Cardiovascular Risk Factors in HIV-Infected Adults in North- Central Nigeria: Prevalence, Associated Factors and Assessment of Risk Using the Framingham Risk Score – Preliminary Results from a Survey

HIV/AIDS is an increasingly important cause of cardiovascular (CVD) morbidity world-wide. We sought to evaluate the prevalence of CVD risk factors in HIV positive (HIV+) adults and assessment of these risks using the Framingham risk score (FRS). A cross-sectional study of adult clients of the HIV clinic at Jos University Teaching Hospital. One hundred and fifty HIV+ selected randomly with 50 (age and sex matched) HIV negative (HIV-) participants were enrolled. Relevant history, physical examination and biochemical investigations and 12-lead electrocardiography were performed. Data was analyzed using Epi-info 7.2 statistical software and P value < 0.05 was considered significant. The prevalent major CVD risk factors were dyslipidaemia (30.0% versus 6.0%), hypertension (34.0% versus 10.0%) and diabetes mellitus (10.0% versus 2.0%) among the HIV+ and HIVparticipants respectively. The FRS of the HIV+, 3 (IQR 3-28) were statistically significantly higher than that of the HIVparticipants, 2 (IQR 1-13); P=0.001. Furthermore, 32% of the HIV+ had moderate-high FRS compared to 2% of HIVparticipants. CD4 count ≤ 200 cells/ml, use of anti-retroviral (ART), ART use ≥ 2 years and use of protease inhibitors (PI) emerged as predictors of moderate-high FRS among the HIV+ participants. In conclusion, a high prevalence of CVD risk factors exists among HIV+ population in our local environment. These risk factors can be identified early by baseline/periodic cardiovascular work-up which should include use of CVD risk tools. Early diagnosis and treatment will significantly reduce morbidity and mortality in these patients.


INTRODUCTION
he acquired immune deficiency syndrome (AIDS) Thas been described as a global pandemic by the United nations (UN) and World Health Organization 1,2 (WHO). The burden is disproportionately high in sub-Saharan Africa where it causes significant morbidity and mortality mostly in the young and productive age 1,2 groups. Over 20 million deaths globally have been attributed to HIV infection, with about 75% of these 1,2 occurring in sub-Saharan Africa alone. Nigeria has the second largest population of HIV infected [1][2][3] persons in the world. According to the 2018 Nigeria HIV /AIDS Indicator and Impact Survey (NAIIS), the 3 current national prevalence is 3.1%. With a population of more than 140 million people, this represents over 10% of [1][2][3] the global pandemic in terms of absolute numbers . The introduction of anti-retroviral therapy (ART) in 1996 has led to a gradual decline in morbidity, mortality rate and a change in causes of death in persons living with 4,5 HIV/AIDS (PLWHA). Significant numbers of PLWHA 4,5 are estimated to be over 50 years of age.
This has brought to pre-eminence other causes of morbidity and mortality particularly cardiovascular diseases among [1][2][3][4][5] PLWHA. HIV/AIDS is known to have a significant negative impact on the cardiovascular system in protean [6][7][8][9] ways. The possible mechanistic pathways are the direct cardiotoxic effects of the virus, metabolic changes induced by the virus and ART, and certain predisposing lifestyles like smoking, alcohol abuse and drug abuse [6][7][8][9] commoner among PLWHA. In PLWHA, concerns have been expressed about the emergence of a cardiovascular [1][2][3][4][5]10 disease epidemic within the AIDS pandemic.
A preponderance of studies report that CVD increasingly account for significant proportion of morbidity and mortality (30% cause of mortality and more than 50% [11][12][13][14][15][16] increase in all-cause mortality) among PLWHA. Others have reported that certain classes of ART i.e. nucleoside reverse transcriptase inhibitors (NRTIs) and protease inhibitors (PIs) are associated with inducing metabolic syndrome which is a cluster of cardiovascular [17][18][19][20][21][22][23] risk factors. Furthermore, some have advocated for HIV infection to be classified as a major CVD risk 10 factor. Collectively these studies indicate that theshort and long term risks of ART may differ but there is no question that the use of ART to ensure adequate viral suppression is paramount to the successful clinical management of HIV infected patients despite its long term attendant risk. There is a growing call for baseline/periodic assessment of CVD risk factors in PLWHA. The FRS has been found useful in assessing CVD risk in PLHWA and has been validated for use [24][25][26][27] among different races. This study aims to add to the available evidence in literature on the burden of CVD risk factors and assessment of CVD risk using the FRS in Nigeria where similar studies are very few.

MATERIALS AND METHODS
This was a hospital based cross sectional analytical study conducted in the HIV clinic of JUTH in North-Central Nigeria. The study aimed to determine the prevalence of specific CVD risk factors and assessment of these risks with the FRS. One hundred and fifty adult HIV+ participants (90 on ART, 60 not ART) with fifty age and sex-matched HIV-as controls were enrolled into the study in a simple random manner over a period of four months. Those with acute illnesses and pregnancy were excluded. The HIV-controls were enrolled from clients of the HIV clinic who underwent HIV counselling and testing and were tested negative for HIV. The ethical approval for the study was obtained from the Human Research and Ethics Committee of JUTH. The nature of the study was explained to each participant in the language they best understand. Consenting participants were required to sign the consent forms, or append their thumb print where appropriate. Participants were at liberty to withdraw from the study at any stage without consequence and information obtained was treated as confidential.
Each participant was interviewed to obtain relevant demographic and clinical history. They were also grouped into social class using a Nigerian method by 28 Olusanya et al. Physical activity was graded according 2 9 to the WHO guidelines. Anthropometric and cardiovascular examinations were performed and documented. Weight was recorded in kilograms to the nearest 0.1kg using a flat scale on a firm horizontal surface with subjects wearing only light clothes and 30 without shoes and head gears. Height was recorded in metres with a stadiometer positioned on a flat surface and recorded to the nearest 0.01 metre. Height measurements 30 were taken without shoes and head gears. Body mass index (BMI) was calculated as the weight divided by the 2 30 square of the height and recorded in Kg/m . The body surface area (BSA) was calculated using the Mosteller (Kg)/3600] ). Waist circumference (WC) was measured in centimetres using a non-stretch metric tape on bare skin at the horizontal level at a point mid-way between the lowest rib and the iliac crest at the end of a 3 0 gentle expiration. Hip circumference (HC) was measured in centimetres using a non-stretch metric tape on bare skin at the horizontal level of maximum circumference around the buttocks and the pubic 30 symphysis. Blood pressure was measured using a standard mercury sphygmomanometer with appropriate cuff size in both in both arms in sitting position after a 5-10 minutes rest. Values corresponding to the first and fifth Korotkoff sounds were considered as the systolic and diastolic blood pressures respectively. Two additional recordings were made using the limb with the higher value; the average was taken to give a representative value.
On completion of physical examination, 10mls of fasting venous blood was drawn from each patient and put into appropriate specimen bottles for the following investigations; serum lipid profile, plasma glucose, serum creatinine and uric acid.  24,40 or low risk (10-year risk < 10%). Individuals with clinical form of cardiovascular disease or 24,40 with diabetes belong in the high risk category. Moderate risk also includes patients with 2 or more major risk factors and 10-year risk < 10%. Low risk patients are those with 1 major risk 24,40 factor or none and 10-year risk < 10%.

Statistical analysis
Data was analyzed using the EPI-Info version 7.2.2.6 (CDC Atlanta, Georgia USA) statistical software. Quantitative variables were summarized using mean and standard deviation (SD). Categorical variables were expressed using frequencies and percentages with confidence intervals stated. The student t -test or non-parametric Mann-Whitney/Kruskal-Wallis tests was used to compare means of 2 groups. 2 The Chi Square (X ) test was used to test the significance of association between categorical variables. Where the expected frequency of a cell was <5, Fisher's exact test was used. Multiple logistic regression analysis was performed to determine which HIV parameters (CD4 ≥ 200 cells/ml, detectable viral load, ART use, ART use ≥ 2 years and protease inhibitors use) are predictors of moderate-high FRS using variables that had a p-value of <0.05 on univariate analysis. In all cases, p-value of <0.05 was considered statistically significant.

RESULTS
One hundred and fifty HIV+ participants (90 ART experienced and 60 ART naive) and 50 age and sex matched HIV-controls were enrolled for this study. Everyone completed the study.     did not show statistically significant difference across the groups, however there was a statistically significant difference in the mean diastolic blood pressure across the two groups, P=0.008. There was statistically significant difference in the total cholesterol, HDL, triglycerides, atherogenic index, serum creatinine and estimated glomerular filtration rate across the two groups. The difference in the fasting blood sugar and LDL was not statistically significant across the groups.  Table 4 depicts the Framingham risk score and grades and relative risks of the participants. There was statistically significant difference in the FRS and relative risks across the 2 groups, P=0.001. 34% of the HIV+ have moderatehigh FRS compared to 2% of the HIV-. Table 5 depicts the results of multiple logistic regression analysis to determine which HIV parameters (CD4 ≤200 cells/ml, detectable viral load (≤200 copies/ml), ART use. duration of ART use ≥ 2 years and use of protease inhibitors) emerged as predictors of moderate-high FRS. CD4 count ≤200 cells/ml, use of ART, ART duration ≥ 2 years and use of protease inhibitors emerged as predictors of moderate-high FRS in the HIV+ participants.

DISCUSSION
HIV/AIDS is a global pandemic with preponderance of burden in sub-Saharan Africa where a significant amount [1][2][3] of morbidity and mortality occurs. Coupled with the burdens of other communicable and non-communicable diseases; HIV/AIDS predominantly affects the young and productive age-group in SSA with consequent socioeconomic devastation in societies where the [1][2][3] prevalence is high. The virus affects every system in the human body often causing multiple morbidities in 1 -3 , 6 -1 6 i n f e c t e d a n d a ff e c t e d i n d i v i d u a l .
T h e cardiovascular system in particular is affected by HIV/AIDS in protean ways with consequent increase in [6][7][8] morbidity and mortality. This study set out to evaluate the prevalence of specific cardiovascular disease risk factors, assessment of these risks using the Framingham risk score and the predictors of moderate-high Framingham risk scores in persons living with HIV/AIDS seen at the HIV clinic of Jos University Teaching Hospital. A total of 150 HIV+ and 50 HIV-participants who met the study criteria were enrolled. There were more females than males and majority were in the fourth decade of life. This was expected and is comparable to what obtains generally in Nigeria and in most parts of sub-Saharan Africa where majority of PLWHA are [1][2][3] young and in the productive age group. About 57% of the subjects in this study were single (20% due to loss of spouse), the disease tends to decimate families with loss of a either or both spouses common in affected [1][2][3] families. In this study most of the subjects were either unemployed or self-employed, reside in urban areas and belonged to the low socioeconomic class. This was expected and is in agreement with studies which identify HIV infection in Africa as a disease that in most cases is associated with low socioeconomic class and [1][2][3]41 societal economic retardation. The HIV+ participants and HIV-participants in this study were age and sex matched, they were also of comparable socio-demographic status. Therefore, the difference in the findings between them may to a large extent be due to the difference in their HIV status. The human immuno-deficiency virus causes pathologic cardiovascular and metabolic changes that may be [6][7][8][9] detected even in the early stages of the disease. This in addition to adverse lifestyle predisposes PLHWA to [41][42][43][44] cardiovascular disease.

Prevalence of specific cardiovascular disease risk factors in PLWHA
Low socio-economic class: The difference in social class of HIV+ and HIV-participants in this study did not attain statistical significance. This shows that the differences in the prevalence of cardiovascular risk factors and Framingham risk scores found may have been due to the virus and its associated factors. Studies have identified extremes of social class as a risk factor for cardiovascular 24,28,35,37,40,41 diseases in the general population. Furthermore in PLWHA, low socio-economic class has been identified as a significant predictor of increased cardiovascular risk score and disease. In a large multicenter study involving 931 men and 1455 women with HIV infection in the United states. Robert Kaplan et al concluded that having a low socio-economic class (income less than 10,000 dollars/year) was associated with increased prevalence of moderate-high coronary heart disease risk scores in 4 1 PLWHA. Infact, among the factors considered; belonging to a lower socio-economic class was the strongest risk factor identified to having a moderate-high 41 CHD risk score (OR 2.32, 95%CI 1.51-3.36). Similar findings have been obtained by several other studies on risk factors for cardiovascular abnormalities in HIV [17][18][19]41,46 infected persons.
These habits significantly promote cardiovascular diseases and are [41][42][43][44] commoner in those who experience stigma. Recent evidences suggest that smoking and alcoholism reduces the efficacy of anti-retroviral therapy, reduces immunity and promotes vascular disease, atherosclerosis, [41][42][43] hypertension and other co-morbidities. They have also 1-been shown to increase all-cause mortality in PLWHA. 6,[10][11][12][13][14] Inadequate physical activity was more prevalent in the HIV+ participants; this may be due to the chronicity and debilitating nature of the disease. HIV infection predisposes its victim to long periods of physical inactivity with its attendant consequences. Lack of exercises has been shown to have an adverse impact on blood glucose levels, blood pressure and lipid profiles; it also raises the risk of a cardiovascular event by two- [8][9][10][11][12][13][14][15][16][17][18]42,43,53 fold.
Overweight/Obesity: Excess body mass index (BMI ≥25) was found in 26.7% of HIV+ participants and 32% had increased waist circumference. The HIVparticipants however had higher mean values of weight and BMI; this may be due to the fact that HIV is a chronic infection which results in progressive weight [6][7][8][9][10][11]15,16,42,43 loss especially in those not yet on ART.
Among the HIV+ participants, those on ART had statistically significant mean values of weight, body mass index, waist circumference and waist-hip ratio compared to those not on ART. A recent study in fact found that being overweight or obese is now more prevalent than wasting 42,43 in PLWHA.
Increased mortality rates have been identified as an exponential function of increasing body 39,42,43,54 weight.
The risk of coronary heart disease doubles with BMI greater than 25 and increases nearly fourfold 39,54 when it is above 29. The risk of developing type 2 diabetes also increases with increasing weight such that individuals with a BMI above 35 have about 40-fold higher risk of developing diabetes when compared to 37,39,42,43,54 non-obese individuals. In people with normal BMI, increased waist circumference has a positive correlation with abdominal fat content. Fat located in the abdominal region is associated with a greater health risk than 37,39,42,43,54 peripheral fat.
This study also showed that increased wait-hip ratio is commoner in the HIV- and size in HIV infected persons with its attendant 37,39,42,43,54 consequences.
Dyslipidaemia: Dyslipidaemia is a significant predictor of endothelial dysfunction and atherosclerosis in the 35,39,40 general population.
Dietary modification may 15,16 therefore be needed as part of management modalities.
It is known that the impact of hypertension on morbidity and overall mortality rate is much higher among PLWHA than in the general population, hypertension increases the all-cause 45,47,50,51,[54][55][56][57][58] mortality in PLWHA by more than 50%. Furthermore, studies have shown that hypertension in PLWHA is related to the nadir CD4 count and by 55,56,57 extension the severity of the disease.
Hypertension has been described as a harbinger of other CVD risk factors, this may be due to a central pathophysiologic pathway for CVD risk factors; it is easily detectable and 9,15,16,35,39,50-is thus central in management of CVD risks. 52,[54][55][56][57][58] Experts in resource poor setting like ours must take advantage of this to manage cardiovascular risks in PLWHA and the general population.
Metabolic syndrome: About 25% of the HIV+ participants had at least 4 identified cardiovascular disease risks, those on ART had higher clustering of risks compared to those not on ART. Metabolic syndrome using the IDF criteria was found in almost 9.3% of the HIV+ participants, majority of who were on ART. Although the exact prevalence is not known in [11][12][13][15][16][17][18][19][20][21][22][23] PLWHA, some have estimated it as high as 32%. Metabolic syndrome refers to a cluster of certain specific CVD risks which occurs in different 35,39 combinations and is not due to chance alone. The individual risks sum up and increases risk exponentially resulting in a chronic pro-thrombotic and pro- 35,39 inflammatory state with attendant morbidity.

Cardiovascular risk assessment using the Framingham risk scores
In this study, the HIV+ participants had statistically significantly higher values of Framingham risk scores (FRS) compared to the HIV-controls. Likewise, those on ART also had statistically significantly higher values of FRS compared to those not on therapy. Among those on ART, those on protease inhibitor-based therapy had statistically significantly higher values of FRS compared to those not on protease inhibitor-based therapy. These findings are in keeping with reports from 17-19.25,26,41,46,49 several other studies.
A major and landmark study, 'Data on Adverse Effects of Antiretroviral therapy' survey reported similar findings of increased cardiovascular risk in persons on ART especially the PI 18,19 based one. Reasons suggested for this is that HIV and ART may have elaborative influence on traditional and non-traditional CVD risk factors by multifactorial [6][7][8][9][10]15,16,21 pathways.
However, it is generally agreed that knowing the CVD risk and acting on it is imperative to long term survival in HIV-infected persons. There is a growing call for HIV infection to be classified as a major CVD risk factor on account of the significant effect of the 10 virus and ART on the cardiovascular system.

Predictors of moderate-high Framingham risk scores in PLWHA
This study set out to look at the specific HIV-related parameters i.e., ART use, ART duration, use of proteaseinhibitor based regimen, CD4 count ≤200 cells/ml, and detectable viral load (≥200 copies/ml) rather than the traditional CVD risk factors as independent predictors of moderate-high FRS among PLWHA. Moderate-high FRS connotes significant increase in the likelihood of occurrence of a cardiovascular event or disease hence 24,35 infers adverse prognosis. The presence of moderate-high Framingham risk score was significantly associated to all the HIV-related parameters examined on univariate analysis. However, on multiple logistic regression analysis; CD4 count ≤200 cells/ml, ART duration ≥ 2years and use of protease inhibitor-based regimen emerged as independent predictors of moderate-high FRS in the HIV+ group. Paradoxically, use of ART generally appears to be protective to getting moderate-high FRS (OR 0.0942, 95%CI 0.0094 -0.9487, P= 0.0450), perhaps due to the overall initial improvement in the state of health when ART is started i.e., marked reduction in chronic inflammatory markers, reduction in viral load etc. The initial effects of ART become attenuated over time with emergence and accentuation of chronic complications A R T u s e ( d y s l i p i d a e m i a , l i p o d y s t r o p h y , m e t a b o l i c / m i t o c h o n d r i a l r e -p r o g r a m m i n g , cardiovascular toxicity etc.). The results in this study appears to suggest ART being initially protective against elaboration of CVD risks and thence moderatehigh FRS, the initial protection is lost after 2 years when the complications of ART use sets in (OR 2.3729, 95%CI 1.3589-12.6606, P= 0.0256). Reports from studies appears to be ambivalent with regards to the role [11][12][13][14][15][16][17][18][19][20][21][22][23]41,[45][46][47][48][49][50][51][52][53][54][55][56][57][58] of ART on CVD risks, some studies however [17][18][19][20][21][22][23][54][55][56][57][58] align with the finding in this study; the reasoning behind the results in this study also appears [4][5][6][7][8][9][10] scientifically rational and plausible. Detectable viral load (≥200 c/ml) didn't emerge as an independent predictor perhaps due to the rapid flux rate of viral load, other studies also report similar findings. CD4 count ≤200 cells/ml was found to be a predictor of moderatehigh FRS, low CD4 count signifies chronic inflammatory state (with consequent re-setting of immunologic memory) and severity of HIV infection [6][7][8][9][10] and unlike viral load is not prone to rapid flux rate. Baseline (nadir or trough) and recent CD4 count have been reported in some studies to be related to the development of hypertension and other CVD risk [17][18][19][20][21][22][23][54][55][56][57][58] factors in PLWHA.
Similarly, use of protease inhibitor-based regimen was found to predict moderatehigh FRS in this study perhaps due to the consequences of pathologic mitochondrial and metabolic changes induced by protease inhibitors. The finding is in [17][18][19][21][22][23]41 agreement with reports from other studies.

CONCLUSION
Cardiovascular disease risk factors are prevalent in PLWHA in our local environment. Early identification and assessment of these risk using a CVD risk assessment tool such as the Framingham risk score is advocated with prompt intervention to decrease morbidity, improve quality of life and reduce mortality these patients.

Recommendations
It is recommended that baseline and routine screening for CVD risk factors and assessment of these risks using