Dissociation and Delayed Conduction in the Canine Right Bundle Branch

Transcription

1 Dissociation and Delayed Conduction in the Canine Right Bundle Branch ABE WALSTON, II, M.D., JOHN P. BOINEAU, M.D., JAMES A. ALEXANDER, M.D., AND WILL C. SEALY, M.D. SUMMARY Bipolar electrograms were recorded from five to six sites on the septal right bundle branch (RBB) using a multielectrode patch in ten mongrel dogs. Antegrade activation was recorded along the right bundle during sinus rhythm and during right atrial pacing at varying heart rates. Retrograde activation was produced by pacing from right ventricular epicardium. Right bundle conduction velocities varied from 1.4 to. m/sec (mean.0 m/sec) for antegrade conduction and from 1.8 to.7 m/sec (mean.1 m/sec) for retrograde activation. Varying degrees of conduction delay were elicited by premature stimulation and mechanical pressure to the right bundle. The conduction delays occurred over small segments of the right bundle branch with regeneration of normal propagation velocity distal to the region of block. The conduction delays were associated with fragmentation of the RBB electrograms with initiation of delayed activation waves which traveled variable distances antegrade and retrograde. These data show that stress-induced longitudinal dissociation may produce delayed activation waves within the right bundle branch which may simulate re-entrant arrhythmias. PROPAGATION THROUGH MYOCARDIUM and specialized conduction tissue is complex and incompletely understood. The concepts derived from intracellular electrophysiology form a basic foundation, but are inadequate to fully describe the behavior of an electrical wavefront spreading through a three-dimensional matrix of cardiac cells. The cellular or cable theory of cardiac depolarization describes propagation along single fibers.' This theory postulates that fiber-to-fiber excitation is through low resistance connections confined anatomically to the intercalated discs. Weidmann has concluded that these areas are of relatively low resistance compared to membrane resistance. Previous investigators, however, have suggested that cell-to-cell activation in cardiac tissue may not occur exclusively over these sites. I Recent studies have suggested that interaction between fibers may be an important factor in determining some tissue propagation characteristics.8 The present study was undertaken to examine the spread of activation in the canine right septal bundle branch under a variety of conditions. The data show that longitudinal dissociation of the right bundle branch can be produced in areas of delayed conduction during stress. Methods Ten mongrel dogs underwent electrophysiological studies under total cardiopulmonary bypass using sodium pentobarbital anesthesia (5 mg/kg). Atrial and venous pressures were held constant using continuous monitoring. The experimental model was surgically prepared by exposing the right septal myocardium and suturing a specially designed 0 point electrode grid on the right septal surface in the area of the right bundle branch. The electrode grid consisted of six rows of electrodes, each row mm apart, with an interelectrode distance of 1.5 mm, allowing recording of bipolar From the Veterans Administration Hospital, and the Departments of Pediatrics, Medicine and Surgery, Duke University Medical Center, Durham, North Carolina. Supported in part by USPHS Grants HL-1107, HL-1109, HL-57, HL-5716, HL-0576, and HL-1898, and a grant from the North Carolina Heart Association. Dr. Boineau is the recipient of USPHS Research Career Development Award HL VA Program No Address for reprints: Abe Walston, II, M.D., C-800, Veterans Administration Hospital, Durham, North Carolina Received September, 1975; revision accepted for publication November 1, electrograms at mm intervals along the right bundle branch (fig. 1). A switching system allowed rapid determination of those points in contact with the right bundle. The electrode grid was placed as near parallel to the right bundle as possible, but because of the inability to localize precisely the right bundle without special staining, this alignment was not identical in every case. Therefore, the activation wave of the RBB encountered the electrode grid at different angles in different experiments causing the polarity of the electrograms to differ slightly from one experiment to another. In one single experiment, however, the electrograms for each bipolar point were constant in form during the entire recording period. The RBB and muscle electrograms recorded from the electrode grid were fed through field effect transistor buffer amplifiers (input impedance of 1011 ohms) and recorded on FM magnetic tape using an Ampex Data Acquisition System-100. The frequency response of the system was linear to 000 Hz. During the electrophysiological studies, pacing was performed using bipolar pacing electrodes sutured to the right atrium and right ventricular epicardium. The data were reproduced on Kodak photographic paper using a Honeywell Visicorder Oscillograph at a paper speed of 1000 mm/sec to facilitate analysis. Slight fractionation of the activation wave caused the most rapid slope of the waveforms to appear in different portions of the right bundle electrograms (fig. 1). Calculations of conduction velocity using the most rapid slope of the electrogram varied markedly from point to point and were not reproducible, whereas calculations using the midpoint of the electrogram gave consistent and reproducible values. Therefore, the activation time was determined using the midpoint between the onset and termination of the RBB electrogram. After termination of the experiment, the electrode grid was removed and the exact distances (nearest 0.1 mm) between selected points were determined to allow accurate computation of conduction velocities. The velocity (in meters per second) was computed as the product of the activation time and the measured distance between the specific grid locations recorded. The conduction velocity of each animal was obtained by averaging the data from ten right bundle depolarizations. All conduction velocities were computed on measurements over distances of 5 9 mm. Velocity measurements on segments shorter than 9.0 mm were found to contain error, and there- 605

2 606 CIRCULATION VOL. 5, No. 4, APRIL 1976 R BUNDLE COMPLEXES 50 msec. FIGURE 1. The electrode grid. On the left is depicted the electrode grid with six rows of recording points. The center shows the right ventricular septal surface with the right bundle emerging just inferior to the papillary muscle of the conus and dividing near the anterior papillary muscle. On the right is an actual recording of RBB and muscle electrograms at the points along the right bundle branch indicated in the center. fore calculations of conduction velocity between adjacent electrode positions were used only to indicate general directions of velocity change. Statistical computations were performed using Student's t-test for paired data on an IBM 110 computer. Results Figure shows representative right bundle branch electrograms during antegrade and retrograde activation in two separate experiments. The right bundle depolarization is seen as the rapid bipolar complexes preceding the slower muscle electrograms (fig. ). During antegrade conduction the right bundle activates from row 1 to row 5, and the right ventricular septal myocardium activates in an opposite direction from row 5 to row 1 in the normal manner. The normal activation of the right septal myocardium occurs from row 5 to row 1 (opposite from the RBB) because activation of the right ventricle is initiated at the termination of the right bundle branch. No activation of the adjacent myocardium occurs during the passage of the activation wave down the early and midportions of the RBB. The slight fragmentation of the electrograms at rows and 5 may represent local inhomogeneity of the wavefront or be due to the geometrical relationship between the RBB and the electrode patch. The distal end of the electrode patch frequently overlay the arborization of the right bundle as it terminated in the apex of the right ventricle, causing the electrogram at row 5 to show multiple small spikes. Activation elicited by right ventricular epicardial pacing shows both RBB and muscle activation proceeding in a retrograde manner from row 5 to row 1; therefore, the direction of activation of the right septal myocardium is similar during both antegrade and retrograde conduction (fig. ). Sufficient data to allow accurate calculation (10 RBB depolarizations) of conduction velocity was present in seven experiments for antegrade velocity and in four of these experiments for retrograde velocity (fig. ). Mean antegrade velocity was.0 m/sec (range = 1.4 to. m/sec) and the mean retrograde velocity was.1 m/sec (range = 1.8 to.7 m/sec); however, in the experiments in which both antegrade and retrograde velocity were measured, retrograde was significantly faster (P < 0.0). The effect of manual pressure exerted on the right bundle branch in one experiment is demonstrated in figure 4, panel - - lmean l Meon Antegrade.0 m/sec. Retrograde.1 m/sec. 100 msec. FIGURE. Normal antegrade and retrograde Purkinje propagation. The top tracing is lead! ofthe electrocardiogram. The electrograms recorded from the right bundle branch are labeled RBB. The small rapid RBB electrograms are followed by the slower ventricular septal muscle electrograms. The left panel shows a recording during antegrade conduction and the right panel shows a recording during retrograde activation FIGURE. Mean antegrade and retrograde propagation velocities. On the ordinate, propagation velocity is shown in m/sec, and on the abscissa is indicated the experiment number. The stippled bars represent antegrade conduction; the striped bars represent retrograde conduction. Mean velocity is indicated by the height of the bar. Two standard deviations are shown by the brackets.

3 607 DISSOCIATION IN THE RIGHT BUNDLE BRANCH/Walston, Boineau, Alexander, Sealy A. C. FIGURE 4. Conduction delay due to manual on the right bundle. In panel A is represented the effect ofmanual pressure on the right bundle resulting in marked slowing of propagation between rows I and. In rows to 5 normal propagation velocity has been regenerated distal to the block. Panel B represents partial recovery of propagation velocity after the pressure has been released in the same experiment. The delay is less but there is still slowing between rows I and. Panel C reveals the conduction delay during retrograde activation. This conduction delay was induced by pressure and primarily affects conduction between rows and 1. pressure A. The pressure caused a localized delay in antegrade conduction at row with fragmentation of the electrogram into two components and marked slowing. Conduction velocity distal to the region of block from row to row 5 is normal. Panel B (fig. 4) shows data from the same experiment after the pressure was released. There was only a slight delay in activation between row and row, and the electrogram at row shows a more normal appearance. The activation time from row to row 5 is unchanged from the previous recording. Similar delays in retrograde conduction could also be elicited by manual pressure (fig. 4, panel C). There is delay in retrograde conduction occurring between row and row and prolongation of the electrogram at row. In both experiments conduction velocity returned to control levels after the pressure was released. Additional examples of dissociation with fragmentation of the electrograms induced by manual pressure to the right bundle are shown in figure 5. Panel A shows marked fragmentation of the right bundle electrograms at row 1 with the initiation of a second antegrade wave from row 1 to row which fires off the trailing edge of the electrogram at row 1. It is possible that this delayed spike at row is simply a result of the fragmentation at row ; however, the length of time between the primary electrogram and the secondary spike at row makes this interpretation unlikely. In panel B the desynchronization occurs at row 5 with the initiation of a retrograde activation of the RBB which proceeds in a retrograde direction from row 5 to row 1 (seen as the small spikes just preceding the muscle electrograms). In this secondary retrograde pathway, there is additional conduction delay at row. The effect of a premature stimulus on right bundle conduction is shown in figure 6. In the first beat (panel A) there is sinus rhythm (R-R interval = 778 msec) with a right bundle conduction velocity of 1.7 m/sec. After the second beat a premature stimulus of the atrium (SA) with a coupling interval of 10 msec resulted in a decrease in RBB conduction velocity in the next beat from 1.7 to 1.5 m/sec. Panel B shows on an expanded time base the identical electrograms at rows,, 4, and 6 which occurred after the premature stimulus in panel A. In addition to the decreased conduction velocity, the premature stimulus caused fragmentation of the electrogram at row with retrograde activation back to row. This initiation of a separate wave of activation in the RBB (as shown in figs. 5 and 6) was seen in multiple experiments following supraventricular or ventricular premature beats and could be elicited in an antegrade or retrograde direction. The anatomical site of dissociation induced by pressure or premature stimulation was not constant and varied within each experiment. Discussion The concept of functional dissociation originated from the studies of Mines,'1' 1 Garrey," Drury,1" Schmitt and Erlanger," and Lewis.16 The above studies were performed on isolated muscle strips with mechanical pressure and cooling used to produce the dissociation. More recently, Moe et al.,10 Watanabe and Dreifus,7 and Myerberg et al have extended these observations to the A-V node and distal A-V conduction system in the intact heart. The conduction delay seen in the present study, whether due to premature stimulation or mechanical pressure, was associated with desynchronization of the RBB electrograms. Longitudinal dissociation with desynchronization may be explained by normal antegrade activation in part of B A / RBB 5 50 msec. FIGURE 5. Longitudinal dissociation. Panel A shows longitudinal dissociation at row I with initiation of late antegrade activation. Panel B shows longitudinal dissociation in the antegrade pathway beginning at row S with late retrograde activation from row S to row 1. Note that in the late retrograde activation in panel B there is marked conduction delay between rows and.

4 608 CIRCULATION A B VOL. 5, No. 4, APRIL Prop Vel. 1.7m/sec. 500 msec. 4 6 I.fn4sec 5m/sec. 1.5m/secc. 50 msec It K FIGURE 6. Effect of premature stimulus on RBB conduction. In panel A, rows 1 to 6 demonstrate the six RBB and muscle electrograms recordedfrom the electrode patch. The premature atrial stimulus is indicated as SA. Panel B shows on an expanded time base the identical complexes from rows,, 4, and 6 which occurred after the premature stimulus in panel A. The conduction velocity after the premature stimulus has decreasedfrom 1.7 to 1.5 m/sec and desynchronization has occurred with initiation of a secondary activity wave retrograde from row to row. the right bundle with absent activation in adjacent cells (fig. 5). Subsequently, (possibly due to summation) the depressed pathway becomes partially excitable and allows retrograde or antegrade conduction in the previously unexcited cells, initiating a delayed secondary activation wave. In figure 4 panel A, note that the conduction delay occurs mainly at row. Conduction velocity below this level is normal and fires off of the trailing component of the slow activity in row. Previously reported neurophysiologic observations' and more recent cardiac electrophysiologic data", 0 suggest that this may represent an example of electrotonic summation. Ascribing these phenomena to summation would imply that the depressed area caused fragmentation of the RBB electrogram and the first stimulus reaching row is subthreshold and does not provide the necessary current needed to produce normal excitation of these fibers. However, it may interact electrotonically and slowly raise the resting potential of the fibers below this region. The elevation in potential of slowly activating fibers may result in summation of two or more impulses as long as the intervals between stimuli do not exceed the duration of the action potentials. Summation may then result in normal conduction once the wavefront reaches an area able to generate normal conduction velocity as is seen in figure 4, panel A. Depression of excitability may be distributed inhomogeneously and potentiate the asynchrony in activation, decrease the current density, and allow the underlying electrotonic currents to be manifest. The authors point out that this explanation is speculative and cannot be proven by the methods used in this experimental model. However, this concept is in agreement with the hypothesis of others that failure of normal propagation may allow the slower electrotonic currents to summate and generate normal conduction velocities downstream from areas of block.6 1' The term reentry, as originally used to mean reexcitation, may be inappropriate when applied to the secondary activation wave in the right bundle branch shown here. The duration of the refractory period of the right bundle is too long to permit reexcitation of tissue as rapidly as would be necessary to explain the secondary activation. It seems likely that those fibers carrying the delayed activation were not excited during the initial RBB depolarization. The concept of dual pathways has been well documented in both A-V node and distal A-V conducting systems,'0 17 and the data shown here would support this concept as applied to the RBB. Myerburg et al. and others have demonstrated separate longitudinal bundles with transverse connecting bridges of low electrical resistance within the bundle branches.'7, 4 This microanatomical arrangement would facilitate the transverse spread of electrotonic currents and thereby potentiate summation. These crossover regions were described by James and Sherf as transverse bundles smaller than the longitudinal conduction fibers.' A second type of crossover region described by Sommer and Johnson consists of long segments in which the cells' membranes are in close proximity to each other. If cable theory derived from nerve tissue5 is applicable to cardiac tissue, then the smaller crosssectional area would mean a higher resistance for the transverse bridges relative to the adjacent longitudinal bundles. The conduction delay caused by premature stimulation or manual pressure therefore may appear first in the transverse bridges (because of their higher resistance) and thereby favor dissociation of the RBB into separate longitudinal conduction tracts (figs. 4-6). The data presented show that manual pressure and premature depolarization may cause conduction delay and longitudinal dissociation of the RBB. This longitudinal dissociation facilitates formation of a secondary wave of activation antegrade or retrograde for variable distances. The time relationships suggest that the cells excited by the secondary wave were not previously depolarized during the primary wave of activation and, therefore, reentry did not occur. The demonstration here of multiple conducting pathways within the right bundle branch may explain the mechanism of some atrial and ventricular arrhythmias previously labeled as reentrant in origin. These data demonstrate another mechanism whereby arrhythmias may originate from within the right bundle branch under various types of stress. Acknowledgment The authors gratefully acknowledge the valuable contribution of Dr. Joseph C. Greenfield, Jr., who critically reviewed the manuscript during preparation. We are indebted to William Joyner, J. H. Kasell, and C. B. Clark for invaluable technical assistance. The Department of Medical Illustration at the Durham Veterans Hospital rendered valuable support. The secretarial assistance of Mrs. Brenda Haley and Mrs. Rosa Ethridge is gratefully acknowledged.

5 HIS BUNDLE AND BUNDLE BRANCHES/Massing, James 609 References 1. Clark J, Plonsey R: A mathematical evaluation of the core conductor model. Biophys J 6: 95, Weidmann S: The diffusion of radiopotassium across the intercalated disks of mammalian cardiac muscle. J Physiol 187:, Sperelakis N, Hoshiko T, Berne RM: Nonsyncytial nature of cardiac muscle: membrane resistance of single cells. Am J Physiol 198: 51, Plonsey R: Bioelectric Phenomena. New York, McGraw-Hill Book Company, Pruitt RD, Essex HE: Potential changes attending the excitation process in the atrioventricular conduction system of bovine and canine hearts. Circ Res 8: 149, Wennemark JR, Ruesta VJ, Brody DA: Microelectrode study of delayed conduction in the canine right bundle branch. Circ Res : 75, Watanabe Y, Dreifus LS: Inhomogeneous conduction in the A-V node. A model for re-entry. Am Heart J 70: 505, Clark JW Jr, Plonsey R: Fiber interaction in a nerve trunk. Biophys J 11: 81, Myerburg RJ, Steward JW, Hoffman BF: Electrophysiological properties of the canine peripheral A-V conducting system. Circ Res 6: 61, Moe GK, Preston JB, Burlington H: Physiologic evidence for a dual A-V transmission system. Circ Res 4: 57, Mines GR: On circulating excitations in heart muscles and their possible relation to tachycardia and fibrillation. Trans Roy Soc Canada Sec 4: 4, Mines GR: On dynamic equilibrium in the heart. Proc Physiol Soc (Lond) 46: 49, Garrey WE: The nature of fibrillary contraction of the heart: its relation to tissue mass and form. Am J Physiol : 97, Drury AN: Further observations upon intra-auricular block produced by pressure or cooling. Heart 1: 14, Schmitt FO, Erlanger J: Directional differences in the conduction of the impulse through heart muscle and their possible relation to extra-systolic and fibrillary contractions. Heart 1: 6, Lewis T: Mechanism and Graphic Registration of the Heart Beat, ed. Chicago, Chicago Med Book Company, Myerburg RJI, Nilsson K, Befeler B, Castellanos A Jr, Gelband H: Transverse spread and longitudinal dissociation in the distal A-V conducting system. J Clin Invest 5: 885, Myerburg RJ, Stewart JW, Hoffman BF: Electrophysiological properties of the canine peripheral A-V conducting system. Circ Res 6: 61, Cranefield PF, Klein HO, Hoffman BF: Conduction of the cardiac impulse. 1. Delay, block, and one-way block in depressed Purkinje fiber. Circ Res 8: 199, Cranefield PF, Hoffman BF: Conduction of the cardiac impulse. II. Summation and inhibition. Circ Res 8: 0, Singer DH, Lazzara R, Hoffman BF: Interrelationship between automaticity and conduction in Purkinje fibers. Circ Res 1: 57, Anderson GJ, Greenspan K, Bandura JP, Fisch C: Asynchrony of conduction within the canine specialized Purkinje fiber system. Circ Res 7: 691, Sommer JR, Johnson EA: Comparative ultrastructure of cardiac cell membrane specializations. Am J Cardiol 5: 184, James TN, Sherf L: Fine structure of the His bundle. Circulation 44: 9, Hodgkin AL, Rushton WAH: The electrical constants of a crustacean nerve fiber. Proc R Soc Lond B 1: 444, 1946 Anatomical Configuration of the His Bundle and Bundle Branches in the Human Heart GEORGE K. SUMMARY The relationships among the His bundle, the origin of both bundle branches, and the interventricular (IV) septum were examined histologically in human hearts, and the entire bundle branch systems were delineated in 1 of these. The His bundle in five hearts traversed the right IV septal crest, and the LBB origin was a very narrow stem (maximum 1.5 mm in cross-section) crossing from right to left through the inferior margin of the membranous septum. Proximal LBB anatomy was extremely variable, PRECISE KNOWLEDGE of the anatomy and distribution of the human atrioventricular (A-V) conduction system is important in understanding the sequence of ventricular activation and thereby the anatomical basis of a variety of conduction disorders. Such knowledge is also valuable for the cardiac surgeon who sometimes must place sutures or make incisions very near (or into) the A-V conduction system in the course of certain intracardiac operations. The anatomy From the Department of Medicine, University of Alabama Medical Center, Birmingham, Alabama. Supported by the National Heart and Lung Institute (MIRU Contract PH , SCOR on Ischemic Heart Disease No. 1 P17 HL 17,667 and Program Project Grant HL 11,10) and by the Rast Fund for Medical Research. Address for reprints: Thomas N. James, M.D., Department of Medicine, University of Alabama Medical Center, Birmingham, Alabama 594. Received October, 1975; revision accepted for publication November 19, MASSING, M.D., AND THOMAS N. JAMES, M.D. demonstrating multiple fiber groups which fanned out over the entire left septal surface. The LBB did not divide into two discrete divisions without multiple interconnections. The RBB formed an obtuse angle with the His bundle in 7 of hearts. In those five hearts with "right-sided His bundles," the right bundle branch was a direct continuation. The clinical, electrophysiologic, and electrocardiographic implications of these anatomical observations are discussed. of the human A-V node has been carefully studied in the past1' but less attention has been given to important anatomical relationships among the His bundle, the origin and course of both bundle branches, and the interventricular (IV) septum. Previous studies of human left bundle branch (LBB) anatomy have variously described the LBB: 1) to be divided into two discrete divisions without proximal interconnections; ) to have three rather than two separate divisions;4 and ) to be a diffuse fanlike structure broadly distributed over the left septal surface.5' 6 Rosenbaums proposed a trifascicular concept of A-V conduction based upon his anatomical studies supporting bidivisional LBB anatomy, and then described electrocardiographic criteria to identify impaired conduction within the anterior and posterior divisions of the LBB. Uhley7 has subsequently