MEASUREMENT OF SINO-ATRIAL CONDUCTION TIME BY PROGRAMMED ELECTROSTIMULATION

Measurement of Sinoatrial Conduction Time by Premature AtrialStimulation in the Rabbit

By Hugh C. Miller and Harold C. Strauss

ABSTRACTThe premature atrial stimulation technique was investigated as a method of

measuring sinoatrial conduction time in the rabbit. Fifteen studies wereperformed in which intracellular recordings were obtained from a sinus nodecell and atrial electrical activity was recorded from the crista terminalis by asurface electrogram. An additional ten studies were performed without micro-electrode recordings. Atrial premature depolarizations late in the cardiac cycleproduced compensatory atrial return cycles, but earlier premature depolariza-tions produced less than compensatory return cycles. Compensatory returncycles only occurred with atrial premature depolarizations that failed tocapture the sinus node cell. The transition from compensatory to less thancompensatory return cycles occurred with late atrial premature depolariza-tions that failed to capture the sinus node. Therefore, the transition fromcompensatory to less than compensatory atrial return cycles failed to indicatesinus node capture by an atrial premature depolarization. Although thesepremature depolarizations were too late to capture the sinus node cell, theystill shortened the sinus node and atrial return cycles to make the atrial returncycle less than compensatory. This shortening of the sinus node return cyclewas due to a shortening of the sinus node action potential by electrotonicinteraction between sinus node and adjacent cells during repolarization. Theelectrotonic effect resulted in shortening of the sinus node action potential andaccounted for the poor correlation (?• = 0.64) between estimated and measuredvalues of sinoatrial conduction time. These data indicate that there aresignificant limitations to the use of the premature atrial stimulation techniquefor estimating sinoatrial conduction time.

KEY WORDSelectrotonic effectscrista terminalis

sinus node action potential shorteningreturn cycle repolarization

compensatory return cycle

• A satisfactory technique for measuringconduction time between the sinus node andthe atrium would represent a considerableadvance in our ability to evaluate sinus nodefunction. It has been suggested that sino-atrial conduction time can be derived byassessing the atrial response to atrial prema-ture depolarizations, but this procedure is

From the Department of Medicine, Duke UniversityMedical Center, Durham, North Carolina 27710.

This investigation was supported in part by U. S.Public Health Service Grants HL-08845, HL-05736, andHL-15190 from the National Heart and Lung Institute,the Walker P. Inman Fund, and a grant-in-aid from theNorth Carolina Heart Association. Dr. Miller is therecipient of a Fulbright Scholarship and was supportedby grants from G. D. Searle & Company, Ltd., andWinthrop Laboratories.

This study was presented in part at the Federation ofAmerican Society of Experimental Biology Meetings,April, 1974, Atlantic City, New Jersey.

Please address reprint requests to H. C. Strauss, M.D.,Box 3845, Duke University Medical Center, Durham,North Carolina 27710.

Received May 1, 1974. Accepted for publication July31, 1974.

Circulation Research, VoL .15, December 1971,

still unsubstantiated (1) even though severalstudies have analyzed the atrial response toatrial premature depolarizations (1-19). Anatrial premature depolarization elicited latein the atrial cycle is followed by a returncycle that is compensatory. Progressivelyearlier premature depolarizations are fol-lowed by return cycles that lengthen propor-tionally to remain compensatory. At somecritical coupling interval, an atrial prema-ture depolarization is followed by a returncycle that is just less than compensatory,and thereafter progressively earlier prema-ture depolarizations are followed by lessthan compensatory return cycles. It hasbeen postulated that late atrial prematuredepolarizations are followed by compensa-tory return cycles because these depolariza-tions fail to capture the sinus node (1, 2, 7);somewhere between the site of stimulationand the sinus node they collide with theemerging sinus node impulse and areblocked. It has similarly been postulated

935

by guest on October 31, 2015http://circres.ahajournals.org/Downloaded from

http://circres.ahajournals.org/

936 MILLER, STRAUSS

that the atrial premature depolarizationsthat result in the transition from compensa-tory to less than compensatory return cyclescapture the sinus node and "reset" the sinuspacemaker. If it is assumed, as Wencke-bach (2) did, that a premature beat capturingthe sinus node resets it in the same way thata spontaneous discharge of the node itselfdoes, then the duration of a noncompensa-tory atrial return cycle equals the basic cyclelength plus the time necessary for retro-grade conduction of the atrial prematuredepolarization from its site of origin to thesinus node and for antegrade conductionfrom the sinus node to the atrium. If "sino-atrial conduction time" is considered to com-prise the sum of retrograde and antegradeconduction times, then the sinoatrial conduc-tion time equals the noncompensatory re-turn cycle duration minus the basic cyclelength.

Analysis of the atrial response to atrialpremature depolarizations has been used toevaluate human sinus node function (1, 7, 12-14, 17), to describe sinoatrial entrance blockwith late premature depolarizations (1), andto delineate the effects of drugs on sinoatrialconduction time (17), even though no experi-mental verification of the technique is avail-

able. It was therefore the purpose of thisstudy to investigate the atrial response toatrial premature depolarizations in the iso-lated rabbit heart and to examine the rela-tion between the atrial response and thesinoatrial conduction time. Changes in con-duction time and changes in automaticitycan occur with early atrial premature depo-larizations, and these changes can affectreturn cycle duration (5, 10, 11, 16, 19).Therefore, the response to the latest atrialpremature depolarization consistently givingnoncompensatory return cycles was used inthe derivation of sinoatrial conduction timein the present study (Fig. 1). The resultsshowed that in this experimental model theatrial response to atrial premature depolari-zations did not accurately reflect sinoatrialconduction time due to previously unsus-pected effects of the premature depolariza-tions; atrial premature depolarizationscaused shortening of sinus node action poten-tials and, consequently, shortening of thesinus node return cycle duration.

MethodsStudies were performed on 25 rabbits (1.5-3 kg).

They were anesthetized with sodium pentobarbi-tal (100-150 mg/kg, ip), and their hearts were thenrapidly removed and dissected in cool, modified

FK3URE 1

Derivation of sinoatrial conduction time from the atrial response to atrial prematuredepolarizations. A-A = basic cycle length, Ay = atrial premature depolarization, AH =atrial return cycle response, A-A,, = atrial premature cycle, Ap-AR = atrial return cycle,and SAN = sinoatrial node. Compensatory line is the plot of (A-Ar)l(A-A) versus (A,.-An)l(A-A) tchen (A-A,-) + (Ar-AH) = 2(A-A). Following Ar a less than compensatory atrialresponse falls below this line and a greater than comjyensatory response falls above theline. Ap® marks the transition betiveen atrial premature depolarizations giving compensa-tory (Ap®) and less than compensatory (AP <S>) responses. Sinoatrial conduction time(between atrium and SAX + SAN and atrium) can be derived from Ar® as the differencebetween Ar^AR and A^A. See text for further details.

Circulation Research, VoL 35, December 1974



SINOATRIAL CONDUCTION 937

Tyrode's solution. The right atrium, including thesinus node but excluding the atrioventricularnode, was dissected free and pinned to the waxbottom of a Lucite tissue chamber with the endo-cardial surface uppermost (20). In some prepara-tions with ectopic rhythms, the atrial appendagewas trimmed.

The modified Tyrode's solution had the follow-ing millimolar composition: NaCl 130.0, KC1 4.0,NaH,PO4 1.8, CaCl, 2.7, MgCL 0.5, dextrose 5.5,and NaHCOn 18.0 in deionized, distilled water. Thesolution was bubbled with a 95% O2-5% CO, gasmixture in both the reservoir bottles and thetissue bath to establish a pH of 7.40. The bath wasmaintained at 36.0 ± 0.5°C and perfused withTyrode's solution at approximately 10 ml/min.

Transmembrane potentials were recordedthrough glass microelectrodes filled with 3M KC1and having tip resistances of 18-35 Mohms. Themicroelectrodes were connected by silver-silverchloride wires to a high-input impedance, capaci-tance-neutralizing amplifer,1 and the transmem-brane potentials were displayed on a Tektronixtype 565 dual-beam oscilloscope. Calibration wasperformed by applying a 50-mv signal in serieswith the ground of the tissue bath.

A bipolar surface electrogram was recordedfrom the crista terminalis, usually at its upperend, through silver wires insulated except at thetip. This signal was amplified by a Tektronix 26A2differential amplifier and displayed on the oscillo-scope screen.

The preparations beat spontaneously, and afterevery eighth to tenth beat a premature depolari-zation at a predetermined coupling interval wasevoked by stimulating the atrium through bipolarsilver electrodes placed immediately adjacent tothe surface electrogram electrodes. Stimuli wererectangular constant-voltage pulses 4 msec in du-ration and less than twice diastolic threshold volt-age in intensity. The atrial premature depolariza-tions were evoked late in the atrial cycle and thenat progressively earlier intervals, usually in 5-10-msec decrements.

The action potential, the atrial electrogram, andthe signal from a 50-msec time-mark generator(Tektronix model 2901) were recorded on magnetictape at 7.5 inches/sec. Records for analysis weresubsequently obtained by playback to an ElemaMingograf 800 recorder (frequency response 0-750Hz) at a paper speed of 200 mm/sec or by Polaroidphotography of a storage oscilloscope display(Tektronix model Dll). During playback, intervalsbetween depolarizations on the atrial electrogramwere measured using a Hewlett-Packard model5304A interval counter coupled to a voltage-trig-gered time circuit2 and a Hewlett-Packard model5055A digital recorder. This system was accuratewithin ±1 msec.

Before commencing the studies, preparations

1 Designed by Dr. William New, Jr., Stanford Univer-sity School of Medicine, Palo Alto, California.

1 Designed by Jack Kasell, Duke University MedicalCenter, Durham, North Carolina.Circulation Research, VoL .15, December 1971

were allowed to stabilize for at least 1 hour follow-ing dissection until the basic cycle length wassteady. Action potentials were only accepted assinus node recordings if they were at least 50 mvin total amplitude and had a slow phase 0 depolar-ization which developed smoothly from a sponta-neous phase 4 depolarization. The action potentialalways preceded the atrial electrogram by at least25 msec and during any one experiment was al-ways recorded from the same cell. In 15 experi-ments, action potentials meeting these require-ments were obtained for a sufficient period of timeto allow scanning of the diastolic period withatrial premature depolarizations. In many experi-ments, it was not possible to locate cells withsinus node action potentials meeting these rigidcriteria in spite of systematic search of the endo-cardial surface; these studies were thereforeabandoned. In an additional 10 experiments, pre-mature stimulation was performed in the absenceof microelectrode recordings.

Time intervals measured from the sinus nodeaction potential were taken from the junction oftangents drawn to the steepest portion of phase 0and the most prominent part of phase 4 depolari-zation. Capture of the sinus node by a prematurebeat occurred when the basic sinus cycle meas-ured in this way was shortened.

DEFlNtnONB

A premature cycle is a cycle ending in an atrialpremature beat, and a return cycle is a cyclefollowing an atrial premature beat. The sinusnode premature and return cycles are those corre-sponding to the relevant atrial cycles.

Results

Only the responses to late atrial prema-ture depolarizations occurring in the last50% of the atrial cycle were analyzed. Meancycle length of the preparations was 533msec (400-650 msec) for the 15 experimentswith microelectrode recordings (Table 1) and509 msec (400-650 msec) for the 10 experi-ments without microelectrode recordings. Inall experiments the durations of the prema-ture and the return cycles were normalizedusing the duration of the last spontaneouscycle as the denominator (5). The normalizedreturn cycle was then plotted as a function ofthe normalized premature cycle using thesame format as that shown in Figure 1.Normalization allowed the ready identifica-tion of the atrial response to atrial prema-ture depolarizations as compensatory or non-compensatory, permitted comparison of re-sponses in spite of changes in basic cyclelength, and facilitated comparison of datafrom different experiments.



938 MILLER, STRAUSS

TABLE 1

Results of 15 Experiments in which Microelectrode Recordings Were Made

Expt.

123456789

101112131415

MEAN

Cyclelength

600420525575620635400500550550465620520530470

533

Conductiontime: crista

to SAN

184512306525786336183016405030

37

Conductiontime: SANto crista

342549759043627978353925656344

54

Measuredconduction

time

527061

10515568

140142114

536941

105113

74

91

Estimatedconduction

time

345026296226

113743520621634

12031

49

Actionpotential

shortening

192223485422

63232312325352715

27

SAN = sinoatrial node.A TRIAL RETURN CYCLE DURATION

The results of two experiments which illus-trate the range and the variability of re-sponse to atrial premature depolarizationsare shown in Figure 2. In experiment A, theintroduction of premature depolarizationslater than 87% of basic cycle length resulted

in compensatory return cycles. An atrial pre-mature depolarization at 87% of basic cyclelength marked the transition between com-pensatory and noncompensatory return cy-cles. Earlier premature depolarizations thencaused progressively longer return cycles,but they remained less than compensatory.

140

130

1 2 0 -

110 —

KX)I

Compensatory Line

CL 500 msec A •

CL 550 msec. B o o o

O 2 0 3 0 40

A-Ap/A-A(%)FIGURE 2

Atrial response to atrial premature depolarizations in tico experiments. Abbreviations arethe same as in Figure 1. Cycle length (CL) in experiment A and experiment B are shown.In experiment A, less than compensatory return cycles occurred ivith premature depolari-zations less than 87% of basic cycle length and sinus node capture occurred ivithpremature depolarizations earlier than 70% of basic cycle length (vertical line). Forexperiment B, corresponding values ivere 95% and 8T7c, respectively. In experiment A, AP-AH increased with earlier atrial premature depolarizations (typical of 20 experiments), buti7i experiment B this value remained constant (typical of 5 experiments).

Circulation Research, VoL 35, December lS?i




Curves with a similar configuration wereobtained in 20 experiments.

In experiment B (Fig. 2) the plot of thenormalized return cycle versus the normal-ized premature cycle showed a plateau overwhich the return cycle duration was vir-tually constant with increasing prematurityof the atrial premature depolarizations. Aplateau-shaped curve was obtained in 5 ex-periments. The position of the curve for all25 experiments was usually between A andB.

The transition from compensatory to non-compensatory return cycles with progres-sively earlier atrial premature depolariza-tions was similar for curves with or withouta plateau and varied between 72% and 97%of basic cycle length. However, this valuevaried with the site of stimulation. Stimula-tion further from the sinus node on the atrialappendage caused a shift of the curve up-ward, and noncompensatory return cyclesfirst occurred with earlier premature depo-larizations compared with results obtainedwhen the crista terminalis was stimulated.

SINUS NODE RESPONSE TO ATRIAL PflEMATURE DEPOLARIZATIONS

Microelectrode recordings from experi-ment A in Figure 2 showed that atrial pre-mature depolarizations earlier than 70% ofbasic cycle length captured the sinus nodebut that later premature depolarizationsblocked somewhere between their point oforigin on the crista terminalis and the re-cording site in the sinus node. Althoughatrial premature depolarizations between70% and 87% of the basic cycle length failedto capture the sinus node, they caused non-compensatory return cycles, demonstratingthat the occurrence of noncompensatory re-turn cycles is not a reliable indicator of sinusnode capture. On the other hand, compensa-tory responses only occurred with prematuredepolarizations that failed to capture thesinus node. Similar observations were madein 13 other experiments both in the presenceand the absence of plateau-shaped curves.Sinus node capture first occurred with atrialpremature depolarizations between 53% and92% of basic cycle length in these experi-ments in contrast to the transition from com-pensatory to noncompensatory return cycleswhich occurred with later premature depo-larizations between 72% and 97% of basic

Circulation Research, Vol. 35, December 1971,

cycle length. In one experiment, sinus nodecapture coincided with the appearance ofnoncompensatory return cycles.

In experiment A (Fig. 2), atrial prematuredepolarizations later than 87% caused com-pensatory return cycles, those between 70%and 87% caused noncompensatory return cy-cles without sinus node capture, and thoseearlier than 70% caused sinus node captureand noncompensatory return cycles. Thesinus node response to atrial premature de-polarizations in these different ranges isshown in Figure 3. In Figure 3A, an atrialpremature depolarization at 90% of basiccycle length had no effect on the sinus nodeaction potential, and a compensatory returncycle occurred. In B, a premature depolariza-tion at 80% of basic cycle length failed tocapture the sinus node but caused early andmore rapid phase 3 repolarization of thesinus node action potential, followed by theearly onset of phase 4 depolarization, ashortened sinus node return cycle, and anearly noncompensatory atrial response. Theunchanged slope of phase 4 depolarization isnoteworthy. In C, an even earlier prematuredepolarization at 75% also failed to capturethe sinus node; the same phenomena seen inB were present to a more marked degree. InD, an atrial premature depolarization at 60%of basic cycle length captured the sinus node,reset the node, and resulted in a noncompen-satory atrial response.

Figure 3 shows that, for atrial prematuredepolarizations that failed to capture thesinus node, shortening of the sinus nodeaction potential approximately equaledshortening of the sinus node return cycleand accounted for the amounts by which theatrial return cycles were less than compen-satory. Similar results were found in 13other experiments (Fig. 4). The maximumshortening seen in 14 experiments was 54msec, and the mean maximum shorteningwas 27 msec. In the one experiment in whichsinus node capture coincided with the ap-pearance of noncompensatory atrial returncycles, shortening of action potentials oc-curred but there was no change in returncycle duration (see Discussion). The maxi-mum amount of shortening of the sinus nodeaction potential in milliseconds was not re-lated to the basic cycle length (Table 1).

Following sinus node capture by earlieratrial premature depolarizations, several



940 MILLER, STRAUSS

B

Ap"A-H-

Q , 1 I I I I I > 1

50 mV

0.5 sec.FK1UHE3

Sinus node response to atrial premature depolarizations. Abbreviations are the same as inFigure 1. In A-D, a sinus node action potential is shown in the top tracing and the cristaterminalis electrogram in the bottom tracing. Each trace comprises several oscilloscopesweeps which were guperimposed by triggering the sweep at the same voltage level. A:Compensatory responses follow two late atrial premature depolarizations which fail tocapture the sinus node. B and C: Noncompensatory atrial resjwnse follows a prematuredepolarization which failed to capture the sinus node. D: Noncompensatory atrial responsesfollow premature depolarizations which did capture the sinus node. See text for furtlierdetails.

patterns of response of the sinus node returncycle occurred. In four experiments, thesinus node return cycle, which had beenshortened by late premature depolarizationsthat failed to capture the sinus node, re-mained short with earlier atrial prematuredepolarizations that did capture the sinusnode (Fig. 5A). In eight experiments, thesinus node return cycle rapidly lengthenedwith atrial premature depolarizations thatcaptured the sinus node and returned towithin 2% of the control value (Fig. 5B). Intwo experiments, depression of sinus nodeautomaticity occurred with early prematuredepolarizations, and the sinus node returncycle lengthened to 106% and 114% of con-trol, respectively (Fig. 5C).

There was no evidence that shortening ofthe sinus node action potential was mediatedby humoral factors. Stimulation at ten timesthreshold voltage did not affect the returncycle response to premature beats. Also stim-uli during the atrial refractory period did not

50

40-

s.30-

•S 2 C -

10-

Yr

n

.

= 0.83X= 0.85= 99

• •:'A2 *

+ 0.49 /

/ m

•

•

20 30 40 50 60

Storttratj of Action Potential (msec.)

F IGURE 4

Results of lr> exjyeriments with plot of the shortening ofsinus node action potential duration (measured at 909cof amplitude) and shortening of sinus node return cyclefollowing atrial premature depolarizations that failed tocupture the sinus node. There loas a tendency for actionpotential shortening to be more marked than shorteningof cycle length. The line of identity is shoicn.

Circulation Research. Vol. 35, December 197i




I 3 0 I -

I MO

I1 00

<XHI-50 60 70 80 90 DO

A-Ap/A-A(%)

l30r

90H/50 60 70 80 90

A-AP/A-A(%)

00

FIGURE 5

50 60 70 80 90 100

A-AP/A-A(%)

Plot of the normalized sinus node return cycle (open circles) and the normalized atrialpremature cycle (solid circles) in three experiments shoiving three different types ofresponse, A, B, and C. The compensatory lines are shown. In all three examples, sinus nodereturn cycle shortening occurred with late atrial premature depolarizations that failed tocapture the sinus node, but with earlier premature depolarizations the sinus node returncycle remained short in A, returned to control values in B, and increased above controlvalues in C, indicating some depression of sinus node automaticity.

affect the basic rhythm of the preparation.In the presence of 3.4 x lO-'M propranolol or 2x 1(HM atropine, comparable shortening ofthe sinus node action potential occurred andsimilarly accounted for shortening of theatrial return cycle.

ELECTROTONIC INTERACTION

Demonstration of shortening of the sinusnode action potential by electrotonic interac-tion between the sinus node and adjacentcells is shown in Figure 6. In this experimentsimultaneous microelectrode recordingswere made from a sinus node cell meeting allof the criteria for this study (Methods) andfrom a cell 0.3 mm closer to the crista termi-nalis. The figure shows the smooth onset ofphase 0 depolarization of the sinus node ac-tion potential which preceded the action po-tential of the adjacent cell and preceded theatrial electrogram by 32 msec. The actionCirculation Research, Vol. -J-5, December I97i

potential of the peripheral cell had spontane-ous phase 4 depolarization but abrupt onsetof phase 0 depolarization typical of a latentpacemaker. An atrial premature depolariza-tion at a coupling interval of 442 msec cap-tured the peripheral cell but failed to pene-trate and capture the more central sinus nodecell as shown by its undisturbed basic sinuscycle length. The action potential of theperipheral cell on the other hand had a morerapid upstroke and an early repolarizationfollowing capture by the premature depolari-zation. Although the central sinus node cellwas not captured, shortening of the sinusnode action potential accompanied shorten-ing of the action potential of the peripheralcell. The shortening of the two action poten-tials ran a parallel course in keeping withelectrotonic interaction between the cells.The shortening of the sinus node action po-tential was followed by early spontaneous



942 MILLER. STRAUSS

484

200 msec.'

50 msec.

FIGURE 6

Simultaneous action potential recordings from a sinus node cell and a latent pacemaker0.3 mm closer to the crista terminalis. Top: Action potential and crista terminaliselectrogram responses to an atrial premature depolarization at 1,1,2 msec, shoiving thevalues of the sinus node (ujri)er values) and atrial cycles (lower values). Bottom Left:Sinus node action potential before (broken line) and after (solid line) the prematuredepolarization. Bottom Right: Action potential of the latent pacemaker before (dotted line)and after (solid line) the atrial premature depolarization. The crista terminalis electro-gram common to both action potentials is shown in both bottom sections. See text forfurther details.

phase 4 and phase 0 depolarizations andan early atrial response like that seen inFigure 3.CONDUCTION TIME: CRISTA TERMINALIS TO SINUS NODE

Retrograde conduction time from thecrista terminalis to the sinus node rangedfrom 12 msec to 78 msec (mean 37 msec) forthe latest atrial premature depolarizationthat just captured the sinus node (Table 1).In 13 of 15 experiments, this value was lessthan the value for antegrade conductiontime from the sinus node to the crista termi-nalis. With progressively earlier prematuredepolarizations the retrograde conductiontime from the crista terminalis to the sinusnode increased strikingly, the increase beingmore marked in experiments with longerconduction times (Fig. 7, solid circles) andless marked in experiments with shorter con-duction times (Fig. 7, open circles).

CONDUCTION TIME: SINUS NODE TO CRISTA TERMINALIS

The antegrade conduction time from thesinus node to the crista terminalis rangedfrom 25 msec to 90 msec (mean 54 msec)(Table 1). In seven experiments atrial prema-ture depolarizations that captured the sinus

90,

80-

i 60-COo- 50-

s -.5 30-

I 20-10-

CL 550 msec.

CL 550 msec.

_L J_ _L I

o°

I _L0 0 20 30 40 50 60 70 80 90

A-AP/A-A(%)FIGURE 7

Plot of retrograde conduction time from the crista termi-nalis to the si nun node versus premature atrial cycleduration. One experiment in which conduction time teaslonger than average and one in which it was shorter thanaverage are shoivn. The retrograde conductioti time in-creased icith increasing prematurity of the atrial prema-ture depolarization, but the increase teas more marked inexperiments with longer conduction times. Abbreviationsare the same as in Figures 1 and 2.

Circulation Research. Vol. J5, December 1974




node caused shortening of antegrade conduc-tion time, and in one experiment shorteningoccurred with premature depolarizationsthat were not early enough to capture thesinus node. There was no other evidence ofpacemaker shift. Shortening of antegradeconduction time was only detected in experi-ments with conduction times greater than 45msec, and in these experiments shorteningbecame more marked with early prematuredepolarizations (Fig. 8).ESTIMATION OF SINOATRIAL CONDUCTION TIME

Estimation of sinoatrial conduction timefrom the latest atrial premature depolariza-tion giving a noncompensatory return cycle(Fig. 1) consistently underestimated the truesinoatrial conduction time. The consistentunderestimation reflected the fact that non-compensatory return cycles occurred withatrial premature depolarizations that failedto capture the sinus node. In the one experi-ment in which this phenomenon did not oc-cur, the estimated and measured values forsinoatrial conduction time agreed closely.

IOOI—

9 0 -

8 0 -

S 70

40-

CL 550 msec.

3 0 - CL 550 msec

20-

10

I I I I I0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 0 0

FIGURE 8

Plot of antegrade conduction time from the sinus node tothe crista terminalis versus premature atrial cycle dura-tion in one experiment in which antegrade conductiontime teas longer than average and one in which it wasshorter than average. Abbreviations are the same as inFigures I and 2. The vertical lines indicate the latestatrial premature depolarizations that captured the sinusnode. Following capture the antegrade conduction timedecreased ivith earlier premature depolarizations. Thiseffect was not seen in experiments rcith short conductiontimes.Circulation Research. Vol. ,f.5, December m?i

The correlation coefficient for estimated ver-sus measured values of sinoatrial conductiontime for 15 experiments was 0.64. Resultsfrom each experiment are shown in Table 1.The results in experiments with a plateau-shaped curve did not differ significantly fromthose in the other experiments. Also use ofthe plateau value for the duration of thenoncompensatory return cycle in these ex-periments did not improve our ability to pre-dict sinoatrial conduction time.

Discussion

The most important observation in thisstudy is that late atrial premature depolari-zations which fail to capture the sinus nodecan cause shortening of the sinus node ac-tion potential by as much as 54 msec. Short-ening of the action potential in turn causesshortening of the sinus node return cycle sothat atrial premature depolarizations whichfail to capture the sinus node can shortenthe sinus node return cycle duration.

There is good evidence that action poten-tial shortening is not mediated indirectly bycatecholamines or acetylcholine released asa result of electrical stimulation; action po-tential shortening occurred even in the pres-ence of effective beta-receptor (21) and cho-linergic receptor blockade (22).

Mendez and Moe (23) have demonstratedthat electrotonic interaction can occur in therabbit atrioventricular (AV) node and causealteration of action potential duration of AVnodal cells. We demonstrated that atrial pre-mature depolarizations could similarly causealteration of action potential duration in thesinus node. Depending on their prematurity,late atrial premature depolarizations pene-trated a variable distance into the sinusnode before blocking at the site of collisionwith the emerging sinus node impulse. Nor-mally the impulse formed in the sinus nodewould proceed toward the crista terminalisand sequentially depolarize the cells lying inbetween. Repolarization would occur in asimilar sequence. However, when an atrialpremature depolarization was introduced,the sequence of activation peripheral to thesite of block was reversed. This reversal ofactivation permitted repolarization periph-eral to the block to precede its expected timeof occurrence and caused disparity of repo-larization between cells on either side of the



944 MILLER, STRAUSS

site of block. The early repolarization periph-eral to the block then caused early repolari-zation central to the block by electrotonicinteraction. The early repolarization causedshortening of the sinus node action potential.As the site of the block moved closer to thepacemaker cell, shortening of the sinus nodeaction potential increased. This proposedmechanism for electrotonic interaction isdemonstrated in Figure 6.

In the majority of experiments, shorteningof the sinus node action potential was accom-panied by similar shortening of the sinusnode and atrial return cycles, but this simi-larity was not always the case. When Bonke(24) applied hyperpolarizing and depolarizingcurrent to sinus node cells during phase 4depolarization, he found that only in someinstances did the cycle length of the prepara-tion change appropriately. In those cases inwhich no change in cycle length occurred, heconcluded that the true pacemaker was dis-tant from the polarizing electrode and wastherefore unaffected. He reemphasized thatseveral areas within the sinus node can haveaction potential morphology typical of pace-maker cells but that these areas do not nec-essarily drive the heart and can in fact depo-larize after atrial depolarization (24, 25). Inthe present study, records were only takenfrom cells discharging at least 25 msec beforethe atrial electrogram. Also in 14 of 15 exper-iments, shortening of the sinus node actionpotential was followed by shortening of thesinus return cycle duration, indicating thatthe cell being recorded from was either thetrue pacemaker or close to it. When someaction potential shortening occurred with-out a change in return cycle duration (Fig.4), the true pacemaker was presumablyslightly more central in the sinus node thanthe cell from which the recordings werebeing made. Electrotonic effects could thenshorten the recorded action potential with-out affecting the more centrally placed truepacemaker, but slightly earlier atrial prema-ture depolarizations would penetrate far-ther and affect both. In one experiment, therecorded cell and the true pacemaker werepresumably farther apart, since action po-tential shortening occurred with late atrialpremature depolarizations but no change incycle length occurred until the prematuredepolarizations were early enough to cap-ture the recorded cell. This phenomenon pre-

sumably coincided with the onset of electro-tonic effect on the pacemaker due to deeperpenetration by the atrial premature depolari-zation. In several experiments, the electro-tonic effect did not completely account forthe disparity between estimated and meas-ured values for sinoatrial conduction time.In these experiments, measured values ofsinoatrial conduction time were high; thetrue pacemaker might have been nearer thecrista terminalis than the recorded cell.

Electrotonic effects provide a significantsource of error in determining sinoatrialconduction time by premature atrial stimula-tion, as shown in Table 1. A further short-coming relates to the uncertainty of the rela-tive contributions of intra-atrial conductionas opposed to conduction between the atriumand the sinus node. When the atrial compo-nent was increased by placing stimulatingand recording electrodes on the atrial appen-dage farther from the sinus node, the estima-ted sinoatrial conduction time increased ap-propriately. Also, antegrade and retrogradeconduction time differed (Table 1), suggest-ing that the antegrade and retrograde con-duction pathways differed slightly (26). Boththis result and the effect of stimulation fromdifferent atrial sites indicate that the atrialcomponents of sinoatrial conduction time areimportant and that their relative contribu-tions are not assessed by premature atrialstimulation. The atrial component of the si-noatrial conduction pathway in the ante-grade direction can be effectively elimi-nated by using a surface P wave, since thebeginning of the P wave coincides with acti-vation of the crista terminalis (27).

The diagram in Figure 1 is drawn on theassumption that antegrade and retrogradeconduction times are equal; however, thisassumption was incorrect for most experi-ments. Retrograde conduction time is us-ually less than antegrade conduction timefor an atrial premature depolarization justcapturing the sinus node (10, 16). With ear-lier atrial premature depolarizations, retro-grade conduction time is prolonged (10, 16).The slowing of conduction which this phe-nomenon must reflect indicates a prolongedrelative refractory period of cells betweenthe atrium and the sinus node. In a previousstudy, the recovery of phase 0 Vmax of rab-bit perinodal fibers has been shown to bestrikingly time dependent, suggesting that

Circulation Research, Vol. J5, December 1971,




the fibers have a prolonged relative refrac-tory period (28).

Shortening of the sinus node return cyclewith late premature beats occurred in allexperiments except one. Shortening per-sisted in four experiments with atrial prema-ture depolarizations as early as 50% of thebasic atrial cycle length, although with thisprematurity there was a tendency for thesinus return cycle duration to increase to-ward the control value. In eight experi-ments, sinus return cycle duration returnedto near the control value following the initialshortening. In such experiments, shorteningof the return cycle length due to action po-tential shortening was presumably counter-balanced by a depression of automaticityresulting from the atrial premature depolari-zation. In two experiments, this depressionwas the predominant effect, and, followinginitial shortening, sinus return cycle dura-tion increased, in one case to 114% of normal.Eccles and Hoff (5) thought that the majorresponse of the sinus node to an atrial pre-mature depolarization was a depression ofautomaticity. This depression, in fact, ap-pears to be an unusual response, and short-ening of the return cycle length or no changein rhythm is much more common.

The shortening of conduction time fromthe sinus node cell to the crista terminalisfollowing atrial premature depolarizationsconfirms earlier studies (10, 16). These stud-ies showed convincing evidence that shorten-ing resulted from a shift in the pacemakertoward the origin of the premature beat; theshift could persist for several cycles and oc-curred because latent pacemakers at the pe-riphery of the sinus node were depolarizedearly by an atrial premature depolarization.Although the natural cycle length of thelatent pacemaker would normally be slightlylonger than that of the true pacemaker, fol-lowing an atrial premature depolarization itsdepolarization precedes sinus node depolari-zation and with this early start it has anadvantage over the true pacemaker. A fur-ther factor may be shortening of action po-tential duration which also permits earlyonset of phase 4 depolarization. Both factorsare exaggerated by earlier atrial prematuredepolarizations, thus accounting for moremarked pacemaker shifts with early prema-ture depolarizations. However, with the latepremature depolarizations used in the pres-

Circttlation Re$earch, Vol. .I.',, December 1UTA

ent study, pacemaker shifts were only de-tected in one experiment, and this phenome-non did not otherwise contribute to actionpotential shortening.

Comparison of the graphs obtained fromexperiments on rabbit atria with those ob-tained from other animal species and manshows similarities and differences. In some ofthese studies the transition from a compen-satory return cycle to a constant but lessthan compensatory return cycle has beeninterpreted as signifying the transition fromnoncapture to capture of the sinus node bythe atrial premature depolarization (1, 17).In studies in man and dogs, this transitionoccurs over a relatively long diastolic inter-val (5, 19). Although sinus node action poten-tials were not recorded in these studies, wenow believe that this long transition periodreflects electrotonic shortening of the sinusnode action potential at a time when thepremature depolarization has still not cap-tured the sinus node. Hence, in these stud-ies curves with long transition periodsshould preclude an accurate estimation ofsinus node capture by the atrial prematuredepolarization and therefore preclude an ac-curate estimation of sinoatrial conductiontime.

In the human studies reported thus far,the curves depicting the atrial return cycleresponse to atrial premature depolarizationsdiffer significantly from those obtained inanimal studies (1, 12, 13). Typically, a fullycompensatory response in the later portion ofdiastole is followed on the graph by a sharptransition to a constant, less than compensa-tory return cycle, the transition occurringover a narrow range of atrial prematuredepolarization coupling intervals. Since wedid not obtain such responses in our rabbitstudies, the sinus node response to atrialpremature depolarizations in this situationis unknown. One can speculate that the dif-ference in response may in part reflect thelonger cycle lengths of human studies. If theabsolute shortening of sinus node return cy-cles occurring with atrial premature depolar-izations that fail to capture the sinus node iscomparable at long and short cycle lengths,then the relative shortening of the returncycle at long cycle lengths would be small,possibly undetectable.

Another difference between the curves ob-tained in rabbit and human studies is the



946 MILLER, STRAUSS

infrequent occurrence of a plateau responsein the rabbit studies. In five such experi-ments, the pattern of sinus node responsewas inconsistent. In the majority of our ex-periments, the increasing value of the atrialreturn cycle was due to an increase in theconduction time of the premature depolariza-tion into the sinus node as the couplinginterval of the atrial premature depolariza-tion decreased. The long cycle lengths re-corded in human studies may permit greaterrecovery from the refractory state and hencea more uniform conduction time of the atrialpremature depolarization into the sinusnode.

In conclusion, we were unable to validatethe premature atrial stimulation techniqueas an accurate method of assessing sino-atrial conduction time in the rabbit, primarilybecause atrial premature depolarizationscaused shortening of the sinus node actionpotential. This shortening in turn causedshortening of the sinus return cycle durationand the appearance of noncompensatoryatrial responses in the absence of sinus nodecapture. The premature atrial stimulationtechnique therefore remains unsubstan-tiated as a means of accurately measuringsinoatrial conduction time in man.

Acknowledgment

We gratefully acknowledge the technical assistance ofMiss Anne Winfree and the electronic expertise of JackKasell. We thank Ayerst Laboratories for their generoussupply of Inderal.

References1. STRAUSS HC, SAROFF AL, BIGGER JT JR, GIARDINA

EGV: Premature atrial stimulation as a key to theunderstanding of sinoatrial conduction in man.Circulation 47: 86-93, 1973

2. WENCKEBACH KP: Uber die Dauer der compensato-richen Pause nach Reizung der Vorkammer desSaugethierherzens. Arch Anat Physiol PhysiolAbt, pp 57-64, 1903

3. CUSHNY AR, MATTHEWS SA: On the effects ofelectrical stimulation of the mammalian heart. JPhysiol (Lond) 21:213-230, 1897

4. SANSUM WD: Extrasystoles in the mammalian heartcaused by the stimulation of the Keith-Flack node.Am J Physiol 30:421^128, 1912

5. ECCLES JC, HOFF HE: Rhythm of the heart beat: II.Disturbance of rhythm produced by late prema-ture beats. Proc R Soc Lond [Biol] 115:327-351,1934

6. HlRSCHFELDER AD, EYSTER JAE: Extrasystoles inthe mammalian heart. Am J Physiol 18:222-249,1907

7. LANGENDORF R, LESSER ME, PLOTKIN P, LEVIN BD:Atrial parasystole with interpolation: Observa-tions on prolonged sinoatrial conduction. AmHeart J 63:649-658, 1962

8. FLEISCHMANN P: Interpolation of atrial prematurebeats of intra-atrial origin due to concealed A-Sconduction: Report of a case of A-V nodal parasys-tole and of a case of premature impulses emergingfrom a preexcitation bypass. Am Heart J 66:309-320, 1963

9. HAN JH, MALOZZI AM, MOE GK: Sino-atrial recipro-cation in the isolated rabbit heart . Circ Res22:355-362, 1968

10. BoNKE FIM, BoUMAN LN, VAN RUN HE: Change ofcardiac rhythm in the rabbit after an atrial pre-mature beat. Circ Res 24:533-544, 1969

11. BONKE FIM, BoUMAN LN, SCHOPMAN FJG: Effect ofan early atrial premature beat on activity of thesinoatrial node and atrial rhythm in the rabbit.Circ Res 29:704-715, 1971

12. GOLDREYER BN, DAMATO AN: Sinoatrial-node en-trance block. Circulation 44:789-802, 1971

13. NARULA OS, SAMET P, JAVIER RP: Significance ofthe sinus node recovery time. Circulation 45:140-158, 1972

14. PAULAY KL, VARGHESE PJ, DAMATO AN: Atrialrhythms in response to an early atrial prematuredepolarization in man. Am Heart J 85:323-331,1973

15. PAULAY KL, VARGHESE PJ, DAMATO AN: Sinusnode reentry: An in vivo demonstration in thedog. Circ Res 32:455-^163, 1973

16. KLEIN HO, SINGER DH, HOFFMAN BF: Effects ofatrial premature systoles on sinus rhythm in therabbit. Circ Res 32:480-491, 1973

17. BOND RC, ENGEL TR, SCHAAL SF: Effect of digitalison sinoatrial conduction in man (abstr). Am JCardiol 33:128, 1974

18. CHILDERS RW, ARNSDORF MF, DE LA FUENTE DJ,GAMBETTA M, SVENSON R: Sinus nodal echoes. AmJ Cardiol 31:220-231, 1973

19. TICZON AR, STRAUSS HC, GALLAGHER JJ, WALLACEAG: Sinus node function in the intact dog heart(abstr). Circulation 48(suppl IV):IV-62, 1973

20. P A E S DE CARVALHO A, DE MELLO WC, HOFFMANBF: Electrophysiological evidence for specializedfiber types in rabbit atrium. Am J Physiol 196:483-488, 1959

21. BLINKS JR: Evaluation of the cardiac effects ofseveral beta adrenergic blocking agents. Ann NYAcad Sci 139:673-685, 1974

22. AMORY DW, WEST TC: Chronotropic response follow-ing direct electrical stimulation of the isolatedsinoatrial node: A pharmacologic evaluation. JPharmacol Exp Ther 137:14-23, 1962

23. MENDEZ C, MOE GK: Some characteristics of trans-membrane potentials of AV nodal cells duringpropagation of premature beats. Circ Res 19:993-1010, 1966

24. BOXKE FIM: Electrotonic spread in the sinoatrialnode of tho rabbit heart. Pfluegers Arch 339:17-23, 1973

25. HOFFMAN BF, CRANEFIELD PF: Physiological basisof cardiac arrhythmias. Am J Med 37:670-684,1964

Circulation Research, Vol. 35, December 197i



SINOATRIAL CONDUCTION 9 4 7

26. S A N O T , YAMAGISHI S: Spread of excitation from the and their relationship to atrial potentials in thesinus node. Circ Res 16:423-430, 1965 dog. Circ Res 30:393-405, 1972

27. KING TD, BARR RC, HERMAN-GIDDENS GS, BOAZ 28. STRAUSS HC, BIGGER JT JR: ElectrophysiologicalDE, SPACH MS: Isopotential body surface maps properties of the rabbit sinoatrial perinodal fibers.

Circ Res 31:490-506, 1972

Circulation Research, Vol. -h">, December 1974



HUGH C. MILLER and HAROLD C. STRAUSSMeasurement of Sinoatrial Conduction Time by Premature Atrial Stimulation in the Rabbit

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1974 American Heart Association, Inc. All rights reserved.is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation Research

doi: 10.1161/01.RES.35.6.9351974;35:935-947Circ Res.

http://circres.ahajournals.org/content/35/6/935World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circres.ahajournals.org//subscriptions/

is online at: Circulation Research Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer about this process is available in the

located, click Request Permissions in the middle column of the Web page under Services. Further informationEditorial Office. Once the online version of the published article for which permission is being requested is

can be obtained via RightsLink, a service of the Copyright Clearance Center, not theCirculation Research Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:


http://circres.ahajournals.org/content/35/6/935

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circres.ahajournals.org//subscriptions/


MEASUREMENT OF SINO-ATRIAL CONDUCTION TIME BY PROGRAMMED ELECTROSTIMULATION

Documents