78
VOLUME 10 NUMBER 2 • JUNE 2013
CONFERENCE REPORT
SA JOURNAL OF DIABETES & VASCULAR DISEASE
Interpretation:
Hypertension and its complications is the com-
monest cardiovascular disease in the Nigerian population in Abuja,
and unlike in Soweto, coronary artery disease is not common.
AETIOLOGY OF PULMONARY HYPERTENSION IN
AFRICA: PRELIMINARY DATA ANALYSIS AFTER ONE
YEAR OF RECRUITMENT: THE PAN AFRICAN PULMO-
NARY HYPERTENSION COHORT STUDY (PAPUCO)
Thienemann F*, Blauwet L, Dzudie A, Karaye KM, Mahmoud S,
Mbakwem A, Udo P, Mocumbi AO, Sliwa K
The PAPUCO Group, Hatter Institute for Cardiovascular Research in
Africa, University of Cape Town, South Africa
Introduction:
Pulmonary hypertension (PH) is a devastating, pro-
gressive disease, with increasingly debilitating symptoms and, usu-
ally, shortened overall life expectancy.
Subjects and methods:
A prospective observational study of
patients with newly diagnosed and previously untreated PH based
on echocardiography. Preliminary data analysis after 1 year of
recruitment is presented.
Results:
Among the 107 recruited cases, the median age was
41 years (range 1–86 years) with a female-to-male ratio of 1.5:1.
Cardiovascular (CV) risk factors were family history of CVD (34%),
hypertension (38%), hypercholesterolaemia (7%), diabetes (8%) and
smoking (5%). The HIV prevalence of the cohort was 25% with a
median CD4 count of 352 cells/µl (interquartile range(IQR) 201–516
cell/µl) at presentation. Twenty-five per cent of patients had previ-
ous episodes of TB. Presenting symptoms were shortness of breath
(SOB) (93%), fatigue (81%), palpitation (69%), cough (54%), cya-
nosis (14%) and syncope (or near syncope) (5%); 64% of patients
presented at WHO-FC III or IV. The mean right ventricle systolic pres-
sure (RVSP) was 56 mmHg (IQR 46–68 mmHg), whereas the median
RVSP in HIV-PH was 60 mmHg (53–73 mmHg,
p
=0.08). At the time
of writing, 6-month follow-up data were available for 33 patients.
Of those, 9 (27%) died within the first 6 months. All but one were
HIV-positive. Median time from diagnosis of PH to death was 3.0
months (IQR 1.5–3.7). The median RVSP at baseline was 72 mmHg
(IQR 60–83 mmHg) of patients who died within 6 months compared
to a median RVSP of 55 mmHg (IQR 45–65) at baseline of patients
alive at 6-month follow-up (
p
=0.03). Final diagnosis according to
WHO classification: idiopathic PH (2%), HIV-PH (11%), congenital
heart disease PH (5%), PH due to left heart disease (56%), PH due to
lung disease and hypoxia (17%), chronic thromboembolic PH (2%)
and unclear and/or multifactorial mechanisms (8%).
Interpretation:
Left heart disease, HIV, chronic lung disease and con-
genital heart disease are common contributors to PH in Africa. Disease
targeted therapy is not routinely available in the public sector. Out-
come of HIV-PH is very poor with RVSP being a prognostic marker.
DIABETES AND OBESITY
THE THIKA DIABETES STUDY: DIABETES IN NEWLY
ADMITTED PATIENTS AT A REFERRAL COUNTY HOSPI-
TAL IN KENYA
Kamotho C*
International Clinic, Nairobi, Kenya
Introduction:
Diabetes is estimated to increase by 161% by the year
2030 in sub-Saharan Africa as compared with 54% in the estab-
lished market economies of the West. INTERHEART Africa showed
that the odds ratio (OR) of diabetes leading to a new myocardial
infarction in Africans was 3.55 (2.53–4.99), higher than in the overall
INTERHEART Study, 3.07 (2.84–3.33). A study of black Africans with
angiographically evident coronary artery disease showed that diabe-
tes was significantly more prevalent in such individuals (38.5%) than
in those with normal coronary arteries (12%,
p
=0.0002).
Subjects and methods:
This prospective, descriptive study of
disease prevalence, which was preceded by a pilot study, involved
blood sugar measurements of serially admitted patients at Thika
Level 5 Hospital, Kenya, in January 2010. A total of 145 patients
were studied and results of 141 were analysed. Random blood
sugar measurements were done for patients on admission followed
by fasting blood sugar (FBS) assessments.
Results:
Of the 141 patients, a total of 46 (32.6%) patients had FBS
of 7.0 mmol/l and above, hence had diabetes mellitus; 13 (9.2%)
were known diabetics and 33 (23.4%) of the 141 were newly diag-
nosed diabetics. Nineteen (13.5%) patients had impaired fasting
glucose, defined by FBS from 6.1 to 6.9 mmol/l. Only one of these
19 patients was a known diabetic, all the others being newly diag-
nosed with the abnormality. Therefore an important 65 (46%) of all
the patients in the study had abnormal glucose metabolism, and a
total of 51 (36%) were newly diagnosed with hyperglycaemia.
Interpretation:
The prevalence of diabetes mellitus (and impaired
fasting glucose) in newly diagnosed patients admitted at this
county hospital in Kenya was found to be considerably high. More
concerning still was the very high prevalence of newly diagnosed
diabetics. It is recommended therefore that aggressive screening be
done at the admission points of such hospitals to pre-empt cardio-
vascular complications.
DO ENVIRONMENTAL FACTORS, INCLUDING SLEEPING
PATTERNS, AFFECT THE PREVALENCE OF OBESITY IN AN
URBAN AFRICAN POPULATION?
Pretorius S*, Sliwa K, Stewart S, Carrington M, Crowther NJ
Soweto Cardiovascular Research Unit, University of the Witwa-
tersrand, Johannesburg, South Africa
Introduction:
Urbanisation and the nutrition transition in South
Africa are accompanied by an increase in CVD risk factors, particu-
larly obesity, in African populations. Furthermore, environmental
factors other than food intake and urbanisation are also known
to affect the prevalence of obesity. The aim of this study was to
examine the relationship of these environmental factors (educa-
tion, employment, exercise, smoking, sleeping patterns) with obes-
ity in an urban population of African subjects.
Subjects and methods:
We systematically collected data on 1 311
consecutive patients attending two pre-selected primary health
care clinics in Soweto, South Africa. Education, employment status,
level of exercise, smoking habits and variables related to sleep dura-
tion were obtained using questionnaires. Weight and height were
measured using standard procedures and body mass index (BMI)
was calculated.
Results:
The prevalence of obesity was 45.4% in females and
11.3% in males. A multiple regression model was designed using
variables that correlated with BMI in a univariate analysis with
p
<0.20. These variables were: age, gender, smoking, education,
