experimental amitriptyline intoxication: treatment of cardiac toxicity with sodium bicarbonate

sPECIAL CONTRIBUTION amitriptyline, intoxication, sodium bicarbonate; drug overdose, amitriptyline, sodium bicarbonate; tricyclic antidepressants, amitriptyline

Experimental Amitriptyline Intoxication: Treatment of Cardiac Toxicity With Sodium Bicarbonate

Overdose with amitriptyline and other tricyclic antidepressants can result in ventricular conduction abnormalities as well as severe ventricular arrhythmias. The arrhythmogenic effects of these compounds may be attributed to their direct local anesthetic actions in blocking sodium channels in cardiac membranes. Thus tricyclic-induced ventricular arrhythmias usually do not respond well to therapy with standard Class I antiarrhythmic drugs that also have the same direct local anesthetic action and may poten- tiate the adverse effects of tricyclic antidepressants. Cardiac toxicity was produced in dogs by the administration of amitriptyline, both orally and IV. At serum concentrations less than 2,000 ng/mL, sinus tachycardia occurred with widened QRS complexes. At higher concentrations, QRS duration became more markedly prolonged and was followed by ventricular tachyarrhythmias. Occurrence of ventricular tachyarrhythmias was associated with QRS durations of more than 0.Ii second. Sodium bicarbonate (18 to 36 mEq) administered IV over either 30 seconds or two minutes rapidly con- verted ventricular tachycardia to normal sinus rhythm. Conversion was associated with abbreviation of the QRS complex and was accompanied by a rise in both systolic and diastolic pressures. The duration of sodium bicarbonate effect paralleled the duration of the changes in arterial pH and plas- ma bicarbonate concentrations. In vitro studies in cardiac Purkinje fibers suggested that reversal of amitriptyline-induced cardiac membrane effects by sodium bicarbonate may be attributed not only to alkalinization but also to increases in extracellular sodium concentration, diminishing the local anesthetic action of amitriptyline and resulting in less sodium channel block. The direct depressant effect of amitriptyline on conduction is frequency dependent, and thus would be expected to be potentiated by the anticholinergic-mediated sinus tachycardia that accompanies toxicity. Counteracting the sinus tachycardia with either propranolo] or physostigmine may reduce the direct cardiotoxic effect of the drug, but may exacer- bate the hypotensive effects. Our results suggest that alkalinization with sodium bicarbonate provides an effective therapeutic modality for the treatment of tricyclic-induced cardiac toxicity. [Sasyniuk BL Jhamandas V, Valois M: Experimental amitriptyline intoxication: Treatment of cardiac toxicity with sodium bicarbonate. Ann Emerg Med September 1986;15:1052-1059.]

INTRODUCTION The tricyclic antidepressants (TCAs) are local anesthetic drugs with poten-

tially useful antiarrhythmic as well as potentially harmful arrhythmogenic effects. Overdoses with amitriptyline and other tricyclic antidepressants can result in ventricular conduction abnormalities and severe ventricular arrhythmias. The generation of ventricular axrhythmias is due predominantly to a slowing of conduction in the His-Purkinje system and ventricular muscle. u3 This slowing has been shown to result almost entirely from a direct local anesthetic action in blocking fast sodium channels in cardiac membranes.a-7

In addition to their direct local anesthetic actions, these drugs also have potent anticholinergic actions and block reuptake of norepinephrine into adrenergic neurons. This latter effect probably does not play a major role in the induction of ventricular arrhythmias. However, sinus tachycardia, due ,to vagal blockade, can facilitate ventricular arrhythmias by enhancing the local

Betty I Sasyniuk, PhD Vija Jhamandas Maria Valois Montreal, Canada

From the Department of Pharmacology and Therapeutics, McGill University, Montreal, Canada.

Received for publication February 14, 1986. Revision received June 17, 1986. Accepted for publication June 23, 1986.

Supported by grants from the Medical Research Council of Canada and the Quebec Heart Foundation.

Presented at the UAEM/IRIEM Research Symposium on Toxicology in San Francisco, California, February 1986.

Address for reprints: Betty I Sasyniuk, PhD, Department of Pharmacology and Therapeutics, Faculty of Medicine, McGill University, 3655 Drummond Street, Montreal, H3G 1Y6, Canada.

15:9 September 1986 Annals of Emergency Medicine 1052/97

AMITRIPTYLINE INTOXICATION Sasyniuk, Jhamandas & Valois

o, ia r = 0 . 838

®

o.1, @ • A " "

~o.09 d" C~3

A A A ~

o e.o7 I ~ ~

0 . 0 5

0 2 0 0 0 4000 G 0 0 0

AMITRIPTYLINE SERUM LEVEL (ng/mf)

260

220

&

LL] 180

~ 1,1o

03

100

2

q

1 r = 0 . 589 L,

-i A

i 0 400 B00 1200 1600 2 0 0 0

AMITRIPTYLINE SERUM LEVEL (ng/ml)

FIGURE 1. Corre la t ion b e t w e e n serum levels of ami tr ip ty l ine and QRS duration. Solid symbols indicate QRS widening in animals with ventricular tachyarrhythmias . Results were obtained from six dogs during both oral and IV dosing.

FIGURE 2. Correlat ion be t w een serum levels of ami tr ip ty l ine and sinus rate,

anesthetic block of sodium channels and causing further slowing of con- duction7 Thus tricyclic-induced ventricular arrhythmias usually do not respond to therapy with standard Class I ant iarrhythmic drugs, which them- selves possess the same direct local anesthetic effects. In fact, TCA arrhy thmias may be exacerbated by Glass I antiarrhythmics, s

Twenty years ago French investiga- tors recommended using sodium bicarbonate for the t reatment of tricyclic-induced arrhythmias. 9ll Sim- ilar therapy has been used to treat quinidine toxicity. The effectiveness of sodium bicarbonate now has been confirmed in both animalsh12-14 and human beings. 15-21 The mechanism by which alkalinization with sodium bicarbonate is beneficial, however, has been debated.h11,2o,21 It has been suggested that sodium bicarbonate may be e f fec t ive on ly in ac idot ic patients, ~2 suggesting that it is the reversal of acidosis rather than alkalotic therapy per se that is important. Oth- ers have suggested an increase in protein binding resulting in a decrease of

biologically active free drug.iS, 2o We have shown previously that the

efficacy of sodium bicarbonate therapy is related to a direct cardiac effect in normalizing conduction parameters.7 We review these results and report our more recent work.

METHODS Our study was designed to deter-

mine the relationship between tricyclic antidepressant drug concentration and cardiotoxic effects and to evaluate the efficacy of sodium bicarbonate treatment. Studies were per- formed in awake mongrel dogs pre- t rea ted wi th 5 mg d iazepam and breathing room air spontaneously. Di- azepam was used to prevent signs of central nervous system toxicity that occurred at similar TCA concentrations as cardiac toxicity.

Eight animals were treated wi th three oral doses of 30 mg/kg amitriptyline administered during a 24-hour period. If cardiac toxici ty was not present after oral dosing, an additional IV infusion at a rate ranging from 0.3 to 1 m g / k g / m i n was administered. Several days prior to the experiment, the animals were anesthetized and catheters were inserted into the left carotid artery for recording of blood pressure; in the left femoral artery and vein for withdrawing blood samples for blood gas and drug analysis, re- spectively; and in the left cephalic vein for infusion of drug. Drug samples usually were taken after the second and third oral doses and during IV infusion at the t ime ventricular ar- r h y t h m i a s had occurred. P l a sma

amitriptyline and nortriptyline concentrations were measured by high- pressure liquid chromatographic tech- niques. I3 Arterial blood pressure and an ECG, usually leads II and aVr, were monitored throughout.

The significance of changes in acid- base parameters, QRS duration, ~heart rate, and blood pressure following sodium bicarbonate treatment were analyzed by one-way analysis of variance. Changes in potassium concentration were analyzed using Student's t test. A P < .05 was considered significant.

RESULTS Characteristics of Toxicity

Pretreatment of animals with oral doses of ami t r ip ty l ine resulted in serum drug levels ranging from 1,140 ng/mL to 1,960 ng/mL. Total serum drug levels (both amitr iptyl ine and nortriptyline) ranged from 1,601 ng/ mL to 2,390 ng/mL. All of the animals had sinus tachycardia with widened QRS complexes and some degree of hyper tension. None, however, had ventricular arrhythmias despite serum levels exceeding 1,000 ng/mL.

Ventricular t achyar rhy thmias occurred only after additional IV infu- sions and were preceded by more marked widening of the QRS complex (ie, longer than 0.11 second). A bundle branch block pattern usually preceded initiation of arrhythmias. Ventricular tachyarrhythmias occurred at serum levels of ami t r i p ty t i n e exceeding, 2,000 ng/mL (range, 2,091 to 5,9t5 ng/ mL). Total drug levels associated with ventricutar tachycardia ranged from 2,431 to 6,217 ng/mL.

98/1053 Annals of Emergency Medicine 15:9 September 1986

AFTER PRETREATMENT AMIT iNFUSION 3o ~l/'kl tool @ I mll/kg/mln

2 I / ; / f'. r ,:' "

6 MIN AFTER START OF AMIT INFUSION

o

! t

7 1 MIN AFTER NaCHO 3 ADMINISTRATION (36 rnEq)

BP

0 a

, , , , , r , '

' ¢" ~ '¢ ¢'t ' xF" " . / - ' '

3

QRS ' = - - DURATION ;,3

(reset) 55

7.54 1

pH 7,4s

7.3B "l H C O ~ - 27

(mEq/L )

21

BASE Js 1

EXCESS s (rnEq/L) .-

-4

A I / A A i

/ A ~ A - - A

> "NA N A ~ / A

'N'N& ~ A ""A ~ A

f . . . . . . . . . t . . . . . . . . . t . . . . . . . . . l . . . . . . . . . i . . . . . . . . . i . . . . . . . . . ~ . . . . . . . . . i . . . . . . . . . L - 1 2 -8 - 4 0 4 8 t2 16 2 0

TIME (mln)

There was a significant correlation between the QRS duration and serum levels of amitriptyline (r = 0.838) (Fig- ure 1) as well as with total serum drug levels (r = 0.838). The correlation between QRS duration and serum levels of nortriptyline was poor (r = 0.245). The range of concentrations after oral dosing was 375 to 566 ng/mL.

The sinus rate correlated with serum amitriptyline concentration up to a concentration of approximately 2,000 ng/mL (r = 0.589)(Figure 2). At higher concentrations there was no further increase in sinus rate and, in fact, the rate began to decrease at the highest concentrations. Thus maximum changes in sinus rate occurred at much lower concentrations than those usually associated with marked prolongation of QRS duration and ventricular arrhythmias.

With low drug concentrations the QT interval was shortened in accordance with the drug's effect in abbre- viating action potential duration of both ventricular muscle and the spe- cialized ventricular conducting system.6,7 With higher concentrations the QT interval was prolonged as a result of the increase in QRS duration. At high heart rates and high concentrations, QT interval was not mea- surable.

All our animals had normal pH values prior to and following admin-

istration of amitriptyline. In a previous study using anesthesized dogs that were artificially ventilated, it was more difficult to produce arrhythmias when the animals were hyperventi- lated. 13,14

T r e a t m e n t W i t h S o d i u m B i c a r b o n a t e

In a previous study we showed that sodium bicarbonate is extremely effective in reversing conduction slowing and ar rhythmias produced by amitriptyline in awake and pentobar- bital-anesthetized dogs33 In this study we evaluated further the efficacy of sodium bicarbonate treatment.

In eight animals, either ventricular a r r h y t h m i a s or s u p r a v e n t r i c u l a r tachycardia with widened QRS complexes were produced by either oral doses or addi t ional infus ions of amitriptyline at doses ranging from 0.3 to 1 mg/kg/min. Ventricular arrhythmias reverted to a regular rhythm following administration of 18 to 36 mEq sodium bicarbonate (Figure 3). In every case reversion to normal rhythm was accompanied by an abbreviation of the QRS complex and an improvement in blood pressure without any consistent effect on the sinus rate. Su- praventricular tachycardias usually re- sponded with a slowing of the heart rate and an abbreviation of the QRS complex.

FIGURE 3. Abolition of ventricular arrhythmia following administration of NaHCO 3. During amitriptyline infusion heart rate increased from 140 to 200 beats per m i n u t e and ventricular tachycardia occurred wi th wide, bizarre QRS complexes. Thirty- six mEq NaHCO 3 were administered as two boluses during a one-minute period. By the end of the injection the QRS complexes narrowed and the rhy thm became clearly regular although the heart rate remained high.

FIGURE 4. Effects of administration of NaHCO 3 on electrical and acid base parameters in a dog with sinus tachycardia and marked widening of the QRS complex following oral administration of amitriptyline. Eigh- teen m E q N a H C O 3 were a d m in - istered during a two-minute period as ind ica ted by hor izonta l bar. De- creases in QRS duration paralleled increases in pH and b i c a r b o n a t e concentration.

A typical example of the effect of sodium bicarbonate administration on electrical and acid-base parameters in an an imal wi th supraven t r icu la r tachycardia and wide QRS complexes is shown (Figure 4). Abbreviation of the QRS complex paralleled changes in pH and bicarbonate levels, suggest-

15:9 S e p t e m b e r 1986 Anna ls of Emergency Medicine 1054/99

AMITRIPTYLINE INTOXICATION Sasyniuk, J h a m a n d a s & Valois

TABLE 1. Time course of changes in pH, bicarbonate ion, and base excess following administration of sodium bicarbonate to amitriptyline-toxic dogs*

Time After Beginning of Bicarbonate Administration (min)

Parameter Control 1-2 4-5 6-7 8-10 pH 7.38 ± .01t 7.50 -+ .01 7.46 _+ .02 7.45 _+ .02 7.44 ± .02

HCO~ mEq/L 20.14 _+ .99 29.13 _+ 1.62 25.35 + 1.66 25.02 ± 1.11 23.30 -+ .92

Base excess

15 20 7.42 ± .02 7.38 _+ .01

23.42 ± .99 21.30 _+ .50

mEq/L -2.78 ± .95 12.08 ± 1.66 5.88 ± 2.19 4.94 ± 1.68 3.34 _+ 1.49 2.16 _+ 1.68 12.13 ± .43

Data were obtained from five dogs. Changes in all parameters significantly different from control according to one-way ANOVA. *Either 18 or 36 mEq NaHCO3 were administered over two minutes during either a supraventricular or ventricular tachycardia. Average bicarbonate dose was 1.96 ± 0.27 mEq/kg.

1Mean _+ SE.

5

AMiT INFUSION o.a9 ~/kg/~

NoHCO 3 ADMINISTRATION (tin ~q l

2 ~ 9 m~

~o

AMIT INFUSION o.s mg /kg /m ln

PROPRANOLOL 0.1 mt l / ~

I min ~ 2 mln 12S ] :

O~

6

FIGURE 5. Comparison of effects of physostigmine with those of NaHCO 3 administration. Physostigmine 1 mg f a i l e d to a b o l i s h a r r h y t h m i a . NaHCO 3 18 mEq administered during a two-minute period regularized the rhy thm before the end of the injection.

FIGURE 6. Abolition of ventricular arrhythmia by propranoloI 0.1 mg/kg administered during a one-minute period was accompanied by profound slowing of the heart rate (from 205 to 105 beats per minute) and a marked fall in blood pressure. QRS duration declined almost to predrug values despite an amitriptyline concentration of 3,399 ng/mL.

ing tha t the impor t an t improvemen t in conduc t ion was related largely to a l k a l i n i z a t i o n . T h e t i m e cour se of changes in pH, b icarbonate ion, and base excess fol lowing admin is t ra t ion

of 18 to 36 mEq sodium bicarbonate over two minutes in five of the ani- ma l s is s u m m a r i z e d (Table 1). The m e a n b ica rbona te dose was 1.96 -+ 0.27 mEq/kg (mean + SE).

There was a significant increase in all parameters by the end of the injection; these then declined and reached control values wi th in 15 minutes . The t ime course of changes in QRS duration, hear t rate, and systol ic and diastolic pressures is summarized (Table 2). Maximal shortening of QRS dura- t ion is apparent by the end of the so- d ium bicarbonate injection and is accompanied by a decrease in the heart rate and an increase in blood pressure. Al l p a r a m e t e r s rever ted toward pre- b icarbonate values after 15 minutes , and thus paralleled the changes in pH and b i c a r b o n a t e levels . Se rum potassium, measured in three of the animals before adminis t ra t ion of bicarbonate and two to five minutes after the beginning of the injection, decreased

f rom 3.6 -+ 0.2 to 2.8 + 0.1 m M (mean -+ SE; P < .05).

In five animals, 1 mg physost igmine also was adminis tered to test its effectiveness. On two occasions physostig- m i n e was a d m i n i s t e r e d dur ing ven- t r i c u l a r t a c h y c a r d i a and on th ree occas ions dur ing a supraven t r i cu la r t achycard ia . P h y s o s t i g m i n e was not e f fec t ive in a b o l i s h i n g v e n t r i c u l a r tachycardia (Figure 5). Both rate and pressure were unchanged. In the animals wi th supraventr icular tachycardia, there was a decline in heart rate of 30 + 13 beats per minu te and a de- cl ine in blood pressure of 15 _+ 3.0 m m Hg. QRS durat ion was shortened in only one of the three animals in which there was a 74 beats-per-min- u te dec l ine in hea r t rate. However, t h e r e was on ly a 0 .005- second decrease in QRS duration. In the other two animals, QRS duration remained unchanged.

Propranolol 0.1 mg/kg was admin-


pH

~' Max

AMITRIPTYLINE INTOXICATED FIBER

High pH-High Na-Normal pCO 2 High pH-Normal Na- Low pC02 ~

Normal i f I

7. 5 i i i W I I I I I II I n I l I I I I I • • & • & j • • • • & & • • • & 7 .5

7 . 4 • • 7 . 4

7 . 3 ~ ' e • • , , == ~' 7 . 3

9 5 0

9 0 0

8 5 0

800

7 5 0

7 0 0

mid

I i

• m = i i • •

&

w@ @@mmmw•

.

• & A & • & & & &

6 5 0

6 0 0

= o o o

f

O 5 10 15 2 0 2 5 3 0 5 4 5

950

900

850

8 0 0

7 5 0

7 0 0

650

6 0 0

T , m e ( m i n ) 7

istered to two animals during ventricular tachyarrhythmia. In both animals sinus rhythm was restored, but at a much slower rate despite the presence of high TCA levels (Figure 6). In both animals, however, the beta block- er produced profound and progressive hypotension leading to death.

Mechanism of Reversal of Toxic Effects

Is improvement in conduction following administration of sodium bicarbonate due to alkalinization alone, or does the increase in extracellular sodium concentrat ion contribute to the beneficial effect? To answer this question we superfused canine cardiac Purkinje fibers in vitro with 500 ng/ mL amitriptyline. We then recorded intracel lular action potent ia ls and monitored the action potential parameters that correlate with conduction changes in vivo. 7

The fibers were exposed to the following three superfusates: 1) normal sodium, high pH-Iow PCO 2 (respiratory alkalosis), pH was increased by decreasing the CO 2 concentra t ion from 5% to 3% (Na, 144 mM; HCO3,

25 mM); 2) high sodium, normal pH- normal PCO 2 (hypernatremia), sodium was increased by increasing the concentration of NaCI by 15 mM (Na, 159 m2"Vi; HCO3, 25 mM); and 3) high sodium, high pH-normal PCO 2 (metabolic alkalosis), both sodium and pH were increased by increasing the concentration of NaHCO 3 by 15 mM (Na, 159 raM; HCO3, 40 raM).

The effects of the three superfusates on the upstroke velocity (Vmax) of the action potential, one of the major determinants of conduction velocity in vivo, are compared (Figures 7 and 8). Both hypocapnia and increases in extracellular sodium concentrat ion cause an i m p r o v e m e n t in ;¢max. However, the improvement in Vmax in the presence of sodium bicarbonate markedly exceeds that of either of the other two interventions, suggesting that the beneficial effects of sodium bicarbonate may be a t t r ibuted not only to atkalinization, but also to the presence of the sodium ion. These results suggest that hypervent i la t ion alone would be less effective in treat- ing amitriptyline toxicity than is administration of sodium bicarbonate.

FIGURE 7. Comparison of the effects of hypocapnea versus NaHCO 3 on the upstroke velocity (~7max) of a cardiac Purkinje fiber action potential super- f u s e d in v i t r o w i t h 500 n g / m L amitriptyline.

Frequency-Dependent Effects of Amitriptyline in Determining Arrhythmogenesis

It is well known that frequency has important effects on action potential characteristics and that drug effects are modified by changes in frequency.2S The important relevance to the arrhythmogenic effect of amitriptyline is the progressively greater degree of sodium channel block with increases in stimulation frequency This is man- ifested as a progressive decrease in the Vmax of the cardiac action potential (Figure 9). Thus, in the presence of drug, Vmax is markedly frequency dependent over the range of heart rates tha t would be expected to occur clinically. This result would be man- ifested as a progressive prolongation of the QRS duration clue to progressive conduction delay. In accordance with our in vitro data, we found that in in- dividual animals the QRS duration varied with fluctuations in the sinus rate. Thus, in the presence of amitriptyline, the degree of QRS prolongation depended on both the frequency of activation and the serum level of the drug.

Frequency-dependent effects with a resultant decrease in sodium channel block may account for the shortening of QRS duration that may accompany administration of either propranolol or physostigmine. If there is no significant change in sinus rate the QRS duration may be largely unaffected by either drug. On the other .hand, sodium bicarbonate improves Vmax and conduction regardless of the heart rate (Figure 9). In the presence of acidosis, amitriptyline is much more depressant at all heart rates, suggesting a greater susceptibility to arrhythmia.

DISCUSSION Three major mechanisms have been

identified as responsible for generation of ventricular arrhythmias, that is, enhancement of automaticity, initiation of triggered activity, or induction of reentry.24 Tricyclic antidepressants do not enhance automaticity; in fact, all available data demonstrate that these drugs depress spontaneous rate

15:9 September 1986 Annals of Emergency Medicine 1056/101

AMITRIPTYLINE INTOXICATION Sasyniuk, dhamandas & Valois

pH

~max

AMITRIPTYLINE INTOXICATED PURKINJE FIBER

High pH-High Na- Normal pCO 2 Hi~lh Na-Normal pH

Normal' 7.6 '

7.5 f • • • S A A A A , A A A AAA

~ 4 7.3 • e e e j l i B = = , . = m • I l i a

8 0 0

AA

750

70OI

eS01

eO01

550 I

== = i • • •

j l l 0 e =

AA A A A A A *,A I

5 10 15 20 25 3 0 35

Normal

OQ Q 0 0 0 0

h 4 0

Time (min)

7.6 7.S 7,4 7,3

8OO

75O

7OO

650

6OO

550

x

E .>

8 0 0

730

6 6 0

590

520

450

z, control = alkalotic

I , , , , l l , r V l , l r , , , , , , i , , i , r r r , i i 5 0 1 0 0 1 5 0 2 0 0

STIMULATION RATE (pulses/rain)

FIGURE 8. Comparison of the effects of increases in sodium concentration in the superfusate versus increases in N. aHCO 3 on the upstroke velocity Vmax of cardiac Purkinje fiber action

potential superfused in vitro with 500 ng/mL amitriptyline.

FIGURE 9. Effects of increases in the stimulation rate on the upstroke velocity (~Zmax) of the action potential recorded from a cardiac Purkinje fiber superfused in vitro with 500 ng/mL amitriptyline. Triangles represent control condition; squares, after addition of NaHCO 3 to the superfusate.

in the His-Purkinje system even when it is enhanced by catecholamines. 4-7 Because these drugs abolish triggered act ivi ty induced by digitalis-glyco- sides, the mechanism of the toxic arrhythmias is unlikely to be induction of triggered automaticity. 6

All TCAs studied to date slow conduction in Purkinje fibers and ventricular muscle and abbreviate action potential duration and refractory period in these fibers, particularly in the former. 4-z Marked slowing of conduction, coupled with abbreviation of the refractory period, would favor generation of arrhythmia by a reentry mechanism.

The effects of amitriptyline on conduction are highly frequency dependent. This action is related closely to the t ime-dependen t depress ion of 9max of cardiac action potentials produced by amitriptyline as well as by other local anesthetic blocking drags. The mechanism of frequency-dependent drug effects on cardiac sodium c h a n n e l s has been de f ined by a number of models.23, 25 These models predict that drug-receptor association and dissociation rates depend on the state of the sodium channel, while ac- cess of drug to the receptor depends on the drug's biophysical properties and the state of the channel.

With each ventricular activation, amitriptyline binds to sodium channels and begins to unbind during diastole. If diastole is sufficiently long, sodium channels become completely unblocked between activations and conduction is unaltered. At high heart rates, the diastolic period is shortened and there is insufficient time for com- plete drug unbinding between beats. Thus c u m u l a t i v e sod ium channel block occurs until an equilibrium between drug association and dissociation is established. Frequency-dependent b lock is p o t e n t i a t e d in the presence of acidosis because drug unbinding is slowed. This will result in persistent blocked channels and re-

duction of conduction velocity with resultant QRS prolongation and induction of a reentrant mechanism.

Slowing heart rate in the presence of amitriptyline would allow more diastolic time for drug unbinding from sodium channels and result in an improvement in ventricular conduction, which may be sufficient to abolish a reentry mechanism. Thus a decrease in sinus rate may itself be effective in abolishing ventricular arrhythmias by el iminating rate-dependent conduction slowing. This may be the mechanism for the occasional effectiveness of beta blockers or physostigmine in reversing cardiac toxicity.

The occurrence of rate-dependent conduction slowing may account for the lack of correlation between serum drug levels and occurrence of ventricular arrhythmias in the study of Bochnert and Lovejoy. 26 QRS duration may be markedly prolonged at lower serum levels if the sinus rate is high, particularly if metabolic acidosis also is present. Thus QRS duration may be a better indication of cardiac toxicity than is serum drug level.

Our results suggest that systemic alkalinization wi th sodium bicarbonate should be the first line of therapy for cardiac conduction abnormalities and ventricular tachyarrhythmias associated with antidepressant intoxica-


TABLE 2. Time course of changes in QRS duration, heart rate, and systolic and diastolic pressures following administration of NaHCO a to amitriptyline-toxic dogs*

Time After Beginning of Bicarbonate Administration (min)

Parameter Control 1 2 3 4 5 6 8 10 15 QRS duration

(msec) 108 _+ 9t 82 ± 5 79 ± 4 80 - - 4 86 -+ 5 83 ± 4 85 ± 4 87 ± 4 91 _+ 5 96 ± 8

Heart rate (beats/min) 203 ± 11 198 ± 9 184 ± 9 185 ± 9 185 ± 9 180 ± 10 184 __ 9 183 ± 9 194 _+ 1t 201 ± 9

Systolic pressure (mm Hg) 140 ± 9 148 ± 6 163 _+ 5 169 ± 5 166 _+ 7 160 ± 6 160 ± 6 154 ± 6 147 _+ 7 138 ± 6

Diastolic pressure (mm Hg) 95 ± 7 99 ± 6 112 ± 7 109 ± 8 109 _+ 10 102 ± 9 102 +_ 8 102 ± 7 99 _+ 8

Data were obta ined from five dogs. NaHCO3 was administered on 12 occasions in the five dogs. QRS durat ion and systolic pressure signif icant ly different from control accord ing to one-way ANOVA, *Either 18 or 36 mEq NaHCO3 was administered over two minutes dur ing either a supraventr icular or ventr icular tachycard ia . 1-Mean _+ SE.

95 ± 5

tion. Administration of sodium bicarbonate is effective whether or not underlying metabolic acidosis is present and whether or not there is a significant change in the sinus rate.

Several possible mechanisms have been suggested for the beneficial effects of sodium bicarbonate. Increases in protein binding resulting in a decrease in the amount of biologically active free drug have been suggested by Brown and associates, is This mechanism has been debated. Al- though amitriptyline is bound exten- sively to serum proteins and this binding is pH dependent, redistribution of amitr iptyl ine from peripheral sites might diminish the reduction in free drug concentration in serum caused by alkalinization. 21

A more plausible explanation for sodium bicarbonate efficacy is a direct cardiac effect in normalizing conduction parameters, as demonstrated in our study. Reversal of conduction slowing would be expected to abolish the conditions permitting reentrant arrhythmias. Increases in pH have been shown to accelerate the recovery of s o d i u m c h a n n e l s b locked by amitriptyline, presumably by decreasing protonation of the drug-receptor complex. 27 An increase in pH would increase the proportion of amitriptyline that is nonionized, thus facili- tating egress of drug from the receptor site in the sodium channel. The direct cardiac effect also would be dimin- ished as a result of the lowering of serum potassium levels in the presence of increased bicarbonate levels. The resultant membrane hypopolari-

15:9 September 1986

zation would diminish block of sodium channels by amitriptyline because block is voltage dependent.7

Although alkalinization plays a major role in reversal of conduction abnormalities, increases in extracellular sodium concentration contribute also. Increases in extracellular sodium concentration have been shown to diminish the local anesthetic action of drags. 2s Thus sodium bicarbonate administration would be expected to be more effective than hyperventilation, as has been borne out in a recent study. 14

Sodium bicarbonate reverses conduction abnormalities regardless of heart rate. Furthermore, unlike most other treatment modalities (eg, physostigmine, beta blockers), alkalinization with sodium bicarbonate also reverses the hypotensive effects of the drug. Sodium bicarbonate is effective not only in patients with acidosis, but also in patients with respiratory alkalosis. 20 It is effective immediately after administration and, although its effects do not last long (a maximum of ten to 20 minutes), repeated administration is always effective.

Concern has been expressed about the hazard of sodium overload, which may predispose to pulmonary ede- ma. 29 If this is a concern in a particu- lar patient, we suggest administration of sodium bicarbonate for rapid effective conversion of a ventricular ta- chyarrhythrnia and the use of hyperventilation to prevent the develop- ment of arrhythmias. Several case reports support the beneficial effect of hyperventilation in tricyclic-induced

Annals of Emergency Medicine

cardiac arrhythmias.29, 30

C O N C L U S I O N Amitriptyline and other TCAs are

local anesthetic drugs that result in ventricular tachyarrhythmias at high serum concentrations. Like other local anesthetic drags, they produce frequency-dependent slowing of intra- ventricular conduction. The mechanism of amitriptyline toxic arrhythmias is probably a form of ventricular reentry due to the severe conduction slowing. Our s tudy showed tha t NaHCO 3 effectively reverses these ar- r h y t h m i a s . A l k a l i n i z a t i o n w i t h NaHCO 3 partially reversed QRS. prolongation in vivo and improved Vmax of the cardiac action potential in vitro. These findings are compatible with the concept that NaHCO 3 reverses ar- r hy thmias by accelera t ing intra- ventricular conduction, abolishing the conditions that allow reentry to occur. In vitro studies suggest that the bene- fit from NaHCO 3 can be attributed both to alkalinization and to the sodium moiety. Both diminish the local anesthetic action of amitriptyline, result ing in less sodium channel block. Our study supports the use of NaHCO 3 for the effective treatment of TCA-induced cardiac toxicity.

REFERENCES 1. Preskorn SH, Irwin H: Toxicity of tricyclic antidepressants: Kinetics, mechanism, intervention: A review. J Clin Psy- chiatry 1982;43:151-156.

2. Marshall JB, Forker AD: Cardiovascu- lar effects of tricyclic antidepressant drugs: Therapeutic usage, overdose and

1058/103

AMITRIPTYLINE INTOXICATION Sasyniuk, Jhamandas & Valois

management of complications. A m Heart J 1982; 103:401-414,

3. Vohra J, Burrows G, Hunt D, et al: The effect of toxic and therapeutic doses of tricyclic antidepressant drugs on intracar- diac conduction. Eur J Card io l 1975; 3:219-227.

4. Rawlings DA, Fozzard HA: Effects of imipramine on cellular electrophysiological properties of cardiac Purkinje fibers. J Pharmacol Exp Ther 1979;209: 371-375.

5. Weld FM, Bigger JT, Jr: Electrophysi- ological effects of imipramine on ovine cardiac Purkinje and ventricular muscle fibers. Circ Res 1980;46:167-174.

6. Muir WW, Strauch SM, 8chaal SF: Ef- fects of tricyclic antidepressant drugs on the electrophysiological properties of dog Purkinje fibers• J Cardiovasc Pharmacol 1982;4:82-90.

7. Sasyniuk BI, Jhamandas V: Mechanism of reversal of toxic effects of amitriptyline on cardiac Purkinje fibers By sodium bicarbonate. J Pharmacol Exp Ther 1984; 231:387-394.

8. Nymark M, Rasmussen J: Effects of certain drugs upon amitriptyline induced electrocardiographic changes. Acta Phar- macol Toxicol 1966;24:148-156.

9. Bismuth C, Bodin F, Pebay-Peroula F, et al: Intoxication par l'imipramine avec in- suffisance cardiaclue aigue. La Presse Medicale 1968;76:2277-2278.

10. Prudhommeaux JL, Lechat P, Auclair MC: Etude experimentale de l 'influence des ions sodium sur la toxicite cardiaque de l'imipramine. Therapie (Paris) 1968;23: 675-683•

11. Gaultier M: Sodium bicarbonate and tricydic antidepressant poisoning (letter)• Lancet 1976;2:1258.

12. Brown TCK: Tricyclic antidepressant overdosage: Experimental studies in the management of circulatory complications. Clin Toxicol 1976;9:255-272.

13. Nattel S, Keable H, Sasyniuk BI: Ex- perimental amitriptyline intoxication: Electrophysiologic manifestations and management• J Cardiovasc Pharmaco l 1984;6:83-89.

14. Nattel S, Mittleman M: Treatment of ventricular tachyarrhythmias resulting from amitriptyline toxicity in dogs. J Pharmacol Exp Ther 1984;231:430-435.

15. Brown TCK, Barker GA, Dunlop ME, et al: The use of sodium bicarbonate in the treatment of tricyclic antidepressant induced arrhythmias. A n e s t h In t ens i ve Care 1973; 1:203-210.

16. Brown TCK: Sodium bicarbonate treatment for tricyclic antidepressant arrhythmias in children. Med J A u s t 1976;2: 380-382.

17. Hoffman JR, McElroy CR: Bicarbo- nate therapy for dysrhythmia and hypotension in tricyclic antidepressant overdose. West J Med 1981;134:60-64.

18. Bergman RN, Watson WA: Cardiac toxicity associated with acute mapro- tiline self-poisoning. A m J Emerg M e d 1984;2:144-146.

19. Shepherd KE: Prevention of cardiovas- cular and renal complications in tricyclic drug overdose. Arch Intern Med 1984;144: 413-414.

20. Molloy DW, Penner SB, Rabson J, et al: Use of sodium bicarbonate to treat tricyclic antidepressant-induced arrhyth- rnias in a patient with alkalosis. Can Med Assoc J 1984; 130:1457-1459.

21. Pollack BG, Perel JM: Sodium bicarbonate in tricyclic antidepressant-induced

arrhyhmias (letter). Can M e d A s s o c J 1984;131:717.

22. Sodium bicarbonate and tricyclic antidepressant poisoning (editorial). Lan- cet 1976;2:838.

23. Hondeghem LM, Katzung BG: Antiar- rhythmic agents: The modulated receptor mechanisms of action of sodium and cal- cium channel blocking drugs. A n n Rev Pharmacol Toxicol 1984;24:387-423.

24. Sasyniuk BI: Concept of reentry versus automaticity. A m J Cardiol 1984;54: 1A-6A.

25. Grant AO, Starmer CF, Strauss HG: Antiarrhythmic drug action. Blockade of the inward sodium current. Circ Res 1984;55:427-439.

26. Boehnert MT, Lovejoy FH: Value of the QRS duration versus the serum drug level in predicting seizures and ventricular arrhythmias after an acute overdose of tricyclic antidepressants. N Engl J Med 1984;313:474-479•

27. Sasyniuk BI, Jhamandas V: Frequen- qy-dependent effects of amitriptyline on Vmax in canine Purkinje fibers and its al- teration by alkalosis. Proc West Phar~ macol Soc 1986;29:73-75.

28. Kolhardt M: A quantitia.tive analysis of the Na + dependence of Vmax of the fast action potential in maintaining ven- t r icular myocardium. Pf lugers A r c h 1982;392:379-387.

29. Kingston ME: Hyperventilation in tricyclic antidepressant poisoning. Crit Care Med 1979;7:550-551.

30, Strom J, Sloth M, Nielsin NN, et al: Acute self-poisoning with tricyclic antidepressants in 295 conscious patients treated in an ICU. Ac ta A n a e s t h e s i o l Scand 1984;28:666-670.

104/.1059 Annals of Emergency Medicine 15:9 September 1986

experimental amitriptyline intoxication: treatment of cardiac toxicity with sodium bicarbonate

Documents