Association between visceral adipose tissue measured by deep neural network architecture and chronic kidney disease

The relationship between abdominal body composition and chronic kidney disease (CKD) is well-documented. In this study, we aimed to investigate the association between CKD and abdominal fat volume assessed using deep neural network architecture. This study used the health check-up data of 14,105 patients with available computed tomography (CT) images of the abdomen in a Korean population. The volumes of body segments, including visceral adipose tissue (VAT) and subcutaneous adipose tissue (SAT), were measured using an artificial intelligence (AI)-based image analysis software. Of the 14,105 participants, the prevalence of CKD was 2.3% in males and 1.3% in females. In the multivariable analysis, the volumes of VAT were significantly associated with an increased risk of CKD in both males (odds ratio [OR], 1.25, 95% confidence interval [CI], 1.18-1.33, P < 0.001) and females (OR, 1.18, 95% CI, 1.10-1.27, P < 0.001). The volumes of SAT were significantly associated with a decreased risk of CKD in females (OR, 0.77, 95% CI, 0.71-0.83, P < 0.001) and an increased risk of CKD in males (OR, 1.09, 95% CI, 1.03-1.16, P < 0.001). The VAT/SAT volume ratio was significantly associated with an increased risk of CKD in males (OR, 1.09; 95% CI, 1.04-1.14, P < 0.001). Abdominal fat volumes were independently associated with CKD risk, showing positive associations of VAT in both sexes and inverse associations of SAT in women. These findings highlight the importance of considering sex-specific abdominal fat distribution in CKD risk evaluation. The use of AI-based analysis of abdominal CT images may help improve early detection and risk stratification of CKD.