CARDIOVASCULAR JOURNAL OF AFRICA • Vol 23, No 7, August 2012
AFRICA
387
ventricular filling time to cardiac cycle was 32.6
±
3.9%. Six
patients had a first-degree atrio-ventricular block (AVB1). There
was no relationship between the existence of an AVB1 and AVD
(Table 2). There was no correlation between the magnitude of
AVD and the duration of the PR interval (from the beginning of
the P wave to the beginning of the QRS complex) (
r
²
=
0.02,
p
=
0.37) or the QRS width (
r
²
=
0.01,
p
=
0.38).
Mean IVMD was 38.6
±
19.3 ms. The IVMD was significantly
higher in patients from group 1 than in group 2 (54.4
±
10.8 vs
36.2
±
18.7 ms,
p
=
0.01) (Table 1). Inter-VD was present in 19
patients (47.5%) with a mean IVMD of 55.8
±
10.6 ms. A greater
proportion of patients with inter-VD was observed in group 1
(QRS
≥
120 ms) compared to those in group 2 (87.5 vs 60%,
p
=
0.03). There was a trend of increasing importance of IVMD
according to QRS width (
r
²
=
0.09,
p
=
0.06).
The mean
Δ
S
was 66.5
±
46.9 ms. The ESD was similar in the
two groups. Intra-LVD was present in 28 patients (70%) with a
mean
Δ
S
of 84.8
±
46.2 ms. The proportion of intra-LVD was
similar in all groups, 87.5% in group 1 and 65.6% in group 2
(Table 3). In patients with intra-LVD, seven had an ESD
≥
100
ms (17.5% of the total population and 25% of the patients with
intra-LVD). Three of these seven patients had a QRS duration
≥
120 ms.
Discussion
The study shows that in dilated cardiomyopathy, NYHA III
to IV, cardiac dyssynchrony is common, regardless of QRS
duration. Only two patients (5%) did not exhibit dyssynchrony.
Intra-LVD seemed to be the most frequent type of mechanical
dyssynchrony. The prevalence of AVD was 40%. There was no
increase in this prevalence in the presence of AVB1.
Jurcut
et al.
6
found the prevalence of AVD was 45% among
58 patients hospitalised for DCM of different aetiologies.
Abnormal conduction of the atrio-ventricular node resulted in
delay between the atrial and ventricular contractions. Too much
delay means that the valve leaflets are open in the mid-plane
position as ventricular systole starts, resulting in pre-systolic
mitral and tricuspid regurgitation. Prolonged atrio-ventricular
delay also means that the diastolic filling period is abbreviated,
limiting diastolic volume.
7
In the CARE-HF study,
8
prolonged
PR interval and right bundle branch block (RBBB) were
predictors of favourable outcome in patients with NYHA stage
III to IV heart failure.
The prevalence of inter-VD was 47.5% in our study. Schuster
et al.
reported a prevalence of 45% for inter-VD in DCM.
9
This
prevalence was higher than in others series. Bader
10
reported a
prevalence of 23% in a population optimally treated for stable
heart failure. This difference may be explained by the stage of
heart failure. In the CARE-HF study, the prevalence of inter-VD
among patients recruited by the conventional criteria of cardiac
resynchronisation therapy was 62%.
8
In our study the exclusive cause of wide QRS was LBBB.
A number of patients (12) in our study population with normal
QRS duration showed inter-VD. A greater proportion of patients
with inter-VD was observed in group 1 (QRS
≥
120 ms)
compared to patients in group 2 (
p
=
0.03) (Table 4). This is in
accordance with other series of patients.
6,8,10,11
The IVMD was not significantly related to QRS duration even
if there was a trend of increasing importance of IVMD according
to QRS width (
p
=
0.06). Data in the literature showed a good
correlation between the IVMD and the width of the QRS.
6,10-12
This correlation improved slightly when patients with pulmonary
hypertension or right ventricular dysfunction were excluded.
12
We have shown a prevalence of 70% of intra-LVD, which
seemed to be the most frequent type of mechanical dyssynchrony,
irrespective of the QRS width (Table 3). Similar results were
observed in other studies.
6,10
Hospitalisation for heart failure
is a predictor of the existence of left ventricular dyssynchrony
and is significantly correlated with a high risk of early cardiac
decompensation.
10
There was a substantial proportion of patients
with normal QRS duration who also showed intra-LVD (21/32)
(Table 3). These findings point to the fact that the QRS width is
not an accurate predictor of mechanical dyssynchrony.
Inter-VD is the principal factor associated with contractile
impairment and is affected by CRT.
1
The evaluation of
dyssynchrony is done with CRT.
We noted that 21 patients with narrow QRS presented with
intra-LVD. They could potentially have benefitted from CRT
but if we rely on conventional criteria,
13
these patients would
have been excluded because they did not exhibit electrical
dyssynchrony (defined by a QRS duration
>
120 ms). Perez
de Isla
et al
.
14
noted that 38.4% of patients with QRS
<
120 ms
presented with intra-LVD. Ghio
et al
.
12
found 29.5% of patients
with QRS
<
120 ms had left ventricular dyssynchrony. This
prevalence was 72% in the study by Jurcut
et al
.
6
In our study, in patients with intra-LVD, seven had an ESD
≥
100 ms. Three of these seven had a QRS duration
≥
120 ms.
This cut-off value of 100 ms is an indicator of a good response
to CRT.
1
Improvement of left ventricular function after CRT is
predicted by tissue Doppler imaging echocardiography. Penicka
et al
.,
15
using a composite index of inter-VD and intra-LVD
longer than 100 ms, achieved 88% accuracy in identifying
patients who responded to CRT.
There were some study limitations. This study was performed
TABLE 2. RELATIONSHIP BETWEEN FIRST-DEGREE
ATRIOVENTRICULAR BLOCKAND
ATRIOVENTRICULAR DYSSYNCHRONY (
p
=
0.85)
Patients with
AVB1
Patients without
AVB1
Total
Patients with AVD
4
12
16
Patients without AVD
2
22
24
Total
6
34
40
TABLE 3. RELATIONSHIP BETWEEN QRS DURATIONAND
INTRAVENTRICULAR DYSSYNCHRONY (
p
=
0.39)
Group 1
(QRS
≥
120 ms)
Group 2
(QRS
<
120 ms) Total
Patients with intra-LVD
7
21
28
Patients without intra-LVD
1
11
12
Total
8
32
40
TABLE 4. RELATIONSHIP BETWEEN QRS DURATIONAND
INTERVENTRICULAR DYSSYNCHRONY (
p
=
0.03)
Group 1
(QRS
≥
120 ms)
Group 2
(QRS
<
120 ms) Total
Patients with inter-VD
7
12
19
Patients without inter-VD
1
20
21
Total
8
32
40
