Abstract
The association between the ratio of uric acid relative to high-density lipoprotein cholesterol (UHR) and abdominal aortic calcification (AAC), and the mediating effect of diabetes, is not fully understood. The ultimate assessment encompassed 2,731 participants (average age: 58.64 years; 51.59% women) from the 2013-2014 National Health and Nutrition Examination Survey. The presence of abdominal aortic calcification (AAC) was evaluated employing Kauppila's (1997) semi-quantitative scoring system, with the results expressed in Kauppila semi-quantitative units. In the analysis, AAC scores were treated as continuous variables. AAC was also categorised as a binary variable, with non-zero scores assigned a value of 1 and zero scores assigned a value of 0. Similarly, SAAC was categorised as a binary variable, with scores greater than to 6 assigned a value of 1 and scores less than or equal to 6 assigned a value of 0. The present study employed weighted multiple logistic regression and linear regression analysis to investigate the association between UHR and AAC scores, as well as between AAC and SAAC. Subgroup and interaction analyses were conducted in order to investigate whether these associations varied by different confounders. The present study employed causal mediation analysis to assess the mediating effect of diabetes between UHR and abdominal aortic calcification (AAC). Sensitivity analysis was performed to determine the robustness of the association results between UHR and AAC. A one-unit rise in the log2-transformed UHR led to a 0.53 increase in the AAC scores [β(95% confidence interval, CI): 0.53 (0.31, 0.75)] and a 43% higher risk of AAC [odds ratio (OR) (95% CI): 1.43 (1.22, 1.67)], and the risk of SAAC increased by 60% [OR (95% CI): 1.60 (1.21, 2.12)]. The findings of this study indicate that diabetes mediated 7.5% of the association between UHR and AAC scores, and 14% of the association between UHR and SAAC risk. This study found significant positive correlations between UHR and AAC scores, and between UHR and the risk of AAC and SAAC. It also found that diabetes partially mediated the association between UHR and AAC scores, as well as between UHR and SAAC. These results imply that UHR could be a useful clinical biomarker for predicting AAC risk and identifying AAC and SAAC, while diabetes partly explains this association.