Complete Atrioventricular Block Due to Potassium

Transcription

1 Complete Atrioventricular Block Due to Potassium By Charles Fisch, M.D., Kalman Greenspan, Ph.D., and Robert E. Edmandi, M.D. ABSTRACT Intravenous infusion of KC1 at a rate of 1.22 or 2.45 meq per minute to anesthetized dogs frequently resulted in complete A-V block at a time when P waves were still recorded in the ECG. Furthermore, some of the observations suggest that in hyperkalemia a regular idioventricular rhythm may not be essential for the diagnosis of complete atrioventricular (A-V) block. Thus, with complete A-V block induced during a rapidly changing plasma K +, the ventricular pacemaker may be irregular. In addition, evidence was obtained that some parts of the atrial tissue, the automatic ventricular focus and the ventricular myocardium, are more resistant to K + than is the A-V conduction tissue. The relative sensitivity of the various tissues of the heart seems to depend, among other factors, on the rate of KC1 infusion. ADDITIONAL KEY WORDS A-V dissociation A-V conduction A-V nodal sensitivity to K + S-A nodal resistance to K + ventricular resistance to K + atrial sensitivity to K* irregular idioventricular rhythm with complete A-V block effects of K + on atrial segments ectopic pacemaker rate and K + concentration anesthetized dogs In 1911 Mathison 1 observed that in some decerebrate, vagotomized cats potassium (K) produced a gradual slowing of the heart rate but "In other cases an abrupt slowing of pulse rate occurs, due to dissociation of the auricular and ventricular rhythms, a conduction of 2:1 or of complete heart block being produced." This early observation that K + is capable of producing a high degree of atrioventricular (A-V) block has been confirmed by a number of subsequent studies 2 " 4 and has been the subject of a recent review. 5 However, the question whether hyperkalemia is capable of producing complete A-V block has not been settled. In fact, in recent studies, Hoffman and his associates 0 "" 8 indicated that the ventricles may be driven by the sinus pacemaker From the Krannert Heart Research Institute, Marion County General Hospital and the Department of Medicine, Indiana University School of Medicine, Indianapolis, Indiana. Supported in part by the Herman C. Krannert Fund, U. S. Public Health Service Grants HE-6308 and HTS 5363, and the Indiana Heart Association. Accepted for publication March 22, at plasma K + levels sufficiently high to depress atrial activity and to eliminate P waves from the electrocardiogram (ECG). Such direct sinoventricular propagation was ascribed to the specialized atrial conduction pathway, which has been shown, under certain conditions, to be particularly resistant to depolarization by K +. Persistence of sinoventricular conduction also implies high resistance of A-V conduction tissue to the depressing effect of hyperkalemia. The purpose of this report is (1) to present evidence that under certain specific experimental conditions hyperkalemia frequently produces complete A-V block when P waves are still recorded in the ECG and (2) to suggest that under conditions of changing plasma K +, a regular idioventricular rhythm may not be essential for the diagnosis of complete A-V block. Method Mongrel dogs of either sex weighing 10.9 to 15.3 kg were anesthetized with 30 mg of sodium pentobarbital per kg of body weight given in- CircmUlion Restmrcb, Vol. XIX, August

2 374 FISCH, GREENSPAN, EDMANDS travenously. After thoraeotomy, electrodes were implanted on the epicardial surface of the right atrium (RA) and the right ventricle (RV) and in the area of the sinoatrial node (SA). During this procedure, ventilation maintained by a Harvard respirator set at a rate of 12 to 18 cycles per min and a tidal volume of 300 to 500 cc, provided for each animal a minute volume of 4 to 6 liters. In addition lead II (L-2) of the ECG was recorded. After implantation of the electrodes, the chest was closed and artificial respiration was continued at a minute volume of 4.5 liters. The experiment was monitored by an Electronics for Medicine recorder with an eight trace switched beam oscilloscope. Tracings were recorded on 7-inch paper moving at 25 or 50 mm per sec. After control tracings were recorded, an isotonic solution of KC1 (160 meq per liter) in demineralized, triple-distilled water was infused through a femoral vein at a rate of either 1.22 or 2.45 meq per min until large, diphasic ventricular complexes appeared. The infusion was then stopped. At that time, 75 to 120 ml of solution had been infused and the plasma K + level was 8.4 meq per liter (SD±0.74). The ECG and plasma K + level were then allowed to return to normal. Then KC1 was reinfused. Twenty-three infusions were given to 6 dogs. Plasma K + levels were determined from arterial blood samples, with a Technicon autoanalyzer. Results Administration of KC1 resulted in unequivocal complete A-V dissociation in 17 of the 23 infusions. In the other six infusions either the P wave or QRS complexes were eliminated or the P wave became obscured by the T wave so that a diagnosis of complete A-V block from L-2 was impossible. The effect of K + on the SA and RA complexes was similar to that described by Vassalle and Hoffman. 8 The RA complex widened, the upstroke became slower and the amplitude gradually became lower and frequently disappeared at a time when the P wave in L-2 was still present. Under the conditions of our experiment, the SA activity persisted throughout the infusion, frequently after RA activity was no longer discernible. At such plasma K + levels, however, the SA complexes were prolonged in duration, lower in amplitude and not infrequently composed of "multiple oscillations." 8 Figures 1 through 5 are representative of the changes observed during this study. In the experiment illustrated by figure 1, the control tracing was recorded when the plasma K* level was 3.8 meq per liter and the heart rate was 115 per min. When the plasma K + level was 8.4 meq per liter, the rate was 93, the amplitude of the SA complex was decreased and its duration was markedly prolonged; the RA activity was no longer recorded, but distinct P waves synchronous with SA activity were visible in L-2. The ventricular rate was 41 and perfectly regular with a QRS morphology identical to that recorded during the control period. There was no relation between the P and QRS complexes. These changes are diagnostic of complete heart block with an idioventricular pacemaker located above the bifurcation of the bundle of His. The failure to register RA activity in the presence of P waves indicates that other areas of the atria were still excitable. Such a differential sensitivity of various parts of the atria to K + was pointed out by Vassalle and Hoffman. 8 The complete heart block in this case indicates that, under the conditions of our experiments, the A-V conducting tissue was more sensitive to K + than were either the ventricles or parts of the atrial myocardium. Figure 2 is an example of complete A-V block with an idioventricular rate much faster than that in figure 1. During hyperkalemia the FIGURE 1 In this and figures 2-5, the tracings from top to bottom are from the sinoatrial node (SA), right atrial epicardium (RA), right ventricular epicardium (RV) and L-2 of an ECG. Here, the control heart rate is 115 per min and the plasma K* is 3.8 meq per liter. After infusion of KCl (right), ventricular rate is 41 per min and plasma K* is 8.4 meq per liter. CircmUiio* Rts rcb, Vol. XIX, Amgnsl 1966

3 POTASSIUM ION AND A-V BLOCK 375 SA and RA leads show pronounced aberration as compared with the control. The SA, RA and P waves are synchronous at a rate of 107. The QRS complexes show no change in morphology from the control tracing; their rhythm is absolutely regular at a rate of 70. There is no temporal relation between the atrial and ventricular complexes. Figure 3 demonstrates gradual depression of an idioventricular focus from a control rate of 158 to 44 per min with a concomitant decline of the SA rate from the control of 158 to 103 per min. Complete heart block FIGURE 2 Control plasma K* was 3.2 meq per liter; after infusion of KCl it was 8.2 meq per liter. (See text.) EXr-AVt 11 /«FIGURE 3 These tracings were recorded during a single infusion as the plasma K* was rising and demonstrate complete heart block with different idioventricular rates. In panel B the RA activity is absent and in panel C it is of low amplitude and prolonged. The P waves (L-2) and SA complexes are synchronous, with the rate gradually declining from the control of 158 to 136, 120 and 103 in panels A, B and C respectively. The idioventricular rate also declines gradually from the control rate of 158 to 94, 70 and 45 in A, B and C respectively. The plasma K* concentration, in milliequivalents per liter, was 3.4 (control), 7.6 (A), 8.1 (B) and 8.6 (C). Circulation Rts rcb, Vol. XIX, August 1966 is recorded in panels A, B and C. The morphology of the QRS complexes remains normal, indicating that the ventricular focus is located above the bifurcation of the bundle of His. The gradual depression of the ventricular rate in complete A-V block consequent to changing plasma K + levels, suggests that under conditions of unstable electrolyte concentrations, the classical concept of absolute regularity of the ventricular activity may not be prerequisite to the diagnosis of complete heart block. Plasma K + at the time figure 4 was recorded was 8.0 meq per liter. The SA rate is 100 and absolutely regular; the SA complex is of smaller amplitude and longer duration than in the control tracing, but not so aberrant as that in figure 1. The RA complex is of extremely low amplitude and has a duration of 0.30 sec as compared with a control of 0.08 sec. Again, well defined P waves synchronous with the SA and RA deflections are recorded in L-2, and the QRS complexes are morphologically identical to those observed during the control period. In this tracing, however, the ventricular rhythm is irregular, the rate varying from 37 to 43 per min, with an R-R interval of 1.60 to 1.40 sec. There is no temporal relation between P and QRS complexes. In spite of the irregularity of the ventricular rhythm, the lack of a temporal relation between atrial and ventricular complexes and the presence of a regular SA rhythm indicates the presence of complete A-V block. Figure 5 is a composite of tracings recorded as the plasma K + level was decreasing and demonstrates (1) persistence of P waves in the absence of ventricular (L-2, RV) and FIGURE 4 The control tracing for this record is the same as in figure 1 (plasma K* of 3.8 meq per liter). (See text.)

4 376 FISCH, GREENSPAN, EDMANDS FIGURE 5 The plasma K* level before KCl infusion was 4.8 meq per liter. In panel A the RA activity is absent (plasma K* equaled 7.6 meq per liter). (See text.) RA activity, (2) synchronous but irregular SA- and P-wave rhythm, (3) reappearance of RA activity as the plasma K decreased and (4) presence of complete A-V block. Panel A shows an abnormal SA complex, absence of RA activity, disappearance of the P wave in the preceding T wave, marked prolongation and finally disappearance of ventricular complexes, with reappearance of P waves. The fact that the aberrant QRS waves are SA driven until the ventricular complexes disappear suggests that the failure of conduction is somewhere below the A-V junctional area. In panel B, both RA and ventricular activity are absent. SA rhythm is irregular. The P rhythm is also irregular but synchronous with the SA rhythm, indicating that the P waves are SA driven. Panel C shows a regular SA rhythm with a rate of 100 per min with the SA complexes of greater amplitude than in panels A or B. Both the RA and P waves are SA driven. The QRS complexes are identical in appearance to those recorded during the control period, with an irregular rhythm, with two R-R cycles measuring 1.60 sec, and one somewhat shorter measuring 1.52 sec. There is no relation between the P and QRS complexes. Discussion The diagnosis of complete A-V block in 17 of the 23 infusions was based on the classical criteria, which include (1) presence of P waves in limb lead, (2) different P and QRS rates, (3) lack of a fixed relation between P and QRS and (4) a regular idioventricular rhythm (figs. 1-3). The idioventricular complexes originated above the bifurcation of the bundle of His, as indicated by a morphology identical to that recorded during the control period. Block along the A-V junctional tissue is obviously different from failure of ventricular excitation due to block of conduction along the intraventricular specialized tissue, block at the Purkinje-myocardial junction, a failure of myocardial cells per se to conduct, or a combination of any of these. 6 An example of such a block, which is characterized by gross aberrancy of the QRS prior to failure of the ventricles to respond, is shown in figure 5. Some of our observations suggest that when the plasma K + varies it may not be necessary to demonstrate a regular idioventricular rhythm before a diagnosis of complete A-V block can be made. Figure 3 demonstrates complete A-V block with a regular idioventricular rhythm which varies in rate as the plasma K + is rising; this indicates that the discharge rate of the idioventricular focus is dependent on the level of plasma K*. It is reasonable, therefore, to assume that in complete A-V block with rapidly changing plasma K + the ventricular pacemaker may be irregular (figs. 4 and 5c). The persistence of P waves in L-2 and the presence of an idioventricular rhythm originating above the bifurcation of the bundle of His indicates that under the conditions of our experiment some parts of the atrial tissue, the idioventricular focus and the ventricular myocardium are more resistant to K* than is the A-V conduction tissue. Potassium, depending on the plasma level and the rate of infusion, has been shown to have quantitatively different, and at times an opposite, effect on A-V 2 " 5 ' 0 - n and intraventricular 3 ' 6 ' 12 ~ 15 conduction, automaticity 1 '' 5 ' and excitability It is not unexpected, therefore, that with a varying rate of infusion such electrophysiologically different tissues as the atrial muscle and the specialized conduction system exhibit different degrees of sensitivity to K +. In the experiments of Hoffman and associates 6 ' 8 demonstrating direct sino- Circulaiiom Rtsttrcb, Vol. XIX, August 1966

5 POTASSIUM ION AND A-V BLOCK 377 ventricular conduction in the absence of P waves in the surface ECG, the rate of KC1 infusion varied from 0.19 to 0.79 meq per min. In our experiments, in which complete A-V block was observed, the rate of infusion was either 1.22 or 2.45 meq per min. This difference in rate of infusion of KC1, we believe, may explain the difference in the results of the two studies. References 1 MATHISON, G. C: Effect of potassium salts upon the circulation and their action on plain muscle. J. Physiol. (London) 42: 471, MCLEAN, F. C, BAY, E. B., AND HASTING, A. B.: Electrical changes in the isolated heart of the rabbit following changes in the potassium content of the perfusing fluid. Am. J. Physiol. 105: 72, WlNKLER, A. W., HOFF, H. E., AND SMITH, P. K.: Electrocardiographic changes and concentration of potassium in serum following intravenous injection of potassium chloride. Am. J. Physiol. 124: 478, FISCH, C, FEICENBAUM, H., AND BOWERS, J. A.: Effect of potassium on atrioventricular conduction of normal dogs. Am. J. Cardiol. 11: 487, FISCH, C, KNOEBEL, S. B., FEICENBAUM, H., AND GREENSPAN, K.: Potassium and the monophasic action potential, electrocardiogram, conduction and arrhythmias. Prog. Cariovasc. Dis. 8: 387, VASSALLE, M., CREENSPAN, K., JOMAIN, S., AND HOFFMAN, B. F.: Effects of potassium on automaticity and conduction of canine hearts. Am. J. Physiol. 207: 334, HOFFMAN, B. F.: The pathophysiology of failure of impulse transmission to the ventricles. In Sudden Cardiac Death, edited by B. Surawicz and E. D. Pellegrino, New York, Grune & Stratton, 1964, p VASSALLE, M., AND HOFFMAN, B. F.: The spread of sinus activation during potassium administration. Circulation Res. 17: 285, SPEALMAN, C. R.: Actions of ions on the mammalian heart. Am. J. Physiol. 136: 332, PAES DE CARVALHO, A., AND LANCAN, W. B.: Influence of extracellular potassium levels on atrioventricular transmission. Am. J. Physiol. 205: 375, GREENSPAN, K., WUNSCH, C, AND FISCH, C: The relationship between potassium and vagal action on atrioventricular transmission. Circulation Res. 17: 39, BAMMER, H., AND ROTHSCHUH, K. E.: Uber die Erregungsleitung in Froschherzstreifen unter der Wirkung von Kalium-Ionen und anderen herzmuskeleigenen Substanzen. Z. Ges. Exptl. Med. 119: 402, SWAIN, H. H., AND WEIDNER, C. L.: A study of substances which alter intraventricular conduction in the isolated dog heart. J. Pharmacol. Exptl. Therap. 120: 137, MENDEZ, C, ERLIJ, D., AND MOE, G. K.: Indirect action of epinephrine on intraventricular conduction time. Circulation Res. 14: 318, WlNKLER, A. W., HOFF, H. E., AND SMITH, P. K.: Factors affecting the toxicity of potassium. Am. J. Physiol. 127: 430, NAHUM, L. H., AND HOFF, H. E.: Observations on potassium fibrillation. J. Pharmacol. Exptl. Therap. 65: 325, WINXLER, A. W., HOFF, H. E., AND SMITH, P. K.: Cardiovascular effects of potassium, calcium, magnesium and barium. Yale J. Biol. Med. 13: 123, WIGCERS, C. J.: Studies on ventricular fibrillation complexes resulting from surface applications of potassium salts. Am. Heart J. 5: 346, SIEBENS, A. A., HOFFMAN, B. F., ENSEN, Y., FERRELL, J. E., AND BROOKS, C. M.: Effects of 1-epenephrine and 1-nor-epenephrine on cardiac excitability. Am. J. Physiol. 175: 1, WALKER, W. J., ELKINS, J. T., AND WOOD, W. L.: Effect of potassium in restoring myocardial response to a subthreshold cardiac pacemaker. New Engl. J. Med. 271: 597, SURAWICZ, B., CHLEBUS, H., REEVES, J. T., AND GETTES, L. S.: Increase of ventricular excitability threshold by hyperpotassemia. J. Am. Med. Assoc. 191: 1049, GrcmUimm Rtsurcb, Vol. XIX, AmfMSI 1966