Predictors of echocardiographic abnormalities among people living with hiv on anti-retroviral therapy in Kumasi, Ghana Authors: Isaac Nana Ogyefo, Isaac Kofi Owusu, Daniel Ohemeng Minkah, Betty Roberta Norman, Bernard Nkum, Abdul Jalil Inusah, Collins Kokuro Published: 2024 ABSTRACT BackgroundCardiac abnormalities are common among patients living with human immunodeficiency virus (PLWH) on anti-retroviral therapy (ART). Echocardiograms represent a useful modality in determining cardiac abnormalities among PLWH. In Ghana, the predictors of echocardiographic abnormalities have not been extensively elucidated. This study was designed to determine HIV-specific and traditional risk factors associated with echocardiographic abnormalities among PLWH on ART.MethodsOne-hundred and seventeen (117) PLWH attending the HIV clinic at the Komfo Anokey Teaching Hospital (KATH), Kumasi were consecutively recruited between March and August 2021. Age and sex-matched control group of 117 individuals who were HIV-negative blood donors were recruited. Relevant baseline demographic and clinical characteristics were obtained from all participants using a standardized questionnaire. Transthoracic echocardiography was performed for all participants and the predictors of echocardiographic abnormalities were determined using binary logistic regression modelling.ResultsIncreasing age (aOR = 5.8; 95% CI, 3.4–9.8; p = 0.04), female gender compared to male gender (aOR = 4.8; 95% CI, 2.4–9.4; p = 0.033), and viral load > 50 copies/mL compared to viral load ≤ 50 copies/ml or undetectable (aOR = 4.4; 95% CI, 2.2–9.0; p < 0.001) were significantly associated with the presence of echocardiographic abnormalities.ConclusionMarkers of immunosuppression such as HIV viral load > 50 copies/ml and traditional cardiovascular risk factors such as increasing age and female gender were useful in predicting cardiac abnormalities among PLWH in this Ghanaian study. HIV disease and traditional cardiac risk factors must be recognized and aggressively managed to reduce cardiovascular morbidity and mortality among PLWH especially among females given the faster progression to AIDS even at similar levels of viremia compared to males.Supplementary InformationThe online version contains supplementary material available at 10.1186/s12879-024-10246-x. Introduction According to the Joint United Nations Programme on Human Immuno-deficiency Virus (HIV) and Acquired Immuno-deficiency Syndrome (AIDS) (UNAIDS), 37.7 million people are living with HIV (PLWH) worldwide as of 2020 [1]. There were about 1.5 million new HIV infections globally in 2020 [1, 2]. The first-ever case of HIV/AIDS in Ghana was reported in 1986 [3, 4]. According to the UNAIDS, 320 000 people are living with HIV and AIDS as of 2020 in Ghana [3]. Over 90% of this number are persons aged 15 and older. HIV remains an epidemic in Ghana, with a prevalence rate of 2.4% in 2016 from a prevalence of 1.8% in 2015, representing a 0.6% rise [3]. Some studies have shown that subclinical echocardiographic abnormalities can independently predict adverse cardiovascular outcomes among PLWH and allow the targeting and identification of high-risk patients for early treatment and intervention [5]. Recently, in a cross-sectional multi-centre study, Vasconcellos and colleagues examined the association between HIV serostatus and cardiac structure and function by 2-dimensional (2D) echocardiography among over 1000 males living with HIV in the United States of America (USA) which found a stronger association between HIV seropositivity and increased left ventricular mass index (LVMI), larger left atrial diameter and right ventricular (RV) end-diastolic area, lower RV function and higher left ventricular (LV) diastolic dysfunction in men with HIV infection compared to age-matched controls [6]. In Indonesia, Hidayat and colleagues performed a cross-sectional study on 52 people with HIV in 2018 [7]. They found that cluster of differentiation-4 (CD4) count of fewer than 200 cells/mm3 was associated with a higher incidence of increased LVMI, LV diastolic dysfunction and pericardial effusion [7]. Adebola and colleagues, in an cross-sectional pilot study in Nigeria using 80 HIV-positive patients and 40 healthy controls, observed larger LV cavity and mass in the HIV-infected cohort than in controls [8]. Like other developing and sub-Saharan African (SSA) countries, Ghana is experiencing a rapid epidemiological transition from predominantly diseases of infectious origin such as HIV and tuberculosis to a double burden of infectious and chronic non-communicable diseases, including cardiovascular diseases, type 2 diabetes mellitus, and cancers [9–11]. Appiah et al. [12] demonstrated a high prevalence of cardiovascular risk factors, including type 2 diabetes mellitus (5%), hypertension (9%), and disorders of lipid metabolism (29%) among PLWH in Ghana. This study sought to determine the predictors of echocardiographic abnormalities among PLWH on anti-retroviral therapy (ART) at Komfo Anokye Teaching Hospital (KATH), Kumasi, Ghana. Additionally, new data on prevalence, predictors, and correlates of echocardiographic abnormalities among PLWH can be obtained in this study to build on the already existing data. Furthermore, additional knowledge will be obtained to help bridge the gap on the association between HIV-specific factors such as viral load, CD4 count, duration and type of ART, and duration of HIV illness on cardiac structure and function using echocardiograms among PLWH on ARTs in Kumasi, Ghana. Materials and methods Study design This was a hospital-based cross-sectional study involving one-hundred and seventeen (117) consecutive PLWH aged ≥ 18 years adults reporting for care at the HIV clinic at KATH and one-hundred and seventeen (117) age-and-sex-matched HIV negative blood donors as controls from the KATH blood donor clinic. This study was conducted over a period of six (6) months, from 1st March 2021 to 31st August 2021. All PLWH on ART presenting for review at the KATH HIV clinic who gave written informed consent were included in the study. Acutely ill PLWH, those with a history of congenital heart disease, pregnant women or PLWH in the puerperium and PWLH with a previous history of cardiac disease or cardiac surgery were excluded. PLWH with a history of conditions causing autonomic dysfunction such as hypertension and diabetes mellitus were also excluded. The required sample size was calculated using the formula below. Sample size, n = Z2 PQ/d2. Where, n is the minimum sample size of the study population, Z is the Z score at 95% confidence interval (1.96), P is the estimated prevalence of ECG abnormalities among PLWH on ART in a similar study (93%) [13], Q = 1 – P, d = margin of error (0.05). Therefore, N = 1.962 × 0.93 × 0.07 / 0.052=100.03. The minimum sample size determined from the formula above was 100 was calculated from the formula above. One-hundred and seventeen (117) participants were recruited into each arm of the study. Study methods All participants were interviewed using a standardized questionnaire to obtain socio-demographic information such as age, gender, marital status, religion, occupation, and level of education. After the interview, all participants were evaluated for clinical and anthropometric parameters such as height (cm), weight (kg), and blood pressure (mmHg) by the researcher and trained senior nurses in the cardiology unit of KATH as research assistants. Height was measured to the nearest 0.5 cm while weight was measured to the nearest 0.5 kg without shoes while wearing light clothing using a Secca digital height and weight scale. Body mass index (BMI) was calculated as the weight of the body in kilograms divided by the square of the height in meters. Abdominal circumference was measured in the standing position as the smallest circumference between the lower rib margin and the anterior superior iliac crest when the participant exhaled. Blood pressure was measured with an Omron Hem 907 electronic sphygmomanometer in a seated position after the participant had rested for at least 5 min at the left arm followed by the right arm and repeated in the arm with the highest value. The average of the last two recordings was recorded. For all participants, 5 mls of venous blood was taken by a trained phlebotomist into a gel-separator tube for fasting lipid profile. As per the laboratory protocol, all participants were advised to come in after an overnight fast and samples were taken early in the morning before commencement of clinic or on a separate day if convenient with appropriate arrangements in place to support transportation if not a routine clinic day, as per recommendations from the Institutional Review Board (IRB). Samples were examined using a Mindray fully automated chemistry analyzer. For PLWH, two 5 mls of venous blood were taken for CD4 count and viral load. CD4 count was analyzed at the KATH laboratory using a BD (Becton, Dickinson, and Company) Facscount machine by the principle of flow cytometry. HIV viral load was analyzed at the KATH laboratory using a Cobas protocol AmliPrep/Taqman 48 Version 2.0 (Roche Molecular System, Branchburg, New Jersey, USA) by polymerase chain reaction. Trans-thoracic echocardiographic examination was performed for all participants by the researcher using a SIEMENS Acuson x700 cardiac ultrasound machine equipped with a 3.5-MHz phased-array transducer. All subjects were counselled on the non-invasive nature of the procedure to allay fear and anxiety. Two-dimensional (2D), M-mode, pulse-wave, continuous-wave, colour doppler, and tissue Doppler echocardiography assessment were done with the patient lying in the left lateral decubitus position. The procedure was done following the joint European Association of Cardiovascular Imaging (EACVI) and the American Society of Echocardiography 2015 guidelines (ASE) [14]. LV dimensions such as LV internal diameter at end-diastole (LVIDd), LV internal diameter at end-systole (LIVDs), interventricular septal diameter at end-diastole (IVSd), interventricular septal diameter in systole (IVSs), LV posterior wall thickness at end-diastole (LWPWd) were measured in 2-D in the parasternal long-axis view with ECG gating. End-diastolic measurements were taken at the peak of the QRS complex while end-systolic measurements were taken at the end of the T-wave [15]. Care was taken to obtain the measurements perpendicular to the LV long axis at the level just distal to mitral valve leaflet tips. Electronic calipers were positioned on the interface between the myocardial wall and cavity and the interface between the myocardial wall and pericardium [14]. RV systolic function was assessed using tricuspid annular plane systolic excursion (TAPSE) [14]. It was measured by M-mode in the apical 4-chamber view by aligning the cursor along the direction of the tricuspid lateral annulus [14]. TAPSE has shown a good correlation with other parameters estimating RV global systolic function, such as radionuclide-derived RV ejection fraction (EF) and RV 2-D fractional area change (FAC) [14]. Left atrial and LV volumes were estimated in 2-D using the biplane method of disks summation (modified Simpsons rule) [14]. Cardiac chamber dimensions and volumes were indexed by the body surface area (BSA), which was calculated using the Mosteller formula defined below [16]. \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{mathrsfs} \usepackage{upgreek} \setlength{\oddsidemargin}{-69pt} \begin{document}$$BSA\left( {{m^2}} \right) = {\left( {height\left( {cm} \right){\text{ }}x{\text{ }}weight\left( {kg} \right)/3600} \right)^{{\raise0.5ex\hbox{$\scriptstyle 1$}\kern-0.1em/\kern-0.15em\lower0.25ex\hbox{$\scriptstyle 2$}}}}$$\end{document} Echocardiographic parameters were defined as follows. I.LV systolic dysfunction was defined as LVEF less than 52% for male and less than 54% for females [14]. Based on current recommendations, LV end-diastolic and end-systolic volumes were estimated using the modified Simpsons biplane method and was used to calculate LV ejection fraction (LVEF) [14]. The LVEF was calculated from the formula: LVEF = (LVEDV – LVESV)/LVEDV, where LVEDV is LV end-diastolic volume, and LVESV is LV end-systolic volume. Mild LV systolic dysfunction was determined as LVEF of 41 − 51% for males and 41 -53% for females [14]. Moderate LV systolic dysfunction was defined as LVEF between 30 – 40%.14 Severe LV systolic dysfunction was defined as LVEF of less than 30%.14II.Dilated cardiomyopathy was defined based on three criteria: LV end-diastolic diameter ≥ 5.5 cm; LVFS < 28% and global hypokinesia (LVEF < 50%). [17].III.LV hypertrophy (LVH) was defined as LVMI greater than 95 g/m2 in women and greater than 115 g/m2 in men [14]. LV mass was estimated using the formula below and indexed to body surface area to determine LVMI [14]. LV mass = 0.8 × 1.04 x [(IVSd + LVIDd + LVPWd)3 − LVIDd3] + 0.6 g [14, 16]. where, IVSd is interventricular septum thickness at end-diastole, LVIDd is LV internal diameter at end-diastole, and LVPWd is LV posterior wall diameter at end-diastole. BSA was calculated using the formula by Mosteller [16]. Normal ranges for LVMI was 43–95 g/m2 for women and 49–115 g/m2 for men [14]. IV.LV diastolic dysfunction was diagnosed in participants who had the following: Average E/e’ > 14; septal e’ velocity < 0.07 m/s; peak tricuspid regurgitant (TR) velocity > 2.8 m/s; left atrial volume index (LAVI) > 34 ml/m2, where E is mitral inflow velocity and e’ is mitral annular tissue doppler velocity [18]. Grade 1 LV diastolic dysfunction was defined as mitral E/A ratio ≤ 0.8, average E/e ratio < 10, peak TR velocity < 2.8, and a normal or increased LAVI [18]. Grade 2 LV diastolic dysfunction was defined as mitral E/A ratio that ranged from > 0.8 to < 2, average E/e ratio > 14, peak TR velocity > 2.8 m/sec and increased LAVI [18]. Grade 3 LV dysfunction was defined as mitral E/A ratio > 2, average E/e ratio > 14, peak TR velocity > 2.8, and increased LAVI [18]. LV diastolic dysfunction was determined according to the 2016 joint update from the American Society of Echocardiography (ASE) and the European Association of Cardiovascular Imaging (EACVI) recommendations for the evaluation of LV diastolic function [18]. Left atrial volume was measured in the apical 4-chamber and 2-chamber view using the biplane area-length method and indexed to body surface area per recommendations from ASE/EACVI.18V.Pericardial effusion was considered present if there was the presence of echo-free space between visceral and parietal pericardium that persists throughout the cardiac cycle [19]. Pericardial effusion was further graded as mild if echo-free space is < 1.0 cm, moderate when it was 1.0–2.0 cm and large when it is > 2.0 cm on two-dimensional images during diastole [19]. VI.Pulmonary hypertension was determined based on an estimated pulmonary artery systolic pressure (PASP) > 35 mmHg on echocardiogram [20]. In the absence of RV outflow tract obstruction, right ventricular systolic pressure (RVSP) = systolic pulmonary artery pressure (sPAP in mmHg) = trans-tricuspid gradient + right atrial pressure (RAP) [20]. The trans-tricuspid gradient is 4 V [2], where V = peak velocity of tricuspid regurgitation in m/second. RVSP was determined using this formula: RVSP = 4V2 + RAP (mmHg). RAP was estimated using recommendations from the ASE endorsed by the European Society of Echocardiography according to the following: Normal RAP of 3 mmHg (range, 0–5 mm Hg) = inferior vena cava (IVC) IVC diameter ≤ 2.1 cm that collapses > 50% with a sniff. RA pressure of 15 mm Hg (range, 10–20 mm Hg) = IVC diameter > 2.1 cm that collapses < 50% with a sniff. In scenarios in which IVC diameter and collapse do not fit the description above, an intermediate value of 8 mm Hg (range, 5–10 mm Hg) was used [20]. VII.RV systolic dysfunction was defined as TAPSE < 1.7 cm [14]. I.LV systolic dysfunction was defined as LVEF less than 52% for male and less than 54% for females [14]. Based on current recommendations, LV end-diastolic and end-systolic volumes were estimated using the modified Simpsons biplane method and was used to calculate LV ejection fraction (LVEF) [14]. The LVEF was calculated from the formula: LVEF = (LVEDV – LVESV)/LVEDV, where LVEDV is LV end-diastolic volume, and LVESV is LV end-systolic volume. Mild LV systolic dysfunction was determined as LVEF of 41 − 51% for males and 41 -53% for females [14]. Moderate LV systolic dysfunction was defined as LVEF between 30 – 40%.14 Severe LV systolic dysfunction was defined as LVEF of less than 30%.14 LV systolic dysfunction was defined as LVEF less than 52% for male and less than 54% for females [14]. Based on current recommendations, LV end-diastolic and end-systolic volumes were estimated using the modified Simpsons biplane method and was used to calculate LV ejection fraction (LVEF) [14]. The LVEF was calculated from the formula: LVEF = (LVEDV – LVESV)/LVEDV, where LVEDV is LV end-diastolic volume, and LVESV is LV end-systolic volume. Mild LV systolic dysfunction was determined as LVEF of 41 − 51% for males and 41 -53% for females [14]. Moderate LV systolic dysfunction was defined as LVEF between 30 – 40%.14 Severe LV systolic dysfunction was defined as LVEF of less than 30%.14 II.Dilated cardiomyopathy was defined based on three criteria: LV end-diastolic diameter ≥ 5.5 cm; LVFS < 28% and global hypokinesia (LVEF < 50%). [17]. Dilated cardiomyopathy was defined based on three criteria: LV end-diastolic diameter ≥ 5.5 cm; LVFS < 28% and global hypokinesia (LVEF < 50%). [17]. III.LV hypertrophy (LVH) was defined as LVMI greater than 95 g/m2 in women and greater than 115 g/m2 in men [14]. LV mass was estimated using the formula below and indexed to body surface area to determine LVMI [14]. LV mass = 0.8 × 1.04 x [(IVSd + LVIDd + LVPWd)3 − LVIDd3] + 0.6 g [14, 16]. where, IVSd is interventricular septum thickness at end-diastole, LVIDd is LV internal diameter at end-diastole, and LVPWd is LV posterior wall diameter at end-diastole. BSA was calculated using the formula by Mosteller [16]. Normal ranges for LVMI was 43–95 g/m2 for women and 49–115 g/m2 for men [14]. LV hypertrophy (LVH) was defined as LVMI greater than 95 g/m2 in women and greater than 115 g/m2 in men [14]. LV mass was estimated using the formula below and indexed to body surface area to determine LVMI [14]. LV mass = 0.8 × 1.04 x [(IVSd + LVIDd + LVPWd)3 − LVIDd3] + 0.6 g [14, 16]. where, IVSd is interventricular septum thickness at end-diastole, LVIDd is LV internal diameter at end-diastole, and LVPWd is LV posterior wall diameter at end-diastole. BSA was calculated using the formula by Mosteller [16]. Normal ranges for LVMI was 43–95 g/m2 for women and 49–115 g/m2 for men [14]. IV.LV diastolic dysfunction was diagnosed in participants who had the following: Average E/e’ > 14; septal e’ velocity < 0.07 m/s; peak tricuspid regurgitant (TR) velocity > 2.8 m/s; left atrial volume index (LAVI) > 34 ml/m2, where E is mitral inflow velocity and e’ is mitral annular tissue doppler velocity [18]. Grade 1 LV diastolic dysfunction was defined as mitral E/A ratio ≤ 0.8, average E/e ratio < 10, peak TR velocity < 2.8, and a normal or increased LAVI [18]. Grade 2 LV diastolic dysfunction was defined as mitral E/A ratio that ranged from > 0.8 to < 2, average E/e ratio > 14, peak TR velocity > 2.8 m/sec and increased LAVI [18]. Grade 3 LV dysfunction was defined as mitral E/A ratio > 2, average E/e ratio > 14, peak TR velocity > 2.8, and increased LAVI [18]. LV diastolic dysfunction was determined according to the 2016 joint update from the American Society of Echocardiography (ASE) and the European Association of Cardiovascular Imaging (EACVI) recommendations for the evaluation of LV diastolic function [18]. Left atrial volume was measured in the apical 4-chamber and 2-chamber view using the biplane area-length method and indexed to body surface area per recommendations from ASE/EACVI.18 LV diastolic dysfunction was diagnosed in participants who had the following: Average E/e’ > 14; septal e’ velocity < 0.07 m/s; peak tricuspid regurgitant (TR) velocity > 2.8 m/s; left atrial volume index (LAVI) > 34 ml/m2, where E is mitral inflow velocity and e’ is mitral annular tissue doppler velocity [18]. Grade 1 LV diastolic dysfunction was defined as mitral E/A ratio ≤ 0.8, average E/e ratio < 10, peak TR velocity < 2.8, and a normal or increased LAVI [18]. Grade 2 LV diastolic dysfunction was defined as mitral E/A ratio that ranged from > 0.8 to < 2, average E/e ratio > 14, peak TR velocity > 2.8 m/sec and increased LAVI [18]. Grade 3 LV dysfunction was defined as mitral E/A ratio > 2, average E/e ratio > 14, peak TR velocity > 2.8, and increased LAVI [18]. LV diastolic dysfunction was determined according to the 2016 joint update from the American Society of Echocardiography (ASE) and the European Association of Cardiovascular Imaging (EACVI) recommendations for the evaluation of LV diastolic function [18]. Left atrial volume was measured in the apical 4-chamber and 2-chamber view using the biplane area-length method and indexed to body surface area per recommendations from ASE/EACVI.18 V.Pericardial effusion was considered present if there was the presence of echo-free space between visceral and parietal pericardium that persists throughout the cardiac cycle [19]. Pericardial effusion was further graded as mild if echo-free space is < 1.0 cm, moderate when it was 1.0–2.0 cm and large when it is > 2.0 cm on two-dimensional images during diastole [19]. Pericardial effusion was considered present if there was the presence of echo-free space between visceral and parietal pericardium that persists throughout the cardiac cycle [19]. Pericardial effusion was further graded as mild if echo-free space is < 1.0 cm, moderate when it was 1.0–2.0 cm and large when it is > 2.0 cm on two-dimensional images during diastole [19]. VI.Pulmonary hypertension was determined based on an estimated pulmonary artery systolic pressure (PASP) > 35 mmHg on echocardiogram [20]. In the absence of RV outflow tract obstruction, right ventricular systolic pressure (RVSP) = systolic pulmonary artery pressure (sPAP in mmHg) = trans-tricuspid gradient + right atrial pressure (RAP) [20]. The trans-tricuspid gradient is 4 V [2], where V = peak velocity of tricuspid regurgitation in m/second. RVSP was determined using this formula: RVSP = 4V2 + RAP (mmHg). RAP was estimated using recommendations from the ASE endorsed by the European Society of Echocardiography according to the following: Normal RAP of 3 mmHg (range, 0–5 mm Hg) = inferior vena cava (IVC) IVC diameter ≤ 2.1 cm that collapses > 50% with a sniff. RA pressure of 15 mm Hg (range, 10–20 mm Hg) = IVC diameter > 2.1 cm that collapses < 50% with a sniff. In scenarios in which IVC diameter and collapse do not fit the description above, an intermediate value of 8 mm Hg (range, 5–10 mm Hg) was used [20]. Pulmonary hypertension was determined based on an estimated pulmonary artery systolic pressure (PASP) > 35 mmHg on echocardiogram [20]. In the absence of RV outflow tract obstruction, right ventricular systolic pressure (RVSP) = systolic pulmonary artery pressure (sPAP in mmHg) = trans-tricuspid gradient + right atrial pressure (RAP) [20]. The trans-tricuspid gradient is 4 V [2], where V = peak velocity of tricuspid regurgitation in m/second. RVSP was determined using this formula: RVSP = 4V2 + RAP (mmHg). RAP was estimated using recommendations from the ASE endorsed by the European Society of Echocardiography according to the following: Normal RAP of 3 mmHg (range, 0–5 mm Hg) = inferior vena cava (IVC) IVC diameter ≤ 2.1 cm that collapses > 50% with a sniff. RA pressure of 15 mm Hg (range, 10–20 mm Hg) = IVC diameter > 2.1 cm that collapses < 50% with a sniff. In scenarios in which IVC diameter and collapse do not fit the description above, an intermediate value of 8 mm Hg (range, 5–10 mm Hg) was used [20]. VII.RV systolic dysfunction was defined as TAPSE < 1.7 cm [14]. RV systolic dysfunction was defined as TAPSE < 1.7 cm [14]. Ethical consideration Ethical approval was obtained from the KATH IRB with reference number KATH IRB/AP/147/20. Written informed consent was also obtained from all participants. All methods were carried out in accordance with the Helsinki Declaration. Statistical analysis Descriptive and inferential statistical analytical methods were employed. In describing the prevalence and patterns of echocardiographic abnormalities among PLWH and controls, continuous parametric variables were presented by measures of central tendency and measures of spread such as mean ± standard deviation. Alternatively, continuous non-parametric variables were presented by the median and interquartile range (IQR). Categorical variables were reported as frequency counts with percentages. Statistically significant differences between categorical variables were tested through the chi-square statistical method. The statistically significant differences between continuous variables were tested through student’s t-test statistical method. A p value < 0.05 was considered significant. Inferentially, echocardiographic abnormalities were redefined as dichotomized categorical variables as follows: Echocardiographic abnormality (Normal = 0, Abnormality = 1) and were evaluated or tested using the binary logistic regression method. Both unadjusted (univariable) and adjusted (multivariable) binary logistic regression methods were employed. The HIV-specific and traditional cardiovascular risk factors considered independent variables in the univariable binary logistic regression analyses included age, sex, BMI, waist circumference, total cholesterol, HIV duration, ART duration, CD4 Count, and viral load. BMI was dichotomized into normal body weight and overweight for binary logistic regression analyses. The Odds Ratio (OR) derived from the binary logistic regression analyses was considered significant if p-value < 0.05. Only independent variables significantly associated with echocardiographic abnormalities in the univariable analysis (p < 0.05) were included in the multivariable logistic regression models. Results A total of 234 patients were recruited and enrolled in this study as shown in Table 1 which demonstrates the socio-demographic, clinical, anthropometric and laboratory characteristics of the participants. They comprised 117 PLWH recruited as cases and 117 age and sex-matched HIV-negative controls. Majority of participants, 92 (78.6%) of PLWH and 80 (68.4%) of the controls, were females as shown in Table 1. Age was similarly distributed among both cases (47.1 ± 11.6) and controls (45.8 ± 11.9) (p = 0.554). Table 1Socio-demographic, clinical, anthropometric and laboratory characteristics of participantsCharacteristicStudy participantsp-valuePLWHN = 117ControlsN = 117 Gender 0.075 Male25(21.4)37(31.6) Female92(78.6)80(68.4) Religion 0.610 Moslem7(6.1)7(6.0) Christian110(93.9)109(93.2) Others0(0.0)1(0.9) Marital Status ≤ 0.001* Married37(31.6)78(66.4) Single22(18.8)31(26.7) Divorced27(23.1)2(1.7) Widowed31(26.5)6(5.2) Occupation 0.231 Employed90(77.2)98(83.5) Unemployed27(22.8)19(16.5) Education ≤ 0.001* No Formal9(7.7)15(12.8) Primary37(31.6)12(10.1) Secondary59(50.4)28(23.9) Technical11(9.4)5(4.6) University1(0.9)57(48.6) Mean ± SD Mean ± SD Age, in years47.1 ± 11.645.8 ± 11.90.554Weight (kg)70.9 ± 14.972.0 ± 14.70.565Height (cm)160 ± 10160 ± 100.099BSA (m2)1.8 ± 0.21.8 ± 0.20.171BMI (kg/m2)26.8 ± 5.826.5 ± 5.00.667Waist circ. (cm)91.3 ± 12.888.4 ± 10.10.064Systolic BP (mmHg)128.1 ± 20.3124.3 ± 17.90.133Diastolic BP (mmHg)82.9 ± 11.780.2 ± 10.70.073Pulse rate (bpm)85.1 ± 17.181.5 ± 12.60.066Total cholesterol (mmol/L)4.2 ± 1.24.4 ± 1.00.121LDL cholesterol (mmol/L)2.5 ± 1.02.7 ± 1.20.551HDL cholesterol (mmol/L)1.2 ± 0.81.3 ± 0.50.231Triglycerides (mmol/L)1.4 ± 0.91.5 ± 0.70.671*p < 0.05 = statistically significantBMI: body mass index; BP: blood pressure; BSA: body surface area; HDL: high-density lipoprotein; LDL: low-density lipoprotein; PLWH: people living with HIV: Waist circ.: waist circumferenceReference: Adapted from [21] Table 1Socio-demographic, clinical, anthropometric and laboratory characteristics of participantsCharacteristicStudy participantsp-valuePLWHN = 117ControlsN = 117 Gender 0.075 Male25(21.4)37(31.6) Female92(78.6)80(68.4) Religion 0.610 Moslem7(6.1)7(6.0) Christian110(93.9)109(93.2) Others0(0.0)1(0.9) Marital Status ≤ 0.001* Married37(31.6)78(66.4) Single22(18.8)31(26.7) Divorced27(23.1)2(1.7) Widowed31(26.5)6(5.2) Occupation 0.231 Employed90(77.2)98(83.5) Unemployed27(22.8)19(16.5) Education ≤ 0.001* No Formal9(7.7)15(12.8) Primary37(31.6)12(10.1) Secondary59(50.4)28(23.9) Technical11(9.4)5(4.6) University1(0.9)57(48.6) Mean ± SD Mean ± SD Age, in years47.1 ± 11.645.8 ± 11.90.554Weight (kg)70.9 ± 14.972.0 ± 14.70.565Height (cm)160 ± 10160 ± 100.099BSA (m2)1.8 ± 0.21.8 ± 0.20.171BMI (kg/m2)26.8 ± 5.826.5 ± 5.00.667Waist circ. (cm)91.3 ± 12.888.4 ± 10.10.064Systolic BP (mmHg)128.1 ± 20.3124.3 ± 17.90.133Diastolic BP (mmHg)82.9 ± 11.780.2 ± 10.70.073Pulse rate (bpm)85.1 ± 17.181.5 ± 12.60.066Total cholesterol (mmol/L)4.2 ± 1.24.4 ± 1.00.121LDL cholesterol (mmol/L)2.5 ± 1.02.7 ± 1.20.551HDL cholesterol (mmol/L)1.2 ± 0.81.3 ± 0.50.231Triglycerides (mmol/L)1.4 ± 0.91.5 ± 0.70.671*p < 0.05 = statistically significantBMI: body mass index; BP: blood pressure; BSA: body surface area; HDL: high-density lipoprotein; LDL: low-density lipoprotein; PLWH: people living with HIV: Waist circ.: waist circumferenceReference: Adapted from [21] Socio-demographic, clinical, anthropometric and laboratory characteristics of participants *p < 0.05 = statistically significant BMI: body mass index; BP: blood pressure; BSA: body surface area; HDL: high-density lipoprotein; LDL: low-density lipoprotein; PLWH: people living with HIV: Waist circ.: waist circumference Reference: Adapted from [21] Table 2 shows data on HIV-related factors such as HIV type, HIV infection and ART duration, ART types, viral load, and CD4 count. Most, 114 (97%) of the PLWH, were infected with HIV type 1, as shown in Table 2. The mean ± SD duration of ART use in years was 6.7 ± 5.8. Seventy-five (64.4%), representing a majority of the PLWH had a CD4 count > 200 cells/mm3. Similarly, 70 (60%) of the PLWH had viral load levels less than or equal to 50 copies/ml or undetectable. Table 3 compares echocardiographic dimensions between the cases and controls. HIV patients had significantly increased IVSd, mitral E/A, mitral valve deceleration time, LA volume, LAVI, average E/e, LVM, and LVMI. The IVSd (cm) was 0.9 ± 0.2 among the HIV patients and was 0.8 ± 0.1 among the controls (p < 0.000). LAVI (mL/m2) was 18.5 ± 4.8 among the HIV patients and was 16.3 ± 3.0 among the cases (p < 0.000). The average E/e was 9.1 ± 3.0 among the cases and was 7.9 ± 2.6 among the controls (p < 0.000). LVMI (g/m2) was 72.1 ± 21.1 among the cases and was 62.7 ± 13.0 among the controls (p < 0.000). There was no significant difference between patients with HIV and controls on parameters such as LVEF, RVSP, TRVmax, LVESV, LVEDV, LVFS and TAPSE. LVEF was 65.3 ± 8.3 among patients with HIV and 66.1 ± 6.2 among the controls (p = 0.431). RVSP (mmHg) was 26.8 ± 10.0 among PLWH and was 24.8 ± 8.0 among controls (p = 0.107). LVESV (mL) was 25.2 ± 10.1 and 24.0 ± 7.6 among cases and controls respectively (p = 0.285). Table 4 shows unadjusted (univariable) and adjusted (multivariable) binary logistic regression of potential risk factors (HIV specific risk factors and traditional risk factors) and echocardiographic abnormalities. Increasing age (in years) was significantly associated with echocardiographic abnormalities on both univariable (OR = 6.7; 95% CI, 4.8–9.3; p = 0.018) and multivariable analyses (aOR = 5.8; 95% CI, 3.4–9.8; p = 0.040). Female gender compared to male gender was significantly associated with echocardiographic abnormalities on both univariable (OR = 5.7; 95% CI, 3.9–8.4; p = 0.005) and multivariable analysis (aOR = 4.8; 95% CI, 2.4–9.4; p = 0.033). Being overweight, abdominal obesity and increased fasting total cholesterol were, however, not significantly associated with echocardiographic abnormalities on univariable analysis. Viral load > 50 copies/mL compared to viral load < 50 copies/mL or undetectable was the only HIV-specific risk factor significantly associated with echocardiographic abnormalities on both univariable (OR = 5 ; 95% CI, 2.8–8.8; p < 0.001) and multivariable analyses (aOR = 4.4; 95% CI, 2.2–9.0 ; p < 0.001). Table 2HIV specific characteristicsCharacteristics (Cases)FrequencyN = 117Percentage (%) HIV Type Type 111497.4 Type 221.7 Both Type 1 & 210.9Mean HIV Duration, in years (mean ± SD)6.9 ± 6.0 Antiretroviral therapy NRTI + PI53.9 NRTI + NNRTI8673.8 NRTI + DTG2622.3Mean ART Duration, in years (mean ± SD)6.7 ± 5.8CD4 Count (cells/mm3) ≤ 2004235.6 > 2007564.4Viral Load (copies/ml) ≤ 50 or undetectable7060.0 > 504740.0ART: antiretroviral therapy; CD4: cluster of differentiation-4; DTG: dolutegravir; HIV: human-immuno-deficiency virus; IQR: interquartile range; NNRTI: non-nucleoside reverse transcriptase inhibitor; NRTI: nucleoside reverse transcriptase inhibitor; PI: Protease inhibitorReference: Adapted from [21] Table 2HIV specific characteristicsCharacteristics (Cases)FrequencyN = 117Percentage (%) HIV Type Type 111497.4 Type 221.7 Both Type 1 & 210.9Mean HIV Duration, in years (mean ± SD)6.9 ± 6.0 Antiretroviral therapy NRTI + PI53.9 NRTI + NNRTI8673.8 NRTI + DTG2622.3Mean ART Duration, in years (mean ± SD)6.7 ± 5.8CD4 Count (cells/mm3) ≤ 2004235.6 > 2007564.4Viral Load (copies/ml) ≤ 50 or undetectable7060.0 > 504740.0ART: antiretroviral therapy; CD4: cluster of differentiation-4; DTG: dolutegravir; HIV: human-immuno-deficiency virus; IQR: interquartile range; NNRTI: non-nucleoside reverse transcriptase inhibitor; NRTI: nucleoside reverse transcriptase inhibitor; PI: Protease inhibitorReference: Adapted from [21] HIV specific characteristics ART: antiretroviral therapy; CD4: cluster of differentiation-4; DTG: dolutegravir; HIV: human-immuno-deficiency virus; IQR: interquartile range; NNRTI: non-nucleoside reverse transcriptase inhibitor; NRTI: nucleoside reverse transcriptase inhibitor; PI: Protease inhibitor Reference: Adapted from [21] Table 3Echocardiographic characteristics of participantsParameterPLWHControlp-valueMean ± SDMean ± SDLA diameter (cm)3.1 + 0.43.1 ± 0.40.452AR (cm)2.5 ± 0.32.5 ± 0.30.716IVSd (cm)0.9 ± 0.20.8 ± 0.1 ≤ 0.001* IVSs (cm)1.2 ± 0.21.2 ± 0.20.403LVPWd (cm)0.9 ± 0.20.8 ± 0.1 ≤ 0.001* LVIDd (cm)4.3 ± 0.54.4 ± 0.50.203LVIDs (cm)2.6 ± 0.42.7 ± 0.40.242LVFS, 2D (%)39.8 ± 7.039.8 ± 5.80.920Mitral valve E (m/s)0.8 ± 0.60.8 ± 0.20.390Mitral valve A (m/s)0.1 ± 0.020.1 ± 0.020.173Mitral valve E/A1.2 ± 0.41.4 ± 0.4 ≤ 0.001* Mitral valve Dec time (msec)210.1 ± 47.2182.5 ± 43.5 ≤ 0.001* RVSP (mmHg)26.8 ± 10.024.8 ± 8.00.107Mitral valve e’, medial (m/s)0.1 ± 0.00.1 ± 0.00.322Mitral valve a’, medial (m/s)0.1 ± 0.00.1 ± 0.00.173Mitral valve e’, lateral (m/s)0.1 ± 0.00.1 ± 0.00.505LA volume (mL)32.2 ± 7.629.3 ± 5.4 0.002* LAVI (mL/m2)18.5 ± 4.816.3 ± 3.0 ≤ 0.001* LVEDV, (mL)73.2 ± 22.170.6 ± 17.30.310LVESV, (mL)25.2 ± 10.124.0 ± 7.60.285LVEF, modified Simpson (%)65.3 ± 8.366.1 ± 6.20.431Average E/e’9.1 ± 3.07.9 ± 2.6 0.004* TR Vmax (m/s)2.0 ± 0.61.9 ± 0.60.462TAPSE (cm)2.4 ± 2.32.2 ± 0.30.458LVM (g)127.3 ± 35.2113.5 ± 25.9 ≤ 0.001* LVMI (g/m2)72.1 ± 21.162.7 ± 13.0 ≤ 0.001* *p < 0.05 = Statistically significantA: Late diastolic mitral peak velocity; AR: aortic root; Dec Time: deceleration time; E: Early diastolic mitral peak velocity; e’: Early diastolic mitral annulus velocity by tissue Doppler; LVEF: left ventricular ejection fraction; FS: fractional shortening; IVS: interventricular septum; IVSd: interventricular septal diameter in end-diastole; IVSs: interventricular septal diameter in end-systole; LA: Left Atrium; LVPW: left ventricular posterior wall; LVIDd: Left Ventricular Internal Diameter in end-diastole; LVIDs: left ventricular internal diameter in end-systole; LAVI: left atrial volume index; LVM: left ventricular mass; LVMI: left ventricular mass index; RVSP: right ventricular systolic pressure; PLWH: people living with human immune-deficiency virus; TR Vmax: Maximal tricuspid regurgitation velocity; TAPSE: tricuspid annular plane systolic excursions Table 3Echocardiographic characteristics of participantsParameterPLWHControlp-valueMean ± SDMean ± SDLA diameter (cm)3.1 + 0.43.1 ± 0.40.452AR (cm)2.5 ± 0.32.5 ± 0.30.716IVSd (cm)0.9 ± 0.20.8 ± 0.1 ≤ 0.001* IVSs (cm)1.2 ± 0.21.2 ± 0.20.403LVPWd (cm)0.9 ± 0.20.8 ± 0.1 ≤ 0.001* LVIDd (cm)4.3 ± 0.54.4 ± 0.50.203LVIDs (cm)2.6 ± 0.42.7 ± 0.40.242LVFS, 2D (%)39.8 ± 7.039.8 ± 5.80.920Mitral valve E (m/s)0.8 ± 0.60.8 ± 0.20.390Mitral valve A (m/s)0.1 ± 0.020.1 ± 0.020.173Mitral valve E/A1.2 ± 0.41.4 ± 0.4 ≤ 0.001* Mitral valve Dec time (msec)210.1 ± 47.2182.5 ± 43.5 ≤ 0.001* RVSP (mmHg)26.8 ± 10.024.8 ± 8.00.107Mitral valve e’, medial (m/s)0.1 ± 0.00.1 ± 0.00.322Mitral valve a’, medial (m/s)0.1 ± 0.00.1 ± 0.00.173Mitral valve e’, lateral (m/s)0.1 ± 0.00.1 ± 0.00.505LA volume (mL)32.2 ± 7.629.3 ± 5.4 0.002* LAVI (mL/m2)18.5 ± 4.816.3 ± 3.0 ≤ 0.001* LVEDV, (mL)73.2 ± 22.170.6 ± 17.30.310LVESV, (mL)25.2 ± 10.124.0 ± 7.60.285LVEF, modified Simpson (%)65.3 ± 8.366.1 ± 6.20.431Average E/e’9.1 ± 3.07.9 ± 2.6 0.004* TR Vmax (m/s)2.0 ± 0.61.9 ± 0.60.462TAPSE (cm)2.4 ± 2.32.2 ± 0.30.458LVM (g)127.3 ± 35.2113.5 ± 25.9 ≤ 0.001* LVMI (g/m2)72.1 ± 21.162.7 ± 13.0 ≤ 0.001* *p < 0.05 = Statistically significantA: Late diastolic mitral peak velocity; AR: aortic root; Dec Time: deceleration time; E: Early diastolic mitral peak velocity; e’: Early diastolic mitral annulus velocity by tissue Doppler; LVEF: left ventricular ejection fraction; FS: fractional shortening; IVS: interventricular septum; IVSd: interventricular septal diameter in end-diastole; IVSs: interventricular septal diameter in end-systole; LA: Left Atrium; LVPW: left ventricular posterior wall; LVIDd: Left Ventricular Internal Diameter in end-diastole; LVIDs: left ventricular internal diameter in end-systole; LAVI: left atrial volume index; LVM: left ventricular mass; LVMI: left ventricular mass index; RVSP: right ventricular systolic pressure; PLWH: people living with human immune-deficiency virus; TR Vmax: Maximal tricuspid regurgitation velocity; TAPSE: tricuspid annular plane systolic excursions Echocardiographic characteristics of participants *p < 0.05 = Statistically significant A: Late diastolic mitral peak velocity; AR: aortic root; Dec Time: deceleration time; E: Early diastolic mitral peak velocity; e’: Early diastolic mitral annulus velocity by tissue Doppler; LVEF: left ventricular ejection fraction; FS: fractional shortening; IVS: interventricular septum; IVSd: interventricular septal diameter in end-diastole; IVSs: interventricular septal diameter in end-systole; LA: Left Atrium; LVPW: left ventricular posterior wall; LVIDd: Left Ventricular Internal Diameter in end-diastole; LVIDs: left ventricular internal diameter in end-systole; LAVI: left atrial volume index; LVM: left ventricular mass; LVMI: left ventricular mass index; RVSP: right ventricular systolic pressure; PLWH: people living with human immune-deficiency virus; TR Vmax: Maximal tricuspid regurgitation velocity; TAPSE: tricuspid annular plane systolic excursions Table 4Predictors of echocardiographic abnormalities Risk factors UnadjustedAdjusted OR (95% CI) Z P-value OR (95% CI) Z P-value Traditional risk factors Age 6.7 (4.8–9.3)2.190.0185.8 (3.4–9.8)2.06 0.040* Gender - Female5.7 (3.9–8.4)-3.210.0054.8 (2.4–9.4)-2.14 0.033* BMI - Overweight2.60 (1.2–5.5)2.500.112 Waist Circumference - Abdominal Obesity0.83 (0.44 − 1.6)-0.560.576 Total cholesterol - ≥5.21.25 (0.84–1.87)1.100.270 HIV-related risk factors HIV Duration, in years1.02(0.96 − 1.1)0.680.496ART Duration, in years1.08 (0.43 − 2.7)0.160.871 Medication - NNRTI0.08 (0.01-0.63)-2.390.0170.11(0.01-1.1)-1.880.060 CD4 count - ≤ 200 cells/mm31.09 (0.56 − 2.1)0.250.802 Viral Load - > 50 copies/ml5 (2.8–8.8)3.740.0014.4 (2.2-9.0)3.68 0.001* * p < 0.05 = Statistically significantART: antiretroviral therapy, BMI: body mass index, HIV: human immune-deficiency virus, therapy NNRTI: non-nucleoside reverse transcriptase inhibitor Table 4Predictors of echocardiographic abnormalities Risk factors UnadjustedAdjusted OR (95% CI) Z P-value OR (95% CI) Z P-value Traditional risk factors Age 6.7 (4.8–9.3)2.190.0185.8 (3.4–9.8)2.06 0.040* Gender - Female5.7 (3.9–8.4)-3.210.0054.8 (2.4–9.4)-2.14 0.033* BMI - Overweight2.60 (1.2–5.5)2.500.112 Waist Circumference - Abdominal Obesity0.83 (0.44 − 1.6)-0.560.576 Total cholesterol - ≥5.21.25 (0.84–1.87)1.100.270 HIV-related risk factors HIV Duration, in years1.02(0.96 − 1.1)0.680.496ART Duration, in years1.08 (0.43 − 2.7)0.160.871 Medication - NNRTI0.08 (0.01-0.63)-2.390.0170.11(0.01-1.1)-1.880.060 CD4 count - ≤ 200 cells/mm31.09 (0.56 − 2.1)0.250.802 Viral Load - > 50 copies/ml5 (2.8–8.8)3.740.0014.4 (2.2-9.0)3.68 0.001* * p < 0.05 = Statistically significantART: antiretroviral therapy, BMI: body mass index, HIV: human immune-deficiency virus, therapy NNRTI: non-nucleoside reverse transcriptase inhibitor Predictors of echocardiographic abnormalities * p < 0.05 = Statistically significant ART: antiretroviral therapy, BMI: body mass index, HIV: human immune-deficiency virus, therapy NNRTI: non-nucleoside reverse transcriptase inhibitor Discussion In this study, increasing age, female gender, and viral load > 50 copies/ml were predictors of echocardiographic abnormalities on multivariable analysis. The pathophysiology of heart disease among PWLH is complicated and multifactorial. The combination of traditional cardiovascular risk factors and emerging risk factors all play contributory roles [22]. Establishing definite causative mechanisms has proven somewhat illusory despite the abundance of literature establishing a robust association between HIV and heart disease. The exact underlying mechanisms and biological processes remain elusive and incompletely defined. This has led to speculative and hypothetic explanations [23]. Again, the interrelation between HIV itself, ART, and traditional risk factors further potentiate the difficulty in delineating risk specifically attributable to each of them [22]. Some mechanisms and pathobiological pathways that have been proposed to elucidate the increased incidence of heart disease in PLWH include chronic inflammation, coagulation disorders, immune activation, renal disease, HIV disease itself and ART [24]. PLWH are known to have a high incidence of conventional risk factors such as hypertension, metabolic diseases including diabetes mellitus, obesity, smoking, substance abuse, and dyslipidemia. Aside from these factors, HIV-specific risk factors that have been suggested include ART, immune activation, and chronic inflammation in the setting of suppressed and treated infections in HIV [22]. Wise et al. in a study looking at sex differences in incident atherosclerotic cardiac disease among males and females with HIV found the advantage of women against men that is commonly seen in the general population tends to be diminished in high risk population such as patients with HIV [25]. Reasons purported for this difference included increased immune activation relative to viral load in women compared to men. Others include higher ART-induced inflammation in females compared to men [25, 26]. These could potentially explain female gender being predictive of echocardiographic abnormalities compared to men. Addo et al. found that following infection with HIV type 1, males may have up to 67% higher HIV viral loads and lower CD4 cell count than women, however, even at similar levels of viremia, females progress faster to AIDS than men [26, 27]. These may be attributable to gender-based differences in innate immunity as a result of differences in the ability of plasmacytoid dendritic cells production of interferon alpha [26]. Despite significant gaps in available literature and our general understanding of the pathobiology of gender based differences in HIV and AIDS disease progression, Solomon et al. also postulate that notwithstanding a higher prevalence of some cardiometabolic risk factors in women living with HIV compared to their male counterparts, there is poor risk factor management of conditions like hyperlipidemia among the female cohort with HIV infection [27]. Hamadou et al. [28] in Cameroon found WHO stage 3 or 4 and CD4 count < 200 cells/mm3 to be factors independently associated with echocardiographic abnormalities. Luo et al. [29] in China in a prospective multicenter cohort study involving over 300 PLWH found age to be the only risk factor associated with LV systolic dysfunction. They also found HIV infection, age, and hypertension and significant independent predictors of LV diastolic dysfunction and HIV infection and hypertension as a significant predictor of LVH. Risk factors that were considered in the multivariable analysis included male gender, lipid profile, systolic and diastolic blood pressure, CD4 count, and viral load. Roozen et al. [30] in South Africa found no relation between HIV or ART and LV systolic or LV diastolic heart function. In multivariable analysis, Mondy et al. [31] in the SUN study found a history of myocardial infarction, elevated hs-CRP, and current tobacco smoking as significant risk factors for LV systolic dysfunction. For LV diastolic dysfunction, the significant predictors were elevated hs-CRP, and hypertension while the current use of ritonavir was found as a significant risk factor for pulmonary hypertension. For LVH, the significant predictors were hypertension, diabetes, non-white race, and female sex. Vasconcellos et al. [6] also determined that CD4 cell count > 400 cells/mm3 was associated with smaller LV diastolic volume and RV area. Furthermore, they reported HIV-positive men had higher LVMI, left atrial area and more significant LV diastolic dysfunction than controls – findings comparable with our study. Hsue et al. [32] reported a higher prevalence of LVH among PLWH than controls and determined that LVH was independently associated with lower nadir CD4 cell count. This corroborates evidence that suggests immunodeficiency contributes to the occurrence of increased LVMI. Hsue et al. [32] also found that patients with HIV have greater odds of having LV diastolic dysfunction than controls even after adjusting for age and traditional risk factors. Mankwe et al. [33] demonstrated in a cross-sectional study involving 100 newly diagnosed ART naïve to 100 age and sex matched controls and in Port Harcourt, Nigeria and found PLWH with CD4 count ≤ 200 cell/ml were more likely to have LV diastolic dysfunction and increased LVMI although the differences were statistically not significant. Isiguzo et al. [17] in a hospital based cross-sectional study sought to investigate the determinants of HIV-related cardiac disease among 200 PLWH in North-Central Nigeria. The most common cardiac abnormalities found in this study were LV systolic dysfunction (10.5%), diastolic dysfunction (10%) and pericardial disease (8.5%). The study authors more importantly ascertained duration of HIV diagnosis and degree of immunosuppression as significant determinants of HIV-related cardiac disease [17]. The relationship between markers of HIV disease severity such as low CD4 count and high viral load and the consequent increased risk of cardiovascular disease is believed to be mediated through heightened inflammation and persistent immune dysfunction although the exact pathophysiological mechanisms have not been elucidated [34]. The prevalence of dilated cardiomyopathy among people living with HIV is reported to be about 4% [5, 35, 36]. This rate is believed to increase with increased duration of illness and CD4 counts of less than 400 cells/mm3 [5, 35]. Survival is reduced with a higher hazard ratio of death in HIV-associated cardiomyopathy compared to idiopathic dilated cardiomyopathy. The major pathogenic pathway through which HIV causes dilated cardiomyopathy among PLWH is believed to be myocarditis [5]. It is hypothesized that HIV virions directly invade and infect myocardial cells in a patchy distribution [35]. This process is characterized by lymphocytic infiltration of the myocardium with concomitant necrosis and degeneration of adjacent myocytes in a manner that is not typical of ischemic damage, which is observed in coronary artery disease [35]. The specific pathway through which HIV virions enter CD4 negative cells such as myocytes is obscure; however, dendritic cells are believed to act as reservoir cells and are considered major pathogenic players because of their role in the interaction between HIV and the myocytes [35]. Multifunctional cytokines are subsequently activated, leading to progressive and subsequent tissue destruction. This happens when activated macrophages initiate proapoptotic signaling through mitochondrial injury, activation of caspases, Fas ligand expression and tumor necrosis factor alpha receptor mediated signaling [37]. HIV viral load and duration of immunosuppression all contribute to the incidence and progression of HIV-associated dilated cardiomyopathy [5]. Limitations The cross-sectional descriptive study design limited the ability of this study to make causal inferences between HIV disease and any HIV-specific risk factors or traditional risk factors as the cause of the echocardiographic abnormalities. Secondly, this was a single-centre study and so the findings may not be generalizable to the whole of Ghana, the West-african sub-region or SSA considering the social, demographic and fiscal heterogeneity of the region. Henceforward, longitudinal multi-center and multi-national studies are required to adequately and comprehensively describe predictors of adverse cardiovascular outcomes among PLWH in the SSA region. Additionally, this study focused on PLWH on ART and controls. The findings are not generalizable to PLWH who are not on ART. Conclusion This study determined that cardiovascular disease was associated with markers of immunosuppression. Viral load > 50 copies/ml was found on multivariable analysis to be an independent predictor of echocardiographic abnormalities. This study also showed that traditional risk factors such as increasing age and female gender contribute to and are independent predictors of cardiac abnormalities among PLWH. HIV disease and traditional cardiac risk factors must be recognized and aggressively managed to reduce cardiovascular morbidity and mortality among PLWH especially among females given the faster progression to AIDS even at similar levels of viremia compared to men. Electronic supplementary material Below is the link to the electronic supplementary material. Supplementary Material 1 Supplementary Material 1