Association of HIV-infection, antiretroviral treatment and metabolic syndrome with large artery stiffness: a cross-sectional study

Effective use of combined antiretroviral therapy (cART) has improved life expectancy among people living with HIV-1infection [1]. Despite immunological modulation, aging with HIV-1 infection is linked with an increased prevalence of metabolic syndrome (MS), both in the absence and presence of cART, which can accelerate the development of vascular abnormalities, particularly endothelial dysfunction [2] and increased carotid-artery media thickness [3]. In fact, large elastic artery stiffness, a marker of heightened cardiovascular risk and an independent predictor of adverse vascular events [4], is elevated in HIV-1-infected patients in North America and Europe [5] and appears to be further increased the presence of cART [6, 7]. We have previously shown that first-line cART with non-nucleoside and nucleoside reverse transcriptase inhibitors and time since HIV-1 diagnosis were independent predictors of increased large elastic artery stiffness in sub-Saharan African HIV-1-infected patients [8], largely free of clustered cardiometabolic risk factors that comprise MS. Data are limited and findings have been inconsistent in sub-Saharan African HIV-1-infected patients [9, 10], even though antiretroviral treatment and arterial stiffening could contribute to accelerated atherosclerotic disease. It is possible that the inconsistencies arise in part because HIV-1 infection and cART are both associated with a spectrum of metabolic complications such as dyslipidemia, insulin resistance, changes in body fat distribution, and increased blood pressure [9], which can impact arterial stiffening [11]. Despite sub-Saharan Africa having the highest rates of HIV-1 infection [10], MS in relation to arterial stiffness in HIV-1-infected patients is poorly explored.

The present study aimed to determine: 1) if MS is associated with increased large elastic artery stiffness in HIV-1-infected patients; and 2) if HIV-1 treatment with combination non-nucleoside and nucleoside reverse transcriptase inhibitors is associated with increased arterial stiffness compared to untreated patients.

Participant characteristics according to sub-groups are shown in Table 1. The point prevalence of MS in the HIV-1 patients was 28%. Subjects without MS were generally well matched, however, HIV-1-cART patients had lower body mass and BMI compared with corresponding group controls. As expected, body mass, BMI, percent body fat, and waist circumference were higher (all P < 0.05) in subjects with MS compared to corresponding between-group subjects without MS. There were no significant differences in indices of adiposity, blood pressure, or plasma lipids, lipoproteins, and glucose among the controls, HIV-1 untreated patients, and HIV-1-cART patients with MS.

No differences were seen in the most recent CD4+ T cell counts between untreated and cART HIV-1 patients regardless of MS. There were no significant differences in the duration of HIV-1 infection between untreated and cART HIV-1 patients with MS compared to those without. There were no differences in the median duration of cART between the HIV-1 patients with MS (24 months; IQR: 15.9 to 42.3 months, P = 0.07) compared with HIV-1 patients without MS (30.9 months; IQR: 23.2 to 66.4 months, Table 1).

Hemodynamic results from the time of pulse wave analysis are displayed in Table 2. Peripheral systolic and mean arterial pressures and pulse rates were higher, and peripheral diastolic pressure lower in comparison to central values in both the MS group and in patients without MS (P < 0.001 for all). No differences were seen among the groups in central systolic, diastolic, mean arterial, or pulse pressures. Similarly, corresponding peripheral blood pressures were not different among the groups.

The main findings of the present study were that MS and other risk variables (i.e., blood pressure, duration of cART, and age) are associated with increased large elastic artery stiffness in in sub-Saharan African HIV-1-infected patients. Importantly, the degree of arterial stiffness in HIV-1 patients with MS is worse in the presence of combination non-nucleoside and nucleoside reverse transcriptase inhibitors. This suggests that perhaps treated HIV-1 infection coupled with clustering of risk factors of MS may increase aPWV and hence arterial stiffness. Increased arterial stiffness may represent an important link contributing to atherosclerotic vascular disease associated with HIV-1 infection and cART. It would benefit practitioners, and be clinically advantageous to treated HIV-1 infected individuals with MS to discriminate between those with and without increased arterial stiffness, in sub-Saharan Africa HIV-1 patients.

Our findings showed that aPWV was significantly elevated in individuals with MS compared to their counterparts without MS. Other studies also reported elevated aPWV in individuals with MS compared to those without [14, 15]. Further analysis demonstrated that HIV-1 cART patients compared to their cART-naive counterparts had significantly higher aPWV. These findings corroborate similar studies [7, 8, 16]. HIV-1 cART with MS compared to HIV-1 cART-naive and controls with MS had a significant elevated aPWV even after inclusion of mean arterial pressure as a covariate and after adjustment for sex. Furthermore, a significant difference was observed in arterial stiffness when HIV-1 cART patients with MS were compared to HIV-1 cART patients without MS. However, no significant elevation in aPWV was noted between untreated HIV-1 patients with MS and their counterparts without MS. Similar findings have been reported in a study done in Italy [17], though they did not find significant differences between HIV-1 cART and HIV-1 cART-naive patients with MS. The fact that there was no difference in aPWV in untreated HIV-1 patients with MS compared to their untreated counterparts without MS would imply that cART together with MS impacts aPWV. It is argued that systemic inflammation of any cause could initiate the process of endothelial damage which is characterized by abnormal orientation of endothelial cells in the direction of blood flow in contact with the vessel wall [18]. According to Monsuez et al. [19] the molecular mechanisms by which HIV-1 induces endothelial dysfunction are multifactorial with several theories linking them with combinations of high viral load, elevated inflammatory markers and adhesion molecules, pro-atherogenic lipid profile and the effects of cART. Contrary to our findings, a study in sub-Saharan Africa found higher prevalence of MS and increased arterial stiffness in cART-naive HIV-infected individuals compared to healthy individuals in Cameroon [20]. However, the study did not involve either HIV-infected individual on cART or stratify the groups according to MS status. Also, a review of arterial stiffness and HIV infection in adult Africans concluded that arterial stiffness is elevated in HIV-infected individuals regardless of antiretroviral treatment status [21].

Upon stepwise multivariate analysis, independent predictors of aPWV were found to be presence of MS, age, duration of cART and mean arterial pressure. Further subgroup analysis revealed that in the HIV-1 cART group, MS remained the only predictor of aPWV while in the HIV-1 cART-naive group only age predicted aPWV. This could imply that treated HIV-1 infection coupled with the presence of MS is associated with increased aPWV and hence arterial stiffness. Similar to our study findings, the study by Maloberti et al. [17] demonstrated that pulse wave velocity was elevated in HIV patients on treatment with MS than in their counterparts without MS indicative of independent effect of MS on pulse wave velocity. Also, consistent to our study findings, Rider et al. [16] found that MS had an additive effect to HIV-1 on elevated aPWV. However, they concluded that, independent of MS, patients with treated HIV-1 have reduced vascular function and that the magnitude of the effect of treated HIV infection is the same as that of the MS. One possible explanation of the effect of cART is on lipid metabolism, endothelial and adipocyte cell function and mitochondrial dysfunction [22]. Also, cART may activate pro-inflammatory cytokines which are a result of infection and inflammation. These abnormalities are believed to lead to antiretroviral-associated metabolic syndrome [23]. Different organs and organ systems are affected by systemic complications as a result of MS. One characteristic of MS is that it is associated with alterations of the arterial vasculature, especially of the endothelium, and basal membrane and also polymorphonuclear activation, increased oxidative stress, and changes in the expression of matrixmetalloproteinases (MMPs)– family of enzymes essential for degradation of extracellular matrix during the embryonic development, morphogenesis and tissue remodeling [24, 25]. Medley et al. [26] point out that, MMP-9 in particular, could be associated with large artery stiffening and hence vascular disease. It is suggested that, the extracellular matrix of the vessel walls that are made up of collagen and elastin [27] are closely related to structural strength and elasticity of the blood vessels and are regulated by catabolic MMPs. MMPs degrade the extracellular matrix by affecting the production of weaker collagen and worn elastin fibers leading to arterial stiffening [28]. On the other hand, vascular stiffening may be a result of a complex interplay between various independent and inter-dependent factors which include age, the hormonal situation, the individual’s glycemic state and salt intake, as well as the global decline in cellular systems and function [27].

Contrary to aPWV, we did not observe differences in AIx@HR75 between HIV-1 cART patients with MS compared to their cART counterparts without MS. Several studies have shown a dissociation of aortic pulse wave velocity and augmentation index in patients with MS and healthy individuals. For example, Vlachopoulos et al. [29] found that circulating MMPs in healthy persons were inversely associated with large artery stiffness determined by aPWV, but not with wave reflections, which are measured by AIx. In their earlier randomized, double blind, sham procedure-controlled study in healthy subjects, Vlachopoulos et al. [30] found that acute systemic inflammation leads to increases large artery stiffness and a decrease in aortic AIx. Similar findings on the dissociation between aPWV and AIx were revealed in a study done by Vágovicová et al. [31] regarding the association of MS with arterial properties. They reported that aPWV was higher in subjects with MS compared with those without. In addition, aPWV was elevated in subjects with a greater number of MS risk factors. In contrast, AIx was lower in subjects with a higher number of MS risk factors, even after adjustment for age gender heart rate and mean arterial pressure. The possible explanation for the dissociation between aPWV and AIx as indicators of arterial stiffness could be that inflammation increases PWV resulting in an early return of wave reflections to central arteries. At the same time, it induces vasodilation and thus decreases the magnitude of reflected waves culminating in a neutral or decreased net result of wave reflection indices. On the other hand, Cheng and Holewijn et al. [32, 33] pointed out the dissociation could be due to the discrepancy in the AIx formula; arguing that, the AIx formula has other factors apart from central second peak and augmentation pressure and hence cannot be solely attributed to changes in central second peak and augmentation pressure. However, this needs further validation.

In this study we found the point prevalence of MS among HIV-1-infected patients was 28%. This is consistent with prevalence data reported in other studies in Tanzania [34, 35] and similar to studies in the developed world [36‐38]. However, great variations in prevalence of MS have been reported in other sub-Sahara African countries ranging from 11% in Botswana [39] to 58% in Uganda [40] among self-reported adherent HIV-infected patients using at least two MS components. The differences could be attributed by the sample size and/or difference in MS criteria used. The reported high prevalence of MS in HIV-1 infected in the Ugandan study could be due to involvement of only cART-adherent study population with no substitutions which other studies do not specify.

This study shows that metabolic syndrome (MS) in treated HIV patients further contributes to CVD risk as measured by arterial stiffness. It should make the clinician extra aware of the importance of CVRM (smoking, blood pressure, lipids, and lifestyle) in HIV patients with MS.

During screening our study excluded patients with clinical evidence of cardiometabolic disease, known to stiffen arteries. By excluding these patients we aimed at removing variables that could have directly influenced the outcome variable. The observational nature of the study limits the establishment of the direct causative effect MS on arterial stiffness. Moreover, the small sample size and even smaller sub-groups with/without MS limits the power to detect important associations as well.

In conclusion, in stark contrast to HIV-1 treatment-naive patients with MS, the presence of MS was associated with more severe large elastic arterial stiffening in HIV-1 patients treated with combination non-nucleoside and nucleoside reverse transcriptase inhibitors. In addition, the large artery stiffness was greater in HIV-1 cART patients with MS compared to treated patients free of MS. These findings suggest that while cART may contribute to the development of MS, it may also exaggerate the development of arterial remodeling.