15 / 66
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 30, No 6, November/December 2019
AFRICA
323
significance level to detect a coefficient of determination (
R
2
) of
0.042 or greater from the linear regressionmodel comprising three
predictors. Categorical variables are presented as frequencies and
percentages, while continuous variables are presented as mean
±
standard deviation (SD) for normally distributed variables,
and median and 25–75th percentiles for skewed variables. Data
were tested for normality using the Kolmogorov–Smirnov and
Shapiro–Wilk statistic.
The women were also split into two groups based on
menopausal age, which was estimated to be 50 years in this
population.
41
Group comparisons were made using the Mann–
Whitney
U
-test or chi-squared test. Robust regression analyses
were used to investigate the associations between body fat
distribution and cardiometabolic risk factors (insulin resistance,
lipid levels, blood pressure and inflammatory markers), adjusting
for age and gender. In addition, we explored the interactions
between gender and body composition on cardiometabolic risk
factors, adjusting for age, and in women, between menopausal
age and body composition.
To investigate whether one body compartment was more
closely associated with the risk factor than the other, coefficients
of determination were used from robust regressions. We
calculated the
R
2
for the model with covariates only (age and
gender), then the
R
2
for models containing covariates and each
of the adiposity measures.
Results
In all, 253 participants (18% men and 82% women) were
included. The average age of the participants was 55 years, and
was similar between men and women (
p
=
0.630). Differences in
body composition and body fat distribution between women and
men, as well as between pre- and post-menopausal women are
presented in Table 1. On average, the women were obese (mean
BMI
=
32.6
±
7.2 kg/m
2
), whereas the men were overweight
(mean BMI
=
27.4
±
6.1 kg/m
2
) (
p
<
0.001).
Men had higher fat-free soft tissue mass compared to women,
(
p
<
0.001), but body fat mass (kg and %) was significantly higher
in the women than men (
p
<
0.001). As a percentage of total fat
mass, women had significantly less central fat mass (
p
<
0.001)
and greater peripheral fat mass (arm, leg and gynoid fat %,
p
≤
0.003 for all) than men. VAT area was not different between men
and women (
p
=
0.474), but SAT area was higher in women than
men (
p
<
0.001).
When examining differences in body composition between the
pre- and post-menopausal women, we found that although there
were no differences in BMI, more post- than pre-menopausal
women were obese (68.9 vs 57.3%), and post-menopausal
women had greater fat mass (
p
=
0.026) and %FM (
p
<
0.001)
than pre-menopausal women. Although trunk fat mass (%)
and android fat mass (%) did not differ between pre- and post-
menopausal women (both
p
≥
0.415), post-menopausal women
had greater waist circumference and VAT (both
p
≤
0.004), and
less gynoid %FM (
p
=
0.001) than pre-menopausal women.
Differences in cardiometabolic risk factors between mixed-
ancestry men and women and between pre- and post-menopausal
women are described in Table 2. While blood pressure, fasting
glucose, insulin and lipid levels were not different between men
and women (all
p
≥
0.085), two-hour post-prandial glucose (
p
<
0.05) and HDL-C (
p
<
0.001) concentrations were higher
in women than men. The majority of the sample had normal
glucose tolerance (NGT) (men 63.1% and women 57.3%). The
Table 1. Comparison of body composition and body fat distribution between mixed-ancestry men and women, and pre- and post-menopausal women
Men
Women
Men vs women
Women
Pre- vs post-
menopausal women
Parameters
n
Total sample
n
Total sample
p
-value
n
20
–
49 years
n
≥
50 years
p
-value
Age (years)
46 53.5 (44.8–65.3)
207 55.0 (45.0–63.0)
0.630
75 39.0 (31.0–45.0)
131 61.5 (56.0–67.0)
<
0.001
Anthropometry
Height (cm)
46 168.0 (163.3–173.6) 206 156.0 (151.5–160.5)
<
0.001
75 157.5 (153.0–160.5) 131 155.0 (151.0–160.5)
0.043
Weight (kg)
46 75.6 (63.6–89.1)
206 78.2 (66.3–90.4)
0.443
75 74.5 (62.1–92.2)
131 79.8 (67.6–90.2)
0.196
BMI (kg/m
2
)
46
27.4
±
6.1
206
32.6
±
7.2
<
0.001
75
31.4
±
7.7
131
33.4
±
6.9
0.083
Waist (cm)
46
96.2
±
17.5
206
99.3
±
15.0
0.284
75
95.0
±
17.1
131
101.9
±
13.2
0.004
BMI category
46
%
207
% of total sample Pearson chi-squared
75
%
132
%
Pearson chi-squared
Underweight
2
4.3
3
1.4
<
0.001
2
2.7
1
0.8%
0.010
Normal
14
30.4
29
14.0
18
24
11
8.3
Overweight
16
34.8
41
19.8
12
16
29
22
Obese
14
30.5
134
64.7
43
57.3
91
68.9
DXA-derived body composition and body fat distribution
Fat-free soft-tissue
mass (kg)
46 50.4 (43.8–56.8)
207 38.9 (35.6–44.7)
<
0.001
75 38.9 (34.3–44.7)
132 37.3 (34.2–42.2)
0.243
Body fat (kg)
46 16.4 (12.7–27.8)
207 31.2 (24.4–40.0)
<
0.001
75
30.0
±
12.5
132
34.1
±
11.9
0.026
Body fat (%)
46 26.5 (19.9–32.5)
207 44.0 (39.8–48.6)
<
0.001
75 41.5 (35.3–46.8)
132 44.9 (41.4–49.6)
<
0.001
Trunk fat (%FM) 46
57.1
±
5.2
207
50.7
±
6.01
<
0.001
75
50.1
±
6.9
132
51.0
±
5.4
0.415
Arm fat (%FM)
46
10.7
±
1.5
207
12.5
±
1.97
<
0.001
75
12.1
±
1.8
132
12.7
±
2.0
0.059
Leg fat (%FM)
46 31.7 (29.0–34.9)
207 36.1 (31.9–40.5)
<
0.001
75 37.5 (32.0–41.5)
132 35.8 (31.5–40.2)
0.180
Android (%FM) 46
10.8
±
2.0
207
8.9
±
1.57
<
0.001
75
8.8
±
1.8
132
9.1
±
1.4
0.445
Gynoid (%FM)
46 15.6 (14.6–17.2)
207 17.2 (15.4–19.1)
0.003
75 18.1 (16.0–20.7)
132 16.9 (14.9–18.5)
0.001
VAT (cm
2
)
46 167.0 (101.2–260.7) 207 180 (135–236)
0.474
75 154.5 (93.2–211.0) 132 197.2 (149.4–244.1)
<
0.001
SAT (cm
2
)
46
263.8
±
143.7 207
451
±
142
<
0.001
75
432.6
±
160.6 132 461.0
±
129.5
0.220
Values presented as means
±
standard deviations (SD), median and 25–75th percentiles, or %. BMI (WHO classification), body mass index; WC, waist circumference; FM,
fat mass expressed as a percentage relative to sub-total fat mass; VAT, visceral adipose tissue; SAT, subcutaneous adipose tissue.