Electrocardiographic changes and cardiac arrhythmias in patients receiving psychotropic drugs

Electrocardiographic

Changes and Cardiac Arrhythmias

Patients Receiving Psychotropic

NOBLE 0. FOWLER, MD, FACC DAVID MCCALL, MD TE-CHUAN CHOU, MD, FACC JOHN C. HOLMES, MD, FACC IRWIN B. HANENSON, MD Cincinnati,

Ohio

in

Drugs

Eight patients had cardiac manifestations that were life-threatening in five while taking psychotropic drugs, either phenothiazines or tricyclic antidepressants. Although most patients were receiving several drugs, Mellaril (thioridazine) appeared to be responsible for five cases of ventricular tachycardia, one of which was fatal in a 35 year old woman. Supraventricular tachycardia developed in one patient receiving Thorazine (chlorpromazine). Aventyl (nortriptyline) and Elavil (amitriptyline) each produced left bundle branch block in a 73 year old woman. Electrocardiographic T and U wave abnormalities were present in most patients. The ventricular arrhythmias responded to intravenous administration of lidocaine and to direct current electric shock; ventricular pacing was required in some instances and intravenous administration of propranolol combined with ventricular pacing in one. The tachyarrhythmias generally subsided within 48 hours after administration of the drugs was stopped. Five of the eight patients were 50 years of age or younger; only one clearly had antecedent heart disease. Major cardiac arrhythmias are a potential hazard in patients without heart disease who are receiving customary therapeutic doses of psychotropic drugs. A prospective clinical trial is suggested to quantify the risk of cardiac complications to patients receiving phenothiazines or tricyclic antidepressant drugs.

Alexander and Niiiol reported cardiovascular complications in seven young patients who were receiving psychotropic drugs. Data from hospital-based surveys have both confirmed2 and refuted3 the cardiotoxicity of tricyclic antidepressant drugs. This paper describes cardiac manifestations produced by psychotropic drugs in eight patients observed within the last 6 years. These complications were often lifethreatening arrhythmias that developed in patients with emotional disorders who were receiving customary therapeutic doses of these agents. The potential dangers of psychotropic drugs, especially in large doses, deserve greater emphasis and suggest the need for a prospective study to quantitate the risk involved. Case

From the Division of Cardiology and the Division of Clinical Pharmacology, Department of Medicine, University of Cincinnati College of Medicine, Cincinnati, Ohio. Manuscript accepted July 2, 1975. Address for reprints: Noble 0. Fowler, MD, Division of Cardiology, College of Medicine, University of Cincinnati, 231 Bethesda, Cincinnati, Ohio 45267.

Descriptions

The cases described are summarized in Table I. Case 1: A 35 year old woman,was admitted to the Cincinnati General Hospital on August 17, 1968 after attempted suicide. A depressive reaction was diagnosed. There was no history of neurologic or cardiac disease. She was given the following medications orally: MellariP (thioridazine hydrochloride), 250 mg four times daily; Stelazinea (trifluoperazine), 10 mg four times daily; Cogenti@ (benztropine mesylate), 1 mg twice daily; Thorazinea (chlorpromazine hydrochloride) Spansules Q, 300 mg at bedtime. Thirty-six days later, while still in the hospital, she had seizures with urinary and fecal incontinence. A physician observed her pulse to be irregular during a seizure. An electrocardiogram revealed a prolonged Q-T interval with prominent U waves, especially in leads II, III, aVF and Vq to Va (Fig. 1).

February

1976

The American

Journal of CARDIOLOGY

Volume

37

223

ARRHYTHMIAS AFTER PSYCHOTROPIC DRUGS-FOWLER

TABLE

ET AL.

I

Summary

of Eight Cases

Drugs Mellaril fthioridazine); Thorazine Ichlorpromazine); Stelazine (trifluoperazine)

Daily Dose (Ills) 1000 300

Prior Heart Disease

Manifestations

None

VT; fatal VF

24 {death)

IVCD

VT

24

40 400800

2

24

Mellaril

3

30

Mellaril; Serentil (mesoridazinc)

1600 800

None

VF; RBBB; VT; abnormal T-U complexes

4

64

Mellaril; Serentil

1200 300

None

VF: A-V block

48

5

19

Tofranil fimipramine); Stelazine

Overdose

None

Bizarre T-U complexes

24

6

73

1st episode: Thorazine; Aventyl (nortrrptyline); 2nd episode: Elavil (amitriptyline)

150

HCVD

LBBB; heart failure

<5 days

HCVD

LBBB

<48

75 12.5

5 days

7

50

Thorazine

400

None

SVT

48

8

71

Mellaril; Serentil

200 75

Not documented

VT

48

HCVD = hypertensive cardrovascular disease; IVCD = intraventricular conduction defect; LBBB = left bundle branch block; bundle branch block; SVT = supraventricular tachycardia; VF = ventricular fibrillation; VT = ventricular tachycardia.

Physical examination, including cardiac examination, was normal. Chest roentgenogram was normal. Serum potassium was 3.9 mEq/liter and other serum electrolytes, as well as serum enzymes, were normal. She had repeated episodes of documented ventricular tachycardia and fibrillation, some of which were terminated by a precordial blow (Fig. 1). Despite intravenous administration of Xylocainea (lidocaine) and Pronestyle (procainamide hydrochloride), cardiac pacing and direct current countershock, she died on the same day. A coroner’s autopsy revealed no gross or microscopic cardiac abnormality. Case 2: A 24 year old man had been treated in the psychiatric clinic of the Cincinnati General Hospital since November 22, 1969 because of a schizophrenic reaction. An electrocardiogram 2 days before admission had shown a QRS interval of 0.12 second with a pattern of nonspecific intraventricular conduction defect. On admission, he was treated with Mellaril, 100 mg four times a day, although he admitted taking 600 to 800 mg a day on occasion. He was taken by ambulance to the emergency room on February 23, 1970 because of weakness, dizziness, nausea and vomiting. He was found to have ventricular tachycardia (rate 186/min) (Fig. 2). A bolus injection of 100 mg of lidocaine was given intravenously and was followed by direct current countershock, which converted the rhythm to sinus rhythm. The patient was said to have had a cardiac murmur since childhood. He denied having any symptoms of cardiovascular disease before this episode. Physical examination during sinus rhythm revealed a blood pressure of 110/70 mm Hg. The heart was not enlarged and the heart sounds were normal. A grade l/6 systolic ejection murmur heard over the pulmonary area was thought to be innocent. The lungs were clear. Laboratory examination revealed values for sodium and potassium of

224

Duration (hours)

February 1976

The American Journal of CARDIOLOGY

RBBB

= right

135 and 5.7 mEq/liter, respectively. A chest roentgenogram was normal. The electrocardiogram after the resumption of normal sinus rhythm (Fig. 2) showed a P-R interval of 0.20 second and a QRS duration of 0.12 second. There were prominent U waves in all leads. A follow-up tracing on September 24, 1970 showed a normal S-T segment and T waves. The prominent U waves were no longer present. The QRS duration remained the same. Case 3: A 30 year old woman was admitted to a community hospital on December 2, 1971 shortly after an episode of syncope at Longview State Hospital. The patient, a schoolteacher, was receiving the following medications at the time of syncope: Mellaril, 400 mg four times daily; Prolixine (fluphenazine hydrochloride), 2 cc intramuscularly every 2 weeks; Serentile (mesoridazine), 200 mg four times daily; and Cogentin. There was no past history of cardiac disease. When the patient collapsed on the day of admission, an electrocardiogram disclosed ventricular tachycardia, followed by ventricular fibrillation. Direct current countershock successfully restored sinus rhythm, and the patient was sent to a local private hospital. On admission, the general physical findings were normal except for a flattened emotional affect. Blood count, urinalysis and chest X-ray film were normal. On admission, serum levels of glutamic oxaloacetic acid (GOT) and lactic dehydrogenase (LDH) were elevated but those of creatine phosphokinase (CPK) were normal. The serum enzyme elevations were believed to be due to the arrhythmia and hepatic damage. Serum potassium was 3.4 mEq/liter on admission. She was treated with an intravenous infusion of lidocaine; other antiarrhythmic agents were not used. The admission electrocardiogram of December 2 showed ventricular tachycardia and, later the same day, borderline first degree atrioventricular (A-V) block and intermittent

Volume 37

ARRHYTHMIAS AFTER PSYCHOTROPIC DRUGS-FOWLER

Lead

ET AL.

II

2/24/70 Lead

FIGURE 1. Case 1. Electrocardiogram revealing S-T and T wave abnormalities and large U waves in a 35 year old woman with a fatal ventricular arrhythmia after treatment with Mellaril, Thorazine and Stelazine. Response of a ventricular tachycardia to a sternal blow is shown at bottom.

I

Lead

II

Lead

III

aVR

aVL

aVF

FIGURE 2. Case 2. Ventricular tachycardia and T and U wave normalities in a 24 year old man after Mellaril therapy.

ab-

right bundle branch block with abnormal T-U complexes (Fig. 3). Electrocardiograms of December 4 and 5 showed no bundle branch block but continued T-U abnormalities. A recording on January 11, 1972 was normal, and the P-R interval was decreased from 0.20 to 0.13 second. Case 4: A 64 year old woman had been confined in Longview State Hospital for 34 years because of manicdepressive psychosis. In April 1971, Mellaril, 400 mg, and Serentil, 100 mg, were given orally three times daily. On December 30, the patient suddenly lost consciousness. An electrocardiogram after her resuscitation showed long periods of sinus arrest. Atropine was given intravenously; ventricular bigeminy developed, followed by ventricular fibrillation. Direct current countershock and intravenous injections of lidocaine were given; there was intermittent complete A-V block interrupted by bursts of multifocal ventricular premature beats and ventricular tachycardia (Fig. 4). On admission to this hospital, she was treated with temporary pervenous pacing of the right ventricle and lidocaine, intravenously, 3 mg/min. On the 3rd hospital day, she regained consciousness and had normal sinus rhythm. She had no further cardiac arrhythmias and on January 7,1972 was returned from this hospital to Longview State Hospital. Case 5: A 19 year old woman was admitted to Cincinnati General Hospital on February 23, 1972 after ingestion of 50 tablets of Tofranile (imipramine hydrochloride), 50 mg, and 20 tablets of Stelazine, 5 mg, in a suicide attempt. She had been discharged from the hospital the previous day with the diagnosis of chronic depressive psychosis. The patient was semicomatose. Blood pressure was 110/70 mm

J,

VI

FIGURE 3. Case 3. Right bundle branch block and S-T and T-U segment abnormalities in a 30 year old woman after therapy with Mellaril and Serentil tracing (tracing retouched).

Hg, pulse 80/min. She responded to light sensory stimuli. General examination was normal otherwise. Chest roentgenogram, serum electrolytes, blood chemistry values and complete blood count were normal; serum potassium was 3.7 mEq/liter. An admission electrocardiogram showed sinus tachycardia with a large T-U complex. The patient was observed to have several brief episodes of generalized tremor without loss of consciousness. During these periods the electrocardiogram showed gross abnormalities of the T-U complex (Fig. 5). After each seizure, the electrocardiogram was normal except for premature ventricular systoles (Fig. 5). Later that day, the electrocardiogram remained normal. On the following day, the patient was alert and was transferred to the psychiatric service. Case 6: A 73 year old woman had been treated for hypertensive cardiovascular disease. She was admitted to a local

VENTRICULAR TACHYCARDIA

c*

FIGURE 4. Case 4. Ventricular tachycardia in a 64 year old woman rhythmia was suppressed by an electronic ventricular pacemaker.

v6

v3

ELECTRONICALLY PACED VENTRICULAR RHYTHM

after Mellaril and Serentil therapy.

February 1976

In the latter portion of the tracing, the ar-

The American Journal of CARDIOLOGY

Volume 37

225

ARRHYTHMIAS AFTER PSYCHOTROPIC DRUGS-FOWLER

ET AL

of March 21 and 25 showed left axis deviation with normal intraventricular conduction time. Electrocardiograms on April 1 and 4 indicated left bundle branch block, and she had clinical signs of early heart failure. The patient was given digitalis and a diuretic agent. Administration of Thorazine and Aventyl was discontinued on April 1, and on April 6 the signs of heart failure had cleared and an electrocardiogram no longer indicated bundle branch block. On October 13, 1973, because of another episode of de: pression, the patient was given Elavil@ (amytriptyline hydrochloride), 12.5 mg daily. An electrocardiogram showed left axis deviation with normal intraventricular conduction time. Eight days later, left bundle branch block appeared. Administration of Elavil was stopped, and 5 days later an electrocardiogram no longer indicated left bundle branch block.

LIL Pre seizure

LII Post -5eizure

L AVR Preseizure

L AVR Post-seizure FIGURE 5. Case 5. Bizarre T-U segment abnormalities and ventricular premature beats in a 19 year old woman with convulsive seizures after overdoses of Stelazine and Tofranil.

private hospital on March 21, 1972 with agitated depression and was given Thorazine, 50 mg orally three times daily; Valium@ (diazepam), 10 mg four times daily; and Tuinal@ (sodium secobarbital-amobarbital) and Phenergan@ (promethazine hydrochloride) at bedtime. On March 22, 1972, she began to receive Aventyl@ (nortriptyline hydrochloride), 25 mg three times daily. Electrocardiograms

ADMISSION

Case 7: A 50 year old woman had chronic emotional difficulties. On July 8, 1972, before admission to a private general hospital, she had been in another hospital, where she was given Thorazine, 100 mg orally four times daily. She was hospitalized on August 18 because of 2 days of chest pain followed by circulatory collapse. Her blood pressure could not be measured; the heart rate was 240/min. An electrocardiogram revealed regular supraventricular tachycardia with electrical alternans of the QRS complexes. Sinus rhythm was restored by carotid sinus pressure. The paroxysmal tachycardia recurred and responded to elevation of systemic blood pressure produced by infusion of Levophede (levarterenol bitartrate); digoxin and quinidine therapy was begun and there were no further episodes of tachycardia. Because of her psychotic behavior, Thorazine therapy was reinstituted in smaller dosage. Electrocardiograms during sinus rhythm showed apparently prolonged Q-T intervals and prominent U waves. These changes might have been related either to Thorazine or to quinidine. The patient was discharged from the hospital on September 2. The electrocardiogram at this time and on September 16 was normal. An electrocardiogram 1 year later showed nonspecific S-T and T wave changes, possibly related to medication.

10-23-74

LEAD

FOLLOWING

DC

LEAD

I

SHOCK

i

!

LEAD II

i ._. / I..

LIDOCAINE

7--““r-ii.

4mg/min

FIGURE 6. Case 8. Electrocardiograms of a 71 year old woman receiving Mellaril. Upper record, ventricular fibrillation. Middle record, a brief period of sinus rhythm with a period of ventricular tachycardia after direct current countershock. Bottom record, sinus rhythm interrupted by abnormal QRS complexes, during lidocaine infusion

226

February

1976

V,

The American

Journal of CARDIOLOGY

LEAD v,

/:/I. ,.t

:’

~p.1.

LEAD IU LEAD V6 FIGURE 7. Case 8. Sinus rhythm with large T-U complexes and multifocal ventricular ectopic beats after control of the ventricular tachycardia.

Volume

37

ARRHYTHMIAS AFTER PSYCHOTROPIC DRUGS-FOWLER

Case 8: A 71 year old woman had been in Longview State Hospital for 2 years because of depression. She was ambulatory and was receiving Mellaril, 50 mg orally four times daily; Serentil, 25 mg orally three times daily; Valium, 5 mg orally four times daily; and Cogentin, 1 mg orally daily. On October 23, 1974 she was found unconscious and pulseless and was resuscitated. An electrocardiogram revealed multifocal premature ventricular beats with bigeminy. Tracheal intubation was performed and an intravenous infusion of Araminee (metaraminol bitartrate) was used to maintain

her blood pressure. On her arrival at Cincinnati General Hospital she had no respiration or pulse. During resuscitation she received direct current countershocks on 20 to 30 occasions and the following intravenously administered medications: Lidocaine, 300 mg; Pronestyl, 500 mg; sodium bicarbonate; Dilantine (sodium diphenylhydantoin), 500 mg; and potassium chloride, 60 mEq/liter. Multifocal premature ventricular beats with runs of ventricular tachycardia continued (Fig. 6 and 7). After intravenous administration of 2 mg Inderal@ (propranolol hydrochloride), junctional rhythm (rate 35/min) was noted. Right ventricular pacing was instituted while she received lidocaine intravenously, 4 mg/min. The next morning ventricular tachycardia recurred, requiring cardioversion five times in the next hour; administration of Inderal, 0.1 mg/min intravenously, controlled the tachycardia. A semiquantitative study of her urine for phenothiazines4 yielded an estimated daily dose. of 100 to 400 mg/day. The next morning the pacemaker was turned off and the patient was alert and responsive and continued to have normal sinus rhythm (Fig. 7). Prominent U waves and occasional premature ventricular contractions were present. Serial electrocardiograms revealed no evidence of myocardial infarction. Discussion

Actions of the Phenothiazines and Tricyclic Antidepressants on the Cardiovascular System

The actions of the phenothiazines (such as thioridazine, chlorpromazine, trifluoperazine and mesoridazine) on the cardiovascular system are complex, resulting from an interaction of the direct effects of the drugs on the heart and blood vessels and indirect effects mediated by way of the central nervous system and autonomic reflexes. Hypotension, the most common cardiovascular effect of the phenothiazines, is due primarily to an inhibition of centrally mediated pressor reflexes5 but, in addition, the drugs produce alpha adrenergic blockade, which contributes to the hypotensive effect. The hypotension is therefore principally the result of peripheral vasodilatation due to the effects on the autonomic nervous system combined with the direct action of the drugs on the peripheral vasculature. The drugs also produce a direct depressant action on the myocardium and papillary muscle.5 Relatively small concentrations of chlorpromazine and thioridazine have been shown to produce a negative inotropic response in guinea pig hearts.6 Thioridazine and other phenothiazines have antiarrhythmic properties similar to those of quinidine.7*8 In canine cardiac fibers9 and in tissue obtained from human atria,lO both chlorpromazine and thioridazine decreased the overshoot of the action potential, decreased the maximal rate of rise during phase zero of the action potential, decreased the du-

ET AL.

ration and amplitude of phase 2 and prolonged phase 3, or the repolarization phase of the action potential. Both drugs shifted the curve relating maximal rate of rise of phase 0 to membrane potential downward and to the right, indicating a decreased membrane responsiveness. These electrophysiologic effects of phenothiazines are identical to those of quinidine.i’ The tricyclic antidepressants, desipramine, imipramine, amitriptyline, protriptyline and nortriptyline, also have direct pharmacologic effects upon the heart. These agents block norepinephrine reuptake by the heart12 and exert a quinidine-like action in preventing experimental arrhythmias in rat, guinea pig and dog hearts. l2 As with the phenothiazines, small concentrations may produce an anticholinergic effect.13 The tricyclic antidepressants may also produce a negative inotropic response. In dogs the intravenous infusion of imipramine (7.5 mg/kg per hour) produced hypotension, decreased the maximal rate of rise of left ventricular pressure (dP/dt max) and reduced cardiac output. I4 In rats and guinea pigs, administration of five tricyclic antidepressant compounds produced both negative inotropic and negative chronotropic responses. In addition, electrophysiologic effects similar to those produced by the phenothiazines have been described.15 Electrocardiographic

Effects

In addition to explaining the antiarrhythmic actions of the phenothiazines and tricyclic antidepressants, the electrophysiologic studies explain the electrocardiographic effects of these drugs, namely, the prolongation of the Q-T interval, decrease in T wave amplitude and prominent U wave and, at larger concentrations, prolongation of the QRS interval. Similarity between the electrophysiologic effects of quinidine and the phenothiazines and tricyclic antidepressants does more than explain the antiarrhythmic activity of the latter group. It also suggests that the mechanism of the arrhythmogenicity of these drugs is similar to that seen with quinidine toxicity, namely, facilitation of reentrant excitation as a result of decreased conduction velocity and temporal dispersion of action potentiation duration in different types of cardiac fibers.g In man, phenothiazine derivatives prolong the P-R and Q-T intervals, depress the S-T segments, and cause flattening and other deformities of the T wave.16J7 Repolarization abnormalities resemble those produced by quinidine.l* In the case,of thioridazine, approximately 50 percent of patients receiving 100 to 400 mglday were found to have some electrocardiographic abnormality.ig The U wave is frequently increased in size.‘* The electrocardiographic effects of tricyclic antidepressants are similar to those produced by phenothiazines and occur in approximately 20 percent of patients receiving therapeutic doses.i6 Imipramine prolongs the Q-T interval, depresses the S-T segment, widens the QRS complex, decreases T wave amplitude and produces supraventricular, as well as ventricular, arrhythmias.18

February 1976

The American Journal of CARDIOLOGY

Volume 37

227

ARRHYTHMIAS

AFTER PSYCHOTROPIC

DRUGS-FOWLER

ET AL.

Cardiac arrhythmias: In accidental or deliberate overdosage of phenothiazines or tricyclic antidepressants, cardiac arrhythmias are common. These include sinus tachycardia, atria1 fibrillation, ventricular fibrillation, A-V block and bizarre QRS complexes.16,20 Review

of Reported

Cardiac

Toxicity

Phenothiazine drugs: Thioridazine toxicity may cause A-V conduction disturbances, ventricular ectopic beats, ventricular tachycardia and ventricular fibrillation.21,22 In 1963, fatal heart block and ventricular tachycardia were reported in two patients receiving large doses of thioridazine,2s and a nonfatal case was described in 1964.24 Hollister and Kosekz5 described six examples of sudden death in otherwise healthy patients receiving phenothiazine drugs. Richardson et al.2s implicated phenothiazines in sudden death, and this concept was supported by Leestma and Koenig27 and by Alexander.28 Leestma and Koenig27 reported fatal ventricular tachycardia in a 24 year old woman receiving Thorazine, Mellaril and Stelazine. Moore and Book2g reported 12 instances of unexpected death without adequate anatomic explanation in patients receiving phenothiazine drugs. No electrocardiograms were available. Giles and Modlin2i described fatal ventricular tachycardia in a 19 year old man receiving Mellaril. Crane’s review’s included two patients who had probable ventricular fibrillation while receiving Thorazine. Alexander and Nifiol described bundle branch block in one patient and heart failure in two. The review of Davis et al.sO included 10 deaths from accidental or deliberate phenothiazine overdosage. Tricyclic antidepressants: Complete A-V block in one patient and intraventricular conduction defect in another, associated with imipramine therapy, were reported by Kristiansen”’ in 1961. Edwardss2 reported atria1 fibrillation and ventricular tachycardia in a 33 year old man after ingestion of 400 mg of Tofranil, 30 mg of Librium (chlordiazepoxide) and 25 mg of Elavil. The patient had no previous heart disease, and the electrocardiogram returned to normal after 80 hours. Smith and Rusbatchss reported complete A-V block probably related to Aventyl therapy in a 73 year old woman. Alexander and Nifioi reported atria1 flutter in a 30 year old man receiving tricyclic antidepressants. In a fatal case of imipramine poisoning, atria1 flutter, ventricular tachycardia and an intraventricular conduction defect developed in an 11 month old boy. s4 Williams and SherterzO described two patients with cardiac complications of tricyclic antidepressant therapy. A 27 year old woman who took 1.6 g of Elavil in a suicide attempt had sinus bradycardia followed by refractory cardiac standstill. In a 37 year old man given imipramine, 25 mg three times daily, while receiving Ismelin@ sulfate (guanethidine) for hypertension, profound bradycardia was followed by fatal cardiac standstill. Roy and Scherrers” described two patients who had severe left heart failure apparently caused by therapeutic doses

228

February 1976

The American

Journal of CARDIOLOGY

of imipramine, although both had heart disease before the drug was given. 2o A 2 l/2 year old boy ingested 500 to 600 mg of imipramine and manifested a supraventricular tachycardia that responded to practo101 therapy. 36 In two patients suicide was accomplished by an overdose of a tricyclic antidepressant drug. In each case an intraventricular conduction defect was followed by cardiac arrest.37 A 58 year old man with no history of heart disease experienced ventricular tachycardia, apparently related to Elavil therapy.38 Present Series

Seven of our eight patients described had cardiac arrhythmias or conduction disturbances, or both, while receiving therapeutic doses of psychotropic drugs. The one patient with a drug overdose had bizarre abnormalities of the T-U complex. Only one of the eight patients had definite underlying heart disease although the possibility of its presence is difficult to discard in two other patients who were more than 60 years of age. The other five patients were 50 years old or younger; one of them had a preexisting intraventricular conduction defect without other evidence of heart disease. Our eight patients included five receiving therapeutic doses of Mellaril who had ventricular tachycardia or ventricular fibrillation, or both, that was fatal in one case. Two of these patients were in their 60’s but three were in their 30’s; none had documentation of associated heart disease. One patient had heart failure and left bundle branch block related to the tricyclic antidepressant Elavil. One had paroxysmal supraventricular tachycardia while receiving the phenothiazine Thorazine. The lack of associated heart disease or history of previous or later cardiac arrhythmias in these patients indicates that these abnormalities were probably drug-related. Three patients with major arrhythmias were younger than age 35 years, thus further suggesting that the psychotropic drugs were responsible for the arrhythmia. Most of these patients were receiving more than one psychotropic drug when the arrhythmia appeared. It is difficult to estimate the cardiac effects of combinations of these agents. Surveys

The Boston Collaborative Drug Surveillance Program examined the relation between therapy with tricyclic antidepressant drugs and cardiovascular complications in 260 of 11,526 hospitalized patients receiving these drugs.” There was no significant increase of cardiovascular complications, either in these 260 patients or in 80 of 4,074 patients with cardiovascular disease who had received these agents. Coull et a1.3g used a hospital-based drug information system to study the relation between tricyclic antidepressant drugs and unexpected death in patients with cardiac disease. Six of 53 patients with cardiac disease died suddenly while receiving amitriptyline; none of 53 matched control subjects died suddenly.

Volume

37

ARRHYTHMIAS AFTER PSYCHOTROPIC DRUGS-FOWLER

This high frequency of unexpected death was not found in patients who received imipramine, nor in patients without cardiac disease who received amitriptyline. A further analysis indicated that imipramine might be important in causing cardiac arrhythmias, in cardiac patients, but the numbers were too small for statistical analysis.2,40 Treatment of Cardiovascular Phenothiazines

Toxicity of the

Once the phenothiazines are absorbed, little can be done to enhance their elimination because these drugs are (1)highly bound to plasma proteins, and (2) their principal metabolic pathways are in the liver and the metabolites are excreted by way of the biliary system. The urine contains relatively little of these drugs. Thus, both hemodialysis and peritoneal dialysis have been unsuccessful in enhancing the removal of these agents. This lack of specific treatment for phenothiazine toxicity requires that therapy be directed toward control of the cardiovascular complications until the natural course of metabolism and excretion has reduced tissue levels to the nontoxic range. Arrhythmias: Ventricular tachyarrhythmias consequent to phenothiazine toxicity should be treated in the same manner as quinidine-induced ventricular tachycardia. Since the electrophysiologic actions of both quinidine and procainamide are identical to those of the phenothiazines, it is unlikely that either drug would abolish the tachyarrhythmia and, indeed, each may aggravate the already existing underlying electrophysiologic abnormality. Because the phenothiazine-induced arrhythmias may result from enhanced reentrant excitation secondary to reduced conduction velocity, it would appear logical to use a drug that enhances conduction velocity in myocardial fibers. In this respect lidocaine appears to be a drug of choice, and ventricular tachyarrhythmias should be treated with repeated bolus injections or continuous intravenous infusions of this agent. Abolition of ventricular ectopic activity with lidocaine may unmask an underlying sinus bradycardia or sinus arrest with slow idioventricular response, in which case a transvenous pacemaker may have to be inserted. If the tachyarrhythmia fails to respond to lidocaine and there are repeated episodes of ventricular fibrillation or ventricular tachycardia, a transvenous pacemaker should be used in an attempt to overdrive the ventricular tachycardia. With the pacemaker in place, further antiarrhythmic drugs can be given without the risk of adding the hemodynamic complications of bradycardia to an already precarious situation. In one of our eight patients, the ventricular tachyarrhythmia failed to respond to lidocaine and could be controlled only by intravenous administration of propranolol, of 0.1 mg/min, given with a transvenous pacemaker in place. Because of the tendency of propranolol to produce a bradyarrhythmia, its use in this situation should probably be reserved for those situations resistant to other therapy and then only in con-

February

ET AL.

junction with a transvenous endocardial pacemaker. Ventricular fibrillation should, of course, be treated with immediate direct current countershock. Hypotension: In addition to causing tachyarrhythmias, phenothiazine toxicity may result in significant hypotension, which may improve with correction of the tachyarrhythmia. If the blood pressure does not increase after control of the arrhythmia, the hypotension should be treated with pressor agents after initial intravenous administration of fluids to correct the relative intravascular hypovolemia. Since psychotropic drugs have a direct effect on the myocardium, rapid volume replacement may precipitate acute left heart failure or pulmonary edema. When rapid volume replacement is contemplated, it is advisable to monitor the pulmonary arterial wedge pressure with a Swan-Ganz catheter. Failure of the hypotension to respond to volume replacement may necessitate the use of a pressor agent such as norepinephrine or methoxamine. Treatment of the cardiovascular toxicity of the tricyclic antidepressants should be similar to that suggested for phenothiazine toxicity. Since some of the arrhythmias induced may reflect cholinergic blockade, the effect of intravenously administered physostigmine should be assessed before using more vigorous therapy such as intracardiac pacing. Cardiac failure: Both groups of drugs have negative inotropic effects that, at least in the case of the tricyclic antidepressants, are responsive to the administration of a cardiac glycoside.41 Therefore, cardiac failure supervening during the course of toxicity of agents of either group should be treated in the usual fashion with digitalis and diuretic drugs. Conclusions

association between psychotropic drugs and electrocardiographic changes, cardiac arrhythmias and sudden death has been well documented. However, the experience reported here emphasizes certain aspects of the relation. Previous communications have dealt primarily with the risk of major arrhythmias or sudden death in patients who receive a toxic drug overdose or who have underlying heart disease. Our experience demonstrates that major and even fatal ventricular arrhythmias can occur in young adults, without previous heart disease, who are receiving usual therapeutic doses of psychotropic agents. Most writings have emphasized the arrhythmogenic potential of the tricyclic antidepressant agents; in our experience, the phenothiazine Mellaril was the principal offender. The experience reported here demonstrates the need for further studies to quantify the risk of therapy with large doses of psychotropic agents. Two kinds of studies are suggested: retrospective studies comparing the incidence of sudden death in hospitalized patients receiving large doses of psychotropic agents and in a matched control population as well as prospective studies in similar groups. Previous hospital surveys have examined primarily the risk of tricyThe

1976

The American

Journal of CARDIOLOGY

Volume

37

229

ARRHYTHMIAS

AFTER

PSYCHOTROPIC

DRUGS-FOWLER

ET

AL.

clic antidepressant agents in a general hospital population of patients with heart disease. Our data indicate that the problem should be studied in large mental hospitals where large doses of these agents are

likely to be used. Further, the proposed study should include the phenothiazine drugs, especially thioridazine, and should not be restricted to elderly patients or those with heart disease.

References 1. Alexander CS, Niiio A: Cardiovascular complications in young patients taking psychotropic drugs. A preliminary report Am Heart J 78~757-769. 1969 2. Moir DC, Cornwell WB, Dingwall-Fordyce I, et al: Cardiotoxicity of amitriptyline. Lancet 2: 561-564, 1972 3. Collaborative Drug Surveillance Program. Report from Boston. Adverse reactions to the tricyclic-antidepressant drugs. Lancet 1:529-531, 1972 4. Forrest IS, Forrest RM: Urine color test for the detection of phenothiazine compounds. Clin Chem 6:362-363, 1960 5. Jarvik ME: Drugs in the treatment of psychiatric disorders. In, The Pharmacological Basis of Therapeutics, fourth edition (Goodman LS, Gilman A). New York, Macmillan, 1970, p 162 6. Hollander PB, Cain RM: Effects of thioridazine on transmembrane potential and contractile characteristics of guinea pig hearts. Eur J Pharmacol 16:129-135, 1971 7. Madan BR, Pendse VK: Antiarrhythmic activity of thioridazine hydrochloride (Mellaril). Am J Cardiol 11:78-81, 1963 8. Arora RB, Madan BR: Chlorpromazine in experimental cardiac arrhythmias. J Indian Med Assoc 26:262-264, 1956 9. Arita M, Surawlcz B: Electrophysiologic effects of phenothiazines on canine cardiac fibers. J Pharmacol Exp Ther 184: 619-630, 1973 10. Arita M, Surawlcz B: Electrophysiologic effects of phenothiazines on human atrial fibers. Jap Heart J 14:398-405, 1973 11. Bassett AL, Hoffman BF: Anti-arrhythmic drugs: electrophysiological actions. Annu Rev Pharmacol 11: 143- 170, 197 1 12. Schmitt H, Cheymol G, Gilbert JC: Effets anti-arythmisants et hemodynamiques de I’imipramine et de la chlorimipramine. Arch Int Pharmacodyn Ther 184: 158-l 74. 1970 13. Raisfetd IH: Cardiovascular complications of antidepressant therapy. Interactions at the adrenergic neuron. Am Heart J 83: 129-133, 1972 14. Van de Ree JK, Zimmerman ANE, Van Heljst ANP: In!oxication with tricyclic antidepressants. Hemodynamic consequences. Eur J Toxicoi 5:302-305, 1972 15. Langstet A, Johansen WG, Ryg M, et al: Effects of dibenzipine and imipramine on the isolated rat heart. Eur J Pharmacol 14: 333-339, 1971 16. Crane GE: Cardiac toxicity and psychotropic drugs. Dis Nerv Syst 31534-539, 1970 17. Arita M, Mashiba H: Effects of phenothiazine and propranolol on EGG. Jap Circ J 34:391-400, 1970 18. Surawicz B, Lasseter KC: Effect of drugs on the electrocardiogram. Prog Cardiovasc Dis 13:26-55, 1970 19. Huston JR, Bell GE: The effect of thioridazine hydrochloride and chlorpromazine on the electrocardiogram. JAMA 198: 16-20, 1966 20. Williams RB Jr, Sherter C: Cardiac complications of tricyclic antidepressant therapy. Ann Intern Med 74: 395-398, 1971

230

February

1976

The American

Journal of CARDlOLOGY

2 1. Giles TD, Modltn RK: Death associated with ventricular arrhythmia and thioridazine hydrochloride. JAMA 205:108-l 10, 1968 22. Schoonmaker RW, Osteen RT, Greenfield JC Jr: Thioridazine (Mellaril)-induced ventricular tachycardia controlled with an artificial pacemaker. Ann Intern Med 65:1076-1078, 1966 23. Kelly HG, Fay JE, Laverty SG: Thioridazine hydrochloride (Mellaril): its effect on the ECG and a report of two fatalities with ECG abnormalities. Can Med Assoc J 89:546-554, 1963 24. Desautels S, Fllteau C, St-Jean A: Ventricular tachycardia associated with administration of thioridazine hydrochloride (Mellaril). Can Med Assoc J 90:1030-1031, 1964 25. Hollister LE, Kosek JC: Sudden death during treatment with phenothiazine derivatives. JAMA 192:1035-1038, 1965 26. Richardson HL, Graupner KE, Richardson ME: lntramyocardial lesions in patients dying suddenly and unexpectedly. JAMA 195:254-260. 1966 27. Leestma JE, Koenig KL: Sudden death and phenothiazines. Arch Gen Psychiatry 18:137-148, 1968 28. Alexander CS: Cardiotoxic effects of phenothiazines and related drugs. Circulation 38:1014-1015, 1968 29. Moore MT, Book MH: Sudden death in phenothiazine therapy. Psychiatr Cl 44:389-402, 1970 30. Davis JM, Bartlett E, Termini BS: Overdosage of psychotropic drugs. A review. Parts I and II. Dis Nerv Syst 29: 157-164 and 246-256, 1968 3 1. Kristiansen ES: Cardiac complications during treatment with imipramine (Tofranil). Acta Psychiatr Stand 36:427-442, 1961 32. Edwards AL: lmipramine myocardial toxicity. NY State J Med 64:1979-1982, 1964 33. Smith RB, Rusbatch BJ: Amitriptyline and heart block. Br Med J 3:311, 1967 34. Fouron J-C, Chicoine R: ECG changes in fatal imipramine (Tofranil) intoxication. Pediatr 48:777-781, 1971 35. Roy A, Scherrer P: A propos des complications myocardiopathiques induites par I’imipramine et les thymoleptiques tricycliques. Ann Med Psycho1 2:260-265, 1972 36. Brown KGE, McMichen HUS, Briggs DS: Tachyarrhythmia in severe imipramine overdose controlled by practolol. Arch Dis Child 47:104-106, 1972 37. Sedat L, Korman MG, Williams PO, et al: Overdosage of tricyclic antidepressants. Med J Aust 2:74-79, 1972 38. Scollins MJ, Robinson DS, Nies A: Cardiotoxicity of amitriptyline. Lancet 2:1202, 1972 39. Coull DC, Dingwall-Fordyce I, Crooks J, et al: Amitriptyline and cardiac disease. Lancet 2:590-59 1, 1970 40. Moir DC: Annotations. Tricyclic antidepressants and cardiac disease. Am Heart J 86:841-842, 1973 41. Laddu AR, Somani P: Desipramine toxicity and its treatment. Toxicol Appl Pharmacol 15:287-294, 1969

Volume

37