HIV Infection: An Independent Risk Factor of Peripheral Arterial Disease

Abstract

To the Editors: Highly active antiretroviral therapy (HAART) had significantly reduced the death rate resulted from HIV infection. However, HAART had also increased the risk of coronary heart disease in HIV-infected patients, especially in those on protease inhibitors (PIs).1 Whether HIV infection itself can promote the atherosclerosis is still controversial. Ankle brachial index (ABI) has been validated against lower extremity contrast angiography to determine its sensitivity, specificity, and accuracy as a diagnostic tool for lower extremity peripheral arterial disease,2 and the presence of a low ABI was predictive of total and cardiovascular mortality.3 Pulse wave velocity (PWV) is noninvasive parameter directly proportional to arterial wall stiffness. Our study aimed to determine whether the HIV-infected patients with or without receiving HAART are more likely to develop atherosclerosis in comparison to the general population, using ABI and PWV; and to assess the associated factors of peripheral arterial atherosclerosis. Our study had been approved by Peking Union Medical College Hospital Institutional Review Board. Eighty-two patients with HIV infection were divided into 2 groups: antiretroviral therapy (ART) naïve group comprising 41 antiretroviral-naive patients and HAART-treated group comprising 41 patients on HAART for more than 12 months (34.7 ± 17.1 months). Forty-three healthy people whose HIV screen tests were negative were enrolled as control subjects. Total cholesterol, triglyceride, low-density lipoprotein cholesterol (LDL-C), and high-density lipoprotein cholesterol (HDL-C) were determined by standard methods using Olympus AU5400 chemistry autoanalyzer (Olympus, Tokyo, Japan). CD4+ lymphocyte count and CD8+ lymphocyte count were measured by Epics XL flow cytometry (Beckman Coulter, Ramsey, MN). Plasma HIV RNA level was determined using a branched DNA assay (lower limit of quantification, 50 copies RNA/mL, version 3.0; Bayer Health Care, Leverkusen, Germany). ABI and PWV were determined using a pulse pressure analyzer (model BP-203RPE II; Nihon Colin, Komaki, Japan). The ABI was performed by measuring the systolic blood pressure from both brachial arteries and from both posterior tibial arteries; ABI was then calculated by the device as the ratio of systolic blood pressure in the leg to that in the arm on each side, and the average value was used for analysis. Pulse waves were recorded using sensors placed on both posterior tibial arteries. The time intervals required for the pulse waves to travel from the heart to both posterior tibial arteries were measured, and the distances between the heart and both posterior tibial arteries were estimated from the patient's height. The PWV was calculated by dividing the distance by the time interval. For statistical analysis, between-group differences were compared by 2-sample t tests or Mann-Whitney tests (non-normal distribution) for continuous variables and by χ2 analysis for categorical variables. Correlation was tested with Spearman rank order or Pearson correlation coefficient. Multiple linear regression analysis was used to test for independent associations between ABI and various factors. No significant statistical differences were observed among 3 groups considering age, sex, and family history of coronary heart disease or current smoking (Table 1). The ART-naive patients were more likely to be hypertensive and had lower level of total cholesterol and LDL-C than control subjects (P < 0.01). A lower HDL-C level was also noted in our ART-naive patients (P < 0.01). ART naïve HIV-infected patients have a lower ABI (P < 0.001) and a higher PWV (P = 0.010) when compared with control subjects. We included 41 ART-naive patients and 43 control subjects in a single group for linear regression analysis (R2 = 0.362; P = 0.001). We found that factors associated with reduced ABI were age (B = 0.03; 95% confidence interval: 0.000-0.005; P = 0.039) and HIV infection (B = −0.069; 95% confidence interval: −0.120 to −0.017; P = 0.01). That meant, after adjustment for other cardiovascular risk factors, HIV infection group was associated with a 0.069 lower ABI compared with control subjects. However, Pearson relation analysis shows not any association between ABI and common HIV infection parameters [CD4+ lymphocyte count (r = −0.161; P = 0.314), CD8+ lymphocyte count (r = −0.026; P = 0.871), CD4+/CD8+ ratio (r = −0.055; P = 0.731), and HIV RNA (r = −0.047; P = 0.770)] in patients with HIV. Patients in HAART-treated group mostly were treated with nonnucleoside reverse transcriptase inhibitors (NNRTIs, 40 of 41, 98%) and nucleoside reverse transcriptase inhibitors (NRTIs, 41 of 41, 100%). Only 7.3% (3 of 41) of the patients were on PIs-containing regimen. There were no differences in total cholesterol, triglyceride, LDL-C, HDL-C, and ABI between HAART-treated group and ART-naive group. Compared with ART-naive patients, the HAART-treated patients showed a lower body mass index, blood pressure, and PWV, as well as the HIV RNA level (P < 0.05).TABLE 1: Comparison of Traditional Risk Factors and Subclinical Atherosclerosis Parameters Among ART Naive Patients With HIV, Control subjects, and HAART-Treated Patients With HIVThe idea that HIV-infected patients are likely to have an atherogenic lipid profile, such as high levels of total cholesterol, LDL-C, and triglyceride, has been reported by several studies.4,5 HAART is thought to be associated with these metabolic abnormalities. However, our ART-naive patients had lower total cholesterol and LDL-C levels, which is an atheroprotective lipid profile. This result supports the assumption that HAART may be contributory to dyslipidemia in HIV-infected patients. In this study, we proved that ART-naive patients have lower ABI compared with HIV-uninfected control subjects, suggesting that HIV-infected patients are more likely to develop peripheral arterial disease. We also found in ART-naive HIV-infected patients having higher PWV level, which is also a marker of subclinical atherosclerosis. Multiple variant analysis confirms that HIV infection is a risk factor of reduced ABI, independent of other traditional cardiovascular risk factors, suggesting that HIV infection itself can promote the atherosclerosis of peripheral artery. HAART used to be considered as the major cause of premature atherosclerosis in HIV-infected patients. Now, the result of our study implies that this accelerated atherosclerosis should be contributed to both HAART and HIV infection. This is of great significance, and henceforth, the therapeutic strategies of atherosclerosis in HIV-infected patients should not only focus on the metabolic complication of HAART but also on the HIV infection. However, we did not find associations between ABI and HIV infection parameters including CD4+ lymphocyte count, CD8+ lymphocyte count, CD4+/CD8+ ratio, and HIV RNA. The exact mechanism of this atherogenic process is unclear. Unlike the previous study, HAART do not aggravate the peripheral arterial disease in our patients with HIV and the ABI is not significantly different between ART-naive and HAART-treated groups. PWV is surprisingly lower in HAART-treated patients with HIV, accompanying with lower levels of body mass index and systolic blood pressure. This is mainly because most of our patients are treated with NNRTIs and NRTIs, not PIs. Actually, NRTIs have been proved to provide protection from atherogenic side effect of PIs by downregulation of protein kinase C alpha in animal models6. Second, we have demonstrated that HIV infection is a risk factor of peripheral arterial disease, so treatment of the underlying HIV infection may prevent the process of atherosclerosis. However, because our study is a cross-sectional study, it has its limitation and does not estimate the true magnitude of the contribution of variables such as disease activity and therapy. In conclusion, we have found that HIV infection itself is a risk factor of peripheral arterial disease, independent of traditional cardiovascular risk factors and HAART. NRTIs/NNRTIs-containing HAART regimens do not seem to accelerate the atherosclerosis, and switch to a NNRTIs and/or NRTIs-containing regimen from a PIs-containing regimen may be reasonable in some kind of patients. ACKNOWLEDGMENT We thank Dr. Tao Xu for his advice of statistical analysis. Yicong Ye, MD* Yong Zeng, MD* Xiaomeng Li, MD* Shuyang Zhang, MD* Quan Fang, MD* Ling Luo, MD† Zhifeng Qiu, MD† Yang Han, MD† Taisheng Li, MD† *Departments of Cardiology; and †Infectious Diseases, Peking Union Medical College Hospital, Peking Union Medical College and Chinese Academy of Medical Sciences, Beijing, China