Advertisement
Browse Subject Areas
?

Click through the PLOS taxonomy to find articles in your field.

For more information about PLOS Subject Areas, click here.

  • Loading metrics

HIV infection as vascular risk: A systematic review of the literature and meta-analysis

  • Jose Gutierrez ,

    jg3233@cumc.columbia.edu

    ‡ These authors are co-first authors on this work.

    Affiliation Department of Neurology, Columbia University Medical Center, New York, NY, United States of America

  • Ana Letícia A. Albuquerque ,

    ‡ These authors are co-first authors on this work.

    Affiliation School of Medicine, Federal University of Alagoas, Maceió, AL, Brazil

  • Louise Falzon

    Affiliation Center for Behavioral Cardiovascular Health, Columbia University Medical Center, New York, NY, United States of America

HIV infection as vascular risk: A systematic review of the literature and meta-analysis

  • Jose Gutierrez, 
  • Ana Letícia A. Albuquerque, 
  • Louise Falzon
PLOS
x

Abstract

Importance

The vascular risk attributable to HIV infection is rising. The heterogeneity of the samples studied is an obstacle to understanding whether HIV is a vascular risk across geographic regions.

Objective

To test the hypothesis that HIV infection is a vascular risk factor, and that the risk conferred by HIV varies by geographical region.

Data sources

A systematic search of publications was carried out in seven electronic databases: PubMed, The Cochrane Library, EMBASE, Web of Science, LILACS, ClinicalTrials.gov, and WHO International Clinical Trials Registry Platform from inception to July 2015.

Study selection

We included longitudinal studies of HIV+ individuals and their risk of vascular outcomes of ≥ 50 HIV+ cases and excluded studies on biomarkers of vascular disease as well as clinical trials.

Data extraction and synthesis

Data was extracted by one of the authors and independently confirmed by the other two authors. We used incidence rate (IR), incidence risk ratio (IRR) and hazard ratio (HR) with their 95% confidence intervals as measures of risk.

Main outcome

All-death, myocardial infarction (MI), coronary heart disease (CHD), any stroke, ischemic stroke (IS) or intracranial hemorrhage (ICH).

Results

We screened 11,482 references for eligibility, and selected 117 for analysis. Forty-four cohorts represented 334,417 HIV+ individuals, 49% from the United States. Compared with their European counterparts, HIV+ individuals in the United States had higher IR of death (IRR 1.78, 1.69–1.88), MI (IRR 1.61, 1.29–2.01), CHD (IRR 2.27, 1.92–2.68), any stroke (IRR 1.94, 1.59–2.38), IS (IRR 1.56, 1.23–1.98), and ICH (IRR 4.03, 2.72–6.14). Compared with HIV- controls and independent of geographical region, HIV was a risk for death (HR 4.77, 4.55–5.00), MI (HR 1.60, 1.49–1.72), any CHD (HR 1.20, 1.15–1.25), any stroke (HR 1.82, 1.53–2.16), IS (HR 1.27, 1.15–1.39) and ICH (HR 2.20, 1.61–3.02). Use of antiretroviral therapy was a consistent risk for cardiac outcomes, while immunosuppression and unsuppressed viral load were consistent risks for cerebral outcomes.

Conclusions

HIV should be considered a vascular risk, with varying magnitudes across geographical and anatomical regions. We think that strategies to reduce the HIV-related vascular burden are urgent, and should incorporate the disparities noted here.

Introduction

Vascular disease is the leading cause of global death, [1] and it has become a particular concern in the aging HIV population. The World Health Organization estimates that approximately 37 million people are currently living with HIV/AIDS. [2] Non-infectious comorbidities have emerged as important sources of morbidity and mortality in HIV+ populations, [3] with vascular disease being the leading cause of death among older HIV+ individuals. [4] Whether the rates of vascular disease are a reflection of aging and aging-comorbidities per se or HIV-related is an area of ongoing investigation. Furthermore, disparities in rates of vascular disease are influenced by geographical factors that not only reflect greater prevalence of vascular risk factors but also socioeconomic factors that vary regionally.[5] HIV in the United States (US) affects predominantly underserved populations, often living in circumstances that may also influence vascular disease.[6] Additionally, evidence suggests that HIV-related variables such as immunological factors,[7] use of cART, [8] and co-infection with other viruses may modify the risk of vascular disease seen in HIV+ populations.[9] Because the prevalence of these confounders of risk vary depending on the population studied,[1012] it is plausible that some of disparities in vascular risk seen in HIV+ populations may be attributable to geographical confounders.

Therefore, we conducted a systematic review and a meta-analysis of published medical literature to assess the risk of death and vascular disease among HIV+ individuals to test the hypothesis that HIV infection is a vascular risk factor, and that the risks conferred by HIV varies by geographical region.

Methods

Data sources and searches: Our first search was conducted on PubMed on May 26th, 2015 by a librarian (LouF), and 7,203 articles were found. Our second search was carried out in EMBASE, Web of Science and LILACS, within a week (from July 27th to July 31st). 1,172 articles were found to be duplicates from our first search, so we carefully excluded them leaving 2,859 new articles, of which 1,500 were from EMBASE, 1,293 from Web of Science and 66 from LILACS. Our third search was conducted in The Cochrane Library, ClinicalTrials.gov and WHO International Clinical Trials Registry Platform on July 29, 2015, and 248 studies were found. After the exclusion of the non-relevant studies and the duplicates, 37 new articles remained, of which 31 were from The Cochrane Library, 2 from ClinicalTrials.gov and 4 from WHO International Clinical Trials Registry Platform. Therefore, a total of 11,482 studies were screened for inclusion in our meta-analysis (Fig 1, S1 File and S2 File). One publication from a cohort identified through our search was brought to our attention by one of the study investigators.[13].

Longitudinal studies of HIV+ individuals were selected for analysis. The studies were included if they met the following criteria: (1) observational study design; (2) sample size ≥ 50 HIV+ cases, and (3) all-death, myocardial infarction (MI), any coronary heart disease (CHD), any stroke, ischemic stroke (IS) or intracranial hemorrhage (ICH) reported as primary outcome. Studies were excluded if: (1) not about HIV; (2) not about vascular disease; (3) related to cognition and/or dementia; (4) animal studies; (5) in vitro studies; (6) cross-sectional design; (7) small samples (< 50 patients); (8) meta-analysis, systematic reviews and reviews; (9) studies that were not in English, Spanish, French or Portuguese; (10) editorials; (11) studies exploring only pathogenesis; (12) clinical trials (to improve more representative samples of the general population with HIV) and (13) studies from before cART era (<1994).

Study Selection: The first screening in each database was carried out by looking at the title and the abstract of the articles. The second screening consisted of retrieving the full text or extended abstract from pre-screened citations. One of us (AnaA) independently classified the studies as appropriate for further consensus review (between AnaA and JosG) to select studies for this analysis.

Data extraction and quality assurance: We obtained from each of the selected articles the demographic, clinical and laboratory data reported by HIV status. We noted whether the study provided a measure of risk (i.e. Hazard ratios, risk ratio, etc.), and whether the given measure was adjusted for confounders. If the study did not report measures of risks, we calculated crude rate ratios with the provided number of events and sample size. We extracted crude incidence rates when provided. If not provided, we estimated incidence rate and their confidence intervals with the data provided by the authors. If the author did not provide total person-year follow-up, we estimated it by multiplying the follow-up time by the number of subjects and contrasted the estimated incidence rate with the reported in the paper. In the cohorts where HIV negative controls were included, we prioritized obtaining the adjusted hazard ratios, but if not available, we obtained the unadjusted hazard ratio, the risk ratio, or calculated the unadjusted hazard ratio as indicated elsewhere.[14] We used exchangeable incidence rate ratio (IRR), risk ratio, relative risk, with hazard ratios.[15] We noted whether the population study reflected first-time event, recurrent events in populations with known vascular disease, and the geographical region. Finally, we extracted the hazard ratio or risk ratio in cohorts that included HIV only models.

An ideal study would include a random representative sample of the population of HIV in the geographical area of study, compared to a representative or random sample from HIV uninfected controls from the same geographical area. The study must present adjusted hazard ratios by traditional vascular risk factors and ideally would control not only by the presence or absence or vascular risk factors, but also by the degree of control as defined by established clinical guidelines or by using continuous longitudinal measurements of the intensity of such vascular risk factors. A good study would include commodity samples (e.g. administrative datasets, hospital-based, etc.) that included matched (at least by age and sex) HIV- controls, and that reported hazard ratios adjusting for at least demographics and vascular risk factors (expressed categorically or with some continuous measurements measured at baseline). A fair study would report only unadjusted rates of a given outcome.

Data synthesis and Analysis: Demographic, clinical and laboratory data were summarized with simple averages from all the studies that provided them. When a given study provided stratified data, a weighted average was obtained accordingly to be able to harmonize with other studies. Incidence rate and 95% CI (when not provided by the authors) were calculated from raw data using MedCalc Statistical Software version 16.2.1 (MedCalc Software bvba, Ostend, Belgium; https://www.medcalc.org; 2016).

Statistical analysis

Fixed effects were used for all outcomes. Random effects were used to test robustness of results.[16, 17] The natural logarithm of the HR and their 95% CI for demographic, clinical and laboratory variables were calculated for each study.[14] The average HR and 95% CI for each outcome was obtained using Inverse Variance statistical method provided by Review Manager (RevMan) [Computer program]. Version 5.3. Copenhagen: The Nordic Cochrane Centre, the Cochrane Collaboration, 2014. Heterogeneity was tested only in geographic subgroups and it was quantified with Q test and I2. A two-tailed alpha value of ≤ 0.05 was used to determine statistical significance.

Results

a) Characteristics of the HIV populations studied

Forty-four cohorts represented 334,417 HIV+ individuals, 49% from the US (Tables A-C in S1 File). The majority of the studies with extractable data and well-defined vascular outcomes came from the US and Europe, which facilitated the comparison between the US and Europe. Using simple linear regression with each study weighted by their sample size, HIV+ individuals in the US had a higher mean age (B = 10.0, P = 0.001), were more likely non-white (B = 27.3, P = 0.005), had viral loads < 1000 copies (B = 5.05, P = 0.001), hepatitis C (B = 24.33, P = 0.01) or higher CD4 counts (B = 48.1, P = 0.01) than their European counterparts. Among the 11 studies that included HIV- controls, HIV+ were more commonly men, non-white, and had a higher prevalence of hypertension, diabetes, dyslipidemia, smoking, hepatitis C, drug use, and baseline cardiac disease (Table D in S1 File).

b) Incidence rate of death and vascular disease in HIV+ versus HIV- populations (Table 1)

thumbnail
Table 1. Incidence rate (per 1,000 person-year) by HIV status and geographical regions.

https://doi.org/10.1371/journal.pone.0176686.t001

Among HIV+ individuals, the MI incidence was higher than stroke incidence (IRR 1.31, 1.05–1.63). By defining regions more homogenously, however, HIV+ individuals in the US had a higher stroke incidence than that of MI (IRR 1.37, 1.15–1.64) but less so in Europe (IRR 1.13, 95% CI 0.89–1.44) and in low-to-middle income countries (IRR 1.33, 95% CI 0.94–1.89. Among HIV- individuals in the US, the incidence rates of stroke versus MI were similar (IRR 1.09, 0.93–1.28).

Compared with their European counterparts, HIV+ individuals in the US had a higher IRR for death (1.78, 1.69–1.88), MI (1.61, 1.29–2.01), CHD (2.27, 1.92–2.68), stroke (1.94, 1.59–2.38), IS (2.46, 1.98–3.07), and ICH (4.03, 2.72–6.14). The same geographic disparity was noted among HIV- controls so that those in the US had a higher IRR for death (4.56, 4.08–5.11), MI (1.42, 1.19–1.70), CHD (1.34, 1.14–1.59), stroke (1.43, 1.21–1.71), IS (4.45, 3.37–5.96), and ICH (6.18, 3.68–11.00) than their uninfected European or Canadian counterparts.

c) Outcomes by HIV status (Tables E-K in S1 File)

-Death.

HIV infection conferred a four-fold increase in the risk of death compared with HIV- controls (HR 4.77, 4.55–5.00). The point estimate was the highest in the US, but the significance remained similar across geographical regions (Fig 2). Of the six included studies, five did not adjust for confounders; the only study that did adjust for confounders showed a significant 4-fold increase risk of death attributable to HIV.[18] The results were heterogeneous (I2 = 98%), but the six individual point estimates were independently significant. Varying the estimation method from fixed effect to random effect did not change the statistical significance but it reduced the point estimate by 23%.

thumbnail
Fig 2. Risk of death in HIV+ individuals vs. HIV- controls.

Red squares represent the point estimate for HIV-related vascular risk by study, and the size of each red square represents the weight of each study in the average effect size. Horizontal lines across red squares represent the 95% confidence intervals for the point estimate. The diamonds represent the weighted average point estimate.

https://doi.org/10.1371/journal.pone.0176686.g002

-Myocardial infarction.

HIV+ individuals had a higher risk of MI compared with HIV- controls (HR 1.60, 1.49–1.72, Fig 3). The seven studies included adjusted for at least sex and age, and the studies from administrative datasets in the US and Europe also controlled for vascular risks. The results were significant across regions, but the point estimates were lower for those in the US. The two smallest studies that reported a non-significant increased risk of MI with HIV studied recurrent MI as outcome. The results were moderately heterogeneous (I2 = 64%), mostly due to the two studies that showed a point estimate greater than two.[19, 20] The results were insensitive to the effect estimation method.

thumbnail
Fig 3. Risk of myocardial infarction in HIV+ individuals vs. HIV- controls.

Red squares represent the point estimate for HIV-related vascular risk by study, and the size of each red square represents the weight of each study in the average effect size. Horizontal lines across red squares represent the 95% confidence intervals for the point estimate. The diamonds represent the weighted average point estimate.

https://doi.org/10.1371/journal.pone.0176686.g003

-Any CHD.

Nine studies had data for any CHD, five referring to MI and 4 to CHD. HIV+ individuals had a higher risk of CHD than HIV- controls (HR 1.20, 1.15–1.25, Fig 4). Taking out the five studies with MI as outcome reduced significantly the strength of the association between HIV and CHD, but it remained significant (HR 1.05, 1.00–1.10). Among the four studies reporting CHD, only one adjusted for multiple confounders, reporting twice the risk of CHD with HIV.[21] The results were heterogeneous as the point estimate ranged from 1.02 to 2.40 (I2 = 94%). The results were significant across regions, but the point estimates were lower for those in the US. Furthermore, using a random model rather than fixed model increased the strength of association by 40%.

thumbnail
Fig 4. Risk of any coronary artery disease in HIV+ individuals vs. HIV- controls.

Red squares represent the point estimate for HIV-related vascular risk by study, and the size of each red square represents the weight of each study in the average effect size. Horizontal lines across red squares represent the 95% confidence intervals for the point estimate. The diamonds represent the weighted average point estimate.

https://doi.org/10.1371/journal.pone.0176686.g004

-Any stroke.

HIV+ individuals had a higher risk of any stroke than HIV- controls (HR 1.82, 1.53–2.16, Fig 5). The results were consistent across the Atlantic, with higher point estimates in the US. The risk of stroke among individuals with established CHD was numerically but not nominally higher among HIV+ individuals. The results were homogenous (I2 = 0%) and insensitive to the effect estimation method.

thumbnail
Fig 5. Risk of cerebrovascular events in HIV+ individuals vs. HIV- controls.

Red squares represent the point estimate for HIV-related vascular risk by study, and the size of each red square represents the weight of each study in the average effect size. Horizontal lines across red squares represent the 95% confidence intervals for the point estimate. The diamonds represent the weighted average point estimate.

https://doi.org/10.1371/journal.pone.0176686.g005

-Ischemic stroke.

HIV+ individuals had a higher risk of IS than HIV- controls (HR 1.27, 1.15–1.39, Fig 4). The point estimate was higher for the European study included in this comparison, but this study used rate ratio (American studies used HR) and adjusted only for age, sex and country of origin but not for vascular risk factors.[22, 23] The results were homogenous (I2 = 28%) and insensitive to the effect estimation method.

-Intracranial hemorrhage.

The risk of hemorrhagic stroke was higher in HIV+ individuals compared to uninfected controls (HR 2.20, 1.61–3.02, Fig 4). The risk was adjusted for confounders in all studies. The results were homogenous (I2 = 7%) and the point estimate did not change using a random model. The point estimates were higher for the non-American studies than for the American study included in this outcome.

d) Vascular risks among HIV+ individuals (Tables 2 and 3)

thumbnail
Table 2. Meta-analysis of demographic and vascular risk among HIV infected cohorts.

https://doi.org/10.1371/journal.pone.0176686.t002

thumbnail
Table 3. Meta-analysis of HIV-related variables and coinfections among HIV infected cohorts.

https://doi.org/10.1371/journal.pone.0176686.t003

Older age was the more consistent risks for outcomes among HIV+ individuals. While male sex was a significant predictor of MI (HR 1.97, 1.51–2.57) and CHD (HR 1.76, 1.33–2.32); it was not for stroke (HR 1.11, 0.83–1.49). Traditional vascular risks predicted MI/CHD and stroke, but not IS. The study used to extract HR for IS used atrial fibrillation, structural heart disease and anticoagulant/antiplatelet together with traditional vascular risks,[23] which confounds the role of risk factors into a given stroke etiology.

While the risk of death was higher among cART naïve patients (HR 5.60, 3.18–9.86), the risk of MI (HR 1.20, 1.09–1.33) was higher with the use of cART, especially abacavir (HR 1.77, 1.50–2.10) and protease inhibitors (HR 1.09 per year of use, 1.08–1.47). While abacavir was also identified as a risk factor for stroke (HR 2.10, 1.20–3.66), the use of cART or different subtypes of cART was less consistently reported as a risk factor for IS or ICH. While the role of CD4 counts was equivocal for MI/CHD, CD4 counts < 200 cells was a risk factor for death (HR 2.98, 2.67–3.32), stroke (HR 1.91, 1.44–2.54) and ICH (HR 4.61, 2.09–10.2). Similarly, the risk of MI/CHD due to unsuppressed viral replication is equivocal, while unsuppressed viral replication is a risk for death (HR 1.65 if >10-100K, 1.50–1.81), stroke (HR 3.97 if > 400 copies, 1.90–8.31), and IS (HR 1.10 per log, 1.04–1.17). While the use of drug use and hepatitis b were less consistently related with vascular outcomes, Hepatitis C emerged as a consistent risk for death (HR 1.74, 1.55–1.96), CHD (HR 3.94, 1.00–15.5) and stroke (HR 1.22, 1.06–1.40).

Discussion

The cumulative evidence presented here suggests that HIV+ individuals have a higher risk of death and vascular disease than uninfected populations. The results were consistent even in studies that used adjusted models of good quality and in those that used referent uninfected populations with similar socioeconomic background.[12, 24, 25] Although there exist unmeasured variability in vascular risk factor control and other possible confounders, this deficit in methodology does not change the public health implication that HIV+ populations are at a greater risk of vascular disease than uninfected controls. Furthermore, the increased vascular risk is consistent across geographical regions, which adds strength to our main results. The magnitude of the risk, however, differs by region, with greater rates of vascular disease noted in HIV+ individuals in the US compared to those living in Europe. Similar geographic disparities were also noted for HIV- controls living in the US compared to those in Europe, and this disparity is not exclusive to the HIV- controls included in this meta-analysis, but also noted in studies more representative of the general population in US and Europe.[26, 27] Together, these data suggest that individuals in the US included in this meta-analysis (predominantly from matched underserved populations) and in general have poorer vascular health than those in Europe. The infection with HIV accentuates even further such disparity.

The disparities in risk attributed to HIV may have several explanations. First, the incidence rates used in our analysis are averages of several years. As the care for those with HIV has improved, the rates of vascular risk have decreased and the relative risk compared with uninfected controls has attenuated. [13, 2830] Also, because greater risk of vascular events is often noted after initiation of cART in patients with low CD4 counts compared to starting it when CD4 counts are higher, [21, 31] more recent event rates may be lower as greater emphasis is given to start cART earlier in the course of the infection.[32] The higher prevalence of vascular risk factors among HIV+ individuals included in this study (Table D in S1 File) or HIV+ individuals in the US (Table L in S1 File) compared to uninfected referent populations may be another contributor to the disparity in vascular risks. Traditional vascular risks confer a higher risk of death and vascular disease (Table 3), and thus represent a clear target for intervention. It is a major limitation that several studies focused on vascular disease did not report rates of vascular risk factors, and that only a minority of studies reported the degree of control of such risks, so an even greater proportion of the HIV-related disparity in vascular risk attributable to traditional vascular risk factors cannot be excluded. Arterial inflammation may also play a role in HIV-related vascular disease. For example, evidence exists of arterial inflammation in the aorta and brain arteries accompanied by activation of soluble blood biomarkers of inflammation, which relates to biomarkers of vascular risk.[33, 34] Accelerated atherosclerosis has been documented in HIV+ individuals as documented by a higher carotid intima-media thickness and coronary artery calcium rate of progression mostly driven by traditional vascular risk factors, with a lesser role for higher viral loads and CD4 nadir < 200.[35, 36] Determining the triggers for arterial inflammation may offer new therapeutic venues in this population.

Coronary heart disease is predominantly due to atherosclerosis, and the risk is typically related to traditional vascular risk factors and cART, especially abacavir (Table 3).[37] Stroke during HIV infection, on the other hand, is a heterogeneous disease where traditional vascular risk factors play only a partial role. For example, the Framingham risk score is a better predictor of biomarkers for CHD [38] than for stroke risk among HIV+ individuals.[39] Studies that define stroke more broadly do reproduce known associations between stroke and vascular risk factors among those with HIV, but reports that disclose the mechanisms of stroke in more details provide etiologies not typically found in cardiac disease, such as hypercoagulable state, vasculitis, and HIV vasculopathy.[4043] These data suggest the need for greater discrimination in mechanistic studies between cerebral and cardiac outcomes, so that targeted preventive measures can be appropriately applied.

Results of this study should be interpreted in light of the following limitations. The populations studied are heterogeneous, most outcomes are obtained from administrative datasets, and adjudication of vascular events was usually not confirmed by chart review. The studies we included over-represent HIV+ populations on cART and results may not apply to the entire HIV+ population. However, given that HIV suppression, immune competence, and access to care protect against death and vascular events, vascular risks of HIV may be even greater for these underrepresented in this study. This bias is not a minor issue: in the US alone, more than 1.2 million people live with HIV and over 50,000 new cases of HIV infection are diagnosed each year.[44] Of this population, 14% are unaware of their diagnosis, 60% do not engage in care, and only a minority received cART or achieve viral suppression, with the lowest rate of viral suppression noted in the youngest cohort. Because one out of four newly HIV-infected individuals and over 100,000 currently living with HIV are between 13–24 years old, [45] studying HIV-related vascular disease represents a major public health challenge due to high indirect costs of vascular disease in the young such as longer years of lost productivity and premature death attributable to vascular disease.

Supporting information

S1 File.

Table A: Demographic, clinical and immunological characteristics of HIV cohorts studied at baseline enrollment Table B: General characteristics of the studies including HIV+ cases compared with HIV- controls Table C: Weighted averages of study characteristics of the HIV samples studied across geographical regions Table D: Difference in demographic and clinical characteristics in HIV+ compared with HIV- cases Table E: All-death risks among HIV+ samples Table F: Myocardial infarction risks among HIV+ individuals Table G: Coronary Heart Disease risks among HIV+ individuals Table H: Cerebrovascular risks among HIV+ individuals Table I: Vascular death risks among HIV+ individuals Table J: Any vascular disease risks among HIV+ individuals Table K: Other Vascular Outcomes risks among HIV+ individuals Table L: Characteristics of the HIV studies from the US included in this study and the NHANES-derived characteristics of HIV infected individuals in the United States and the general uninfected population.

https://doi.org/10.1371/journal.pone.0176686.s001

(PDF)

Acknowledgments

Dr. Ken Cheung for statistical advising.

Author Contributions

  1. Conceptualization: JG.
  2. Data curation: JG.
  3. Formal analysis: JG ALAA.
  4. Funding acquisition: ALAA LF.
  5. Investigation: JG ALAA.
  6. Methodology: JG LF ALAA.
  7. Project administration: JG.
  8. Resources: JG.
  9. Software: JG LF.
  10. Supervision: JG.
  11. Validation: JG LF ALAA.
  12. Visualization: ALAA JG.
  13. Writing – original draft: ALAA JG.
  14. Writing – review & editing: JG ALAA.

References

  1. 1. Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, et al. Heart disease and stroke statistics—2015 update: a report from the American Heart Association. Circulation. 2015;131(4):e29–322. Epub 2014/12/19. pmid:25520374
  2. 2. World Health Organization. HIV/AIDS 2014 [updated July 2015; cited 2015 July 30]. Available from: http://www.who.int/mediacentre/factsheets/fs360/en/.
  3. 3. Causes of death in HIV-1-infected patients treated with antiretroviral therapy, 1996–2006: collaborative analysis of 13 HIV cohort studies. Clin Infect Dis. 2010;50(10):1387–96. Epub 2010/04/13. PubMed Central PMCID: PMCPmc3157754. pmid:20380565
  4. 4. Sackoff JE, Hanna DB, Pfeiffer MR, Torian LV. Causes of death among persons with AIDS in the era of highly active antiretroviral therapy: New York City. Ann Intern Med. 2006;145(6):397–406. Epub 2006/09/20. pmid:16983127
  5. 5. Howard G, Cushman M, Kissela BM, Kleindorfer DO, McClure LA, Safford MM, et al. Traditional Risk Factors as the Underlying Cause of Racial Disparities in Stroke: Lessons From the Half-Full (Empty?) Glass. Stroke. 2011;42(12):3369–75. pmid:21960581
  6. 6. Pellowski JA, Kalichman SC, Matthews KA, Adler N. A pandemic of the poor: social disadvantage and the U.S. HIV epidemic. Am Psychol. 2013;68(4):197–209. pmid:23688088
  7. 7. Shrestha S, Irvin MR, Grunfeld C, Arnett DK. HIV, inflammation, and calcium in atherosclerosis. Arterioscler Thromb Vasc Biol. 2014;34(2):244–50. Epub 2013/11/23. pmid:24265418
  8. 8. Brouwer ES, Napravnik S, Eron JJ Jr., Stalzer B, Floris-Moore M, Simpson RJ Jr., et al. Effects of combination antiretroviral therapies on the risk of myocardial infarction among HIV patients. Epidemiology. 2014;25(3):406–17. Epub 2014/04/10. PubMed Central PMCID: PMCPmc4159700. pmid:24713880
  9. 9. Gillis J, Smieja M, Cescon A, Rourke SB, Burchell AN, Cooper C, et al. Risk of cardiovascular disease associated with HCV and HBV coinfection among antiretroviral-treated HIV-infected individuals. Antivir Ther. 2014;19(3):309–17. Epub 2014/01/17. pmid:24429380
  10. 10. Triant VA, Lee H, Hadigan C, Grinspoon SK. Increased acute myocardial infarction rates and cardiovascular risk factors among patients with human immunodeficiency virus disease. J Clin Endocrinol Metab. 2007;92(7):2506–12. Epub 2007/04/26. PubMed Central PMCID: PMC2763385. pmid:17456578
  11. 11. Rasmussen LD, Omland LH, Pedersen C, Gerstoft J, Kronborg G, Jensen J, et al. Risk of myocardial infarction in parents of HIV-infected Individuals: a population-based Cohort Study. BMC Infect Dis. 2010;10:169. Epub 2010/06/16. PubMed Central PMCID: PMC2909236. pmid:20546604
  12. 12. Currier JS, Taylor A, Boyd F, Dezii CM, Kawabata H, Burtcel B, et al. Coronary heart disease in HIV-infected individuals. J Acquir Immune Defic Syndr. 2003;33(4):506–12. Epub 2003/07/19. pmid:12869840
  13. 13. Marcus JL, Leyden WA, Chao CR, Chow FC, Horberg MA, Hurley LB, et al. HIV infection and incidence of ischemic stroke. AIDS. 2014;28(13):1911–9. Epub 2014/06/18. pmid:24937309
  14. 14. Parmar MK, Torri V, Stewart L. Extracting summary statistics to perform meta-analyses of the published literature for survival endpoints. Stat Med. 1998;17(24):2815–34. pmid:9921604
  15. 15. Hernán MA. The Hazards of Hazard Ratios. Epidemiology (Cambridge, Mass). 2010;21(1):13–5.
  16. 16. Cochrane Handbook for Systematic Reviews of Interventions: The Cochrane Collaboration; 2011.
  17. 17. Systematic reviews. CRD’s guidance for undertaking reviews in health care. York: CRD, University of York.; 2009. 281 p.
  18. 18. Carballo D, Delhumeau C, Carballo S, Bahler C, Radovanovic D, Hirschel B, et al. Increased mortality after a first myocardial infarction in human immunodeficiency virus-infected patients; a nested cohort study. AIDS Res Ther. 2015;12(1):4. Epub 2015/02/24. PubMed Central PMCID: PMC4336509.
  19. 19. Rasmussen LD, Helleberg M, May MT, Afzal S, Kronborg G, Larsen CS, et al. Myocardial infarction among danish HIV-infected individuals: Population-attributable fractions associated with smoking. Clin Infect Dis. 2015;60(9):1415–23. pmid:25595744
  20. 20. Durand M, Sheehy O, Baril JG, Lelorier J, Tremblay CL. Association between HIV infection, antiretroviral therapy, and risk of acute myocardial infarction: a cohort and nested case-control study using Quebec's public health insurance database. J Acquir Immune Defic Syndr. 2011;57(3):245–53. Epub 2011/04/19. pmid:21499115
  21. 21. Obel N, Thomsen HF, Kronborg G, Larsen CS, Hildebrandt PR, Sorensen HT, et al. Ischemic heart disease in HIV-infected and HIV-uninfected individuals: a population-based cohort study. Clin Infect Dis. 2007;44(12):1625–31. Epub 2007/05/23. pmid:17516408
  22. 22. Sico JJ, Chang CC, So-Armah K, Justice AC, Hylek E, Skanderson M, et al. HIV status and the risk of ischemic stroke among men. Neurology. 2015;84(19):1933–40. Epub 2015/04/12. PubMed Central PMCID: PMCPMC4433456. pmid:25862803
  23. 23. Chow FC, Regan S, Feske S, Meigs JB, Grinspoon SK, Triant VA. Comparison of Ischemic Stroke Incidence in HIV-Infected and Non–HIV-Infected Patients in a US Health Care System. J Acquir Immune Defic Syndr. 2012;60:351–8. pmid:22580566
  24. 24. Mateen FJ, Shinohara RT, Carone M, Miller EN, McArthur JC, Jacobson LP, et al. Neurologic disorders incidence in HIV+ vs HIV- men: Multicenter AIDS Cohort Study, 1996–2011. Neurology. 2012;79(18):1873–80. PubMed Central PMCID: PMC3525315. pmid:23077015
  25. 25. Freiberg MS, Chang CC, Kuller LH, Skanderson M, Lowy E, Kraemer KL, et al. HIV infection and the risk of acute myocardial infarction. JAMA Intern Med. 2013;173(8):614–22. Epub 2013/03/06. pmid:23459863
  26. 26. Bejot Y, Daubail B, Jacquin A, Durier J, Osseby GV, Rouaud O, et al. Trends in the incidence of ischaemic stroke in young adults between 1985 and 2011: the Dijon Stroke Registry. J Neurol Neurosurg Psychiatry. 2014;85(5):509–13. Epub 2013/11/20. pmid:24249786
  27. 27. Howard VJ, Kleindorfer DO, Judd SE, McClure LA, Safford MM, Rhodes JD, et al. Disparities in Stroke Incidence Contributing to Disparities in Stroke Mortality. Ann Neurol 2011;69(4):619–27. pmid:21416498
  28. 28. Helleberg M, Kronborg G, Larsen CS, Pedersen G, Pedersen C, Gerstoft J, et al. Causes of death among Danish HIV patients compared with population controls in the period 1995–2008. Infection. 2012;40(6):627–34. Epub 2012/07/14. pmid:22791407
  29. 29. Klein DB, Leyden WA, Xu L, Chao CR, Horberg MA, Towner WJ, et al. Declining relative risk for myocardial infarction among HIV-positive compared with HIV-negative individuals with access to care. Clin Infect Dis. 2015;60(8):1278–80. pmid:25595743
  30. 30. Bozzette SA, Ake CF, Tam HK, Phippard A, Cohen D, Scharfstein DO, et al. Long-term survival and serious cardiovascular events in HIV-infected patients treated with highly active antiretroviral therapy. J Acquir Immune Defic Syndr. 2008;47(3):338–41. Epub 2008/01/08.
  31. 31. Jensen-Fangel S, Pedersen L, Pedersen C, Larsen CS, Tauris P, Moller A, et al. Low mortality in HIV-infected patients starting highly active antiretroviral therapy: a comparison with the general population. AIDS. 2004;18(1):89–97. pmid:15090834
  32. 32. Cohen MS, Chen YQ, McCauley M, Gamble T, Hosseinipour MC, Kumarasamy N, et al. Prevention of HIV-1 infection with early antiretroviral therapy. N Engl J Med. 2011;365(6):493–505. Epub 2011/07/20. PubMed Central PMCID: PMCPMC3200068. pmid:21767103
  33. 33. Subramanian S, Tawakol A, Burdo TH, Abbara S, Wei J, Vijayakumar J, et al. Arterial inflammation in patients with HIV. JAMA. 2012;308(4):379–86. PubMed Central PMCID: PMC3724172. pmid:22820791
  34. 34. Gutierrez J, Menshawy K, Gonzalez M, Goldman J, Elkind MS, Marshall R, et al. Brain large artery inflammation associated with HIV and large artery remodeling. AIDS. 2016;30(3):415–23. Epub 2016/01/15. pmid:26765935
  35. 35. Volpe GE, Tang AM, Polak JF, Mangili A, Skinner SC, Wanke CA. Progression of carotid intima-media thickness and coronary artery calcium over 6 years in an HIV-infected cohort. J Acquir Immune Defic Syndr. 2013;64(1):51–7. Epub 2013/08/16. PubMed Central PMCID: PMCPmc3815556. pmid:23945252
  36. 36. Hsue PY, Lo JC, Franklin A, Bolger AF, Martin JN, Deeks SG, et al. Progression of atherosclerosis as assessed by carotid intima-media thickness in patients with HIV infection. Circulation. 2004;109(13):1603–8. Epub 2004/03/17. pmid:15023877
  37. 37. Sabin CA, Reiss P, Ryom L, Phillips AN, Weber R, Law M, et al. Is there continued evidence for an association between abacavir usage and myocardial infarction risk in individuals with HIV? A cohort collaboration. BMC Med. 2016;14:61. Epub 2016/04/03. PubMed Central PMCID: PMCPMC4815070. pmid:27036962
  38. 38. Falcone EL, Mangili A, Skinner S, Alam A, Polak JF, Wanke CA. Framingham risk score and early markers of atherosclerosis in a cohort of adults infected with HIV. Antivir Ther. 2011;16(1):1–8. Epub 2011/02/12. pmid:21311103
  39. 39. Mateen FJ, Post WS, Sacktor N, Abraham AG, Becker JT, Smith BR, et al. Long-term predictive value of the Framingham Risk Score for Stroke in HIV-positive vs HIV-negative men. Neurology. 2013;81(24):2094–102. Epub 2013/11/12. PubMed Central PMCID: PMCPMC3863354. pmid:24212385
  40. 40. Ortiz G, Koch S, Romano JG, Forteza AM, Rabinstein AA. Mechanisms of ischemic stroke in HIV-infected patients. Neurology. 2007;68(16):1257–61. Epub 2007/04/18. pmid:17438215
  41. 41. Vinikoor MJ, Napravnik S, Floris-Moore M, Wilson S, Huang DY, Eron JJ. Incidence and Clinical Features of Cerebrovascular Disease Among HIV-Infected Adults in the Southeastern United States. Aids Res Hum Retrov. 2013;29(7):1068–74.
  42. 42. Connor MD, Lammie GA, Bell JE, Warlow CP, Simmonds P, Brettle RD. Cerebral infarction in adult AIDS patients: observations from the Edinburgh HIV Autopsy Cohort. Stroke. 2000;31(9):2117–26. Epub 2000/09/08. pmid:10978040
  43. 43. Gutierrez J, Goldman J, Dwork AJ, Elkind MS, Marshall RS, Morgello S. Brain arterial remodeling contribution to nonembolic brain infarcts in patients with HIV. Neurology. 2015;85(13):1139–45. Epub 2015/09/01. PubMed Central PMCID: PMC4603890. pmid:26320196
  44. 44. Bradley H, Hall HI, Wolitski RJ, Van Handel MM, Stone AE, LaFlam M, et al. Vital Signs: HIV diagnosis, care, and treatment among persons living with HIV—United States, 2011. MMWR Morb Mortal Wkly Rep. 2014;63(47):1113–7. Epub 2014/11/27. pmid:25426654
  45. 45. Vital signs: HIV infection, testing, and risk behaviors among youths—United States. MMWR Morb Mortal Wkly Rep. 2012;61(47):971–6. Epub 2012/11/30. pmid:23190571