TOPLINE

Both traditional obesity measures (BMI, waist circumference [WC], and waist-to-height ratio [WHtR]) and newer measures of adipose tissue moderately predicted type 2 diabetes (T2D) development over 20 years in healthy adults. Body surface area was the strongest predictor even after accounting for lifestyle and health factors; each one-unit increase was linked to more than five times the diabetes risk.

METHODOLOGY

• Researchers analyzed data from 2000 participants (51% women; mean age 43 years) free of atherosclerotic cardiovascular disease and T2D at baseline in the ATTICA cohort study conducted in Greece from 2002 to 2022.
• A total of 22 adiposity indices were measured or calculated at baseline, categorized as total adiposity (BMI, body surface area of Dubois [BSAD], body surface area of Mosteller [BSAM]), central adiposity (WC, WHtR, waist-to-hip ratio [WHR], body adiposity index, body roundness index, and others), or dysfunctional adipose tissue markers (visceral adiposity index, lipid accumulation product [LAP], dysfunctional adiposity index, cardiometabolic index, waist-circumference-triglyceride index, and triglyceride-glucose [TyG]-based indices).
• The primary outcome was 20-year cumulative incidence of T2D, defined as fasting plasma glucose > 125 mg/dL or use of antidiabetic medications.
• Multivariable logistic regression models adjusted for fasting glucose levels, age, sex, smoking status, Mediterranean diet adherence, physical activity, family history of T2D, hypercholesterolemia, hypertension, and changes in BMI or waist/hip circumferences during the first 10 years of follow-up.

TAKEAWAY

• The 20-year cumulative incidence of T2D was 26.3%, with higher rates in men (31.4%) than women (21.4%).
• After full adjustment, BSAD and BSAM demonstrated the strongest associations with T2D incidence (adjusted odds ratio [aOR], 5.52 and 5.90, respectively, per unit increment).
• Among dysfunctional adiposity indices, TyG × WHtR showed strong association with T2D (aOR, 2.28 per unit increment), while LAP showed no significant association in adjusted models.
• Almost all indices exhibited moderate discriminative accuracy (C-statistic 0.70-0.80) for predicting T2D. When tested against a core model of fasting glucose, age, sex, and family history of T2D, obesity-related anthropometric indices added no further predictive value, while dysfunctional adipose tissue indices added up to 4.7%.

IN PRACTICE

"These findings underscore the importance of incorporating adiposity measurements into prevention strategies, focusing on the early identification and management of excess body weight to reduce the burden of T2D," wrote the authors of the study.

SOURCE

The study was led by Ioanna Kechagia, Harokopio University, Athens, Greece. It was published online on July 24 in Clinical Obesity.

LIMITATIONS

The observational design may have introduced residual confounding despite adjustments for multiple covariates. Exact diabetes onset dates were unavailable, precluding time-to-event analyses. Adiposity indices were calculated only at baseline, limiting insights into longitudinal changes. Body composition analyzers were not used to distinguish between sarcopenic obesity and metabolically unhealthy obesity phenotypes. Approximately 25% of participants were lost to follow-up, raising the possibility of selection bias.

DISCLOSURES

The ATTICA study was funded by the Hellenic Cardiological Society and Hellenic Atherosclerosis Society. The authors reported no relevant conflicts of interest.

This article was created using several editorial tools, including AI, as part of the process. Human editors reviewed this content before publication.