Secondary Logo

Journal Logo


The cardiovascular risk management for people living with HIV in Europe

how well are we doing?

Shahmanesh, Maryam; Schultze, Anna; Burns, Fiona; Kirk, Ole; Lundgren, Jens; Mussini, Cristina; Pedersen, Court; De Wit, Stephane; Kutsyna, Galyna; Mocroft, Amanda on behalf of EuroSIDA in EuroCOORD

Author Information
doi: 10.1097/QAD.0000000000001207
  • Free



Cardiovascular disease is the leading cause of death in Europe, accounting for nearly half of all-cause mortality [1]. Prolonged survival because of successful antiretroviral management of people living with HIV (PLHIV) has resulted in a cohort at increased risk of comorbidities and chronic diseases related to ageing such as cardiovascular disease [2–6]. Studies suggest that there is a greater prevalence of some modifiable cardiovascular risk factors in PLHIV and that HIV and/or HIV treatment may be an independent risk factor for cardiovascular disease [7–9]. To sustain the successes of combination antiretroviral therapy (cART) in reducing mortality and morbidity [2] in PLHIV and accessing care in Europe, we need to ensure that there is an equally vigorous approach to the management of the comorbidities associated with HIV and ageing, and in particular cardiovascular disease.

Patterns of cardiovascular risk vary across geographical settings and populations, partly because of genetic factors and partly because of lifestyle and environmental factors [1]. Successful management of cardiovascular diseases requires a stepped care approach: primary prevention, identifying risk factors, nonpharmacological management of modifiable risk factors, pharmacological management of risk factors and finally specialist care [10–12]. Although there has been growing awareness of the clustering of health-related risk and consequently long-term conditions, integrated management frequently fails. For example, a review found that even in well functioning health systems less than half of people with diabetes underwent screening and treatment of modifiable cardiovascular risk factors [13,14].

PLHIV are followed up regularly, however, although this provides the opportunity to manage co-morbidities, we have limited data on stepped care for cardiovascular disease: In particular of those identified to have a modifiable risk factor, what proportion achieves modification of their risk factor and what are the predictors of success?

The aim of this article is to describe the patterns of modifiable cardiovascular risk and explore predictors of successful medical management of modifiable risk in HIV-positive individuals accessing HIV care within the EuroSIDA cohort.



The EuroSIDA study is a prospective observational cohort of 18 791 HIV-1-positive adult patients in 108 clinics across 34 European countries, Israel and Argentina. This study has been described previously [15]. In brief, patients were enrolled from nine cohorts from May 1994 onwards. Information is collected onto standardized data collection forms every 6 months, including all CD4+ cell counts and viral loads measured since the last follow-up and starting and stopping dates for all cART. For the purposes of this analysis, we include all the individuals enrolled in EuroSIDA after 1 January 2000 for whom at least two measurements of predicted cardiovascular risk could be calculated using two different sets of time-updated variables. The baseline date was considered the date when the risk equation could first be calculated.


Time-updated cardiovascular variables contributing to the construction of the predicted risk estimates were carried forward for a maximum of 12 months. Development of moderate-to-high-cardiovascular risk was defined as a predicted 5-year cardiovascular risk of over 5% using both DAD and Framingham equations [16,17]. We used a conservative estimate of a predicted 5-year cardiovascular risk of 5%-equivalent to a 10-year cardiovascular risk of 10%, to capture individuals with both moderate and high cardiovascular risk. Modifiable risks, such as hypertension, smoking, high serum cholesterol, diabetes and being overweight, were defined according to European AIDS Clinical Society (EACS) guidelines (Table 1), and successful cardiovascular risk modification was defined as two consecutive measurements meeting the same guidelines. Weight loss from a BMI of over 25 to one that is less than 25 was used as a surrogate marker of successful lifestyle interventions (exercise and diet).

Table 1:
Definitions of modifiable cardiovascular risk factors and risk modification outcomes.

Statistical methods

Baseline associations between risk factors and predicted cardiovascular risk were evaluated using logistic regression.

Rates of risk development and risk modification were calculated by dividing the total person years of follow-up (PYFU) with the number of events. Person–time with incomplete covariate data were excluded from the calculation of rates and rate ratios. Individuals were followed until they experienced the outcome of interest for the first time If the outcome of interest did not occur, individuals were censored 6 months after their last available modifiable risk factor measurement (which equates their last available risk estimate) or 31 December 2014 whichever occurred first. For the analysis of risk modification, individuals were censored 6 months after their last available measurement of the relevant modifiable risk factor or at 31 December 2014, whichever occurred first. Factors associated with cardiovascular risk development and risk modification were investigated using Poisson regression.

All models were adjusted for sociodemographic variables (sex, ethnicity, risk group and region), calendar year, HIV-related variables (CD4+ cell count, CD4+ nadir, prior AIDS diagnosis, prior AIDS or non-AIDS event, cumulative cART exposure and viral load suppression), hepatitis B and C, and cardiovascular-related variables (hypertension and treatment, hyperlipidaemia and treatment, smoking status, BMI, diabetes and family history of cardiovascular disease). All analyses were conducted using SAS 9.3 (Statistical Analysis Software, Cary, North Carolina, USA).


Prevalence, incidence and pattern of cardiovascular risk

A total of 8762 individuals were included in the analysis. The baseline characteristics can be seen in Table 2. The majority of individuals were men (76.0%), white (87.1%) and had acquired their infection through sex with another man (45.2%).

Table 2:
Characteristics of the study population according to baseline cardiovascular risk.

The prevalence of traditional cardiovascular risk factors as defined in Table 1 was high; 32.0% of individuals were hypertensive, 27.1% overweight, 45.0% had high cholesterol levels, 5% had diabetes mellitus and 47.4% were current smokers. At baseline 1504 (17.2%) of individuals had a moderate-to-high (5% 5 year) cardiovascular risk according to the DAD risk assessment and 1729 (19.7%) according to Framingham's.

After adjustment (Fig. 1), having a high cardiovascular risk at baseline was associated with being man and older age. CD4+ nadir of less than 200 cells/μl, longer duration of cART, current CD4+ count of greater than 200 cells/μl and earlier year of entry into the cohort were also associated with moderate-to-high cardiovascular risk.

Fig. 1:
Baseline factor associated with a high cardiovascular risk (DAD) in multivariable1 logistic regression models.1The model was adjusted for all the factors listed in the figure.

Among the 7258 individuals who had a low predicted baseline risk, 1905 (26.2%) went on to develop a moderate-to-high cardiovascular risk, with an overall incidence rate of 6.83 95% confidence interval (CI) (6.53–7.34)/100 PYFU. Results from this analysis can be seen in Fig. 2. Individuals who developed moderate-to-high cardiovascular risk were more likely to be, older men, with a family history of cardiovascular disease, living in East Europe, with traditional modifiable cardiovascular risk factors (smoking, hypertension, high cholesterol, overweight and diabetes). MSM and those with a longer cumulative exposure to cART, with a CD4+ nadir of less than 200 cells/μl, current CD4+ cell counts of greater than 500 cells/μl and prior AIDS were more likely to develop moderate-to-high cardiovascular risk; whereas there was no relationship between HIV viral load being controlled by cART and the development of cardiovascular risk. Those on antihypertensive treatment and living in central Europe were less likely to develop cardiovascular risk.

Fig. 2:
Factors associated with the development of a high cardiovascular risk (DAD) in multivariable1 Poisson regression model.1The model was adjusted for all the factors listed in the figure.

Risk modification

A total of 1205 of 2077 (58.0%) individuals indicated for BP treatment successfully modified their BP, 1283 of 3919 (32.8%) smokers stopped smoking, 277 of 1394 (19.9%) individuals indicated for lipid-lowering treatment lowered their cholesterol levels, and 543 of 2163 (25.1%) overweight individuals lowered their BMI to below 25.

Blood pressure

In a multivariable Poisson model (Table 3), individuals less likely to modify their blood pressure (BP) were men, older, heterosexual, overweight, with high cholesterol and diabetes. Those with a family history of cardio vascular disease (CVD) and who had a prior cardiovascular event were more likely to modify their BP. There was some evidence of regional differences with individuals from northern Europe less likely to modify their BP as compared with individuals from southern Europe. Those with higher baseline BP, those on antihypertensive medication and those with CD4+ cell counts of greater than 200 cells/μl were less likely to control BP. There was no association between BP control and HIV viral load. BP control improved over time.

Table 3:
Incidence rates and rate ratios for modifying blood pressure.
Table 3:
(Continued) Incidence rates and rate ratios for modifying blood pressure.


Table 4 shows that after adjustment, there was good evidence that the rate of smoking cessation differed according to mode of infection and region, with injecting drug users and individuals from eastern Europe less likely to stop smoking. Those with a CD4+ nadir of less than 200 cells/μl and detectable viral loads were less likely to stop smoking. Those with a prior cardiovascular event were more likely to stop smoking. Rates of smoking cessation improved over time.

Table 4:
Incidence rates and rate ratios for stopping smoking.
Table 4:
(Continued) Incidence rates and rate ratios for stopping smoking.


After adjustment (Table 5), those with higher CD4+ cell counts and undetectable viral load were less likely to reduce their cholesterol. People with higher baseline cholesterol and diabetes were less likely to successfully reduce their cholesterol. Those with a prior CVD event were more likely to control their cholesterol. There was no evidence for sex-related or age-related differences in modification of cholesterol levels, and there was no change in cholesterol modification over time.

Table 5:
Incidence rates and rate ratios for reducing cholesterol.
Table 5:
(Continued) Incidence rates and rate ratios for reducing cholesterol.

Time trends

We found evidence that modification of BP and smoking increased over time. After adjustment, the rate of BP modification increased by 6% per calendar year [adjusted incidence rate ratio (aIRR) = 1.06, 1.03–1.09, P < 0.0001] and the rate of smoking cessation by 5% per year (aIRR = 1.05, 1.02–1.07, P < 0.0001).


The current study has demonstrated that the prevalence and incidence of cardiovascular risk in people accessing HIV care is very high, and only a modest proportion are able to modify any of their cardiovascular risk factors. However, risk modification seems to be improving over time. This needs to be interpreted in the context of a relatively robust HIV cascade of care within the same clinical settings, with more than 80% virologically suppressed on cART and despite compelling evidence of the effectiveness of cardiovascular risk modification in reducing cardiovascular morbidity and mortality. Furthermore, the marked demographic and geographic heterogeneity in cardiovascular risk modification has profound implications on equity of care across Europe.

In keeping with other studies, the prevalence of cardiovascular risk was high [7–9] and heterogeneous across demographics and geography [1]. However, the pattern of modifiable cardiovascular risk in this cohort varies considerably from general population estimates. For example, in this cohort the prevalence of smoking and high cholesterol is much higher than the general population, whereas the prevalence of hypertension and obesity is comparable and possibly even lower than the background prevalence [18]. Specifically, in Europe approximately one in three people smokes, compared with one in two in this European cohort of PLHIV. Similarly in the country with the highest prevalence of hypercholesterolemia – Iceland, the general population prevalence is 29%, and the rest of Europe closer to one in five, suggesting that high cholesterol in this cohort was close to double the background prevalence [18]. These differences compared with general population prevalencemay partly be explained by the characteristics of the EuroSIDA cohort; however, other HIV cohorts have also found similar high prevalence of smoking and high cholesterol. Prevalence of hypertension and obesity of one in three is lower than that recorded in the background populations particularly of north and western Europe [18]. The heterogeneous geographical distribution of modifiable risk factors, however, mirrors that of the background populations [18].

The prevalence and incidence of cardiovascular risk was high with one in five having moderate-to-high cardiovascular risk at baseline and 7% per annum developing moderate-to-high cardiovascular risk. This potentially has huge resource implications for European countries. The WHO Europe region is estimated to have 2.2 million PLHIV [19] of which extrapolating from this study over 400 000 potentially have moderate-to-high cardiovascular risk, with another 120 000 developing moderate-to-high cardiovascular risk per annum and would require primary cardiovascular disease prevention in addition to their ongoing HIV care [19].

On a more positive note, being on antihypertensive treatment was associated with a lower probability of developing cardiovascular risk. This mirrors the findings from general population studies. Europe-wide data have suggested that although there have been guideline-driven increases in the prescription of drugs for primary and secondary prevention of cardiovascular disease, there remains marked geographic variation in prescription practices, particularly for vulnerable groups such as injection drug users, which is also reflected in the geographic variation in risk modification that we have seen [18]. Although earlier diagnosis and more aggressive approaches to cardiovascular risk, including smoking cessation and other lifestyle interventions [20], may reduce longer term morbidity and mortality, this has resource implications that health systems squeezed by austerity may not be able to invest in, potentially exacerbating existing inequalities in quality of care across PLHIV and accessing care in Europe.

More than half modified their BP. The health survey of England suggested that only one-third modified their BP, whereas a review of management of diabetes in Europe found the number closer to one in four, albeit of the more stringent target of 130/80 [21]. However, pilot studies in primary care settings in the United Kingdom found that by setting clinical targets, 62% of patients reached the target of 140/90 or less over 6 months of follow-up, suggesting that although HIV clinicians in Europe are doing well, they could do better [22]. The improvement in smoking and BP control over time mirrors the improvement in BP control and smoking cessation in the general population and may also reflect the emphasis and guidance around cardiovascular risk management in HIV that has emerged over time. The sex, age and geographical variation in management of BP and other risk suggest that there are biological, social, cultural and health-service-related factors that need further investigation. For example, to what degree does the poor control in older age reflect the effect of poly-pharmacy and age-related metabolic changes on pharmacodynamics and side effects, resulting in poor adherence or variable drug levels in older populations? Alternatively is this heterogeneity a reflection of the models of care and in particular the varying expertise and experience of managing cardiovascular risk in primary care settings compared with specialist HIV and/or infectious disease units? For example, it is notable that those with a history of CVD, that is those who are eligible for secondary prevention of CVD were two times more likely to modify their lipids, 50% more likely to stop smoking and 20% more likely to reduce their BP, suggesting room to improve primary prevention.

One aspect that we are unable to directly comment on from this study is the relationship between adherence to cART and successful modification of cardiovascular risk factors. On the one hand, those who had undetectable HIV viral loads were more likely to stop smoking, perhaps an indication of engagement in care. On the other, those who successfully managed their HIV, with undetectable viral load and higher CD4+ cell counts, were less able to reduce their cholesterol to recommended levels, perhaps reflecting the effect of some antiretroviral therapy on cholesterol. On the contrary, we are unable to comment on whether clinicians switched cART to manage lipids. Overall HIV-related factors had a less impact on successful reduction in cholesterol than being on a statins.

The limitation of this study should be noted. Although EuroSIDA is one of the larger observational cohorts, cardiovascular clinical events were uncommon and therefore we did not have the power to look at the effect of risk modification on cardiovascular clinical outcomes. The findings are therefore restricted to cardiovascular risk modification, and we are unable to comment as to whether cardiovascular risk modification is as effective in PLHIV as has been observed in general population cohorts, further combinations of cohorts would be required to answer this question. Second, we have presented the DAD risk prediction tool, which is both a conservative estimate of risk (5%, 5-year comparable with a 10%, 10-year Framingham risk) and cardiovascular risk tool, which is not in common clinical use. However, in this cohort both medium and high DAD cardiovascular risk measurements predicted substantial cardiovascular events that were higher than those predicted by Framingham's (unpublished data). We repeated our analyses using the Framingham equation, and the findings were comparable. Third, we were concerned that there would be a risk of channelling bias, with those with higher cardiovascular risk being more likely to have a second cardiovascular risk measure; however, in fact the opposite was true, and they were less likely to have a second cardiovascular risk measurement (unpublished data available from the authors). Fourth, given the numbers of those who had diabetes mellitus, we were unable to look at the risk modification of this important risk factor for cardiovascular disease. However, we note those with diabetes mellitus were 76% less likely to modify their BP, which, given the risk of cerebral vascular events and chronic renal failure in this group, suggests a need to focus on management of BP in PLHIV and diabetes mellitus. Finally, the patients in the EuroSIDA cohort and in these analyses represent a group of patients who are engaged in care, with good HIV-related outcomes, consequently the actual cascade of cardiovascular care is likely to be worse than what is reported in this study.

In conclusion, to sustain the morbidity and mortality benefits of antiretroviral therapy, we need to tackle the intersecting epidemic of HIV and other comorbidities including cardiovascular risk. Although this study suggests that we are improving in recognition and treatment of cardiovascular risk, particularly BP and smoking, we need to do much more, particularly for older people, men and those with diabetes mellitus. We need to better understand the reasons for the geographical variation, tackle inequities in access to care, and advocate for the resources needed to manage cardiovascular risk. It is important to develop innovative models of patient-centred integrated HIV and cardiovascular disease care and evaluate them with both HIV and cardiovascular end points in mind.


Conflicts of interest

There are no conflicts of interest.


1. Nichols M, Townsend N, Scarborough P, Rayner M. Cardiovascular disease in Europe: epidemiological update. European Heart Journal 2013; 34:3028–3034.
2. Lodwick RK, Sabin CA, Porter K, Ledergerber B, van SA, Cozzi-Lepri A, et al. Death rates in HIV-positive antiretroviral-naive patients with CD4 count greater than 350 cells per microL in Europe and North America: a pooled cohort observational study. Lancet 2010; 376:340–345.
3. Ray M, Logan R, Sterne JA, Hernandez-Diaz S, Robins JM, Sabin C, et al. The effect of combined antiretroviral therapy on the overall mortality of HIV-infected individuals. AIDS 2010; 24:123–137.
4. Havlik RJ, Brennan M, Karpiak SE. Comorbidities and depression in older adults with HIV. Sex Health 2011; 8:551–559.
5. Brent RJ, Brennan M, Karpiak SE. Economic evaluations of HIV prevention in rich countries and the need to focus on the aging of the HIV-positive population. Curr Opin HIV AIDS 2010; 5:255–260.
6. Worm SW, Sabin C, Weber R, Reiss P, El-Sadr W, Dabis F, et al. Risk of myocardial infarction in patients with HIV infection exposed to specific individual antiretroviral drugs from the 3 major drug classes: the data collection on adverse events of anti-HIV drugs (D:A:D) study. J Infect Dis 2010; 201:318–330.
7. De WS, Sabin CA, Weber R, Worm SW, Reiss P, Cazanave C, et al. Incidence and risk factors for new-onset diabetes in HIV-infected patients: the data collection on adverse events of anti-HIV drugs (D:A:D) study. Diabetes Care 2008; 31:1224–1229.
8. Sabin CA, Worm SW, Weber R, Reiss P, El-Sadr W, Dabis F, et al. Use of nucleoside reverse transcriptase inhibitors and risk of myocardial infarction in HIV-infected patients enrolled in the D:A:D study: a multicohort collaboration. Lancet 2008; 371:1417–1426.
9. Worm SW, Friis-Moller N, Bruyand M, D’Arminio MA, Rickenbach M, Reiss P, et al. High prevalence of the metabolic syndrome in HIV-infected patients: impact of different definitions of the metabolic syndrome. AIDS 2010; 24:427–435.
10. Beaglehole R, Epping-Jordan J, Patel V, Chopra M, Ebrahim S, Kidd M, et al. Improving the prevention and management of chronic disease in low-income and middle-income countries: a priority for primary healthcare. Lancet 2008; 372:940–949.
11. Van OJ, Ku GM, Bermejo R, Kegels G, Hermann K, Van DW. The growing caseload of chronic life-long conditions calls for a move towards full self-management in low-income countries. Global Health 2011; 7:38.
12. Allotey P, Reidpath DD, Yasin S, Chan CK, de-Graft AA. Rethinking health-care systems: a focus on chronicity. Lancet 2011; 377:450–451.
13. Seddon ME, Marshall MN, Campbell SM, Roland MO. Systematic review of studies of quality of clinical care in general practice in the UK, Australia and New Zealand. Qual Healthcare 2001; 10:152–158.
14. Bodenheimer T, Chen E, Bennett HD. Confronting the growing burden of chronic disease: can the U.S. healthcare workforce do the job?. Health Aff (Millwood) 2009; 28:64–74.
15. Mocroft A, Phillips AN, Gatell J, Horban A, Ledergerber B, Zilmer K, et al. CD4 counts and viral load-specific rates of AIDS, non-AIDS and deaths according to current antiretroviral use. AIDS 2013; 27:907–918.
16. Friis-Moller N, Thiebaut R, Reiss P, Weber R, Monforte AD, De WS, et al. Predicting the risk of cardiovascular disease in HIV-infected patients: the data collection on adverse effects of anti-HIV drugs study. Eur J Cardiovasc Prev Rehabil 2010; 17:491–501.
17. Sabin CA, D’Arminio MA, Friis-Moller N, Weber R, El-Sadr WM, Reiss P, et al. Changes over time in risk factors for cardiovascular disease and use of lipid-lowering drugs in HIV-infected individuals and impact on myocardial infarction. Clin Infect Dis 2008; 46:1101–1110.
18. Nichols M, Townsend N, Luengo-Fernandez R, Leal J, Gray A, Scarborough P, Rayner M. European cardiovascular disease statistics. 2012; Brussels, Sophia Antipolis: European Heart Network, European Society of Cardiology, www:// [accessed 30 July 2014].
19. European Centre for Disease Prevention and Control/WHO Regional Office for Europe. HIV/AIDS surveillance in Europe 2012. 2013; Stockholm: European Centre for Disease Prevention and Control, [accessed 30 July 2014].
20. Rasmussen LD, Helleberg M, May M, Afzal S, Kronborg G, Larsen CS, et al. Myocardial infarction among Danish HIV infected individuals: population attributable fraction associated with smoking. Clin Infect Dis 2015; p.civ013.
21. Sanchez RG, Cuixart CB, Carcia JM, Taomilehto J, Ryden L. Euroheart II project evaluation of diabetes guidelines for cardiac patients Work package 9. 2013. [accessed 4 August 2014].
22. Caulfield M, MacDonald T, Brady A, McCormack T, Poulter N, MacGregor G. Letter to DoH and NICE regarding hypertension targets in the Quality Outcomes Framework. 2014; Leicester, UK: British Hypertension Society, [accessed 1 August 2014].

antiretoviral therapy; cardiovascular risk; cohort; Europe; HIV

Copyright © 2016 Wolters Kluwer Health, Inc.