Effect of selective intracoronary antiarrhythmic drug administration in sustained ventricular tachycardia

ARRHYTHMIAS AND CONDUCTION DISTURBANCES

Effect of Selective Intracoronary Antiarrhythmic Drug Administration in Sustained Ventricular Tachycardia Peter L. Friedman, MD, PhD, Andrew P. Selwyn, MD, Elazer Edelman, MD, PhD, and Paul J. Wang, MD

The effect of selective intracoronary antiarrhythmic drug inhrsion on inducibility of cardiac arrhythmias was studied in 3 patients wRh recurrent sustained monomorphic ventricular tachycardia referred for comprehensive electrophysiologic studies. Each patient had evidence of prior myocardial infarction, 1 or more occluded coronary arteries and a readily identifiable collateral vessel that provided collateral flow to the infarct-related artery. In each patient, the clinical arrhythmia was reproducibly inducible by programmed stimulation in the control state. After positioning a small infusion catheter in the coilateral vessel, selective intracoron ary lidocaine 0.3 to 0.6 mg/min (patients 1 and 2) or procainamide 0.1 to 1.4 mg/min (patient 3) was hrfused for a lominute period. In each patient the clinical arrhythmia was rendered noninducible during selective intracoronary drug infusion. The arrhythmia was again inchrcible after a lo-minute drug-washout period and also after standard intravenous doses of antiarrhythmic drug. Selective intracoronary antiat-rhythmic drug inhrsion may help to localize the site of origin of some cardiac arrhythmias, may provide a means of testing the effects of several drugs during a single study and may be a new method for studying mechanisms of action of antiarrhythmic drugs. (AmJCardid 1989;64:475-499)

From the Clinical Electrophysiology and Cardiac Catheterization Lab oratories, Brigham and Women’s Hospital, and the Department of Medicine, Harvard Medical School, Boston, Massachusetts. Manuscript received March 9, 1989; revised manuscript received May 26, 1989, and accepted May 30. Address for reprints: Peter L. Friedman, MD, PhD, Cardiovascular Division, Brigham and Women’s Hospital, 75 Francis Street, Boston, Massachusetts 02115.

entricular tachycardiac(VT) often cannotbemanaged satisfactorily with conventional antiarrhythmic drug therapy. All available antiarrhythmic drugs can cause debilitating side effects that limit their use.l Even when the drugs are well tolerated they frequently fail to prevent recurrence of arrhythmia.2y3The most common substrate for sustainedmonomorphic VT is coronary artery diseasewith prior myocardial infarction.4 In such a setting, VT usually is due to reentry in surviving but electrophysiologically abnormal subendocardialor subepicardial fibers within or adjacent to the region of infarction5-l3 Becauseall regions of the heart are nourished by the coronary arterial circulation, delivery of an antiarrhythmic drug to the site of origin of VT via the coronary artery supplying such a site might be an effective method for suppressingthe arrhythmia. Theoretically, this method of delivery could achievevery high local tissue concentrationsof the drug where its effects are most needed,without risk of noncardiac toxicities that accompanysystemic antiarrhythmic drug administration. This hypothesiswas testedin 3 patients with recurrent sustainedVT due to prior myocardial infarction. We describe the results of selective intracoronary antiarrhythmic drug infusion in these patients and discuss the possible applications of this approach to the localization and therapy of cardiac arrhythmias.

V

METHODS Three patients with documented sustained mono morphic VT referred to the Brigham and Women’s Hospital for electrophysiologic studies were enrolled in a researchprotocol approved by the hospital’s Committee for the Protection of Human Subjects from ResearchRisks. Informed consentwas obtained from each at the time of enrollment. Before their electrophysiologic studies each patient underwent routine diagnostic cardiac catheterization with coronary arteriography to define coronary anatomy and determine possiblesites of prior myocardial infarction. Electrophyslologlc studies: All patients underwent comprehensive electrophysiologic studies while they were in the fasting state, lightly sedatedwith diazepam as needed to allay anxiety. Patient 1, who had been found to have inducible sustained monomorphic VT during an electrophysiologic study 1 year previously, was being treated with flecainide, 100 mg twice daily,

THE AMERICAN JOURNAL OF CARDIOLOGY SEPTEMBER 1, 1989

475

TABLE I Patient Characteristics

PI

Age (yrs), Sex

Clinical Arrhythmia

1

70. M

VT

252W

VT

363,M

VT

LV Wall Motion Inferior hypokinesis Anteroapical dyskinesis lnferoapical dyskinesis Anterior akinesis

Occluded Coronary Artery

Collaterals

Right

Distal LC

IAD

B

PDA

LC marginal

LAD

B

B = brid&g collaterals; LAD = kft anterior descending artery; LC = left circumflex; LV = left ventricular: FDA = posterior descending artery: VT = ventricular tachycardia.

and digoxin, 0.125 mg orally once daily, when he had an episodeof syncopewithout documented arrhythmia. This patient was therefore restudied on his regimen of flecainide plus digoxin. Patients 2 and 3 had receivedno antiarrhythmic medication for 72 hours before study. Electrode catheters were inserted via sheathsplaced in both femoral veins, advanced under fluoroscopic guidance and positioned in the high right atrium, right ventricular apex and acrossthe tricuspid anulus; the latter position was used to record a His bundle electrogram. An 8Fr sheath with side arm was also introduced into the femoral artery. Once the catheters and sheathswere in place, 5,000 U of heparin was administered intravenously. Programmed electrical stimulation of the atria (patients 2 and 3) and of the ventricles (all patients) was performed according to a standard protocol as previously described from our laboratory.i4 The ventricular stimulation protocol consistedof delivery of progressively earlier single, double and then triple ventricular premature stimuli to the point of ventricular refractoriness. It was done first during sinus rhythm (patients 2 and 3) or atria1 fibrillation (patient l), and then after 8-beat ventricular-paced trains at 2 or more basic cycle lengths (600 ms or 500 and 400 ms) in all 3 patients. In all 3 of these patients sustained monomorphic VT corresponding to the patient’s clinical arrhythmia was induced by programmed stimulation from the right ventricular apex. Patients 1 and 2 had VT that was easily inducible in a baseline state. In patient 3, tachycardia was only inducible during isoproterenol infusion 0.8 to 1.0 pg/ min. This infusion was continued throughout the entire study. Once sustained tachycardia had been induced it was terminated by programmed stimulation. In each patient the stimulation sequencethat had initiated arrhythmia was then repeated at least twice to confirm that induction of the tachycardia was reproducible. Research protocok Following induction and termination of tachycardia in the control state, an 8Fr left Judkins guiding catheter was introduced through the femoral arterial sheath and positioned in the ostium of the appropriate coronary artery. At this point an additional 5,000 U of heparin was administered intravenously. In each patient the branch of the coronary artery supplying collateral blood supply to the infarct-related 476

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 64

artery was identified from the patient’s diagnostic coronary arteriogram and then selectively catheterized. In patient 1 the collateral vesselwas the distal circumflex artery (Table I). In patient 2 the collateral vessel was the stump of the left anterior descendingartery (Table I). In patient 3, who had 2 occluded coronary arteries, the 2 collateral vesselsidentified were the stump of the left anterior descendingand a marginal branch of the circumflex artery (Figure 1). Each collateral vessel in this patient was selectively catheterized separately. A O.OlCinch diameter guidewire (Hi-Torque Floppy, Advanced Cardiovascular Systems Inc.) was advanced through the coronary guiding catheter and positioned in the study vessel.After the guidewire was positioned, an infusion catheter with a 3Fr proximal shaft diameter, 2.2Fr distal shaft diameter and radiopaque tip marker (Tracker, Target Therapeutics Inc.) was then advanced over the guidewire and also positioned in the appropriate study vessel.Proper positioning of the infusion catheter was confirmed by injection of radiographic contrast material through the guiding catheter (Figure 1). Once the infusion catheter was properly positioned the guidewire was removed,enabling the proximal end of the infusion catheter to be connected to a Harvard infusion pump. In patients 1 and 2 lidocaine dissolvedin normal saline was infused into the collateral vesselvia a Harvard infusion pump at a constant rate in dosesranging between 0.3 and 0.6 mg/min using flow rates between 0.075 and 0.15 cc/min. Patient 1 received 2 different intracoronary doses:first 0.3 and then 0.6 mg/min. Patient 2 receivedonly a single intracoronary doseof 0.35 mg/min. Patient 3 receivedan intracoronary infusion of procainamide rather than lidocaine in normal saline at a dose of 1.4 mg/min and flow rate of 0.175 cc/mm for the stump of the left anterior descending artery and then 1.0 mg/min and flow rate of 0.125 cc/mm for the marginal branch of the circumflex artery. In each patient at each dose the intracoronary antiarrhythmic drug infusion was administered for 10 minutes, after which a complete sequenceof programmed stimulation was performed severaltimes according to the sameprotocol used in the control state. Upon completion of the stimulation protocol the intracoronary drug infusion was discontinued, 10 minutes were allowed for the effects of intracoronary antiarrhythmic drug infusion to dissipate, and then a recontrol complete stimulation protocol was repeated to confirm that tachycardia was again inducible as it had been during the initial control period. Finally, antiarrhythmic drug was administered in standard doses intravenously. Patients 1 and 2 re ceived bolus injections of 100 and 75 mg, respectively, of lidocaine followed by a 1.0 mg/min infusion. Patient 3 received a bolus injection of 1,000 mg of procainamide. The complete programmed stimulation protocol was again repeatedseveraltimes after intravenous drug administration using the samestimulation protocol as in the control state. Plasma lidocaine and procainamide concentrations were determined by EMIT enzyme immunoassayfrom venous blood samplesobtained before

initiation of intracoronary drug infusion, at the end of the 1O-minute period of intracoronary drug infusion and after administration of intravenous drug.

descending artery via its collateral vessel, VT was no longer inducible. Thus, abolition of VT in this patient was critically dependenton which of the 2 occludedvessels selective drug infusion was delivered into.

RESULTS

Pertinent clinical features of the 3 patients and the results of their diagnostic angiographic studies are listed in Table I. All 3 patients had angiographic evidenceof prior myocardial infarction with an easily identifiable infarct-related occluded coronary artery. Patient 3 actually had 2 sites of prior infarction, a calcified inferoapical infarct related to an occluded right coronary artery and a more recent anterior wall infarct causedby occlusion of the left anterior descendingartery. The effects of selective intracoronary lidocaine infusion and conventional intravenous lidocaine administration on inducibility of ventricular tachycardia in patient 1 are shown in Figure 2. In the control state, monomorphic sustained VT with a right bundle branch block QRS morphology and average cycle length of 460 ms (rate 124 beats/min) was easily and reproducibly initiated by 3 consecutive programmed ventricular premature stimuli delivered during atria1 fibrillation (Figure 2A). The tachycardia was sustained and required overdrive ventricular pacing for termination. Plasma lidoCaine concentration in the control state was undetectable. During infusion of lidocaine at a rate of 0.3 mg/ min into the distal left circumflex artery, which supplied collateral blood flow to the occluded right coronary artery, VT was still inducible, although the cycle length of the tachycardia was lengthened to 500 ms (not shown). However, during selective intracoronary infusion of lidocaine at a rate of 0.6 mg/min, VT was no longer inducible by programmed ventricular stimulation (Figure 2B). Systemic plasma lidocaine concentration during the intracoronary infusion was still undetectable. Ten minutes after cessation of the intracoronary lidocaine infusion, recontrol programmed ventricular stimulation was performed. Again, sustainedmonomorphic VT was easily inducible, this time by delivery of only 2 premature stimuli. The tachycardia had the same QRS morphology and rate as in the initial control period (Figure 2C). Administration of intravenous lidocaine failed to prevent induction of the patient’s VT, even though a plasma lidocaine concentration of 2.6 pg/ml was achieved (Figure 2D). Table II summarizes the effects of intracoronary antiarrhythmic drug infusion compared to intravenous drug administration on VT induction in patients 2 and 3. In each case,induction of VT could be abolished by selective intracoronary infusion of the antiarrhythmic drug but not by intravenous administration, even though the latter route of administration resulted in a much higher plasma drug concentration. The results obtained in patient 3 were particularly noteworthy. When procainamide was infused selectively into the stump of the left anterior descending artery and into the distal vesselby bridging collaterals, VT was still inducible as it had been in the control state. However, when procainamide was infused selectively into the occluded posterior

DISCUSSION

Patients with cardiac arrhythmias that have proven unresponsive to antiarrhythmic drug therapy or in whom drug therapy causesintolerable side effects require alternative forms of therapy to prevent recurrent arrhythmia. Ablation of the arrhythmogenic tissue, whether surgically or by means of a catheter, offers such an alternative. In a recent study in dogs by Inoue et al,15 focal VT due to intramyocardial injection of aconitine could be abolished by infusion of phenol or ethyl alcohol into the coronary artery supplying this region. These investigators15postulated that the coronary

FIGURE 1. Rii anterior oblique arkriogram of ths left coronary artery in patisnt 3, who had an occkkd IeR anterior detsceding artery (straight arrow, A) that filled by bridging collateralsaswellasanocdudsd postehr descending artery (broad arrow, B) that received collateral Row from a marginal branch of ths circumitex artery. The radiopaqw tip of the infusion cathetsr can be seen positioned in the proximal left anterior descending artery (curved arrow, A) and then in the ciruanttsx marginal branch (thin arrow, 4.

THE AMERICAN JOURNAL OF CARDIOLOGY SEPTEMBER 1, 1989

477

INTRACORONARY DRUG TREATMENT OF VENTRICULAR TACHYCARDIA

arterial circulation might provide a vehicle for delivery of ablative agents to sites of origin of cardiac arrhythmias no matter where in the heart they might be located. The validity of this hypothesishas now beenverified in humans by Brugada et a1,t6who demonstrated that sustained tachycardias of various types could be interrupted by bolus intracoronary injections of iced saline or by temporary occlusion of the coronary artery thought to supply the site of origin of arrhythmia. Furthermore, in selectedpatients with refractory VT the arrhythmia could be abolished permanently by injection of ethyl alcohol into the appropriate coronary arterial branch.”

CONTROL

The presentstudy differs from those of Brugada et al in that we choseto study the effects of intracoronary antiarrhythmic drugs, using inducibility of arrhythmia by programmed stimulation to assessthe effects of local drug administration. Our study demonstrates that recurrent sustainedVT in the setting of chronic myocardial infarction can be rendered noninducible by infusion of an antiarrhythmic drug selectively into a branch of the coronary arterial circulation supplying flow to the presumedsite of origin of the arrhythmia. Theseresults complementthose of Brugada et al, and further support the hypothesisthat cardiac arrhythmias can be manipu-

iASMA [L] < ASSAY

A

I.C. LIDOCAINE 0.6 mg/min

.ASM A [L]
RECONTROL

C

1.v. LIDOCAINE 100 mg bolus

ASMA [Ll= 2.6ugIml

I.Omg+min

D lar rpcx (RV). The bHation.V-artaehycardia(VT)waslnducedby3premahmstimuli w52s3)hthccontrol~andby2stirrmli (slsz)lIurhgand atler intravenous (I.V.) lidocaine. VT could not be ilnluced~~~ (l.c.) lidocaine Mush, 6. [L] = Mocaine unwentratien. So/id bars ad mmnbers at lower right in 4 C and D indicate VT cycle length (in ms). See text. 478

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 64

lated and potentially ablated no matter where in the heart they arise-provided that the coronary arterial supply to the site of origin of arrhythmia can be identilied and catheterized selectively. Procainamide and lidocaine were chosenempirically for intracoronary administration in the present study, partly becausethey are among the few agents available for intravenous use and, as a consequence,are the agents most often tested acutely during electrophysiologic studies. Assuming normal rates of coronary blood flow of 70, 60 and 40 to 50 ml/min in the left anterior descending,circumflex and right coronary arteries, respectively, the intracoronary dosesof lidocaine and procainamide were chosento achieve concentrations in the local coronary circulation at least twice what is ordinarily consideredto be “therapeutic” in systemicblocdnsJ9 In all likelihood, abolition of arrhythmia by selectiveintracoronary antiarrhythmic drug infusion as observedin the present study was due to attainment of much higher local myocardial drug concentrations at or near the site of origin of the arrhythmia than were achievedby intravenousdrug administration. This is despite the fact that the latter route of administration yielded a much higher systemicblood concentration of the drug. In view of the extremely slow infusion rates that were used in the present study relative to normal rates of coronary flow, it is very unlikely that selective intracoronary infusions causedabolition of arrhythmia by cooling or by replacing a significant fraction of the arterial blood flow, thereby rendering the site of origin ischemic. This supposition is supported by the fact that none of the patients developedchest pain or electrocardiographic evidence of ischemia during intracoronary drug infusion. In the present study, selective infusion of antiarrhythmic drug into the vesselsupplying collateral flow to the infarct-related artery rather than into proximal segmentsof other major coronary arteries should have resulted in delivery of drug predominantly if not exclusively to surviving fibers within or at the border of the infarcted zone, the likely site of origin of VT. This route of drug delivery may afford a unique method for assessing mechanisms of action of antiarrhythmic drugs in arrhythmogenic fibers. Observationsin our first patient, who received 2 different doses of intracoronary lidoCaine,are particularly noteworthy in this regard. In this patient an infusion rate of 0.3 mg/min, which failed to prevent induction of VT, did result in appreciable slowing of the rate of VT. In fact, the tachycardia was slower during this intracoronary infusion than after conventional intravenous lidocaine dosing, despite the fact that plasma drug concentration was not detectable during intracoronary infusion and was within “therapeutic” range after the intravenous dose. This observation can most easily be explained by slowing of conduction velocity in all or a part of the reentrant circuit by high local concentrations of lidocaine achieved at that site during intracoronary infusion. When a larger doseof intracoronary lidocaine was administered, induction of sustained tachycardia was prevented altogether. It seemsreasonable to conclude that abolition of sustained tachycardia at the larger dosewas due to further slowing of conduc-

TABLE II Drug Effects on Arrhythmia

Patient 2 Control IC lidocaine Recontrol IV lidocaine Patient 3 Control IC procainamide IC procainamide Recontrol IV procainamide

(LAD) (PDA)

Induction

VT

CL (ms)

+ 0 + +

270 0 260 260

+ + 0 + +

270 250 0 280 325

Plasma Drug Level &g/ml) -

2.2

8.4

CL = tachycardia cycle length; IC = intracoronary; IV = intravenous; detectable: + = present; 0 = absent; other abbreviations as in Table I.

= not

tion velocity in the reentrant circuit, perhaps also accompaniedby prolongation of refractory period, thereby reducing the safety factor for conduction to such a degree that repetitive conduction around the circuit was no longer possible.In support of this conclusion was the observation that premature stimulation during high doseintracoronary lidocaine was still able to provoke a single nonstimulated ventricular beat identical in morphology to the patient’s VT (Figure 2B), but not sustained arrhythmia. One interesting observation in the present study was the apparent dissipation of drug effects that occurred after no more than a lo-minute washout period following intracoronary drug infusion. This period of time was chosen empirically. However, in each of the patients studied, by the end of this lo-minute period electrophysiologic parameters had returned to control values and tachycardia was again inducible. This observation raises the possibility that intracoronary antiarrhythmic drug infusion might enable one to assessthe effects of several drugs in rapid successionduring a single procedure, thereby sparing the patient multiple studies on successivedays. Although none of the patients in our study had complications or ill effects from intracoronary antiarrhythmic drug infusion, it is conceivablethat such a route of administration could be proarrhythmic rather than antiarrhythmic. Further studies in a greater number of patients will be essentialto determine optimum dosesand infusion rates of intracoronary antiarrhythmic drugs and to assessmore fully the risks and possiblebenefits of this new approach. Acknowledgment: We are grateful to Drs. Hussein Hasan Rizk Hasan and Hong Sheng Guo as well as the nursesin our cardiac catheterization laboratory who assisted in the conduct of these studies and to Mary Gillan for her help in preparation of the manuscript.

REFERENCES 1. Nygaard TW, Sellers TD, CookTS, DiMarw JP. Adverse reactions to antiarrhythmic drugs during therapy for ventricular arrhythmias. JAMA 1986;256:5557. 2. Skale BT, Miles WM. Heger JJ, Zipes DP, Prystowsky EN. Survivors of cardiac arrest: prevention of recurrence by drug therapy as predicted by electro-

THE AMERICAN

JOURNAL

OF CARDIOLOGY

SEPTEMBER

1, 1989

479

INYRACORONARY DRUG TREATMENT OF VENTRICULAR TACHYCARDIA

physiologic testing or electrocardiographic monitoring. Am J Cardiol 1986; 57:113-119.

3. Wilber DJ, Garan H, Finkelstein D, Kelly E, Newell J, McGovern B, Ruskin JN. Out-of-hospital cardiac arrest. Use of electrophysiologictesting in the prediction of long-term outcome. N Engl J Med 1988;318:19-24. 4. Wellens HJJ, Duren DR, Lie KI. Observationson mechanismsof ventricular tachycardia in man. Circulation 1976;54:237-244. 5. FriedmanPL, Steward JR, FenoglioJJ Jr, Wit AL. Survival of subendocardial Purkinje fibers after extensivemyocardial infarction in dogs:in vitro and in viva correlations. Circ Res 1973;33:597-61 I. 6. Friedman PL, Stewart JR, Wit AL. Spontaneousand inducedcardiac arrhythmias in subendocardialPurkinje fibers surviving extensivemyocardial infarction in dogs. Circ Res 1973:33x512-626. 7. JosephsonME, Horowitz LN, Farashide A, Kastor JA. Recurrent sustained ventricular tachycardia. 1. Mechanisms.Circulation 1978;57:431-440. 8. JosephsonME, Horowitz LN, Farshidi A, Spear JF, Kastor JA, Moore GV. Recurrent sustainedventricular tachycardia: 2. Endocardial mapping. Circulation 1978;57:440-447.

9. FenoglioJJ Jr, Harken AH, Horowitz LN, JosephsonME, Wit AL. Recurrent sustainedventricular tachycardia: structure and ultrastructure of subendocardial regions where tachycardia originates. Circulation 1983.68.518-533. 10. El-Sherif N, Smith A, Evans K. Canine ventricular arrhythmias in the late myocardial infarction period: epicardial mapping of reentrant circuits. Circ Res

correlation of reentrant circuits. Circulation 1983,67:1 l-23. 12. Wit AL, Allessie MA, Bonke FIM, Lammers W, SmeetsJ, Fenoglio JJ Jr. Electrophysiologicmappingto determinethe mechanismof experimentalventricular tachycardia initiated by premature impulses:experimental approach and initial results demonstrating reentrant excitation. Am J Cardiol 1982;49:166185.

13. Kramer JB, Saflitz JE, Witkowski FX, Corr PB. Intramural reentry as a mechanismof ventricular tachycardia during evolving myocardial infarction. Circ Res 1985;56:736-748. 14. Ludmer PL, McGowan NE, Antman EM, Friedman PL. Efficacy of propa-

fenonein Wolff-Parkinson-White syndrome:electrophysiologicfindings and longterm follow-up. JACC 1987.9:1357-1363. 15. Inoue H, Wailer BF, Zipes DP. Intracoronary ethyl alcohol or phenolinjection ablates aconitine-inducedventricular tachycardia in dogs.JACC 1987;lO: 1342-1349. 16. Brugada P, de Swart H, SmeetsJLRM, Bar FWHM, Wellens HJJ. Termination of tachycardiasby interrupting blood flow to the arrhythmogenicarea.Am J Cardiol 1988,62:387-392. 17. BrugadaP, de Swart H, SmeetsJLRM, WellensHJJ. Tramcoronary chemical ablation of ventricular tachycardia. Circulation 1989;79:475-482.

18. Ganz W, Tamura K, Marcus HS, DonosoR, Yoshida S, Swan HJC. Mcasurement of coronary sinus blood flow by continuous thermodilution in man. Circulation

1971:1S4:181-195.

1981;49:255-265. 11. Mehra R, Zieler RH, Gough WB, El-Sherif N. Reentrant ventricular ar-

19. Cannon PJ, Weiss MB, Sciacca RR. Myocardial blood flow in coronary artery disease:studiesat rest and during stresswith inert gaswashouttechniques.

rhythmias in the late myocardial infarction period: elcctrophysiologic-anatomic

Prog Cardiouasc Dis 1977;20:95-120.

480

THE AMERICAN JOURNAL OF CARDIOLOGY VOLUME 64