(Circulation. 1995;91:2566-2572.)
© 1995 American Heart Association, Inc.
Articles |
From the Academic Medical Center, University of Amsterdam, Department of Clinical and Experimental Cardiology, Amsterdam (A.R.R.M., T.O., J.T.V., J.M.T.d.B., F.J.L.v.C., M.J.J.); the Department of Cardiology, St Antonius Hospital, Nieuwegein (A.R.R.M., N.M.v.H., J.J.A.M.D.); and the Interuniversity Cardiology Institute, Utrecht (J.M.T.d.B.), the Netherlands.
Correspondence to Tobias Opthof, Academic Medical Center, University of Amsterdam, Department of Clinical and Experimental Cardiology, PO Box 22700, 1100 DE Amsterdam, Netherlands.
| Abstract |
|---|
|
|
|---|
Methods and Results We measured local fibrillation intervals at 32 to 64 sites in the noninfarcted part of the left ventricle in patients undergoing antiarrhythmic surgery for symptomatic, drug-refractory, postinfarction ventricular tachyarrhythmias. The grid of electrodes (interelectrode distance, 7 mm) was attached to the epicardium of the left ventricle remote from the infarcted tissue. Group 1 consisted of 7 patients with hemodynamically tolerable sustained VT (VT group). Group 2 consisted of 7 patients with cardiac arrest and documented VF (VF group). With the patients on cardiopulmonary bypass, VF was induced by multiple premature stimulation. The VF interval was not significantly different in the two study groups (VT group, 136±5.5 ms; VF group, 129±3.4 ms, mean±SEM). However, spatial dispersion of the VF intervals (remote from the infarcted area) expressed as the coefficient of variation of VF intervals (SDx100/mean VF interval in each heart) was significantly larger in the VF group. It was 3.63±0.56 in the VF group and 1.55±0.40 in the VT group (mean±SEM; P<.01). Differences between the shortest and longest VF intervals in one and the same heart and the largest difference between two adjacent sites were also larger in the VF group (P<.02 and P<.05, respectively).
Conclusions This study shows larger dispersion in VF intervals and therefore suggests larger dispersion of refractory periods in parts of the myocardium remote from the infarction in patients with postinfarction VF than in patients with postinfarction VT.
Key Words: death, sudden fibrillation tachycardia
| Introduction |
|---|
|
|
|---|
The degree of left ventricular dysfunction, ischemia, and rate and stability of the tachycardia are probably factors involved in the transition of VT into VF in postinfarction patients.18 19 20 21 Moreover, hypertrophied, noninfarcted myocardium has altered electrophysiological characteristics.22 We hypothesized that noninfarcted ventricular tissue may provide heterogeneities in refractory periods that contribute to the transition from VT into VF in postinfarction patients. However, the role of inhomogeneity of refractoriness in patients has not been defined because refractory periods can only be assessed with sequential and time-consuming measurements with the classical extrastimulus technique. Fibrillation intervals have been used as an index for local refractoriness.23 24 25 26 27 28 These fibrillation intervals have been shown to correlate well with local atrial and ventricular refractoriness in several species, including humans.23 24 25 26 27 28 This method allows assessment of "refractory periods" simultaneously at multiple sites. The purpose of this study was to determine spatial dispersion of ventricular refractoriness in patients referred for evaluation of postinfarction ventricular tachyarrhythmias.
| Methods |
|---|
|
|
|---|
|
A detailed clinical and physical examination was obtained from all patients. Cardiac catheterization had been performed according to standard techniques. Significant coronary artery disease was defined as a >50% reduction in luminal diameter of the left main coronary artery or a >75% reduction in luminal diameter of any other major coronary artery. An aneurysm was defined as a dyskinetic segment of myocardium producing a distinct distortion of the diastolic left ventricular contour. The location of an infarction required an akinetic or dyskinetic area supplied by a major coronary artery with a >70% stenosis and corresponding pathological Q waves on the 12-lead ECG. Left ventricular ejection fraction at rest was quantified during sinus rhythm using radionuclide ventriculography. All antiarrhythmic drugs had been discontinued for at least five drug half-lives before programmed ventricular stimulation or surgery. All patients underwent programmed ventricular stimulation according to a previously described protocol.29 The end point of the electrophysiological study was completion of the pacing protocol, the induction of a ventricular arrhythmia requiring DC countershock, or the reproducible initiation of a sustained ventricular arrhythmia that did not require countershock. The diagnosis of VT was made according to standard criteria.30 31 VT was defined as sustained if it lasted more than 30 seconds or if hemodynamic deterioration necessitated termination; VT was defined as nonsustained if it ended spontaneously within 30 seconds and did not provoke hemodynamic compromise. VF was defined as a rapid ventricular rhythm with totally disorganized electrical activity in the surface ECG leads without recognizable QRS complexes.32 33 The protocol of this study was approved by the institutional committee on human research. All patients gave informed consent to the study protocol.
Epicardial Signal Acquisition
A multiterminal grid electrode
was used for simultaneous
recording of epicardial electrograms. The grid electrode consisted of
rectangular sheets of silicone rubber on which terminals (stainless
steel hemispheres, 2-mm diameter) were fixed and arranged in a square
lattice of either 8 columns and 4 rows or 8 columns and 8 rows. The
interelectrode distance was 7 mm. The grid electrode was always placed
remote from the infarcted myocardium of the left ventricle, as judged
by means of the cineangiogram and by visual inspection during the
operation. Recordings were made in the unipolar mode, with the signal
of a needle in the left shoulder as reference. A hook electrode
attached to either the right or left ventricle was used to induce VF by
multiple premature stimulation. All measurements were made at least 1
minute after the induction of VF during normothermia on cardiopulmonary
bypass. Epicardial electrograms were filtered (low cutoff, 1 Hz; high
cutoff, 1 kHz) and amplified (x200) together with surface ECG leads I,
II, and III. Subsequently, these signals were fed into an LS 11based
computer system by means of an 80-channel multiplexing
analog-to-digital converter. Samples were taken every 4 ms. Twelve
epicardial signals, the surface ECG leads, and a stimulation marker
were continuously registered on a 16-channel recorder. With each
registration, a 4-second period of VF was stored. Data were transmitted
to and stored in a PDP-11/73 computer for analysis.
Signal Analysis
Stored signals were displayed on a graphic
display. An
interactive computer program indicated local activation times in each
electrogram. The intrinsic deflection of the unipolar electrogram,
indicating local activation during VF, was arbitrarily defined as a
negative deflection with a steepness of at least 0.5 V/s over two
consecutive 4-ms periods. Subsequently, histograms of the intervals
between local activations were made. The "VF interval" was
defined as the mean of the first (or only) peak of the
histogram.24 28 A VF interval was assigned to a site
only
if the SEM was <2.5 ms.24 28 The reason for
exclusion of
sites with VF intervals with SEM values >2.5 ms was as follows. Local
mean VF intervals are correlated with local refractory periods. This
relation has been established in canine atrium,23 in
canine ventricle,24 in porcine ventricle,25
and in human atrium.26 Obviously, the local VF interval
needs to be accurate to serve as an index for local refractoriness.
Sites with mean local VF intervals with large SEM values are observed
during prolonged periods of acute ischemia. They probably are the
reflection of large excitable gaps during fibrillation. In another
series of patients (not described in this study; see also
"Discussion"), we observed that measurements from the infarcted
area always led to wide histograms of local VF intervals with, as a
consequence, mean local VF intervals with large SEM values. Although it
is easily possible to measure these intervals, it is hard to assign a
meaningful physiological significance to them. Therefore, we restricted
ourselves to the myocardium remote from the infarcted region. All data
were analyzed by observers unaware of the index arrhythmia.
Statistics
Data are mean±SEM unless SD values are
indicated. Differences
in patient characteristics were tested by Student's t test.
Spatial dispersion in refractoriness was assessed by three (comparable)
parameters. We calculated the coefficient of variation (SDx100/mean)
of the mean VF intervals at all sites in each heart. Further, we
calculated the difference between the site with the longest mean local
VF interval and the site with the shortest mean local VF interval in
each heart. Finally, we measured the largest difference in mean local
VF interval at adjacent sites in each heart. The spatial dispersion in
refractoriness in both groups was compared by the nonparametric
Wilcoxon test applied to data from the two patient groups.
| Results |
|---|
|
|
|---|
Electrophysiological Data Before Surgery
During
electrophysiological study, before surgery, sustained
monomorphic VT was induced in all 14 patients (Table 1
). The
mean cycle
length of the VT was 300±80 ms in the VT and 255±34 ms in the VF
group (mean±SD). This difference was not statistically
significant.
Induction of VF During Surgery
In all 14 patients who had
developed either VT or VF spontaneously
after myocardial infarction (the index arrhythmia) that had occurred
between 2 weeks and 9 years before the operation, VF was induced during
antiarrhythmic surgery with programmed ventricular stimulation. Fig
1A
shows a 4-second recording of surface leads I and II
during VF induced in a patient from the VT group. Figs 1B through 1D
show maps (8x8 grid electrode) of three consecutive cycles and provide
evidence that hearts of patients from the VT group indeed fibrillated
during our measurements during antiarrhythmic surgery, because of the
presence of multiple wave fronts.34
|
VF Interval as an Index of Refractoriness
Fig
2
shows a 4-second recording of an electrogram
during VF (Fig 2A
) with the activation moments determined by an
interactive computer program superimposed (Fig 2B
; see
"Methods"). Fig 2C
shows the corresponding
histogram. The VF
interval at this particular site was 151.7±1.63 ms. This procedure was
performed at all recording sites.
|
Spatial Dispersion in VT Group Versus VF Group
Fig
3
shows VF intervals at multiple sites in 1
patient from each group. The mean VF interval in the patient from the
VT group was 155.3±1.37 (SD) ms (Fig 3A
), and in the
patient with
aborted sudden death, it was 131.9±8.34 (SD) ms (Fig
3B
). Table 2
summarizes data from all 14
patients. Reliable
ventricular signals were recorded at 195 sites in the VT group and at
207 sites in the VF group. The corresponding mean VF intervals were
136±5.5 and 129±3.4 ms in the two groups, respectively
(n=7; Table 2
). This difference was not statistically
significant. However, these
SEM values describe only the variability of the mean values of
individual hearts in the two groups. They do not take into account the
spatial differences within hearts (for further explanation, see the
legend of Table 2
). Fig 4
shows the coefficients
of
variation, indicating spatial dispersion in VF intervals, in the
individual hearts. On average, these coefficients of variation were
1.55±0.40 in the VT group and 3.63±0.56 in the VF group (Table
2
;
Wilcoxon test; P<.01). The largest difference between VF
intervals at two adjacent sites was 5.9±1.74 ms in the VT patients and
10.3±2.20 ms in the VF group (Table 2
; Wilcoxon test;
P<.05). The difference between the longest and shortest VF
intervals in one and the same heart was 8.9±2.27 ms in the VT group
and 17.4±2.66 ms in the VF group (Table 2
; Wilcoxon
test;
P<.02). Thus, for each of the three parameters by which
spatial dispersion of VF intervals was assessed, the value was
associated with the index arrhythmia. The values were significantly
larger in the VF group than in the VT group.
|
|
|
| Discussion |
|---|
|
|
|---|
We realize that the patients in this study form a highly selected group. Not all survivors of cardiac arrest nor all patients who develop VT after infarction are eligible for surgery. Monomorphic sustained VT can be induced in only 41% of survivors of cardiac arrest,44 whereas VT was inducible in all patients of the present study. Therefore, our findings may not apply to all patients who develop postinfarction VT or VF.
Dispersion of Refractoriness
We found that the amount of
dispersion in VF intervals measured in
noninfarcted myocardium remote from the infarct area, expressed as the
coefficient of variation, the difference between the longest and
shortest VF intervals in the same heart, or the largest difference
between adjacent sites, was associated with the index arrhythmia. All
values were significantly larger in the VF group than in the VT group.
Noninfarcted myocardium is important because this part of the infarcted
heart will be subjected to hypertrophy, which may cause inhomogeneous
electrophysiological alterations. In a clinical study,45
inhomogeneity in left ventricular refractory periods was the same in
patients with normal hearts and those with prior myocardial infarction
and VT. But in contrast to our study, refractory periods were
determined successively and at limited sites.
Little is known about the differences in myocardial refractoriness between survivors of cardiac arrest and patients with recurrent sustained VT after myocardial infarction. The significance of nonuniformity of refractory periods in the genesis of tachyarrhythmias has been emphasized in several experimental studies.24 26 46 47 48 Inhomogeneity of refractory periods decreases the fibrillation threshold in the canine ventricle under various experimental conditions.49 50 This suggests that disparity of refractoriness facilitates VF, presumably by an increasing opportunity for reentry of an impulse from a region with a shorter refractory period into a region with a longer refractory period. However, VF threshold is measured with very strong stimuli.49 50 51 A decrease in VF threshold associated with increased dispersion in refractoriness does not prove that the same amount of dispersion in refractoriness would also be arrhythmogenic when the premature stimulus has a physiological intensity instead of the very high value during measurement of the VF threshold. Measurement of refractory periods with premature stimulation constitutes a time-consuming procedure, because the measurement disturbs the variable under study.24 52 53 As an alternative, averaged local fibrillation intervals have been used as an index of local refractoriness in atrium and ventricle.23 24 25 26 27 28 These intervals correlate well with refractory periods in atria and ventricles of pigs, dogs, and humans during normal activation.23 24 25 26 This method therefore allows simultaneous assessment of refractory periods at multiple sites.24 It has to be stressed, however, that the process of fibrillation requires a minimum period of at least 1 minute before steady-state values are reached. During the first seconds of the fibrillation process, the fibrillation cycle length has been reported to be as long as 120 ms in the canine ventricle, with a concomitant large excitable gap.54 Under these circumstances, local fibrillation intervals are probably meaningless as an index of local refractory periods. Very soon, the fibrillation cycle length shortens to a steady-state value (about 80 ms in the canine ventricle24 55 ). Thus, in open-chest studies in ventricles, this technique can be applied only during cardiopulmonary bypass.
An excitable gap has been demonstrated during atrial fibrillation in dogs,56 and leading circle excitation need not be the only mode of reentrant activation, especially during VF in larger species. Several studies57 58 59 have shown spiral or scroll waves during reentrant arrhythmias in isolated cardiac muscle. The presence of an excitable gap is not incompatible with the relation between mean local VF intervals and local refractory periods. As long as an excitable gap has a roughly similar probability (and duration) during the period of measurement of VF intervals, the local mean VF interval may equal the minimum local refractory period plus a constant.
Limitations
Epicardial areas remote from the infarcted
myocardium were
selected visually and by cineangiography. Myocardial biopsies have not
been taken from these sites, and thus, there is no guarantee that the
measurements were restricted to normal myocardium. However, in a
preliminary set of observations on other patients not included in this
study, we made observations with the grid of electrodes both on normal
and infarcted myocardium. In that series, we also observed larger
dispersion in VF intervals in patients with VF as index arrhythmia than
in patients with VT as index arrhythmia. We also observed that
measurements at sites in the infarcted area yielded histograms with SEM
values considerably larger than 2.5 ms (compare with Fig 2C
).
Under
these circumstances, VF intervals can probably no longer be regarded as
an index of refractory periods.60 The absence of sites
with such wide histograms in the two groups of patients described in
this study is, apart from visual inspection during surgery and
cineangiography, another warrant that our measurements were indeed
restricted to noninfarcted myocardium.
Conclusions
Our study shows larger dispersion in VF intervals
and therefore
suggests larger dispersion of refractory periods in noninfarcted
ventricular myocardium of patients with VF than in patients with VT
after myocardial infarction. Although the significance of other factors
as dimensions and architecture of infarcts may not be neglected,
electrophysiological properties of noninfarcted myocardium may be
an additional factor for deterioration of VT into VF.
| Acknowledgments |
|---|
Received August 30, 1994; revision received November 29, 1994; accepted December 3, 1994.
| References |
|---|
|
|
|---|
This article has been cited by other articles:
![]() |
T. Opthof In vivo dispersion in repolarization and arrhythmias in the human heart Am J Physiol Heart Circ Physiol, January 1, 2006; 290(1): H77 - H78. [Full Text] [PDF] |
||||
![]() |
V. S. Chauhan, E. Downar, K. Nanthakumar, J. D. Parker, H. J. Ross, W. Chan, and P. Picton Increased ventricular repolarization heterogeneity in patients with ventricular arrhythmia vulnerability and cardiomyopathy: a human in vivo study Am J Physiol Heart Circ Physiol, January 1, 2006; 290(1): H79 - H86. [Abstract] [Full Text] [PDF] |
||||
![]() |
A. E. Buxton, A. J. Moss, A. E. Buxton, and A. J. Moss Should everyone with an ejection fraction less than or equal to 30% receive an implantable cardioverter-defibrillator? Circulation, May 17, 2005; 111(19): 2537 - 2549. [Full Text] [PDF] |
||||
![]() |
M. J Janse Electrophysiological changes in heart failure and their relationship to arrhythmogenesis Cardiovasc Res, February 1, 2004; 61(2): 208 - 217. [Abstract] [Full Text] [PDF] |
||||
![]() |
A.J. Camm, M.J. Janse, D.M. Roden, M.R. Rosen, J. Cinca, and S.M. Cobbe Congenital and acquired long QT syndrome Eur. Heart J., August 1, 2000; 21(15): 1232 - 1237. [PDF] |
||||
![]() |
A. Patwardhan, S. Moghe, K. Wang, and F. Leonelli Frequency modulation within electrocardiograms during ventricular fibrillation Am J Physiol Heart Circ Physiol, August 1, 2000; 279(2): H825 - H835. [Abstract] [Full Text] [PDF] |
||||
![]() |
J. B. Moubarak, P. E. Karasik, R. D. Fletcher, and M. R. Franz High dispersion of ventricular repolarization after an implantable defibrillator shock predicts induction of ventricular fibrillation as well as unsuccessful defibrillation J. Am. Coll. Cardiol., February 1, 2000; 35(2): 422 - 427. [Abstract] [Full Text] [PDF] |
||||
![]() |
J.P. Amlie Electrophysiological mechanisms involved in the increased mortality of patients with cardiac insufficiency Eur. Heart J., September 2, 1999; 20(18): 1290 - 1292. [PDF] |
||||
![]() |
F. L Burton and S. M Cobbe Effect of sustained stretch on dispersion of ventricular fibrillation intervals in normal rabbit hearts Cardiovasc Res, August 1, 1998; 39(2): 351 - 359. [Abstract] [Full Text] [PDF] |
||||
![]() |
L Oikarinen, L Toivonen, and M Viitasalo Electrocardiographic measures of ventricular repolarisation dispersion in patients with coronary artery disease susceptible to ventricular fibrillation Heart, June 1, 1998; 79(6): 554 - 559. [Abstract] [Full Text] |
||||
![]() |
M. Nabauer and S. Kaab Potassium channel down-regulation in heart failure Cardiovasc Res, February 1, 1998; 37(2): 324 - 334. [Abstract] [Full Text] [PDF] |
||||
![]() |
L. Gepstein, G. Hayam, and S. A. Ben-Haim Activation-Repolarization Coupling in the Normal Swine Endocardium Circulation, December 2, 1997; 96(11): 4036 - 4043. [Abstract] [Full Text] |
||||
![]() |
A. Tosaki and D. K. Das Extracellular Mg++ Manipulation Prevents the Proarrhythmic Activity of Cromakalim in Ischemic/Reperfused Diabetic Hearts J. Pharmacol. Exp. Ther., July 1, 1997; 282(1): 309 - 317. [Abstract] [Full Text] |
||||
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||