RESUMO
CONTENT: The correlation between visceral obesity index (VAI) and diabetes and accuracy of early prediction of diabetes are still controversial. OBJECTIVE: This study aims to review the relationship between high level of VAI and diabetes, and early predictive value of diabetes. DATA SOURCES: The databases of PubMed, Cochrane, Embase, and Web of Science were searched until October 17, 2023. STUDY SELECTION: After adjusting for confounding factors, the original study on the association between VAI and diabetes was analyzed. DATA EXTRACTION: We extracted odds ratio (OR) between VAI and diabetes management after controlling for mixed factors, and the sensitivity, specificity and diagnostic four grid table for early prediction of diabetes. DATA SYNTHESIS: 53 studies, comprising 595,946 participants were included. The findings of the meta-analysis elucidated that in cohort studies, a high VAI significantly increased the risk of diabetes mellitus in males (OR = 2.83 (95% CI: 2.30-3.49)) and females (OR = 3.32 (95% CI: 2.48-4.45)). The ROC, sensitivity, and specificity of VAI for early prediction of diabetes in males were 0.64 (95% CI: 0.62-0.66), 0.57 (95% CI: 0.53-0.61), and 0.65 (95% CI: 0.61-0.69), respectively, and 0.67 (95% CI: 0.65-0.69), 0.66 (95% CI: 0.60-0.71), and 0.61 (95% CI: 0.57-0.66) in females, respectively. CONCLUSIONS: VAI is an independent predictor of the risk of diabetes, yet its predictive accuracy remains limited. In future studies, determine whether VAI can be utilized in conjunction with other related indicators to early predict the risk of diabetes, to enhance the accuracy of prediction of the risk of diabetes.