CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 1, February 2012
30
AFRICA
administered in order to identify participants’ preferred coping
strategies. The CSI is a factor analytically derived measure of
coping where the following three fundamental coping strate-
gies are revealed: problem solving, seeking social support, and
avoidance. Prevalence of smoking and alcohol consumption was
determined. Physical activity index (PAI) was determined, where
a score of 3 indicated vigorous and 1, indicated low activity.
Biochemical analysis
A registered nurse sampled fasting blood from brachial vein
branches with a winged infusion set, for serum oestrogen as well
as sodium fluoride (NaF) glucose levels. Analyses were done
with the sequential multiple analyser computer, KonelabTM
20i, Finland. Salivette
®
cortisol analyses were done with a high-
sensitivity enzyme-linked immunosorbant assay (ELISA). The
intra- and inter-assay coefficients of variation for cortisol were
7.7 and 9.8%, respectively. The antibody test to indicate positive
HIV status was done for each participant, including the First
response kit (Premier Medical Corporation LTD, Daman, India)
and Confirmatory Pareekshak test (Bhat Bio-tech India (P) LTD,
Bangalore, India).
Statistical analysis
A 2
×
2 analysis of covariance (ANCOVA) adjusted for age,
BMI and resting values (blood pressure and cortisol) was
conducted to determine interaction between the main effects,
i.e. gender and blood pressure status and each of the different
variables. Gender groups were stratified according to the ESH
guidelines (RivaRocci Korotkoff 2nd measurements) into BP
=
120–139/80–89 mmHg and BP
≥
140/90 mmHg,
20
hereafter
referred to as normotensive and hypertensive African men and
women. Coping scores were not normally distributed, therefore
median splits stratified participant into low and high respond-
ers. Subsequent one-way ANCOVAs followed, adjusted for age,
BMI, resting cortisol and blood pressure values.
Multivariate regression analyses were done. Firstly, partial
correlations, adjusting for age, BMI and resting values (blood
pressure and cortisol), identified independent variables for
forward stepwise regression analyses. Secondly, in forward
stepwise regression analyses we determined associations in
two separate models, i.e. model 1 (for each sampling time) and
model 2 (in all hypertensive groups). CIMT was the dependent
variable, and independent variables included age, resting and
reactivity blood pressure and cortisol values. Multivariate regres-
sion values with
p
≤
0.05,
r
≥
0.35 and adjusted
R
2
³ 0.25 were
regarded as significant.
Results
A single 2
×
2 ANCOVA showed significant interactions on
the main effects for low problem solving [df (1, 181)
=
4.01,
p
=
0.05] and TPR responses during the CWC test [df (1, 181)
=
4.31,
p
=
0.004]. In Fig. 1, the resting cortisol values were
significantly higher (
p
=
0.001) during cortisol sampling time 1:
06:00–07:00, than during sampling time 2: 08:30–09:00, in all
men and women.
FromTable 1 and Fig. 2, it is clear that hypertensive men were
older and showed increased vascular blood pressure during rest
(DBP and TPR as well as decreased Cw reactivity) and increased
mental stress during sampling time 1 (DBP and decreased Cw
reactivity) when compared to their normotensive counterparts.
In forward stepwise regression analyses (Table 2), age was asso-
ciated with progression of sub-clinical atherosclerosis in hyper-
tensive men during sampling time 1. Independent of sampling
times though, only vascular responses during the cold pressor
test predicted progression of sub-clinical atherosclerosis.
In Table 1 and Fig. 2, similar blood pressure trends to that
found in hypertensive men were found in hypertensive women,
namely they were older with increased vascular blood pressure
during rest (DBP, TPR and lower Cw) and increased mental stress
(DBP and TPR) compared to their normotensive counterparts in
time 1. Additionally, hypertensive women showed more central
obesity (WC:
p
=
0.01) compared to their normotensive coun-
terparts. No significant associations for hypertensive women
existed to predict progression of sub-clinical atherosclerosis.
Discussion
The aim of the study was to assess the relationship between
cardiovascular, cortisol and coping responses in urban Africans.
More men than women were hypertensive (63, 34%, respec-
tively) with no differences in their CIMT levels. As their CIMT
levels did not exceed 0.9 mm, which is the cut-off point for
diagnosed sub-clinical atherosclerosis, we only indicated trends
Figure 1. Resting cortisol values for men and women during cortisol sampling time 1 and 2 adjusted for age, BMI and
oestrogen.
3.5
3
2.5
2
1.5
1
0.5
0
Cortisol (ng/lml)
Cortisol sampling time 1:
06:30 – 07:00
Cortisol sampling time 2:
08:30 – 09:00
Resting cortisol values for sampling times 1 and 2 in men
3.07
1.39
p
≤
0.001
3.5
3
2.5
2
1.5
1
0.5
0
Cortisol (ng/lml)
Cortisol sampling time 1:
06:30 – 07:00
Cortisol sampling time 2:
08:30 – 09:00
Resting cortisol values for sampling times 1 and 2 in women
2.71
0.87
p
≤
0.001
