Intravenous adenosine triphosphate during wide QRS complex tachycardia: Safety, therapeutic efficacy, and diagnostic utility

Intravenous Adenosine Triphosphate during Wide QRS Complex Tachycardia: Safety, Therapeutic Efficacy, and Diagnostic Utility ARJUN D. SHARMA, M.D., F.A.c.c., F.R.c.P.c., Sacramento, Ca/ifornia, F.R.c.P.c., RAYMONDYEE, M.D., F.A.C.C., F.R.C.P.C., London, Canada

PURPOSE:Inappropriate administration of intravenous verapamil to patients with wide QRS complex tachycardia due to ventricular tachycardia or atrial fibrillation with Wolff-Parlcinson-White syndrome occurs frequently because of n&diagnosis, and may precipitate a cardiac arrest. We evaluated the safety and the diagnostic and therapeutic utility of adenosine triphosphate administered to a consecutive series of 34 patients during wide QRS complex tachycardia due to a variety of mechanisms. PATIENTSANDMETHODS: Patientswhohadahemodynamically and electrically stable, monomorphic, wide (greater than 120 msec) QRS complex tachycardia induced during an invasive cardiac electrophysiologic test were studied. Hemodynamic stability was defined by a systolic blood pressure greater than 80 mm Hg and no clinical evidence of cerebral or myocardial ischemia. Adenosine triphosphate, 20 mg, was administered as a rapid intravenous bolus via a peripheral vein during wide QRS complex tachycardia. Five surface electrocardiogram leads, at least three intracardiac electrograms, and blood pressure were monitored. RESULTS Ventricular tachycardia was present in 14 patients (mean age 50.6 f 19 years, cycle length 326 f 67 msec) and adenosine triphosphate terminated the arrhythmia in one case. Ventricular tachycardia cycle length did not change. Among 10 patients with supraventricular tachycardia with mechanisms not involving the AV node (average ventricular cycle length 346 f 82 msec), one case of ectopic atrial tachycardia was terminated. The ventricular rate was transiently increased in patients with Wolff-Parhinson-White syndrome and atrial fibrillation (average R-R interval 351 f 84 msec in control and 317 f 82 msec after adenosine triphosphate, p
From the Mercy General Hospital, Sacramento, Calrfornra. and the Department of Pharmacolonv and Medicine. Universitv of Western Ontario. London, Canada. This w;k was supported by the Mkdical Research Council of Canada and the Heart and Stroke Foundabon of Ontario, and was presented in part at the Annual Scienhfic Sessron of the North American Society for Pacing and Electrophysrology. Toronto, May 1989. Requests for reprints shouldbeaddressed to Arjun D. Sharma. M.D., F.A.C.C.. F.R.C.P.C., Cardiac Electrophysiology, Mercy General Hospital, 3941 J Street, Suite 260, Sacramento, Calrfornia 95819. Manuscript submitted May 25, 1989, and accepted fn rewed form January 10. 1990.

GEORGEJ. KLEIN, M.D., F.A.c.c.,

CONCLUSION: In the setting of electrophysiology testing, adenosine triphosphate is a safe agent, even when administered inappropriately during arrhythmias for which it is relatively ineffective, such as ventricular tachycardia, and Wolff-Parhinson-White syndrome with atrial fibrillation. It is an effective agent in terminating supraventricular tachycardia involving the AV node. Tachycardia termination following adenosine triphosphate, when used as a diagnostic test to indicate obligatory participation of the AV node, had a sensitivity of 70%, specificity of 92%, and a positive predictive accuracy of 85%. Thus, adenosine triphosphate also has diagnostic utility, but should be used after the appropriate arrhythmia diagnosis has been made based on the clinical history and analysis of the 12lead electrocardiogram. erapamil has proved to be highly effective in the termination of reentrant supraventricular tachyV cardias involving the atrioventricular (AV) node [l], and this has led to its widespread use in the emergency room therapy of these cardiac arrhythmias. However, recent reports have highlighted the danger of inappropriate use of verapamil in patients with wide QRS complex tachycardia due to ventricular tachycardia [2,3], or atria1 fibrillation with Wolff-ParkinsonWhite syndrome [4,5]. Furthermore, misdiagnosis of ventricular tachycardia as supraventricular tachycardia with aberrancy is relatively frequent [6,7]. Verapamil usually fails to terminate ventricular tachycardia but has vasodilating and negative inotropic effects that may lead to subsequent hypotension and cardiac arrest [2,3]. Potential alternative therapeutic agents for acute termination of supraventricular arrhythmias involving the AV node are the purinergic agents adenosine and adenosine triphosphate [8-lo]. We hypothesized that the short duration of action of these compounds may make them relatively safe even if administered inappropriately to patients with tachycardias not dependent on AV nodal conduction. Therefore, the purpose of this study was to examine the safety and efficacy of intravenous adenosine triphosphate administered to a consecutive series of patients with stable, wide QRS complex tachycardias induced during electrophysiologic testing. The electrophysiologic testing was used to establish tachycardia mechanisms.

PATIENTS AND METHODS A consecutive series of 34 patients with sustained, hemodynamically and electrically stable wide QRS complex tachycardia induced during invasive electroApril 1990

The American

Journal

of Medicine

Volume

88

337

WIDE QRS TACHYCARDIA AND ADENOSINE TRIPHOSPHATE / SHARMA ET AL

2, patients with supraventricular tachycardia that did not require AV nodal conduction for maintenance of arrhythmia; and Group 3, tachycardia in which AV nodal conduction was obligatory.

TABLEI Diagnostic Utility of Adenosine Triphosphate to Indicate Obligatory Role of AV Nodal Conduction Positive test = tachycardia terminatron Negative test = continuing tachycardia True-positive test = tachycardia termination of an arrhythmia with obligatory AV nodal conduction (e.g., AV nodal reentry) False-positive test = tachycardia termination of an arrhythmia that does not require AV nodal conduction (e.g., ventricular tachycardia) Sensitivity

= $

Specificity

= E = 92%

Positive predictive accuracy

= $

= 70%

= g5%

physiologic testing were studied between June 1986 and October 1987. All patients provided informed consent as approved by the University of Western Ontario Human Ethics Committee. Hemodynamic stability was defined by a systolic blood pressure greater than 80 mm Hg and no clinical evidence of cerebral or myocardial &hernia. Only patients with a QRS complex greater than 120 msec in duration were studied. The electrophysiologic study was carried out as described previously [ll] with four catheters used in patients with supraventricular tachycardia, and three catheters in patients with suspected ventricular arrhythmias (high right atrial, His bundle, and right ventricular apex). Five surface electrocardiogram leads and the high right atrial, His bundle, and right ventricular electrograms were monitored continuously. Arterial pressure was monitored noninvasively using an automated arm cuff device with oscillometric determination of systolic, diastolic, and mean pressure recorded up to every 15 seconds at the maximum frequency. In two patients, arterial pressure was also monitored with a femoral arterial cannula. During stable tachycardia (longer than 5 minutes), adenosine triphosphate was administered as a 20-mg intravenous bolus via a peripheral vein. This was immediately followed by a rapid lo-mL flush of normal saline. This dosing regimen is commonly used and reportedly effective in termination of supraventricular tachycardia involving the AV node [12]. The intracardiac electrograms and surface electrocardiogram were recorded for a 5-minute period following administration of the drug. In cases in which the arrhythmia failed to terminate, standard therapeutic maneuvers were utilized to terminate the tachycardia. Quantitative data for cycle lengths, age, and blood pressure are expressed as mean f SD. Comparisons of quantitative data between groups were made using a one-way analysis of variance followed by an unpaired Student’s t-test. Differences in proportions of patients showing tachycardia termination were determined using a chi-square test with the Yates correction. Sensitivity, specificity, and positive predictive accuracy were determined as described in Table I. RESULTS The patient population was classified into three groups based on tachycardia mechanism. Group 1 consisted of patients with ventricular tachycardia; Group 338

April

1990

The American

Journal

of Medicine

Volume

88

Group 1 Fourteen patients with a mean age of 50.6 f 19 years and ventricular tachycardia with cycle length 326 f 67 msec were studied. In one patient, ventricular tachycardia related to underlying ischemic heart disease terminated in response to adenosine triphosphate (Figure 1). In the other 13 patients, no significant change in ventricular tachycardia cycle length was observed (control 330 f 64 msec versus 326 f 68 msec, p = NS). The greatest change in ventricular tachycardia cycle length was a 30-msec decrease. Ventriculoatrial (VA) conduction block occurred transiently following adenosine triphosphate in two of four patients in whom 1:l conduction was present in the control state. Sinus cycle length was observed to prolong in patients in whom there was VA dissociation (Figure 2). Two patients had ventricular tachycardia that was terminated by intravenous verapamil (5 mg), but did not show any tachycardia cycle length prolongation or termination in response to adenosine triphosphate. Both patients subsequently were treated successfully with oral verapamil. Group 2 Group 2 consisted of 10 patients with atria1 arrhythmias including atria1 flutter, atria1 fibrillation, and ectopic atria1 tachycardia. The mean age of the pa-

0A

SE362066

I”

I SECOND

I

H RA !-j-,-,1’-

Figure 1. Effects of adenosine triphosphate (20 mg intravenous) on ventricular tachycardia. Surface electrocardiogram leads I and VI, right ventricular electrograms (RV), and high right atrial electrograms (HRA) are shown. A, ventricular tachycardia with VA dissociation and a uniform QRS morphology and stable cycle length of 275 msec. B, recorded 10 seconds after adenosine triphosphate administration Instability of cycle length and QRS morphology are observed before tachycardia termination.

WIDE QRS TACHYCARDIA

AND ADENOSINE

TRIPHOSPHATE

/ SHARMA

ET AL

Figure 2. Effects

of adenosine triphosphate (20 mg intravenous) on ventricular tachycardia. Surface electrocardiogram leads I, II, Ill, the right ventricular electrogram (RV), and the high right atrial electrogram (HRA) are shown. Ventricular cycle length remains constant at 270 msec. There is VA dissociation and atrial cycle length prolongs from 470 msec to 710 msec. CONTROL

Figure 3. Adenosine

triphosphate effect on atrial fibrillation with antegrade conduction via a right free wall AV accessory pathway. Surface electrocardiographic leads I and Ill are shown in control and 20 seconds after administration of adenosine triphosphate (ATP) (20 mg intravenous). The shortest R-R interval between consecutive preexcited QRS complexes decreased from 265 msec in control to 235 msec after adenosine triphosphate.

MS 384307

I

c

, I SECOND

POST ATP 20mg I AJJJuAMA/1wn

tients was 32.8 f 16 years and the mean ventricular cycle length was 346 f 82 msec. Antegrade accessory pathway conduction was present in nine patients. One patient had an ectopic atria1 tachycardia with bundle branch block. In this group of patients, tachycardia termination was observed only in the patient with ectopic atria1 tachycardia of cycle length 300 msec. Antegrade accessory pathway conduction was transiently enhanced by adenosine triphosphate in all nine patients with preexcitation. The shortest R-R interval in control was 243 f 100 msec and this shortened to 203 f 64 msec (p <0.02) after adenosine triphosphate (Figure 3). Similarly, the average R-R interval between all QRS complexes shortened from 351 f 84 msec to 317 f 82 msec (p
TABLE II Patient Population

Number Age (years) (mean f SD) Tachycardia cycle length (msec) Arrhythmia mechanism (number of patients)

Group 1

Group 2

Group 3

50.61: 19

32.81: 16

31? 6

326 f 67

346 f 82

302 f 52

Ventricular tachycardia

(14)

(8) Atrial flutter with preexcitation (1) Ectopic atrial tachycardia with LBBB

Group 3 Group 3 patients had reentrant tachycardias in which there was obligatory involvement of the AV node, although a number of different arrhythmia mechanisms were involved (Table II). The mean age of these patients was 31 f 6 years, with a mean tachycardia cycle length of 302 f 52 msec. Tachycardia terminated with AV nodal block in three of five patients with orthodromic AV reciprocating tachycardia. AV nodal reentry terminated in the antegrade slow pathway in two of two patients. One patient had a tachycardia utilizing a nodoventricular accessory pathway. Criteria for nodoventricular accessory pathway diagnosis were standard electrophysiologic criteria [13]. In this single patient with antegrade conduction via a nodoventricular accessory pathway during tachycardia, block occurred within the antegrade limb of the tachycardia circuit, as suggested by tachycardia termination following an atria1 depolarization (Figure 4). The one patient with antidromic tachycardia involving a left free wall AV accessory pathway had retrograde AV nodal block leading to tachycardia termination.

Atrial fibrillaAVRT with BBB tion with ven(5) tricular preexcitation

(1) Associated cardiac disease lschemic heart disease Idiopathic cardiomyopathy Arrhythmogenic RV dysplasia Tetralogy of Fallot Ebstein’s anomaly None

AVRT with multipleAP(1) AV nodal reentry with BBB (2) Nodoventricular pathway (1) Antidromic AVRT (1)

9

1 1 1

2

A

10

LBBB = left bundle branch block; AVRT = atrloventricular reciprocating tachycardia; Al = atrioventricular accessory pathway; BBB = bundle branch block.

Failure of tachycardia termination was observed in three cases. Two patients had orthodromic AV reciprocating tachycardia associated with bundle branch block. In one of these patients with a posteroseptal April

1990

The American

Journal

of Medicine

Volume

88

339

MB 374226

HBE I SECOND

Figure 4. Termination of tachycardia associated with a nodoventricular accessory pathway by adenosine triphosphate. Surface electrocardiographic leads I, II, and III and right ventricular (RV), high right atrial (HRA), and His bundle electrograms (HBE) are shown. The antegrade AV interval prolonged progressively, culminating in block of antegrade AV conduction. Figure 5. Adenosine triphosphate (ATPQnduced normalization of bundle branch block and cycle length alternans in a patient with AV nodal reentry. During induced and spontaneous AV nodal reentry, a stable rate-related left bundle branch block was present. Following adenosine triphosphate this normalized (A). Star indicates normalization of the rate-related left bundle branch block. B, recorded 5 seconds later, and alternans of AH intervals is observed prior to tachycardia termination. HBE = His bundle electrograms.

accessory pathway, the antegrade AH interval prolonged by 45 msec and there was concomitant shortening of the VA interval of 45 msec resulting in an unchanged cycle length. The mechanism of the VA interval decrease is uncertain, as ventricular premature stimulation did not provide evidence of decremental retrograde conduction in this patient. In the other patient there was no change in the intervals. In the final case in which the arrhythmia failed to terminate, the arrhythmia mechanism was orthodromic tachycardia with bystander participation of a second AV accessory pathway. After adenosine triphosphate, the tachycardia abruptly changed to an antidromic tachycardia involving the second AV accessory pathway. Other observations included slowing of tachycardia (AH prolongation) followed by normalization of a rate-related bundle branch block prior to tachycardia termination in one of seven cases (Figure 5), and marked cycle length alternans due to AH interval alternans after adenosine triphosphate in both patients with AV nodal reentry (Figure 5). Blood Pressure Changes Hemodynamic collapse was not observed in any patient. The side effects included transient (5 to 10 seconds) flushing, chest pain, and hyperpnea. In Group 1 patients, the blood pressure was 137185 f 29115 mm Hg (mean systolic/mean diastolic, f SD) in sinus rhythm at a rate of 78 f 17 beats/minute. The minimum blood pressure during ventricular tachycardia was 100/71 f 29/21 mm Hg. During adenosine triphosphate bolus, the blood pressure was 110/81 f 35/18 mm Hg. No significant change in blood pressure was observed in the group. However, in three patients with initial pressures between 80 and 90 mm Hg, the systol340

April

1990

The American

Journal

of Medicine

Volume

88

ic pressure dropped to between 70 and 80 mm Hg for less than 1 minute after the adenosine triphosphate bolus. In two patients, intra-arterial lines were used to monitor pressure. This confirmed the absence of transient hypotension between arm cuff measurements. Patients with supraventricular arrhythmias (Group 2) had a mean blood pressure of 137/78 f 20/9 mm Hg in sinus rhythm at a rate of 75 f 15 beats/minute. During tachycardia the blood pressure dropped to 106/71 f 32/15 mm Hg. At the time of adenosine triphosphate administration, the blood pressure was 115/68 f 33/19 mm Hg (p = NS, versus during tachycardia). The termination of tachycardia with its resultant change in blood pressure precluded comparative analysis of data in Group 3. In all but three patients, the minimum blood pressure was observed within seconds of the induction of tachycardia and adenosine triphosphate did not produce any detectable further hypotension. Diagnostic Utility As a diagnostic test to indicate participation of the AV node, tachycardia termination with this dose of adenosine triphosphate had a sensitivity of 70%, specificity of 92%, and a positive predictive accuracy of 85% (Table I). Further observations that have diagnostic utility include the induction of VA block in two of four patients with ventricular tachycardia and 1:l VA conduction, and the normalization of a rate-related bundle branch block (Table III). COMMENTS Purinergic nucleotides have potent effects on cardiac electrophysiologic properties and may also produce

WIDE QRS TACHYCARDIA AND ADENOSINE TRIPHDSPHATE / SHARMA ET AL

vasodilatation. Of these agents, adenosine has been investigated the most. It has negative chronotropic effects both in vitro [14] and in uivo [15]. However, the effect of the purinergic nucleotides on the AV node is of particular relevance to the termination of supraventricular tachycardias [16]. In the guinea pig atrium depolarized by high extracellular potassium concentrations, adenosine has a depressant effect on action potentials dependent on slow inward calcium currents [17], and this action could lead to block of conduction. The negative dromotropic effects of adenosine on the AV node are mediated by a direct action on extracellular purinergic Ai receptors, rather than indirect metabolic effects [16]. In contrast, adenosine and adenosine triphosphate have minor effects in vitro on action potentials in ventricular tissue [18]. Adenosine triphosphate is rapidly hydrolyzed to adenosine, which is in turn converted to inosine, which is electrically inactive. This rapid metabolism leads to a very short duration of action following an intravenous bolus injection [lo]. On the basis of these properties, adenosine triphosphate would be anticipated to have similar electrophysiologic effects as verapamil, but with a much shorter half-life. Consistent with this, several studies have shown that adenosine and adenosine triphosphate are highly effective in terminating supraventricular tachycardia [8-lo] and offer a therapeutic alternative to verapamil. A more serious problem in the therapy of supraventricular arrhythmias occurs when ventricular tachycardia is misdiagnosed as supraventricular tachycardia with aberrancy. In a recent study, 35% of patients with ventricular tachycardia were initially misdiagnosed in this manner [19]. Misdiagnosis of patients with atria1 fibrillation and antegrade accessory pathway conduction was also common. This misdiagnosis currently may then lead to the administration of verapamil, which is usually ineffective in terminating ventricular tachycardia [20] or atria1 fibrillation [4]. Furthermore, in this setting, verapamil may result in hemodynamic collapse [2,3] due to its vasodilating and negative inotropic effects, which are greater in the presence of preexisting ventricular dysfunction [21], as often occurs in patients with ventricular tachycardia. This misdiagnosis continues to occur despite welldefined electrocardiographic features for diagnosing ventricular tachycardia [22]. Therefore, agents such as the purinergic nucleotides, which have a similar therapeutic profile as verapamil but lack the negative inotropic effects and have an ultra-short duration action, may be much safer to use in the treatment of a previously undiagnosed wide QRS complex tachycardia that is believed to be supraventricular. To assess this hypothesis, the safety, therapeutic efficacy, and diagnostic utility of adenosine triphosphate were tested in 34 consecutive patients with a wide QRS complex tachycardia undergoing invasive electrophysiologic testing. To simulate the events that would occur in the emergency room setting, a single dose of adenosine triphosphate was administered, and no attempt was made to determine the reproducibility of the response. Ventricular tachycardia was terminated in only one case (7%). In this case, the cycle length and morphology changed before termination (Figure 1). The mechanism of this tachycardia termination is uncertain. Although the effects of adenosine triphosphate on ventricular tachycardia have not been systematically assessed, Griffith et al [23] reported termi-

TABLE III Termination of Wide QRS Complex Tachycardia by Adenosine Triphosphate Group 1 Termination (number of patients) Antegrade AV block RetrogradeVA block

Group 2

Group 3

l/14

l/10

7/10

2/4

-

6/10 l/10

p‘Co.001

nation of one of 17 cases of ventricular tachycardia with adenosine. Similarly, Lerman et al [24] observed adenosine to terminate exercise-induced ventricular tachycardia. In contrast, in our study, the single patient with ventricular tachycardia terminated by adenosine triphosphate differed from the cases of adenosine-terminated ventricular tachycardia described by Lerman et al [24]. Our patient was 48 years old and had ventricular tachycardia related to a previous myocardial infarction, and the arrhythmia was not inducible by exercise. Two patients in our study had ventricular tachycardia that was terminated by intravenous verapamil, but lacked the other features described by Lerman et al [24], and did not have termination of tachycardia by adenosine triphosphate. This is not unexpected, as there is a spectrum of patients with ventricular tachycardia terminated by verapamil [25,26]. As a consequence, termination of a wide QRS complex tachycardia by adenosine triphosphate does not absolutely rule out ventricular tachycardia. Administration of this agent was safe in the electrophysiology laboratory setting, as no hemodynamic collapse was observed, and in only three cases did the systolic blood pressure drop transiently below 80 mm Hg. However, this does suggest that blood pressure should be monitored frequently immediately after the administration of adenosine triphosphate to any patient with a systolic pressure between 80 and 90 mm Hg. The infrequent and small decrease in blood pressure observed in this study is consistent with the findings of Biaggioni et al [27]. They observed that adenosine in conscious humans increases systolic pressure in doses that are clinically tolerated. Adenosine triphosphate transiently terminated the one case of ectopic atria1 tachycardia. The tachycardia was otherwise incessant in the absence of therapy, but was suppressed by intravenous propafenone. Unfortunately, the mechanism of this tachycardia was not further delineated, but a triggered mechanism as proposed by Lerman et al [24] would be consistent with the termination by adenosine triphosphate. Termination of atria1 flutter by adenosine has been observed [23], but did not occur in this series. However, the ventricular response (average and shortest R-R) to atria1 fibrillation and flutter with antegrade accessory pathway conduction was enhanced, suggesting that the refractory period of the accessory pathway was transiently shortened. Potential mechanisms for this electrophysiologic change include a direct effect of the drug on the accessory pathway, or an indirect effect via enhanced sympathetic activity [27]. Adenosine enhances outward potassium current in atria1 tissue [14], but has relatively minor effects on ventricular action potentials [18]. Thus, a direct shortening of accessory pathway refractoriness by adenosine triphosphate would suggest that the accessory pathway has properApril 1990 The American Journal of Medicine

Volume 88

341

WIDE QRS TACHYCARDIA

AND ADENOSINE

TRIPHOSPHATE

/ SHARMA

ties more akin to atria1 muscle. This observation also raises the potential for induction of ventricular fibrillation by the enhanced ventricular response. Although this has been noted with verapamil [4,5], the short duration of action of adenosine triphosphate makes this a less likely event, which has yet to be observed. The findings on supraventricular tachycardia are consistent with previous studies [8-10,121, namely that adenosine triphosphate is highly effective in inducing AV nodal block. Failure to induce AV block in all cases was likely related to the dose used. In one case, however, failure to terminate AV reciprocating tachycardia was related to the presence of two accessory pathways that both participated in maintenance of tachycardia when AV nodal block occurred. Tachycardia termination with block within the antegrade limb of the tachycardia circuit in the one patient with a “nodoventricular” accessory pathway was unexpected. The anatomic substrate for these arrhythmias has recently been described to be due to an AV accessory pathway on the right ventricular wall that has unidirectional antegrade decremental conduction to the right ventricular apex [13]. In a previous study, nondecremental antegrade accessory pathway conduction was shown to rarely be blocked by adenosine triphosphate [28]. The potent AV nodal blocking effects of adenosine triphosphate suggest that tachycardia termination following adenosine triphosphate may be used diagnostically to indicate obligatory participation of the AV node in the tachycardia circuit. The sensitivity was only 70%. However, one patient with multiple accessory pathways had AV nodal block without tachycardia termination. In addition, higher doses of adenosine triphosphate have greater negative dromotropic effects on the AV node [lo], and may have increased sensitivity. The specificity at this dose of adenosine triphosphate was good at 92% and it is possible that higher doses would have reduced the specificity of the response. Among the patients with regular rates and wide QRS complex tachycardias, only four of 24 continued to have a 1:l relationship between the atria and ventricles (two ventricular tachycardia and two AV reciprocating tachycardia) and thus could not be diagnosed by the presence of AV dissociation. Normalization of a rate-related bundle branch block also suggested the diagnosis of a supraventricular tachycardia in one case. The inappropriate use of verapamil or purinergic agonists in patients with arrhythmias not dependent on AV nodal conduction is to be avoided. Careful analysis of the 12-lead electrocardiogram and clinical history usually can enable the diagnosis of ventricular tachycardia [3,22], in which case verapamil and purinergic agonists should not be administered. The presence of irregular R-R intervals suggests atria1 fibrillation and should also militate against the use of these agents. However, this study suggests that the purinergic agonists may be safer to use for acute termination of arrhythmias than verapamil in settings in which misdiagnosis may occur. This hypothesis is based on the reports of adverse responses to verapamil in patients with ventricular tachycardia [2] or Wolff-Parkinson-white syndrome [4], even when verapamil is administered in the electrophysiology laboratory. The

342

April 1990

The American

Journal

of Medicine

Volume

88

ET AL

failure to terminate a wide QRS complex tachycardia with 20 mg of intravenous adenosine triphosphate should lead the physician to reconsider the arrhythmia diagnosis to include an arrhythmia mechanism not dependent on AV nodal conduction and then administer the appropriate therapy. These data suggest that adenosine triphosphate is a useful drug for primary therapy and diagnosis of wide QRS complex supraventricular tachycardia, even when the mechanism is uncertain. However, empiric pharmacologic testing should not replace accurate diagnosis made on the basis of the clinical history and 12-lead electrocardiogram. These data suggest that a controlled clinical trial of the emergency room use of adenosine triphosphate for reentrant supraventricular tachycardia is warranted.

ACKNOWLEDGMENT We are grateful to Mrs. D. Vigna, Mr. S. Ting. and Miss S. Paradis for assistance in preparing the manuscript, to A. MacDonald and N. Smith for nursing care, and to the research fellows who have assisted in patient care.

REFERENCES 1. Sung RJ, Elser B, McAllister RG: Intravenous verapamil for termination of reentrant supraventricular tachycardias. Ann Intern Med 1980; 93: 682-689. 2. Buxton AE, Marchlrnski FE, Doherty JU. Flores B. Josephson ME: Hazards of intravenous verapamil for sustarned ventricular tachycardia. Am J Cardiol 1987; 59: 1107-1110. 3. Stewart RB, Bardy GH, Greene HL: Wide complex tachycardia: mrsdragnosis and outcome after emergent therapy. Ann Intern Med 1986; 104: 766-771. 4. Gulamhusein S. Ko P, Carruthers SG. Klein GJ: Acceleration of the ventricular response during atrial fibrillation in the Wolff-Parkinson-White syndrome after verapamrl. Circulation 1982; 65: 348-354. 5. McGovern B. Garan H, Ruskin JN: Precipitation of cardiac arrest by verapamrl In patients with Wolff-Parkinson-White syndrome. Ann Intern Med 1986; 104: 791794. 6. Dancy M, Camm AJ, Ward D: Misdiagnosis of chronic recurrent ventricular tachycardia. Lancet 1985; 2: 320-332. 7. Nathan AW, Elstob JE, Camm AJ: The mrsdiagnosis of ventricular tachycardia. Br Heart J 1986; 55: 513-514. 8. DiMarco JP, Sellars TD, Berne RM. West GA, Bellardinelli L: Adenosine: electrophysiologic effects and therapeubc use for terminating paroxysmal supraventricular tachycardia. Circulatron 1983; 68: 1254-1263. 9. Belhassen B, PellegA, Shoshant D, Geva B, Laniado S: Electrophysiologic effects of adenosine-5’Yriphosphate on atrioventricular reentrant tachycardia. Circulation 1983; 68: 827-833. 10. Sharma AD, Klern GJ: Comparative quantitative electrophysiologic effects of adenosine triphosphate on the sinus node and atrioventncular node. Am J Cardiol 1988; 61: 330-335. 11. Sharma AD, Klein GJ, Guiraudon GM, Milstein S: Atrial fibrillation in pabents with Wolff-Parkinson-White syndrome: incidence after surgical ablation of the accessory pathway. Circulation 1985; 73: 161-169. 12. Puech P. Sassine A, Munoz A, et al: Electrophysrologic effects of punnes: clinlcal applrcations. In: Zipes DP. Jalife J. eds. Cardiac electrophysiology and arrhythmias. Orlando: Grune and Stratton, 1985; 4433450. 13. Klein GJ. Guiraudon GM, Kerr CR, eta/:“Nodoventricular” accessory pathway: evidence for a distinct atrioventricular pathway with atrloventricular node-like properties. J Am Coll Cardiol 1988: 11: 1035-1040. i4.‘West GA, Belardinelli L: Sinus slowing and pacemaker shift caused by adenosine in the rabbit SA node. Pflugers Arch 1985; 403: 66-74. 15. PellegA, Belhassen B. llia R. Laniado S: Comparativeelectrophysiologiceffects of adenosine triphosphate and adenosrne in the canrne heart: influence of atropine, propranolol. vagotomy, dipyridamole and aminophylline. Am J Cardiol 1985; 55: 571-576. 16. Belardinelli L, Shryock J, WestAG, Clemo HF, DiMarco JP. Berne RM: Effects of adenosine and adenosine nucleotides on the atrioventricular node of the Isolated guinea pig heart. Circulation 1984; 70: 1083-1091. 17. Knabb MT, Rubio R. Berne RM: Potentiation of slow acbon potentials with theophylline or “macro” adenosrne deaminase. Am J Physiol 1983; 13: H454H457. 18. Rosen MR, Danrlo P. Weiss RM: Actions of adenosrne on normal and abnormal impulse inrtiation in canine ventricle. Am J Physrol 1983; 13: H715-H721. 19. Akhtar M. Shenasa M, Jazayeri M, Caceres J, Tchou PJ: Wide QRS complex tachycardia. Reappraisal of a common clinical problem. Ann Intern Med 1988; 109: 905-912.

WIDE QRS TACHYCARDIA 20. Mason JW, Swerdlow CD, Mrtchell LB: Efftcacy of verapamil in chronic, recurrent ventricular tachycardra. Am J Cardrol 1983; 51: 1614-1617. 21. Chew CYC. Hetch HS, Collett JT. McAllister RG, Singh EN: Influences of severity of ventrrcular dysfunctron or hemodynamic responses to rntravenously administered verapamrl In rschemrc heart disease. Am J Cardtol 1981; 47: 917-922. 22. Wellens HJJ: The wade QRS tachycardias (letter). Ann Intern Med 1986; 104: 879. 23. Griffith MJ, Lanker NJ, Ward D, Camm Al: Adenosrne rn the dragnosis of broad complex tachycardia. Lancet 1988; 1: 672-675. 24. Lerman BB, Belardinelli L, West AG, Berne RM, DiMarco JP: Adenosine sensitive ventrrcular tachycardia: evidence suggesting cyclic AMP-mediated triggered

AND ADENOSINE

TRIPHOSPHATE

/ SHARMA

ET AL

actmty. Circulation 1986; 74: 270-280. 25. Klein GJ, Millman PJ, Yee R: Recurrent ventrtcular tachycardia responsrve to verapamrl. PACE 1984; 7: 938-948 26. Nguyen BX, Yang P-T, Huycke E. Sung RJ: Verapamil and ventricular tachyarrhythmias. PACE 1987; 10: 571-578. 27. Biaggroni I, Onrot J, Hollister AS, Robertson D: Cardiovascular effects of adenosine rnfusron In man and their modulation by dipyridamole. Life SCI 1986; 39: 2229-2236. 28. Rinne C, Sharma AD, Klern GJ, Yee R: Comparative effects of adenosine tn. phosphate on accessory pathway and atrioventricular nodal conductron. Am Heart J 1988; 115: 1042-1047.

April 1990

The American

Journal

of Medicine

Volume

88

343