© 2001 by European Society of Cardiology
Copyright © 2001, European Society of Cardiology
Dispersion of ventricular repolarization and refractory period
Department of Medical Cardiology, Glasgow Royal Infirmary, 10, Alexandra Parade, Glasgow G31 2ER, Scotland, UK
* Corresponding author. Tel.: +44-141-211-4722; fax: +44-141-552-4683 stuart.cobbe{at}clinmed.gla.ac.uk
Received 25 August 2000; accepted 27 December 2000
KEYWORDS Arrhythmia (mechanisms); Hypertrophy; Infarction; Repolarization; Ventricular arrhythmias
 |
1 Introduction |
|---|
 Top 1 Introduction 2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
In their classic studies published in 1964, Han, Moe and co-workers [1,2] established an association between nonuniform recovery of excitability and lowered fibrillation threshold. They concluded that "those agencies known to favour the development of ventricular fibrillation were found to increase the temporal dispersion of recovery of excitability, whether the average refractory period was reduced ... or increased.... The results emphasise the importance of nonuniformity of excitability and conduction velocity during the relative refractory period in the induction of turbulent impulse propagation." The purpose of this review is to describe the basis of dispersion in recovery of excitability in the ventricle and its association with arrhythmogenesis.
Ventricular tachyarrhythmias are readily generated not only in acute ischaemia/infarction but also in hearts that have undergone remodelling following myocardial infarction [3–5]. Life-threatening arrhythmias are commonly seen in patients with previous myocardial infarction in the absence of new ischaemic events, as evidenced by the ability to initiate sustained ventricular tachycardia by programmed stimulation [6]. The risk of ventricular arrhythmias and sudden death in heart failure is inversely proportional to the left ventricular ejection fraction [7]. However, the Veterans Heart Failure Trial and other studies suggest that the proportion of deaths that are sudden is higher in patients with less severe LV dysfunction [8]. Such individuals are less likely to die from pump failure, hence they are at greater relative (but not absolute) risk of sudden death in comparison with patients with advanced heart failure.
|
2 Conditions for reentry |
|---|
|
Top1 Introduction2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
The classical prerequisites for the development of reentry are the presence of a potential circuit around an anatomical obstacle, unidirectional block, and sufficiently slow conduction to enable recovery of excitability in time for reexcitation by the depolarizing wavefront [9]. A feature of this type of reentry is the presence of an excitable gap between the repolarising tail and the head of the next depolarizing wave. Impulses originating outside the reentry circuit may depolarise the tissue in the excitable gap and influence the rhythm. Even in the absence of an anatomical obstacle, differences in conduction velocity due to altered cellular connections or asymmetrical damage and depression of electrophysiological properties in myocardial tissue may result in unidirectional block [10,11]. This mechanism, known as anisotropic reentry, also demonstrates an excitable gap. In contrast, Allessie et al. [12] showed that reentry could arise in the absence of an anatomical obstacle or pre-existing differences in conduction velocity, solely on the basis of temporal differences in recovery of excitability. In their study, a dispersion of refractory period of 11–16 ms was sufficient to produce reentry around a line of conduction block of approximately 5 mm following premature stimulation. This type of mechanism is described as functional or ‘short excitable gap’ reentry [9,13] In this instance, the unidirectional block is transient and slow conduction is dependent on encroachment into the relative refractory period.
|
3 Action potential duration, repolarization time and refractory period |
|---|
|
Top1 Introduction2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
Action potential duration (APD) is the time interval from the onset of phase 0 to the return of membrane potential to the resting level. Because completion of repolarization is difficult to identify precisely, it is common to measure APD to the point of 90% repolarization. We shall refer to this moment as the repolarization time. Refractory period refers to the interval from depolarization to the recovery of excitability. In normally polarised cells there is a close temporal relationship between APD and effective refractory period such that the two are often used as estimates of each other. The close relationship between repolarization and the recovery of excitability is disturbed in depolarised and/or ischaemic myocardium (see Section 8).
|
4 Measurement of repolarization time and refractory period |
|---|
|
Top1 Introduction2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
Repolarization time and refractory period may be measured in a variety of ways. Although an exhaustive account of techniques is beyond the scope of this article, their applicability to the measurement of dispersion is discussed below and summarised in Table 1.
|
4.1 Measurement of repolarization time
4.1.1 Intracellular recording
Intracellular microelectrode recording is the traditional method for measuring APD. However, the difficulty in maintaining multiple stable impalements severely limits its value in the measurement of dispersion of repolarization in whole tissue. Measurements on isolated cells are more straightforward. Myocytes are normally isolated from the whole left ventricle, but if regional differences are to be studied, tissue is cut from defined areas of the heart following enzymatic digestion. In general, the quoted values of APD are derived from measurements on a number of selected cells. The mean value represents the spatial average of the tissue specimen, and makes the assumption that cells studied are a randomly and unbiased sample. There is a trade-off between the tissue sample size needed to produce adequate numbers of live cells (dependent on ‘yield’) and spatial resolution. If the tissue volume is large, significant gradients in APD may be present within the sample. In principle, information on dispersion is available in the spread (e.g. standard deviation) of APD. While this is often quoted, little use is made of it in practice.
4.1.2 Optical mapping
The problems of achieving multiple microelectrode impalements in whole tissue for measurement of APD can be overcome by the technique of optical mapping with voltage-sensitive dyes. Using photodiode arrays or CCD cameras, hundreds to tens of thousands of simultaneous recordings can be made. Fidelity of optical action potentials depends on light being collected from fixed locations, and is therefore compromised by movement artefact. Repolarization is more likely to be distorted than activation, because the latter precedes muscle shortening. Movement artefact may be reduced using E–C uncouplers, such as diacetyl monoxime or cytochalasin-D, but these agents may affect APD [14–16].
4.1.3 Monophasic action potential recording
Monophasic action potentials (MAPs) have been shown to reflect accurately the time-course of the transmembrane action potential in single cells adjacent to the zone depolarised by the contact electrode [17,18]. To date, electrode numbers have been limited to a dozen or so by the technical difficulty of maintaining adequate contact pressure in all the electrodes. Furthermore, the tip diameter of catheter electrodes (1–2 mm) and the size of the underlying depolarised region impose a limit on the density of coverage.
4.1.4 Activation–recovery intervals
The activation–recovery interval (ARI) is the time between the minimum first derivative (
min) of the QRS and maximum derivative (max) of the T wave in a unipolar electrogram. Evidence for the relationship between the ARI and transmembrane APD was provided under conditions of local epicardial warming, sympathetic nerve stimulation and global ischaemia [19]. In most circumstances, mean differences between ARI and APD were small. However, for some paired determinations, particularly in ischaemia, the differences were large. In ischaemia, the development of ST segment elevation and changes in T wave morphology leads to difficulties in definition of repolarization time purely on the basis of T wave max. The use of criteria for signal acceptance as proposed by Ejima et al. [20] allows ARI measurements to be made in the majority of sites in the ischaemic border zone. In the central ischaemic zone, the proportion of acceptable sites falls below 50% after 20–30 min. In sites meeting the criteria, there is an excellent correlation between ARI and APD, but individual values may still differ by 50 ms or more.
4.1.5 Projection of repolarization to the body surface
QT dispersion (QTd) from the 12-lead ECG has been proposed as a non-invasive measure of dispersion of ventricular repolarization time and of arrhythmia risk [21]. However, the theoretical basis of QT interval dispersion and its association with dispersion of repolarization are poorly understood. For example, although QT measurements from all six limb leads are commonly included in measurement of dispersion, all of the information in the frontal plane is contained in any two leads, since the values of the others can be derived mathematically [22]. The entire information from ventricular electrical activity is contained in the QRST vector loop, which is projected onto the X, Y and Z leads of the vectorcardiogram. In theoretical terms it is hard to envisage how a single spatial loop can display ‘dispersion’. In a simulation study, a clear association was shown between the measured QT interval dispersion and the variation in ECG lead amplitudes derived from a simple heart vector model with no regional dispersion of repolarization. The authors concluded that measured QT dispersion is related mostly to a projection effect and does not truly measure dispersion of repolarization [23].
Zabel et al. [24] recorded multiple monophasic action potentials simultaneously with a 12-lead ECG from isolated Langendorff-perfused rabbit hearts. QT and JT dispersion correlated significantly but weakly with dispersion of APD90 and repolarization time (r = 0.58 and 0.64, respectively, P<0.001). In a more recent study using isolated dog hearts, there was a poor correlation between epicardial activation–recovery intervals and surface QT intervals. The surface recordings were insensitive to shortening of regional repolarization in the presence of prolonged repolarization elsewhere [25].
In addition to the theoretical uncertainties surrounding QT interval dispersion as a non-invasive index of dispersion of repolarization, there are major methodological problems. The intra- and interobserver variability and reproducibility of QT dispersion measurements are poor, with relative errors ranging from 17 to 44% [26–29].
4.2 Measurement of refractory period
4.2.1 Extrastimulus technique
The classic technique for the measurement of effective refractory period (ERP) is with regular trains of pacing stimuli followed by premature stimuli at progressively shorter coupling intervals [1]. Measurements may be made from epicardial, endocardial or intramural sites [30]. Refractory period can be determined only one site at a time, hence consecutive measurements are required for estimates of spatial dispersion. The stability over time of myocardial electrophysiological properties must therefore be demonstrated, or else assumed. This is clearly not possible for interventions that produce progressive changes, such as acute ischaemia.
4.2.2 Ventricular fibrillation intervals
Inter-activation intervals obtained from electrograms during VF in perfused hearts may be used to derive an index of local refractoriness [31,32]. The assumption is that during fibrillation, cells will become re-excited as soon as (or very soon after) they have recovered from the previous activation [33,34]. This is supported by the demonstration of a close correlation between VF interval and effective refractory period [31,35] and the absence of a diastolic interval during fibrillation [31,36,37]. However, excitability is restored before full repolarization so that a fully or partially excitable gap may exist during established VF [38]. This is supported by experimental data, at least in thin-walled structures such as the atria and right ventricle [39,40].
The VF interval method allows spatial dispersion to be estimated from a large number of simultaneously recorded sites. However, temporal resolution is sacrificed because VF intervals recorded over a period of many seconds are required for a statistical estimate of local refractory period. The presence of VF itself will alter the electrophysiological properties of the myocardium due to the faster rate, variations in interval between successive depolarizations and differences in direction of wavefront propagation.
|
5 Estimation of dispersion |
|---|
|
Top1 Introduction2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
5.1 Global dispersion
Classically, dispersion has been defined as the difference between longest and shortest repolarization time or refractory period within a set of measurements. This is particularly appropriate in studies where few (two to three) electrodes are used, and where electrodes are sited both inside and outside a defined pathological (e.g. infarct border) or experimentally altered (e.g. ischaemic) area. More recently, a variety of alternative measures of dispersion have been developed to quantify heterogeneity of repolarization time and/or refractory period. These may be classified as ‘global’ or ‘local’. Global dispersion is derived from measurements at multiple sites without regard to their location; it does not express how values are distributed spatially. Measures in this category include maximum difference (range), standard deviation of the mean (
) and coefficient of variation (/
x100). These are roughly equivalent, except that the range is more sensitive to spuriously low or high values.
5.2 Local dispersion
In contrast, measures of local dispersion take into account how repolarization time/refractory period varies from site to site. Fig. 1 illustrates how identical values for global dispersion of refractory period can mask markedly different patterns of local dispersion. Fig. 1a–c represent grids of 16x16 electrode sites of given refractory period which differ only in their distribution. All three panels have identical values for the global mean, standard deviation and coefficient of variation, yet the differences in local dispersion are obvious. The values of global and local dispersion are listed in Table 2.
|
), P50 (
)(median) and P95 (
).
|
The simplest measure of local dispersion is the maximum difference between any two adjacent sites and has the values of 2, 16 and 30 ms in Fig. 1a–c, respectively. Like the global maximum difference, this measure is subject to distortion by extreme and possibly artefactual values. Furthermore, it does not summarise the characteristics of the sites as a whole. In order to do this, the mean or the median of differences between adjacent sites may be used. The method proposed by Lammers et al. to quantify spatial inhomogeneity in atrial conduction [41] may be adapted to provide alternative measures of local dispersion, by substituting estimates of refractory period for local activation times. From the histogram of these estimates (Fig. 1 lower panels), median (P50), absolute inhomogeneity (P5–95) and inhomogeneity index (P5–95/P50) are calculated. An alternative approach involves counting the number of adjacent differences that exceed a threshold value. This number may then be expressed as the percentage of the total number of adjacent sites [37]. Spatial autocorrelation may also be used to estimate local dispersion and can be calculated separately for orthogonal directions [42].
Few studies compare the value of different measures of dispersion in predicting propensity to arrhythmias. Ogawa et al. [43] compared the ability of several indices of dispersion to discriminate between groups of animals with different propensities to reentrant VT. They reported that measures of global dispersion of ERP (standard deviation, coefficient of variation, range) were superior to local measures (maximum adjacent difference, infarct–normal zone difference). In contrast, the authors [37] found no significant differences between normal and infarcted rabbit hearts in global dispersion (expressed as coefficient of variation) of VF intervals, despite lower VF threshold in infarcted hearts. However, local dispersion within the area adjacent to the infarct was significantly greater in infarcted hearts with low ejection fraction than in normal hearts.
It is likely that none of the measures of dispersion discussed above have a uniquely high predictive accuracy for the induction of reentry. The reason for this is that the dispersion of repolarization time or refractory period represents only part of the substrate for reentry, and this can be modified instantaneously and unpredictably by a critically timed extrasystole. Premature stimulation may result in the development of an arc of functional conduction block, which occurs only at the moment of induction of reentry.
|
6 Physiological dispersion of APD |
|---|
|
Top1 Introduction2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
In healthy hearts, the left and right ventricles are structurally heterogeneous, with regional differences in shape, wall thickness and fibre angle. There are regional differences in APD, observed both in the intact heart and in myocytes isolated from different regions (see below). In general, action potentials are longer at early activation sites and shorter at late activation sites [44], resulting in relatively homogeneous recovery times. Thus, a degree of nonuniformity in conduction and refractory period is physiological. Spontaneous ventricular tachyarrhythmias do not normally arise in healthy hearts. However, high intensity electrical stimuli delivered in the vulnerable period can produce sufficient nonuniformity in conduction and refractory period to result in fibrillation [45].
6.1 Heterogeneity of cell types: transmural dispersion
Four distinct excitable cell types have been recognised in the ventricular myocardium of most mammalian species studied, including man: epicardial, endocardial, midmyocardial (M) cells and Purkinje fibres [46,47]. These differ on the basis of relative ion channel densities and responses to drugs/agents. M cells differ from epicardial and endocardial cells with respect to two ionic currents. They display a weaker slowly inactivating delayed rectifier current IKs [48], producing longer action potentials, steeper APD-rate relations and a greater sensitivity to agents with class III actions [46]. M cells also have a more prominent late INa [49]. Their distribution throughout the heart is thought to contribute to the heterogeneity of APD between subepicardium and subendocardium. However, the presence of M cells is not a prerequisite for transmural heterogeneity [50]. For example, in guinea-pig hearts, a transmural APD gradient has been observed in the absence of identifiable M cells [51].
There is heterogeneity of APD in the Purkinje system. In the canine ventricle, APD is short proximally, lengthens distally, before shortening again within the free wall [52]. Electrotonic effects influence Purkinje fibre APD: if a fibre bundle is cut away from its myocardial insertion its APD increases [53].
It is important to emphasise that the transmural differences in APD described above are most readily demonstrated in isolated myocytes or non-perfused tissue slices, and are least evident in the intact heart [50,54]. The mostly likely explanation for this phenomenon is that the intrinsic differences in APD in cells from different layers are minimised in the intact heart by the presence of electrotonic interaction. Under pathological conditions, this interaction may be reduced (see Section 6.4).
6.2 Base–apex dispersion
In addition to transmural gradients due to spatial variation in cell type, there are also transepicardial gradients in ion channel expression and action potential characteristics. HERG, the gene which encodes the major channel protein responsible for IKr, is less abundantly expressed in epicardial cells from the base than elsewhere in the ferret ventricle [55]. Using optical mapping of guinea pig hearts, Laurita et al. [56] observed systematic gradients in epicardial APD running in an oblique direction from the left ventricular apex to the right atrio-ventricular groove. Action potential dispersion was larger in this axis than that reported between epicardium and endocardium: 25 ms apex–base (independent of pacing rate and location) versus 16 ms endo–epi [57].
Recordings from pig hearts in situ showed that epicardial monophasic APD differences between the base (longer) and apex (shorter) were accentuated by an increase in load produced by aortic clamping [58]. In contrast, in work on rabbits, Cheng et al. [59] found that APD was significantly longer in cells isolated from the left ventricular apex compared to the base. This inversion may be due in part to the fact that the cells in the latter study were isolated from transmural rather than epicardial LV samples. A relatively large variability in APD was seen in both apical and basal samples, presumably due to the presence of M cells in the samples. In this study, the largest component of IK in apical myocytes was IKr, while in basal myocytes IKs was the largest component. Results obtained from isolated human myocytes, using tail current density measurements to estimate the relative contribution of IKr and IKs, show a similar gradient [60]. Salata et al. [61] also report a base–apex gradient in IKr/IKs ratio in rabbit ventricle. However, considerable care must be taken in comparing their magnitudes because of the difficulties in separating K currents for tail current density measurement [62]. Our own observations in the whole rabbit heart using the ventricular fibrillation interval as an index of refractory period did not confirm a consistent apex to base dispersion either under basal conditions or in response to acute ventricular distension [32].
6.3 Interventricular dispersion
Repolarization times in left and right ventricle measured with six MAP electrodes, placed at the same level around the heart [63], were not significantly different. However, the inflation of a balloon inside the left ventricle produced an increase in dispersion due to a significantly greater reduction in MAPD90 at left versus right ventricular sites, probably due to the disparity in load. In canine ventricular epicardium, the Ito1-mediated action potential notch is larger in the right versus left ventricle [64]. Volders et al. showed markedly shorter APDs in M cells from RV than those from LV; Ito1 density and IKs (but not IKr) tail currents were significantly larger in RV than LV [65].
6.4 Role of intercellular coupling
The repolarization properties observed in different regions of the ventricles are the result of the interaction between the spatial distributions of (a) the intrinsic properties of cells, and (b) the coupling resistance among the cells of the cardiac syncytium. In normal hearts, electrotonic interactions between cells, mediated by gap junctions, act to attenuate and spatially average (blur) the differences between individual cells [66–68]. A corollary is that when cells are isolated from one another, in vitro, they are free to express their intrinsic repolarization properties. Experimental interventions that reduce intercellular coupling therefore lead to enhanced heterogeneity in repolarization time. For example, global dispersion in rabbit hearts, estimated as the standard deviation of ARI from 256 epicardial electrodes, was increased nearly six-fold (from 6 to 35 ms) by infusion with 20 µM palmitoleic acid [69]. An increase in local dispersion of repolarization time (spatial autocorrelation), but not global dispersion (standard deviation), was observed in optically recorded action potentials from guinea pig LV epicardium when the heart was made hypoxic or hypothermic [42].
6.5 Differences in autonomic response
Ventricular repolarization time and refractory period are affected by the autonomic nervous system. Sympathetic stimulation shortens refractory period in both epicardium and endocardium of the left ventricular free wall [70,71], while in the same studies, vagal nerve stimulation exerted minimal effects on ventricular refractory period. However, vagal nerve stimulation has also been reported to prolong ventricular ERP [72]. Measuring ARI and MAPD at epicardial, endocardial and midmyocardial sites, Takei et al. [73] showed that sympathetic stimulation shortened repolarization times more in the M cell than in the other regions (by 54 ms vs. 27 ms in epi and 26 ms in endo), resulting in a decrease in transmural dispersion.
6.6 Restitution
In cells and tissues, APD generally decreases as the diastolic interval preceding it is shortened. This rate dependent phenomenon, termed restitution, arises as a result of continuing changes in the state of myocyte ion channels after membrane repolarization is complete. Heterogeneity of ion channel distribution occurring across the wall of the ventricle, trans-epicardially and between ventricles (Sections 6.1–6.3 above) would be expected to lead to heterogeneity of rate dependence and restitution in these domains. The most direct evidence for this is provided by optical mapping. In guinea pig, Laurita et al. [56] observed a base–apex gradient of restitution kinetics with a similar orientation to the gradient in APD.
Even at constant heart rate, repolarization time at any location may vary from beat to beat, either with fixed or varying spatial patterns of repolarization (concordant and discordant alternans, respectively). The latter in particular is associated with transient steep gradients in repolarization time and subsequent development of conduction block (when diastolic interval is too short to generate an action potential) and reentry [74]. The dispersion of restitution in infarcted and hypertrophied hearts, and how restitution interacts with pre-existing slow conduction to form an arrhythmogenic substrate, remain to be elucidated.
|
7 Changes in dispersion associated with acute myocardial ischaemia |
|---|
|
Top1 Introduction2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
7.1 Heterogeneity of metabolic changes in ischaemia
Reduction in coronary blood flow entails a sequence of pathophysiological changes which lead eventually to myocardial infarction and remodelling. Although the metabolic changes in acute ischaemia are complex, the majority of the changes seen in resting membrane potential, action potential amplitude, duration and refractoriness can be simulated by a combination of hypoxia, acidosis and increased extracellular [K+] [75]. Studies of regional ischaemia have demonstrated a sharp demarcation in epicardial pO2 between the normal and ischaemic zones. In contrast, the concentrations of extracellular K+ and H+ increase more gradually from the normal zone towards the central ischaemic area [76,77]. Regions of ischaemia may be patchy and/or interdigitate with normal non-ischaemic tissue. Ionic and electrical crosstalk may occur between these regions. For example, there is evidence for diffusion of K+ from the ischaemic toward the normal zone [77]. Interestingly, heterogeneity of [K+]o, measured from up to 64 sites in an ischaemic pig heart was greater in the border region (8 mmol/l) than in central zone (2 mmol/l) [77]. In the presence of hypoxia, small variations in [K+]o may lead to marked effects on recovery of action potential upstroke velocity after premature stimulation [78]. The variations of K+, H+ and pO2 form the basis of the spatial heterogeneity in electrophysiological properties in regions of acute ischaemia.
7.2 Action potential duration
Although some studies have shown transient lengthening of APD within 2 min of coronary occlusion, the major changes in ischaemia are a shortening in APD associated with a reduction in resting membrane potential, action potential amplitude and upstroke velocity. The mechanisms responsible for these changes have been reviewed in detail [79]. The degree of shortening in APD is variable in the epicardial ischaemic zone as a result of the metabolic heterogeneity described above. In addition, electrophysiological changes due to ischaemia are more pronounced in epicardium than in endocardium or Purkinje fibres [80,81], possibly due to diffusion of oxygen from intracavitary blood.
7.3 Refractory period
In contrast to the shortening of APD which occurs to a variable degree throughout the ischaemic area, refractory periods may shorten, lengthen or remain unchanged. Although under normal circumstances, refractory periods shorten in parallel with APD, during ischaemia the recovery of excitability may lag behind the completion of repolarization [82,83]. This phenomenon, termed ‘post-repolarization refractoriness’ is due to delayed reactivation of Na+ channels which first occurs at membrane potentials between –70 and –60 mV, when approximately half of Na+ channels are inactivated [84]. The greatest degree of depolarization occurs in the central ischaemic zone [77]. Progressive depolarization leads to opposite changes in APD (shortening) and refractory period (lengthening) [83].
Following the onset of ischaemia, there is a temporary decrease in diastolic stimulation threshold, which is followed in the central ischaemic zone by a rapid increase. In these areas, local activation may require high intensity stimulation. The measurement of refractory period at these sites may therefore be confounded by excitation of tissue with a lower stimulation threshold remote from the point of stimulation [85]. A more serious limitation in the assessment of refractory period and its dispersion is that the stimulation threshold and refractoriness are not at steady state in acute ischaemia. This effectively precludes the measurement of dispersion by the conventional extrastimulus technique.
Using the ventricular fibrillation interval technique which avoids the problems described above (Section 4.2.2), Opthof et al. [86] observed an increase in the mean interval in the ischaemic zone after coronary occlusion. The dispersion of refractoriness, estimated by the standard deviation of VF intervals, was also increased in the ischaemic zone.
|
8 Changes in dispersion associated with myocardial infarction and hypertrophy |
|---|
|
Top1 Introduction2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
8.1 Subacute (healing) phase
Changes in action potentials and ionic currents in the early post-infarction phase have been comprehensively reviewed [87,88]. In the context of dispersion of repolarization time and refractory period, the most important published results relate to differences between the infarct border zone and the non-infarcted zone. These are summarised in Table 3. The subacute phase is characterised by a shortening in APD with a loss of the plateau in surviving cells in the infarct zone. Despite these changes, slowing of the recovery kinetics of the Na+ channel results in post-repolarization refractoriness and an overall increase in refractory period compared with the normal zone. In studies on dogs, 3–5 days post-infarction [89,90], El Sherif et al. observed prolongation of effective refractory period which increased in a graded manner from border zone to the centre of the infarct. In normal hearts, mean dispersion of ERP was 30 ms, and conduction block did not occur with premature stimulation. In contrast, mean dispersion was 160 ms in infarcted hearts. In these hearts, conduction block in response to premature stimulation occurred when the difference in refractory period between adjacent sites (5–10 mm apart) increased from 10 to 20 ms [89]. In another study [90], a difference of 10 ms in ERP between sites spaced 1 mm apart was sufficient for the occurrence of functional conduction block and reentry.
|
8.2 Chronic (healed) phase
Remodelling of the left ventricle after myocardial infarction involves ventricular dilatation and eccentric hypertrophy accompanied by an increased amount of fibrous tissue. The arrhythmic substrate develops in the first 2 weeks after myocardial infarction, and, once established, appears to remain indefinitely [91,92]. Data from other models of hypertrophy show that hypertrophied myocardium can generate arrhythmias more readily than normal tissue [3,4,93,94]. Prolongation of APD is the most consistent electrical abnormality described in cells and tissues isolated from ventricles of animals or patients with heart failure independent of the mechanism [93,95–97].
Table 4 summarises data on changes in current density and APD in chronic myocardial infarction. In contrast to the subacute phase of infarction, where the non-infarcted left ventricle is normal, there is compensatory hypertrophy of the non-infarcted myocardium in the chronic infarction phase. The degree of hypertrophy varies within the left ventricle: a gradient exists from the infarct towards more distant cells [98,99]. It is likely that the consequent electrophysiological changes vary in the same way in the left ventricle and between left and right ventricles. Work from our own group using monophasic recordings on the epicardial surface of rabbit hearts, 8 weeks after coronary artery ligation, has shown not only an increase in mean APD compared with sham-operated controls but also an increase in the dispersion of APD [100] (Fig. 2). In the same model, we have also demonstrated directionally opposite transmural changes, with lengthening in APD in hypertrophied epicardial and M-cells but shortening in endocardial myocytes [101] (Fig. 3).
|
|
|
) rabbit hearts. Myocyte sub-types are shown above the records. Reprinted from Cardiovasc. Res. (McIntosh et al., 45 (2000) 397–409 8.3 Effect of ventricular dilatation
In addition to the enhanced predisposition to arrhythmias in chronically infarcted hearts, due in part to a raised baseline dispersion of refractory period, volume loading or ventricular dilatation has been found to lead to a further increase in arrhythmia inducibility. For example, in rabbit hearts [102], increasing load had no effect on VF thresholds in normal hearts, but produced a lowering of VF threshold in chronically infarcted hearts. Calkins et al. [103] reported that, of eight chronically infarcted dog hearts in which VT/VF was not inducible during programmed electrical stimulation at low volume, four showed inducible tachyarrhythmias at high volume.
Sustained ventricular dilatation shortens APD and refractory period in parallel [58,104–107]. However, the effects of dilatation on refractory period are spatially heterogeneous. Measuring MAPD at 5–6 epicardial sites, Zabel et al. [63] reported a decrease in mean ERP of 8% (from 198 to 183 ms), due to MAPD shortening at 58% and lengthening at 24% of LV sites. Ventricular dilatation has a greater effect on shortening refractory period at rapid pacing rates [108]. Using VF intervals as an index of refractory period, we showed both a shortening in VF intervals and an increase in their dispersion in normal rabbit hearts with sustained stretch. This was associated with a reduction in current threshold for VF induction [32]. The spatial distribution of VF interval shortening with stretch varied from heart to heart. Although the explanation for heterogeneous changes in refractory period is not known, Halperin et al. [109] showed that regional changes in refractory period correlated with changes in left ventricular end diastolic wall stress. As regional differences in wall thickness and compliance are greater in chronically infarcted hearts, it is not surprising, therefore, that dispersion of refractory period is exaggerated in these hearts under conditions of increased load. For example, Calkins et al. [103] showed that the change in refractory period with load at infarcted sites was greater than at control sites, resulting in an increase in dispersion. Our own recent study using the VF interval method demonstrated a difference in local dispersion between myocardium bordering a well-defined apical infarct and remote myocardium (Fig. 4).
|
8.4 Sympathetic innervation
Sympathetic nerve fibres may be damaged during infarction, and changes in sympathetic innervation may play a role in the propensity of chronically infarcted hearts to arrhythmia. Heterogeneity of sympathetic innervation has been correlated with refractory period in patients with ventricular tachyarrhythmias [110]. In a study on dogs, 1 week post-infarct [111], sympathetic denervation in the surviving layer of epicardium overlying the infarct produced a decrease in response to ansae subclaviae stimulation which led to an increase in ERP dispersion and arrhythmia (VT/VF) inducibility. Experimental and clinical studies have demonstrated regional sympathetic denervation in the peri-infarct zone following acute myocardial infarction [112,113]. Such areas would be unresponsive to sympathetic nerve stimulation in contrast to normally innervated areas.
8.5 Gap junction remodelling
Within a few days of coronary occlusion, changes in the number and distribution of gap junctions in cells bordering the infarct become apparent (for review see [114]). There is a 30–40% reduction of gap junction area per intercalated disk, limited to a few cells layers around the affected area. In contrast, in remodelled hypertrophied ventricles, the reduction is more widespread. The effects of changes in gap junction expression on anisotropy and conduction velocity are of importance in arrhythmogenesis but are beyond the scope of this review. A reduction in intercellular coupling would also be expected to unmask intrinsic differences in APD as discussed in section 6.4. Cellular uncoupling due to gap junction remodelling may contribute to dispersion of refractory period in hypertrophied myocardium particularly given the intrinsic transmural heterogeneity in APD recently demonstrated [101].
|
9 Is increased dispersion of repolarization time arrhythmogenic? |
|---|
|
Top1 Introduction2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
Kuo et al. [115] produced marked dispersion in monophasic APD in canine hearts by a combination of generalised hypothermia and coronary artery perfusion with warm blood. Although dispersion (range) was increased almost 10-fold to 111±16 ms, ventricular arrhythmias did not occur spontaneously. However, when maximal dispersion reached a critical value, repetitive ventricular responses progressing to VF could be induced by premature stimuli, but only at the site where the monophasic action potential was shortest. It was assumed that propagation of the premature impulse encountered a block in the area with long monophasic APD.
In optical recordings from normal guinea-pig hearts, as the coupling interval of the premature beat was shortened, global dispersion of repolarization time decreased to a minimum (at 255 ms), before increasing again at shorter coupling intervals. VF threshold was inversely related to dispersion, with a maximum at the coupling interval where dispersion was lowest [116].
As mentioned in Section 8.1, in the early post-infarction phase in dogs, steep gradients in refractory period were revealed by premature stimulation which corresponded to arcs of conduction block observed in isochronal activation maps during reentrant excitation [89,90]. The authors proposed that it was the spatial distribution of refractory periods rather than the overall inhomogeneity that was necessary for reentry.
As argued by Allessie et al. [12], differences between shortest and longest refractory periods are not the sole indicators of the risk of developing reentry. When the size of sites of prolonged refractory period is small, even in the presence of large disparities in refractory periods, reentry will not occur unless conduction is also grossly slowed. For reentry to arise, the combined effect of three variables must exceed a threshold: the zone (arc) of unidirectional block must be large enough, conduction around this zone must be slow enough, and refractory periods proximal to the zone of block must be short enough. The shape of the blocked zone may also be important; in models of ‘vortex shedding’, formation of spiral or scroll waves depend on the obstacle having a sufficiently sharp corner to allow separation and to give the detached wave sufficient space to turn without reattaching to the obstacle [117,118].
In summary, increased dispersion of repolarization time is not by itself arrhythmogenic, but it increases the likelihood of conduction block leading to initiation of reentry when conduction velocity is slowed.
|
10 Conclusion |
|---|
|
Top1 Introduction2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
It is clear that dispersion of refractory period is a fundamental mechanism in most reentrant arrhythmias. Techniques for the demonstration of dispersion offer limited predictive value for the subsequent development of arrhythmias. However, no current technique can anticipate the precise electrophysiological changes required at the moment of reentry. Dispersion of refractory period is a necessary but not sufficient condition for initiation of reentry. Nevertheless, dispersion represents an important property of the arrhythmia substrate, but a trigger in the form of a premature beat is necessary to achieve the critical degree of conduction delay for reentry.
Time for primary review 34 days.
|
References |
|---|
|
Top1 Introduction2 Conditions for reentry3 Action potential duration,...4 Measurement of repolarization...5 Estimation of dispersion6 Physiological dispersion of...7 Changes in dispersion...8 Changes in dispersion...9 Is increased dispersion...10 ConclusionReferences |
|---|
- Han J, Moe G.K. Nonuniform recovery of excitability in ventricular muscle. Circ. Res. (1964) 14:44–60.
[Abstract/Free Full Text] - Han J, Garcia DeJalon P.D, Moe G.K. Adrenergic effects on ventricular vulnerability. Circ. Res. (1964) 14:516–524.
[Abstract/Free Full Text] - Pye M.P, Cobbe S.M. Mechanisms of ventricular arrhythmias in cardiac failure and hypertrophy. Cardiovasc. Res. (1992) 26:740–750.
[Free Full Text] - Aronson R, Ming Z. Cellular mechanisms of arrhythmias in hypertrophied and failing myocardium. Circulation (1993) 87:76–83.
[Abstract/Free Full Text] - Hart G. The cardiac ventricular myocyte as a substrate for sudden death. The regional lecture in Bath 1993. J.R. Coll. Physicians Lond. (1993) 27:428–434.[ISI][Medline]
- Marchlinski F.E, Buxton A.E, Waxman H.L, et al. Identifying patients at risk of sudden death after myocardial infarction: value of the response to programmed stimulation, degree of ventricular ectopic activity and severity of left ventricular dysfunction. Am. J. Cardiol. (1983) 52:1190–1196.[CrossRef][ISI][Medline]
- Goldman S, Johnson G, Cohn J.N, et al. Mechanism of death in heart failure: the vasodilator-heart failure trials. Circulation (1993) 87:24–31.[ISI]
- Cohn J.N, Archibald D.G, Ziesche S, et al. Effect of vasodilator therapy on mortality in chronic congestive heart failure. Results of a Veterans Administration Cooperative Study. New Engl. J. Med. (1986) 314:1547–1552.[Abstract]
- Janse M.J. The heart and cardiovascular system. Scientific foundations. Fozzard H.A, Haber E, Jennings R.B, Katz A.M, Morgan H.E, eds. (1991) 2nd ed. New York: Raven Press. 2055–2094.
- Spach M.S. Changes in the topology of gap junctions as an adaptive structural response of the myocardium. Circulation (1994) 90:1103–1106.
[Free Full Text] - Fast V.G, Kléber A.G. Role of wavefront curvature in propagation of cardiac impulse. Cardiovasc. Res. (1997) 33:258–271.
[Abstract/Free Full Text] - Allessie M.A, Bonke F.I.M, Schopman F.J.G. Circus movement in rabbit atrial muscle as a mechanism of tachycardia. I.I. The role of non-uniform recovery of excitability in the occurrence of unidirectional block as studied with multiple microelectrodes. Circ. Res. (1976) 39:168–177.
[Abstract/Free Full Text] - Task Force of the Working Group on Arrhythmias of the European Society of Cardiology. The Sicilian gambit. a new approach to the classification of antiarrhythmic drugs based on their actions on arrhythmogenic mechanisms. Circulation (1991) 84:1831–1851.
[Abstract/Free Full Text] - Coulombe A, LeFevre I.A, Deroubaix E, et al. Effect of 2,3-butanedione 2-monoxime on slow inward and transient outward currents in rat ventricular myocytes. J. Mol. Cell. Cardiol. (1990) 22:921–932.[CrossRef][ISI][Medline]
- Biermann M, Rubart M, Moreno A, et al. Differential effects of cytochalasin d and 2,3 butanedione monoxime on isometric twitch force and transmembrane action potential in isolated ventricular muscle: implications for optical measurements of cardiac repolarization. J. Cardiovasc. Electrophysiol. (1998) 9:1348–1357.[ISI][Medline]
- Baker L.C, London B, Choi B.R, et al. Enhanced dispersion of repolarization and refractoriness in transgenic mouse hearts promotes reentrant ventricular tachycardia. Circ. Res. (2000) 86:396–407.
[Abstract/Free Full Text] - Franz M.R, Burkhoff D, Spurgeon H, et al. In vitro validation of a new cardiac catheter technique for recording monophasic action potentials. Eur. Heart J. (1986) 7:34–41.
[Abstract/Free Full Text] - Ino T, Karagueuzian H.S, Hong K, et al. Relation of monophasic action potential recorded with contact electrode to underlying transmembrane action-potential properties in isolated cardiac tissues: a systematic microelectrode validation study. Cardiovasc. Res. (1988) 22:255–264.
[Abstract/Free Full Text] - Haws C.W, Lux R.L. Correlation between in vivo transmembrane APDs and activation-recovery intervals from electrograms: effects of interventions that alter repolarization time. Circulation (1990) 81:281–288.
[Abstract/Free Full Text] - Ejima J, Martin D, Engle C, et al. Ability of activation recovery intervals to assess action potential duration during acute no-flow ischemia in the in situ porcine heart. J. Cardiovasc. Electrophysiol. (1998) 9:832–844.[ISI][Medline]
- Day C.P, McComb J.M, Campbell R.W.F. QT dispersion: An indication of arrhythmia risk in patients with long QT intervals. Br Heart J (1990) 63:342–344.
[Abstract/Free Full Text] - Coumel P, Maison-Blanche P, Badilini F. Dispersion of ventricular repolarization: reality? illusion? significance? Circulation (1998) 97:2491–2493.
[Free Full Text] - Di Bernardo D, Langley P, Murray A. Dispersion of QT intervals: a measure of dispersion of repolarization or simply a projection effect? Pacing Clin. Electrophysiol. (2000) 23:1392–1396.[CrossRef][Medline]
- Zabel M, Portnoy S, Franz M.R. Electrocardiographic indexes of dispersion of ventricular repolarization: an isolated heart validation study. J. Am Coll Cardiol (1995) 25:746–752.[Abstract]
- Lux R.L, Fuller M.S, MacLeod R.S, et al. QT interval dispersion: dispersion of ventricular repolarization or dispersion of QT interval? J. Electrocardiol. (1998) 30(Suppl):176–180.[CrossRef][ISI][Medline]
- Kautzner J, Yi G, Camm A.J, et al. Short- and long-term reproducibility of QT, QTc, and QT dispersion measurement in healthy subjects. Pacing Clin. Electrophysiol. (1994) 17:928–937.[CrossRef][Medline]
- Glancy J.M, Weston P.J, Bhullar H.K, et al. Reproducibility and automatic measurement of QT dispersion. Eur. Heart J. (1996) 17:1035–1039.
[Abstract/Free Full Text] - Savelieva I, Yap Y.G, Yi G, et al. Comparative reproducibility of QT, QT peak, and T peak-T end intervals and dispersion in normal subjects, patients with myocardial infarction, and patients with hypertrophic cardiomyopathy. Pacing Clin. Electrophysiol. (1998) 21:2376–2381.[CrossRef][Medline]
- Gang Y, Guo X.H, Crook R, et al. Computerised measurements of QT dispersion in healthy subjects. Heart (1998) 80:459–466.
[Abstract/Free Full Text] - Naimi S, Avitall B, Mieszala J, et al. Dispersion of effective refractory period during abrupt reperfusion of ischemic myocardium in dogs. Am. J. Cardiol. (1977) 39:407–412.[CrossRef][ISI][Medline]
- Opthof T, Ramdat Misier A.R, Coronel R, et al. Dispersion of refractoriness in canine ventricular myocardium. Effects of sympathetic stimulation. Circ. Res. (1991) 68:1204–1215.
[Abstract/Free Full Text] - Burton F.L, Cobbe S.M. Effect of sustained stretch on dispersion of ventricular fibrillation intervals in normal rabbit hearts. Cardiovasc. Res. (1998) 39:351–359.
[Abstract/Free Full Text] - Moe G.K, Abildskov J.A. Atrial fibrillation as a self-sustaining arrhythmia independent of focal discharge. Am. Heart J. (1959) 58:59–70.[CrossRef][ISI][Medline]
- Moe G.K. On the multiple wavelet hypothesis of atrial fibrillation. Arch. Int. Pharmacodyn Ther. (1962) 140:183–188.[ISI]
- Wang L.X, Li C.Y, Yong A.C, et al. Fast Fourier transform analysis of ventricular fibrillation intervals to predict ventricular refractoriness and its spatial dispersion. Pacing Clin. Electrophysiol. (1998) 21:2588–2595.[CrossRef][Medline]
- Czarnecka M, Lewartowski B, Prokopczuk A. Intracellular recording from the in situ working dog heart in physiological conditions and during acute ischemia and fibrillation. Acta Physiol. Pol. (1973) 24:331–337.[ISI][Medline]
- Burton F.L, McPhaden A.R, Cobbe S.M. Ventricular fibrillation threshold and local dispersion of refractoriness in isolated rabbit hearts with left ventricular dysfunction. Basic Res. Cardiol. (2000) 95:359–367.[CrossRef][ISI][Medline]
- Winfree A.T. The electrical thresholds of ventricular myocardium. J. Cardiovasc. Electrophysiol. (1990) 1:393–410.[CrossRef]
- Cha Y.M, Birgersdotter Green U, Wolf P.L, et al. The mechanism of termination of reentrant activity in ventricular fibrillation. Circ. Res. (1994) 74:495–506.
[Abstract/Free Full Text] - Kenknight B.H, Bayly P.V, Gerstle R.J, et al. Regional capture of fibrillating ventricular myocardium: evidence of an excitable gap. Circ. Res. (1995) 77:849–855.
[Abstract/Free Full Text] - Lammers W.J.E.P, Schalij M.J, Kirchhof C.J.H.J, et al. Quantification of spatial inhomogeneity in conduction and initiation of reentrant atrial arrhythmias. Am. J. Physiol. (1990) 259:H1254–H1263.[ISI][Medline]
- Salama G, Kanai A.J, Huang D, et al. Hypoxia and hypothermia enhance spatial heterogeneities of repolarization in guinea pig hearts: analysis of spatial autocorrelation of optically recorded action potential durations. J. Cardiovasc. Electrophysiol. (1998) 9:164–183.[ISI][Medline]
- Ogawa S, Furuno I, Satoh Y, et al. Quantitative indexes of dispersion of refractoriness for identification of propensity to reentrant ventricular tachycardia in a canine model of myocardial infarction. Cardiovasc. Res. (1991) 25:378–383.
[Abstract/Free Full Text] - Franz M.R, Bargheer K, Rafflenbeul W, et al. Monophasic action potential mapping in human subjects with normal electrocardiograms: direct evidence for the genesis of the T-wave. Circulation (1987) 75:379–386.
[Abstract/Free Full Text] - Han J, Garcia De Jalon P.D, Moe G.K. Fibrillation threshold of premature ventricular responses. Circ. Res. (1966) 18:18–25.
[Abstract/Free Full Text] - Antzelevitch C, Sicouri S, Litovsky S.H, et al. Heterogeneity within the ventricular wall: electrophysiology and pharmacology of epicardial, endocardial, and M cells. Circ. Res. (1991) 69:1427–1449.
[Free Full Text] - Drouin E, Charpentier F, Gauthier C, et al. Electrophysiologic characteristics of cells spanning the left ventricular wall of human heart: evidence for presence of M cells. J. Am. Coll. Cardiol. (1995) 26:185–192.[Abstract]
- Liu D.W, Antzelevitch C. Characteristics of the delayed rectifier current (IKr and IKs) in canine ventricular epicardial, midmyocardial, and endocardial myocytes: A weaker IKs contributes to the longer action potential of the M cell. Circ. Res. (1995) 76:351–365.
[Abstract/Free Full Text] - Eddlestone G.T, Zygmunt A.C, Antzelevitch C. Larger late sodium current contributes to the longer action potential of the m cell in canine ventricular myocardium. Pacing clin electrophysiol (1996) 19(II):569. Abstract.
- Näbauer M. Electrical heterogeneity in the ventricular wall-and the M cell. Cardiovasc. Res. (1998) 40:248–250. (Editorial).
[Free