97 the interruption of collateral blood flow to the...

97

The Interruption of Collateral Blood Flow tothe Ischemic Canine Myocardium by

Embolization of a Coronary Artery withLatex: Effects on Conduction Delay and

Ventricular Arrhythmias

DAVID E. EULER, CHARLES E. PROOD, J.F. SPEAR, AND E. NELL MOORE

SUMMARY We studied the importance of collateral blood flow in the genesis of epicardial conductiondelay and ventricular arrhythmias by creating a model of myocardial ischemia in which coronarycollateral channels were embolized with a rapidly hardening vinyl latex. In 33 open-chest anesthetizeddogs, heart rate was held constant (18O-200/min) and the proximal left anterior descending coronaryartery (LAD) was occluded for 10 minutes, followed by 30 minutes of reperfusion. The LAD then wasligated and latex was injected just below the ligature. The latex filled coronary vessels branching fromthe LAD down to a diameter of 20-30 jun and produced transmural infarcts averaging 42% of the leftventricular mass (range 35-58%). The amount of epicardial conduction delay was measured duringatrial rhythm at multiple electrode sites within the ischemic zone. The maximal delay followingcoronary occlusion was 97 ± 48 msec (mean ± SD), whereas latex embolization increased the delay to203 ± 51 msec (P < 0.001). In response to coronary occlusion, 8% of the electrograme demonstratedWenckebach-like alternation, whereas 58% showed alternation following latex embolization (P < 0.001).Compared to coronary occlusion, coronary embolization with latex significantly increased the numberof dogs that developed ectopic beats during the first 10 minutes of ischemia (66% vs. 94%; P = 0.02).However, coronary occlusion evoked ventricular fibrillation in 21% of the dogs, whereas no animaldeveloped fibrillation at any time following coronary embolization (P = 0.002). Ventricular fibrillationalso was absent in an additional 18 animals in which the LAD was embolized without an earlier periodof occlusion. The results indicate that embolization of the LAD with vinyl latex increases the severityof local conduction abnormalities while reducing the incidence of lethal arrhythmias. The interruptionof collateral blood flow to the ischemic myocardium may depress conduction in a homogeneous mannersuch that the ischemic zone may be unable to support a sufficient number of reentrant circuits toinitiate fibrillation. Ore Res 49: 97-108, 1981

ALTHOUGH abrupt occlusion of a major coronaryartery consistently produces a myocardial infarct,the size of the infarct generally is considerablysmaller than the area of myocardium normally per-fused by the occluded vessel (Schaper et al., 1979;Jugdut et al., 1979). The survival of jeopardizedmyocardium is dependent upon the delivery of alimited amount of arterial blood by preexisting in-tercoronary collateral vasculature; the amount ofmyocardium surviving is related directly to thevolume of collateral flow available to the acutelyischemic myocardium early after coronary occlu-sion (Rivas et al., 1976; Bishop et al, 1976). Inaddition to producing myocardial necrosis, acutecoronary artery occlusion is followed by severe elec-trophysiological alterations manifested by changes

From the University of Pennsylvania, School of Veterinary Medicine,Philadelphia, Pennsylvania.

Supported by Grants HL 16076, HL 23071, and HL 190*6 from theNational Heart, Lung, and Blood Institute.

Address for reprints: David E. Euler, Ph.D., University of Pennsyl-vania, Veterinary School, 3800 Spruce Street, Philadelphia, Pennsylvania19104.

Received September 19, I960; accepted for publication January 5,1961.

in myocardial activation. Several studies have pro-vided quantitative data relating changes in extra-cellular bipolar electrograms to the development ofventricular arrhythmias (Scherlag et al., 1970;Waldo and Kaiser, 1973; Boineau and Cox, 1973;Cox et al., 1973; Hope et al., 1974; Williams et al.,1974). It has been shown that the volume of collat-eral blood flow to the ischemic myocardium is in-versely related to the depression in amplitude ofbipolar electrograms recorded from the subepicar-dium and subendocardium and to the increase induration of subepicardial electrograms (Ruffy et al.,1979). Furthermore, ventricular fibrillation follow-ing acute coronary artery occlusion in the dog hasbeen reported to occur only in those animals thathave a limited number of preexisting collateralchannels (Meesman et al., 1970).

To elucidate further the importance of collateralblood flow in the genesis of conduction abnormali-ties and ventricular arrhythmias, it would be usefulto have a method of producing myocardial ischemiain which retrograde flow through preexisting collat-eral channels was blocked. Such a method has beendescribed for producing transmural infarction of the

by guest on June 24, 2018http://circres.ahajournals.org/

Dow

nloaded from

http://circres.ahajournals.org/

98 CIRCULATION RESEARCH VOL. 49, No. 1, JULY 1981

SA node (Loeb et al., 1980). In these studies, arapidly hardening vinyl latex was injected into theSA nodal artery preventing both antegrade andretrograde perfusion of the SA nodal region. Thepurpose of the present experiments was to deter-mine the effects of acute embolization of the leftanterior descending coronary artery with vinyl latexon local patterns of myocardial activation and ven-tricular arrhythmias. The changes in local conduc-tion and ectopic activity induced by latex emboli-zation were compared to the changes induced bycoronary occlusion without embolization. In com-parison to a period of acute coronary occlusion,coronary embolization with latex resulted in moreconduction abnormalities and a greater number ofectopic beats. However, latex embolization neverevoked ventricular fibrillation in the canine heart.

Methods

Animal PreparationExperiments were performed on 53 adult male

dogs weighing 15-23 kg. All animals were free ofheart worms and respiratory disease and were ingenerally good health. The dogs were anesthetizedwith sodium pentobarbital (30 mg/kg, iv) and ven-tilated with room air via a positive pressure venti-lator at a minute-volume determined from a bodyweight nomogram. Supplemental doses of sodiumpentobarbital (2 mg/kg) were administered every1-2 hours thereafter to maintain a surgical plane ofanesthesia. A femoral artery was cannulated so thatarterial blood pressure could be monitored contin-uously throughout each experiment. A left thora-cotomy was performed in the 5th and 6th intercostalspaces and the 5th rib removed.

The temperature of the ventricular epicardiumwas monitored with a small thermistor (YSI 425),and cardiac temperature was held constant (37-38°C) by means of a heating lamp placed above thethoracic cavity. The left anterior descending coro-nary artery was dissected free 5-10 mm from itsorigin, proximal to all diagonal branches. Closeintramural plunge-type bipolar electrodes were in-serted into the left atrial appendage for pacing (180-200/min) and into the left ventricular subepicar-dium and subendocardium for recording local elec-trograms. The electrograms were filtered at settingsof 40-500 Hz. The electrograms along with a lead IIECG were stored on a 7-channel FM tape recorder(Honeywell 5600C). At a later time, the electricalsignals were retrieved from tape and displayed onan eight-channel Siemens-Elema ink-writing min-gograph at paper speeds of 25-250 mm/sec. Plungeelectrodes were also inserted into the left and rightcervical vagosympathetic trunks.

Experimental ProtocolIn 33 dogs, the left anterior descending coronary

artery was occluded with an atraumatic clamp fora period of 10 minutes and then released. The

cervical vagi were stimulated for about 30 seconds(20 Hz, 100 fisec, 10 V) to produce asystole whilethe clamp was applied to the artery. The vagi alsowere stimulated just before release of the occludedvessel to minimize the severity of reperfusion ar-rhythmias (Murdock et al., 1980). A 20- to 30-min-ute recovery period was allowed to elapse beforethe injection of latex. In those animals in whichventricular fibrillation occurred during either occlu-sion or reperfusion, the heart was rapidly defibril-lated by DC countershock and a 45-minute recoveryperiod was allowed before latex was injected. Justprior to latex injection, the left anterior descendingartery was ligated and the vagi were stimulated toproduce asystole. A small incision was made in theartery just distal to the ligature and a blunt 20-gauge hypodermic needle with a 90° bend was in-serted into the vessel. The needle was connected toa 5-ml syringe containing about 3-4 ml latex (Car-olina Biological Supply). To minimize retrogradeflow of latex through the incision in the artery, theartery was grasped with a pair of atraumatic forcepsand pressure was applied around the cannula. Latexthen was injected rapidly until the distal vascula-ture was filled and the latex flowed retrogradethrough the arterial incision. A volume of 1.5-2.0ml of latex usually was required to fill the leftanterior descending artery and its distal branches.The high viscosity of the latex prevented it fromtotally penetrating the collateral channels and fill-ing the distal branches of the left circumflex coro-nary artery. Once in the vasculature, the latexhardened in about 10-20 seconds. Vagal stimulationwas maintained throughout the entire injection pe-riod (20-30 seconds). In 18 additional dogs, latexwas injected into the left anterior descending coro-nary artery without an earlier period of coronaryocclusion. The ECG and blood pressure were mon-itored in all dogs for 1-6 hours after latex emboli-zation. Dogs that were not used for the evaluationof infarct size or collateral blood flow were killed atthe end of the 1-hour monitoring period.

Evaluation of Infarct Size and CollateralFlow

Fifteen dogs were monitored for a period of 6hours after intracoronary latex injection. The heartfrom each animal was rapidly excised and the rightventricle and great vessels were removed. The leftventricle and septum were cut transversely fromapex to base into 0.5- to 1-cm sections. The sectionswere incubated in para-Nitro Blue Tetrazoleum (p-NBT 0.25 g/liter, 0.01 M phosphate buffer) for 30minutes at room temperature to stain normal tissuefor lactate dehydrogenase (Schaper et al., 1979).The unstained necrotic tissue was separated fromthe normal tissue and the infarct size was quantifiedgravimetrically. In two additional animals not in-jected with latex, hearts were excised, placed inRinger's solution for 6 hours, and then sectionedand incubated with p-NBT.


Dow

nloaded from


CORONARY EMBOLIZATION, CONDUCTION DELAY, ARRHYTHMIAS/Euteretal. 99

To evaluate the effectiveness of latex in blockingretrograde collateral flow, hearts were excised fromeight animals 1 hour after latex injection; the leftmain coronary artery was cannulated and indocy-anine green was infused at a pressure of 100-150mm Hg. These hearts were sectioned perpendicularto the apex-base direction and the distribution ofgreen dye was determined relative to the latex. Ineight more dogs, hearts were excised 1 hour afterlatex injection and the myocardial tissue was di-gested with concentrated sulfuric acid. The remain-ing latex cast was examined microscopically andthe diameters of the most distal branches weredetermined.

Evaluation of Conduction Abnormalitiesduring Ischemia

In 23 animals, three to six subepicardial elec-trodes were placed randomly within the distributionof the left anterior descending coronary artery andelectrograms were monitored following both coro-nary occlusion and coronary embolization with la-tex. In six additional animals, three to six endocar-dial electrograms were monitored for the same pe-riod. Conduction delay was determined by measur-ing the interval from the onset of the Q-wave in theECG to the end of the latest complex associatedwith ventricular activation in the local electro-grams. In six of the 23 animals in which epicardialelectrograms were recorded, brief periods (2-3 sec-onds) of vagal stimulation were initiated during thefirst 30 minutes of ischemia produced by latex todetermine the effects of changes in heart rate onischemic zone conduction delay. In 12 additionaldogs, epicardial mapping of the ischemic and adja-cent normal zone was performed 30 minutes afterthe intracoronary injection of latex. The mappingelectrode consisted of two insect pins (size 000)insulated except at the tips and embedded in acrylicto give an inter-electrode distance of 2 mm. Record-ings were obtained from about 25-30 locations ineach dog. In the central portion of the ischemiczone, recording sites were 0.5-1 cm apart, while inthe border zone sites were 3-5 mm apart. The atriawere paced at a constant rate of 180/min during themapping procedure. In addition to mapping acti-vation, the mapping electrode was used to stimulatethe myocardium at multiple sites (five to 10) withinthe ischemic and normal zone. The ischemic myo-cardium was stimulated with constant currentpulses (4 msec) of 1-30 mA intensity at a rate about20% faster than the sinus rate.

Data AnalysisWhen possible, all results were expressed as the

mean ± SD. The comparison of means was accom-plished with a paired t-test. When data were ex-pressed as the frequency of occurrence (incidence)of an event, a x2 analysis or Fisher's exact test wasemployed. All differences were considered to bestatistically significant when P ^ 0.05.

ResultsFigure 1 shows a typical latex cast of the left

anterior descending coronary artery and its termi-nal branches. Latex was injected in vivo at point A,and points B and C indicate the first diagonalbranch and apical branch, respectively. Five to 10of the distal branches arising from each diagonalvessel were examined under a microscope and foundto have diameters of 20-30 jtm. Retrograde bloodflow was always observed when the left anteriordescending coronary artery was ligated and cannu-lated just prior to latex injection. However, afterlatex had hardened within the vessel, no retrogradeflow could be observed at the arterial incision. Thelatex filled epicardial and intramural vessels, butwas never observed on the endocardial surface, inthe coronary veins, or within the ventricular cavity.In the eight excised hearts in which indocyaninegreen was injected into the left main coronary ar-tery, no dye could be seen within the area of latexdistribution. The boundary between green dye andblue latex was very sharp on both the epicardium

C~FIGURE 1 A latex cast of a left anterior descendingcoronary artery and its distal branches. Latex was in-jected into the intact heart at point A. Point B indicatesthe first diagonal branch of the LAD, while point Cmarks the apical branch. Microscopic examination ofthe distal latex branches showed minimal diameters of20-30 fim.


Dow

nloaded from


100 CIRCULATION RESEARCH VOL.49, No. 1, JUXY 1981

and the intramural slices. There was also a sharpboundary between normal and necrotic tissue in thehearts which had been stained for lactate dehydro-genase 6 hours after latex injection. Figure 2 illus-trates a typical infarct produced by latex emboli-zation of the left anterior descending coronary ar-tery. The necrosis was transmural in each section,and involved the entire apex, a large portion of theanterior free wall, and part of the intraventricularseptum. Although the blue latex in Figure 2 cannotbe seen in the unstained portions of the myocar-dium, all of the infarcted muscle contained intra-mural vessels filled with the latex. The infarctshown in Figure 2 occupied 42% of the left ventric-ular mass. The average infarct mass in 15 of thelatex-injected hearts was 42 ± 8% (range 36-58%).The average left ventricular mass was S3 ± 11 g(range 65-98 g). The large variability in the relativeamount of necrotic tissue among animals probably

reflects an anatomic variability in the perfusionarea of the coronary vasculature. Although all in-farcts were transmural, only dogs with infarctagreater than 45% (four animals) showed hemody-namic deterioration during the observation period.To maintain a mean arterial blood pressure greaterthan 70 mm Hg in these animals, it was necessaryto infuse methoxamine (1 fig/kg per min) after thefirst hour of ischemia.

Ventricular Arrhythmias during MyocardialIschemia

Figure 3 shows rhythm strips (lead II) from atypical experiment during and after the injection ofvinyl latex into the left anterior descending coro-nary artery. Panels A and B show the onset (arrowpanel A) and termination (arrow panel B) of a 25-second period of vagal stimulation that was neces-sary for arterial cannulation and latex injection.

FIGURE 2 An infarct produced by latex embolization of the left anterior descending coronary artery. This figureshows five transverse sections of the left ventricle and septum which were stained withp-NBT. The most apical sectionis shown in the upper left, whereas the most basilar section is in the lower right. The unstained necrotic tissue wastransmural in each section, and involved the entire apex, a large portion of the anterior free wall, and part of theinterventricular septum. The unstained tissue was highlighted with an opaque whitener to improve the contrast of thephotograph. Although blue latex cannot be seen in the necrotic unstained portions of the myocardium, all the infarctedmuscle contained intramural vessels filled with latex. The infarct shown here occupied 42% of the left ventricularmass.


Dow

nloaded from


CORONARY EMBOLIZATION, CONDUCTION DELAY, ABRHYTHMIAS/Euler et al. 101

i -LLU-1- l l l - !^ fB jA.... <A. ^ I

• ' '

I I | ;

I ! I

FIGUHE 3 Changes in cardiac rhythm resulting from the intracoronary injection of latex. Panels A and B showrhythm strips (lead II) during the onset (arrow, Panel A) and termination (arrow, panel B) of vagally inducedbradycardia that was necessary for LAD cannulation and latex injection. Two PVC's (asterisks in panel B) occurredwhile latex was being injected. Panel C, recorded 3 minutes after the onset of the vagal stimulation, shows recurringepisodes of runs of multiform PVC's. Panel D was recorded at 14.5 minutes of ischemia and shows a brief run ofventricular tachycardia. The time lines at the bottom of the figure indicate 1-second intervals.

The two PVC's (asterisks in panel B) that inter-rupted the idioventricular escape rhythm probablywere due to mechanical stimulation of the left ven-tricle as latex distended the coronary vasculature.Runs of two to 20 PVC's were commonly observedas latex filled the vasculature in all animals. How-ever, the ectopic activity ceased immediately uponremoval of the pressure applied to the syringe con-taining latex. Within 1 second of the termination ofvagal stimulation (panel B), the atrial rhythm re-turned and remained stable for the next 2.5 minutes.Panel C, recorded 3 minutes after the onset of vagalstimulation, shows recurring episodes of runs ofmultiform PVC's. The ectopic activity in this ani-mal reached a peak rate of 80 PVC's/min anddeclined so that, by 6 minutes of ischemia, the earlyectopic activity had ceased. A second surge of ec-topic activity appeared after 13 minutes of ischemia.Panel D was recorded at 14.5 minutes of ischemiaand shows a brief run of ventricular tachycardia.The ectopic activity during the delayed periodreached a peak rate of 100 PVC's/min and declinedto result in a quiescent period by 21 minutes of

ischemia. Continued monitoring of this animal fora total period of 1 hour revealed no further ventric-ular arrhythmias.

Table 1 shows the incidence and severity of ven-tricular arrhythmias observed during myocardialischemia produced by both coronary occlusion andcoronary embolization with latex. Latex emboliza-tion significantly increased the number of animalsthat developed ectopic activity during the first 10minutes of ischemia. The latency from the onset ofcoronary occlusion (onset of vagal stimulation) tothe appearance of ventricular arrhythmias was 3.0±1.0 min and not significantly different (P > 0.1)from the latency between embolization and theonset of arrhythmia (3.2 ± 0.8 min). Within 10minutes of ischemia, induced by either occlusion orembolization, all ectopic activity had ceased. Theincidence of ventricular tachycardia (four or morerepetitive PVC's) was similar for both occlusion andembolization. The most striking finding shown inTable 1 was the absence of ventricular fibrillationin any of the 33 latex-injected hearts in contrast toa 21% incidence of fibrillation after coronary occlu-

TABLE 1 Comparison of the Arrhythmogenic Effects of Coronary Occlusion andCoronary Embolization

Coronary occlusionCoronary embolization (1st

10 minutes)

P'

NS — not significant.• Fisher's exact test n - 33.

Incidence ofectopic activity(»1 PVC/min)

66%94%

0.02

Incidence ofventricular tachycardia(==4 repetitive PVC's)

45%52%

NS

Incidence ofventricularfibrillation

21%0%

0.0O2


Dow

nloaded from



sion. It is possible that the initial period of ischemiaresulting from coronary occlusion may have alteredthe electrophysiological response of the myocar-dium to subsequent periods of ischemia, therebyaccounting for the absence of fibrillation after latexinjection. However, latex embolization did notevoke ventricular fibrillation in 18 additional dogs(not included in Table 1) that did not sustain a 10-minute period of coronary occlusion. The absenceof ventricular fibrillation in these dogs represents astatistically significant difference compared to a21% incidence of fibrillation after coronary occlu-sion (P = 0.04; Fisher's exact test). In general, nodifferences in the severity of arrhythmias were ob-served in these 18 animals compared to the arrhyth-mias observed in the 33 animals in which a 10-minute period of coronary occlusion preceded latexinjection.

Following the period of ectopic activity observedduring the first 10 minutes of ischemia, a delayedsurge of arrhythmias appeared in the latex-injectedhearts (Fig. 3). These delayed ectopic beats ap-peared with an average latency of 16.7 ± 3 minutesand disappeared by 30 minutes. The incidence ofdelayed arrhythmias (74%) was not significantlydifferent (P > 0.05) from the incidence of earlyarrhythmias (88%). Furthermore, the maximum fre-quency of the delayed ectopic beats (42 ± 53 PVC's/min) was not significantly different (P > 0.1) fromthe maximum frequency of early ectopic beats (43± 38 PVC's/min). After the disappearance of thedelayed arrhythmias, no further episodes of signifi-cant ectopic activity were observed for up to 6 hoursfollowing latex injection.

Changes in Local Myocardial ActivationFigure 4 illustrates data from a typical experi-

ment in which coronary embolization resulted inmore severe conduction abnormalities than did cor-onary occlusion. The records show a surface ECG(lead II) and three local electrograms recorded fromthe ischemic zone. One electrogram (Rl) was re-corded from the subendocardium, whereas theother two electrograms (R2 and R3) were recordedfrom the subepicardium. Panel A shows a controlrecord while panels B and C show records at 4minutes <B) and 8 minutes (C) after coronary occlu-sion. Little conduction delay was evident in thesubendocardium (Rl), although there was a markedreduction in electrogram amplitude. One of theepicardial electrograms (R2) showed fractionationwith a maximum delay of 46 msec at 4 minutes ofischemia (panel B). There was little change in theamount of delayed activation between panels B andC. Panel D was recorded after 30 minutes of reper-fusion of the ischemic myocardium and shows areturn of the control activation pattern. Panels Eand F were recorded at 4 minutes (E) and 8 minutes(F) after coronary embolization with latex. Again,the endocardial electrogram (Rl) showed a de-crease in amplitude with only minimal conduction

FIGURE 4 Changes in local activation following coro-nary occlusion and coronary embolization. The recordsshow a surface ECG (lead II) and three local electro-grams recorded from the ischemic zone. One electrogram(Rl) was recorded from the subendocardium, whereasthe other two electrograms (R2 and R3) were recordedfrom the subepicardium. Panel A shows a control recordand panels B and C show records at 4 minutes (B) and8 minutes (C) after coronary occlusion. Panel D wasrecorded after 30 minutes of reperfusion of the ischemicmyocardium. Panels E and F were recorded at 4 minutes(E) and 8 minutes (F) after coronary embolization withlatex.

delay (<10 msec). Both epicardial electrogramsshowed marked conduction delay with late activa-tion in panel E extending beyond the T-wave of theECG (196 msec delay). Eight minutes after embo-lization (panel F), one epicardial electrogram (R2)showed no intrinsic deflection, whereas the ampli-tude of the response in the other (R3) was markedlydiminished. It is interesting to note that there wereno arrhythmias in this dog during the 10 minutes ofcoronary occlusion, whereas ectopic activity follow-ing embolization reached a maximum of 83 PVC's/min.

Results similar to those shown in Figure 4 wereobtained for all animals in which local electrogramswere monitored during both occlusion and emboli-zation. Only minimal delays (<20 msec) were ob-served in the subendocardium during ischemia in-duced by either occlusion or latex embolization. Inthe subepicardium, latex embolization consistentlyresulted in more severe conduction abnormalitiesthan did coronary occlusion. Table 2 gives a quan-titative comparison of the conduction changes in-duced by coronary occlusion and embolization. Thecontrol measurements obtained just prior to occlu-sion and embolization showed no significant differ-ences in the magnitude of conduction delay. Epi-cardial delay began to appear 1.5-2 minutes afterthe onset of either occlusion or embolization andbecame maximal by 6 minutes. As shown in Table2, significantly greater delays occurred after embo-lization than after occlusion. In addition, coronary


Dow

nloaded from


CORONARY EMBOLIZATION, CONDUCTION DELAY, ARRHYTHMIAS/£ukr et al. 103

TABLE 2 Comparison of Changes in Local Conduction Produced by CoronaryOcclusion and Coronary Embolization

Incidence of Incidence ofMaximal MuTiniiim electrograms electrogramscontrol iflchemic showing post showing

conduction conduct]on T-wave activa- Wencltebach-likedelay (msec)'' delay (msec)* tion!

Coronary occlusionCoronary embolization

31 ± 73O ± 6

NS

97 ±48203 ± 51

<0.001

63.28%

<0.001

8%58%

<0.001

' Values are mean ± SD for measure ments in 23 dogs.t x1 analysii based on 113 electrograms recorded in 23 dogH NS — not significant.

embolization resulted in more electrograms showingdiscrete epicardial activation occurring after the T-wave of the ECG (Fig. 4). Furthermore, a signifi-cantly greater number of electrograms demon-strated Wenckebach-like alternation in response tocoronary embolization.

Embolization of the proximal LAD always pro-duced unresponsiveness in electrograms recordedfrom the center of the ischemic region within 10-15minutes. In more peripheral regions, ischemicchanges took longer to occur and were less exten-sive. Spontaneous improvement in either the elec-trogram amplitude or conduction delay in the is-chemic zone was never observed. Epicardial map-ping after 30 minutes of ischemia showed that themajor portion of the ischemic zone was totallyunresponsive. Figure 5 shows a diagram of a heartthat was mapped 30 minutes after latex emboliza-tion. The stippled area on the anterior surface ofthe left ventricle represents the area of electricalunresponsivenese. Also included in Figure 5 are anECG and electrograms recorded from three differ-ent mapping locations. Electrogram Rl was taken

:

• . . . .

FIGURE 5 An epicardial map obtained 30 minutes afterlatex embolization of the left anterior descending coro-nary artery (LAD). The vessels filled with latex areindicated by the solid lines branching from the LAD.The stippled area on the anterior surface of the leftventricle (L V) represents the area of electrical unrespon-siveness. Also shown are an ECG (II) and three localelectrograms recorded from different locations. Electro-gram Rl was taken from the normal zone, whereas R2was from the region of unresponsiveness and R3 wasfrom the border zone.

from the normal zone, whereas R2 was from theregion of unresponsiveness and R3 was from theborder zone. The small deflections in R2, synchro-nous with the QRS complex, represent an extrinsicdeflection. Extrinsic deflections were consistentlyrecorded at all mapping locations within the zoneof unresponsiveness. The fractionated, delayed ac-tivity in R3 (conduction delay; 100 msec) was re-corded about 5 mm from border between cyanoticand normal-appearing tissue. The conduction delaywas stable over a 30-minute observation period.

Results similar to those shown in Figure 5 wereobserved in all 12 dogs in which the epicardium wasmapped 30 minutes after latex injection. The areaof electrical unresponsiveness was always slightlyless than the area of latex distribution. Low ampli-tude delayed potentials were consistently recordedon the apex and lateral free wall within 5-10 mm ofthe ischemic border. The maximum conduction de-lay averaged 108 ± 39 msec, which was significantlyless than the conduction delay (203 ± 51 msec)observed in more central regions of the ischemiczone during the first 10 minutes of ischemia (P <0.001). The delayed conduction observed in thelateral and apical border zone remained stable overa 30-minute observation period. Normal tissue, im-mediately adjacent to the ischemic border, showedhigh-voltage electrograms with no activation delaybeyond the end of the QRS complex. In addition tomapping local activation, the mapping electrodewas used to deliver current to the central portion ofthe ischemic zone in six of the 12 animals. Pacing atmultiple sites within the central zone (15 mm fromthe border) with current intensities of 30 mA for aperiod of 90 seconds never resulted in a propagatedventricular response. Stimulation of the normalzone with current intensities of 1 mA always re-sulted in propagated responses. Stimulation of theborder zone with 30-mA pulses consistently resultedin ectopic ventricular beats. However, 1-mA pulseswere frequently unsuccessful.

hi several animals, coronary embolization re-sulted in unstable electrograms showing markedbeat-to-beat variations in conduction delay duringthe first 10 minutes of ischemia. Figure 6 illustratesan example of unstable conduction patterns devel-oping after latex embolization. The records showa surface ECG (lead II) and three local electrograms


Dow

nloaded from



nAJLJLB

R3

FIGURE 6 Unstable conduction patterns developing after latex embolization. The records show surface ECG (leadII) and three local electrograms (Rl, R2, R3) recorded from the ischemic subepicardium. Panel A shows a control cycleduring atrial pacing (cycle length 300 msec), whereas panel B was recorded at 4 minutes of ischemia. Panel C wasrecorded 5 seconds after panel B and panel D was recorded 6 minutes after embolization. See text for furtherexplanation.

(Rl, R2, R3) recorded from the ischemic subepicar-dium. Panel A shows a control cycle during atrialpacing (cycle length 300 msec) and panel B showsrecords obtained at 4 minutes of ischemia. In thefirst three cycles of panel B, conduction delay wasevident in all of the local electrograms with themaximum delay occurring in R2 (175 msec). In the

fourth cycle, the delay in Rl suddenly increased to290 msec, whereas little change occurred in theother electrograms. Over the next three cycles, thedelay in Rl progressively decreased so that by theseventh cycle the maximal delay was again observedin R2 (185 msec). In panel C, recorded 5 secondsafter panel B, another paroxysmal increase in delay

R2-

R3-

110 120 120 150 150 150

FIGURE 7 The effect of heart rate on ischemic zone conduction delay. The records show a surface ECG (lead II) andthree local electrograms recorded from the ischemic zone (Rl, R2, R3). Panel A shows a control cycle during atrialpacing (cycle length 333 msec). Panel B, recorded 6 minutes after embolization, shows the onset of a 2-second periodof vagal stimulation (arrow). Panel C shows the first six cycles after termination of vagal stimulation.


Dow

nloaded from


CORONARY EMBOLJZATION, CONDUCTION DELAY, ARRHYTHMIAS/£«/er et al. 105

occurred during the second cycle (150-300 msec inRl). The marked activation delay (300 msec) wassustained with little variation over the next 3 cycles(cycles 3-5). In addition, panel C shows a 2:1 alter-nation of the delayed response in R3 (125-msecdelay). In panel D, recorded at 6 minutes of is-chemia, a 2:1 alternation appeared in Rl with adelay (310 msec) that exceeded the basic cyclelength (300 msec). Total unresponsiveness in allthree electrograms developed in this experiment by9 minutes of ischemia. Although no arrhythmiasoccurred in the records shown in Figure 5, ectopicactivity reached a maximum of 33 PVC's/min be-tween 4 to 5 minutes of ischemia.

In six dogs, brief periods of atrial arrest wereinduced by vagal stimulation during the first 30minutes following latex injection to determine theeffects of heart rate on epicardial conduction delay.Figure 7 shows records from a typical experimentin which vagally induced slowing of heart rate im-proved conduction. The records show a surfaceECG (lead II) and three local electrograms recordedfrom the ischemic zone (Rl, R2, R3). Panel A showsa control cycle during atrial pacing (cycle length333 msec). The electrograms in panel B, recordedat 6 minutes of ischemia, show marked fractionation(Rl, R3), 2:1 alternation (R2), and conduction delayof 150 msec (R2). A 2-second period of vagal stim-ulation was instituted at the arrow in Panel B.Panel C shows the first six recovery beats aftertermination of vagal stimulation. In the first recov-ery beat, all of the electrograms showed an in-creased amplitude, less fractionation, and a de-creased conduction delay (40-msec decrease in R2).However, there was a progressive increase in theamount of fractionation and conduction delay in allthe electrograms during the first four cycles of panelC, with little further change occurring between the4th and 6th cycles. Although the conduction delayin R2 returned to 150 msec by the 4th cycle, the 2:1 alternation observed prior to vagal stimulation(panel B) required 10 seconds (30 cycles) after thetermination of vagal stimulation to reappear. Dur-ing the first 10-15 minutes of ischemia, there was amarked improvement in amplitude and delay ofischemic zone electrograms following brief periodsof vagally induced atrial arrest in all six animals.However, by 30 minutes of ischemia, vagal stimu-lation had no effect on regions of the ischemic zonewhich showed no intrinsic deflection.

DiscussionOne of the purposes of the present study was to

develop a model of myocardial infarction in whichcollateral blood flow to the ischemic myocardiumwas eliminated. The intracoronary injection of vinyllatex has proven to be an effective method of inter-rupting collateral blood flow to the sinoatrial nodalregion (Loeb et al., 1980). In the present study, vinyllatex was injected into the proximal left anteriordescending coronary artery. Although we did not

quantify collateral blood flow to the ischemic my-ocardium, we obtained qualitative evidence for theinterruption of collateral blood flow. The diameterof functional collateral vessels has been shown tobe greater than 40 jim (Blumgart et al., 1950). Inthe present study, latex entered distal arteriolesbranching from the LAD down to a diameter of 20-30 fan. Thus, it is highly likely that latex also filledthe collateral vasculature between the left anteriordescending coronary bed and the other coronaryvascular beds. Furthermore, we injected indocy-anine green into the left main coronary artery toevaluate the patency of collateral channels. Thefailure of green dye to appear in the zone of distri-bution of latex provides indirect evidence for thetotal embolization of collateral channels. Finally,the massive transmural infarcts which corre-sponded precisely to the area of latex distributionindicate that any overlapping flow from other vas-cular beds was insufficient to prevent necrosis.

In spite of the fact that latex appeared to inter-rupt collateral blood flow to the ischemic myocar-dium, the ischemic zone still must have receivedsome residual perfusion. In a condition of zero per-fusion, it would be impossible to completely removelactate dehydrogenase from the ischemic myocar-dium and thus, p-NBT staining would not differ-entiate infarcted from normal muscle. In two exper-iments, a state of zero perfusion was created artifi-cially by soaking excised hearts in Ringer's solutionfor 6 hours prior to incubation with p-NBT. Thezero perfusion period did not alter staining of thesehearts, indicating little if any loss of lactate dehy-drogenase. We were unable to determine the sourceof the residual blood flow responsible for removinglactate dehydrogenase from the latex-injectedhearts. It has been demonstrated that arterial bloodmay reach all layers of the ventricular myocardiumthrough direct vasculature communications be-tween the left ventricular cavity and the coronarycirculation (Moir, 1969). Presumably, the only vas-cular channels not blocked by latex in the presentstudy would have been the Thebesian veins whichconnect the cardiac chambers with the coronaryvenous system (Bloor and Liebow, 1965).

Conduction Abnormalities and Arrhythmias

Experimental arrhythmias induced by one-stageocclusion of the left anterior descending coronaryartery have been classified into an early (0-20 min)and a late (16-72 hours) period of ectopic activity(Harris, 1950). Ventricular fibrillation, when pres-ent, is confined exclusively to the early period (Har-ris and Rojas, 1943). Scherlag et al. (1974) demon-strated that the early occlusive arrhythmias wereenhanced by rapid atrial pacing and abohshed byvagally induced atrial arrest These rate-relatedarrhythmias were associated with diminution anddelay of epicardial activation in the ischemic zone.Several studies have confirmed the relationshipbetween fractionation and delay of epicardial po-


Dow

nloaded from



tentials and the occurrence of early ventriculararrhythmias (Waldo and Kaiser, 1973; Boineau andCox, 1973; Downar et al., 1977; Elharrar et al., 1977;Kaplinsky et al., 1979; Murdock et al., 1980). Fur-thermore, there appears to be a quantitative rela-tionship between the magnitude of subepicardialconduction delay and the severity of early occlusivearrhythmias (Hope et al., 1974; Elharrar et al., 1977;Kaplinsky et al., 1979).

Although all of the determinants of the extentand time course of delayed activation of the is-chemic zone have not yet been elucidated, the se-verity of ischemia appears to be a major determi-nant of the degree of epicardial conduction delay(Elharrar et al., 1977; Ruffy et al., 1979). In thepresent study, the interruption of collateral bloodflow by the intracoronary injection of latex resultedin more than a 2-fold increase in epicardial conduc-tion delay (Fig. 4; Table 1). Furthermore, the in-crease in conduction delay following embolizationwas of sufficient magnitude to result in epicardialactivation occurring after the T-wave of the ECGin 87% of the latex-injected hearts (Fig. 4, 6, and 7).The delays extending beyond the refractory periodof normal myocardium were observed during bothperiods of stable atrial rhythm and of ectopic activ-ity. Although some of the delayed wavefronts mayhave caused reentrant excitation of the normal my-ocardium, delayed activation of the ischemic zonefollowing the T-wave did not guarantee that reen-tery would occur. Because of the large size of theischemic area and the severe depression of conduc-tion, it is possible that many of the delayed wave-fronts were extinguished before they could propa-gate into normal myocardium.

In contrast to the dramatic effect on epicardialdelays, the interruption of collateral blood flow hadonly a minimal influence on the timing of endocar-dial activation (Fig. 4). A differential effect of is-chemia on subepicardial and subendocardial delayhas been reported previously (Cox et al., 1973;Scherlag et al., 1974; Ruffy et al., 1979). Subepicar-dial delay has been shown to exceed subendocardialdelay even when there is a comparable reduction inregional myocardial blood flow (Ruffy et al., 1979).The preservation of endocardial conduction in thepresence of severe ischemia may depend upon su-perfusion of endocardial fibers with cavitary blood(Lazzara et al., 1973) or electrotonic interactionswith ischemia-resistant Purkinje fibers (Gilmour etal., 1980).

In addition to the degree of ischemic insult, heartrate also appears to be a major determinant of themagnitude of epicardial conduction delay (Scherlaget al., 1974; Elharrar et al., 1977; Downar et al.,1977). The rate-dependency of epicardial conduc-tion abnormalities may be mediated by a prolon-gation of the refractory period of ischemic cellsrelative to their action potential durations (Lazzaraet al., 1975; Downar et al., 1977). In the presentstudy, we found a marked improvement in the

amplitude and delay of ischemic zone electrogramsfollowing brief periods (2-3 seconds) of vagally in-duced atrial arrest (Fig. 7). The amount of improve-ment depended upon the severity of the ischemic-induced depression of conduction. By 30 minutes ofischemia induced by coronary embolization, noelectrical activity could be recorded from the ma-jority of the ischemic zone, and slowing the heartrate did not restore epicardial activation.

The absence of an intrinsic deflection in theelectrograms recorded from a major portion of theischemic subepicardium 30 minutes after coronaryembolization indicated a severe depression of elec-trical responsiveness of the epicardial fibers. Withthe recording techniques employed in the presentstudy, it was impossible to determine if the subep-icardial fibers were totally unresponsive. It is likelythat low-amplitude, slowly propagating wavefrontswould not have been detected by the close bipolarelectrodes (2 mm) and high pass filter (40 Hz).However, we also observed that 30 mA stimulidelivered to multiple sites within the central is-chemic zone failed to result in a propagated ven-tricular response. These findings indicate that thedepression of responsiveness in the ischemic sub-epicardium was sufficient to prevent any localwavefronts from propagating to the normal myo-cardium.

In addition to increasing the magnitude of epi-cardial conduction delay, interruption of collateralblood flow increased the number of dogs that de-veloped ectopic beats during the initial 10 minutesof ischemia (Table 1). However, the most significantfinding in the present experiments was the absenceof ventricular fibrillation during ischemia inducedby latex embolization. This finding is somewhatsurprising in view of the massive transmural infarctscreated by the intracoronary injection of latex (Fig.2). It has been reported that canine hearts, whichhave few collateral channels demonstrable by cor-onary angiography, develop a higher incidence ofventricular fibrillation following acute coronary oc-clusion than hearts with multiple collateral chan-nels (Meesman et al., 1970). Furthermore, porcinehearts, which have only a minimally functioningcoronary collateral circulation (Schaper, 1971; Fe-dor et al., 1978; Most et al., 1978), show a 75%incidence of ventricular fibrillation following acuteocclusion of the proximal left anterior descendingcoronary artery (Downar et al., 1977). However,collateral blood flow to the ischemic porcine myo-cardium has been shown to be distributed in a non-uniform manner (Most et al., 1978). In the presentstudy, it is likely that the intracoronary injection oflatex resulted in a relatively uniform reduction incollateral blood flow to the ischemic myocardium.Thus, the tendency for spontaneous ventricular fi-brillation to develop during acute myocardial is-chemia may be related to both the absolute mag-nitude and the heterogeneity of distribution of cor-onary collateral blood flow.


Dow

nloaded from


CORONARY EMBOLIZATION, CONDUCTION DELAY, AKRHYTHMIAS/Eu^r et al 107

The disparate effects of coronary embolizationon the incidence of PVC's and ventricular fibrilla-tion (Table 1) suggest that the mechanism respon-sible for the genesis of ventricular fibrillation duringacute ischemia may not be the same as the mech-anism responsible for the initiation of prematurebeats. A similar conclusion was reached by Janse etal. (1980) in coronary occlusion experiments onisolated porcine and canine hearts. In these studies,extensive epicardial and intramural mapping indi-cated that Purkinje fiber activation, recorded fromthe normal or ischemic zone, consistently precededmyocardial activation during single PVC's or theinitial beats of ventricular tachycardia. There wasno evidence for reentrant circuits within the is-chemic myocardium during these non-lethal ar-rhythmias. However, the development of ventricu-lar fibrillation depended upon multiple reentrantcircuits following tortuous pathways through theischemic myocardium (Janse et aL, 1980). The de-velopment of multiple reentrant pathways throughthe ischemic myocardium presumably dependsupon the existence of multiple unidirectional blocks.The establishment of these conduction blocks maydepend upon a non-uniform depression of excitabil-ity and conduction among adjacent fibers withinthe ischemic myocardium. In the present experi-ments, it is possible that the depression of electricalresponsiveness of adjacent fibers was not suffi-ciently heterogeneous to allow the formation ofenough reentrant circuits to initiate ventricular fi-brillation.

The absence of ventricular fibrillation in the pres-ence of pronounced fractionation and delay of epi-cardial activation suggests that the magnitude ofepicardial delay has only limited value in predictingthe severity of early ventricular arrhythmias. Inaddition to epicardial delay, previous studies haveshown a coincidental relationship between the ap-pearance of alternation in the duration and ampli-tude of ischemic transmembrane action potentialsand subsequent occurrence of ventricular fibrilla-tion (Downar et al., 1977; Russell et al., 1979).Although we did not record transmembrane actionpotentials in the present study, the alternation ofelectrograms frequently observed in the ischemiczone of latex-injected hearts (Fig. 5 and 6; Table 2)probably reflects alternation of transmembraneevents. Thus, we must also conclude that the pres-ence of alternation does not necessarily foreshadowthe appearance of lethal arrhythmias.

In addition to the arrhythmias observed in thefirst 10 minutes after coronary embolization, weobserved a second surge of ectopic activity occur-ring 12-30 minutes after the onset of ischemia.Previous reports have shown a similar period ofdelayed arrhythmias occurring 12-30 minutes aftercoronary ligation (Kaplinsky et aL, 1979; Kaplinskyet aL, 1980). These arrhythmias were found to occurin the absence of marked epicardial fractionationand delay and resulted in a significant incidence

(27%) of ventricular fibrillation (Kaplinsky et al.,1979). It was proposed that the mechanism of thesedelayed arrhythmias might involve microreentrantcircuits at Purkinje-muscle junctions (Kaplinsky etal., 1979). In contrast to the findings of Kaplinskyet al. (1979), we never observed ventricular fibril-lation during the delayed arrhythmic period. Thedelayed arrhythmias observed in the present studyoccurred at a time when the central regions of theischemic zone were electrically unresponsive. Thus,PVC's or ventricular tachycardia arising from amicroreentrant circuit might not be unable to frac-tionate the ischemic myocardium into a sufficientnumber of wavefronts to produce sustained fibril-lation.

ReferencesBishop SP, White FC, Bloor CM (1976) Regional myocardial

blood flow during acute myocardial infarction in the consciousdog. Circ. Res. 38: 429-438

Bloor CM, Liebow AA (1966) Coronary collateral circulation.Am J Cardiol 16: 238-252

Blumgart HL, ZoU PM, Freedburg S, Gilligan DR (1950) Theexperimental production of intercoronary arterial anastomosesand their functional significance. Circulation 1: 10-27

Boineau JP, Coi JL (1973) Slow ventricular activation in acutemyocardial infarction: A source of reentrant premature ven-tricular contractions. Circulation 48: 702-713

Cox JL, Daniel TM, Boineau JP (1973) The electrophysiologictime course of acute myocardial ischemia and the effects ofearly coronary artery reperfusion. Circulation 48: 971-983

Downar E, Janse MJ, Durrer D (1977) The effect of acutecoronary artery occlusion on subepicardial transmembranepotentials in the intact porcine heart. Circulation 58: 217-224

Elharrar V, Gaum WE, Zipes DP (1977) Effect of drugs onconduction delay and incidence of ventricular arrhythmiasinduced by acute coronary occlusion in dogs. Am J Cardiol39: 544-549

Fedor JM, Mclntosh DM, Rembert JC, Greenfield JC Jr (1978)Coronary and transmural myocardial blood flow responses inawake domestic pigs. Am J Physiol 235: H435-H444

Gilmour RF, Zipes DP (1980) Different electrophysiologicalresponses of canine endocardium and epicardium to combinedhyperkalemia, hypoxia, and acidosis. Circ Res 46: 814-825

Hams AS (1950) Delayed development of ventricular ectopicrhythms following experimental coronary occlusion. Circula-tion 1: 1318-1328

Harris AS, Rojas AG (1943) The initiation of ventricular fibril-lation due to coronary occlusion. Eip Med Surg 1: 105-122

Hope RR, Williams DO, El-Sherif N, Lazzara R, Scherlag BJ(1974) The efficacy of antiarrhythmic agents during acutemyocardial ischemia and the role of heart rate. Circulation 50:507-514

Janse MJ, van CapeUe FJL, Morsink H, Kleber A, Wilms-Schopman F, Cardinal R, D'Alnoncourt CN, Durrer D (1980)Flow of "injury" current and patterns of excitation duringearly ventricular arrhythmias in acute regional myocardialischemia in isolated porcine and canine hearts: Evidence fortwo different arrhythmogenic mechanisms. Circ Res 47: 151-165

Jugdutt BA, Hutchins GM, Bulkley BH, Becker LC (1979)Myocardial infarction in the conscious dog: Three-dimensionalmapping of infarct, collateral flow, and region at risk. Circu-lation 60: 1141-1150

Kaplinsky E, Ogawa S, Balke W, Dreifus L (1979) Two periodsof early ventricular arrhythmia in the canine acute myocardialinfarction model. Circulation 60: 397 t̂O3

Kaplinsky E, Ogawa S, Kmetzo J, Balke W, Dreifus LS (1980)Intramyocardial activation in early ventricular arrhythmiasfollowing coronary artery ligation. J Electrocardtol 13: 1-6


Dow

nloaded from



Lazzara R, El-Sherif N, Scherlag BJ (1973) Electrophysiologicalproperties of canine Purkinje cells in one-day-old myocardialinfarction. Circulation 33: 722-734

Lazzara R, El-Sherif N, Scherlag BJ (1975) Disorders of cellulareleetrophysiology produced by ischemia of the canine Hisbundle. Circ Res 36: 444-453

Loeb JM, Euler DE, Randall WC, Moran JF, Brynjolfeson G(1980) Cardiac arrhythmias after chronic embolization of thesinus node artery: Alterations in parasympathetic pacemakercontrol. Circulation 61: 192-198

Meesman W, Schulz FW, Schley G, Adolphsen P (1970) Uber-lebensquote nach akuten eiperimentellem Coronarverachlussin Abhangigkeit von Spontankollarteralen des Herzens. Z Ges-amtc Exp Med 153: 246-264

Moir TW (1969) Study of luminal coronary collateral circulationin the beating canine heart. Circ Res 24: 735-744

Most AS, Williams DO, Millard RW (1978) Acute coronaryocclusion in the pig: Effect of nitroglycerin on regional myo-cardial blood flow. Am J Cardiol 42: 947-953

Murdock DK, Loeb JM, Euler DE, Randall WC (1980) Electro-physiology of coronary reperfusion: A mechanism for reper-fusion arrhythmias. Circulation 61: 175-182

Kivas F, Cobb FR, Bache RJ, Greenfield JC Jr (1976) Relation-ship blood flow to ischemic regions and extent of myocardialinfarction: Serial measurement of blood flow to ischemic re-gions in dogs. Circ Res 38: 439-447

Ruffy R, Lovelace DE, Mueller TM, Knoebel SB, Zipes DP(1979) Relationship between changes in left ventricular bipolarelectrograms and regional myocardial blood flow during acutecoronary artery occlusion in the dog. Circ Res 45: 764-770

Russell DC, Smith HJ, Oliver MF (1979) Transmembrane po-tential changes and ventricular fibrillation during repetitivemyocardial ischemia in the dog. Heart J 42: 88-96

Schaper W (1971) Collateral circulation of the heart. Amster-dam, North-Holland, pp 156-159

Schaper W, Frenzel H, Hort W, Winkler B (1979) Experimentalcoronary artery occlusion. II. Spatial and temporal evaluationof infarcts in the dog heart. Basic Res Cardiol 74: 233-239

Scherlag BJ, Helfant FH, Haft JA, Damato AN (1970) Electro-physiology underlying ventricular arrhythmias due to coro-nary ligation. Am J Physiol 219: 1665-1671

Scherlag BJ, El-Sherif N, Hope R, Lazzara R (1974) Character-ization and localization of ventricular arrhythmias resultingfrom myocardial ischemia and infarction. Circ Res 35: 372-383

Waldo AL, Kaiser GA (1973) A study of ventricular arrhythmiasassociated with acute myocardial infarction in the canineheart. Circulation 47: 1222-1228

Williams DO, Scherlag BF, Hope RR, El-Sherif N, Lazzara R(1974) The pathophysiology of malignant ventricular arrhyth-mias during acute myocardial infarction. Circulation 50:1163-1172

Negative Diastolic Pressure in the IntactCanine Right Ventricle

Evidence of Diastolic Suction

HANI N. SABBAH AND PAUL D. STEIN

SUMMARY To determine whether the canine right ventricle (RV) can develop a negative diastolicpressure indicative of suction, RV pressure was measured in 15 dogs, with catheter-tip mlcroma-nometers. Six dogs were studied only with the chest closed. In these dogs intrapleural pressure wasmeasured medially (near the heart) in four and laterally in two. In nine dogs, RV pressure wasevaluated with the chest closed and after the chest had been opened. In all dogs, with the cheat closed,minimal RV diastolic pressure during expiration was negative, —4,8 ± 0.3 mm Hg. The lowest diastolicpressures occurred during early diastole. Intrapleural pressure during expiration was never this low.Intrapleural pressure measured medially in four dogs was positive during expiration (1.0 ± 0.6 mmHg). In the two dogs in which it was measured in the lateral pleura! space, it was somewhat negativeduring expiration (—2.5 and —1.3 mm Hg, respectively). After the chest had been opened (nine dogs)minimal RV pressure during early diastole was negative in six dogs and positive in three (range: —1.8to 0.8 mm Hg). These results indicate that the negative RV diastolic pressure during expiration did notresult from a negative intrathoracic pressure. It appears that the RV during early diastole can createa sucking effect which may contribute to the filling process. Circ Res 49: 108-113, 1981

A NEGATIVE pressure is often measured in thehuman right ventricle during early diastole (Steinet al, 1980a; Sabbah et al, 1980a), and is generallythought to reflect a negative intrathoracic pressure

From the Department of Medicine (Division of Cardiovascular Med-icine), and Department of Surgery, Henry Fond Hospital, Detroit, Mich-igan.

Supported in pan by U.S. Public Health Service-National Heart,Lung, and Blood Institute Grant HL23669-02.

Address for reprints Paul D. Stein, M.D., Henry Ford Hospital, 2799West Grand Boulevard, Detroit, Michigan 48202.

Received October 16, 1980; accepted for publication January 14,1981.

rather than diastolic suction, although the latterhas been suggested (Sabbah et al., 1980a). If ven-tricular diastolic suction exists, the implication isthat ventricular filling in early diastole may beassisted by a process of active ventricular relaxation(Katz, 1930; Rushmer et aL, 1953; Brecher, 1956).In general, it is thought that the ventricle is filledonly due to the pressure in the atrium during dias-tole and by atrial contraction. Studies in animalshave demonstrated that the left ventricle is capableof generating a negative diastolic pressure under


Dow

nloaded from


D E Euler, C E Prood, J F Spear and E N Moorearrhythmias.

embolization of a coronary artery with latex: effects on conduction delay and ventricular The interruption of collateral blood flow to the ischemic canine myocardium by

Print ISSN: 0009-7330. Online ISSN: 1524-4571 Copyright © 1981 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.49.1.971981;49:97-108Circ Res.

http://circres.ahajournals.org/content/49/1/97.citationWorld 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:


Dow

nloaded from

http://circres.ahajournals.org/content/49/1/97.citation

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

http://www.lww.com/reprints

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


97 the interruption of collateral blood flow to the...

Documents