Selective Functional Properties of Dual Atrioventricular Nodal Inputs
Role in Nodal Conduction, Refractoriness, Summation, and Rate-Dependent Function in Rabbit Heart
Background The atrioventricular node receives its activation signal from the low crista terminalis and low interatrial septum, the summation of which is believed to favor conduction. A functional asymmetry between the inputs is also believed to be involved in nodal reentrant rhythms. We studied the selective functional characteristics of nodal inputs and determined their role in nodal conduction, refractoriness, summation, and rate-dependent function.
Methods and Results The nodal properties of recovery, facilitation, and fatigue were characterized with stimulation protocols applied with varying phases between the two inputs in isolated rabbit heart preparations. The effects of the input phase, nodal functional state, and input reference on the nodal conduction time, recovery time, and refractory periods were assessed with multifactorial ANOVAs. It was found that the phase of stimulation significantly affected nodal conduction time but not the refractory periods or the time constant of the recovery. Each input could show longer and shorter conduction time than the other depending on the stimulation phase, input reference, and coupling interval. These effects were similar for different nodal functional states. However, pacing and recording from the low crista resulted in similar conduction and refractory values than did pacing and recording from the low septum. Input summation did not increase the otherwise equal efficacy of individual input in activating the node. Nodal surface recordings confirmed this functional symmetry and equivalent efficacy of the inputs and showed that input effects were confined to the proximal node.
Conclusions The two nodal inputs have equivalent functional properties and are equally effective in activating the rate-dependent portion of the node. Input interaction affects perinodal activation but not the rate-dependent nodal function.
The AV node is activated by atrial wave fronts traveling via the low crista terminalis and interatrial septum. These wave fronts converge and activate the central part of the node.1 2 3 Several studies have indicated that the interaction between these inputs affects nodal function. Zipes et al4 separated the two inputs with an incision in the floor of the coronary sinus and showed that in a narrow range of short atrial coupling intervals, the stimulation of either input alone resulted in a blocked beat, whereas their combined stimulation resulted in a propagated beat, a phenomenon called nodal summation. Watanabe and Dreifus5 similarly observed that successful nodal propagation of early premature beats occurred only when two independent wave fronts simultaneously arrived at a junctional point. The difference between the two inputs in sustaining 1:1 nodal conduction has also been interpreted as evidence for summation within the activation wave.1 The input interaction also affects nodal cell action potential morphology and activation pattern.1 6 7 8 Similarly, summation of the activity from different cells is required for a local nodal stimulation to result in a propagated beat.9 Based on these observations, input summation is frequently postulated to be involved in a variety of clinically observed nodal responses. However, functional studies have indicated that input summation plays a minor role in normal AV nodal function.10 11 There also is abundant indirect evidence that the two nodal inputs are functionally asymmetrical.12 13 14 15 16 17 The low septum input is part of a fast pathway with a long refractory period, and the low crista is part of a slow pathway with a short refractory period. A shift in conduction from the fast to the slow pathway is manifested by a sudden jump in the nodal recovery curve and the occurrence of reentrant beats or tachycardia. However, local stimulation and recording studies have not confirmed this asymmetry.11 18 19 The present study was designed (1) to assess the selective functional characteristics of the two nodal inputs, (2) to determine the differences in their conduction and refractory properties, (3) to characterize the effects of input summation on nodal rate-dependent function, and (4) to determine the intranodal origin of input-related changes in NCT.
Preparation and Apparatus
Experiments were performed in two groups of six superfused, isolated rabbit heart preparations. Animal care was conducted according to the guidelines of the American Physiological Society and Universite´ de Montre´al. The preparation, perfusion system, stimulation techniques, and recording system were as described previously.11 20 21 Briefly, the preparation (Fig 1A⇓), which included the right atrium, AV node area, and upper portion of the ventricular septum, was mounted in a tissue bath perfused at 200 mL/min with a 6-L volume of oxygenated (95% O2/5% CO2) Tyrode's solution maintained at 37°C, pH 7.38. Its composition (in mmol/L) was NaCl 128.2, KCl 4.7, CaCl2 2.0, MgCl2 1.0, NaHCO3 20, NaH2PO4 0.7, and dextrose 11.1. In all preparations, bipolar platinum-iridium stimulation electrodes were placed on the upper atrium (crista terminalis near sinus node region), low crista terminalis beneath the opening of the coronary sinus, and low interatrial septum.11 Unipolar electrograms were recorded from the upper atrium, low crista, low septum, and His bundle. The indifferent electrode was positioned 3 cm from the recording electrodes in the perfusion bath. Stimulation and recording electrodes were ≈1 mm from each other. In the second group of preparations, another electrogram was also taken from the surface of the AV node. The nodal electrode was positioned with a micromanipulator under visual control through a dissecting microscope to obtain an activation complex that occurred at approximately one third of the nodal delay. All recording electrodes consisted of a sharply cut Teflon-insulated 0.25 mm silver wire. Electrograms were recorded on a videotape with the stimulation pulse, a time code, and a tachogram and were analyzed off-line. Bandwidth was 0.1 Hz to 3 kHz. Stimulation sequences were generated with 1-ms resolution and 0.47-μs precision with a locally developed computer algorithm. Stimulation voltage pulses were twice threshold and had a 2-ms duration.
The nodal properties of recovery, facilitation, and fatigue20 21 22 were determined for each of five phases (−15, −7, 0, +7, and +15 ms) between input stimuli (Fig 1A and 1B⇑⇑). Negative and positive phases correspond to those in which the low crista is stimulated before the low septum, and vice versa, respectively (Fig 1B⇑). The three protocols required for the characterization of the nodal properties were as described previously.20 21 22 The pulse sequences involved are as follows—recovery: 20 L and 1 P; facilitation: 20 L, 1 S, and 1 P; and fatigue: 20 S, 1 L, and 1 P*, where L is long cycle, P is test premature cycle, and S is short cycle (*first P after 5 minutes of S).
The long cycle (L) was imposed with a His-stimulus interval (343±83 ms) that yielded a His-atrial interval 30 ms shorter than the His-atrial interval observed during sinus rhythm. The short cycle (S), determined with an incremental pacing protocol, was equal to the minimum His-stimulus interval (70±10 ms) plus 30 ms that consistently resulted in persistent 1:1 nodal conduction at all pacing sites. The premature His-stimulus interval (P) was reduced by 20-, 10-, and 2-ms steps in its long, intermediate, and short range, respectively. A recovery protocol with stimuli applied to the upper atrium was also performed at the beginning, middle, and end of the experiments to detect temporal drifts. Mean NCT changes between the beginning and end of experiments were <2 ms.
To assess the relative efficacy of phase stimulation compared with single input stimulation, a recovery protocol was performed at the control basic cycle length (L) for each phase (as described for group 1) and for each atrial pacing site. The dissociation of the roles of the proximal and central node in the input-related changes in nodal function was achieved with the use of measurements made on the continuously recorded nodal surface electrogram. Stability was similar to that of group 1 experiments.
Interval Measurements and Statistical Analyses
Activation times at the three atrial (A) recording sites and at the His bundle (H) were determined with 0.2-ms precision. For this purpose, the electrograms were digitized at 5 kHz per channel with the Asyst program (Keithley) and analyzed with the Data Pack program (Run Technologies). Nodal responses obtained during each stimulation protocol were represented as a recovery curve constructed by plotting each premature NCT (AH) against the preceding His-atrial (HA) interval.20 21 22 Recovery curves arising from different protocols, stimulation phases, and reference sites are identified as such with different symbols on the graphs. Changes in parameters representing nodal function were assessed with multifactorial ANOVAs for repeated measures.23 For group 1 data, the differences related to the stimulation phase (−15, −7, 0, +7, and +15 ms), protocol (control, facilitation, and fatigue), and reference (low crista and low septum) were determined through a single analysis for each parameter. For group 2 data, the differences related to the input phase, atrial pacing site, and reference were also assessed through a single analysis for each parameter. Group 2 also involved the measurement of the proximal conduction time (AECT or AEIAS) and distal conduction time (EH). Data are given as mean±SD.
Our definition of the AV node includes anatomic structures corresponding to transitional, midnodal, and lower nodal cells.8 24 25 Nodal recovery, facilitation, and fatigue properties were as defined previously.20 21 22 25 Briefly, recovery refers to the slow and progressive recovery of excitability that causes the NCT to increase with prematurity.9 Facilitation refers to NCT shortening observed in the short coupling interval range when the premature cycle is preceded by a short cycle. Fatigue causes a rate- and time-dependent prolongation of NCT for any given recovery time or facilitation level. ERPN is the longest AA resulting in a nodal block or, when an atrial block occurred before the nodal block, the shortest AA resulting in a conducted beat.26 27 28 Because ERPN increases with the NCT that precedes the premature beat regardless of nodal refractoriness,29 30 ERPNc was also determined. ERPNc is equal to ERPN minus the increase over control value in the last NCT preceding the premature beat.30 The FRPN is the minimum interval reached between two His-bundle responses. The RTC is the time constant resulting from the single exponential approximation of the recovery curve.
Effects of Input Phase on Rate-Dependent Nodal Conduction and Refractory Properties
Control Recovery Property
The effects of the input phase on the NCT measured from both inputs are illustrated in Fig 2⇓, showing the five input phases (−15, −7, 0, +7, and +15 ms). Each section shows the electrograms and NCT values obtained from one preparation at the longest coupling interval (HS 280 ms) during the recovery protocol. As shown by the time delay between CT and IAS activation signals, the phase progressively changes the activation sequence of the two inputs. The resulting NCT measured from the low crista (AHCT) decreases from 77 to 55 ms while the NCT measured from the low septum (AHIAS) increases from 61 to 70 ms.
The input phase also produced opposite shifts of the nodal recovery curves constructed with the two input references (Fig 3⇓). The five recovery curves superimposed in Fig 3A⇓, and constructed using the low crista reference, show a progressive downward shift with varying input phase. The same nodal responses assessed from the low septum resulted in an upward shift of the recovery curve (Fig 3B⇓). The increasing input phase also shifts the recovery curve to the right in Fig 3A⇓ and to the left in Fig 3B⇓. The recovery curves cross and thus show opposite phase effects in the long and short coupling interval ranges. Low septum curves also show a reduced dispersion in the short coupling interval range. Shortest NCT values were obtained at phases +15 and −15 ms when measured from low crista and low septum, respectively. Phase 0, which presumably results in the most effective nodal input, never resulted in the shortest NCT. Mean values of AHmin (AH obtained at longest coupling interval) and HAmin (shortest HA reached) are shown in Table 1⇓ and Fig 4A and 4B⇓⇓. When measured from low crista, AHmin decreased from 66±8 to 46±14 ms between phases −15 and +15 ms. Corresponding values measured from low septum increased from 53±8 to 59±10 ms. The changes in HAmin were opposite in direction as compared to those of AHmin. When measured from low crista, HAmin increased from 40±16 to 62±14 ms between phases −15 and +15 ms but decreased from 59±13 to 49±16 ms on low septum measurements. These effects were all statistically significant (Table 2⇓). Notably, the effects of the phase on AHmin varied significantly with the reference (φ versus R), a finding reflecting a slight difference in the magnitude of the effects seen at the two inputs. The effects of the reference taken globally were not statistically significant because opposite variations from the two inputs canceled each other. Thus, the input phase has opposite effects on NCT and recovery time; these effects are also opposite at the two references and at long and short coupling intervals.
The effects of the input phase on the other control nodal parameters (RTC, ERPN, ERPNc, and FRPN) were not statistically significant. Mean values of these parameters and significance of corresponding statistical comparisons are given in Tables 1 and 2⇑⇑, respectively. Mean changes in the ERPN and FRPN are also illustrated in Fig 4C and 4D⇑⇑, respectively. As reflected by resulting horizontal curves, control nodal refractoriness was insensitive to the input phase.
Nodal Facilitation and Fatigue
To assess whether the effects of the input phase change with the nodal functional state, the facilitation and fatigue protocols were imposed for each input phase in group 1 preparations. Results obtained in a typical preparation at phase −15, 0 and +15 ms are illustrated in Fig 5⇓ for both the low crista (panels A, B, and C) and low septum (panels D, E, and F) references. Each panel is a superimposition of the recovery, facilitation and fatigue curves obtained at the specified phase. Beside small differences observed mainly in the short HA range, the recovery, facilitation and fatigue protocols resulted in similar relative changes in NCT for the different input phases. Mean values (Table 1⇑) also show that the NCT and other nodal parameters change in the expected manner with the facilitation and fatigue protocols,11 20 21 22 25 and that these changes are similar for the different input phases. This is also supported by the absence of statistically significant interaction between the input phase and the protocol (φ versus P in Table 2⇑); the nodal functional state did not alter the effects of the input phase. The low crista and low septum reference yielded similar results. In summary, the effects of the input phase are independent of the nodal functional state.
Nodal Origin of Input Effects
The nodal origin of the phase effects on NCT was studied in group 2 preparations. An electrogram was recorded from the surface of the node while we repeated the control recovery protocol for the 5 input phases. This electrogram permitted the division of the AH into a proximal and a distal component. Nodal electrograms (E) obtained in one preparation at longest coupling interval for three input phases (−15, 0 and +15 ms) are illustrated in Fig 6⇓. These recordings and the conduction intervals listed show that the input phase changes the interval from the input to the nodal complex (AE) while the interval from the nodal complex to the His bundle (EH) remains nearly constant. For instance, AECT (AE measured from low crista) decreased from 32 to 14 ms between phases −15 and +15 ms while AEIAS (AE measured from low septum) increased from 15 to 27 ms. Corresponding EH variations were negligible. Mean data confirms these observations (Fig 7⇓ and Table 3⇓). Mean AE obtained at the longest coupling interval decreased from 18 to 4 ms between phases −15 and +15 ms. The inverse was observed at the low septum reference. Fig 7A⇓ shows mean minimum AH measured from the two references for the 5 input phases. Fig 7B⇓ shows corresponding mean AE. The pattern of changes of AE with input phase and reference is remarkably similar to that of AH. Conversely, Fig 7C⇓ shows no input-related changes in mean EH. When the same preparations were subjected to upper atrial, low crista and low septum pacing one at a time, the resulting changes in AH also arose mainly from changes in the AE (data columns at the right of Table 3⇓). While AH and AE similarly vary with atrial pacing site, EH remains constant. For instance, the mean AE measured from the low crista was 19 ms for low crista pacing and −2 ms for low septum pacing while the corresponding EH was 43 and 40 ms, respectively. These findings show that the effects of the input phase and atrial pacing site on NCT arise from the proximal third of the node (likely the transitional region).
Effects of Synchronous Versus Asynchronous Inputs on Nodal Function
The effects of input synchrony on NCT were assessed by comparing the mean minimum AH values obtained during phase stimulation to those obtained while pacing the low crista, low septum or upper atrium alone (Table 3⇑). Mean AH measured from low crista at phase −15 ms (61±4 ms) was very similar to that obtained with low crista pacing alone (63±6 ms). Phase 15 ms stimulation resulted in an AH (60±10 ms) that was very close to the value obtained with single low septum stimulation (60±13 ms). Upper atrial and phase 0 ms stimulations also yielded similar AH values. These findings show that equivalent nodal inputs result in similar NCT even when they arise from a different number of pacing sites.
The possibility that the input summation resulted in improved conduction in the proximal node was also assessed. Data from Table 3⇑ shows that the AH decreases from 61±4 ms to 56±6 ms in going from phase −15 to 0 ms when measured from the low crista. A similar decrease is observed in going from phase 15 to 0 ms when measuring from the low septum. Stimulation from the upper atrium results in shorter AH (55±6 ms) than either of the single inputs (CTR, 63±6 ms) and (IASR, 60±13 ms). The data also show that the AE portion of the AH is the main contributor to these changes. These effects were small and at the limit of statistical significance. The EH and other nodal parameters did not change significantly. These findings suggest that synchronous as compared to asynchronous nodal inputs may marginally shorten conduction time in the proximal third of the node.
The present study selectively characterizes the functional properties of the low crista and low septum inputs to the AV node and determines their role in nodal function. The findings demonstrate that the two nodal inputs have equivalent conduction and refractory properties, and are equally effective in activating the rate-dependent portion of the node. Local stimulation of the inputs either individually or in different phasic relationships did not significantly affect nodal recovery and refractory parameters nor their rate-dependent variations. These findings indicate that input interaction is a minor determinant of normal nodal properties. Moreover, nodal surface potentials recorded during input modulations confirmed the functional symmetry of the two inputs and their similar efficacy in activating the rate-dependent portion of the node. These recordings also help to establish that the input-related changes in NCT come mainly from changes in perinodal activation. Furthermore, these NCT variations were found to depend on the input phase, reference site, and coupling interval and to simulate both slow or fast conduction at either input without actual changes in input conduction velocity. Our results suggest that the proximal node acts as a matching gate that enables widely varying atrial wave fronts to result into equally effective activation signals for the rate-dependent portion of the node.
Functional Symmetry of Dual AV Nodal Inputs
A previous study showed that pacing and recording from the low crista result in similar NCT and refractory values as pacing and recording from the low septum.11 The present study formally establishes this functional symmetry of the inputs with modulations of the time relationship between the inputs and nodal surface recordings (Table⇑s 1 and 3). The AE initiated and measured from the low crista did not differ significantly from AE initiated and measured from the low septum (Fig 7⇑ and Table 3⇑). Moreover, phase-related changes in AE, which accounted for almost all AH changes, were nearly symmetric at the two inputs. The continuity, symmetry and bidirectional nature of AE and AH changes demonstrate that both inputs have equivalent properties and initiate similar responses from the rate-dependent portion of the node. This is in agreement with recent anatomical and functional studies that failed to identify a genuine difference in the anatomical and electrophysiological properties of the inputs.11 18 19 31 However, reports are not yet unanimous on this issue.17
Origin of Apparent Slow and Fast Input Conduction
Despite their functional symmetry, both nodal inputs showed apparent characteristics of slow and fast conduction depending on the pacing site, reference site and recovery interval (Figs 2 through 4⇑⇑⇑ and Table 1⇑). The NCT resulting from a given input stimulation could both appear short or long depending on the reference from which it was measured. The effect was inverted at the two inputs and at short compared with long coupling intervals. This multiform input asymmetry originated from measurement biases introduced by changes in perinodal activation pattern (Figs 6 and 7⇑⇑ and Table 3⇑). This can be easily understood from changes in AE. An example is the marked decrease in AECT observed between phase −15 and +15 ms. At phase −15 ms, the low crista recording electrode is located on the main stream of nodal activation and accurately detects its beginning. The resulting AECT closely reflects the real transit time of the impulse in the proximal node. Conversely, a measurement taken from the low septum in the same circumstances detects the activation after it has already entered the node11 indicated by a markedly shortened or negative AEIAS (Fig 7B⇑ and Table 3⇑). The same nodal response and thus the same conduction velocity is then seen differently from the two inputs. A similar phenomenon is observed when the low crista is paced alone and NCT is measured from the low septum. Another analogous situation occurs when an impulse reaches the node from the posterior input and NCT is measured from the atrial complex of the His bundle derivation. A shorter than real NCT is then expected. An impulse entering the node from the low septum (phase +15 ms or low septum pacing) results in an inverse bias. While the low septum electrode then accurately detects the beginning of nodal activation, the low crista electrode detects the impulse after it has entered the node indicated by a shortened AECT. For yet unknown reasons, CT measurements were somewhat more affected than IAS measurements by this reference bias (Fig 3⇑). The extent to which this bias contributed to previously reported differences between nodal inputs1 7 8 12 13 17 32 cannot be rigorously established; major differences in methods, NCT baseline values and lack of comparable statistical analyses prevent objective comparisons. In summary, AE and thereby AH can change in different directions with perinodal activation pattern and simulate slow or fast conduction at either input without actual changes in input conduction velocity.
Nodal Input Interaction and Rate-Dependent Function
The independence of the rate-dependent nodal function from the inputs is supported by several observations. The NCT, recovery time constant, and effective refractory period did not vary significantly with the input from which they were assessed (Table⇑s 1 and 2). Synchronous versus asynchronous inputs had no effects on these parameters either. This remained true for different nodal functional states. Similarly, the functional refractory period of the node did not vary significantly either with the input phase and number of inputs. Nodal surface recordings also show that the rate-dependent portion of the NCT was insensitive to input interaction as indicated by a constant EH for widely different input conditions (Fig 7⇑ and Table 3⇑).
The only effect that was compatible with input summation was the slight AE and AH shortenings observed in going from phase −15 to 0 ms (Table⇑s 1 and 3) and during upper atrial pacing as compared to low crista or low septum pacing. However, this trend was small and at the limit of statistical significance. Thus, if there was any summation, it was small and confined to the proximal node. The safety margin of atrionodal coupling may explain this lack of effect of input interaction on rate-dependent nodal function; summation is manifest only when the ratio of available input energy to the energy necessary to activate the node is close to one. In the present study, the reduced nodal excitability produced by early prematurity or by fatigue did not result into manifest summation. It remains possible that further impairment of the conductivity caused by drugs or hypoxia would do so.33 This may also be the case in patients with severely depressed nodal conductivity.
Relationship to Human AV Node Physiology
The application of the present findings to human AV node physiology will obviously require specifically designed studies. Comparison of our data with that of previous clinical studies is limited by major differences in stimulation and measurement approaches. Nevertheless, it might be useful to discuss the potential implications of our findings for the interpretation of studies on nodal rate-dependent function and dual pathway physiology.
Assessment of Rate-Dependent AV Nodal Function
The results on the effects of input interaction on rate-dependent nodal function showed that the AE fraction of the AH changes with perinodal activation pattern while nodal recovery and refractory properties remain unaffected. A practical consequence of this finding is that the nodal recovery and refractory properties can be accurately assessed for different input conditions. If applicable in human studies, it will mean that the atrial activation pattern should be a concern only to the extent that it affects NCT baseline. Our findings also show that input-related changes in NCT values are largely due to an inaccurate detection of the real beginning of nodal activation, a problem previously recognized in some clinical studies.14 34 35 36 This bias is easily controlled by pacing and recording from the same input. However, in the more standard condition of upper atrial pacing, the atrial and thereby nodal input activation pattern vary and prevent such a control. We suggest that the resulting measurement bias will be minimized by recording from the two inputs and measuring NCT from the input which shows the earliest activity and is thus likely dominant.9 11 Another potential clinical development that may arise from the present study is the selective determination of the contribution of the proximal and central nodes to nodal function. If improvements in recording techniques give access to nodal potentials equivalent to those achieved in the present study, it will become possible to dissociate input and rate-dependent factors in human AV nodal responses.
Dual Pathway Physiology
During sinus rhythm, the human AV node is usually activated from the septal input.37 38 Our findings suggest that an NCT measurement made from the atrial complex of the His bundle derivation during such an activation pattern closely approximates real nodal activation time. In other words, fast pathway NCT is then truly reflecting nodal activation. However, an NCT measured from the posterior input in the same circumstances is bound to be artificially shorter; the posterior input is activated after the real beginning of nodal activation. Although rarely acknowledged, such shortening can be readily seen in many records which include a proximal coronary sinus electrogram.39 40 Our findings also suggest that, in a case where the posterior input is dominant, a shorter than real NCT will be obtained from a measurement made from the His bundle derivation as reported by Suzuki et al.41 This bidirectional bias renders local measurement of conduction velocity from both inputs necessary before concluding to their slow or fast nature. This applies to both normal and dual pathway physiology. If, as recently suggested by McGuire and Janse,19 the two main atrial inputs are the anatomically relevant structures for slow and fast conduction in dual pathway physiology, it would be important to dissociate the contribution of measurement biases from that of true differences in input conduction velocity.
Our findings predict that, even for symmetric functional properties of the inputs, an impulse entering the node from the low septum could be blocked at the anterior nodal input but propagated from the posterior input, thus resulting in an NCT prolongation when measured from the His bundle lead. This prolongation is due to the extra traveling time of the impulse toward and in the low crista input. This will be manifested by an upward shift of the recovery curve. The potential contribution of such a phenomenon to the jump in the recovery curve observed during dual pathway physiology remains to be assessed. Such a shift could, for yet unknown reasons, be amplified in patients suffering from dual pathway physiology and reentry. As the recovery curves of these patients frequently differ from those of normal subjects, it is possible that additional factors are involved. However, there is evidence for both jump in the recovery curve not producing nodal reentry and nodal reentry not associated to jump39 so that it is difficult to establish the boundary between normal and abnormal physiology in this reentry. Alternatively, one may speculate that such a shift in atrial activation pattern simply does not occur in normal subjects or affects the recovery curve differently. The steep rise without discontinuity of the curves observed at short coupling intervals may be such a manifestation (Fig 3⇑). These issues can only be resolved in human studies designed to achieve a rigorous control over the effects of the pacing and recording sites on NCT measurements during both normal and dual pathway physiology.
Although nodal surface electrograms clearly show that the input effects are confined to the proximal node, the exact intranodal origin of these electrograms remains uncertain. Their mean activation time corresponds to that of the late atrionodal cells but this link remains to be confirmed.35 Another unknown is the exact change in perinodal activation pattern produced by the different input modulation procedures used in the present study. The determination of these changes will require mapping techniques performed for selected beats reproducing typical effects observed with an approach analogous to our experimental one. The present study also does not resolve the issue of the exact location of the input convergence or the nature of their electrical connection. This uncertainty was in fact at the origin of our decision not to surgically separate the inputs as previously done by others.4 Such a cut alters nodal input and rate-dependent function, as suggested by markedly prolonged NCT even in the presence of synchronous inputs. This prolongation was a major concern, as we wanted to study input effects on normal rate-dependent nodal function. Our conclusions therefore only apply for conditions of normally interconnected inputs.
Selected Abbreviations and Acronyms
|AECT||=||time from low crista reference to the nodal complex|
|AEIAS||=||time from low septum reference to the nodal complex|
|EH||=||time from the nodal complex to the His bundle|
|ERPN||=||effective refractory period of the atrioventricular node|
|ERPNc||=||corrected effective refractory period of the atrioventricular node|
|FRPN||=||functional refractory period of the atrioventricular node|
|NCT||=||nodal conduction time|
|RTC||=||recovery time constant|
The research was supported by the Medical Research Council of Canada, Quebec Heart and Stroke Foundation, and Fonds de la Recherche en Sante´ du Que´bec. We thank Dr Alvin Shrier for his comments on the manuscript and Maurice Tremblay and Lise Plamondon for their technical assistance. We also thank Dr Michel Lamoureux and Dr Alain Vinet for their contribution to statistical analyses.
- Received December 27, 1995.
- Revision received February 12, 1996.
- Accepted February 17, 1996.
- Copyright © 1996 by American Heart Association
Janse MJ. Influence of the direction of the atrial wave front of A-V nodal transmission in isolated hearts of rabbits. Circ Res. 1969;25:439-449.
Van Capelle FJL, Janse MJ, Varghese PJ, Freud GE, Mater C, Durrer D. Spread of excitation in the atrioventricular node of isolated rabbit hearts studied by multiple microelectrode recording. Circ Res. 1972;31:602-616.
Spach MS, Lieberman M, Scott JG, Barr RC, Johnson EA, Kootsey JM. Excitation sequences of the atrial septum and the AV node in isolated hearts of the dog and rabbit. Circ Res. 1971;29:156-172.
Zipes DP, Mendez C, Moe GK. Evidence for summation and voltage dependency in rabbit atrioventricular nodal fibers. Circ Res. 1973;32:170-177.
Cranefield PF, Hoffman BF, Paes De Carvalho A. Effects of acetylcholine on single fibers of the atrioventricular node. Circ Res. 1959;7:19-23.
Mazgalev T, Dreifus LS, Iinuma H, Michelson EL. Effects of the site and timing of atrioventricular nodal input on atrioventricular conduction in the isolated perfused rabbit heart. Circulation. 1984;70:748-759.
Merideth J, Mendez C, Mueller WJ, Moe GK. Electrical excitability for atrioventricular nodal cells. Circ Res. 1968;23:69-85.
Sanchis J, Chorro FJ, Such L, Matamoros J, Monmeneu JV, Cortina J, Merino L. Effects of site, summation and asynchronism of inputs on atrioventricular nodal conduction and refractoriness. Eur Heart J. 1993;14:1421-1426.
Amellal F, Billette J. Effects of the atrial pacing site on rate-dependent AV nodal function in the rabbit heart. Am J Physiol. 1995;269:H934-H942.
Moe GK, Preston JB, Burlington H. Physiologic evidence for a dual A-V transmission system. Circ Res. 1956;4:357-375.
Mendez C, Moe GK. Demonstration of a dual A-V nodal conduction system in the isolated rabbit heart. Circ Res. 1966;19:378-393.
Ross DL, Brugada P, Bar FWHM, Vanagt EJDM, Weiner I, Farre J, Wellens HJJ. Comparison of right and left atrial stimulation in demonstration of dual atrioventricular nodal pathways and induction of intranodal reentry. Circulation. 1981;64:1051-1058.
Akhtar M, Jazayeri MR, Sra J, Blank Z, Deshpande S, Dhala A. Atrioventricular nodal reentry. Clinical, electrophysiological and therapeutic considerations. Circulation. 1993;88:282-295.
Stein KM, Lerman BB. Evidence for functionally distinct dual atrial inputs to the human AV node. Am J Physiol. 1994;:H2333-H2341.
Billette J, Me´tayer R. Origin, domain and dynamics of rate-induced variations of functional refractory period in rabbit atrioventricular node. Circ Res. 1989;65:164-175.
Dixon WJ. BMDP: Biomedical Computer Program. Berkeley, Calif: University of California Press; 1975:711-760.
Anderson RH, Janse MJ, Van Capelle FJL, Billette J, Becker AE, Durrer D. A combined morphological and electrophysiological study of the atrioventricular node of the rabbit heart. Circ Res. 1974;35:909-922.
Billette J, Shrier A. Atrioventricular nodal activation and functional properties. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology: From Cell to Bedside. Philadelphia, Pa: WB Saunders; 1995:216-228.
DuBrow IW, Fischer EA, Amat y Leon F, Denes P, Wu D, Rosen KM, Hastreiter AR. Comparison of cardiac refractory periods in children and adults. Circulation. 1975;51:485-491.
Nattel S, Talajic M, Quantz M, Deroode M. Frequency-dependent effects of amiodarone on atrioventricular nodal function and slow-channel action potentials: evidence for calcium channel-blocking activity. Circulation. 1987;76:442-449.
Zhao J, Billette J. Beat-to-beat changes in AV nodal refractory and recovery properties during Wenckebach cycles. Am J Physiol. 1992;262:H1899-H1907.
Batsford WP, Akhtar M, Caracta AR, Josephson ME, Seides SF, Damato AN. Effect of atrial stimulation site on the electrophysiological properties of the atrioventricular node in man. Circulation. 1974;50:283-292.
Waldo AL, Vitikainen KJ, Kaiser GA, Malm JR, Hoffman BF. The P wave and P-R interval: effects of the site of origin of atrial depolarization. Circulation. 1970;42:653-671.
McGuire MA, Bourke JP, Robotin MC, Johnson DC, Meldrum-Hanna W, Nunn GR, Uther JB, Ross DL. High resolution mapping of Koch's triangle using sixty electrodes in humans with atrioventricular junctional (AV nodal) reentrant tachycardia. Circulation. 1993;88:2315-2328.
Josephson ME. Supraventricular tachycardias. In: Clinical Cardiac Electrophysiology: Techniques and Interpretations. Philadelphia, Pa: Lea & Febiger, 1993:181-274.
Jazayeri MR. Atrioventricular nodal reentrant tachycardia: characterization of the reentrant circuit before and after selective fast or slow pathway ablation. In: Shenasa M, Borggrefe M, Breithardt G. eds. Cardiac Mapping. Mount Kisco, NY; Futura; 1993:411-434.
Billette J. Atrioventricular nodal activation during periodic premature stimulation of the atrium. Am J Physiol. 1987;252:H163-H177.