Gabriella Malfatto, Francesca Ciambellotti, Roberto Chianca, Renato Bragato, Giovanna Branzi, Gastone Leonetti*, Mario Facchini.
Divisione di Cardiologia, Istituto Scientifico Ospedale San Luca, Istituto Auxologico Italiano IRCCS, *Universita di Milano, Italy
|
|
The dispersion of the QT interval (QT dispersion=QTd)
on the surface ECG is thought to reflect the spatial and temporal recovery of repolarization in the
underlying myocardium1. In myocardial infarction and ischemia, and in long QT syndromes, an
increased QTd is associated with a propensity for malignant ventricular arrhythmias (VA) and
sudden cardiac death2-5. No data exist about this particular feature of the repolarization
process in subjects with frequent ventricular arrhythmias and otherwise structurally normal
hearts, although some data existed of an increased QTd in subjects with mitral valve prolapse6.
From patients referred to our arrhythmia unit for the evaluation of frequent, non sustained
ventricular arrhythmias, we retrieved the data of 49 of them meeting the following criteria:
1) the presence of >100 VA/h in at least 12 hours of two non consecutive Holter monitoring
(to ensure that the arrhythmic phenomenon was stable); 2) normal blood pressure; 3) no
heart disease, included mitral valve prolapse or left ventricular dilatation (end-diastolic left
ventricular diameter >55 mm), as proven by history, physical examination, 2D and M-mode
echocardiogram and colorDoppler, maximal exercise stress test; 4) no ventricular late potential,
determined by the routinely performed time-domain signal averaged ECG (CORAZONIX LdT, OK).
Patients did not assume antiarrhythmic drugs; the few who have been on 7, in the
10 cases when this occurred QT was counted 3 sinus cycles afterwards. We used QTd obtained
from the precordial leads for two reasons. The T waves in the precordial leads were always well
distinguishable, while T waves in the limb leads were sometimes flat and had to be discarded for
the analysis; moreover, the limb lead electrodes probably reflect vector components of overall
cardiac events rather than local phenomena1,8. Data are reported as mean ±1 SD. The unpaired
T test was used to compare patients and controls. The chi-square test was used to compare
differences in prevalence. A p <0.05 was considered significant.
The median number of ventricular premature beats in the patients group was 9953±9687
(range 1075-45185). Couplets (up to 650/24 h) were observed in 25 patients (51%).
Nonsustained ventricular tachycardia (4 to 13 beats at rates from 110 to 180 b/min) was
observed in 18 patients (37%). Extrasystoles were of LBBB morphology with posterior
axis in the majority of patients (34 of 49, 69%). During exercise stress test, VA disappeared
in all patients above a heart rate of 126±15 b/min, and returned in the recovery phase:
this is the usual behavior in normal populations9. In controls, Holter monitoring resulted in
a very low number of VA (in three subjects, <10 VA/24 h). In patients, QTc routinely measured in lead II was similar to that of controls (395±21 v 386±20). However, QTd was greater in patients than in controls: 49±20 msec v 32±14 msec, p<0.01. Eleven patients (20%) had QTd exceeding 65 msec, a value reported as abnormal in a recent metanalysis10: in this group, we found a significantly higher number of arrhythmias/24
h than in patients with QTd<65 msec (13885±13085 v 7613±6963, p<0.05). In fact, arrhythmia number increased significantly from patients with QTd <40 msec (n="28," QTd 35±8 msec, VPBs/24 h 7866±7255), to patients with QTd between 41 and 65 msec (n="10," QTd 57±5 msec, VPBs/24 h 11792±10150) and patients with QTd>
65 msec (n = 11, QTd 78±6 msec, VPBs/24 h 13593±3549) (Fig. 1). As a consequence,
a slight but significant correlation was found between the arrhythmia number/24 h and
QTd (R=0.33, p<0.02). There was no difference in QTd between patients having VA with LBBL, posterior axis morphology (n="34," QTd 51±22 msec) with respect to patients having other arrhythmia morphologies (n="15," QTd 46±17 msec). Moreover, QTd was similar in patients with couplets and nonsustained VT and those with isolated VA: 51±20 ms v 47±18 msec respectively, who had indeed a similar total VA number/24 h (10321±9765 and 8976±8763 respectively).
Fig. 1: Progressive increase in the number (mean ± SE) of ventricular premature beats/24 h,
going from patients with QTd <40 msec, to patients with QTd between 41 and 65 msec and patients with QTd> 65 msec. * = p <0.05 among the three groups (ANOVA).
Thus, patients with frequent, non repetitive ventricular arrhythmias have a greater QTd
than controls. The values of QTd were only slightly increased, not enough to support a
reentrant circuit11. In patients with mitral valve prolapse and frequent arrhythmias, the
increase in QTd has been attributed to the presence of delayed afterdepolarizations
(DAD)6: however, the rate dependence of arrhythmias in our population (they
disappeared during tachycardia) does not fill the criteria for DAD-induced arrhythmias12.
The observation of a significant relationship between the amount of QTd and the
number of ventricular premature beats introduces an alternative hypothesis. In our
opinion, an increased QTd would not indicate any specific electrophysiologic mechanism
for arrhythmias, but it would represent their consequence throughout a
mechanoelectrical feedback. Thus, QTd in this population could be a peculiar aspect of
cardiac “memory”13-15. The abnormal sequence of excitation/contraction originated
each time by very frequent premature beats – of whichever origin – may perturb the
process of repolarization and QT interval duration in adjacent ventricular areas. Instead
of the marked changes in T wave morphology and polarity, detectable after constant
ventricular pacing in the classic memory effect13,14, frequent VA may modulate
repolarization in a subtler way: spatial dispersion of both action potential duration and
QT may be increased by the frequent changes in mechanical load, a stimulus known to
affect action potential duration13. In fact, if QTd was a marker of a given
electrophysiological moiety, more dangerous VA should indicate a different substrate, and
thus they would be associated with a greater QTd: our data showed that this was not
the case. Thus, in the clinical setting of normal hearts with frequent premature beats,
QTd and ventricular premature beats may bear a cause-effect relationship that cannot be
easily dissected, with a sort of positive mechanoelectrical feed back process linking them.
A limitation in testing this hypothesis is that it is difficult to trace back the beginning of the
arrhythmic phenomenon in healthy subjects who often are unaware of their rhythm
disturbance. Actually, time is presumably a relevant factor in creating irregularities in the
repolarization sequence and consequently in the contraction process: in children with
normal hearts and frequent ventricular arrhythmias, who obviously have a short-living
abnormality in the excitation-contraction process, QTd is normal16. Therefore, whether
our hypothesis on the underlying mechanism of QTd in normal hearts with frequent non
sustained arrhythmias holds true can only be demonstrated by an experimental investigation.
|
