic changes over 5 years in people initially without T2D. Researchers used hierarchical clustering on 953 participants based on IGI, fasting insulin secretion rate, HOMA-IR, and insulin clearance; 488 had their lipid profiles analyzed by LC/MS-QTOF	Four clusters (LS, PGS, ID, IR) with distinct dysglycemia risk were found. Although metabolic features were similar for both sexes, threshold differences led to sex-specific lipid profiles. Women had higher dihydroceramide and SM levels, while men showed a greater ceramide-to-SM ratio. All clusters except LS showed unique lipid signatures linked to metabolic dysfunction, with clear sex-based differences	Distinct insulin-associated metabolic characteristics and sex delineate phenotypes with unique lipidomic profiles, highlighting the importance of considering prediabetes within a comprehensive metabolic framework that extends beyond glycaemic measures	Low	Pina et al[62]

T2D: Type 2 diabetes; SLD: Steatotic liver disease; OGTT: Oral glucose tolerance test; BMI: Body mass index; IHL: Intrahepatic lipids; MRI: Magnetic resonance imaging; SHBG: Sex hormone-binding globulin; APN: Serum adiponectin; AUC: Area under the curve; PPG: Postprandial glucose; PC: Principal component; PRSs: Polygenic risk scores; IGI: Insulinogenic index; HOMA-IR: Homeostasis model assessment of insulin resistance; LC: Liquid chromatography; QTOF: Quadrupole time-of-flight; MS: Mass spectrometry; OR: Odds ratio; CI: Confidence interval; LRA: Logistic regression analysis; eGFR: Estimated glomerular filtration rate; LS: Liver sensitive; PGS: Pancreas glucose sensitive; ID: Insulin deficient; IR: Insulin resistant; SM: Sphingomyelin(s).