Journal Article > ResearchFull Text
J Am Heart Assoc. 15 June 2021; Volume 10 (Issue 12); e019994.; DOI:10.1161/JAHA.120.019994
Siender M, Bibangambah P, Kim JH, Lankowski A, Chang JL, et al.
J Am Heart Assoc. 15 June 2021; Volume 10 (Issue 12); e019994.; DOI:10.1161/JAHA.120.019994
BACKGROUND
Although ≈70% of the world's population of people living with HIV resides in sub-Saharan Africa, there are minimal prospective data on the contributions of HIV infection to atherosclerosis in the region.
METHODS AND RESULTS
We conducted a prospective observational cohort study of people living with HIV on antiretroviral therapy >40 years of age in rural Uganda, along with population-based comparators not infected with HIV. We collected data on cardiovascular disease risk factors and carotid ultrasound measurements annually. We fitted linear mixed effects models, adjusted for cardiovascular disease risk factors, to estimate the association between HIV serostatus and progression of carotid intima media thickness (cIMT). We enrolled 155 people living with HIV and 154 individuals not infected with HIV and collected cIMT images at 1045 visits during a median of 4 annual visits per participant (interquartile range 3-4, range 1-5). Age (median 50.9 years) and sex (49% female) were similar by HIV serostatus. At enrollment, there was no difference in mean cIMT by HIV serostatus (0.665 versus 0.680 mm, P=0.15). In multivariable models, increasing age, blood pressure, and non-high-density lipoprotein cholesterol were associated with greater cIMT (P<0.05), however change in cIMT per year was also no different by HIV serostatus (0.004 mm/year for HIV negative [95% CI, 0.001-0.007 mm], 0.006 mm/year for people living with HIV [95% CI, 0.003-0.008 mm], HIV×time interaction P=0.25).
CONCLUSIONS
In rural Uganda, treated HIV infection was not associated with faster cIMT progression. These results do not support classification of treated HIV infection as a risk factor for subclinical atherosclerosis progression in rural sub-Saharan Africa.
REGISTRATION
https://www.ClinicalTrials.gov; Unique identifier: NCT02445079.
Although ≈70% of the world's population of people living with HIV resides in sub-Saharan Africa, there are minimal prospective data on the contributions of HIV infection to atherosclerosis in the region.
METHODS AND RESULTS
We conducted a prospective observational cohort study of people living with HIV on antiretroviral therapy >40 years of age in rural Uganda, along with population-based comparators not infected with HIV. We collected data on cardiovascular disease risk factors and carotid ultrasound measurements annually. We fitted linear mixed effects models, adjusted for cardiovascular disease risk factors, to estimate the association between HIV serostatus and progression of carotid intima media thickness (cIMT). We enrolled 155 people living with HIV and 154 individuals not infected with HIV and collected cIMT images at 1045 visits during a median of 4 annual visits per participant (interquartile range 3-4, range 1-5). Age (median 50.9 years) and sex (49% female) were similar by HIV serostatus. At enrollment, there was no difference in mean cIMT by HIV serostatus (0.665 versus 0.680 mm, P=0.15). In multivariable models, increasing age, blood pressure, and non-high-density lipoprotein cholesterol were associated with greater cIMT (P<0.05), however change in cIMT per year was also no different by HIV serostatus (0.004 mm/year for HIV negative [95% CI, 0.001-0.007 mm], 0.006 mm/year for people living with HIV [95% CI, 0.003-0.008 mm], HIV×time interaction P=0.25).
CONCLUSIONS
In rural Uganda, treated HIV infection was not associated with faster cIMT progression. These results do not support classification of treated HIV infection as a risk factor for subclinical atherosclerosis progression in rural sub-Saharan Africa.
REGISTRATION
https://www.ClinicalTrials.gov; Unique identifier: NCT02445079.
Journal Article > ResearchFull Text
Ann Glob Health. 12 March 2015; Volume 81 (Issue 1); 207.; DOI:10.1016/j.aogh.2015.02.977
Kim JH, Hemphill LC, Boum Y II, Bangsberg DR, Siedner MJ
Ann Glob Health. 12 March 2015; Volume 81 (Issue 1); 207.; DOI:10.1016/j.aogh.2015.02.977
BACKGROUND
Non-communicable diseases (NCDs) account for the majority of adult deaths worldwide, and 80% of these deaths occur in
low and middle-income countries (LMICs). The burden of NCDs in LMICs is predicted to grow with improvements in sanitation and infectious disease control, and will be altered by local diet, smoking rates, and HIV co-infection. There is a critical need to identify and implement low-cost, well-validated diagnostic tests to elucidate the epidemiology of NCDs, and enable diagnostic monitoring and ther- apeutic interventions. Moreover, tests that enable non-healthcare professionals to lead care provision will augment the scalability of this strategy. We recently completed implementation and evaluation of a bundle of point-of-care, low-cost diagnostics for NCD measurement in rural Uganda.
METHODS
We performed a cross-sectional cohort study in rural, southwestern Uganda of HIV-infected persons on antiretroviral therapy at the Mbarara Regional Referral Hospital and a control group of HIV-uninfected persons from the clinic catchment area. Three non-healthcare professional Ugandan staff completed a two- week intensive course to perform a series of point-of-care cardiovas- cular assessments, including portable electrocardiogram (EKG), ankle-brachial index (ABI), hemoglobin A1c testing (HbA1c), auto- mated blood pressure, and anthropometric measurements. An American medical student was trained through the University of Wisconsin Atherosclerosis Imaging Research Program to perform measurement of carotid intima-media thickness (CIMT). We assessed the quality and feasibility of each measurement by: 1) proportion of valid hemoglobin A1c results; 2) proportion of interpretable carotid ultrasound images as graded by a board-certified vascular cardiologist using the University of Wisconsin CIMT image quality assessment scale; and 3) correlation between brachial blood pressure measure- ments and automated systolic blood pressure measurements. The study received ethics approval from the Mbarara University of Science & Technology and Partners Healthcare. All participants provided written informed consent.
FINDINGS
105 HIV-infected and 90 HIV-uninfected individuals were enrolled in the study. None of the HbA1c tests were invalid (0/ 195). Of the 96 CIMT images reviewed, 86 (90%) were found to be of adequate quality, and 10 (10.4%) were not suitable for measure- ment. The right and left brachial blood pressure measurements had coefficients of determination of 0.79 and 0.72, respectively, with the automated systolic blood pressure measurements. Based on an esti- mate patient volume of 1,000 patients per year and measurement for 3 years, the cost for this array of tests, including capital equipment, would be approximately $28 per patient.
INTERPRETATION
Low-cost, portable, and well-validated point-of-care tests can be implemented by non-medical professionals in LMICs. Implementation evaluations should be pursued to assess the large- scale feasibility, scalability, and impact of this strategy.
Non-communicable diseases (NCDs) account for the majority of adult deaths worldwide, and 80% of these deaths occur in
low and middle-income countries (LMICs). The burden of NCDs in LMICs is predicted to grow with improvements in sanitation and infectious disease control, and will be altered by local diet, smoking rates, and HIV co-infection. There is a critical need to identify and implement low-cost, well-validated diagnostic tests to elucidate the epidemiology of NCDs, and enable diagnostic monitoring and ther- apeutic interventions. Moreover, tests that enable non-healthcare professionals to lead care provision will augment the scalability of this strategy. We recently completed implementation and evaluation of a bundle of point-of-care, low-cost diagnostics for NCD measurement in rural Uganda.
METHODS
We performed a cross-sectional cohort study in rural, southwestern Uganda of HIV-infected persons on antiretroviral therapy at the Mbarara Regional Referral Hospital and a control group of HIV-uninfected persons from the clinic catchment area. Three non-healthcare professional Ugandan staff completed a two- week intensive course to perform a series of point-of-care cardiovas- cular assessments, including portable electrocardiogram (EKG), ankle-brachial index (ABI), hemoglobin A1c testing (HbA1c), auto- mated blood pressure, and anthropometric measurements. An American medical student was trained through the University of Wisconsin Atherosclerosis Imaging Research Program to perform measurement of carotid intima-media thickness (CIMT). We assessed the quality and feasibility of each measurement by: 1) proportion of valid hemoglobin A1c results; 2) proportion of interpretable carotid ultrasound images as graded by a board-certified vascular cardiologist using the University of Wisconsin CIMT image quality assessment scale; and 3) correlation between brachial blood pressure measure- ments and automated systolic blood pressure measurements. The study received ethics approval from the Mbarara University of Science & Technology and Partners Healthcare. All participants provided written informed consent.
FINDINGS
105 HIV-infected and 90 HIV-uninfected individuals were enrolled in the study. None of the HbA1c tests were invalid (0/ 195). Of the 96 CIMT images reviewed, 86 (90%) were found to be of adequate quality, and 10 (10.4%) were not suitable for measure- ment. The right and left brachial blood pressure measurements had coefficients of determination of 0.79 and 0.72, respectively, with the automated systolic blood pressure measurements. Based on an esti- mate patient volume of 1,000 patients per year and measurement for 3 years, the cost for this array of tests, including capital equipment, would be approximately $28 per patient.
INTERPRETATION
Low-cost, portable, and well-validated point-of-care tests can be implemented by non-medical professionals in LMICs. Implementation evaluations should be pursued to assess the large- scale feasibility, scalability, and impact of this strategy.
Journal Article > ResearchFull Text
J Acquir Immune Defic Syndr. 11 April 2018; Volume 78 (Issue 4); 458-464.; DOI:10.1097/QAI.0000000000001696
Muiru AN, Bibangambah P, Hemphill LC, Sentongo R, Kim JH, et al.
J Acquir Immune Defic Syndr. 11 April 2018; Volume 78 (Issue 4); 458-464.; DOI:10.1097/QAI.0000000000001696
BACKGROUND
The utility and validity of cardiovascular diseases (CVD) risk scores are not well studied in sub-Saharan Africa. We compared and correlated CVD risk scores with carotid intima media thickness (c-IMT) among HIV-infected and uninfected people in Uganda.
METHODS
We first calculated CVD risk using the (1) Framingham laboratory-based score; (2) Framingham nonlaboratory score (FRS-BMI); (3) Reynolds risk score; (4) American College of Cardiology and American Heart Association score; and (5) the Data collection on Adverse Effects of Anti-HIV Drugs score. We then compared absolute risk scores and risk categories across each score using Pearson correlation and kappa statistics, respectively. Finally, we fit linear regression models to estimate the strength of association between each risk score and c-IMT.
RESULTS
Of 205 participants, half were females and median age was 49 years [interquartile range (IQR) 46-53]. Median CD4 count was 430 cells/mm (IQR 334-546), with median 7 years of antiretroviral therapy exposure (IQR 6.4-7.5). HIV-uninfected participants had a higher median systolic blood pressure (121 vs. 110 mm Hg), prevalent current smokers (18% vs. 4%, P = 0.001), higher median CVD risk scores (P < 0.003), and greater c-IMT (0.68 vs. 0.63, P = 0.003). Overall, FRS-BMI was highly correlated with other risk scores (all rho >0.80). In linear regression models, we found significant correlations between increasing CVD risk and higher c-IMT (P < 0.01 in all models).
CONCLUSIONS
In this cross-sectional study from Uganda, the FRS-BMI correlated well with standard risk scores and c-IMT. HIV-uninfected individuals had higher risk scores than HIV-infected individuals, and the difference seemed to be driven by modifiable factors.
The utility and validity of cardiovascular diseases (CVD) risk scores are not well studied in sub-Saharan Africa. We compared and correlated CVD risk scores with carotid intima media thickness (c-IMT) among HIV-infected and uninfected people in Uganda.
METHODS
We first calculated CVD risk using the (1) Framingham laboratory-based score; (2) Framingham nonlaboratory score (FRS-BMI); (3) Reynolds risk score; (4) American College of Cardiology and American Heart Association score; and (5) the Data collection on Adverse Effects of Anti-HIV Drugs score. We then compared absolute risk scores and risk categories across each score using Pearson correlation and kappa statistics, respectively. Finally, we fit linear regression models to estimate the strength of association between each risk score and c-IMT.
RESULTS
Of 205 participants, half were females and median age was 49 years [interquartile range (IQR) 46-53]. Median CD4 count was 430 cells/mm (IQR 334-546), with median 7 years of antiretroviral therapy exposure (IQR 6.4-7.5). HIV-uninfected participants had a higher median systolic blood pressure (121 vs. 110 mm Hg), prevalent current smokers (18% vs. 4%, P = 0.001), higher median CVD risk scores (P < 0.003), and greater c-IMT (0.68 vs. 0.63, P = 0.003). Overall, FRS-BMI was highly correlated with other risk scores (all rho >0.80). In linear regression models, we found significant correlations between increasing CVD risk and higher c-IMT (P < 0.01 in all models).
CONCLUSIONS
In this cross-sectional study from Uganda, the FRS-BMI correlated well with standard risk scores and c-IMT. HIV-uninfected individuals had higher risk scores than HIV-infected individuals, and the difference seemed to be driven by modifiable factors.
Journal Article > ResearchFull Text
J Glob Health. 1 June 2020; Volume 10; DOI:10.7189/jogh.10.010407
Yang IT, Hemphill LC, Kim J-H, Bibangambah P, Sentongo R, et al.
J Glob Health. 1 June 2020; Volume 10; DOI:10.7189/jogh.10.010407
Background
Cardiovascular disease (CVD) morbidity and mortality are increasing in sub-Saharan Africa (sSA), highlighting the need for tools to enable CVD risk stratification in the region. Although non-HDL-cholesterol (nHDL-C) has been promoted as a method to measure lipids without a requirement for fasting in the USA, its diagnostic validity has not been assessed in sSA. We sought to estimate: 1) the association between LDL-cholesterol (LDL-C) and nHDL-C, 2) the impact of fasting on their measurement, and 3) their correlation with carotid atherosclerosis, within a rural Ugandan population with high HIV prevalence.
Methods
We collected traditional CVD risk factors, blood for serum lipid levels, self-reported fasting status, and performed carotid ultrasonography in 301 participants in rural Uganda. We fit regression models, stratified by fasting status, to estimate associations between carotid intima media thickness (cIMT), LDL-C, and nHDL-C.
Results
Median age was 50 years (interquartile range = 46-54), 49% were female, 51% were HIV-positive, and at the time of blood collection, 70% had fasted overnight. Mean LDL-C, nHDL-C, and triglycerides in the non-fasting and fasting groups were 85 vs 88 mg/dL (P = 0.39), 114 vs 114 mg/dL (P = 0.98), and 130 vs 114 mg/dL (P = 0.05) mg/dL, respectively. In unadjusted models, mean cIMT (mm) was associated with both increased LDL-C (β = 0.0078 per 10mg/dL, P < 0.01) and nHDL-C (β = 0.0075, P < 0.01), and these relationships were similar irrespective of fasting status. After adjustment for traditional CVD risk factors, we observed similar associations, albeit with muted effect sizes within the fasting group.
Conclusions
We found a high correlation between LDL-C and nHDL-C, and both were correlated with cIMT, irrespective of fasting or HIV serostatus in rural Uganda. Our findings support use of either fasting or non-fasting serum lipids for CVD risk estimation in rural sSA.
Cardiovascular disease (CVD) morbidity and mortality are increasing in sub-Saharan Africa (sSA), highlighting the need for tools to enable CVD risk stratification in the region. Although non-HDL-cholesterol (nHDL-C) has been promoted as a method to measure lipids without a requirement for fasting in the USA, its diagnostic validity has not been assessed in sSA. We sought to estimate: 1) the association between LDL-cholesterol (LDL-C) and nHDL-C, 2) the impact of fasting on their measurement, and 3) their correlation with carotid atherosclerosis, within a rural Ugandan population with high HIV prevalence.
Methods
We collected traditional CVD risk factors, blood for serum lipid levels, self-reported fasting status, and performed carotid ultrasonography in 301 participants in rural Uganda. We fit regression models, stratified by fasting status, to estimate associations between carotid intima media thickness (cIMT), LDL-C, and nHDL-C.
Results
Median age was 50 years (interquartile range = 46-54), 49% were female, 51% were HIV-positive, and at the time of blood collection, 70% had fasted overnight. Mean LDL-C, nHDL-C, and triglycerides in the non-fasting and fasting groups were 85 vs 88 mg/dL (P = 0.39), 114 vs 114 mg/dL (P = 0.98), and 130 vs 114 mg/dL (P = 0.05) mg/dL, respectively. In unadjusted models, mean cIMT (mm) was associated with both increased LDL-C (β = 0.0078 per 10mg/dL, P < 0.01) and nHDL-C (β = 0.0075, P < 0.01), and these relationships were similar irrespective of fasting status. After adjustment for traditional CVD risk factors, we observed similar associations, albeit with muted effect sizes within the fasting group.
Conclusions
We found a high correlation between LDL-C and nHDL-C, and both were correlated with cIMT, irrespective of fasting or HIV serostatus in rural Uganda. Our findings support use of either fasting or non-fasting serum lipids for CVD risk estimation in rural sSA.