Sarcopenic obesity predicts negative health outcomes among older patients with type 2 diabetes: The Ageing and Body Composition of Diabetes (ABCD) cohort study.

Background And Aims: The definition of and diagnostic criteria for sarcopenic obesity (SO) remain unclear, hindering the assessment of its prevalence as well as its clinical relevance to negative health outcomes, especially in diabetic patients, who are more prone to body composition changes. The aim of this study was to investigate the prevalence of SO and its impact on negative health outcomes among elderly patients with type 2 diabetes (T2DM) from the Ageing and Body Composition of Diabetes (ABCD) cohort.

Methods: This retrospective cohort study included 386 elderly patients with T2DM (177 males and 209 females, mean age of 67.91 ± 6.10 years). SO was defined as the coexistence of sarcopenia defined by the 2019 Asian Working Group for Sarcopenia up-to-date consensus and obesity identified by five alternative measurements as follows: body mass index (BMI)≥28 kg/m (BMI28), BMI≥25 kg/m (BMI25), body fat percentage (BF%)≥25% for men or 35% for women, visceral fat area (VFA)≥100 cm, or android fat mass (AF) higher than the sex-specific median. The primary endpoint was all-cause death or fragility fracture, and the secondary endpoint was a composite of cardiovascular diseases (CVDs). Cox proportional hazards regression analysis was used to estimate the association between SO and negative health outcomes.

Results: The prevalence of SO was 0.2% (BMI28), 2.5% (BMI25), 9.8% (AF), and 18.7% (BF% or VFA) among elderly patients with T2DM, according to the different obesity surrogate markers. During a mean follow-up period of 3.46 ± 1.15 years, 50 patients reached the primary endpoint, and 33 patients had incident CVD. SO classified using BF% was significantly associated with the primary endpoint [hazard ratio (HR) = 2.94, 95% CI = 1.25-6.92] and incident CVD (HR = 6.02, 95% CI = 1.56-23.15), even after comprehensive adjustment for bone-, comorbidity-, and diabetes-specific confounding variables. When SO was classified using BMI25, VFA and AF, similar results were found for adverse outcomes. However, SO classified using BMI25 resulted in misclassification of SO for 61 participants, 19 of whom experienced adverse events during follow-up, and SO classified using VFA or AF was not significantly associated with incident CVD.

Conclusions: SO is not uncommon in geriatric patients with T2DM, and its prevalence varies widely depending on the diverse surrogate indices of body fat excess. Furthermore, SO may be a better independent risk factor for negative health outcomes when classified using BF%.