Abstract
BACKGROUND: Cardiovascular disease (CVD) is a leading cause of mortality among individuals with type 2 diabetes mellitus (T2DM), with visceral adiposity being a key contributor to increased CVD risk. Novel visceral obesity indices (NVOI), including the lipid accumulation product (LAP), visceral adiposity index (VAI), and metabolic score of visceral fat (METS-VF), offer improved visceral adipose tissue assessment and may enhance CVD risk prediction. This study aimed to evaluate the association of these indices with 10-year CVD risk and their predictive performance in adults with T2DM. METHODS: A cross-sectional study was conducted in the diabetes outpatient clinic and family medicine units of Suez Canal University in Ismailia, Egypt over 15 months starting in February 2023. A total of 397 randomly selected patients with T2DM participated. A structured interview questionnaire was used to collect demographics, medical, family, and lifestyle-related data. Clinical data such as blood pressure, body mass index (BMI), waist circumference (WC), and laboratory data such as fasting blood glucose (FBG) and lipid profile were obtained. NVOIs were calculated using standardized equations, and 10-year CVD risk was determined using the 2019 WHO/ISH CVD risk-laboratory-based chart. Logistic regression was used to assess the associations between NVOIs and high CVD risk, while receiver operating characteristic (ROC) curve analysis was used to evaluate its predictive accuracy. RESULTS: High CVD risk (≥ 20% 10-year risk) was identified in 40.5% of participants and was significantly associated with higher LAP, VAI, and METS-VF levels (p < 0.001). VAI was associated with 3.18 times higher odds of having a high 10-year CVD risk (95% CI 1.61-6.26, p < 0.001) in males and 4.16 (95% CI 1.26-13.68, p = 0.019) in females. METS-VF had the highest predictive ability, with an adjusted odds ratio (aOR) of 7.39 (95% CI 1.03-52.85, p = 0.046) in males and 7.80 (95% CI 1.53-39.92, p = 0.014) in females while, LAP showed no significant association. The area under the curve (AUC) values indicated acceptable to excellent predictive accuracy for all indices, with METS-VF and VAI generally outperforming LAP. VAI performs best in males and METS-VF in females. Sensitivity ranged from 63.92 to 87.5%, while specificity varied between 73.79% and 94.51%. Positive predictive values (PPVs) were higher in males (77-92.5%), whereas negative predictive values (NPVs) were higher in females (88.9-93%). CONCLUSIONS: High CVD risk was significantly associated with elevated VAI, METS-VF, and LAP; however, only VAI and METS-VF emerged as independent predictors. These indices demonstrated the highest predictive accuracy, reinforcing their clinical relevance. Given their superior discriminative ability, incorporating VAI and METS-VF into routine assessments could enhance CVD risk prediction in adults with T2DM, allowing for earlier intervention and better management strategies.