HIV and heart disease: D:A:D prevalence rates show importance of accessing CVD risk in HIV patients

Simon Collins, HIV i-Base

The objective of the EMEA initiated and pharmaceutical sponsored D:A:D study (Data on collection of Adverse events of anti-HIV Drugs) is to describe the prevalence of risk factors for cardiovascular disease (CVD) in HIV-patients and to investigate whether antiretroviral drugs or HIV itself are associated with an increased risk of CVD. Results from both aspects of the study are critical to management of patients using HAART.

This is an ongoing study and the results from data collected to August 2002 were presented at the 10th CROI (Conference on Retroviruses and Opportunistic Infections) in Boston earlier this year and were reported in the March issue of HTB. From almost 36,500 patient-years the study identified 126 cases of myocardial infarction (90% of cases were men, median age 48 years), 25% of which were fatal. In a multivariate analysis these results indicated a cumulative relative risk of myocardial infarction of 1.26 per year of HAART use. [1, 2]

The focus of this article is the results from the analysis of baseline data for the prevalence of risk factors for the cohort, published in 23 May issue of AIDS. [3] This coincided with distribution of a 175-page Supplement from AIDS focusing on HIV-associated CVD in the HAART era. [4] The results have serious implications for management of patient care in the UK, especially coupled with the knowledge that this has translated into increased clinical events.

The D:A:D study is an observational study involving 11 cohorts and includes data from more than 20,000 patients at 188 clinics mainly in Europe but including one cohort each from USA and Australia. These cohorts provide a very broad patient group across diverse national differences. Median age is 39, with 24% women (range: 3% Australia to 31% Italy), 42% homosexually acquired HIV (range: 20% in Italy to 87% in Australia) and 23% IVDU (range: 4% Australia to 40% Spain). Baseline data from this analysis includes almost 18,000 patients from nine of the cohorts.

At enrolment and every eight months thereafter, completed patient forms based on patient interview, case notes and physical exam are completed. In addition to HIV-related health results (CD4, viral load, use of treatment, total cholesterol, HDL cholesterol, triglycerides), this includes age, patient and family history of CVD, smoking status, BMI, blood pressure, diabetes, lipid-lowering and hypertensive therapy and clinical signs of lipodystrophy. Use of ARVs is recorded as naïve, ever-used and current, by RTI only and by PI and NNRTI use.

Results: prevalence of risk factors

At enrolment 13% of study population were ARV-naïve, 6% were previously exposed but not currently on treatment. Eleven percent were on RTI-only, 20% on NNRTI, 43% on PI (>70% ever used) and 7% included all three classes.

Almost 25% were in an age group constituting a study defined risk factor (>45 men, >55 women). Eleven percent had a family history of heart disease (first degree relative with MI when <50 years old) and 1.4% had a previous history of CVD.

Twenty-two percent of subjects had total cholesterol levels >6.2mmol/l which was associated with PI, NNRTI or PI+NNRTI. People who discontinued treatment, whatever their previous treatment history, had similar cholesterol levels to treatment naïve individuals, suggesting reversibility. Risk for elevated cholesterol increased with CD4 count (24% per two-fold increase in CD4).

Triglycerides were elevated in 28% fasted samples and 35% unfasted samples but 40% samples lacked information about fasting status. Risk of cumulative exposure by drug class for elevated triglycerides was 1.05, 1.28, 1.38 in NRTI, NNRTI and PI use respectively (in univariate and multivariate analysis). For people not on treatment only, risk of increased triglycerides correlated with viral load.

Compared to treatment naïve individuals, all regimens were associated with risk of low HDL cholesterol, except those containing an NNRTI, with risk for decreased HDL being highest in patients with low CD4 or high viral load.

Twenty-five percent of patients were recorded as having lipodystrophy, recognisable by physician assessment of either fat loss, fat accumulation or a combination of both and this data was included in 99% of submissions. As expected this was associated with both ARV use and exposure, and also in multivariate analysis with elevated total cholesterol (OR 1.56), elevated triglycerides (OR 2.16) and decreased HDL (OR 1.56).

Although more than 8% of the study population had hypertension, this correlated with age, sex and BMI and not antiretroviral use, but it highlights the importance of considering hypertension as an additional health concern for cohorts of HIV-positive patients. However, lipodystrophy and diabetes were both associated with hypertension (OR 1.34 and 2.05 respectively). Overall prevalence of diabetes was 2.5%.

Taken together these results highlight the growing range of health concerns that will impact on the health of HIV-positive patients and their primary providers of healthcare (which are increasingly their HIV-treating physicians).

While there are limitations from the observational design and cross sectional nature of this analysis – these results present associations and do not show causality – this is still by far the largest study of its kind to prospectively look at both events and the wide range of contributory risk factors.

In the discussion the authors refer to the limitations from data collection and incidence on missing data current smoker (18%), family history (37%), total cholesterol (18%), HDL cholesterol (54%) and triglyceride (18%). Although there was an uneven distribution of missing data from cohorts this was accounted for in the analysis. Information on other important risk factors such as diet, exercise, alcohol and genetic factors were not included, but by their nature would also be unlikely to standardise with such large patient numbers and international differences.

It is impressive that this study was initiated at a European level as a safety issue, and that it has been funded collectively by all the major manufacturers of antiretroviral drugs. It is also very reassuring that these companies have committed continued funds to extend the data collection for at least a further two years.

Full text from the AIDS article is available at:
http://www.natap.org/2003/may/052803_3.htm

References:

Friis-Møller N, Weber R, D’Arminio Monforte A et al – Exposure to HAART Is associated with an increased risk of myocardial infarction: The D:A:D Study. 10th Conference on Retroviruses and Opportunistic Infections. Abstract 130.
http://www.retroconference.org/2003/cd/Abstract/130.htm
Aberg J – HAART to HEART: cardiovascular risk in HIV. HIV Treatment Bulletin Vol4 No3 April 2003.
https://i-base.info/htb/10979
Friis-Møller N, Weber R, Reiss P – Cardiovascular disease risk factors in HIV patients – association with antiretroviral therapy. Results from the D:A:D study. AIDS 2003; 17(8):1179-1193.
Murphy RL, Stein JH et al – Clinical and biological insights in HIV-associated cardiovascular diseases in the HAART era. AIDS Vol17 Supplement 1 April 2003.
Bozzette S, Ake CF, Tam H et al – Cardiovascular and cerebrovascular events in patients treated for Human Immunodeficiency Virus infection. NEJM 348:702-710.
Kuritzkes DR, Currier J – Cardiovascular risk factors and antiretroviral therapy. NEJM 348:679-680.

Comment

One of the controversial aspects of the results is that these findings are in contrast to a retrospective analysis of the US Veterans Association database of over 36,000 patients, which was presented at Retrovirus Conference in 2002 and which had found no increase in reports of cardiovascular events between pre-HAART years and in the years that HAART has been widely available (January 1993-June 2001).

Presentation of the D:A:D study at Retrovirus this year overlapped with the publication of the VA study with commentary in 20 February edition of the NEJM. [5, 6]

Neither D:A:D nor VA database studies are perfect but the increased risk shown in the European study is already reflected in treatment guidelines that include making an assessment of risk factors for CVD prior to starting patients on antiretroviral therapy.