Cardiovascular Disease and Cardiovascular Disease Risk in HIV-Positive Populations in the Asian Region

Introduction: Cardiovascular diseases (CVD) are becoming more prevalent in HIV-infected populations as they age largely due to improved treatment outcomes. Assessment of CVD risk and CVD risk factors in HIV-positive populations has focused on high income settings, while there are limited studies evaluating CVD in HIV-positive populations in the Asian region. Materials and Methods: We provided an overview of the prevalence and incidence of CVD and its risk factors in adult HIV-positive populations, and of the strategies currently in place for CVD management in the Asian region. Results: Studies from the Asian region showed that CVD and CVD risk factors, such as dyslipidaemia, elevated blood glucose, obesity and smoking, are highly prevalent in HIV-positive populations. A number of studies suggested that HIV infection and antiretroviral therapy may contribute to increased CVD risk. National HIV treatment guidelines provide some directions regarding CVD risk prevention and management in the HIV-infected population, however, they are limited in number and scope. Conclusion: Development and consolidation of guidelines for integrated CVD and HIV care are essential to control the burden of CVD in HIV-positive populations. To inform guidelines, policies and practice in the Asian region, research should focus on exploring appropriate CVD risk screening strategies and estimating current and future CVD mortality and morbidity rates.


INTRODUCTION
Due to largely improved antiretroviral therapy (ART) outcomes, HIV positive individuals are growing older and as a result chronic diseases associated with aging, including cardiovascular diseases (CVD) and cancer, are becoming more prevalent in the HIV-infected population [1]. A recent review of data from high income countries estimated that HIV populations have around a twofold increased risk of experiencing CVD events compared to the general population [2].
Similar to risk factors in the general population, CVD in the HIV-infected population can be caused by nonmodifiable risk factors including older age and family history of CVD, and modifiable risk factors such as medical conditions (e.g. hypertension, diabetes, dyslipidaemia) and behavioural factors (e.g. smoking and diet). Furthermore, the increased risk of CVD might also be attributable to risk factors specific to the HIV-positive population, such as compromised immunity [3,4], immune activation, HIV associated inflammation [5], and receiving ART [6,7].
Assessment of CVD risk and CVD risk factors in HIV populations has focused on high income settings, while there are limited studies evaluating CVD in HIV populations in the Asian region, where income levels greatly differ across countries [8]. This review aims to provide insight on the prevalence and incidence of CVD and its risk factors in adult HIV populations in the Asian region.
We defined CVD as stroke, myocardial infarction and coronary heart disease (CHD) and we considered blood lipid levels, diabetes and blood glucose levels, abdominal obesity, hypertension and smoking as CVD risk factors. We report research findings from the last 10 years of studies conducted in China, India, Indonesia, Japan, Malaysia, South Korea, Taiwan, Thailand, Cambodia, Myanmar, the Philippines and Vietnam.
When required, differences were reported between HIV negative, ART naïve, and ART experienced populations. In addition, we provide an overview of the strategies currently in place for CVD management in HIV-populations in the Asian region. We conclude with future directions for practice and research.

CARDIOVASCULAR DISEASE AND CARDIAC FUNCTIONING
Few studies have evaluated cardiovascular health in HIV-positive populations in the Asian region. Table 1 provides an overview of studies evaluating CVD events and surrogate markers of cardiovascular dysfunction. One recent Taiwanese study assessed the incidence of stroke in patients with HIV compared with non-infected individuals from the Taiwan National Health Insurance Research Database [9]. The study showed that stroke occurred more frequently in those with HIV (0.96% vs. 0.24%).
In fact, independent of other traditional CVD risk factors, HIV infection was associated with a 2.75 fold increased risk (95% Confidence Interval [CI] 1.17-6.44) of having a stroke [9]. Further evidence of CVD risk stems from studies investigating surrogate measures of CVD, such as carotid artery plaque and carotid intima-media thickness. A study from South Korea found that 23.4% of HIV patients receiving ART had carotid artery plaque and that ART use was significantly associated with the existence of carotid artery plaque [10]. In a small Indian study population, no significant difference was found in carotid intima-media thickness between ART experienced and ART naïve patients [11].
Several studies in the Asian region have used two-dimensional echocardiogram to assess cardiac function in HIVpositive populations. In India, echocardiographic abnormalities were identified in 67.0% of HIV-infected patients [12]. Diastolic dysfunction (DD) was the most common abnormality occurring in 48.2% of patients, followed by left ventricular systolic dysfunction occurring in 17.6% of patients.
Similar proportions of DD were found in Chinese patients on ART, which was substantially lower compared to proportions of DD (20%) detected in HIV negative individuals [13]. A Chinese longitudinal study found that cardiac abnormalities, including left ventricular systolic dysfunction, DD, increased left ventricular mass index and pulmonary arterial hypertension, were more prevalent 48 weeks after ART initiation, although this finding was only significant for DD [14]. This study also showed HIV infection to be independently associated with DD and increased left ventricular mass index.

ESTIMATES OF CARDIOVASCULAR DISEASE RISK
Additional information on CVD risk among HIV-positive populations is provided by studies that estimate CVD risk using various risk assessment systems, such as the Rama-EGAT risk score, the Framingham risk score (FRS) and the D:A:D risk score. Risk scores of ≥10% are commonly considered as having high risk of developing CVD in the future. One study investigated various CVD risk outcomes in a multi-country analysis and showed that high 5-year risk of CVD, CHD and myocardial infarction was prevalent in 16, 11 and 6% of the treated population, respectively [15].
In South Korea, 10-year CVD risk was compared among HIV positive and negative patients using the FRS [16]. High CVD risk was more common in HIV positive than in HIV negative individuals (29.3 vs. 19.5%), although this difference was not significant.
Using the Rama-EGAT risk score, one study estimated that 11% of the HIV patients from a clinic in Bangkok had a high 10-year risk of CVD events [17]. Another Thai study estimated the 10-year risk for Coronary Heart Disease (CHD) with each of the scoring systems and found that the prevalence of high 10-year CHD risk was 0.8%, 2.1% and 9.9% as calculated by the D:A:D, Rama-EGAT and FRS, respectively [18]. Table 2 provides an overview of the studies on modifiable risk factors of CVD and HIV.

Blood Lipid Profiles
Blood lipid profiles are generally considered abnormal when there are high levels of triglycerides (≥150 mg/dL), total cholesterol (TC ≥200 mg/dL), low-density lipoprotein cholesterol (LDL ≥130 mg/dL) and/or low levels of HDL (<40 mg/dL).
Abnormalities in one or more of these blood lipids are commonly referred to as dyslipidaemia. Several studies from the Asian region showed that dyslipidaemia was less prevalent in patients who were first presented with HIV compared to their HIV negative counterparts [19,20]. Cholesterol levels were on average lower in HIV positive patients [19 -22], whereas findings were inconsistent regarding triglyceride levels [19 -21].
Moreover, a Thai study showed that the prevalence of abnormal lipid levels was even higher in those on PIcontaining ART regimens [28]. Studies from high, upper-middle and lower-middle income groups all confirmed significant associations between ART use and unfavourable lipid levels [16,19,25,26,29,30].

Diabetes and Abnormal Blood Glucose Levels
Information regarding glucose levels has frequently been described in studies from India. In a study comparing glucose levels among ART experienced patients, ART naïve patients and HIV negative individuals, hyperglycaemia (fasting plasma glucose ≥100 mg/dL) was observed in 34, 21 and 31% of the participants, respectively [19]. In contrast, in a study evaluating ART naïve and experienced patients, no significant difference in hyperglycaemia prevalence was found between the two patient groups [24]. Interestingly, one study showed a significant increase in blood glucose levels in HIV patients after they started ART [29]. In fact, after six months of ART, 4.2% of patients were newly diagnosed with impaired fasting glucose (fasting plasma glucose 100-125 mg/dL), while an additional 2.1% of patients developed diabetes (fasting plasma glucose >125 mg/dL). A cohort study evaluated blood glucose levels in tuberculosis co-infected HIV patients [30]. Twelve months after ART initiation, elevated blood glucose levels (>110 mg/dL) had developed in 11% of patients. However, compared to before ART initiation there was no significant difference in the prevalence of abnormal blood glucose levels, since abnormalities did not persist in all baseline cases [30]. Two studies evaluated insulin resistance and found it more frequent in those on ART than in those who had not yet started treatment, albeit not significantly [19,31].
Additional information on diabetes and blood glucose levels was available from other Asian countries. In China, blood glucose levels were higher in HIV positive than HIV negative men [22]. Furthermore, 20% of Chinese ARTnaïve patients had elevated blood glucose levels, of whom 11% met the criteria for diabetes [32]. Other findings from China showed that diabetes and impaired fasting glucose developed relatively frequently in patients after they initiated ART (incidence of 2.62 and 35.64 per 100 person-years, respectively) [33]. Substantially, lower diabetes incidence rates were found in ART experienced patients in Thailand (5.0 per 1000 person-years) [34] and in high-income country Taiwan (13.1 per 1000 person-years) [35]. In Thailand, the diabetes prevalence was 5.4% in a study population 10 years after HIV diagnosis [27] and (depending on the type of regimen) ranged from 1.7-6.6% in another ART treated study population [28], while impaired fasting glucose was seen in 28% of a non-diabetic population of whom the majority was on ART [36]. In Malaysian patients on ART, the prevalence of diabetes and hyperglycaemia was 13 and 38%, respectively [23]. A lower prevalence of hyperglycaemia was reported among HIV infected patients assessed in Thailand, with the proportion in ART experienced patients being higher than in ART naïve patients (24 vs. 12%, p <0.05) [25]. Lastly, an analysis on the wider Asian region showed a mean diabetes prevalence of 8.3% [15].

Hypertension
There is inconclusive evidence from the Asian region regarding hypertension and differences across HIV or ART status. For example, no significant differences were found between HIV negative and positive men in China [22]. In addition, ART naïve patients in India had a relatively low prevalence of hypertension (12%) compared to HIV negative individuals (36%) and ART experienced patients (17%, p=0.002) [19]. Other findings from India indicate no difference with regard to hypertension between ART naïve and experienced patients [24], while a cohort study showed an increase in average systolic and diastolic blood pressure after 6 months on ART, with 10% of HIV patients having developed hypertension [29]. A Thai study, however, reported no significant difference in the prevalence of hypertension between ART naïve and ART experienced patients (22 vs. 28%) [25]. In Malaysia, hypertension was reported in almost half of HIV patients on ART [37]. Finally, a multi-country study showed that around 20% of ART experienced individuals had hypertension [15].

Metabolic Syndrome
Based on studies from India, Thailand and Taiwan, the prevalence of metabolic syndrome, a set of clinical parameters associated with the development of heart disease and other health problems such as diabetes, varied from 19-43% for HIV patients who were on ART [24,25,31,38]. Findings suggest that ART has a negative effect on metabolic functioning. This is illustrated by two studies that reported a significantly higher prevalence of metabolic syndrome in ART experienced than ART naïve patients [25,31] and by a longitudinal study that described new onset metabolic syndrome in 25% of study participants 6 months after ART initiation [29].

Abdominal Obesity
Abdominal obesity is a well-known contributor to hypertension, diabetes and metabolic syndrome. In ART experienced HIV patients in Thailand abdominal obesity was prevalent in 39% [25], which is similar to findings from an Indian study [19]. In both studies, the prevalence in patients on ART was comparable to ART naïve peers. Interestingly, findings from an Indian cohort study suggest that ART might eventually negatively affect healthy body mass, with study participants having increased BMI and skinfold thickness after 6 months of ART [29]. Likewise, waist circumference and waist-to-hip ratio were significantly higher in ART experienced compared to ART naïve individuals [28].

Tobacco Use
Across the Asian region, the proportion of current smokers ranged from 23-62% [15, 22, 39 -41] and that of former smoker from 10-20% [15,40]. One study specifically compared smoking among Asian populations on ART and found that compared to Thailand, the prevalence of smoking was lower in India and the Philippines, while it was higher in Vietnam, Indonesia, Hong Kong, Taiwan, Malaysia, Japan, South Korea and Singapore [15]. In other studies, the prevalence of current smoking was notably higher in men [40,42]. Furthermore, when comparing across HIV status, in India those with HIV were more likely to smoke (p <0.05) [42], whereas in Chinese men who had sex with men there were no significant differences between those with or without HIV [22].

MANAGEMENT OF CVD IN HIV POPULATIONS
Since HIV populations are at increased CVD risk, screening approaches designed for the general population may underestimate individual CVD risk. Hence, it is recommended that clinicians consider using HIV specific risk scoring systems or take into account that HIV infection is an additional risk factor and provide CVD preventive measures accordingly. As shown by a meta-analysis of observational studies, increased CVD risk might be attributable to specific antiretroviral drugs, such as abacavir, indinavir and lopinavir [43]. Newer generations of drugs could diminish CVD risk [44], however, these drugs might not be available in low-resource settings. It is of particular importance that clinicians carefully select ART regimens, thereby limiting adverse events that increase CVD risk. Moreover, health systems should plan for and ensure access to the most appropriate ART.
Globally, there are limited specific guidance and recommendations on the joint management and prevention of CVD in HIV patients. However, guidelines for HIV treatment do offer some support on the management of CVD and associated risks in HIV-infected patients. For example, in limited resource settings, WHO HIV treatment guidelines recommend the use of statins for people with a 10-year CVD risk exceeding 30%, and caution the use of boosted protease inhibitors with lovastatin and simvastatin as drug interactions may lead to increased risk of developing serious adverse effects [45]. Although national HIV treatment guidelines from the Asian region generally provide recommendations with regard to screening and monitoring of patients prior to ART initiation and afterwards (Table 3), the current routinely collected patient data in some countries might be insufficient to estimate individual CVD risk and provide preventive treatment accordingly. Furthermore, not all guidelines provide information with regard to possible interactions between ART and other drugs (i.e. antiplatelet or statin) that might prevent or treat CVD [46 -48].