The main finding of this study highlights the association of of anthropometric indicators and single cardiovascular risk factors and Framingham Risk Score in men living with HIV, while this result was not observed in women. Furthermore, the correlation of these indicators with isolated biochemical and hemodynamic variables seems to contribute to its composition.
CI, WHR, WHtR, BSI and BRI stand out as indicators that maintained more robust associations for men. Despite being simple (single measurement) and practical for assessing abdominal fat in adults, showing a substantial relationship with the percentage of body fat43, the WC alone was not associated with cardiovascular risk in this study. On the other hand, this measurement was inserted in formulas to calculate anthropometric indicators with largest association with cardiovascular risk factors.,.
Espírito Santo49 studied BMI, CI, WC, WHR, and WHtR as discriminators of cardiovascular risk, defined by the Framingham Risk Score in ART-naïve PLHIV. Of these, the indicators that were best associated with high cardiovascular risk were WHR and WC, where CI was the indicator with the lowest discriminatory power. Beraldo4 studied WC, HC, thigh circumference, BMI, BAI, WHR, and waist-to-thigh ratio with the aim of analyzing their associations with metabolic syndrome in PLHIV, observing that WC was the indicator that presented the best performance to identify there. This was not found in our study, since the CI was the indicator that best explained the variance in risk. The divergence between study results can be justified by the different cutoff points used for WC.
Oliveira et al.32 found that the poor performance of NC in ROC curve analysis for predicting cardiometabolic risk in women with HIV, compared to men, can be attributed to variations in the distribution of body fat. In the present study, although a cardiovascular risk prediction analysis was not conducted, none anthropometric indicator, including NC, revealed a significant association with cardiovascular risk factors in women. Women tend to have subcutaneous fat, while men have a more centralized distribution of body fat32. These differences, together with factors such as hormonal variations, body composition and ethnic characteristics, can negatively influence the discriminatory capacity of NC in predicting cardiometabolic risk in women32.
A significant difference was also observed between men and women for total cholesterol, HDL-c, LDL-c and triglycerides. Previous studies, such Oh and Hegele50, indicate that the disparity in cholesterol levels between men and women can be attributed to hormonal, genetic and behavioral factors. Furthermore, it has been observed that men have slightly higher levels of total cholesterol compared to women. However, LDL and total cholesterol are less predominant in predicting cardiovascular risk compared to HDL and triglycerides, highlighting the need for a more individualized approach in women for effective heart health management51.
The Framingham Risk Score stratifies the sample by sex due to characteristic differences in risk (for example, high levels of HDL-c in women, as well as the hormonal protective factor before menopause, and a higher prevalence of diabetes in men). Stratification allows the creation of specific risk profiles, taking into account physiological and health differences between the sexes35, and is supported by differences in body composition between the sexes12,13,14.
Our findings reflect the sex-stratified cardiovascular risk in PLHIV. Although women present more metabolic dysfunctions, such as abdominal obesity and dyslipidemia, men exhibited a higher percentage of elevated risk – using Framingham Risk Score for this definition. Furthermore, the justification may be the score calculation itself, where there are differences in the points attributed to risk factors stratified by sex, such as age, treated and untreated SBP and smoking habit, considering differences in the rates of cardiovascular events between the sexes35. Analyzing the characteristics of the sample in the present study, it is possible to observe a higher prevalence for the male sex, with a predominance of advanced age36, this was the main factor that possibly influenced a high score on the Framingham Risk Score in male gender.
Previous research has highlighted metabolic and morphological changes that may be due to HIV infection, including marked dyslipidemia (increased triglycerides and reduced HDL-c) that are strongly related to increased cardiovascular risk3,8,52,53. This evidence demonstrates that these factors may have influenced the results of the present study (Fig. 2).
Among the variables that increased cardiovascular risk, the one with the highest prevalence, regardless of sex, was low HDL. Oh and Hegele50 associate low HDL concentration with immune activation in initial HIV infection. The decrease observed in HDL in patients with HIV, treated by antiretrovirals or not, results from impaired cholesterol efflux from macrophages due to HIV interference in ATP binding, contributing to low HDL concentration in plasma3,50. Inflammation stimulates endothelial lipases and phospholipases A2, reducing availability of plasma HDL; Additionally, hypertriglyceridemia enriches triglycerides in HDL, making it more susceptible to hepatic removal by hepatic lipase3,50.
Other metabolic changes have been associated with prolonged use of ART4, especially in combination with PI and nucleoside analogue reverse transcriptase inhibitors, which inhibit the mitochondrial enzyme DNA polymerase γ, resulting in mitochondrial DNA depletion and respiratory chain dysfunction, leading to reduced energy production in mitochondria. This mitochondrial dysfunction can affect different cell types, causing lipoatrophy in adipocytes and insulin resistance in skeletal muscle4. Therefore, secondary dyslipidemia occurs characterized by increase in plasma triglycerides, total cholesterol and LDL, and reduction in HDL, causing an elevated cardiovascular risk in HIV patients undergoing treatment3,50.
The limitations of this study include an absence of technical measurement error calculation intra and inter-evaluators of anthropometric measurements; however, anthropometric measurement training and a pilot study were done prior to data collection. Another limitation of the study is the impossibility of predicting the cardiovascular outcome due to the type of study and the analyzes carried out that do not allow such inference. We also assume the limitation of having chosen Framingham as a predictor of cardiovascular risk, although it is indicated by Brazilian institutions.
The strengths of study include the address cardiovascular risk and a wide range the anthropometric indicators of AVI, BRI, BSI and FMI in PLHIV, which highlights the urgent need for in-depth research. This gap in the literature highlights the importance of specifically addressing the intersection between these indicators and cardiovascular health in this population, providing crucial insights for clinical practice and preventive interventions.