THE EFFECT OF LITHIUM ON THE ELECTROCARDIOGRAM OF ANIMALS AND RELATION OF THIS EFFECT TO THE RATIO OF THE INTRACELLULAR AND EXTRACELLULAR CONCENTRATIONS OF POTASSIUM Victor A. McKusick J Clin Invest. 1954;33(4):598-610. https://doi.org/10.1172/JCI102931. Research Article Find the latest version: https://jci.me/102931/pdf THE EFFECT OF LITHIUM ON THE ELECTROCARDIOGRAM OF ANIMALS AND RELATION OF THIS EFFECT TO THE RA- TIO OF THE INTRACELLULAR AND EXTRACELLULAR CONCENTRATIONS OF POTASSIUNI By VICTOR A. McKUSICK 1 (From the Clinic of Genieral Medicine anid Experimental Therapeutics of the Nationial Hcart Institute [Cardiovascular Cliniic, U.S.P.H.S. Hospital, Baltinmore, Md.] and the De- partnent of Mledicine, Johnis Hopkinis Untiversity and Hospital, Baltimore, Md.) (Submitted for publication July 16, 1953; accepted December 30, 1953) The cardiac toxicity of the potassium ion is fa- three larger species it was given by continuous intrave- miliar from observations both in animals and in nous infusion. The intravenous solutions were adminis- man. With elevation of the serum potassium a vir- tered at rates which delivered 300 mEq. of lithium in a period varying from 15 to 90 minutes in individual ex- tually pathognomonic sequence of electrocardio- periments. In the guinea pigs 20 cc. of isotonic LiCl solu- graphic changes occurs. Of interest is the effect tion was administered intraperitoneally at intervals of on the heart of elements from the same periodic 15 to 20 minutes. Electrocardiograms were recorded at group, e.g., lithium, rubidium, and cesium. With frequent intervals by means of a Sanborn direct-writinig the reporting (1-5) of patients with severe un- electrocardiograph. In dogs, contact electrodes with con- ductant jelly were used, whereas in the smaller ani- toward reactions from lithium salts taken as a so- mals copper wires piercing the limbs and attached to dium salt substitute, it became of practical signifi- "Fahnstock" clips established contacts with the lead cance to evaluate the role of the cardiac toxicity cables. In the case of the dogs, blood samples were taken (if any) in the total picture of lithium intoxication. as the successive electrocardiographic changes developed. Unexpectedly these studies of the effects of lithium In the smaller species, the blood was sampled only at cardiac arrest. All animals were anesthetized at the be- disclosed interesting features of potassium me- ginning of the experiments with 65 mg. of pentobarbital tabolism. Because of ramifications of possible sodium for each 5 lbs. of body wxeight administered theoretical significance in this connection, a miiore intraperitoneally. detailed report than that of four years ago (6) was Concentrations of sodium, potassium and lithium in the deemed desirable. serum were estimated by means of the Beckman DU In Hesse Spectrophotometer with flame attachment. In the case 1875, (7) demonstrated that lithium of lithium, observations were made at a wave length of chloride administered intravenously caused dia- 671 millimicrons. stolic stoppage of the heart while nerves and mus- For purposes of estimating lithium and inuliii spaces cles were still irritable. Krumhoff (8, 9) con- three dogs were subjected to bilateral nephrectomy. Im- firmed this finding in 1884 and made the significant mediately thereafter in two animals 50 cc. of 0.65 per additional observation that much more lithium than cent solution of LiCl and 10 cc. (1 Gm.) of inulin were administered intravenously to each animal. In a third potassium is necessary to produce cardiac arrest. animal only inulin space was determined. The blood was From neuromuscular experiments, MAilheiro (10) sampled one hour after administration of the inulin and concluded that the effect of lithium is an additive lithium solutions and daily thereafter until death. Water one with potassium. and food were withheld in these animals. Estimations of the concentration of inulin in plasma were made by Har- rison's modification (12) of the method of Alving, Rubin, METHODS and Miller (13). In acute experiments modeled after those of Winkler, The effect of lithium on the loss of potassium from in- Hoff, and Smith (11) wvith potassium, lithium chloride, cubated suspensions of human red blood cells was exam- usually in isotonic 0.65 per cent solution, was administered ined in the following manner: suspenisions of red cells (as parenterally to six dogs, one cat, two rabbits and six whole blood) were prepared in various proportions of guinea pigs (cf. Table I A, B, C, D). In the case of the isotonic LiCl (0.65 per cent) solution and of isotonic guinea pigs the solution of lithium chloride was given by NaCl (0.85 per cent) solution, as indicated in Figure 6. intermittent intraperitoneal injection, whereas in the Each suspension was prepared in triplicate and incubated at 370 C. for 24 hours. At 1, 8, and 24 hours one of each 'Present address: The Johns Hopkins Hospital, Balti- set of three was removed from incubation and the super- more 5, Maryland. natant analyzed for potassium, sodium, and lithium. 598 EFFECTS OF LITHIUM ON THE ELECTROCARDIOGRAM 599 ~~~ ~ ~ ~ ~~ '~0 e- 0a 0 c bO0 4 *O U) G - U 4 ,o.~~~~~'o~~~ o'bO U o *0 0 1~~~~~~~~~~~~~~~~~ 0~~~~~~~ ~ ~ ~~~d0 ~ 2 1 + +4 + 0 J & + N d JC 0 c a. 4 U) 0 0 00 - 0 4i 5~-w v- -0 W Cd . 00( N.t 0 0 c N '0 - N a, . u-04N - U U) N 00 V4J j4 (au v6 - 's . .-02.~~~~~~0Cd$ 4 ~ + +~~) CUU- A Z6~ 00~~~~~~~~~~0 CdC)~~ ~ ~ ~~U CU %0 4Go .-00 00 U, ~~~~~~~~~~~~~~-4 ZCU 0 0 I~ U)~~~~~otN 00t 0 o~~~~~~J & U) CU *d z '0-) U)0 0 C! 00 00~~~N) 0' 00 00 U )U A 'oN U)~~~~~~~~~~~~~~~~~~~c - - - - U)~~~~~~~~~~~~~~~0 600 VICTOR A. MCKUSICK * |(: C)' * 40 0 0 CZ 4 .0 .0 CZCZ ._ CJ)c 06 CCv 4-e Q 0c E ,Z C CCZ C) bc 0 CC~~~~~~~~~~0) u OV o V° 4-C- .--z U) ._~1 ._ CCC V> lav9 + 'A C) C -- -e cCt CZ: C)CZC -C~ :T~ .01 C) 1 c 4)C V- > 0 0 0 ,if) 14 44 C 4 (4 76 0 .0 o2 o>0 0 C) U))-CJu>: CZo.W ~ ~ . ;.C)C C.CO - C ~~~~0 C') 0 * 4 .. ->,0 -4)- CO))).° o , o i)C e; CO:C.D0C); 04, CZ -Enc 0 S.5 5o. I-) C D*l S _ 01 (4 OC) oo00 S oc C-11 4 C5 cli 1'r) ll. 0) 4i) cC 0 04 ;o 'o ~-~ ~'t -; t- -: C cO o ~c CN c0 -~~~~~~~~o CO C5 LO t- 04C -o CC -o CO ._ 0 U E-FFECTS OF LITHIUM ON THE ELECTROCARDIOGRAM 6I601 F .. ....... - 2..~~~~~~~~~~~~~~....... IV~ .-. ~~~~~~~ ~~~~~~~~~~~~k. ... 777~~~~~~~~~~4~ :~~~~~~~~~~~~~~~~~~~~~~~~~~... ... I - ....C. A FIG. 1. REPRESENTATIVE ELECTROCARDIOGRAMs-DOG No. 2 OF TABLE IA A. Control tracings (2 :30 p.m.). B. The first change, peaking of T waves, is demonstrated (4:15 p.m.). C. Auricular fibrillation and pronounced slowing of the ventricular rate have developed (7:45 p.m.). The T waves are markedly peaked. RESULTS fibrillation in about an equal number of animals. At autopsy all of these animals showed dilated The electrocardiographic effects of acute parenteral hearts suggesting cessation in diastole. (Loud lithium administration in animals systolic murmurs suggesting cardiac dilatation A consistent series of electrocardiographic events were heard in the later stages.) In some there were in the was observed in all 15 animals of the four species petechial hemorrhages myocardium. Both of these changes were described Krum- studied (Figures 1 and 2). First, the T waves be- by hoff (8). came high, narrow, and peaked. Secondly, atrial Two guinea pigs were given potassium chloride standstill and less frequently atrial fibrillation solution in an exactly comparable manner. These supervened. First, second, and third degree atrio- animals also showed impairment of A-V conduc- ventricular block was observed in some guinea pigs tion before the development of atrial standstill or before the development of the abnormality of atrial atrial fibrillation, both of which occurred in these mechanism. In two dog experiments intra-atrial animals. In general, the same sequence of changes electrocardiograms by electrode catheter corrobo- was seen as has been described for potassium in rated the impression of auricular fibrillation fol- the dog ( 1 1). lowed by atrial standstill at this stage. Thirdly, The significant finding was that lithium pro- widening of the QRS appeared and finally there duced the same electrocardiographic sequence as developed a bizarre, biphasic QRS-T pattern at a did potassium. Hitherto the series of changes slow rate, omen of cardiac arrest. The terminal described has been considered unique for hyper- event was ventricular standstill or ventricular kalemia. 602 VICTOR A. MCKUSWcX Calcium appeared to antagonize the electro- in serum lithium concentration, progressed through cardiographic effects of lithium to some extent. the characteristic changes to ventricular standstill. Sodium was a more effective antagonist. In one A reciprocal relationship between lithium and so- dog carried to end-stage changes with lithium, (iuum concentrations is demonstrated (Figures 4 striking improvement, including the return of P and 5). (There are suggestions from clinical ex- waves, accompanied saline infusion. perience [14] that hyponatremia exaggerates the electrocardiographic effects of hyperkalemiiia. Such DISCUSSION may have been the case to some extent in this situation.) No data was obtained bearing on the The mechanism of electrocardiographic the changes question of whether the hyponatremia was the re- of hyperlithemia sult of dilution, intracellular shift, or natriuresis, Figures 3, 4, and 5, present charts of three rep- or of some combination of these mechanisms. resentative dog experiments. In all of the dogs, Inasmuch as the serum lithium concentration serum potassium concentration rose steadily in the bore no direct relationship to the stage of electro- presence of hyperlithemia, attaining a level of cardiographic change observed and since the se- about 10 mEq.