51 / 92
CARDIOVASCULAR JOURNAL OF AFRICA • Volume 27, No 4, July/August 2016
AFRICA
253
care and provide the much-needed information on this subject
in sub-Saharan Africa. This study analysed the ECG findings
in HIV-infected individuals, including those on HAART, and
HIV-negative subjects residing in Enugu, Nigeria.
Methods
This cross sectional study was carried out from November
2010 to November 2011 at the University of Nigeria Teaching
Hospital (UNTH), Enugu, south-east Nigeria. The procedures
followed were in accordance with the ethical standards of the
Helsinki Declaration (1964, amended most recently in 2008) of
the World Medical Association. This study was approved by the
ethics committee of the UNTH, Enugu, and written consent
was obtained from the subjects. Information obtained was
anonymised as far as possible.
Inclusion criteria were adult Nigerians aged 18 years and
older with confirmed HIV-positive serology. HIV screening was
by enzyme-linked immunoassay (ELISA) and confirmed by
Western-blot electrophoresis, while CD4 T-lymphocytes (CD4
cells) were quantified by flow cytometry.
Sample size was calculated using the Fisher’s formula:
15
n
=
z
2
pq
____
d
2
where
n
=
minimum sample size; z
=
95% confidence level i.e.
1.96; d
=
level of precision (0.075)
13
; p
=
maximum prevalence
reported in a study of a similar population
16
(13.6%); and q
=
1–p.
A minimum sample size of 80 was calculated. For the purpose
of the study, 100 HIV-positive patients who had not taken
HAART were recruited into the HIV-positive, HAART-naïve
group. One hundred patients who had received HAART for at
least three months were enrolled in the group of HIV-positive
patients on HAART. Another 100 controls with already known
HIV-negative serology were recruited from those being screened
for blood donation, marriage and insurance purposes.
Patients in end-stage AIDS disease, classified as category C
by the Centre for Disease Control, 1993, were excluded.
17
Other
subjects excluded were those under 18 years of age, individuals
with arterial hypertension, coronary artery disease or active
symptoms suggestive of ischaemic heart disease or congestive
cardiac failure, cardiomyopathy, peripheral or cerebrovascular
disease or diabetes mellitus. Further exclusion criteria were
patients who were pregnant or in pueperium, as well as those
with a significant history of tobacco and or alcohol use, or those
who used drugs known to affect the cardiovascular system.
All subjects were evaluated clinically and anthropometric
parameters such as height (m), weight (kg), body mass index (kg/
m
2
) and body surface area (m
2
) were assessed. Qualifying subjects
had a resting 12-lead surface ECG recording in the supine
position at a speed of 25 mm/s using a two-channel automated
Techmel ECG machine (USA), ECG-1101 model.
ECG tracings from each participant were analysed in the
standard fashion with the long lead II tracing serving as the
rhythm strip. Parameters analysed were heart rate, rhythm, P
wave (duration, shape), height (paroxysmal atrial complexes), PR
interval, QRS wave (duration, shape, height, axis), paroxysmal
ventricular complexes, QT interval, QTc, Q wave, T wave
(shape), ST-segment (shape), and R and S waves for ventricular
hypertrophy.
Echocardiography was also carried out on each of the
subjects using the SonoScape SS1-5000 machine and transducer
of frequency 3.5 MHz. M-mode, two-dimensional, pulsed-wave,
continuous-wave, tissue Doppler imaging and colour Doppler
assessments were done with the subject in the left lateral
decubitus position. Measurements were taken (in cm) using the
American Society of Echocardiography guidelines (leading-edge
methodology).
18
Statistical analysis
Statistical analysis of data was done using EPI INFO version 6
software. The chi-squared test was used to test the association
between categorical variables. Continuous variables were
analysed using the Student’s
t
-test. Comparison of mean
±
standard deviations of parameters across the three groups was
done using one-way ANOVA, and the Duncan
post hoc
multiple
comparison test was done to indicate means for groups in
homogenous subsets (means not significantly different) (Table 1).
A
p
-value
<
0.05 was taken as statistically significant.
Results
Three hundred adults were recruited for the study, comprising
the group of 100 HIV-positive patients on HAART, made up
of 49 males and 51 females, 100 HIV-positive HAART-naïve
patient group, made up of 48 males and 52 females, and 100
apparently healthy adults (control group), made up of 52 males
and 48 females. There was no significant difference in the gender
distribution of these three groups (
χ
2
=
0.347,
p
=
0.841) (Table 2).
The mean age of the HIV-positive patients on HAART was
35.85
±
8.94 years, that of the HIV-positive HAART-naïve
patients was 34.43
±
9.49 years, while that of the control group
was 35.76
±
9.74 years. There was no significant difference in
the mean age of the three groups (
F
=
0.72,
p
=
0.49). There was
no significant difference in the age groups of the patients and
controls (
χ
2
=
4.74, P
=
0.19) (Table 3).
Table 1. Comparison of mean
±
standard deviations of parameters
across the three groups using one-way ANOVA
Parameters
HIV-positive
on HAART
HIV-positive
HAART-naïve
Control
F
-value
p
-value
Age (years)
35.85
±
8.94
34.43
±
9.49 35.76
±
9.74 0.716 0.490
Weight (kg) 65.77
±
13.92* 62.40
±
12.45 68.69
±
8.67*
7.007 0.001
Height (m)
1.66
±
0.07* 1.66
±
0.09* 1.71
±
0.79 17.886
<
0.001
BMI (kg/m
2
) 24.14
±
4.55* 22.47
±
3.65 24.18
±
3.32* 6.301 0.002
BSA (m
2
)
1.77
±
0.17* 1.75
±
0.18* 1.81
±
0.15 4.420 0.013
Heart rate
(bpm)
82.92
±
14.08* 84.28
±
16.79* 68.77
±
8.02 40.232
<
0.001
*Duncan
post hoc
multiple comparison test indicating means for groups in
homogenous subsets (means not significantly different). BMI
=
body mass index,
BSA
=
body surface area.
Table 2. Gender distribution of the study groups
Groups
Male,
n
(%)
Female,
n
(%)
Total,
n
(%)
HAART
51 (51.0)
49 (49.0)
100 (100)
HAART-naive
48 (48.0)
52 (52.0)
100 (100)
Control
52 (52.0)
48 (48.0)
100 (100)
Total
151 (50.3)
149 (49.7)
300 (100)
χ
2
=
0.347,
p
=
0.841. HAART
=
highly active antiretroviral therapy.