CARDIOVASCULAR JOURNAL OF AFRICA • Vol 24, No 6, July 2013
AFRICA
241
curtailment of the looming epidemic.
This survey, representing patients in a developing country,
indicates that a large proportion of patients on LLDs are not
reaching the accepted LDL-C goals. While these percentages
are not dissimilar to those from other studies in Europe, they are
below what is currently achieved in North America. The percent
of patients at goal in North America has risen over the course
of six years, from the NHANES survey of 1999/2000 and the
follow up conducted in 2005/2006, indicating that increased
awareness and education can result in a greater percentage of
patients reaching LDL-C goals, despite the targets becoming
more stringent.
The current study indicates both ethnic and gender variances
in cardiovascular risk-factor distribution and control in South
Africa. Smoking was less prevalent in those of African ancestry
and very few black females were smokers. In general black
subjects used fewer cigarettes than their Caucasian counterparts
(possibly due to economic constraints). The South African Heart
of Soweto study confirms that African patients have the lowest
smoking prevalence, with patients of mixed ancestry twice as
likely, and Caucasian patients three-fold more likely to be current
smokers.
17
In the current study of patient on LLDs, patients of
mixed ancestry had the highest prevalence of smoking among
both males and females, followed by Asian males.
The majority of black subjects did not have a family history
of premature heart disease, which probably reflects the evolution
of the epidemiological transition in an urbanising population,
compared to Caucasians, Asians and patients of mixed ethnicity.
The Heart of Soweto study also noted that African patients were
least likely to be diagnosed with CAD, and showed similar data
to CEPHEUS SA for the Asian patients, who had the highest
prevalence of a family history of vascular disease.
Control of DM was particularly poor in both male and
female African subjects compared with their ethnic counterparts.
This may have been due to differences in access to guideline-
based management protocols. Despite a high prevalence of the
metabolic syndrome in African females, with poor control of
DM, their TG levels were the lowest of all subjects – male and
female.
The Heart of Soweto study noted that patients of African
descent had significantly lower total cholesterol (TC), LDL-C
and triglyceride (TG) levels compared to other ethnicities.
17
These patients were not receiving LLDs. In CEPHEUS SA, in the
African-ancestry group, the TGs were not elevated despite a high
prevalence of DM with poor control. This would lend credence
to the finding that this population group may inherently have low
TG levels, and the influence of DM on TGs may be muted.
The Heart of Soweto study confirmed a high prevalence
of obesity in patients of African ancestry (43% of the patients
having a body mass index greater than 30 kg/m
2
). This substantial
burden of obesity among African subjects points to an elevated
risk for the future development of DM. Given that DM in this
group is poorly controlled, the ameliorating influence of lipid-
lowering therapy on future cardiovascular risk could potentially
be undermined, or at the very least minimised.
The number of African-ancestry patients who had CAD was
low in proportion to the other ethnic groups; however these
percentages reflect a change in the prevalence of a disease that
was previously considered to be rare in this population. Other
studies from South Africa have indicated a prevalence of CAD
of less than 10% in the African population.
18
The prevalence of
CAD in African subjects receiving LLDs in the current study
was 15.9%.
The INTERHEART Africa study noted that patients of
African ancestry presented with myocardial infarction a mean of
3.8 years earlier than patients from the overall INTERHEART
study, and also found no inter-ethnic or gender differences.
1,19
Although data for the INTERHEART Africa study were drawn
from patients from sub-Saharan countries, more than 80% of
subjects were from South Africa, indicating that the data may be
comparable to the current CEPHEUS SA study.
Limitations
This study had the same limitations that apply to many surveys
that differ fundamentally from formal prospective studies. The
study population was drawn from those already on LLDs and
cannot be extrapolated to the general population. Although
attempts were made to sample patients from as wide a spectrum
as possible, potential selection bias may still have occurred.
All centres were located in urban areas, and the applicability to
patients of rural origin cannot be assumed.
The public sector provides healthcare to about 80% of the
South African population but it made up only about one-third
of the sample. Similarly, the study population does not strictly
reflect the ratios of the different ethnic groups residing in
South Africa. However all previous studies on lipid-lowering
therapy in South Africa were predominantly Caucasian based.
Private-sector patients were recruited from a wide variety of
both specialist and non-specialist practices. The public-sector
patients were predominantly recruited from tertiary-care lipid
and diabetes clinics. The majority of African patients came from
the public sector.
Several private-sector centres had practitioners who dealt
predominantly with patients with DM, and this could have
further swayed the emphasis of the results on the diabetic cohort.
DM is often associated with an increase in body mass index
and other anthropometric measures of obesity, and data from
a cohort with a high prevalence of DM may therefore not be
reflective of the general population.
As the veracity of the patient questionnaires was not tested,
the validity of the CVD history may have been inaccurate.
Measured clinical parameters (such as blood pressure) were from
a single visit and methods of measurement were not standardised
or checked, and therefore inaccuracies could have arisen. Causal
correlations were not established, and relationships should
therefore be interpreted with caution.
Conclusion
Management of lipid-lowering treatment in South Africa is
sub-optimal, and in general lags behind control achieved in
the more developed nations. Furthermore, other cardiovascular
risk factors are not receiving due attention and their prevalence
in this population remains high. For any serious impact to be
made on the looming epidemic of cardiovascular disease in the
underdeveloped world, more attention needs to be focused on
more aggressive treatment of dyslipidaemia as well as the other
cardiovascular risk factors and, in particular, diabetes mellitus
and obesity.
