CARDIOVASCULAR JOURNAL OF AFRICA • Vol 21, No 4, July/August 2010
208
AFRICA
Anthropometric measurements
All measurements were standardised and taken in triplicate to the
nearest 0.1 cm by registered biokineticists. Height was measured
by making use of a stadiometer while the participant’s head was
in the Frankfurt plane.
27
Weight was measured to the nearest 0.1
kg using a Krups scale with the participants wearing minimal
clothing. These measurements were used for the calculation of
body mass index (body mass/height
2
).
28
Physical activity was
measured using an Actical
®
accelerometer (Montréal, Québec).
29
Waist circumferences were measured with a metal tape at the
midpoint between the lower costal border and the iliac crest,
perpendicular to the long axis of the trunk.
27
Cardiovascular measurements
A Cardiotens apparatus (Meditech CE0120
®
) was used for the
24-hour ambulatory blood pressure measurement (SBP and
DBP) and a 12-lead ECG (Norav PC-1200) was applied to
obtain six resting cardiac cycles. Non-invasive rested continu-
ous arterial blood pressure recordings were obtained for five
minutes using the Finometer device (Finapres Medical Systems,
Amsterdam, the Netherlands). The Fast Modelflo computer soft-
ware programme analysed the results to provide: mean arterial
pressure (MAP) and total peripheral resistance (TPR), arterial
compliance (Cw) and heart rate (HR). Left ventricular hypertro-
phy (LVH) was calculated from the 12-lead ECG device using
the following gender-specific formula:
Cornell product: sum of all the leads, (RaVL
+
SV3
≥
2.8 mV in
men and
≥
2.0 mV in women)* QRS
>
244 ms.
30
Statistical analysis
All data were analysed by means of the computer software pack-
age STATISTICA 8 (StatSoft, Inc., Tulsa, OK, USA, 2008). All
data were normally distributed, hence parametric methods were
used. A single 2
×
2
×
2 (hypertension
×
depression
×
gender)
analysis of covariance (ANCOVA) was done to evaluate the main
effects interactions for cardiovascular and psychological distress
data. Subsequent 2
×
2 ANCOVA (24-h HT
×
depression) in
men and women and (gender
×
depression) analyses followed.
The prevalence of smoking, alcohol consumption, hypertension,
hypertension medication, physical activity (PAI) and depression
were computed using the two-way Pearson Chi-square analysis.
One-way ANCOVA was used to compare the psychological and
cardiovascular variables between the hypertensive and normoten-
sive gender groups while independent of confounders (age, BMI,
PAI, smoking and alcohol consumption).
Partial correlations followed to indicate associations between
cardiovascular variables (WC, Cw, MAP, 24-h SBP and the
Cornell product), depression (PHQ-9) and the common mental
health domains of depression (GHQ_DS), anxiety (GHQ_AS),
somatic symptoms (GHQ_SS) and social dysfunction (GHQ_
SD), as well as the GHQ_Total score (GHQ_T) separately in
the hypertensive men and women. Partial correlations were done
while adjusting for confounders (age, BMI, PAI and smoking
and alcohol consumption).
Logistic regression analysis was done using hypertension as
the dependent variable and GHQ_SS, PHQ_TT, Cornell product,
blood pressure, MAP and vascular compliance as the predictor
variables. The odds ratio was determined to measure effect size.
The reliability of GHQ-28 and PHQ-9 were determined by the
Cronbach alpha (
α
) reliability coefficient, which was between
0.77 and 0.83 for GHQ-28 and 0.81 for PHQ-9. Data were
considered statistically significant at
p
≤
0.05.
Results
The 2
×
2
×
2 (24-h HT
×
depression
×
gender) interactions were
not significant for either of the cardiovascular variables, depres-
sion and perception of health data. For exploratory reasons,
cardiovascular variables were evaluated in subsequent two-way
ANCOVAs, (24-h HT
×
depression) in men and women, which
showed significance for own perception of own health [GHQ-T
(F: 1, 91)
=
3.98,
p
=
0.05; GHQ-AS (F: 1, 91)
=
4.02,
p
=
0.05]
and GHQ-DS, [(F: 1, 91)
=
4.17,
p
= 0.05]. The 2
×
2 (gender
×
depression) interaction showed a significant interaction for LVH
[(F: 1, 163)
=
7.30,
p
=
0.01].
Table 1 shows that more men were hypertensive (79%) than
women (57%). The hypertensive men and women were older (
p
=
0.01), more obese (
p
<
0.01), and with larger waist circumfer-
ence (WC) (
p
=
0.05) compared with their normotensive coun-
terparts. The hypertensive groups also revealed a higher Cornell
product value only in HT men (
p
=
0.06) coupled to a lower
arterial compliance (
p
=
0.05), compared with the normotensive
groups.
Table 2 reveals that in HT men, blood pressure (systolic,
r
=
0.24 and diastolic,
r
=
0.30) was associated with perceived
health (GHQ_SS) and target end-organ damage (LVH). LVH
(
r
=
0.32) was positively associated with depression whereas it
was negatively associated (
r
=
–0.26) with vascular compliance
in HT men. In the HT women, HR correlated positively with
the perception of somatic symptoms
(r
=
0.30). To determine
the effect size of these associations in predicting hypertension,
a logistic regression analysis was performed. Hypertension
was used as the dependent variable and SBP, DBP, Cw, LVH,
GHQ_SS and PHQ_TT were independent predictors. In only the
women, depression and LVH showed an odds ratio of 1.15 (95%
CI = 1.01–1.32) and 1.02 (95% CI
=
0.998–1.05), respectively,
as predictors of hypertension.
Discussion
The main aim of this study was to investigate the interaction
between cardiovascular function and psychological distress in
urbanised black South Africans. The main findings of this study
were that in HT men, blood pressure (SBP and DBP) was associ-
ated with the perception of more somatic symptoms and LVH,
while LVH was also associated with depression (
p
=
0.001). In
HT women, HR was associated with the perception of more
somatic symptoms and MAP was associated with depression.
Additionally, this study showed that in HT African men, elevated
BP and lower vascular compliance were both associated with the
development of target end-organ damage.
Based on the DSM-IV depression criteria as measured with
the PHQ-9, more women revealed a trend of experiencing
depression compared to men, which is consistent with results
found in African-Americans.
31
When comparing the five levels
of severity of depression, however, a new trend emerged.
While women were more prone to suffer from minimal (6%)
to moderately severe (36%) depression, substantially more men
