The role of catecholamines in the production of ischaemia-induced ventricular arrhythmias in the

13 Pages·2006·1.96 MB·English

Br. J. Pharmac. (1986), 87, 265-277
The role ofcatecholamines in the production of
ischaemia-induced ventricular arrhythmias in the rat in
vivo and in vitro
Alan Daugherty', Keith N. Frayn*, William S. Redfern*2 & BrianWoodward
PharmacologyGroup,UniversityofBath,ClavertonDown,BathBA27AYandMRCTraumaUnit*,The
MedicalSchool,UniversityofManchester,OxfordRd, ManchesterM139PT
1 Theroleofcatecholaminesintheproductionofischaemia-inducedventriculararrhythmiasinvivo
and in vitro was studied using coronary artery ligation in the rat.
2 Increases in plasma catecholamine concentrations during coronary artery ligation in pentobar-
bitone-anaesthetized animals were prevented by either acute adrenalectomy or chronic adrenal
demedullation,buttheseproceduresdidnotprotectagainsttheoccurrenceofventriculararrhythmias.
Thus plasma catecholamines were not obligatory mediators ofarrhythmogenesis.
3 Threeprotocolswereused in vitrotoevaluatethepossibleinfluenceofintramyocardialreleaseof
noradrenaline, produced by the local conditions of ischaemia, on the production of ventricular
arrhythmias. Duringcoronary artery ligation in isolated perfused hearts, no enhanced output of3H
could be detected from [3H]-noradrenaline loaded hearts, even in the presence of inhibitors of
catecholamine uptake processes, although washout of lactate from ischaemic regions was readily
demonstrable.
4 Bothopticalisomersofpropranololwereequallyeffectiveinreducingtheincidenceofarrhythmias,
implying a non-specific effect, since the (+)-isomer possesses considerably less P-adrenoceptor
blockingactivity.Theequipotencyofopticalisomersofpropranololcombinedwithalackofeffectof
atenololsuggestedthatarrhythmiaproductionwasnotaconsequenceof,-adrenoceptorstimulation.
5 Thea-adrenoceptor blockersphentolamine andprazosin, bothexertedantiarrhythmic actionsof
similarpotency,butphenoxybenzamineandtrimazosinhadnosignificanteffects.Anevaluationofthe
pharmacological properties of the a-adrenoceptor blockers showed that those drugs which had
demonstrable local anaesthetic properties also exerted significant antiarrhythmic effects. No
relationship was found between potency ofa-adrenoceptor blockade and antiarrhythmic efficacy.
6 The overall conclusion from these multifaceted approaches was that catecholamines were not
necessary mediators ofthe early phase ofventricular arrhythmias in the rat.
Introduction
Disturbancesofthehormonalandautonomicnervous The severity ofventricular arrhythmias produced in
systems havebeenobserved duringtheearlyphaseof the early phase ofacute myocardial ischaemia in the
acute myocardial ischaemia produced by coronary dog has been related to increased plasma concentra-
artery ligation (Gillis, 1971; Ceremuzynski, 1981). tions ofcatecholamines (Ceremuzynski et al., 1969)
Attention has been drawn to the possible role of releasedfromadrenalmedullae(Kelliheretal., 1975).
plasma catecholamine concentrations in the produc- Despite reports on detrimental effects of elevated
tion ofventricular arrhythmias in addition to release plasma catecholamine concentrations, other studies
of noradrenaline within the ischaemic myocardium. havedemonstratedareductionofventriculararrhyth-
mias by increasing plasma concentrations of adren-
'Present address and correspondence: Cardiovascular Div- aline(Marshalletal., 1981)andnoradrenaline(Regan
ision,WashingtonUniversitySchoolofMedicine,660South etal.,1970;Parrattetal., 1981).Thepresentstudyhas
Euclid Avenue, Box 8086, St Louis, MO63110, U.S.A. attemptedtoclarifytheroleofplasmacatecholamines
2Present address: Department ofPhysiology, University of on the early phase of ischaemically-induced ven-
Birmingham B15 2TJ. triculararrhythmias, usingcoronaryarteryligationin
© TheMacmillan Press Ltd 1986
266 A. DAUGHERTY et al.
ratsinwhichplasmaconcentrationsofcatecholamines was extracted with an ion-exchange column and
were greatly reduced by either chronic adrenal de- alumina and theeluatewas used foranalysis. Adren-
medullation or acute adrenalectomy. aline, noradrenaline and dopamine were resolved by
The present study also investigated the possible reverse phase ion-pair high pressure liquid
influence of intramyocardial noradrenaline release, chromatography and the catecholamine concentra-
caused byconditions withinischaemiccardiac tissue, tions were determined by electrochemical detection.
ontheproductionofventriculararrhythmias. Release
of noradrenaline within ischaemic myocardium has Adrenaldemedullation andadrenalectomyprocedures
beenfoundpreviously (Shahabetal., 1972),although AdrenaldemedullationwasperformedonmaleWistar
more recent studies have failed to demonstrate this rats (60-90g) under ether anaesthesia. Rats were
phenomenon (McGrath et al., 1981; Riemersma & given physiological saline to drink during the first
Forfar, 1981). Many investigations have used phar- post-operative week tocompensate forpossible tem-
macologicalmeanstodefinewhethertheintramyocar- porary impairment of mineralocorticoid synthesis.
dial-releaseofcatecholamines hasanarrhythmogenic Thereafter, they were given normal drinking water.
effect. However, it should be emphasized that the Ratswere subjected tocoronaryartery ligationwhen
protective effect noted in some of these studies re- greaterthan250g(5to 6weeksafterdemedullation).
quirescarefulinterpretation, sinceconditionsprevail- Acute adrenalectomy was performed onpentobar-
ing within ischaemic myocardium alter phar- bitone-anaesthetized ratsbeforecannulation andtra-
macologicalprofilesofdrugs,particularlywithrespect cheotomy. Adrenal glands were exposed by lumbar
to enhancement of local anaesthetic properties incisions and removed completely. Rats were sub-
(Vaughan Williams, 1978). To investigate whether sequently prepared for coronary artery ligation, and
noradrenaline is released directly by ischaemia and morethan 30minelapsedfromthetimeofadrenalec-
contributes to the production ofventricular arrhyth- tomy to tightening ofthe ligature.
mias, we have used a paradigm involving coronary
arteryligation intheratisolatedheart. A ratisolated 6-Hydroxydopamine-induced sympathectomy 6-Hy-
heartpreparationhastheadvantageofbeingdevoidof droxydopamine (50mgkg-') was dissolved im-
haemodynamic, hormonal and extrinsic neural in- mediatelypriortoinjectioninsalinecontainingascor-
fluences with could obscure the interpretation of bicacid(10% w/v)andadministeredviaatailveinto
results. Preliminary accounts ofthis work have been ether-anaesthetized chronically adrenal medullated
presentedpreviously(Daugherty&Woodward, 1982; rats.Coronaryarteryligationwasperformed6hlater.
Daugherty et al., 1983). This protocol was chosen because this dose of 6-
hydroxydopamine administered intravenously max-
imally depletes myocardial catecholamine content at
Methods this time (Kostrzew & Jacobowitz, 1974), but
problems ofsupersensitivity ofadrenoceptors would
In vivo studies not have arisen.
Coronaryarteryligation inanaesthetizedrats Coron- Myocardialglycogen levels To assess whether 6-hy-
aryarteryligationwasperformedinanaesthetizedrats droxydopamine mimicked the putative protective
by amethod similar to thatdescribed by Clark et al. effect of reserpine on myocardial glycogen stores, 6
(1980), except that the heart was notexteriorized for ratsreceivedthedrugasdetailedabove.Sixhoursafter
the placement of suture. Arrhythmic activity was injection the rats were killed, hearts were rapidly
recorded for 30min post-ligation. Parameters recor- removed and perfused retrogradely. After 15min of
dedwere: prematureventricularcontractions(PVCs), perfusion, hearts were freeze-clamped using Wollen-
percentage incidence and duration of ventricular berger tongs previously cooled to the temperature of
tachycardia (VT), percentage incidence and duration liquid nitrogen. Tissue was digested in alcoholic
of venricular fibrillation (VF) and percentage mor- potassium hydroxide and glycogen was then hy-
tality attributable to ventricular fibrillation. drolysed and assayed as glucose equivalents.
Approximately 3ml ofarterial bloodwasremoved
by cardiac puncture after 30min ofcoronary artery In vitro studies
ligation. All blood samples were heparinized, cen-
trifuged, and the plasma stored at -20°C until Heart perfusion Hearts were obtained from male
catecholamine determinations were performed. Wistar rats (170-340g) and perfused retrogradely
withmodified Krebs-Henseleitsolution (composition
Determinationofplasmacatecholamines Plasmacate- in mM: CaCl2 2.5, KCI 1.3, MgSO4 1.2, NaCl 118,
cholamine concentrations were determined using the KH2PO4 1.2, NaHCO3 24.9, glucose 11.5) at a con-
method ofFrayn&Maycock (1983). Briefly, plasma stantflowof10mlmin-'.Theperfusatewasprevious-
CATECHOLAMINESANDVENTRICULAR ARRHYTHMIAS 267
ly gassed with 95% 02, 5% CO2 and maintained at polyethylenemini-vialscontaining4mlofscintillation
370C. A ligature was loosely sited around the left fluid(Aqua-Luma, LKB). The samples werecounted
coronary artery at the point were the vessel emerged in a LKB 1215 Rack 1i liquid scintillation counter.
from under the left atrium. Quenching of the samples was estimated using the
Surface electrical records were obtained from thin external standard channels ratio method.
wireelectrodesplacedontherightatriumandapexof
leftventricle. Epicardialelectrogramsweremonitored Perfusate lactate concentrations One ml aliquots of
continuously on a Narco storage oscilloscope and perfusate were collected at intervals indicated on
recorded, as required, on a Lectromed MX 216 Figure3b. Perfusate lactate content was assessed by
recorder.Developedtensionwasassessedatadiastolic the method of Bergmeyer (1965) and expressed as
tensionof2g, usingaDevices UFI transducerwhich nmolml-'g ' wetweightofventricularmyocardium.
wasattachedtotheapexoftheheartviaafinepinand
thread. Recordings of developed tension and heart Quantification of ca-adrenoceptor blocking potency
rate were displayed on a Devices M19 recorder. Assessment of the potency of the a-adrenoceptor
blockersonthemyocardiumprovedtobeimpractical
Protocol for coronary artery ligation in vitro All due to the small effect ofca-adrenoceptor stimulation
heartswerealloweda 15minstabilizationperiodfrom on cardiac function. Therefore, their potencies were
theonsetofperfusion. Onlyheartswhichhadastable determinedinratanococcygeusmuscle,whichhaswell
heart rate and contractile activity at the end of the characterized a-adrenoceptors (Gillespie, 1972).
stabilization period were accepted for study. Follow- Anococcygeus muscles from male Wistar rats were
ingthestabilization period, thedrug underinvestiga- placed in a small organ bath through which Krebs-
tion was perfused continuously, beginning 15min Henseleit solution was perfused. Noradrenaline was
before the coronary artery ligature was tightened. administered via an injection port positioned im-
Ventricular arrhythmic activity was recorded for mediately before the organ bath. Antagonists were
30min post-ligation. administeredintheperfusionfluidandwerealloweda
VTwastakenasarunoffiveormorePVCs.VFwas 15min contact time before the noradrenaline dose-
taken as the appearance ofchaotic electrical activity response curves were reassessed. The method of
withthelossofcontractileactivity. Fiveparametersof Arunlakshana & Schild (1959) was used to calculate
ventriculararrhythmiaswerealwaysquantifiedduring thepA2 for competitive blockers, and the method of
the 30min ofligation: number of PVCs; percentage VanRossum(1963)wasusedtocalculatethepD'2for
incidenceofVT;durationofVT;percentageincidence phenoxybenzamine.
ofVFandduration ofVF. When thea-adrenoceptor
blockers were compared for antiarrhythmic effects, Quantification of local anaesthetic potency Sciatic
local anaesthetic and cx-adrenoceptor blocking nerves were removed from pithed frogs, partially
potency, itwasdesirable to have oneoverall indexof desheathed, and placed in an apparatus containing
antiarrhythmic activity. Therefore, an arrhythmia three chambers. The proximal end ofthe nerve was
score was devised on an arbitrary scale as follows: I suspended across two platinum electrodes and
per 100PVCs, I per 10%VT, Iper l0sVT, Iper 10% stimulated with square wave pulses of0.01ms dura-
VF, I per 100s VF. tionand6V(supramaximal)amplitudedeliveredbya
GrassS8stimulator. Thecentralportionofthenerve
[3H]-noradrenaline overflow during coronary artery wasearthedandperfusedwithfrogRinger.Thedistal
ligation Hearts were perfused for a stabilization portion of the nerve was placed on a platinum
period of 15min and instrumented as described electrode, which was connected to a Tektronix 502A
earlier. [3H]-noradrenaline ((-)-7,8-[3H1-noradren- oscilloscope. After a stabilization period, drugs were
aline, totalof17.5jiCi,8CimM- Amersham Radio- administered cumulatively with a contact time of
,
chemical Centre, diluted in physiological saline) was 30minateachconcentration. The EC50 wastaken as
infused for 10min. A 20min washout period was the concentration of drug which reduced the
allowed to remove radioactivity present in the inter- amplitude of the action potential to 50% of its
stitial fluid, with Imin samplescollected every 5min. maximum value.
After the 20min washout period, the perfusate was
collected continuously at Imin intervals for the re- Statisticalanalysis
mainderoftheexperiment, asillustratedinFigure3a.
In experiments in which an uptake inhibitor was All values are represented as mean± s.e.mean or
presentintheperfusate,itwasallowedacontacttime percentage incidence. x2analysis was used to test the
of15minpriortoligation,asmentionedearlierforall data on the incidence ofVT, VF and mortality, with
other drugs. the inclusion of the Yates correction factor for the
Samples of perfusate (Iml) were transferred to special case of a 2 x 2 contingency table. The non-
268 A. DAUGHERTY etal.
parametric Wilcoxon Rank Sum test was used to contrastoftheinvitroresultspresentedlater,theonset
compare drug-treated values, with those ofcontrols, of PVCs in our in vivo study was more delayed
ofthe numberofPVCs, duration ofVTand VF, and compared to that observed by other workers
plasma catecholamine concentrations. To test for (Figure 1a).
statistical significance of changes in mean arterial Neither chronic adrenal demedullation nor acute
blood pressure, heart rate and glycogen levels adrenalectomy significantly affected any of the
Student's unpaired t test (two-tailed) was used. In all parameters of ventricular arrhythmias (Table 1). It
tests P<0.05 was considered to be statistically sig- may be argued that the adrenal demedullated group
nificant. had a reduced incidence of VF and mortality.
However, these differences did not attain statistical
Drugs used significance (X2 value of0.048 and 0.099 respectively
forfibrillation and mortality) and the most pertinent
Drugs used were: atenolol, (-)-propranolol hydro- feature of this surgically-treated group was that
chloride, (+)-propranolol hydrochloride (ICI), de- fibrillation was indeed present.
sipramine hydrochloride (Geigy), phentolamine Changes in heart rate (Chadda et al., 1974) and
mesylate (Ciba), lignocaine hydrochloride (Astra), arterialbloodpressure(Verrieretal., 1974)havebeen
noradrenaline hydrogen tartrate, normetanephrine shown to influence the incidence ofventricular arr-
hydrochloride, quinidine hydrochloride, 6-hydrox- hythmias in the early phase of acute myocardial
ydopamine hydrochloride (Sigma), phenoxyben- ischaemia, but fortunately neitherdemedullation nor
zaminehydrochloride(SKF),prazosinhydrochloride, adrenalectomy significantly affected theseparameters
trimazosin hydrochloride (Pfizer). (Table2).
With the exception of prazosin, trimazosin and Concomitantchemicalsympathectomyandchronic
phenoxybenzamine, all drugs were dissolved in adrenal demedullation abolished the occurrence of
physiological saline. Prazosin and trimazosin were ventricular fibrillation and significantly reduced the
dissolved in dimethylacetamide for all experimental incidence of the other less severe arrhythmias
procedures. Phenoxybenzamine was dissolved in (Table 1). However, this group had appreciable
dimethylacetamidefortheestimationoflocalanaesth- haemodynamic changeswith a significantdecreasein
eticand x-adrenoceptorblockingpotency,andin70% meanarterial bloodpressure(P<0.01)accompanied
v/v ethanol in experiments involving coronary artery by small pulse pressures. Heart rates were reduced,
ligation. although not significantly (Table2). Biochemical
changes were also noted in the myocardium of this
group. Glycogen content in a control group was
Results 15.5 ± I.Opmol glucoseg-' wet weight which was
similartothosefoundbyotherworkers(Gaudeletal.,
In vivo studies 1979). Chemical sympathectomy with 6-hydrox-
ydopamine produced a large increase in myocardial
Developmentofventriculararrhythmiasintheanaesth- glycogen content (29.1 ± 4.31 mol glucoseg- wet
etized rat Left coronary artery ligation ofanaesth- weight, P<0.01).
etized rats produced arrhythmias which were similar
inincidence ofVT, VF, and mortality tothosefound Plasma catecholamine concentrations during coronary
previously (Clark et al., 1980; Marshall et al., 1981) artery ligation Concentrations of plasma cate-
(Table 1). However, as it is of importance for the cholamines were measured in pentobarbitone-anaes-
Table1 Incidence ofventricular arrhythmias during 30min ofcoronaryartery ligation inanaesthetized rats
n PVCs % VT VTduration % VF VFduration % mortality
Control 10 633±170 100 44± 16 60 394± 233 20
Adrenal demedullation 9 546± 196 100 29± 10 44 361 ± 323 11
Adrenalectomy 10 660± 214 100 34± 15 60 640± 274 30
Adrenal demedullation + 8 173± 73* 75 8± 3* 0 0
6-hydroxydopamine
PVCs=premature ventricular contractions; VT=ventricular tachycardia; VF=ventricular fibrillation. Chronic
adrenal demedullation or acute adrenalectomy did not significantly affect any parameter, but the combination of
adrenaldemedullationwith6-hydroxydopaminetreatment(50mgkg-'i.v.)wasprotective.Valuesarerepresentedas
mean±s.e.meanand significantdifferences fromcontrols aredenoted *P<0.05.
CATECHOLAMINES AND VENTRICULAR ARRHYTHMIAS 269
a a (7)
500 - s 100 T
(9) I~
400 - 0 50
(6)
300 - < OL (9) (10)
b
200 F (9) -- 50 -
(10)
(8) -F
E 251-
100 _ (10) Q-
(8) I-
(10)((lo)(0) (8) (8) (8) (8) z
L- o
0
b
400 r c
E f 10
C,, -~
0(u)
E 5
sI
300
0..
Control CAL CAL CAL
demedullation adrenalectomy
200 Figure2 Plasmaconcentrationsof(a)adrenaline(Ad),
(b) noradrenaline (NA) and (c) dopamine (DA) after
coronary artery ligation (CAL) in anaesthetized rats.
Groups are those subjected to CAL alone and those
(9) which were either adrenal demedullated or adrenalec-
100 tomizedinadditiontoCAL.Plasmaconcentrationswere
alsodeterminedin nentobarbitone-anaesthetizedanimals
(9) (9) (9) ..ala r- v/_,-_ass
W) T T which had not been subjected to CAL to assess basal
.J f1 lIlevels(denotedas'control').Numberofdeterminationsis
represented in parentheses.
0
30
Timeafterligation(min) centrations were similar to those seen in the control
Figure 1 (a) The time course of the occurrence of group(Figure2).Inparticular,thetworatswhichdied
premature ventricular contractions (PVCs) during the from ventricular fibrillation in this group had very
30min ofcoronary artery ligation in vivo inthecontrol high concentrations ofadrenaline and noradrenaline
group ofpentobarbitone-anaesthetized rats. Numberof (232 and 139pmolmlP'; 84 and 48pmolml-',
observationsisrepresentedinparentheses anddecreases respectively.
as rats were excluded due to death from ventricular In the groups ofadrenal demedullated and adren-
fibrillation(VF)(b)Thetimecourseoftheoccurrenceof alectomized rats, plasma adrenaline concentrations
PVCs during the 30min ofcoronary artery ligation in
wereat thelowerlimitsofdetection, while noradren-
vitrointhecontrolgroupofperfusedhearts.Numberof
observationsisrepresentedinparenthesesanddecreaseas aline and dopamine concentrations were not sig-
ratswereexcludeddue to occurrence ofVF. nificantly different from those of the group not
subjected to coronary artery ligation. Particularly
thetized rats which were not subjected to coronary pertinent was that plasma catecholamines in these
artery ligation. Plasma concentrations ofadrenaline surgically-treated groups were low even in rats in
(5.1 ±0.1 pmolml-'), noradrenaline (1.2± 0.3 which persistent VF occurred.
pmolml-')anddopamine(1.1 ± 0.4pmolml ')were
similar to those found by other authors in the rat In vitro studies
(Buhler et al., 1978). Coronary artery ligation alone
producedalargeincreaseinplasmaconcentrationsof Occurrence of ventricular arrhythmias Fibure lb
adrenaline and noradrenaline, while dopamine con- shows the time course of the occurrence of PVCs
270 A. DAUGHERTY etal.
following coronary artery ligation in the isolated a
perfusedheartforcomparisonwithresultsobtainedin 6000
the in vivo study. After an initial quiescent period of E 4000
8mintheincidenceofPVCsincreasedtoapeakafter 0
12minandthendeclinedslightly.Theprolongedruns
ofVTandVFalsohadanonsetwhichwassimilarto -(60 2000 F
that ofthe peak ofPVCs. This onset ofarrhythmias Ligation
C)
was similarto thatseeninthepentobarbitone-anaes- 0)
thetized rat, although in the anaesthetized rat there a0C). F
1000
wasamoredramaticreversiontoanelectricallystable at
condition (Figure la).
During ligation, heart rate remained unchanged I
whiledevelopedtensionwasreducedinthefirst5min \ < I*
andthenremainedconstantfortherestofthe30min 200 * *= * * * * . .
ofobservation.
Ligation
160
Releaseofmetabolitesduringcoronaryarteryligation
I
Afterloadingtheheartswith[3H]-noradrenaline,there 0.0): 120
wasagradualdecreaseintheamountof3Hpresentin 0-
'a)
the perfusate over the following 20min period, re- 0- a 80
"
presenting washout of the interstitial space. In the
controlseriesofexperiments,coronaryarteryligation C; E 40
M
produced a small decrease in 3H overflow which _
stabilizedafterapproximately3min,andwasfollowed 0 ,mlI-Il'T
byaperiodinwhichtheoverflowremainedsimilarfor 20 10 5 0 5 10 15 20 25 30
eachtimeinterval.Hence,noreleasecouldbedetected Time (min)
that was temporally related to the onset or peak of Figure3 (a)Theeffects ofcoronary artery ligationon
arrhythmias (Figure3a). the release of 3H from hearts preloaded with [3H]-
Theseexperimentswererepeatedinthepresenceof noradrenaline in vitro.Thelinerepresentsthemeanof5
an uptake 1 inhibitor, desipramine (1IsM), or an observations;s.e.meanareshownbythehatchedarea.(b)
uptake 1 inhibitor, normetanephrine (3;LM), at con- Effectsofcoronaryarteryligationonlactatereleasefrom
centrationswhichsubstantiallyinhibittheirrespective themyocardiuminvitro.Thelinerepresentsthemeanof5
uptake 2 inhibitor, normetanephrine (3ItM), at con- observations; s.e.mean are shown bythehatched area.
centrationswhichsubstantiallyinhibittheirrespective
overflow was identical to that seen in the absence of coronary artery ligation.
drug(datanotshown).Acontrolseriesofexperiments Despite this inability to detect enhanced release of
wasperformedinwhichheartswereloadedwith[3H]- 3Hduringligation,increasedconcentrationsoflactate
noradrenaline but ischaemia was not induced. The were present in the effluent after occlusion, demon-
slopeforthefinalphaseofthewashoutcurveforthis strating that washout of metabolites from the
controlgroupwasthesame asin groupssubjected to ischaemic region occurred (Figure3b).
Table2 Heart rateandmean arterial bloodpressuremeasured immediately before tighteningthecoronary artery
ligature inanaesthetized rats
Meanarterial
Heartrate bloodpressure
n (beatsmin-') (mmHg)
Control 10 375±10 93± 7
Adrenaldemedullation 9 413± 15 88± 5
Adrenalectomy 10 378± 7 87± 5
Adrenal demedullation+ 8 355± 6 68±2*
6-hydroxydopamine
Adrenaldemedullationoradrenalectomydidnotsignificantlyaffecteitherparameter,althoughadrenaldemedullation
combinedwith6-hydroxydopamineadministration(50mgkg-',i.v.)reducedmeanarterialbloodpressure.Valuesare
represented asmean±s.e.mean andsignificantdifferences fromcontrols aredenoted *P <0.01.
CATECHOLAMINES ANDVENTRICULAR ARRHYTHMIAS 271
Desipramine (1 gM)was foundtohave verypotent hythmic effects ofphentolamine and prazosin, their
antiarrhythmic properties, with only a few PVCs potenciesasa-adrenoceptorblockersandlocalanaes-
occurring during ligation. Normetanephrine (3gM) theticswereassessedandcomparedwithphenoxyben-
alsopossessed someminorantiarrhythmic properties zamine and trimazosin.
as shown by a significant reduction in VF duration pA2 and pD'2 values for blocking potency were
(P<0.05), although it was without effect on other determined on rat anococcygeus muscle using
ventricular arrhythmias (Table3). noradrenaline as the agonist. Values for the pA2 of
prazosin (9.1, n = 5), phentolamine (8.1, n = 8) and
Adrenoceptor blockers on development ofventricular trimazosin (6.8, n = 6) were in good agreement with
arrhythmias The potent P-adrenoceptor blocker, thosedeterminedbyotherworkers(Doxeyetal.,1977;
atenolol (I ,uM; a concentration considerably above Constantine & Hess, 1981). The pD'2 value for
the pA2 for cardiac tissue), failed to modify the phenoxybenzamine of8.2 (n = 7) was also similar to
development ofventricular arrhythmias. In contrast, that found by Doxey et al. (1977) (Figure4c).
the (-)-isomer of propranolol (1I M) reduced the Intheabsence offacilities to record cardiac action
number of PVCs (P<0.01) and duration of VT potentials, measurement oflocal anaesthetic potency
(P<0.01) whilst short runs ofVF were seen in only on frog sciatic nerve was taken as an index ofdirect
two experiments. However, the (+)-isomer of membraneeffects. Althoughthispreparationisnotas
propranolol was equally effective in preventing arr- sensitiveascardiactissuetolocalanaestheticproper-
hythmias (Table3), which suggests that the antiarr- tiesofdrugs, itwasusedinthisstudytocompare the
hythmic effect was not due to ,-adrenoceptor block- relativepotencyofthea-adrenoceptorblockers. EC50
ade,butrathertoanon-specificeffectofpropranolol. ofphentolamine was 0.65mM. Othera-adrenoceptor
Phentolamine and prazosin (10 1M), significantly blockers all required dissolution in dimeth-
reducedthenumberofPVCs(P<0.01), theduration ylacetamide. The volume of dimethylacetamide re-
ofVT (P<0.01) and the incidence ofVF (P<0.01 quired to obtain sufficient concentrations of these
for phentolamine, P<0.05 for prazosin). Although drugs (2ml 100ml') itselfreduced the amplitude of
thesetwodrugswereeffective,twootherax-adrenocep- the action potential by 19% which was taken into
tor blockers, trimazosin (IO M) and phenoxyben- accountinthedeterminationofEC_0.EC,0ofprazosin
zamine (10 tM), did not significantly influence the was0.61mMandhencehadasimilarlocalanaesthetic
development ofarrhythmias (Table3). potencytophentolamine. Neitherphenoxybenzamine
None of the drugs used, at the stated concentra- (n =7) nor trimazosin (n = 5) reduced the action
tions, had any significant effect on heart rate or potential sufficiently for an ECo to be determined
developed tension. (Figure4).
When comparing the antiarrhythmic properties
Pharmacologicalprofileof'a-adrenoceptorblockers To with the a-adrenoceptor blocking and local anaesth-
help establish the possible mechanism ofthe antiarr- etic potencies of phentolamine, prazosin, phenoxy-
Table3 Effectofcatecholamine uptakeinhibitorsandadrenoceptorblockersonearlypost-ligation arrhythmiasinvitro
VTduration VFduration
Treatment Conc. (pM) n PVCs % VT (s) % VF (S)
Control 14 607± 93 86 54± 11 64 563± 134
Desipramine 1 6 20+16** 0** 0*
Normetanephrine 3 6 662±133 100 53± 10 67 65± 32*
Atenolol 1 10 636±110 100 52± 9 80 541 ± 168
(-)-Propranolol 1 10 213± 39** 90 16± 4** 20 13± 6*
(+)-Propranolol 1 9 193 44** 67 14± 3** 22 49± 52*
Phentolamine 1 10 747±127 100 65± 10 60 186± 45
10 12 247± 50** 33 8± 1 0* -
Prazosin 1 10 525 144 100 51 ± 4 50 243± 150
10 9 90± 23** 67 6± 2** 11* 5
Trimazosin 10 10 748 219 90 51 ± 13 30 740± 337
tDimethylacetamide 10 697± 79 100 53± 9 70 333± 146
Phenoxybenzamine 10 11 710 167 55 47± 18 27 610± 113
tEthanol 10 628 203 100 54± 17 60 505± 212
Valuesarerepresentedasmean±s.e.meanandsignificantdifferencesaredenoted*P<0.05,**P<0.01.tSolventfor
prazosin andtrimazosin; Isolvent forphenoxybenzamine.
272 A. DAUGHERTY etal.
a oceptor blockers; a concentration which was con-
40r siderably higher than the pA2 for all of the drugs.
Values calculated were: phentolamine, 6.5, prazosin,
Ea)
I- 9.3, phenoxybenzamine, 26.1, and trimazosin, 31.8.
30
0CA Arrhythmiascoreintheabsenceofanydrugwas32.0.
20I- Whencomparingpharmacologicalpropertiesofthese
drugs (Figure4), there was no relationship between
10I.. antiarrhythmicpropertiesand x-adrenoceptor block-
L:
ing potency. However, comparison between local
0L anaesthetic potency and arrhythmia score was more
striking. Phenoxybenzamine and trimazosin were
b
found to have little antiarrhythmic activity and no
1.0r
detectable local anaesthetic activity. In contrast,
-
prazosin and phentolamine were approximately
equipotent for local anaesthetic and antiarrhythmic
uL)O effects.
EU 0.2 ND. N.D.
Local anaesthesia on development of ventricular arr-
c hythmias Quinidineandlignocainewereexaminedto
l0
testwhetherantiarrhythmicdrugswithlocalanaesth-
N C etic properties were effective in this paradigm of
D 6 ventriculararrhythmias, andtoestablish whetherthis
'I . E
property of the x-adrenoceptor blockers could have
L- 4i conferred the protection seen with these drugs.
Praz Phen Tmz Pbz Atthelowerconcentrationofquinidineused(I ftM),
only the duration of VF was significantly reduced
Figure 4 1Effects of a-adrenoceptor blockers on: (a) (P<0.01) while a higher concentration (1I0 tM), re-
Antiarrhythkmic action represented as an arrhythmia ducedthenumberofPVCs(P<0.01)andincidenceof
scoreinthe1presenceof10jMofthedrug.Thearrhythmia VT (P<0.01) and abolished the incidence of VF.
scoreisexplIainedintheMethodssection.(b)Concentra- Lignocaine was less effective in preventing arrhyth-
tion ofdrulg which reduced the amplitude of the frog mias,with 1pMhavingnosignificanteffects, although
sciaticnerveactionpotentialto50%ofitsmaximalvalue. 10 'mreducedtheincidenceofVT(P<0.01) andVF
(ANd.rDe.no-cleopctao1ranbaleosctkhientgicpeoftfeecntcwyaussinnogtdpeAt2ectaanbdle)p.D'(c2)ao-s (1PO<<0.r0e5)dc(eTatbhlee44)).. NNeeiittheerofdrug ssiiggninfiicfaintclaynadffVeFc-
determinantts. Praz=nrazosin Phen=nhentolamine: ted cardiac function at the concentrations used.
La. ,Isi - =phenoxAybena-mine.
Tmz=trim;azosin; Pbz=phenoxybenzamine.
Discussion
benzamine and trimazosin, it was necessary to have
one index of ventricular arrhythmias from the five In vivo studies
parameters recorded. This scalewas devised only for
graphical representation and would have produced a This study has shown that the production of ven-
similar pattern regardless of the criteria. It was ricular arrhythmias during the early phase of acute
determined in the presence of 101iM ofthe a-adren- myocardial ischaemia in pentobarbitone-anaesth-
Table4 Effects oflignocaine and quinidineonearly post-ligation ventriculararrhythmias in vitro
VTduration VFduration
Treatment Conc. (PuM) n PVCs % VT (s) % VF (s)
Control 14 607± 93 86 54±11 64 563± 134
Lignocaine 1 6 522± 166 83 30± 8 50 347± 56
10 9 330±116 22** 19± 5 11* 9
Quinidine 1 7 555± 122 100 35± 7 42 34± 6**
10 9 230± 75* 22** 10± 8 0**
Values are represented as mean±s.e.mean and significant differences from control values are denoted *P<0.05,
**P<0.01.
CATECHOLAMINES AND VENTRICULAR ARRHYTHMIAS 273
etized rats was not dependent upon elevated concen- considered statistically significant. Therefore, the
trations ofplasma catecholamines. apparent protection of adrenal vein ligation on the
Adirectrelationshipbetweenelevatedplasmacate- productionofventriculararrhythmiasobservedinthe
cholamine concentrations and those ventricular arr- cat may not have been a real effect.
hythmiasoccurringduringtheearlyphaseofcoronary Combinationofchronicadrenaldemedullationand
artery ligation has been suggested (Ceremuzynski et chemical sympathectomy greatly reduced the severity
al., 1969). Plasma catecholamine (primarily adren- ofventricular arrhythmias following coronary artery
aline) concentrations in anaesthetized dogs were in- ligation. A superficial explanation ofthis protection
creased during acute myocardial ischaemia, and a may be that depletion of catecholamines from the
relationship was found between blood catecholamine sympathetic nerve terminals reduced the release of
concentrationsandtheseverityofventriculararrhyth- noradrenaline within the ischaemic myocardium.
mias (Ceremuzynski et al., 1969). The present study However, it is clear that other factors must be taken
also demonstrated high plasma concentrations of intoaccount.Reserpine-inducedcatecholaminedeple-
adrenaline and noradrenaline in rats subjected to tion is known to produce morphological and bio-
coronary artery ligation alone. Very high plasma chemical changes which may subsequently lead to
concentrations ofadrenaline and noradrenaline were protective effects which would not be observed with
observed in two control animals, in this series of catecholamine depletion per se. Protective effects of
experiments, which died from ventricular fibrillation. reserpine-induced catecholamine depletion on the
Thesefindingsareingoodagreementwiththedataof ischaemic myocardium have been attributed to the
Ceremuzynski etal. (1969). These workers suggested enhanced glycogen content ofmyocardium (Scheuer
that elevated plasma catecholamine concentrations &Stezoski, 1970;Gaudeletal., 1979),whichmayhave
caused arrhythmias, but the results produced in the beneficial electrophysiological consequences (Brick-
present studyallowanalternativeinterpretation. The nell & Opie, 1978). In accord with the effects of
fact that acute adrenalectomy or chronic adrenal reserpine, 6-hydroxydopamine produced a large in-
demedullation had no significant effect on the de- crease in myocardial glycogen content. The presence
velopment of ventricular arrhythmias, while ofan increased myocardial content ofglycogen may
dramatically reducing the circulating plasma adren- have reduced the ischaemia-induced action potential
aline concentrations, was indicative that adrenaline duration shortening (Cowan & Vaughan Williams,
released in the control group was probably caused 1980). In addition to these metabolic effects, inter-
reflexly;eitherinresponsetotherapidhaemodynamic pretation ofthe 6-hydroxydopamine-induced reduc-
fluctuationscausedbythearrhythmias,oraspartofa tion in ventricular arrhythmias in vivo mustalso take
general physiological response to myocardial accountofthedifferenthaemodynamicsinthisgroup
ischaemia with no obvious detrimental effects on ofanimals.
development ofarrhythmias. Althoughtheresultsobtainedfromourinvivostudy
The findings that adrenal demedullation or adren- provide evidence that plasma catecholamines are not
alectomy abolished increased concentrations of necessarymediatorsoftheproductionofarrhythmias
plasma catecholamines during coronary artery liga- followingcoronaryarteryligation,theydonotexclude
tion, confirmed the observations of other workers the possibility that catecholamines released locally
(Kelliher et al., 1975), who implicated the adrenal within ischaemic myocardium be important arrhyth-
medullae as the source of elevated plasma cate- mogenicmediators. Hencecoronaryarteryligationin
cholamine concentrations observed during the initial theisolated perfusedratheartwasused toinvestigate
stages ofmyocardial ischaemia. this possibility.
Lack ofa significant effect ofacute adrenalectomy
on thedevelopmentofventriculararrhythmias, while In vitro studies
greatly reducing plasma adrenaline concentrations,
complements the findings in the chronically adrenal Useofisolatedperfusedratheartsforthe investigation
demedullated animals. The absence of a beneficial of ventricular arrhythmias The use of an in vitro
effect ofacute adrenalectomy in the present study is paradigm of coronary artery ligation to investigate
contrary to the protective effect of adrenal vein ventriculararrhythmiasassumedthatthearrhythmias
ligationobservedinthecat(Kelliheretal., 1975). The produced in vivo arose from factors inherent to the
inconsistency may also be attributable to the x- myocardium. Insupportofthisassumption,therewas
chloralose used to anaesthetize the cats compared to a close similarity in the onset ofventricular arrhyth-
pentobarbitoneusedintherat.However,theapparent mias in the present in vitro study compared to that
differencemaybeaconsequenceofstatisticalanalysis. observed in vivo, possiblyindicative that thearrhyth-
Ourreassessment ofthedata ofKelliheretal. (1975) mias are initiated by mechanisms within the myocar-
using x2 analysis with the Yates correction factor dium. Although the onset of arrhythmias was very
yields a x2 value of 2.272, which would not be similar in both models, following the peak of ven-
274 A. DAUGHERTYetal.
tricular arrhythmic activity in vitro, hearts did not In themodel used in thepresent study there was a
revert to a stable condition as rapidly as in vivo. The residual flow through at least part of the ischaemic
reason for this difference is unknown, but an extra- regiontofacilitatewashoutofnoradrenalineifrelease
cardiac influence would be implicated from these had occurred, as demonstrated by the presence of
results. lactate in the effluent. Even in the presence of de-
Since the present study attempted to clarify the sipramine (uptake I inhibitor) or normetanephrine
involvement ofthe adrenergic nervous system in the (uptake 2 inhibitor) at concentrations producing
production ofventricular arrhythmias, it was neces- markedinhibitionofuptakeprocesses(Iversen, 1965),
sary to postulate that the proposed involvement of there was no demonstrable enhanced release of 3H
adrenergic mechanisms in the production of ven- duringischaemia.Therefore,thelackofdetectable3H
tricular arrhythmias was due to intramyocardial re- releaseduringischaemia,inthepresenceofinhibitors
lease ofcatecholamines from neuronal stores due to ofthecatecholamine uptakeprocessandwhereother
factors within ischaemic myocardium such as hyper- metabolites produced by the ischaemic myocardium
kalemia and acidosis. This was considered as a are readily demonstrable, does not support the
reasonable postulate for this in vitro study because, speculation that the local conditions of ischaemia
although enhanced sympathetic outflow has been produce noradrenaline release within the myocar-
demonstrated during the early phase of conronary dium.
artery ligation in vivo (Gillis, 1971), both acute Althoughthereisagreementonthelackofdemon-
sympatheticdenervationinthedog(Ebertetal., 1970) strable intramyocardial noradrenaline release during
andganglionblockadewithchlorisond-amineintherat the early phase of ischaemia from in vitro studies
(Abrahamsson et al., 1982) do not influence the (Rochette et al., 1980), this consistency has not been
incidence of ischaemia-induced ventricular arrhyth- found by other workers during coronary artery liga-
mias. Both of these procedures remove effects of tion studies performed in vivo. Earlier studies in
extrinsic neural activity during ischaemia, but were anaesthetizedopen-chesteddogsdemonstratedrelease
ineffectiveatalteringelectricalstabilityofthemyocar- ofcatecholamines after 2min (Shahab et al., 1972),
dium. Both procedures would not affect the possible 15min(Dutta&Booker, 1970)and 1 h(Lammerantet
release of catecholamines due to local conditions al., 1966)ofcoronaryarteryligation.Recentstudiesin
within ischaemic myocardium. dogs,usingmoreprecisemethodsfordeterminationof
catecholamines have uniformly failed to detect
Releaseofcatecholaminesfromischaemicmyocardium elevated concentrations ofnoradrenaline in the local
Toinvestigatereleaseofnoradrenalinefromischaemic coronary vein from the ischaemic myocardium after
myocardium, weemployedthewidelyusedmethodof 10min(McGrathetal., 1981), 15min (Riemersmaet
prelabelling neuronal pools with [3H]-noradrenaline. al., 1981) of coronary occlusion, although all des-
This technique labels noradrenaline pools which are cribed increased plasma concentrations of lactate
readily releasable bynerve stimulation and indirectly which would have leaked from the ischaemic zone.
acting sympathomimetics. It is probable that this Priortothepresentinvitrostudy,onlytwoprevious
readily releasablepoolwould bethemost susceptible investigationswhichwerebothinvivo,haveexamined
toreleasebyischaemia,ifthisphenomenonoccurred. directly the relationship between possible in-
The specific radioactivity of the tritiated noradren- tramyocardial noradrenaline release and the produc-
alineusedwasveryhigh(8CimM -')andconsequently tion ofventricular arrhythmias in the early phase of
onlyasmallenhancementofreleasewouldhavebeen coronary artery ligation and again the results are
detectable. The present study failed to show that conflicting. Hirche et al. (1980) found a release of
ischaemia produced an enhancement of 3H release noradrenaline into a local coronary vein of an
fromperfusedhearts'whichwerepreloadedwith[3H]- ischaemic region after 3min in the anaesthetized pig
noradrenaline. This lack of enhanced overflow whichcoincidedwiththeoccurrenceofthefirstphase
occurred even though a considerable amount of of ventricular arrhythmias. However, in agreement
releasable [3H]-noradrenaline was present during the with the present study, Marshall & Parratt (1980)
ischaemicperiod,asdemonstratedduringreperfusion failed to demonstrate such a relationship in the
studies which used a similar protocol to that ofthe anaesthetizeddog.Thereasonforthesecontradictory
present study (Rochette et al., 1980). Besides the data is not readily apparent, but demonstrates that
interpretation that catecholamines are not released there is considerable doubt whether noradrenaline is
during ischaemia, it may be hypothesized that any released from the ischaemic myocardium, contribut-
liberatednoradrenalinewouldnothavebeendetected ing to the production ofventricular arrhythmias.
due to a combination of the low flow from the During experiments to determine 3H overflow,
ischaemic myocardium and the involvement ofcate- antiarrhythmic effects ofboth desipramine and nor-
cholamine uptake processes, although this is con- metanephrine were noted (Table3). The effects were
sidered unlikely for reasons discussed below. particularly profound in the case of desipramine,