29 / 62
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 31, No 4, July/August 2020
AFRICA
191
We used a structured data-collection tool to record socio-
demographic characteristics, cardiovascular risk factors and
history of ART. Height and weight measurements were used to
determine body mass index (BMI).
Resting blood pressure (BP) was assessed with the
patient seated and after 10 minutes of rest using a digital
sphygmomanometer; a set of three readings was taken and the
average of the last two was used. We defined high BP as systolic
pressure ≥ 140 mmHg, diastolic pressure ≥ 90 mmHg or use of
antihypertensive medication.
Venous blood was collected to determine CD4 counts and the
presence of anaemia was defined as values of haemoglobin (Hb)
below 12 g/dl in women and 13 g/dl in men; severe anaemia was
defined as values of Hb below 7 g/dl. As per standard of care in
this health facility, we could not assess the lipid profile.
For cardiovascular assessment we obtained resting
electrocardiogramandperformedabbreviatedcardiacultrasound.
An experienced cardiologist performed all cardiac ultrasounds,
using a portable battery-powered SONOSITE machine. We
estimated left ventricular systolic function (LVSF) based on
the visual assessment of ventricle contractile performance,
wall motion in multiple bi-dimensional views and shortening
fraction. The subjects were classified as having normal LVSF
if the shortening fraction was between 51 and 72%, minimal
systolic dysfunction if the shortening fraction was between
40 and 50%, moderate systolic dysfunction if the shortening
fraction was 30 to 40%, and severe systolic dysfunction if the
shortening fraction was < 30%).
Pericardial effusion was defined as an echo-free space between
the visceral and parietal pericardia that persisted throughout
the whole cardiac cycle; it was graded as small (≤ 2 cm) or
large (> 2 cm) on two-dimensional pictures during diastole.
We estimated pulmonary arterial pressure by the presence of
tricuspid regurgitation and defined pulmonary hypertension
as a value of over 35 mmHg with or without dilated and/
or hypertrophied right ventricle. We assessed LV dilatation at
echocardiography by measuring the internal diameter in diastole
(LV diastolic diameter/body surface area > 3.1 cm/m
2
).
9
After 24 months we re-evaluated the patients and assessment
of cardiac abnormalities on ultrasound. For patients who did
not come for ultrasound, data were collected from medical files.
At five years’ follow up we confirmed the vital status of all
patients via telephone interview.
Statistical analysis
Data were entered into an Epi Info version 7 data base
and analysed using the Statistical Package for Social Sciences
(SPPS) version 23. We express continuous data as mean (±
standard deviation, SD) and categorical data as number
(%). We summarised participants’ demographics and clinical
characteristics, and conducted univariate and multivariate
binary logistic regressions to determine the predictors of having
the different echocardiographic abnormalities. A
p
-value < 0.05
was considered statistically significant.
Results
We enrolled 264 HIV-positive adult patients with a mean age of
39.3 years (SD 9.8) (range 18–75) with a female predominance
(186, 70.5%) and most of black race (260, 98.5%). The average
BMI was 25.1 (SD 0.29) kg/m² (range 15.8–43.3); 45 (17%)
patients were obese (BMI > 30 kg/m
2
) and 65 (24.6%) were
overweight (BMI 25–29.9 kg/m
2
). High blood pressure was
found in 54 (20.5%) patients. All patients were in sinus rhythm
and none had signs of ischaemic heart disease on resting ECG.
The mean glycaemia was 4.3 mmol/l (range 1.1–17.6 mmol/l).
Diabetes mellitus was found in four (1.5%) patients and all were
on a specific treatment. Anaemia was found in 119/170 (70.0%)
women (range 5.7–11.9 g/dl) and 34/78 (43.5%) men (range
9.9–12.9 g/dl); it was severe in 21 patients.
According to the World Health Organisation (WHO) HIV
clinical classification, 173 (65.5%) patients were in stage I at the
time of the study, 40 (15%) were in stage II, 45 (17%) in stage III,
and the remaining six (2%) were in stage IV. High blood pressure
was present in 54 (20.5%) patients. The mean time on ART
was 46 (SD 36) months. Change of ART regimen had occurred
once in 75 (28.4%) and twice in 21 (7.9%) patients. The mean
CD4 count was 516 cells/ml (range 18.0–1 300) and 28 (10%)
individuals had CD4 counts less than 200 cells/ml. Low viral
load was found in 174 (65%) patients (Table 1).
Electrocardiograms were performed in 261 subjects, of whom
122 had some form of abnormality including LV hypertrophy
in 78 (63.9%), sinus arrhythmia in 31 (11.9%), non-specific
repolarisation pattern in 24 (9.2%), sinus bradycardia in 17
(6.5%), abnormal intraventricular conduction in eight (3.1%) and
early repolarisation in seven (2.7%). Data on X-ray evaluation
in 210 subjects revealed increased cardiothoracic index in 46
patients (21.9%), of whom 14 had confirmation of LV or atrial
dilatation on echocardiography.
Echocardiograms were performed on 252 patients at baseline
and 88 (34.9%) had cardiac abnormalities. The most frequent
abnormalities were mitral valve abnormalities, present in 39
patients (15.5%). Six patients had severe RHD (four mitral, two
with both mitral and aortic valves affected). The remaining 33
patients had minor mitral valve abnormalities, namely moderate
valve thickening and mild functional regurgitation without
definite signs of RHD.
Severe LV systolic dysfunction (SF < 30%) was present in 29
patients and 83 had moderate systolic dysfunction. Abnormal
relaxation pattern or grade I diastolic dysfunction was found
in 13 patients (5.2%); six had concurrent hypertension. Three
patients had mild to moderate degenerative aortic valve stenosis.
Congenital heart disease was found in three (1.2%) patients.
Pulmonary arterial hypertension and tuberculous pericarditis
were each found in two (0.7%) patients.
LV systolic dysfunction was symptomatic in only five patients
who had signs of congestive heart failure. The two patients
who had pulmonary arterial hypertension had dyspnoea. The
six patients with RHD had audible murmurs and signs of
heart failure that had gone unnoticed. The three patients with
congenital heart disease (CHD) had patent foramen ovale,
and restrictive ventricular septal and atrial septal defects. The
two patients with large pericardial effusion had tachycardia,
suggestive chest X-rays and were easily noted on abbreviated
ultrasound. Finally, patients with severe anaemia were not being
treated for its correction.
At the 24-month follow up of the 252 patients who had had
echocardiographic diagnosis, six had died, of whom four had
cardiac abnormalities at baseline. Two of these had RHD, one