Cardiovascular Journal of Africa: Vol 29 No 3 (May/Junel 2018)

57 / 76

CARDIOVASCULAR JOURNAL OF AFRICA • Volume 29, No 3, May/June 2018 AFRICA 187
54.9 ± 10.9 years, found age to be an important predictor of CCA
IMT in both groups of subjects. Similarly, Ibinaiye et al.19 found
among hypertensives that CCA IMT increased with age from 21
to 70 years in a study population sample drawn from northern
Nigeria. Ren et al.6 also showed that middle-aged and older adults
with CVRFs displayed increased CIMT and higher grades of
severity than the younger age groups in Chinese subjects.
We found a relationship between traditional CVRFs and CA
in this study. Increased CIMT was independently predicted by
age ≥ 50 years (six times the unadjusted odds and 0.05 times the
adjusted odds in those under 50 years), hypertension (26 times
the unadjusted odds and 0.04 times the adjusted odds in those
without hypertension), intake of > 2 g/day of alcohol (7.5 times
the unadjusted odds and 0.07 times the adjusted odds in those
who had never drunk or not taken in the past year), obesity
(seven times the unadjusted odds and 0.2 times the adjusted odds
in the non-obese) and dyslipidaemia (13 times the unadjusted
odds and 0.03 times the adjusted odds in the non-obese).
A Brazilian study by Baroncici et al.20 among 533 CVRF
subjects with a mean age of 67.06 ± 12.44 years found male gender
in addition to hypertension and age to be risk factors that increased
CIMT. Gender was however not associated with increased CIMT
in our study. The differences in CVRFs associated with CA in
these studies could be due to ethnoracial differences, which should
be confirmed in larger multiracial studies.
Another key finding of this study was that CIMT increased as
the number or burden of CVRFs increased. Clustering of CVRFs
was seen in our sample population and the value of CIMT
paralleled the number of CVRFs in a linear, dose-dependent
fashion. The risk of atherosclerosis increases with increasing
burden of CVRFs. Previous studies6,21,22 have confirmed the
greater impact of multiple risk factors on CIMT than individual
CVRFs in different population groups despite differences in age,
number of risk factors and race of the subjects and the carotid
segments studied.
The prevalence of CP in this study was 16.1%. The prevalence
of CP reported from other studies varies quite widely, despite
comparability in sonographic methods. This may be explained
by differences in sample characteristics, prominent among which
are racial and environmental differences. Umeh et al.,13 in a study
of normotensive and hypertensive subjects, found an overall
prevalence of CP of 25.8% (29.2 and 22.5% in the hypertensive
and normotensive subjects, respectively), which is higher than
our prevalence despite similarities in the segment of carotid
vessel where CIMT and CP were measured and the location of
the studies. The older age of their sample compared to ours may
partly explain the differences in CP prevalence. Interestingly,
we found in our study that the presence of carotid plaque was
independently predicted by age ≥ 50 years (seven times the
unadjusted odds and 0.2 times the adjusted odds in those < 50
years) and hypertension (11 times the unadjusted odds and 0.3
times the adjusted odds in people without hypertension).
In support of our finding, a study in northern Nigeria, 19
which measured CIMT and CP in the CCA, similar to our study,
but in a slightly younger sample of hypertensive patients (mean
age of 50.62 ± 10.46 years) than ours, expectedly found a lower
CP prevalence of 10%. A hospital-based study similar to ours
and the other two Nigerian studies,13,19done among Brazilians, 20
with a mean age of 67.06 ± 12.44 years, found the prevalence of
CP to be 23.8% in their study, which is not unexpected, as the
mean age of the subjects in their study was higher than in the
three Nigerian studies.
Much higher prevalence of CP has been reported in the
literature in population-based studies such as the Northern
Manhattan Cohort Study (NOMAS),8 with a unique race/
ethnic distribution of community residents aged ≥ 39 years,
which reported CP prevalence of 58% overall, 70% in Caucasian
participants, 52% in Hispanics and 58% in blacks. Also, in
Beijing, China,23 the prevalence of CP was 60.3% among urban
residents aged 43–81 years, almost 70% in subjects ≥ 60 years,
and 80% in those ≥ 70 years. The population-based study setting,
which would have eliminated selection bias, in addition to the
lower age range of the participants in our study (23–81 years)
and the other three hospital-based studies by Umeh et al., 13
Ibinaiye et al.19 and Baroncici et al.,20 compared to the NOMAS
(65–74 years) and Beijing studies (≥ 75 years) might explain
the lower prevalence of CP found in our study and the three
hospital-based studies.
From our study, age ≥ 50 years, hypertension, dyslipidaemia,
obesity and alcohol intake > 20 g/day explained 78.7% of
the variance in CIMT, while age ≥ 50 years and hypertension
explained 38.0% of the variance in CP. This finding suggests that
CIMT and CP may be influenced by different CVRFs, although
age and hypertension influenced both. The relationships between
CVRFs and carotid atherosclerosis could be properly evaluated
in a longitudinal study.
It is not surprising that age and hypertension rank high in
the prediction of CIMT and CP because they happen to be the
most important risk factors for stroke. Age is the most important
non-modifiable risk factor for stroke, while hypertension is
the most important modifiable risk factor. Santos et al.,24 in a
multicentre Brazilian study, found traditional CVRFs explained
14.1 to 37.3% of the CIMT variance. Kuo et al.,8 in the NOMAS
study, found age, SBP, DBP, blood pressure- and lipid-lowering
medications and diabetes to be the traditional risk factors that
predicted CP and they explained 19.5% of variance in CP burden.
This difference can most likely be attributed to different
characteristics of the study populations (race, age), the carotid
segments measured, and study designs. Kuo et al.8 measured
near and far walls of the CCA, the bulb and internal carotid
artery (ICA) on both sides, while Santos et al.24 measured the far
wall of the CCA, similar to us. Our study was a hospital-based
study, in contrast to the population-based studies by Kuo et al.8
and Santos et al.24 Another hospital-based study like ours found
age, gender, pack-years of smoking, SBP, DBP, DM, HDL-C,
and blood pressure- and lipid-lowering medications to be the
most significant determinants of carotid plaque area, explaining
52% of the variance in total plaque area (TPA).25 Apart from
the difference in the predictors of CP between our study and
this hospital-based study, the difference in the measurement of
plaque considered (plaque thickness in our study versus total
plaque area in theirs) may explain the higher percentage of
variance in CP in that study.
Limitations
The evidence from this study is limited by its cross-sectional
design and hospital-based setting. The analysis was also limited
to the CCA, which might not have detected the presence of
atherosclerosis in other vascular beds or the more distal segments