32 / 76
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 28, No 2, March/April 2017
98
AFRICA
development of clinical hypertension.
37
Most physicians describe
it as a pre-sign to clinical hypertension. However, it has also been
established that it presents a high risk for cardiovascular and
coronary heart disease, even without the development of clinical
hypertension and should be given adequate interventional
attention by promoting early lifestyle modification to prevent
progression to blood pressure or other related chronic diseases.
Prehypertension, unlike hypertension, is most often without
symptoms, but both conditions have been shown to share similar
risk factors, such as age, overweight, obesity and hyperlipidaemia,
but to different extents. The odds of the risk factors are
usually higher in hypertension than in prehypertension. The
association of higher cardiac and haemodynamic characteristics
with hypertension and prehypertension has also shown similar
trends.
38
These minor differences may explain the differences
in predictive potentials of some anthropometric indices for
hypertension and prehypertension.
Lee and Kim in their studies suggested the use of combined
anthropometric indices in models to improve the predictive
potential for hypertension or related diseases. Our study shows
WC and WHtR added to the prediction of hypertension using
BMI when included in a model, however, this contribution was
not statistically significant using our decision rule. The association
of WHtR, WC and BMI with hypertension prevalence rate was
statistically significant and similar in magnitude on adjusting for
age, gender, physical activity and alcohol intake, but differed for
prehypertension. WC and BMI had a stronger (significant
p
<
0.05) association with prehypertension relative to PI and WHtR
(
p
>
0.05). This was consistent with the ROC curve analysis
result. This reinforces the fact that BMI, WHtR and WC were
equally good indicators of hypertension in this population; none
significantly outperformed the other.
Overall, the mechanism of association of general obesity
(BMI and PI) with hypertension may differ from that of central
obesity (WHtR and WC) with hypertension. As mentioned, the
changes due to the addition of WC and WHtR were generally
decremental, while that of PI was incremental when WC, WHtR
or PI was included in a model with BMI. This is to be expected
as WC and WHtR were strongly correlated (
τ =
0.95;
p
<
0.0001)
and BMI and PI were also correlated (
τ =
0.98;
p
<
0.0001). The
correlation between BMI, WC and WHtR (
τ
=
0.60,
τ
=
0.64
respectively) was equally strong. The normalisation of BMI by a
factor of 1/height (m) to give PI did not improve the predictive
power of BMI, as traditional BMI outperformed PI in this study.
Limitations
The limitations of the present study include: firstly, the cross
sectional nature of the study precludes conclusion about a
cause–effect relationship. A longitudinal study in this regard
is required; secondly, our study population was drawn from a
particular religious group in a state in south-eastern Nigeria.
This may not be a true representation of the Anambra state
population and therefore limits the application of our findings to
other populations. However, it is noteworthy that the population
not captured represented a negligible proportion of the major
population. Thirdly, our overall sample size was large but on
categorisation by age group, some age groups had small sample
sizes, which may have limited our statistical power to detect
better performance in some age categories.
Despite these limitations, the study has some strengths.
The study population was large and typical of an African
population, where there has been dearth of data of this sort. The
anthropometric and blood pressure measures were standardised.
Conclusion
This study showed BMI, WC, WHtR and PI were strongly
associated with blood pressure and were better potential
predictors of risk for hypertension and prehypertension than
the other indices tested. They performed well independently
and there was no evidence to show that WC, WHtR or PI
outperformed or statistically added to the prediction power of
BMI. Their prediction potentials were better in the male gender
and in predicting risk for hypertension than for prehypertension.
In practice, these anthropometric measures are surrogate
measures of body fat and are cost free, practical and easy to
interpret for healthcare providers and lay people.
39
In the context
of developing countries, indices of obesity (both general and
abdominal) could be used simultaneously but independently to
predict risk for both conditions, since they both performed well
and possibly define different mechanisms of the association of
obesity with hypertension and other cardiovascular disorders.
References
1.
Ekwunife OI., Udeogaranya, PO, Nwatu IL. Prevalence, awareness,
treatment and control of hypertension in a Nigerian population.
Health
2010;
2
: 731–735.
2.
Mohammad NAA. Anthropometric measurements are simple predic-
tors for metabolic syndrome.
E-Int Sci Res J
2011;
3
(4): 320–325.
3.
Sakurai M, Miura K, Takamura T, Ota T, Ishizaki M, Morikawa Y,
et al
. Gender differences in the association between anthropometric
indices of obesity and blood pressure in Japanese
Hypertens Res
2006;
29
(2): 75–80.
4.
Ishikawa-Takata K, Ohta T, Moritaki K, Gotou T, Inoue S. Obesity,
weight change and risk for hypertension, diabetes and hypercholester-
olemia in Japanese men.
Eur J Clin Nutr
2002;
56
(7): 601–607.
5.
Cassani RSL, Nobre F, Filho AP, Schmidt A. Relationship Between
blood pressure and Anthropometry in a cohort of Brazilian man: A
cross-sectional study.
Am J Hypertens
2009;
22
(9): 1–5.
6.
Duren DL, Sherwood RJ, Czerwinski SA, Lee M, Choh AC, Srervogel
RM,
et al
. Body composition methods: comparisons and interpretation.
J Diabetes Sci Technol
2008;
2
(6): 1139–1146
7.
Lee ZB, Kim ZY. A comparison of the predictive power of anthropo-
metric indices for hypertensive and hypotension risk.
PLOS One
2014;
9
(1): 1–11.
8.
Kashyap K. Comparative evaluation and correlation of different
anthropometric indices with blood pressure in adult populations.
Int J
Basic Appl Physiol
2012;
1
(1): 36–41.
9.
Chuang Y-C, Wang M-H, Huang D-H, Yang C-H, Lin J-D. To
construct a forecasting model of the anthropometric chronic disease risk
factor score.
Chang Gung Med J
2006;
29
(2): 135–142.
10. Lee J-W, Lim N-K, Back T-H, Park S-H, Park H-Y. Anthropometric
indices as predictors of hypertension among men and women aged
40–69 years in the Korean population: the Korean Genome and
Epidemiology Study
BMC Public Health
2015;
15
: 140–146.
11. Tostade Almeida R, Guimaraes de Almerda MM, Araujo TM.
Abdominal obesity and cardiovascular risk: performance of anthropo-
metric indexes in women.
Arq Bras Cardiol
2009;
2
(5): 345–350.