CARDIOVASCULAR JOURNAL OF AFRICA • Vol 21, No 6, November/December 2010
318
AFRICA
mass by the volume, using the Siri equation.
23
From Db, the fat
percentage, fat mass, lean body mass and lung volume could be
obtained. Two measurements were taken and the aggregate was
used.
Trained professionals in private rooms used the Tanner stag-
ing scale questionnaire to estimate the physical maturity of the
boys and girls.
24
Blood samples and biochemistry
Registered nurse practitioners obtained fasting blood samples
from the subjects. The vena cephalica was used to draw 20 ml
venous blood for the preparation of EDTA plasma and serum.
Blood was centrifuged for 15 minutes at 4ºC and 2 000
×
g
for
serum and plasma preparations, divided into Eppendorff tubes
and frozen at –84ºC until the analyses were done.
High sensitivity C-reactive protein (hs-CRP) was measured
by immuno-nephelometry (Cardiophase hsCRP, Dade Behring,
2004) at an accredited laboratory (Ampath Laboratories,
Pretoria). Control serum with a concentration of 13.5 mg/l was
used as an external standard. The mean concentration of the
controls was 13.53 mg/l, with a range of 13.0–14.3 mg/l and a
coefficient of variation of 8.6%.
Physical activity and fitness
Trained field workers used the Previous day physical activity
recall questionnaire (PDPAR) to obtain information regarding
subjects’ physical activity of the previous day (24 hours) and a
24-hour day the previous weekend.
25
Cardiovascular fitness was determined by means of the bleep
test. Participants had to run a 20-m distance, timed by a metro-
nome. The running pace increased with each level (levels 1 to 6).
When a participant could no longer complete the distance within
the allocated time, the test was terminated and the correspond-
ing level of fitness was recorded. The levels were converted to
indirect maximum rate of oxygen consumption (VO
2
max) values
using a conversion calculator.
26
Statistical analysis
The Statistica computer data analysis software system from
Statsoft, Inc (2004), version 7 was used to analyse data. CRP
data were not normally distributed and were transformed loga-
rithmically. Descriptive statistics were used to describe the char-
acteristics of the subjects. The Mann-Whitney
U
-test, as well
as Kruskall-Wallis tests were performed to assess differences
between serum CRP levels of the children in the different catego-
ries of habitual physical activity, as well as to compare serum
CRP levels of children with normal or low percentage body fat
with those with high body fat percentages. Spearman correlation
was performed to assess the correlation between body composi-
tion variables, serum CRP (log transformed), reported physical
activity and fitness of the children.
Results
The sample comprised black children living in a poor socio-
economic setting. The type of housing was generally of galva-
nized zinc or brick with partial water and electricity supplies.
Subjects in the two schools were in a similar growth phase and
socio-economic status, with comparable eating habits and physi-
cal activity levels. Only 5.7% of the children admitted to smok-
ing, and one was a girl. The median age of starting to smoke was
15 years (interquartile range 13–17) and they smoked a median
of six cigarettes per day (interquartile range 2–10).
The serum CRP concentrations of the children ranged
between
<
0.2 and 39.8 mg/l. Three girls and three boys with
high serum CRP concentrations, above 10 mg/l, were excluded
from further analysis because this could have been due to acute
inflammation.
16,17
Most of the children had serum CRP concen-
trations within the normal range (
<
3 mg/l), and only a few had
raised levels. The children’s serum CRP concentrations were
then compared with regard to habitual physical activity level
and body fat percentage. Because the data were not normally
distributed, median and interquartile ranges were also calculated.
The descriptive statistics of girls and boys per category of
habitual physical activity are shown in Table 1. For the purpose
of this study, ‘inactive’ was defined as children with a score of
1, moderately active was a score of 2, and ‘most active’ was a
score of 3, according to the coding for the PDPAR, as proposed
by Weston
et al
.
25
There were significant differences between serum CRP
concentrations for the three physical activity categories of girls
and boys. The Kruskall-Wallis value for the difference in log
CRP in the girls was
H
=
7.33 (
p
=
0.025) and for the boys
H
=
7.5 (
p
=
0.02). The difference in CRP between activity groups
showed lower median serum CRP with higher physical activity in
the girls, as seen in Table 1. In the boys there was no clear trend,
with the highest median serum CRP in those with a moderate
physical activity level (Table 1).
With correlation analyses, in the girls there was a significant
positive correlation between the reported physical activity level
and their bleep test results (number of laps completed). The bleep
test result also correlated negatively with all body composition
variables indicating obesity (BMI, waist circumference, skin
folds, body fat percentage).
In the boys, a significant negative correlation was found
between the bleep test result and body fat percentage, as well as
skin fold thickness, but no correlation between bleep test result
and reported physical activity was found (Table 2). There was,
TABLE 1. DESCRIPTIVE STATISTICS OF CHILDREN PER CATEGORY
OF HABITUAL PHYSICALACTIVITY
(MEAN
±
SD/MEDIAN, INTERQUARTILE RANGE
#
)
Variable
n
* Inactive
n
Moderately
active
n
Most active
Girls
BMI (kg/m
2
)
59 20.43
±
2.94 35 19.96
±
3.60 10 20.22
±
3.02
Bleep test score 58 3.27
±
0.90 37 3.64
±
1.31 9 4.47
±
0.93
CRP (mg/l)
#
58 0.33 (0.15–0.94) 38 0.46 (0.26–1.75) 10 0.22 (0.21–0.26)
WC (cm)
60 64.54
±
5.90 38 63.82
±
6.42 10 63.35
±
4.13
TSF (mm)
60 16.60
±
5.91 38 14.50
±
4.47 10 14.12
±
5.13
Boys
BMI (kg/m
2
)
18 19.35
±
2.58 34 18.56
±
2.43 23 19.51
±
2.92
Bleep test score 18 6.43
±
2.49 33 6.2
±
1.75 23 7.00
±
1.91
CRP (mg/l)
#
17 0.19 (0.15–0.38) 32 0.59 (0.28–1.06) 21 0.31 (0.16–0.95)
WC (cm)
18 66.42
±
4.95 33 64.81
±
5.90 22 66.41
±
5.60
TSF (mm)
18 9.12
±
4.09 33 8.41
±
4.11 22 9.36
±
4.94
BMI: body mass index; CRP: C-reactive protein; WC: waist circumference; TSF:
triceps skin fold. *Number of subjects differs due to missing data.
