26 / 76
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 29, No 3, May/June 2018
156
AFRICA
work in concert to promote development of atherosclerotic
disease, with selectins facilitating activated leukocyte rolling
and the adhesion molecules permitting adhesion and passage
of leukocytes into the sub-endothelial space (site for atheroma
formation).
12
Expression of these molecules among virally
suppressed HIV-infected patients has been linked to both
systemic and arterial inflammation as well as immune activation.
HIV-infected patients experience excess arterial
inflammation,
13-16
which has been linked to an increase in
cardiovascular events.
17
One of the main drivers of arterial
inflammation is activated monocyte production of interleukin 6
(IL-6).
18,19
IL-6 has been strongly associated with atherosclerotic
disease
20-22
and all-cause mortality among HIV-infected persons
in Botswana and elsewhere.
22,23
Similarly, soluble CD163, a
haemoglobin scavenger protein released from activated
monocytes, has been linked to arterial inflammation and
all-cause mortality among HIV-infected persons.
24,25
More
recently, activated non-classical monocytes and VCAM-1 were
significantly associated with degree of carotid intima thickness
in a cohort of virally suppressed patients in the United States
of America.
26
Exactly how these markers of inflammation
and monocyte activation impact on endothelial function in
sub-Saharan Africa (SSA) is not widely known.
We therefore sought to compare the degree of endothelial
dysfunction among virally suppressed HIV-infected participants
on ART in Botswana with that of HIV-uninfected controls, who
had similar age and gender distributions, after controlling for
traditional cardiovascular risk factors. Within the same setting,
we sought to further assess whether biomarkers of inflammation
(IL-6) and monocyte activation (sCD163) or carotid intima–
media thickness (cIMT) were associated with biomarkers of
endothelial dysfunction among virally suppressed HIV-infected
participants. We hypothesised that HIV-infected participants
have elevated biomarkers of endothelial dysfunction when
compared with an HIV-uninfected control group. Additionally,
we hypothesised that both IL-6 and sCD163 independently
drive excess endothelial dysfunction among HIV-infected
participants.
Methods
Participants were randomly selected from a larger cross-
sectional study whose main aim was to assess sub-clinical
carotid atherosclerosis and immune activation among adult
HIV-infected patients compared to HIV-uninfected controls in
Gaborone, Botswana. All participants self-identified as black
Africans and were enrolled between February 2014 and April
2015. HIV-infected participants between 30 and 50 years of
age (roughly balanced by gender) were recruited from Princess
Marina Hospital Infectious Disease Care Clinic (PMH-IDCC).
All HIV-infected patients were required to have documented
dual-positive HIV enzyme-linked immunosorbent assay
(ELISA) testing or pre-treatment HIV RNA
>
400 copies/ml,
a minimum six months of documented use of ART, no change
in antiretroviral regimen in the six weeks preceding enrolment,
and documented HIV RNA
<
400 copies/ml throughout the six
months prior to enrolment. ART in all participants consisted
of two nucleoside/tide reverse transcriptase inhibitors (NRTI)
plus either a non-nucleoside reverse transcriptase inhibitor
(NNRTI) or a ritonavir-boosted protease inhibitor (PI). Patients
on national salvage regimen consisting of two NRTIs plus
combination of a ritonavir-boosted PI and an integrase inhibitor
were eligible to participate. We also recorded the lowest CD4
count prior to ART initiation (nadir CD4) and the last recorded
CD4 count prior to ART initiation (baseline CD4 count). All
HIV RNA results were available from laboratory testing during
routine clinic visits. HIV-uninfected participants were enrolled
at the main Gaborone voluntary HIV testing centre and were
required to have documented same-day dual-negative HIV
ELISA testing.
At enrolment, all participants provided written, informed
consent to participate. Each participant had all study-related
procedures completed on the day of enrolment. A targeted
interview and review of medical records were performed to obtain
CVD risk history, including prior CVD events, family history
of CVD, diabetes mellitus or use of antidiabetic medications,
hypertension or use of antihypertensive agents, statin use,
cigarette smoking, and chronic kidney disease. Medical records
were also reviewed to obtain complete HIV disease history,
including diagnoses, pre-antiretroviral therapy and enrolment
HIV-associated laboratory results, and complete anti-retroviral
treatment history.
Furthermore, a focused physical examination was done to
obtain resting bilateral arm blood pressures and calculate body
mass index (BMI). Non-fasting plasma samples previously
obtained and frozen on the day of enrolment were used
to measure levels of biomarkers of endothelial dysfunction,
inflammation and monocyte activation. Non-fasting lipid
profiles and glycosylated haemoglobin levels were available
from prior analysis of samples drawn on the same day as the
enrolment, when all study-related procedures were completed
(including storage of plasma that was thawed for the current
analysis).
The study protocol was approved by the Botswana Ministry
of Health Human Research Ethics Committee, Princess Marina
Hospital Ethics Committee and Partners Human Research
Committee in Boston, MA, USA.
The degree of endothelial injury was assessed using ELISA
commercially available kits fromR&Dsystems
©
products supplied
by Bio-techne
©
, Abingdon, United Kingdom. The ELISA kit
manufacturer’s instructions were followed to ascertain levels of
the following biomarkers: human soluble E-selectin (CD62E)
quantikine ELISA [lower limit of detection (LOD) 0.1 ng/ml,
upper LOD 8 ng/ml], human soluble vascular cell adhesion
molecule 1 (sVCAM-1)/(CD106) quantikine ELISA (lower
LOD 6.3 ng/ml, upper LOD 200 ng/ml) and human intercellular
adhesion molecule 1 (sICAM-1)/(CD54) non-specific allele
quantikine ELISA (lower LOD 0.6 ng/ml, upper LOD 40 ng/
ml) kits.
Soluble CD163 and IL-6 levels were measured using ELISA
(Trillium Diagnostics, Bangor, ME, USA & R&D systems
©
products) according to the manufacturer’s instructions. All
participants had VCAM-1 and ICAM-1 results above the
LOD. All participants had E-selectin levels above the LOD,
however, E-selectin was available in only 90% of the HIV-infected
participants and 58% of the HIV-uninfected controls due
to limited access to testing reagents. All testing was done on
thawed EDTA plasma samples at the Botswana–Harvard HIV
Reference Laboratory, Gaborone, Botswana.
We used mean common cIMT as a summary measure