Abstract
Background: Cardiovascular risk (CVR) assessment helps to identify patients at high CVR. The Framingham CVR score (FRS) is the most widely used methods but may overestimate risk in regions with low incidence of cardiovascular disease. The objective was to compare the 10-year performance of the original and the adapted REGICOR – Framingham CVR functions in HIV-infected individuals.
Methods: We carried out a longitudinal study of HIV-infected patients with CVR evaluation in a hospital in Barcelona between 2003 and 2013. Statistics: Risk probability was calculated using the FRAMINGHAM function and REGICOR adaptation to the Spanish population, and individuals were categorized in three groups (low, 0 < 5%; moderate, 5–10%; and high, >10%). For each risk group, the number of events over 10 years was calculated using the Kaplan–Meier method, and the expected number of events was calculated by multiplying the frequency of participants in the group by the mean of the probabilities from the risk function. We used the X2 goodness-of-fit test to assess agreement between observed and expected.
Results: Six hundred and forty-one patients were followed up for a median of 10.2 years, and 20 ischemic heart events (IHE) were observed. The mean (95% CI) number of IHEs per 1000 person-years was 3.7 (2.06–5.27). The estimates from the Framingham and REGICOR functions were 40 and 14 IHEs, respectively. The estimate from the original Framingham function differed significantly from the observed incidence (p < 0.001), whereas that from the REGICOR-adapted function did not (p = 0.15). In terms of the number of cardiovascular events (38 events observed), the REGICOR function significantly underestimated risk (p = 0.01), whereas the estimate from the Framingham function was similar to observed (p:0.93).
Conclusions: The FRS significantly overestimates risk of IHE events in our HIV-infected patients, while the REGICOR function is a better predictor of these events. In terms of cardiovascular events, the REGICOR function significantly underestimates risk, whereas the FRS is a better estimator. We recommend using CVR scales and adjusting them to the origin of the population being studied.
Keywords:
Introduction
The incidence of cardiovascular disease is lower in Spain than in Northern Europe countries and the United States of America, although it remains the most common cause of morbidity and mortality in this country.Citation1 The incidence of acute myocardial infarction (AMI) in Spain between 1997 and 1998 was 207 and 45 per 100,000 in men and women, respectively.Citation2 These figures remained stable in Spain during the 1990s and have been accompanied by a significant decrease in hospital case-fatality.Citation3 The result is an increase in the number of AMI survivors and, thus, an increase in the number of individuals with coronary heart disease (CHD).Citation4 However, the incidence of Cardiovascular disease (CVD) in Spain is one-half and one-fourth that of Northern, Eastern, and Western European countries, the USA, and other English-speaking countries, even though these countries have a similar prevalence of CV risk factors to that in the Spain. This phenomenon is known as the “southern European paradox.”Citation5,6
The decrease in mortality among HIV-infected patients following the introduction of the highly active antiretroviral treatment (HAART) has allowed this population to age, with a consequent increase in morbidity and mortality due to cardiovascular diseases.Citation7 Several studies have shown a higher incidence of cardiovascular disease in HIV patients than in the uninfected population.Citation8–13 This may due to the effects of the HIV infection itself, which promotes endothelial inflammation and atherosclerosis,Citation14 to the fact that HAART causes dyslipemia, lipohypertrophy, diabetes and insulin resistance, or to other traditional risk factors such as smoking, which is more prevalent in the HIV population.Citation15,16
There is a need to improve the identification in routine practice of persons at high risk, and various algorithms are available for assessing cardiovascular risk (CVR) for primary prevention.Citation1 The Framingham risk score (FRS) is the most widely used, although it is thought to underestimate risk in HIV-infected patients. Some studies have shown that Framingham underestimates risk,Citation17–19 while othersCitation20,21 have shown that it may predict well or overestimate risk. The FRS predicts risk of developing fatal or non-fatal coronary heart disease (CHD) within 10 years in adults.Citation22 There are various calibrations of the FRS, such as the Anderson calibrated scale,Citation23 which was the first to be generalized, and later the WilsonCitation22 and the GrundyCitation24 scales. The utility of FRS has been verified in several studies in different countries, including Spain, and it remains the standard to which new scales are compared.Citation25–27
The D.A.D group recently created an updated global CVD risk equation tailored to HIV-positive persons using routinely collected CVD risk parameters and incorporating markers on immune function (CD4 lymphocyte count), and exposure to antiretroviral therapies. The estimated CVD risk can be used to quantify risk and to guide preventive care.Citation28
The 2013 ACC/AHA Guidelines on the Assessment of CVR estimate 10-year and lifetime risks for atherosclerotic cardiovascular disease (ASCVD), defined as coronary death or non-fatal myocardial infarction, or fatal or non-fatal stroke.Citation29
The REGICOR chart is an adaptation of the Framingham equation to the characteristics of the Spanish population.Citation30 Like the Framingham equation, these functions estimate risk of coronary mortality and morbidity using information about age, gender, total cholesterol, HDL-cholesterol, systolic and diastolic blood pressure and smoking habits, differentiating diabetic from non-diabetic patient.
The consistent overestimation of risk by the original Framingham function in different countries, including Spain,Citation31–36 led the REGICOR researchers (Registre Gironí del Cor [Gerona Heart Registry])Citation27,Citation37 to adapt this tool to the Spanish population in 2003. The result was a function that correctly predicts the rate of coronary events at 5 years in the Spanish population aged 35–74 years.Citation38
The objective of this study was to compare the 10-year performance of the original Framingham algorithm for ischemic heart events and cardiovascular events in HIV-infected patients and the REGICOR adaptation to the Spanish population. In the absence of risk scores that account for immunological factors, our hypothesis is that the REGICOR will be better adapted to the HIV-infected population in Spain because it is better adapted to the Spanish population at large.
Methods
Patients and methods
Study design
We conducted a prospective cohort study based on a cross-sectional descriptive study of the prevalence of CVRFs and CVD, which was used as the starting point for the cohort study.Citation39 This study was carried out at a University Hospital in Barcelona, Spain, and was approved by the local Clinical Research Ethics Committee. All participants were followed up from their baseline date (evaluation from June to December 2003) to a CHD event, CVD, death, or the date of last follow-up (June 2013).
Subject selection
Adult patients managed at the outpatient Infectious Disease Unit at Hospital del Mar were recruited into this study. Clinical data were obtained from an electronic database or electronic/paper medical records. Eligible subjects had HIV infection and were aged 18 years or older, irrespective of antiretroviral therapy. Patients with previous cardiovascular events or who were younger than 35 years were excluded from the comparison of CVR scores. Patients were only included if they completed a minimum of 1 year of follow-up, and patients who were lost to follow up were censored.
Data collection
We collected data on age, gender, HIV disease stage, and HIV exposure (mutually exclusive in the following order: intravenous drug use, men having sex with men, heterosexual). HAART was recorded at baseline and follow-up, and exposure to HAART was coded as a dichotomous variable; the patient was considered to be exposed if they had been treated with the drug for at least 3 months. Weight, height, and waist circumference were measured using standard methods. After the patient had rested seated in a quiet room for 10 min, blood pressure was measured in the left arm with the elbow flexed at heart level by the same physician using a 1042 RIESTER sphygmomanometer (Jungingen, FRG), with diastolic pressures at Korotkoff phase 5 (disappearance of sounds). Three readings were obtained, and the averages of the second and third systolic and diastolic blood pressure readings were used in the analyses. Individuals were considered to have hypertension if their systolic OR diastolic blood pressure ≥140 or ≥90, respectively.
Total cholesterol and triglycerides were determined using enzymatic methods in a Cobas Mira automatic analyzer (Baxter Diagnostics AG, Düdingen, Switzerland). High-density lipoprotein (HDL) cholesterol was measured using separation by precipitation with phosphotungstic acid and magnesium chloride. Low-density lipoprotein cholesterol concentrations were calculated using the Friedewald formula. Glucose was measured using the oxidase method. Subjects were considered to have diabetes if they had fasting plasma glucose >126 mg/dl (7.0 mmol/l) or casual plasma glucose >200 mg/dl (11.1 mmol/l). The nadir CD4 lymphocyte cell count and HIV RNA viral load (Nuclisens Easy Q HIV-1, Biomerieux, Boxtel, The Netherlands) were recorded at the time of the study. All variables were measured at baseline, and blood tests were performed after at least 8 h of fasting.
Assessment of CVR
Two established validated CVR scores were applied to the HIV subjects: the FRS and REGICOR scores.Citation30 The Framingham 10-year CHD risk score computes risk of developing fatal or non-fatal coronary heart disease (CHD) within a 10-year time period in 30- to 74-year-old adults with no known history of CHD, stroke, or peripheral vascular disease. The REGICORCitation30 chart is an adaptation of the Framingham equation validated for the Spanish population aged 35–74.
Cardiovascular events at follow-up were defined as follows. Ischemic heart events: 1. Acute myocardial infarction, fatal or non-fatal if recorded on the hospital discharge report with markers of necrosis, autopsy report, or ICD-9 codes 410–414, 429.9, or 798, or ICD-10 codes I20-I25, I46.1 or R96 on the death certificate. 2. Angina, if the history was compatible, with or without electrocardiographic changes during the episode, plus a positive exercise stress test, scintigraphy or coronary arteriogram. Cerebrovascular events: 3. Stroke, fatal, or non-fatal in the form of a compatible clinical history (focal deficit) with confirmatory computerized tomography or magnetic resonance imaging, an autopsy report or ICD-9 codes 430–434 or 436–438 (excluding 437.4–437.8), or ICD-10 codes I60-I64, I67, I688, or I690-I698 on the death certificate. 4. Peripheral arterial disease in any of the following three situations: clinically compatible intermittent claudication (pain in the lower limbs while walking that improved with rest) plus a diagnostic arteriogram or echo-Doppler and an ankle-brachial index <0.9; pain at rest in the lower limbs not attributable to other causes plus a diagnostic arteriogram or echo-Doppler and an ankle-brachial index <0.9; amputation of the lower limbs or any part of the leg or foot, ischemic ulcers or gangrene of any part of the leg or foot attributable to an ischemic deficit.
The FRSs were calculated at baseline using an algorithm that included the following cardiac risk factors: sex, age, total cholesterol, HDL-cholesterol, low-density lipoprotein (LDL)-cholesterol, smoking status, blood pressure, and diabetes history.Citation22 Subjects for whom we calculated Framingham 10-year CHD risk were classified according to risk scores in three groups: low risk 0–<10%, moderate risk 10–<15%, and high risk >15+%.
Data on the date, time, and specific cause of death were researched and confirmed from hospital records and the municipal registry of Barcelona city. Causes of death were coded using International Classification of Diseases 10th Revision (ICD-10).
Statistical analysis
We computed the mean and standard deviation of continuous variables, and the proportions of categorical variables. We estimated unadjusted hazard ratios to test for differences in sociodemographic variables, co-morbidity, the prevalence of CVR factors, and HIV-specific variables stratified by the incidence of cardiovascular and ischemic events.
Risk was calculated using the REGICOR and FRAMINGHAM functions and categorized in three groups (REGICOR/FRAMINGHAM low cut-points: 0–<5%, 5–<10%,>10%; FRAMINGHAM high cut-points: 0–<10%, 10–<15%, >15%). The NCEP Framingham cut-points are <10, 10–20, and >20%. We chose to use lower cut-points because most of the study population was relatively young, and thus the incidence of CHD was expected to be lower than the general population studied in NCEP Framingham.
For each risk group, the number of events over 10 years were calculated using the Kaplan–Meier method, and the expected number of events was calculated by multiplying the frequency of participants in the group by the mean of the probabilities from the risk function
The AUC describes how the risk probabilities discriminate individuals who will develop the disease from those who will not. This process is performed by making several cut-points at the probabilities and the sensitivity and specificity is then calculated at each cut-point. Since the REGICOR-adapted function is a re-calibration of the original Framingham function, any patient ordered by either function will be in the same percentile. Therefore, the AUC will be the same, regardless of which function is used. We used the X2 goodness-of-fit test to assess agreement between observed and expected.
Results
From a cohort of 850 patients, 748 were able to complete the cardiovascular assessment. After excluding patients who did not meet the inclusion criteria (those with previous cardiovascular events, those aged <35 or >74 years, and those with less than one year of follow-up), 641 patients (all of white ethnicity) were included in the study (mean follow-up 10.2 years). The mean (±SD) age was 45.7 (±9.5) years and most subjects were male (479, 81.65%). The results for CVR factors were: smoking, 298 (54.5%); hypertension, 93 (20.1%); diabetes, 29 (8.8%); mean total cholesterol >200, 268 (49.4%) (Table ).
Table 1 Patient’s characteristics
The baseline median CD4+ T-cell count was 478 cells/μl, and the baseline median nadir CD4+ T-cell count was 199 cells/μl. 80.5% of subjects had an undetectable viral load at baseline.
During a median follow-up of 10.2 years, 38/641 patients (5.93%) had 50 cardiovascular events, of which 20 (3.12%) were IHE, 13 (2.03%) were strokes and 17 (2.61%) were peripheral vascular disease. Twelve percent of patients were lost to follow-up, defined as follow-up for less than 5 years.
Patients with cardiovascular events were significantly older (mean age 47.9 vs. 43.5), more predominantly male (89.5% vs. 79.8%), and more likely to be smokers (63.2% vs. 45.7%), diabetic (13.2% vs. 4.31%), or hypertensive (26.3% vs. 13.8%). We found no significant differences between those with and without events in the other variables, namely cholesterol levels, central obesity, transmission group, immunologic state, or ARV exposure status (Table ).
The mean (95% CI) number of IHEs per 1000 person-years was 3.67 (2.06–5.27), strokes 2.18 (0.95–3.42), and peripheral artery disease 2.37 (1.08–3.66). 564 patients (88%) completed more than 5 years of follow-up.
In terms of the causes of mortality (128 deaths in total), 28 (21.9%) were AIDS-related, 29 due to liver cirrhosis (23.2%), 26 due to non-AIDs-related cancer (20.3%), 13 due to cardiovascular causes (10.1%), and 13 due to other causes (10.1%); the cause of death could not be determined in 19 patients (14.8%). Figure shows the Kaplan–Meier survival analysis, cardiovascular events and ischemic heart events.
Figure 1 Kaplan–Meier survival analysis of death, cardiovascular event. and ischemic heart event.
The observed and the expected number of IHEs according to the coronary risk algorithm are shown in Table . When subjects were assessed according to the FRS, the predicted risk of developing fatal or non-fatal coronary heart disease within 10 years significantly overestimated the true incidence of events. The observed number of cardiovascular events was significantly lower than expected (Figure (A)). In contrast, when individuals are assessed according to the REGICOR risk score, we did not find any significant differences between the observed and expected number of events. (Figure (B))
Table 2 Observed and the expected number of IHEs
Figure 2 Observed and the expected ischemic heart event.
The observed and the expected number of cardiovascular events according to the risk algorithm are shown Table . Applying the FRS and REGICOR risk scores for computing global risk of cardiovascular event, we found that Framingham is an accurate tool (Figure (A)), whereas REGICOR significantly underestimates risk of cardiovascular event (Figure (B)).
Table 3 Observed and the expected number of cardiovascular events
Figure 3 Observed and the expected cardiovascular event.
Discussion
In our cohort of predominantly male HIV-infected patients of white Mediterranean, we observed an IHE rate of 3.7 events per 1000 person-years.
In a US cohort the rate of AMI in HIV patients was 11.13 per 1000 person-years, compared to 6.98 in the non-HIV population, corresponding to a relative risk of 1.75 for the HIV group.Citation12 In another large US cohort, the rate of AMI among HIV-infected varied with increasing age from 0.7 to 13.5 per 1000 person-years.Citation11
The Data Collection on Adverse Events of Anti-HIV Drugs (DAD) studyCitation40 evaluated the duration-related association between ARV therapy and AMI rates among 23,468 HIV patients from 21 countries, and found an incidence of 3.5 events per 1000 person-years, which is very similar to that observed in our cohort. The same study evaluated the association between exposure to 13 ARV drugs and risk of AMI, and reported an overall AMI event rate of 3.2 events per 1000 person-years.Citation41
The profile of CVR factors observed in this study (54.5% smoking, 49.5% dyslipidemia, 20.1% hypertension) adequately reflects the CVR factors of HIV-infected patients in our country. Our results are very similar to those from a cross-sectional study conducted in 10 HIV care units in Spain (54% smoking, 48.6% dyslipidemia, 38.6% hypertension).Citation42 The mean age (45.7 years) and gender ratio were also similar in both studies.
The causes of death in our study were AIDS-related in 21.9%, end- stage liver disease in 23.2%, and cardiovascular disease in 10.1%. The proportion of AIDS-related deaths was lower than in some previous reports (49.5%,Citation43 29%Citation42) but higher than other (16%Citation44). 10.1% of deaths were due to cardiovascular causes, which is higher than in other cohorts such as ATCC (6.5%Citation43). Causes of death vary depending on the time the study was carried out and the proportion of patients with different risk factors for HIV acquisition that are included.
Only 1.7% of patients in our study were categorized at baseline as have >10% CHD risk at 10 years using the REGICOR function compared to 18% risk using the Framingham function. Cross-sectional studies conducted in Spain have compared different scales of CVR in HIV-infected patients,Citation45,46 and have also found that the REGICOR function detects a very small proportion of patients with high risk, while the original Framingham function better identifies these patients. However, most of the events observed occurred in patients with low and moderate risk, 89 and 63% according to the REGICOR and Framingham functions, respectively. REGICOR gave significantly better estimates of the number of ischemic heart events than the original Framingham function in our HIV-infected patients, which somehow reflects the lower risk of coronary disease in the Spanish population.
It has been considered that coronary risk (risk of angina or AMI, fatal or not) is a good approximation to global CVR, but this does not seem to hold true in our setting. Baena et al. found that to estimate global CVR in our setting, coronary risk should be multiplied by 2.1.Citation47 We observed 20 IHE and 38 incident cardiovascular events overall, so it would seem that the global incidence of CVR in our HIV-infected patients is similar to that in the Spanish general population. These findings explain why the REGICOR function, which was developed to estimate risk of ischemic heart disease, underestimates overall CVR, whereas the original Framingham function, which is also used to assess coronary risk, provides a good estimation of global CVR in HIV-infected patients in Spain.
The incidence of CVD in general and of ischemic coronary heart disease in particular is lower in Spain than in other countries,Citation48 and attributable risk is also known to vary between geographical areas. Thus, while dyslipidemia was found to be the most important risk factor for CHD in the USA,Citation22,24,49,50 obesity and especially smoking among men may have a greater impact on the Spanish population.Citation51,52 The “southern European paradox” indicates that classic CVR factors only partly explain CVR, and may not account for the contribution of possible protective factors.
Our study’s limitations include the fact that our population is not representative of HIV-infected patients in the whole of Spain or the Mediterranean area because it is a single-center study. Our population also had a notable percentage of parenteral drug users and patients with psychiatric conditions, which may have influenced the results for cardiovascular events. Notably, the HIV population has changed substantially in the past 10 years, with a marked decrease in parenteral drug users. Another limitation of the study is that while HAART was recorded as a dichotomous variable, exposure or not to the drug, we did not analyze the type of antiretroviral therapy, changes in HAART during follow-up, nor the duration of exposure. Previous studies have shown increased risk of AMI among patients treated with combination antiretroviral therapy (CART) for longer periods,Citation40 particularly those who received protease inhibitors (PIs),Citation53 and those recently exposed to the nucleoside reverse-transcriptase inhibitors (NRTIs) abacavir and didanosine.Citation54 None of the available CV risk scales account for immunological factors or consider other causes of AMI such as cocaine consumption. Due to the small number of events, we were unable to stratify our analyses by sex, which is an important limitation, although the most important limitation was the low number of events detected overall, probably because of the lower incidence of CV in the Spanish population. Some patients were lost follow-up and some had an unknown cause of death, so the total number of events may be underestimated. Some individuals died of causes other than ischemic heart disease, and this competitive risk may explain the relatively low incidence of cardiovascular events.
The strong points of our study include the fact that our sample was followed and treated by the same medical team, hence diminishing intercenter variability.
The Framingham risk function significantly overestimated ischemic heart events in our Mediterranean HIV-infected patients and the REGICOR adapted function was a better predictor of ischemic heart events. Applying the same scales to evaluate risk of cardiovascular events, the REGICOR function significantly underestimated the risk, whereas the Framingham risk was a better estimator.
Further studies with larger numbers of patients are needed to clarify which function is best for assessing CVR in HIV-infected patients. The function should include specific HIV-related factors (like CD4 T-cell count, viral load, and inflammatory markers) and HAART exposure in a specific environment. The practical result of this study is that, in the absence of a specific function, both the REGICOR and Framingham scales should be applied, as they appear to provide complementary information.
Disclosure statement
No potential conflict of interest was reported by the authors.No potential conflict of interest was reported by the authors.
References
- Comín E, Solanas P, Cabezas C, et al. Estimating cardiovascular risk in Spain using different algorithms. Rev española Cardiol [Internet]. 2007;60(7):693–702. http://www.ncbi.nlm.nih.gov/pubmed/1766385310.1157/13108274
- Marrugat J, Elosua R, Aldasoro E, et al. Regional variability in population acute myocardial infarction cumulative incidence and mortality rates in Spain 1997 and 1998. Eur J Epidemiol. [Internet] 2004 [cited March 14, 2016];19(9):831–839. http://www.ncbi.nlm.nih.gov/pubmed/15499893
- Gil M, Martí H, Elosúa R, et al. Analysis of trends in myocardial infarction case-fatality, incidence and mortality rates in Girona, Spain, 1990–1999. Rev española Cardiol. [Internet] 2007 [cited March 14, 2016];60(4):349–356. http://www.ncbi.nlm.nih.gov/pubmed/1752154310.1157/13101638
- Marrugat J, Elosua R, Martí H.. Epidemiology of ischaemic heart disease in Spain: estimation of the number of cases and trends from 1997 to 2005. Rev española Cardiol [Internet]. 2002 [cited March 14, 2016];55(4):337–346.http://www.ncbi.nlm.nih.gov/pubmed/1197589910.1016/S0300-8932(02)76611-6
- Masia R, Pena A, Marrugat J, et al. High prevalence of cardiovascular risk factors in Gerona, Spain, a province with low myocardial infarction incidence. REGICOR Investigators. J Epidemiol Community Health [Internet]. 1998 [cited March 14, 2016];52(11):707–715.http://www.ncbi.nlm.nih.gov/pubmed/1039650310.1136/jech.52.11.707
- Grau M, Subirana I, Elosua R, et al. Trends in cardiovascular risk factor prevalence (1995–2000–2005) in northeastern Spain. Eur J Cardiovasc Prev Rehabil [Internet]. 2007 [cited March 14, 2016];14(5):653–659.http://www.ncbi.nlm.nih.gov/pubmed/1792562410.1097/HJR.0b013e3281764429
- Lewden C, Salmon D, Morlat P, et al. Causes of death among human immunodeficiency virus (HIV)-infected adults in the era of potent antiretroviral therapy: emerging role of hepatitis and cancers, persistent role of AIDS. Int J Epidemiol. 2005;34(1):121–130.
- Althoff KN, McGinnis KA, Wyatt CM, et al. Comparison of risk and age at diagnosis of myocardial infarction, end-stage renal disease, and non-AIDS-defining cancer in HIV-infected versus uninfected adults. Clin Infect Dis. 2015;60(4):627–638.10.1093/cid/ciu869
- Womack JA, Chang C-CH, So-Armah KA, et al. HIV infection and cardiovascular disease in women. J Am Heart Assoc [Internet]. 2014;3(5):e001035. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=4323817&tool=pmcentrez&rendertype=abstract10.1161/JAHA.114.001035
- Schouten J, Wit FW, Stolte IG, et al. Cross-sectional Comparison of the prevalence of age-associated comorbidities and their risk factors between HIV-infected and uninfected individuals: the AGEhIV cohort study. Clin Infect Dis. 2014;59(12):1787–1797.10.1093/cid/ciu701
- Freiberg MS, Chang C-CH, Kuller LH, et al. HIV Infection and the risk of acute myocardial infarction. JAMA Intern Med [Internet]. 2013;173(8):614–622. http://www.ncbi.nlm.nih.gov/pubmed/2345986310.1001/jamainternmed.2013.3728
- 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 [Internet]. 2007;92(7):2506–2512. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2763385&tool=pmcentrez&rendertype=abstract10.1210/jc.2006-2190
- Smith CJ, Ryom L, Weber R, et al. Trends in underlying causes of death in people with HIV from 1999 to 2011 (D:A:D): a multicohort collaboration. Lancet (London, England) [Internet]. 2014 [cited March 15, 2016];384(9939):241–248.http://www.ncbi.nlm.nih.gov/pubmed/2504223410.1016/S0140-6736(14)60604-8
- El-Sadr W, Lundgren J, Neaton J, et al. CD4+count-guided interruption of antiretroviral treatment. N Engl J Med [Internet]. 2006;355(22):2283–2296. http://discovery.ucl.ac.uk/6514/
- Grinspoon SK. Cardiovascular disease in HIV: traditional and nontraditional risk factors. Top Antivir Med. 2014;22(4):676–679.
- Hemkens LG, Bucher HC. HIV infection and cardiovascular disease. Eur Heart J [Internet]. 2014;35(21):1373–1381. http://eurheartj.oxfordjournals.org/cgi/doi/10.1093/eurheartj/eht52810.1093/eurheartj/eht528
- Fiscella K, Tancredi D, Franks P. Adding socioeconomic status to Framingham scoring to reduce disparities in coronary risk assessment. Am Heart J [Internet]. 2009 [cited March 15, 2016];157(6):988–994.http://www.ncbi.nlm.nih.gov/pubmed/1946440810.1016/j.ahj.2009.03.019
- Brindle PM, McConnachie A, Upton MN, Hart CL, Davey Smith G, Watt GCM. The accuracy of the Framingham risk-score in different socioeconomic groups: a prospective study. Br J Gen Pract [Internet]. 2005 [cited 2016 Mar 15]; 55(520):838–845.http://www.ncbi.nlm.nih.gov/pubmed/16281999
- Brindle P, Emberson J, Lampe F, et al. Predictive accuracy of the Framingham coronary risk score in British men: prospective cohort study. BMJ [Internet]. 2003 [cited March 15, 2016];327(7426):1267.http://www.ncbi.nlm.nih.gov/pubmed/1464497110.1136/bmj.327.7426.1267
- Parra S, Coll B, Aragonés G, et al. Nonconcordance between subclinical atherosclerosis and the calculated Framingham risk score in HIV-infected patients: relationships with serum markers of oxidation and inflammation. HIV Med [Internet]. 2010 [cited March 15, 2016];11(4):225–231.http://www.ncbi.nlm.nih.gov/pubmed/1984579210.1111/hiv.2010.11.issue-4
- 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 [Internet]. 2011 [cited March 15, 2016];16(1):1–8.http://www.ncbi.nlm.nih.gov/pubmed/2131110310.3851/IMP1682
- Wilson PWF, D’Agostino RB, Levy D, Belanger AM, Silbershatz H, Kannel WB. Prediction of coronary heart disease using risk factor categories. Circulation [Internet]. 1998;97(18):1837–1847. http://circ.ahajournals.org/content/97/18/1837.abstract10.1161/01.CIR.97.18.1837
- Anderson KM, Wilson PW, Odell PM, Kannel WB. An updated coronary risk profile. A statement for health professionals. Circulation. 1991;83(1):356–362.
- Grundy SM, Pasternak R, Greenland P, Smith S Jr., Fuster V. AHA/ACC scientific statement assessment of cardiovascular risk by use of multiple-risk-factor assessment equations association and the American college of cardiology. Circulation. 1999;1097(99):1348–1359.
- Giampaoli S, Palmieri L, Mattiello A, Panico S. Definition of high risk individuals to optimise strategies for primary prevention of cardiovascular diseases. Nutr Metab Cardiovasc Dis [Internet]. 2005;15(1):79–85. http://www.ncbi.nlm.nih.gov/entrez/query.fcgi?cmd=Retrieve&db=PubMed&dopt=Citation&list_uids=1587185510.1016/j.numecd.2004.12.001
- Wilson PW. Framingham and European risk algorithms: implications for African Americans. Rev Cardiovasc Med. 2004;5(Suppl 3):S34–S41.
- Marrugat J, Solanas P, D’Agostino R, et al. Estimación del riesgo coronario en España mediante la ecuación de Framingham calibrada [Coronary risk estimation in Spain using a calibrated Framingham function]. Rev Española Cardiol [Internet]. 2003;56(3):253–261. http://www.sciencedirect.com/science/article/pii/S030089320376861410.1016/S0300-8932(03)76861-4
- Friis-Moller N, Ryom L, Smith C, et al. An updated prediction model of the global risk of cardiovascular disease in HIV-positive persons; the data-collection on adverse effects of anti-HIV drugs (D:A:D) study. Antivir Ther [Internet]. 2013;18:A23–A24. http://www.intmedpress.com/serveFile.cfm?sUID=bc75b62c-c206-437a-b725-a75baa67f5b2\nhttp://www.embase.com/search/results?subaction=viewrecord&from=export&id=L71281003\nhttp://sfx.library.uu.nl/utrecht?sid=EMBASE&issn=13596535&id=doi:&atitle=An+updated+pre
- Goff DC, Lloyd-Jones DM, Bennett G, et al. 2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American college of cardiology/American heart association task force on practice guidelines. Circulation [Internet]. 2014;129(25 suppl 2):S49–S73.http://circ.ahajournals.org/cgi/doi/10.1161/01.cir.0000437741.48606.9810.1161/01.cir.0000437741.48606.98
- Marrugat J, Solanas P, D’Agostino R, et al. Coronary risk estimation in Spain using a calibrated Framingham function. Rev española Cardiol [Internet]. 2003;56(3):253–261. http://www.ncbi.nlm.nih.gov/pubmed/1262295510.1016/S0300-8932(03)76861-4
- Hense H-W, Schulte H, Löwel H, Assmann G, Keil U. Framingham risk function overestimates risk of coronary heart disease in men and women from Germany–results from the MONICA Augsburg and the PROCAM cohorts. Eur Heart J [Internet]. 2003 [cited March 14, 2016];24(10):937–945.http://www.ncbi.nlm.nih.gov/pubmed/1271402510.1016/S0195-668X(03)00081-2
- Empana JP, Ducimetière P, Arveiler D, et al. Are the Framingham and PROCAM coronary heart disease risk functions applicable to different European populations? The PRIME study. Eur Heart J [Internet]. 2003 [cited March 14, 2016];24(21):1903–1911.http://www.ncbi.nlm.nih.gov/pubmed/1458524810.1016/j.ehj.2003.09.002
- Thomsen TF, McGee D, Davidsen M, Jørgensen T A cross-validation of risk-scores for coronary heart disease mortality based on data from the Glostrup population studies and Framingham heart study. Int J Epidemiol [Internet]. 2002 [cited March 14, 2016];31(4):817–822.http://www.ncbi.nlm.nih.gov/pubmed/1217702810.1093/ije/31.4.817
- Menotti A, Lanti M, Puddu PE, Kromhout D. Coronary heart disease incidence in northern and southern European populations: a reanalysis of the seven countries study for a European coronary risk chart. Heart [Internet]. 2000 [cited March 14, 2016];84(3):238–244.http://www.ncbi.nlm.nih.gov/pubmed/1095628110.1136/heart.84.3.238
- Liu J, Hong Y, D’Agostino RB, et al. Predictive value for the Chinese population of the Framingham CHD risk assessment tool compared with the Chinese multi-provincial cohort study. JAMA [Internet]. 2004 [cited March 14, 2016];291(21):2591–2599.http://www.ncbi.nlm.nih.gov/pubmed/1517315010.1001/jama.291.21.2591
- Ramos R, Solanas P, Cordón F, et al. Comparación de la función de Framingham original y la calibrada del REGICOR en la predicción del riesgo coronario poblacional [Comparison of population coronary heart disease risk estimated by the Framingham original and REGICOR calibrated functions]. Med Clin (Barc). 2003;121(14):521–526.10.1016/S0025-7753(03)74007-X
- Marrugat J. An adaptation of the Framingham coronary heart disease risk function to European Mediterranean areas. J Epidemiol Community Heal [Internet]. 2003;57(8):634–638. http://jech.bmj.com/content/57/8/634.full10.1136/jech.57.8.634
- Marrugat J, Subirana I, Comin E, et al. Validity of an adaptation of the Framingham cardiovascular risk function: the VERIFICA study. J Epidemiol Community Health [Internet]. 2007 [cited March 14, 2016];61(1):40–47.http://www.ncbi.nlm.nih.gov/pubmed/1718301410.1136/jech.2005.038505
- Knobel H, Jericó C, Montero M, et al. Global cardiovascular risk in patients with HIV infection: concordance and differences in estimates according to three risk equations (Framingham, SCORE, and PROCAM). AIDS PatientCare STDS [Internet]. 2007;21(7):452–457. http://www.ncbi.nlm.nih.gov/pubmed/1765102610.1089/apc.2006.0165
- Friis-Moller N, Sabin CA, Weber R, et al. Combination antiretroviral therapy and the risk of myocardial infarction: the data collection on adverse events of anti-HIV drugs (DAD) study group. N Engl J Med. 2003;349:1993–2003.
- Worm SW, Sabin C, Weber R, 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(3):318–330.10.1086/649521
- Salmon-Ceron D, Lewden C, Morlat P, et al. Liver disease as a major cause of death among HIV infected patients: role of hepatitis C and B viruses and alcohol. J Hepatol. 2005;42(6):799–805.10.1016/j.jhep.2005.01.022
- Article M. 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):1387–1396.
- Weber R, Ruppik M, Rickenbach M, et al. Decreasing mortality and changing patterns of causes of death in the Swiss HIV cohort study. HIV Med. 2013;14(4):195–207.10.1111/hiv.2013.14.issue-4
- Masiá M, Pérez-Cachafeiro S, Leyes M, et al. Cardiovascular risk in human immunodeficiency virus-infected patients in Spain. CoRIS cohort, 2011. Enferm Infecc Microbiol Clin [Internet]. 2012; 30(9):517–527. http://www.ncbi.nlm.nih.gov/pubmed/2255165310.1016/j.eimc.2012.02.014
- García-Lázaro M, Rivero Román A, Camacho Espejo Á, et al. Variability in coronary risk assessment in HIV-infected patients. Med Clin (Barc) [Internet]. 2007; 129(14):521–524. http://www.ncbi.nlm.nih.gov/pubmed/1798352910.1157/13111418
- Baena-Díez JM, García-Lareo M, de la Poza-Abad M, Hernández-Ibáñez R, Muñoz-Rubio A, García-Rey Z. Estimación del riesgo cardiovascular global a partir del riesgo coronario. Estudio de cohortes. Med clÝn (Ed impr) [Internet]. 2006; 127(1):8–10.http://pesquisa.bvsalud.org/portal/resource/es/ibc-046347
- Sans S, Kesteloot H, Kromhout D. The burden of cardiovascular diseases mortality in Europe. Task force of the European society of cardiology on cardiovascular mortality and morbidity statistics in Europe. Eur Heart J [Internet]. 1997 [cited March 14, 2016];18(8):1231–1248.http://www.ncbi.nlm.nih.gov/pubmed/945841510.1093/oxfordjournals.eurheartj.a015434
- Levy D, Wilson PW, Anderson KM, Castelli WP. Stratifying the patient at risk from coronary disease: new insights from the Framingham heart study. Am Heart J [Internet]. 1990 [cited March 15, 2016]; 119(3):712–717; discussion 717. http://www.ncbi.nlm.nih.gov/pubmed/213796010.1016/S0002-8703(05)80050-X
- Kannel WB, D’Agostino RB, Sullivan L, Wilson PWF. Concept and usefulness of cardiovascular risk profiles. Am Heart J. 2004:16–26.10.1016/j.ahj.2003.10.022
- Grau M, Subirana I, Elosua R, et al. Why should population attributable fractions be periodically recalculated?. Prev Med (Baltim) [Internet]. 2010 [cited March 15, 2016];51(1):78–84.http://www.ncbi.nlm.nih.gov/pubmed/2036261010.1016/j.ypmed.2010.03.012
- Medrano MJ, Pastor-Barriuso R, Boix R, et al. Coronary disease risk attributable to cardiovascular risk factors in the Spanish population. Rev española Cardiol [Internet]. 2007 [cited March 15, 2016];60(12):1250–1256. http://www.ncbi.nlm.nih.gov/pubmed/1808209010.1157/13113930
- Friis-Møller N, Reiss P, Sabin CA, et al. Class of antiretroviral drugs and the risk of myocardial infarction. N Engl J Med [Internet]. 2007;356(17):1723–1735. http://www.ncbi.nlm.nih.gov/pubmed/17460226\nhttp://www.nejm.org/doi/abs/10.1056/NEJMoa062744
- Sabin CA, Worm SW, Weber R, 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 multi-cohort collaboration. Lancet [Internet]. 2008;371(9622):1417–1426. http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid=2688660&tool=pmcentrez&rendertype=abstract