Abstract
Introduction
The use of highly active antiretroviral therapy (HAART) has reduced mortality among HIV-infected persons; however, long-term adverse effects, such as dyslipidemia, glucose abnormalities, lipodystrophy, and other metabolic complications are emerging. 1 –4 In addition, there are increasing concerns about premature cardiovascular disease and the role of HAART in its development. 5 –8 Since hypertension is a well-known risk factor for cardiovascular disease, understanding the potential association between HAART and hypertension is important.
Prior to the advent of HAART, hypertension was infrequently reported in HIV-infected patients. 9 More recent reports suggest that hypertension is increasingly noted among HIV-infected patients. 10,11 Whether these trends are due to the aging of the HIV population, the accumulation of risk factors for the development of hypertension (eg, weight gain), or a direct effect of the HIV or the antiretroviral (ARV) medications, is unclear. Some studies have suggested that specific HAART medications, such as protease inhibitors (PIs), may be associated with elevated blood pressure. 11,12 We investigated the prevalence of hypertension and factors, including antiretroviral therapy (ART), associated with hypertension in a large group of relatively young, healthy HIV-infected persons. Our study cohort has the advantage of defined dates of HIV infection and HAART use, as well as detailed clinical follow-up.
Methods
We performed a cross-sectional study among all HIV-infected patients who received medical care at the Naval Medical Center San Diego (NMCSD) or the National Naval Medical Center (NNMC) in Bethesda, Maryland. The patient population consisted of Department of Defense active duty members and their beneficiaries who were ≥18 years old. All participants had free access to health care and medications. Given the military drug-testing policies, the rate of illicit drug use in our populations is uncommon. This study was approved by the Institutional Review Boards at each site.
Data abstracted from the medical records included age, gender, ethnicity, body mass index (BMI), presence of hypertension, diabetes mellitus, chronic hepatitis B infection (positive surface antigen), lipid levels, duration of HIV infection (midpoint between the first seropositive and last seronegative test to the time of study initiation or time from first seropositive to study initiation for those without a negative test result), current and nadir CD4 counts (flow cytometry), current plasma HIV RNA level (Roche Amplicor, undetectable as <50 copies/mL), and receipt of HAART.
Blood pressure was measured in a standardized manner using a calibrated automated machine with each participant sitting in a relaxed, upright position. Hypertension was defined using standard definitions by the World Health Organization (WHO)/International Society of Hypertension (ISH) guidelines: diastolic blood pressure (DBP) values of ≥90 mm Hg or systolic blood pressure (SBP) values of ≥140 mm Hg at 2 or more clinic visits. 13 Patients receiving an antihypertensive medication, irrespective of blood pressure, were also defined as hypertensive. Diabetes mellitus was defined as the use of antidiabetic medication or physician diagnosis and obesity as a BMI ≥30 kg/m2 in accordance with classifications by the National Institutes of Health. 14 Highly active antiretroviral therapy was defined as the use of 3 or more ARV medications per the Department of Health and Human Services guidelines. 15
Statistical analyses include descriptive statistics with numbers (proportions) and medians (interquartile ranges [IQR]) for categorical and continuous variables, respectively. Univariate logistic regression models were used to examine the association between hypertension and demographic variables of age, gender, and ethnicity as well as clinical variables of BMI, diabetes, chronic hepatitis B, lipid levels, duration of HIV, CD4 counts, HIV RNA levels, and ARV medication use. The current use of HAART, as well as the use of specific ARV medication classes (nucleoside reverse transcriptase inhibitors [NRTIs], nonnucleoside reverse transcriptase inhibitors [NNRTIs], and PIs) and select individual agents (based on adequate number receiving the agent) were examined for associations with hypertension.
Variables significant in the univariate logistic regression model at a P < .10 were further evaluated using a backward stepwise multivariate logistic regression model. In addition, the final model was adjusted for demographic data and clinical site (NMCSD and NNMC) based on prior data showing site differences. 16 Given that ARV medication use was the primary variable of interest in our study, we also evaluated its association with hypertension in the final model. All P values were 2-sided and a value of <.05 was considered statistically significant. All statistical analyses were performed using the Statistical Analysis Systems (SAS) version 9.1 software (SAS Institute Inc, Cary, North Carolina).
Results
A total of 707 HIV-infected individuals were evaluated with a median age of 41 years (IQR: 36-46); 92% were male; 49% Caucasian, 38% African American, and 14% were other ethnicities (Table 1). Obesity was noted in 18% of the study population, while diabetes mellitus was present in 8%. The median duration of HIV infection was 11.5 years (IQR: 6-17). The median CD4 count was 510 cells/mm3 (IQR: 368-704), and 78% had a current CD4 count >350 cells/mm3. Forty-three percent had an undetectable viral load (<50 copies/mL) and 72% were currently receiving HAART.
Descriptive Characteristics and Factors Associated with Hypertension in HIV Patients
Abbreviations: HIV, human immunodeficiency virus; OR, odds ratio; CI, confidence interval; SD, standard deviation; ART, antiretroviral treatment; HAART, highly active antiretroviral therapy; NNRTI, nonnucleoside reverse transcriptase inhibitors; NRTI, nucleoside reverse transcriptase inhibitors; DBP, diastolic blood pressure; SBP, systolic blood pressure.
a Factors are expressed as number (percentage) for categorical variables and as median (interquartile ranges) for continuous variables.
b Hypertensive defined as DBP ≥ 90 mm Hg or SBP ≥ 140 mm Hg at 2 or more clinic visits or use of antihypertensive medication.
c Odds ratio per 10-year increase in age.
d Odds ratio per 10-year increase in HIV duration.
e Per 100 cells/mm3.
f Includes only antiretroviral medications in which adequate numbers were receiving this agent.
g Among those receiving this antiretroviral medication.
The prevalence of hypertension in our study cohort was 31% (n = 219) and did not significantly differ among those receiving or not receiving HAART (32% vs 29%; P = .47). Characteristics associated with hypertension in the univariate analyses included increasing age (odds ratio [OR]: 2.0 per 10 year increase; P < .001). Compared to HIV-infected persons who were 18 to 39 years of age, those aged 40 to 59 years had a 2.5-fold higher prevalence of hypertension, while those ≥60 years had a 9.4-fold higher prevalence (Table 1).
Other factors associated with hypertension in the univariate analyses included increasing BMI (OR: 1.1, P < .001) and diabetes mellitus (OR: 2.8, P < .001). Among HIV-related factors, longer duration of HIV infection (OR: 2.1 per 10 year increase, P < .001), nadir CD4 counts ≤200 cells/mm3 (OR: 1.4, P = .04), and suppressed HIV RNA level (OR 1.4, P = .03) were associated with hypertension. Examination of the current CD4 count, both as a continuous variable and as a categorized variable, showed no significant associations with hypertension. Current receipt of HAART, individual drug classes, and individual ARV medications were not significantly associated with hypertension (Table 1). We also examined HAART exposure and duration of select NRTIs (ie, abacavir [ABC], tenofovir [TDF]) and found no associations.
Stepwise multivariate regression analyses were performed including factors with a P value < .10 as well as demographics and clinical site. Since age and HIV duration were highly correlated (r = .52, P < .001), we examined these variables in 2 separate models. In the first model with age, factors significantly associated with hypertension among HIV-infected persons included increasing age (OR: 2.1 per 10 year increase, P < .001), African American compared to Caucasian ethnicity (OR: 1.6, P = .02), increasing BMI (OR: 1.1, P < .001), and diabetes (OR 2.3, P = .01). In the second model with HIV duration, factors significantly associated with hypertension included HIV duration (OR: 2.2 per 10 years, P < .001), increasing BMI (OR: 1.1, P < .001), and diabetes (OR 2.9, P < .001). We also examined in our final models, current HAART use and individual ARV agents (shown in Table 1), but none were associated with hypertension.
In order to further evaluate the associations of age and HIV duration with hypertension, we repeated our final multivariate model utilizing categories for age and HIV duration based on their median values (Table 2). Compared to those ≤40 years of age with HIV duration of ≤10 years, the odds of hypertension were significantly increased for both persons with older ages and persons with longer durations of HIV infection. The highest odds of hypertension were noted among those who were both >40 years of age and had an HIV duration of >10 years (OR: 4.0, P < .001). Increasing BMI as well as diabetes also remained as significant factors associated with hypertension, with a borderline association for African American ethnicity (Table 2).
Multivariate Logistic Regression Analysis a of Factors Associated with Hypertension in HIV-Infected Patients
Abbreviations: OR, odds ratio; CI, confidence interval; HIV, human immunodeficiency virus.
a Adjusted for gender and clinical site.
Discussion
Our study cohort of relatively young HIV-infected persons had a high prevalence (31%) of hypertension. The prevalence of hypertension found in our study was similar to other selected HIV-infected cohorts (13%-34%) 10,17,18 but was higher than the reported age-matched historical rates in the general population. 19 For example, the prevalence of hypertension among HIV patients 18 to 39 years of age was 19% compared to the expected rate of 7% in the general population. 19 Given the impact of hypertension on cardiovascular and kidney diseases, which are major causes of morbidity and mortality among HIV patients, understanding the factors associated with elevated blood pressures in this population is important.
Factors associated with hypertension in our study included increasing age, diabetes, and elevated BMI, which are concurrent with other studies among HIV-infected persons 20 –23 and the general population. 19 Thus, the elevated prevalence of hypertension among HIV-infected persons may be related to these traditional risk factors, especially in view of the aging HIV population and their development of metabolic and weight-related issues. 2,24,25
Given the observed elevated prevalence of hypertension, we sought to determine whether there are additional HIV-specific risk factors associated with elevated blood pressures in this population. Our study found that increasing duration of HIV infection, above that associated with increasing age, was associated with hypertension in our multivariate models. 26 HIV infection has been linked to arterial stiffness and proinflammatory responses 27 ; hence, the virus may lead to premature vascular dysfunction causing elevated blood pressure. Beyond endothelial dysfunction and accelerated vascular aging, subclinical HIV-associated kidney effects may also explain the high rates of hypertension among HIV patients 17 ; further studies are needed.
Previous studies exploring the effect of ARV medications on blood pressure have shown conflicting results. 11,12,18,20 –22,28,29 Several studies have found no effects of HAART on blood pressure measurements. 18,22,29 In fact, one study found a lower prevalence of hypertension among HIV-infected participants treated with PIs compared with HIV-uninfected individuals. 28 Yet other investigations implicated PIs (eg, lopinavir [LPV] and indinavir [IDV]) in causing elevated blood pressures 12,20 ; some of these reported effects may be due to concurrent increases in BMI after HAART initiation. 20 The use of TDF has also been cited as a cause of hypertension and may be due, at least in part, to its impact on renal function. 20 Finally, a large cross-sectional study revealed modest elevations in SBP and DBP are associated with NNRTI-based ARV regimens; however, this association attenuated after adjustment for age. 11
Given the variable results of prior studies, we examined the relationship between hypertension and ARV medication use in a large clinical cohort of HIV-infected persons and found no significant associations with current or past ARV medication use. Our lack of associations is concurrent with the D:A:D cohort study, which also did not demonstrate a relationship between any class of ARV medications and the prevalence or incidence of hypertension. 22 Similarly, a second study from the D:A:D showed that the prevalence of hypertension among patients receiving HAART was similar to HAART-naive patients. 29 Regarding associations with current HIV RNA level or CD4 count and hypertension, like other studies, 17,18 we found no significant associations.
Our study had potential limitations. Our primary analyses were based on a cross-sectional study; hence, temporality could not be established between factors and the development of hypertension. In addition, our study did not include an HIV-negative control group for comparison. Blood pressure measurements were performed at 2 different clinical sites and subjected to fluctuations secondary to human and equipment-related factors; however, we adjusted our analyses for the and any biases due to the measurement differences would likely be of a nondifferential type. We also did not have data available on other factors that may influence blood pressure, including tobacco use, alcohol use, and exercise. Finally, the generalizability of our results may be limited due to inherent differences between the military population and the general population, including the paucity of women and intravenous drug users in our cohort. However, our cohort provides important data on the factors associated with hypertension among early diagnosed and treated HIV-infected adults.
Our study had several strengths. It was conducted in a well-characterized population of HIV-infected persons with close clinical follow-up including information on HAART use. In addition, since the majority of our population had defined seroconversion windows, our study adds novel data regarding the association of HIV duration and hypertension.
In summary, hypertension is common among HIV-infected persons, including those of young ages with few comorbidities. In our large cohort of HIV-infected persons, HAART was not associated with the presence of hypertension, suggesting that ART does not play a significant role in the high rates of hypertension in this population. The association of longer duration of HIV infection with hypertension suggests viral-mediated changes in vascular tone may be occurring over time; further studies are needed. Given the negative health consequences of hypertension, HIV-infected persons should undergo routine blood pressure measurements during clinic visits, and therapy should be instituted in an effort to reduce cardiovascular complications in this population.
The content of this publication is the sole responsibility of the authors and does not necessarily reflect the views or policies of the National Institutes of Health (NIH) or the Department of Health and Human Services, the Department of Defense (DoD) or the Departments of the Army, Navy, or Air Force. Mention of trade names, commercial products, or organizations does not imply endorsement by the US Government. The authors have no commercial or other association that might pose a conflict of interest in this work.
The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: support for this work (IDCRP-052) was also provided by the Infectious Disease Clinical Research Program (IDCRP), a Department of Defense (DoD) program executed through the Uniformed Services University of the Health Sciences. This project has been funded in whole, or in part, with federal funds from the National Institute of Allergy and Infectious Diseases, National Institutes of Health (NIH), under Inter-Agency Agreement Y1-AI-5072.