CARDIOVASCULAR JOURNAL OF AFRICA • Vol 24, No 7, August 2013
282
AFRICA
criteria: adult subjects (
>
18 years); cardiovascular symptoms
or conditions such as palpitations, chest pain, dyspnoea,
syncope, hypertension; and ECG done during the study period.
The exclusion criteria were: any other causes of transient
loss of consciousness such as seizure (Table 1); subjects
with low-amplitude ECG waves; subjects with a permanent
pacemaker, as ventricular stimulation alters repolarisation; any
drug that could potentially modify the QRS, Q-T and ST-T
duration and morphology (Table 1); subjects who did not sign
the consent form.
Each subject filled in a questionnaire focusing on a past
history of transient loss of consciousness (TLOC) or family
history of sudden cardiac death (SCD). The diagnosis of syncope
was based on the European Society of Cardiology (ESC)
guidelines.
10
Detailed physical evaluation was performed. The
ethics committees of both hospitals approved the protocol.
A trained nurse recorded the ECGs using a paper speed of
25 mm/s at 10 mm/mV. After a resting period of at least 5 min,
the ECG was registered in the supine position using a numerical
electrocardiograph with the capability to review and modify the
value of the parameters. Resting 12-lead ECG for each subject
was analysed independently by two trained physicians.
A diagnosis of ER was retained if both examiners concluded
that at least two consecutive leads displayed a slurring or
notching pattern of the descending part of the R wave or a
prominent J wave with ST-segment elevation
≥
1 mm in the
lateral or inferior leads. We also paid attention to variation of the
J-T segment as ascending, descending or horizontal morphology.
For a practical understanding, we attributed the term type 1
(t1) to the slurring variant of ER and type 2 (t2) to the notching
variant. We divided our sample into four groups: ERt1
=
early
repolarisation type 1, ERt2
=
early repolarisation type 2, ERt1t2
=
mixed variant, and ER–
=
the normal pattern of repolarisation.
Statistical analysis
Continuous variables are expressed as means
±
SD and statistical
significance was assessed using the unpaired Student’s
t
-test,
or Mann-Whitney
U
-test where used to compare mean values
between two groups of subjects with ER (ER+) and without
(ER–). Categorical variables, were summarised as proportions,
and compared using the
χ
2
test or Fischer’s exact test. ANOVA
and the Kruskal-Wallis test were used to compare mean values
between more than two subgroups (ERt1, ERt2, ERt1t2, ER–).
The Bonferroni correction was used for adjustment in multiple
comparisons. All tests with a two-tailed
p
-value
<
0.05 were
considered statistically significant. Statistical analysis was
performed with SPSS version 11.0.1 software (SPSS Inc)
Results
Over the two-week period, ECGs were performed in 752 subjects,
among whom 248 had cardiac symptoms (palpitations, past
history of syncope, chest pain or dyspnoea) or cardiovascular
morbidity (hypertension, diabetes, heart failure). Two subjects
had low-amplitude ECGs, which did not permit analysis of the
repolarisation and they were excluded.
In the remaining 246 subjects studied, the mean age was 45
±
16 years and 53% were female. The group with ER was younger:
41
±
16 years versus 49
±
16 years in subjects without ER (
p
=
0.0048). Baseline characteristics of the sample are shown in
Table 2. There were 184 ambulatory subjects (75%).
Indications for an ECG were hypertension screening in
57% subjects, and palpitations or dizziness in 41%. Two of
the in-patients were diagnosed with acute heart failure, and 35
subjects (14.5%) had diabetes. TLOC was reported by 45 (18%)
subjects, 19 (41%) in the ER group and 26 (13%) in the group
without ER (
p
=
0.00014). A family history of sudden unexpected
death (SUD) was reported in three cases (6.5%) in the ER group
and in 14 cases (8.5%) in the group with normal repolarisation
(
p
=
ns). Atrial fibrillation was found in four (1.7%) subjects.
ER was observed in 20% of the population, with the following
distribution: slurring pattern in 3.3%, notching pattern in
13%, and both in 3.7%. Among the ER ECGs, we ascertained
ascending J-T morphology in eight, descending in five and
horizontal in 16 patients.
TABLE 1. DIFFERENTIAL DIAGNOSIS BETWEEN
SYNCOPEAND SEIZURE
Syncope
Seizure
Mechanism Global transient cere-
bral hypoperfusion
Abnormal excessive or
synchronous neuronal
activity
Age (years) at
first manifestation
Over 45 if coronary
artery disease
Mainly
<
45 if other
cardiac causes such as
channelopathies
Less than 45 (often, apart
from secondary seizure due
to brain damage)
Symptoms
Before TLOC Nausea
Vomiting
Sweating and body cold
Aura (funny smell)
Crying
During TLOC Brief clonic movement
(
<
15 sec) always
secondary to LOC
Prolonged clonic movement
at the beginning of LOC
Automatism
Blue face
After TLOC Nausea
Pale face
Normal orientation
Prolonged disorientation
Post-event amnesia
Weakness, courbature
TLOC
=
transient loss of consciousness; LOC
=
loss of consciousness.
TABLE 2. DEMOGRAPHIC, CLINICALAND ELECTROCARDIO-
GRAPHIC CHARACTERISTICS OF THE SAMPLE
Aetiology
ER+
(
n
=
49)
ER–
(
n
=
197)
p
-value
Age (years)
–
41 ± 16 49 ± 16 0.0048
Female (%)
–
53
47
ns
BMI (kg/m
2
)
–
29
28
ns
Hypertension,
n
(%)
–
34 (14) 103 (43)
ns
Diabetes,
n
(%)
–
12 (5) 24 (10)
ns
Palpitation,
n
(%)
–
19 (41) 65 (33)
ns
Syncope,
n
(%)
unknown 19 (41) 26 (13) 0.00014
Acute HF
hypertension 0
2
ns
Family history of SUD,
n
(%)
3 (6.5) 14 (8.6)
ns
T(–) wave,
n
(%)
14 (9)
2 (1)
0.00025
Drug therapy
Beta-blockers
0
0
Amiodarone
0
0
Other anti-arrhythmics
0
0
Psychotropic drugs
0
0
ER
=
early repolarisation; BMI
=
body mass index; HF
=
heart failure;
SUD
=
sudden unexpected death; T(–) wave
=
negative T wave.
