Background

A person’s fat accumulation pattern (visceral-, subcutaneous-, and/or liver fat) can determine their cardiometabolic risk profile. This study aims to investigate risk stratification using personalized body fat z-scores in persons with BMI 30-40 kg/m2 from the UK Biobank imaging study.

Methods

Magnetic resonance images of 40174 participants (6-minute neck-to-knee protocol) were analyzed for visceral adipose tissue (VAT), abdominal subcutaneous adipose tissue (aSAT), and liver fat (LF) using AMRA® Researcher. Body fat z-scores (VATz, aSATz, LFz) were calculated for each participant using at least 150 sex- and BMI-matched controls. Associations between z-scores and later incident cardiovascular disease (CVD) and type 2 diabetes (T2D) were investigated using Cox proportional-hazards modelling and Kaplan-Meier curves.

Results

BMI was 25-30/30-40/>40 kg/m2 in n=16361/6716/412 respectively. VATz was positively associated with CVD (crude hazard ratio (cHR) [95% CI]: 1.30 [1.20-1.40], p<0.001), while aSATz and LFz were negatively associated (cHR: 0.91 [0.85-0.99], p=0.028 and 0.88 [0.82-0.95], p=0.002). All z-scores were significant when models were adjusted for sex, age, BMI, but only VATz when also adding previous CVD. VATz and LFz were positively associated with T2D (cHR: 1.53 [1.40-1.67], p<0.001 and 1.35 [1.23-148], p<0.001) while aSATz was negatively associated (cHR: 0.90 [0.81-0.99], p=0.026). All z-scores were significant when models were adjusted for sex, age, BMI, but not when also adding previous T2D.

Conclusions

Personalized body fat z-scores identifies obesity sub-phenotypes with specific cardiometabolic risk profiles. Individuals with a high risk for CVD or T2D may be recommended intensive obesity treatment such as metabolic surgery.