Relationship of the Functional Refractory Period to Conduction in the Atrioventricular Node

Relationship of the Functional Refractory Period to Conduction in the Atrioventricular Node By Gregory R. Ferrier and Peter E. Dresel ABSTRACT The functional refractory period of atrioventricular (AV) transmission has been accepted as a measure of AV nodal refractoriness and has been assumed to be determined solely by conduction of interpolated extrasystoles through the AV node when it is partially refractory. In the present study, we found an important effect of the conduction time of the regular beats by measuring AV nodal conduction times of atrial extrasystoles from the His bundle of isolated, blood-perfused dog hearts. We separated three independent components that determine AV nodal conductivity: (1) a "basal conduction time" measured with a "postmature" extrasystole at low heart rates (<120 /min), (2) a rate-dependent increment in conduction time (previously called "fatigue") that affects both normal and premature cycles equally and (3) an exponential change in conduction time that depends entirely on the immediately preceding interval and, therefore, is not further affected by heart rate. The functional refractory period is one point defining this continuous exponential function. We showed that an important cause of the decrease in the functional refractory period that occurs when heart rate is increased is the change in the conduction time of the regular beats. KEY WORDS electrophysiology His bundle extrasystole conductivity dog heart rate epinephrine Krayer et al. (1) have defined the functional refractory period of the atrioventricular (AV) transmission system as the minimum interval between two ventricular impulses propagated from the atria. They have also demonstrated that both the functional refractory period and the effective refractory period are decreased by administration of epinephrine. This observation implies a relationship between the two refractory periods. Although this relationship was not discussed by Krayer et al. (1), the functional refractory period has been interpreted as an estimate or an equivalent of the effective refractory period (2, 3). The effect of heart rate on AV nodal transmission is not clear. The functional refractory period of the AV transmission system has been shown to decrease when heart rate is increased (4-7). A decrease in the functional refractory period implies an increase in conductivity. However, Lewis and Master (8) have reported that the P-R interval of From the Department of Pharmacology and Therapeutics, Faculty of Medicine, University of Manitoba, Winnipeg, Manitoba, Canada R3E OW3. This study was supported by the Medical Research Council of Canada. Dr. Ferrier's present address is Masonic Medical Research Laboratory, Utica, New York 13501. Received August 3, 1973. Accepted for publication April 18, 1974. 204 early premature beats is decreased as heart rate is increased. Rosenblueth (5) has also reported that increases in heart rate cause increases in conductivity and decreases in the functional refractory period of the AV transmission system. But, when retrograde conduction is examined, no change in conductivity is observed (5). Both of these studies can be criticized for neglecting the role of atrial conduction (9). However, the AV conduction time of a premature response can be prolonged long after the expiration of the functional refractory period of the AV transmission system (1), and the AV nodal tissue can remain relatively refractory for a significant period after the termination of the action potential (). In the present study, we observed an apparent inconsistency between changes in the functional refractory period and changes in AV nodal conduction times of premature responses in experiments in which a single atrial premature response was elicited exactly halfway between two beats of a given basic interval. Under such conditions, the first of the new series of basic beats can be considered to be the second of two equally spaced extrasystoles (11). Since the first extrasystole should have an abbreviated refractory period (4), the second extrasystole should be conducted through the AV node more rapidly than the first. Circulation Research, Vol. 35. August 1974

FUNCTIONAL REFRACTORY PERIOD AND CONDUCTION 205 However, we found that the second extrasystole was conducted more slowly than the first in all hearts examined (5-40 in seven hearts). These results suggest that either the refractory period does not decrease following a short preceding cycle or that conductivity is not directly related to the functional refractory period. The present study tested these possibilities. Methods Isolated dog hearts were perfused with blood by the method of Alanis et al. (12) as modified by Kirk and Dresel (13). Blood was obtained from a large dog of either sex that was anesthetized with sodium pentobarbital (30 mg/kg, iv) and respired with air by a Palmer Ideal respiration pump. Blood from the femoral artery of one limb was pumped through an external perfusion circuit and returned to the femoral vein of the same limb by a double-headed Debakey roller pump. Heparin ( IU/ kg, iv) was administered to the donor dog prior to cannulation of the vessels. Hearts were obtained from small dogs (5-11 kg) anesthetized with sodium pentobarbital (35 mg/kg, iv). Each heart was placed in ice-cold, oxygenated Krebs-Henseleit solution, and the pericardium and other extraneous tissues were removed. The heart was then perfused via the aorta at a pressure maintained at mm Hg by adjusting the speed of the roller pump. The temperature of the perfusing blood was kept at 37 ± 0.5 C with a water-jacketed condenser situated immediately before the perfusion cannula in the external circuit. The right atrium of the isolated heart was incised, and the cut edges were retracted. The contacts of a flexible bipolar probe were placed over the proximal His bundle to record the activity of that structure and surrounding tissues. An additional electrogram from either the right atrial appendage or a site close to the coronary sinus was recorded with a pair of stainless steel clips. In most experiments the sinoatrial (SA) node was destroyed by heat cautery or by ligation of the major vessels supplying the region. Hearts were electrically paced through bipolar stainless steel clips attached to the right atrium near the SA node. Stimuli were 5- rectangular pulses, twice the threshold voltage, generated by a Tektronix stimulator (type 162 and type 161) and passed through an isolation transformer. The output of the stimulator was counted by a scale-of-ten counter which in turn triggered a second Tektronix stimulator to introduce a test stimulus following every ten basic stimuli. The gate of the second wave-form generator was used in some experiments to interrupt the basic stimulator and thus delay the next train of stimuli. Test stimuli could thus be delivered at intervals longer than the basic interval. Basic and test stimuli were delivered through the same electrodes. The recordings from the His bundle and right atrium were displayed on a four-beam double-cathode oscilloscope (Tektronix RM565 with 3A3 amplifiers) and photographed with a 35-mm Shackman camera. The stimulator which generated the basic stimuli triggered one pair of beams. The remaining pair of beams was triggered by the stimulator which generated the test stimuli. Photo- Circulation Research. Vol. 35. August 1974 graphs were taken in two ways: either the electrograms of the tenth basic cycle and the test cycle were photographed for six to ten replicates or the shutter of the camera was kept open to record the photographic means of basic and test cycles. Records were projected on graph paper (Dagmar Super model 35 microfilm viewer) for measurement of conduction intervals; - time markers were photographed and projected the same way. Results LACK OF A RELATIONSHIP BETWEEN THE FUNCTIONAL REFRAC TORY PERIOD OF THE AV NODE AND THE CONDUCTION TIMES THROUGH THE NODE The appearance of the electrogram obtained with extracellular electrodes placed over the His bundle is well known. We used the conventional notations S, A, H, and V to represent the stimulus artifact, the electrical activity of the atrial tissue adjacent to the His bundle, the electrical activity of the proximal His bundle, and the electrical activity of the base of the interventricular septum near the recording site, respectively (9, 11-13). We took the interval to be the conduction time through the AV node and denoted activity due to the interpolated beat as S', A', and H'. The functional refractory period of the AV node was determined by plotting the relationship of the H-H' interval to the A-A' interval (7). This method excludes effects of atrial conduction time (9). The minimum H-H' interval was used as a measure of the functional refractory period. We confirmed in 11 preparations that the functional refractory period decreased when heart rate was increased (Table 1). Data for a representative experiment are shown in Figure 1A. The functional refractory period decreased in response to a decrease in the interval of the basic drive in 29 of 32 cases. No change was seen in 3 cases, all of which TABLE 1 Relationship of AV Nodal Functional Refractory Period and Conduction Time to Changes in Basic Intervals Basic intervals () Long (800-500) Intermediate (500-320) Short (320-250) TOTAL Functional refractory period 7 15 7 29 Number of observations No change 2 1 0 3 6 4 0 Conduction time No change 3 9 0 12 Decrease Decrease Increase 0 3 7

206 FERRIER, DRESEL H-H' 3 SO 250 200-160 140 120 0 60 40 ISO 200 250 A-A' 350 2O0 500 A-A' FIGURE 1 Lack of correlation between changes in the functional refractory period and the AV nodal conduction times of premature responses. A: Functional refractory period decreased as the heart was driven at progressively shorter basic intervals. Solid circles - 500, squares - 320, and open circles - 250. B: AV nodal conduction times (A'-H') of premature responses for same experiment as in A; symbols are the same as in panel A. Conduction times changed little when the basic interval was decreased from 500 to 320 and increased when the basic interval was decreased from 320 to 250. B occurred at relatively long basic intervals. This finding confirmed the results of Mendez et al. (4). The conduction times of the extrasystoles were determined in all of the experiments. Figure IB shows conduction time data from the same representative experiment considered in Figure 1A. The curve relating the A'-H' interval to the A-A' interval at the shorter basic interval would have shifted to the left or downward if a decrease in conduction time occurred. The experiment illustrated in Figure IB shows the opposite result. The results of these experiments are summarized in Table 1. Conduction time decreased in, remained constant in 12, and increased in observations. In all cases in which conduction time was decreased, the change was small ( <1O ). There appeared to be a pattern in the changes in conduction time. At long basic intervals (800-500 ) conduction time most often decreased. At intermediate driving intervals (500-320 ) conduction time usually did not change, and at short basic intervals (320-250 ) conduction time always increased. These observations clearly indicate that the decreases in the functional refractory period which occurred in virtually all of the experiments were not necessarily accompanied by decreases in the conduction times of extrasystoles. In fact, when a short basic interval (320 ) was further abbreviated, the decrease in the functional refractory period was always accompanied by an increase in conduction time. The lack of correlation between changes in the functional refractory period and changes in conductivity raises the question of what property of AV transmission determines the functional refractory period. Figure 2A shows an example of the conventional curve used in determining the functional refractory period of the AV transmission system. The functional refractory period is indicated as the minimum H-H' interval (in this case 270 ). The diagonal line indicates the continuum along which the H-H' interval is equal to the A-A' interval. This line describes the relationship that would exist between the H-H' interval and the A-A' interval if the extrasystoles were conducted with the same conduction time as the basic cycle. The deviations of the actual curve from this diagonal base line must equal the delay in conduction of the extrasystole compared with the conduction of the basic cycle. When the logarithms of the differences between the actual curve and the diagonal line were plotted as a function of the atrial coupling interval, a straight line resulted over most of the range, indicating that the relationship was exponential within limits (Fig. 2B). A change in slope at very short coupling intervals was observed in approximately 50% of the preparations. A similar derivation can be made from the measured AV nodal conduction time (A'-H'). Figure 2C shows the relationship between the A'-H' interval and the A-A' interval of the extrasystole for this experiment. The conduction time of the basic cycle (A-A = 500 ) is indicated by an X. The horizontal broken line represents a base line corresponding to the diagonal line in Figure 2A. Circulation Research. Vol. 35. August 1974

FUNCTIONAL REFRACTORY PERIOD AND CONDUCTION 207 H-H' 500 150 FRP 50 200 H-H' dev. 200 \ A-A' FRP \ 500 B i A C ms< ec r 200 \ A-A' FRP 500 D 200 5O0 A-A' FIGURE 2 200 500 A-A' Analysis of the relationship between the A V nodal conduction times of premature responses and the curve relating His bundle intervals to atrial intervals. A: Relationship between His bundle intervals and atrial intervals. B: Semilogarithmic plot of the deviations of the H-H' interval from the diagonal in A (H-H' dev.) as a function of the atrial interval. C: A V nodal conduction times for the same experiment. Broken line and X indicate the conduction time of the basic cycle (interval - 500 ). D: Semilogarithmic plot of the differences between the conduction times of the premature responses and the basic cycle (ACT.) as a function of the atrial interval. The functional refractory period (FRP) is indicated in each section. This base line was subtracted from the curve for the extrasystole for all coupling intervals tested. When these differences were plotted logarithmically as a function of the A-A' interval, the resulting exponential function (Fig. 2D) was identical to that derived from the curve relating the H-H' interval to the A-A' interval. Thus, the observed curve relating the H-H' interval to the increases in the A-A' interval (Fig. 2A) is the sum of two functions, the decreasing exponential function (Fig. 2B and D) and the increasing straight line function (Fig. 2A); the slope of the straight line is one. The negative slope of the exponential function changes continuously from steeper than minus one to approximately zero. At some intermediate point, the slope of the exponential function must equal minus one. Therefore, the sum of the slopes of the two functions at this point is zero. Thus, the curve relating changes in the H-H' interval to changes in Circulation Research. Vol. 35, August 1974 \ the A-A' interval must reach a minimum point with a slope of zero. This point is the functional refractory period. The points corresponding to the functional refractory period are indicated in each of the graphs in Figure 2. The functional refractory period represents an intermediate point on a continuous function, i.e., a minimum interval generated because of the exponential nature of the relationship between the coupling interval and the change in conduction time. Although this argument implies that the functional refractory period is not a true refractory period, it suggests that the functional refractory period should be a good index of conductivity since it corresponds to a specific point on the exponential function. This supposition might be true when the effect of a specific variable, e.g., a drug, is tested at constant heart rate, but additional complications can arise. Figure 3 shows the effect of

208 FERRIER, DRESEL H-H' 600 500 ACT. B 200 -i 200 500 200 500 A-A' A-A' FIGURE 3 Effect of changing the basic interval on the exponential function relating the difference between conduction times of premature and basic cycles to atrial coupling intervals (ACT.). A: Decrease in the functional refractory period caused by a decrease in the basic interval. Solid circles = 630, crosses -, and open circles» 250. B: Exponential functions for the experiment in A; symbols are the same as in A. increased heart rate on the functional refractory period and the exponential function. The heart used in this experiment was driven at basic intervals of 630,, and 250. The functional refractory period decreased as the heart rate was increased (Fig. 3A). Figure 3B shows the corresponding exponential functions. The slope of the lines increased as the heart rate was increased. The line deviated toward the abscissa at the shortest basic interval, suggesting that the relationship might no longer be a simple exponential function. The exponential functions in Figures 2 and 3 were derived by using the conduction time of the basic cycle as a reference value. By definition, the difference between the conduction time of the extrasystole and the conduction time of the basic cycle reaches zero when the coupling interval (A-A') equals the basic driving interval (A-A). Therefore, it is obvious that the exponential functions must deviate toward the abscissa as the coupling interval approaches the basic interval. This deviation must occur earlier as the basic interval is shortened, suggesting that the conduction time of the basic cycle might not be the best reference value to use in determining the exponential function. It appeared that a better reference value could be obtained in experiments in which coupling intervals longer than the basic intervals were tested. Therefore, it was necessary to modify the experimental method. A long delay (600-0 ) was introduced between each tenth basic stimulus and the beginning of the next train of ten basic pulses to permit coupling intervals longer than the intervals of the basic drive (e.g., A-A' = 750 when A-A = 320 ) to be tested. An example of such an experiment is shown in Figure 4. Figure 4A shows the relationship between the A'-H' interval and the A-A' interval for coupling intervals as long as 800 tested at basic intervals of 500, 320, and 250. The conduction time at long intervals reached a minimum value at each heart rate. This minimum conduction time became significantly longer as the heart rate was increased. A similar phenomenon was described as "fatigue" by Lewis and Master (8). This fatigue effect was found in each of ten hearts in which very long coupling intervals were tested. The exponential functions determined by subtracting the conduction times of basic cycles from the conduction times of the extrasystoles for each of the three heart rates are illustrated in Figure 4B. As in Figure 3B, the slope of these lines changed as the basic interval was decreased. The AV nodal conduction time of regular beats increases as the basic interval is decreased (Fig. 4C). Each of the conduction times of the basic cycles (plus signs [+ ] in Fig. 4A) fell on the appropriate curve describing the conduction times Circulation Research. Vol. 35. August 1974

FUNCTIONAL REFRACTORY PERIOD AND CONDUCTION 209 iraac 200 A-A' FIGURE 4 200 500 BASIC inntvai Role of increases in conduction time of basic cycles in changing the exponential relationship between ACT and the atrial interval. A: Conduction times of responses initiated at intervals up to 800. Conduction time of the basic cycles is indicated by +. Symbols are the same as in Figure 1. B: Effect of changes in basic interval on the exponential relationship between the change in conduction time and the A-A' interval for same experiment. C: Relationship between the conduction time of basic cycles and the change in basic interval. of the extrasystoles. This coincidence was observed in each of ten hearts. Thus, the same relationship between the conduction time and the atrial interval applies to conduction of both extrasystoles and basic cycles. At the slowest heart rate (basic interval = 500 ) the point describing the basic cycle fell on the asymptote of the curve. At the two higher heart rates this point fell on the ascending limbs of the curves. Thus, different values for the conduction time of the basic cycle had been subtracted from the conduction times of the extrasystoles when each of the curves in Figures 4B and 3B were calculated. This change in the base line caused the changes in the slopes of the exponential functions. Most important, as illustrated in Figure 2, these functions are identical to those that determine the functional refractory period. The plots used to determine the functional Circulation Research. Vol. 35. August 1974 refractory period for the data in Figure 4 are shown in Figure 5. As the basic interval was decreased, the curves shifted downward in parallel. This parallel shift was exactly equal to the change in conduction time of the basic cycle. The functional refractory period was also decreased by an amount equal to the increase in the interval for the basic cycle. In other words, when the basic interval was decreased, the functional refractory period decreased because the AV nodal conduction times of the basic cycles increased. The curves in Figure 5 for basic intervals less than 500 crossed the diagonal line. The points that fell below the line corresponded to coupling intervals longer than the basic intervals (320 or 250 ). The AV nodal conduction times of extrasystoles delivered at these long intervals were shorter than the conduction times of the basic intervals. Therefore, the resulting H-H' intervals were less than the A-A' intervals. The curves crossed the diagonal when the A-A' interval was equal to the basic interval, i.e., they were not coincident with the diagonal line at any other point. Thus, the curves relating the H-H' interval to the A-A' interval at different heart rates did not have a common base line and could not be compared directly. Since the conduction time of the basic cycle is different at each heart rate, the basic cycle does not provide a logical reference from which changes in H-H' 500 200 200 A-A 500 FIGURE 6 Plots to determine the functional refractory period from the data in Figure 4. Symbols are the same as in Figure 1.

2 FERRIER. DRESEL conduction times of the extrasystoles should be measured. The minimum conduction time proved to be a better reference. Although it changed with heart rate (Fig. 4), the minimum conduction time provided a natural division of conduction times into a minimum conduction time plus a slowing due to the prematurity of a response. The minimum conduction time of each of the curves in Figure 4A was subtracted from the measured conduction times, and the logarithms of the differences were plotted as a function of the A-A' interval. Figure 6 shows that this relationship was a simple exponential function at all heart rates and that the exponential functions were superimposable, i.e., the relationship between prematurity and the change in conduction time above the minimum conduction time was the same at all three heart rates. This relationship held true for a large range of basic intervals. Figure 7 shows the points obtained in one heart at basic intervals of 800, 630, 500,, 320, and 275. The results shown in Figures 6 and 7 were confirmed in all ten experiments. FATIGUE EFFECT The fatigue effect was examined by testing a single coupling interval at a series of different heart rates. This procedure allowed a wide range of heart rates to be examined within a few minutes; five hearts were tested. Figure 8 shows the results of one ACT. ACT. I 200 A-A' FIGURE 7 Single exponential function (ACT.) relating the increase in conduction time of premature reponses above the respective minimum conduction times to the atrial coupling interval at basic intervals of 800, 630, 500,, 320, and 275. experiment in which the test stimulus was delivered with a coupling interval of 800 while the basic interval was shortened in steps from 800 to 250. The conduction time increased only slightly (2 ) as the basic interval was decreased from 800 to 500. Further decreases of the basic interval caused progressively larger increments in conduction time. 60 50 40 I 2OO A-A' 500 FIGURE 6 Lack of effect of the basic interval on the exponential function (ACT.) if deviations from the minimal conduction limes (Fig. 4A) are plotted instead of deviations from the conduction times of basic cycles {Fig. 4B). Symbols are the same as in Figure 1. 30 I I I 200 600 800 Basic Interval FIGURE 8 Effect of decreasing the basic interval on the conduction time of a test response initiated at an atrial coupling interval of 800, i.e., the fatigue phenomenon. Circulation Rneaixh. Vol. 35. August 1974

FUNCTIONAL REFRACTORY PERIOD AND CONDUCTION 211 EFFECT OF DROMOTROPIC AGENTS ON THE CONDUCTIVITY OF THE AV NODE AV nodal conduction time can be divided into three components. The first component is "basal conduction time," which is the minimum conduction time determined at very slow heart rates (e.g., 800- intervals) at which fatigue is negligible. The second component is the increment due to fatigue. The third component is represented by the exponential function that determines the increase in conduction time which will occur when the coupling interval is shortened (interval-related conductivity). Interval-related conductivity, fatigue factor, and basal conduction time together determine the conduction times of both basic cycles and extrasystoles, and study of the effects of dromotropic agents on each component might help to characterize their types of actions. Therefore, we studied the effects of infusions of epinephrine on the conductivity of the AV node in four hearts. Figure 9A shows the effect of epinephrine (0.02 /ig/ml blood) on conduction times through the AV node for one experiment. The basic interval was 250. Conduction time was decreased at each coupling interval tested. The conduction time of the basic cycle was decreased more (22 ) than was the minimum conduction time (13 ). Figure 9B shows that epinephrine caused a parallel shift of the exponential curve; this effect of epinephrine was clearly different from the effect of changes in heart rate. The change in the minimum conduction time does not differentiate between a change in the basal conduction time and a decrease in the fatigue effect. The fatigue effect was studied in three hearts, and a representative result is shown in Figure. This preparation showed some automaticity; therefore, the maximum basic cycle lengths that could be used were 630 for the control period and 500 during infusion of the drug. The maximum coupling interval that could be used for the extrasystole during infusion of epinephrine was 560. The control basal conduction time estimated by the asymptote of curve 1 was no less than 55. The basal conduction time during infusion of epinephrine estimated by the asymptote of curve 2 was not greater than 33. The increment in conduction time of the late (560 ) extrasystole for each decrease in the basic interval was slightly less during the infusion of epinephrine, i.e., epinephrine decreased the fatigue effect. This process is more easily seen if the two curves are compared from a common base line (curve 3 in Fig. ). The results of the remaining Circulation Research, Vol. 35. August 1974 160 120 roo to 60 ACT. r V A-A 30O msac 0 N r - I 1 200 A-A' FIGURE 9 A: Effect of epinephrine on the AV conduction times of premature and late cycles. Solid circles = control and open circles - epinephrine {0.02 iiglml blood). Conduction time of a basic cycle is indicated by +. B: Effect of epinephrine on the exponential function determined from the minimum conduction time. B two experiments agreed closely with those illustrated in Figure. Thus, the minimum conduction time was decreased by epinephrine by a reduction in both the basal conduction time and the fatigue effect. Discussion We have identified three variables that determine AV conduction time. Basal conduction time, the shortest AV conduction time, was demonstrated by responses initiated 500-0 following trains of regular beats at low frequencies (basic interval 500-0 ). Increasing the heart rate caused slowing of conduction of all beats, independent of the interval between them, so that even late responses were slowed and the

212 FERRIER, DRESEL 80 70 60 50 40 30 200 600 Basic Interval FIGURE Effect of epinephnne on the fatigue phenomenon and basal conduction time. Conduction times of a late response (A-A' - 560 ) at different basic intervals during a control period (curve 1) and during the infusion of 0.01 ng epinephrine/ml blood (curve 2) are shown. Curve 3 is identical to curve 2 except it is shifted so that the point at A -A - 500 is superimposed on the corresponding point in curve 1. minimum conduction time was longer than the basal conduction time. This fatigue effect has been previously reported by Lewis and Master (8). Progressive shortening of the coupling interval of a test response resulted in an exponential increase in AV conduction time (ACT) above the minimum conduction time for a given rate. This relationship between interval and conduction time (interval-related conductivity) was quantitatively the same at all heart rates in a given preparation. The sum of all three components determined the conduction time of all beats whether they were regular, premature, or late beats. Figure 11 schematically shows the interplay between these factors which resulted in the measured conduction times and H-H' intervals. In Figure 11 (top), the conduction time of an extrasystole inserted at 600 was increased from 40 to 50 at the fast rate due to the fatigue effect. Nevertheless, there was a shorter H-H' interval at the fast rate because of ACT of the regular beat. Figure 11 (bottom) shows the conduction of the beats that established, when the A-A' interval was plotted against the H-H' interval, the functional refractory period of the AV node. The values of ACT of the extrasystoles were the same (Fig. 6), but the functional refractory period was shortened because of the differences in ACT of the regular beats (see Figs. 4 and 5). Neither the basal conduction time nor its rate-dependent lengthening to a new minimum conduction time, i.e., fatigue, contributed to the functional refractory period or to any other H-H' interval, because these two factors affected both the regular and the premature beats equally. Thus, a drug that affects only the minimum conduction time would not change the functional refractory period. ACT could be represented by a single exponential function in 50% of our experiments. In the remaining 50%, there was a change in slope at very short coupling intervals. This finding might be explained by the observations of Watanabe and Dreifus (14) who showed that propagation of action potentials through the AV node followed a tortuous route when AV conduction times were prolonged by rapid heart rates. Therefore, the length of the pathway through the AV node might be increased when failure of conduction is approached. The cellular basis of conduction through the AV node has been described extensively in recent years (, 14, 15). The description of conduction in the present study says nothing about these cellular factors. Thus, the measurements we made with extracellular electrodes were analogous to the straight lines in Figure 11 that represent conduction through the node. Both techniques are useful in describing the node as an input-output system, but neither considers changes in path,, conduction times, or refractoriness which might occur within the elements of the system. With this reservation we feel justified in discussing the relation of the functional refractory period to conduction and refractoriness. It has been generally assumed that the functional refractory period is a function of conduction through the AV node of the extrasystoles with which the functional refractory period is determined. In the present study we demonstrated that the conduction times of all beats were increased when heart rate was increased, whereas the functional refractory period was changed in the opposite direction. The cause of the decrease in the functional refractory period Circulation Research, Vol. 35, August 1974

FUNCTIONAL REFRACTORY PERIOD AND CONDUCTION 213 Basic Interval 500 600 Basic Interval 250 600 H-H»600 H-H'-580 I I 40-180^ 40-BASAL-40 0- FATIGUE -0 0-ACT-0 I I 40 II 70 40-BASAL-40 -FATIGUE- 20-ACT-O II 50 I I I I 40 40-BASAL-40 0-FATIGUE-0 O-ACT-60 msec I I I 70 1 40-BASAL-40 -FATIGUE - 20-ACT-60 FIQURE 11 Analysis of transmission through the A V node of a late extrasystole (top) and an extrasystole that establishes the functional refractory-period (bottom). The last driven beats are shown. The conduction of each impulse is broken down into the three factors determining it immediately below each section of the diagram. was traced to the characteristics of conduction of the regular beats. The fatigue effect, which slows conduction through the node, does not affect the functional refractory period. These considerations indicate that statements about the action of physiological or pharmacological influences on conduction through the AV node that are based simply on changes in the functional refractory period must be interpreted with some caution. The functional refractory period has never been a classical refractory period of the system; extrasystoles entering the system before the functional refractory period are in fact conducted through the node. The only dromotropic agent which we studied in detail, epinephrine, caused major changes in all three components of conduction. It shortened the basal conduction time, decreased fatigue, and caused a parallel shift of the ACT exponential relationship. When the effects of epinephrine on the functional refractory period were plotted, the drug not only decreased the functional refractory period but also decreased the coupling interval (A-A') of the extrasystole which determined the functional refractory period. Therefore, it appeared that this shift in the A-A' interval could be used to measure Circulation Retearch. Vol. 35. August 1974 a parallel shift in the ACT curve. However, preliminary results indicate that some manipulations and some agents not only shift the exponential function but also change its curvature, i.e., change the slope of the semilogarithmic plot. Therefore, the functional refractory period is a questionable index of AV nodal conductivity. However, the functional refractory period of the AV node remains an important parameter of cardiac function. By definition the functional refractory period of the AV node is the minimum ventricular interval that can be propagated from the atria and, therefore, is one of the determinants of the ventricular response to rapid atrial rates. It is sometimes desirable to determine how a drug might affect the frequency of ventricular activation in, for example, supraventricular tachycardias or atrial flutter. In these situations the functional refractory period is obviously an important index of AV nodal function. In addition, our results do not discredit the functional refractory period as a measure of refractoriness in atrial or ventricular tissues in which the conduction time of basic cycles changes little over a wide range of physiological frequencies. The relationship between conduction time and

214 FERRIER, DRESEL conductivity of the AV node which emerges from our study is somewhat analogous to the relationship of contractile force to contractility. The latter has become identified with the entire relationship of contractile force or velocity to independent variables such as length, resting tension, etc. We consider conductivity to be definable only by a series of measurements of conduction times of both extrasystoles and basic cycles at various basal rates. Without this technique, it is not possible to determine which of the three determinants of conduction time have been affected by the experimental parameter under investigation. Perhaps this more detailed determination of changes in conductivity will lead to a better understanding of the process of AV nodal conduction. References 1. KRAYER O, MANDOKI JJ, MENDEZ C: Studies on veratrum alkaloids: XVI. Action of epinephrine and of veratramine on the functional refractory period of the auriculo-ventricular transmission in the heart-lung preparation of the dog. J Pharmacol Exp Ther 3:412-419, 1951 2. PRESTON JB, MCFADDEN S, MOEGK: Atrioventricular transmission in young mammals. Am J Physiol 197:236-240, 1959 3. MOE GK, PRESTON JB, BURLINGTON H: Physiologic evidence for a dual A-V transmission system. Circ Res 4:257-275, 1956 4. MENDEZ C, GRUHZIT CC, MOE GK: Influence of cycle length upon refractory period of auricles, ventricles, and A-V node in the dog. Am J Physiol 184:287-295, 1956 5. ROSENBLUETH A: Functional refractory period of cardiac tissues. Am J Physiol 194:171-183, 1958 6. KOHU JD, TUTTLE RR, DRESEL PE, INNES IR: Influence of anesthetics and of arterial blood pressure on the functional refractory period of atrioventricular conduction. J Pharmacol Exp Ther 153:505-5, 1966 7. HAN J, MOE GK: Cumulative effects of cycle length on refractory periods of cardiac tissues. Am J Physiol 217:6-9, 1969 8. LEWIS T, MASTER AM: Observations upon conduction in the mammalian heart: A-V conduction. Heart 12:209-258, 1925 9. FERRIER GR, DRESEL PE: Role of the atrium in determining the functional and effective refractory periods and the conductivity of the atrioventricular transmission system. Circ Res 33:375-385, 1973. MERIDETH J, MENDEZ C, MUELLER WJ, MOE GK: Electrical excitability of atrioventricular nodal cells. Circ Res 23:69-85, 1968 11. SASYNIUK BI, DRESEL PE: Effect of diphenylhydantoin on conduction in isolated, blood-perfused dog hearts. J Pharmacol Exp Ther 161:191-196, 1968 12. ALANIS J, GONZALEZ H, LOPEZ E: Electrical activity of the bundle of His. J Physiol (Lond) 142:127-140, 1958 13. KIRK BW, DRESEL PE: Effects of amodiaquin and quinidine on cardiac conduction. Can J Physiol Pharmacol 43:29-38, 1965 14. WATANABE Y, DREIFUS LS: Inhomogeneous conduction in the A-V node, a model for re-entry. Am Heart J 70:505-514, 1965 15. MENDEZ C, MOE GK: Some characteristics of transmembrane potentials of A-V nodal cells during propagation of premature beats. Circ Res 19:993-, 1966 Circulation Research. Vol. 35. August 1974