CARDIOVASCULAR JOURNAL OF AFRICA • Vol 22, No 3, May/June 2011
AFRICA
135
Methods
This sub-study is nested in the larger international PURE
(Prospective Urban and Rural Epidemiological) study. The inter-
national PURE study is a longitudinal, multinational study that
will address questions regarding the cause and development of
cardiovascular risk factors and chronic disease within popula-
tions in developing countries, including South Africa.
22
The South African leg of the study was performed in the
North West province where a total of 2 000 participants (1 000
urban and 1 000 rural) were randomly recruited from a rural and
urban setting and screened during the baseline phase in 2005.
The inclusion criteria were volunteers older than 35 years who
were non-users of any chronic medication and with no self-
reported diseases. For this case–control sub-study, 300 newly
identified HIV-infected participants of the baseline PURE study
population were individually matched with 300 HIV-uninfected
participants, according to age, gender, body mass index (BMI)
and locality (urban and rural). The protocol appropriate to this
sub-study will be discussed.
All participants provided signed informed consent after all
procedures were explained to them in their home language. The
study protocol complies with the Declaration of Helsinki as
revised in 2004,
23
and was approved by the Ethics Committee of
the North-West University, Potchefstroom, South Africa.
Permission to execute the study was obtained from the
provincial Department of Health, local authorities and from the
tribal chief in the rural area. Over a period of 12 weeks, 30 to 35
participants arrived at the research locality of the rural or urban
areas daily at about 07:00 each morning after a 10- to 15-minute
drive (provided by the research team) from their communities.
The participants were introduced to the setup and after the
procedures were explained, they signed the informed consent
forms and received HIV pre-counselling given by trained coun-
sellors. The HIV status of the participants was revealed during
individual post-counselling and the infected participants were
referred to their local clinics or hospitals for follow up and CD
4
cell count determination.
During the course of the morning, demographic, lifestyle
and food frequency questionnaires were completed with the
help of the specially trained field workers in the subjects’ home
language. Lifestyle data included self-reported current tobacco
use, alcohol intake as well as medical history. Height, weight,
hip and waist circumference (WC) were measured (Precision
Health Scale, A & D Company, Japan; Invicta Stadiometer, IP
1465, UK; Holtain unstretchable metal tape) using standardised
procedures.
24
Systolic (SBP) and diastolic blood pressure (DBP) were
obtained with the validated OMRON HEM-757 device. After
a 10-minute rest period, blood pressure measurements were
performed twice (five minutes apart) on the right arm (brachial
artery), while the participant was seated upright and relaxed
with his/her right arm supported at heart level. Appropriate cuffs
were used for obese participants. MAP pressure was calculated
by diastolic blood pressure plus one-third of pulse pressure.
Carotid-radialis pulse wave velocity (cr-PWV) was measured on
the left side of each participant in the supine position, making use
of the Complior SP (Artech-Medical, Pantin, France) apparatus.
Fasting blood samples were obtained from the antebrachial
vein using a sterile winged infusion set and syringes. Serum and
plasma were prepared according to appropriate methods and
stored at –80°C in the laboratory. In the rural area, serum/plasma
was stored at –18°C (no longer than five days) until it could
be transported to the laboratory facility, where it was stored at
–80°C until analysis.
Biochemical analyses
Quantitative determination of high-density lipoprotein choles-
terol (HDL-C), triglycerides (TG), high-sensitivity C-reactive
protein (hsCRP), glucose and creatinine concentrations in the
serum of the participants was done with the Konelab
TM
auto
analyser (Thermo Scientific, Vantaa, Finland). This is a clinical
chemistry analyser for colorimetric, immunoturbidimetric and
ion-selective electrode measurements. Creatinine clearance rate
was estimated using the Cockcroft-Gault formula.
Serum concentrations of high-sensitivity interleukin-6 (hsIL-
6) were measured using human enzyme-linked immunosorbent
assays (Quantikine
®
HS ELISA, R&D Systems, Minneapolis,
USA). Concentrations of serum intercellular adhesion molecule
1 (sICAM-1) and serum vascular cell adhesion molecule 1
(VCAM-1) were assessed by sandwich ELISAs (human sICAM-
1 and human sVCAM-1 assay, IBL, Hamburg, Germany).
The quantitative determination of fibrinogen
in plasma was
performed by the Multifibren U-test (Dade Behring), a modifica-
tion of the Clauss method on the Dade Behring BCScoagulation
analyser. The quantification of plasminogen activator inhibitor-1
(PAI-1) activity was performed by a chromogenic assay kit,
Spectrolyse
®
/pL PAI-1 (Trinity Biotech plc, Bray Co, Ireland).
HIV status was determined according to the protocol of the
National Department of Health of South Africa. A rapid card
test, First Response (PMC Medical, India) was used for testing,
using whole blood. If tested positive, the result was repeated with
the Pareeshak (BHAT Bio-tech India) card test for confirmation.
The HIV-1 subtype C epidemic prevalent in South Africa has
been established by serotyping and genotyping.
7,14,15
The CD
4
cell counts were obtained from the local clinic or hospital within
three months of the data collection of the baseline phase. CD
4
cell counts could only be obtained for 72 participants, as these
were the only participants who visited their local clinic or hospi-
tal for follow-up. The CD
4
counts were determined (in whole
blood) by the National Health Laboratory using flow cytometric
analysis (Beckman COULTER
®
EPICS
®
XL
TM
, Fullerton, USA).
Statistical analysis
All data were statistically analysed by means of Statistica v.8
(Statsoft Inc., OK, USA, 2008). Mean values, standard devia-
tions and standard errors were calculated. The distributions of
hsCRP, hsIL-6, sICAM-1, sVCAM-1, fibrinogen and PAI-1
were normalised by logarithmic transformation before analysis,
reporting the geometric mean and the fifth and 95th percentile
intervals. Independent
t
-tests were used to compare the unin-
fected group with the infected group and the group with the nadir
CD
4
cell count.
ANOVA and Tukey’s
post hoc
test for multiple comparisons
were used to compare the characteristics of the continuous
variables of the HIV-uninfected, -infected and nadir CD
4
cell
count groups. Chi-square tests were done to compare data of
categorical variables. An analysis of covariance (ANCOVA) and
Bonferoni
post hoc
test were performed to compare the inflam-
