AI Summary & Key Takeaways
AI InsightThese findings support the association of AIP with cardiometabolic characteristics while emphasizing the need for cautious interpretation of composite-index relationships.
📌 3 Key Takeaways
Abstract
Aims: The Atherogenic Index of Plasma (AIP) is a useful marker of atherogenic dyslipidemia and is associated with cardiometabolic risk. However, the relative associations of metabolic and inflammatory factors with AIP remain incompletely characterized. This study investigated the associations between AIP and metabolic and inflammatory biomarkers and evaluated factors independently associated with AIP in adults. Methods: A total of 239 adults with complete clinical, biochemical, and anthropometric data were included. AIP was calculated as log10 (triglycerides/HDL-C), and the Triglyceride-Glucose Index (TyG) was used as a surrogate marker of insulin resistance. C-reactive protein (CRP) and ferritin were log transformed. Pearson and triglyceride-controlled partial correlation analyses were performed, followed by multivariable linear regression and a sensitivity analysis replacing TyG with fasting glucose. Results: AIP showed a very strong positive correlation with the TyG index (r=0.825, p<0.001), as well as moderate correlations with body-mass index (BMI) (r=0.427, p<0.001), CRP_log (r=0.239, p<0.001), and ferritin_log (r=0.326, p<0.001). The association between AIP and TyG was substantially attenuated after controlling for triglyceride concentration (partial r=0.380, p<0.001). Similarly, the unadjusted association between AIP and fasting glucose (r=0.189, p=0.003) was no longer statistically significant after triglyceride adjustment (partial r=0.102, p=0.118). In the primary multivariable regression model, TyG showed the largest standardized coefficient (β=0.754, p<0.001), while BMI (β=0.127, p=0.005) and gender (β=-0.162, p<0.001) were also independently associated with AIP. In the sensitivity model replacing TyG with fasting glucose, BMI (β=0.341, p<0.001) and gender (β=-0.294, p<0.001) remained significantly associated with AIP, whereas fasting glucose, CRP_log, and Ferritin_log were not. Conclusion: The strong unadjusted association between AIP and TyG was substantially attenuated after controlling for triglyceride concentration, highlighting the importance of mathematical coupling when interpreting associations between these composite indices. BMI remained independently associated with AIP across alternative model specifications, whereas fasting glucose and inflammatory markers did not show independent statistical associations after multivariable adjustment. These findings support the association of AIP with cardiometabolic characteristics while emphasizing the need for cautious interpretation of composite-index relationships.