The Pumping Ability of the Left Heart and the Effect of Coronary Occlusion

Transcription

1 97 The Pumping Ability f the Left Heart and the Effect f Crnary Occlusin Gus ELZNGA, M.D., PH.D., AND NCOLAAS WESTERHOF, PH.D. SUMMARY n an islated preparatin f cat heart we studied the pumping capacity f the left heart while left atrial filling pressure was kept cnstant. We used the surce impedance cncept t quantify the pumping capacity. n this surce impedance cncept the relatin between left ventricular utput and left ventricular pressure is given by the frmula Z. (< >)=where Z, = surce impedance, P lv = left ventricular pressure, / = flw in the ascending arta, and ui = ir/, / being frequency. The pressure btained at zer flw is called the hydrmtive pressure (). Only the mean values f pressure and flw were studied. We studied the behavir f 10 hearts in three different experimental situatins and in the fllwing sequence: (1) cntrl cnditins, () after ligating a part f the left crnary arterial system, and (3) after restring left ventricular utput t cntrl level by raising left atrial filling pressure. t was fund that surce resistance was nt significantly different in the three situatins but that mean hydrmtive pressure () was significantly lwer after ligatin f a part f the left crnary arterial system. t was cncluded that the decrease in pumping capacity f the left heart after infarctin can be cmpensated fr almst cmpletely by an increase in left atrial filling pressure. This cmpensating mechanism therefre seems t be very efficient. Dwnladed frm by n Nvember 8, 018 CHARACTERSTCS f fluid pumps ften are given as head capacity curves. n these plts the amunt f fluid that the pump can handle per unit f time is related t the pressure head it has t vercme. The slpe f these pressure-flw plts shws hw "stiff" the pump is. f the pump has n valve and delivers steady flw and pressure, this descriptin is straightfrward. Hwever, when we cnsider the left heart, cmplexity has increased because nw we have t deal with an scillatry pump with valves. We have shwn previusly 1 that in this situatin the understanding f pump behavir may be facilitated by use f an electrical analg as shwn in Figure 1. n this cncept the heart is regarded as a generatr with a cnstant generatr pressure, the s-called hydrmtive pressure (), and a cnstant internal impedance (Z s ). Left ventricular pressure (/*, ) is represented in this analg by the vltage just prximal t the dide (artic valves). The generatr is laded by the input impedance f the arterial system (Z a ). The current thrugh the circuit represents the ttal bld flw (/) thrugh the circulatin. There are tw quantities in this apprach which cannt be measured directly fr the heart, namely, the surce impedance and the generatr pressure. The values f these quantities can be calculated by making use f tw equatins and measuring left ventricular pressure and artic flw. These tw equatins can be btained by tw different settings f the lading impedance (Z a ), using Ohm's law. The equatins are: Subtractin and rearrangement results in: Z,(w) = () u) - /,(«) where > is vf, /being the frequency f the sine wave under study. Frm the Physilgical Labratry, Free University, Amsterdam, The Netherlands. Received August 15, 1975; accepted fr publicatin December 16, (1) () Our present investigatin was made n an islated ejecting cat heart preparatin. We have cncentrated mainly n the zer Herz term f the surce impedance, i.e., the means f left ventricular pressure and artic flw: the surce resistance. This quantity appears t be, within limits, rather independent f the setting f the lading impedance. t was f interest t us t knw in what way the hydrmtive pressure r the surce resistance f the left heart, r bth, change when its pumping actin is altered by ccluding a part f the crnary system. We als determined surce resistance and hydrmtive pressure after we had attempted t restre the functin f the left heart affected by crnary cclusin. This was dne by increasing left atrial filling pressure and, cnsequently, left ventricular end-diastlic vlume. Methds SOLATED EJECTNG CAT HEART PREPARATON Since a mre detailed descriptin f the preparatin is given elsewhere,'- ' 4 nly a shrt descriptin is presented here. Figure shws the experimental arrangement used fr the islated ejecting cat heart. The large reservir (R) cntains an xygenated mixture f Tyrde's slutin and washed bvine erythrcytes. The hematcrit f the mixture is 5, hemglbin cncentratin is 8.6 g/100 ml, ph varies between 7.39 and 7.43, P, between 15 and 300 mm Hg, and PcOj between 5 and 30 mm Hg. The temperature f the fluid, when ffered t the left atrium, is 37.5 ± 0.3 C. Temperature cntrl is achieved by pumping warm water thrugh a cil suspended in the perfusin fluid. The pump is cntrlled by a thermistr that cntinuusly measures the temperature in the reservir R. The perfusin fluid in the reservir R is under pressure and it is frced, via a filtering device (F), t a left atrial supply cntainer (SL). The height f the fluid in SL is cntrlled by an verflw vessel (OL); the verflwing vlume is pumped back frm the small reservir (RL) int

2 CRCULATON RESEARCH 98 FGURE 1 Electrical analg f the arterial part f the systemic circulatin and the left side f the heart. = generatr pressure f hydrmtive pressure; P,v = left ventricular pressure; = the bld flw thrugh the system; Z, = the surce impedance; and Z = the input impedance f the arterial system. The latter system is cmpsed f three elements: Rc, which mdels the characteristic impedance f the first part f the ascending arta; C, which mimics the ttal arterial cmpliance and RD, which is cmparable with the peripheral resistance. Dwnladed frm by n Nvember 8, 018 the large reservir R. The cnnectin f SL with the left atrium is shrt and has a large bre t keep effects f fluid inertia n cardiac filling small. The time needed fr acceleratin f the fluid is less than 50 msec. The left ventricle ejects its cntents int a hydraulic mdel f the input impedance f the systemic arterial tree f the cat. 4 Use f this "lading system" results in patterns f artic flw and pressure which clsely resemble the shapes f thse variables recrded in viv (Fig. 3). The resistr (R c ) is equivalent t the characteristic impedance f the first part f the ascending arta. The capacitance (C), representing the ttal arterial cmpliance, can be changed by changing the air vlume under the pistn. The peripheral resistance (R p ) can be changed by mving a slide that clses the desired number f the many tiny cnduits f which the resistr is cmpsed. The slide is mved by a servmtr. VOL. 38, N. 4, APRL 1976 slated cat hearts were btained frm male cats; the weights f the hearts ranged frm 15.7 t 6. g (n = 10). The cats were anesthetized with sdium thipental (45 mg/kg) administered intraperitneally (ip). Under artificial ventilatin the heart was taken ut f the thrax and cnnected t the input and utput f the system described abve. During the time needed t cnnect the heart t the experimental setup it was kept perfused by the Langendrff technique t cntinue crnary perfusin. When cnnectins were cmplete the left ventricle ejected its cntent int the hydraulic mdel and generated its wn crnary perfusin pressure. The experiments were started abut 0 minutes after the heart had been taken ut f the animal. NSTRUMENTATON AND MEASUREMENTS During the experiments we recrded artic pressure, left ventricular pressure and artic flw (Fig. 3). Pressures were measured thrugh stiff plyvinyl chlride tubing, 30 cm in length, and Statham 3Db pressure transducers. Artic pressure was measured as lateral pressure. Left ventricular pressure was measured thrugh a needle stitched int that cavity at the apex. The damped resnance frequency f the systems was 80 Hz. Artic flw was measured electrmagnetically with a Bitrnex BL-610 pulsed-lgic flwmeter. The frequency respnse f this system was 3 db dwn at 100 Hz; time delay was.3 msec. Since the calculatin f the surce resistance invlves subtractin f large quantities (see Eq. 3), accuracy f measurement has t be high in rder t btain a reliable answer. Therefre we determined mean flw by cllecting the amunt f perfusin fluid ejected by the left ventricle during 30 secnds. The amunt f the fluid cllected was determined by weighing. Left ventricular pressure (the mean f which being the ther variable needed t calculate surce resistance) was pen thrax cat FGURE Layut f the experimental setup. R is a big reservir cntaining an xygenated red cell Tyrde's mixture, F is a filter, SL is the left atrial supply vessel where the height f the fluid is kept cnstant by the verflw system, OL, and RL is a small reservir frm which verflwing fluid is pumped back int R. The islated cat heart is laded with a hydraulic mdel f the input impedance f the arterial system f the cat. Rc is a hydraulic resistr, C is a capacitr (ttal arterial cmpliance), and Rp is the peripheral resistance. Cardiac utput () is measured by cllecting fluid in a glass cylinder. nstantaneus bld flw (/ ), artic bld pressure (P) and left ventricular bld pressure (/*, ) are measured as well. FGURE 3 Upper panel: Artic pressure, left ventricular pressure, and artic flw measured in an pen-thrax cat. Lwer panel: The same variables measured in the islated cat heart preparatin. Pressures are given in mm Hg, flw is given in cm'/sec.

3 PUMPNG ABLTY OF LEFT HEART/Elzingaand Westerhf 99 Dwnladed frm by n Nvember 8, 018 averaged by an analg filter (KEMO VBF/3; setting: -6dB at 0.5 Hz and a fall-ff f 48 db per ctave). The mean left ventricular pressure s btained was, after calibratin, measured with a digital vltmeter. Data als were recrded n an Elema Schnander direct-writing system (EMT 81) and n magnetic analg tape (Hewlett-Packard 355A). EXPERMENTAL APPROACH As described in the intrductin, tw settings f the lading system are enugh t determine the surce resistance and the generatr pressure; hwever, use f mre settings f the lading system gives the advantage f increased accuracy. n the experiments reprted here we changed the setting f the lading system by changing the peripheral resistance f the hydraulic mdel; the capacitance was kept cnstant during the experiments. We avided lw artic pressures and small artic flws, trying t study perfrmance ver a reasnable wrking range. At each setting f the peripheral resistance we measured mean ventricular pressure and mean artic flw. N measurements were made during transient states. We always waited 15-0 secnds after an interventin t make measurements. The values' btained were pltted against ne anther. Examples f these plts are shwn in Figures 4 and 9. The slpe f the line which can be drawn thrugh these pints is the surce resistance (APiv/A/a) (see Eq. 3) and the intercept with the rdinate represents the mean hydrmtive pressure (). The actual experiments ran as fllws: After preparatin f the islated cat heart we started the experiment by establishing the cntrl cnditin: left atrial filling pressure f abut 4 cm H O, mean artic pressure f 80 mm Hg, heart rate fixed by left atrial pacing. We then increased the peripheral resistance in steps. When (lw had fallen t abut 1.5 cm'/sec the resistance was lwered, again in steps, t the cntrl value. When this "run" appeared t be stable, i.e., when the cntrl values fr pressure and flw btained befre and after resistance changes did nt differ mre than : J v 5%, the lwer part f the descending branch f the left crnary artery was tied ff. This resulted in a decline in the functin f the heart. During this decline, mean artic pressure was kept at 80 mm Hg by changing the peripheral resistance, but left atrial filling pressure was kept cnstant. When the heart had stabilized at a lwer level f mechanical activity we repeated the run with stepwise increases and decreases in peripheral resistance. Having dne that we tried t restre cardiac functin by increasing left atrial filling pressure. We did this by trying t duplicate the cntrl cnditin, which we had btained at the start f the experiment, with respect t the values f mean artic pressure and peripheral resistance. This new level f left atrial filling pressure varied between 8 and 1 cm H O (n = 10). After this cmpensatin we executed stepwise increases and decreases f the peripheral resistance fr the third time. The cmplete experiment was perfrmed within 45 minutes. We reprt here n 10 successful experiments n 10 different hearts. Results T give sme insight int the functinal state f the islated heart preparatin, an example f the variables recrded during a cntrl situatin is presented in Figure 3 in cmbinatin with similar recrds frm a cat with pen thrax. Relevant data frm each experiment are given in Figure 4. We pltted here the relatinship between mean left ventricular pressure and mean artic flw fr each setting f the peripheral resistance in the three experimental situatins. Linear regressin analysis was perfrmed n each set f pints btained during each experimental situatin fr each heart. Results f this analysis are given in Table 1. n this table the pumping capacity f the heart is described by its surce resistance (the slpe f the lines in Figure 4) and the mean hydrmtive pressure, given by the intercept f the lines with the vertical axis. The table als includes the a i b 1 '. \ V X ' a ' l> FGURE 4 Graphic presentatin f the experimental results fr 10 experiments n islated cat hearts. Measurements made during the cntrl situatin are represented by O and, measurements after infarctin by V and, and measurements after cmpensatin by A and. Mean left ventricular pressure is given in mm Hg, mean artic flw in cm' sec. Data represented by pen symbls were btained when peripheral resistance was increased, and the slid symbls represent data btained when peripheral resistance was decreased.

4 300 CRCULATON RESEARCH VOL. 38, N. 4, APRL 1976 TABLE 1 Heart Rates(HR), Peripheral Resistances (R r ), Surce Resistances (/?,), and Mean Hydrmtive Pressures () f "Cntrl" Runs, "nfarct" Runs, and "Cmpensated" Runs per Heart Expt n. HR (beats/ miri) R,' R.' Cntrl (mm Hg)f r\ R.' nfarcl (mm Hg)f 4 RS Cmpensated R.' (mm Hg)t a b a b Mean ±SD 5.4 ± ± ± ± ± ± ± ± ±9.5 Peripheral and surce resistance are given in terms f 10' g cm"' sec~'; 10 g cm" 4 sec" 1 ~ 10* kg m~ 4 sec"'. Values fr peripheral resistance are thse measured fr the starting situatin fr each run. t 10 mm Hg ~ 1.33 kpa. Only the mean hydrmtive pressures btained frm the "infarct" and "cntrl" situatin are significantly different at the 5% level (Wilcxn matched pairs signed ranks test). t r = crrelatin cefficient f the linear regressin analysis used t btain R, and. Dwnladed frm by n Nvember 8, 018 crrelatin cefficients fr the lines, the peripheral resistances fr the cntrl settings, and heart rates. When we cmpared the cnditin after infarctin with cntrl we fund a significant decrease in mean hydrmtive pressure. Surce resistance did nt change significantly after infarctin. After the infarct cardiac utput was restred t the cntrl level by an increase in left atrial filling pressure and thus by the Frank-Starling mechanism. n this situatin left atrial filling pressure between hearts varied between 8 and 1 cm H O. n this cmpensated situatin mean hydrmtive pressure was nt significantly different frm that fund during cntrl cnditins. Eight f the 10 hearts shwed a lwer surce resistance after cmpensatin in cmparisn t the cntrl value; in tw hearts the ppsite was true. We culd nt demnstrate a statistically significant change in surce resistance between these tw situatins at the 5% level. T summarize ur findings fr the 10 experiments we determined mean values fr the three experimental situatins and pltted the results in Figure 5. mmhg 80 - cntrl infarct - cmpensated n the experiments described abve, left atrial filling pressure was kept cnstant by an verflw system. This is nt precisely the same as keeping left ventricular end-diastlic pressure r vlume cnstant. 5 With increasing artic pressure we fund a slight increase in left ventricular end-diastlic pressure in spite f the fact that left ventricular filling time decreases with an increase in artic pressure. 8 This is demnstrated in Figures 6 and 7. Figure 6 shws left ventricular end-diastlic pressure pltted as a functin f the end-diastlic pressure in the arta. The pints in this graph were btained frm the cntrl runs f nine experiments. Figure 7 shws tw left ventricular pressure tracings btained in a cntrl run fr tw situatins in which artic pressures differed. The widening f the left ventricular pressure pulse with the increase in ventricular lad and the cnsequent shrtening f left ventricular filling time are illustrated. Discussin The cncept f surce impedance f the heart is particularly suitable fr quantificatin f the pumping ability f the heart and has the advantage f being a quantity purely p«i,v Smr Hg -0 cm 3 sec" 1 FGURE 5 This figure shws the averaged values f Figure 4. 6 i i OO 10 MO FGURE 6 Left ventricular end-diastlic pressure (Ped ta ) pltted as a functin f end-diastlic pressure in the arta (Ped^). Symbls dente data fr different hearts.

5 PUMPNG ABLTY OF LEFT HEART /Elzinga and Westerhf FGURE 7 Ventricular pressure curves (mm Hg) measured at tw different levels f artic pressure. The shrtening f the filling lime with the increase in pressure is shwn. Dwnladed frm by n Nvember 8, 018 dependent n cardiac functin, and nt directly influenced by changes in the peripheral system.7 This cntrasts with, fr instance, ventricular functin curves. Bth surce impedance and ventricular functin curves try t describe a particular aspect f cardiac functin, but the shape f the ventricular functin curve is dependent n the "setting" f the lading system and, therefre, it is nt a specific cardiac quantity. 8 This dependency n the lad is demnstrated by tw ventricular functin curves btained frm an experiment n the islated cat heart and shwn in Figure 8. Here cardiac utput is pltted as a functin f left ventricular end-diastlic pressure under tw different experimental cnditins. n ne situatin mean artic pressure was kept cnstant by changing peripheral resistance, while in the ther peripheral resistance remained unchanged. Relating strke wrk t left ventricular end-diastlic pressure des nt change this dependency essentially, because strke wrk als is a functin f the peripheral resistance.9 Tying ff a part f a crnary artery resulted in a decrease in cardiac utput. Analysis f the data shwed that this was due t a change in hydrmtive pressure and that surce resistance remained unaffected. When cardiac utput was restred t cntrl value by increasing atrial filling pressure, we culd demnstrate n statistically significant difference in surce resistance and hydrmtive pressure between the cntrl and cmpensated situatins. This means that the vlume f bld the heart culd pump per unit f time 301 against different lads in the cntrl situatin was nt diminished after infarctin and subsequent cmpensatin. Thus we culd nt tell frm the hydrmtive pressure r the surce resistance whether the heart was damaged r nt when these tw situatins were cmpared. This demnstrates hw effectively an increase in filling pressure cmpensates fr a certain lss f active muscle in the ventricular wall. One must f curse bear in mind that 8 f the 10 hearts shwed a decrease in surce resistance after cmpensatin as cmpared t cntrl values, and the pssibility that a larger series wuld shw a significant decrease f this value cannt be excluded. Hwever, it appears nt t be a cnsistent finding fr all hearts. We shwed that an increase in artic pressure was related t a slight increase in left ventricular end-diastlic pressure in spite f a cnstant left atrial filling pressure and a decrease in ventricular filling time. This change in left ventricular end-diastlic pressure culd be due t a change in the left ventricular end-diastlic pressure-vlume relatinship which, in turn, was caused by an increased perfusin pressure in the crnary arteries.10 f this is true, left ventricular end-diastlic pressure cannt be used as an index f left ventricular end-diastlic vlume in these experiments and it is then very difficult t estimate the pssible change in left ventricular end-diastlic vlume with a change in artic pressure withut measuring the frmer variable. Our experiments were designed in such a way that pints in the plt f mean left ventricular pressure against mean artic flw were btained by first increasing and then decreasing the periperal resistance f the lad. Pints btained by increasing resistance fell n the same line as pints btained by decreasing ventricular lad. This demnstrates the stability f the preparatin and the reprducibflity f the measurements. Cnsidering a mdel like that presented in Figure 1, ne assumes the hydrmtive pressure t be cnstant and the surce resistance t be cnstant and linear. f the pints in. mmhg OO /! * pa exp LA.filling pressure 8 cm H 0 exp L.A.filling pressure 4cm H cnstant (80mm Hg) R p cnstant mmhg FGURE 8 Tw left ventricular functin curves btained frm the same heart under different lading cnditins. n ne experiment mean artic pressure (Pa) is kept cnstant by altering the peripheral resistance. n the ther experiment the peripheral resistance (Rp) is kept unchanged. 7 = mean flw. 4 6 crr^sec1 FGURE 9 One curved and nejnre linear relatinship between mean left ventricular pressure (P,,) and artic flw (). Difference between pen andfdled-in symbls is the same as fr Figure 4.L.A. = left atrium.

6 30 CRCULATON RESEARCH VOL. 38, N. 4, APRL 1976 the plt f surce resistance fall precisely n a straight line these assumptins are likely t be true. The values f the crrelatin cefficients presented in Table 1 d suggest that in mst cases a straight line fits the data well. Hwever, the degree f linearity seems t vary between hearts. Smetimes a slightly curved line seems t fit the results better than a straight line. Tw extreme examples, measured fr tw different hearts, are shwn in Figure 9. When a mre curved relatinship is fund it is always in the directin shwn in Figure 9. This srt f curvature is in agreement with the bservatin f Bergel (discussin f Elzinga and Westerhf ) that the predicted mean hydrmtive pressure btained by extraplatin is greater than the measured mean hydrmtive pressure determined frm an isvlumic beat. The cause f this is nt yet clear t us. Hwever, in a physilgical wrking range (mean left ventricular pressure f mm Hg and mean artic flw f -5 cm 3 /sec) a linear apprach seems t be reasnable. n cnclusin we wuld like t state that the surce impedance cncept appears t be a useful tl t describe the pumping ability f the heart in quantitative terms. Occlusin f a part f the crnary arterial system affects the hydrmtive pressure but nt the surce resistance. The pumping ability f the left heart can, after infarctin, be restred almst cmpletely by an increase in left atrial filling pressure. The Frank-Starling mechanism which the heart can use t cmpensate fr the effects f a lss in cntractile tissue therefre seems t be very effective in this situatin. Acknwledgments We thank l.t. Gabe, C.J. Mills, and M..M. Nble fr the stimulating discussins n the cncept f surce impedance. References 1. Elzinga G, Westerhf N: Pressure and flw generated by the left ventricle against different impedances. Circ Res 3: , Elzinga G, Westerhf N: End-diastlic vlume and surce impedance f the heart. n The Physilgical Basis f Starling's Law f the Heart. Ciba Fund Symp 4: 51-55, Elzinga G: Crss talk between left and right heart. Ph.D. Thesis. Free University, Amsterdam, Westerhf N, Elzinga G, Sipkema P: An artificial arterial system fr pumping hearts. J Appl Physil 31: , Katz LN: Analysis f the several factrs regulating the perfrmance f the heart. Physil Rev 35: , Buckberg GD, Fixler DE, Archie JP, Hffman JE: Experimental subendcardial ischemia in dgs with nrmal crnary arteries. Circ Res 30:67-81, Milnr WR: Arterial impedance as ventricular afterlad. Circ Res 36: , Snnenblick EH, Dwning SE: Afterlad as a primary determinant f ventricular perfrmance. Am J Physil 04: , Wilcken DEL, Charlier AA, Hffman JE, Guz A: Effects f alteratins in artic impedance n the perfrmance f the ventricles. Circ Res 14: 83-93, Salisbury PF, Crss CE, Rieben PA: nfluence f crnary artery pressure upn mycardial elasticity. Circ Res 8: , 1960 Dwnladed frm by n Nvember 8, 018 Electrphysilgical and Antiarrhythmic Effects f Prpranll in Canine Acute Mycardial schemia JOEL KUPERSMTH, M.D., HOWARD SHANG, D.V.M., ROBERT S. LTWAK, M.D., AND MCHAEL V. HERMAN, SUMMARY T crrelate the antiarrhythmic and electrphysilgical effects f prpranll in acute mycardial ischemia, we examined the effects f temprary (15-minute) ligatins f the left anterir descending crnary artery in studies n 15 dgs. We recrded biplar electrgrams and mnphasic actin ptentials frm the ischemic and nrmal znes and measured the intervals frm the nset f QRS in a standard electrcardigram lead t the majr deflectin f electrgrams recrded frm the ischemic and nrmal znes. We als determined mnphasic actin ptential duratin (APD) and effective refractry perid (ERP). Data fr cntrl ligatins were cmpared t thse during which prpranll, 40 jig/ kg, was administered intravenusly immediately after ligatin. Prpranll reduced the mean number f ventricular beats per minute (frm 15 t 6) (P < 0.01). Prpranll slwed cnductin in the M.D. ischemic zne (by 10 msec at peak effect, P < 0.01) and had n r nly a very slight effect (by 1-msec at 15 minutes, P < 0.05) n cnductin in the nrmal zne. Prpranll als prlnged APD in the ischemic (3-msec) and nrmal (14-msec) znes (P < 0.01), prlnged ERP in the ischemic (41-msec) and nrmal (0-msec) znes (P < 0.01), and reduced the APD/ERP rati in the ischemic (1.6 t 1.47) (P < 0.01) and nrmal (1.6 t 1.55) (P < 0.05) znes. During the cntrl ligatin, APD in the ischemic zne was 5 msec shrter than in the nrmal zne (P < 0.01), but with prpranll the difference was nt significant. The effects f prpranll in slwing cnductin in the ischemic zne, in prlnging refractriness, in reducing APD/ERP, and in reducing the disparity in APD between ischemic and nrmal znes may explain its demnstrated antiarrhythmic effects in acute mycardial ischemia. SPECULATON n the mechanism f actin f antiarrhythmic drugs is based in large measure n micrelectrde Frm the Departments f Medicine and Surgery, The Munt Sinai Schl f Medicine f the City University f New Yrk, New Yrk, New Yrk. Supprted in part by Grants HL and HL frm the Natinal Heart and Lung nstitute. Received August ; accepted fr publicatin December 30, studies f nrmal, islated cardiac tissue. 1 " 4 n these studies, the effects f an agent n varius parameters including cnductin, actin ptential duratin, and refractry perid are determined and the results are extraplated t the arrhythmic, abnrmal heart in situ. 1 '* Hwever, there are limitatins t this methd because the effects f antiarrhythmic drugs n the electrphysilgical prperties f nrmal