Effect of Lithium on Cardiovascular Extended Ambulatory



Report on

and Exercise Testing

Before and During Lithium Therapy








From the Division of Cardiology* and the Department of Psychiatry?, Veterans Administration Hospital, Palo Alto, and Stanford University Medical School, Stanford, Calif. This work was supported in part by the Palo Alto Veterans Administration Hospital and Grants HL-5866. HL-5709 and Program Project Grant l-POl-HL-15833 from the National Institutes of Health, Bethesda, Md. Manuscript received May 6, 1975; revised manuscript received March 18, 1976, accepted March 24, 1976. Address for reprints: John S. Schroeder, MD, Cardiology Division, Stanford University Medical School, Stanford, Calif. 94305.

To assess the effect of long-term lithium therapy on cardiac arrhythmias and cardiovascular performance, extended ambulatory electrocardiographic monitoring was performed in 12 patients, end rest and exercise electrocardiograms in 10 of 12, before and during lithium therapy. Lithium increased the frequency of premature ventricular contractions in three patients, decreased it in one, and produced no change in eight. Three of four patients with atrial arrhythmias showed improvement during lithium therapy. Exercise performance was unchanged. Although 7 of the 12 patients manifested T wave flattening in the resting electrocardiogram, none had S-T segment displacement at rest or on treadmill exercise. Before lithium therapy, arrhythmias on exercise included premature atrial contractions in four patients, ventricular arrhythmias in four (premature ventricular contractions in four, with couplets in two and with ventricular tachycardia in one). During lithium therapy, exercise did not provoke premature atrial contractions or ventricular tachycardia in any of the patients, but three patients had premature ventricular contractions (with couplets in one case). We conclude that lithium at therapeutic levels may precipitate or aggravate ventricular arrhythmias. When administered to patients with heart disease, factors that interfere with renal clearance of lithium (heart failure, salt restriction, long-term diuretic therapy) must be recognized and doses must be adjusted accordingly. Careful follow-up and electrocardiographic monitoring are advisable if lithium is to be used in the presence of ventricular arrhythmias. Cardiovascular performance as assessed by treadmill exercise testing was not affected by long-term lithium therapy.

Lithium was introduced into psychiatric pharmacology in 1949,’ but did not receive widespread attention until the last decade. Numerous studies in the past 10 years have established lithium as a valuable agent in the treatment of mania and for the prevention of a relapse in bipolar manic-depressive illness.’ Beneficial results of lithium have also been reported in various other psychiatric disorders, but its value in disorders other than manic-depressive psychoses remains controversial.:i The presence of cardiovascular disease and arrhythmias is generally considered a contraindication to the use of lithium,“~~i but a search of the literature reveals contradictory and inconclusive reports.” Many psychiatric patients with cardiac disease or arrhythmias have now been exposed to this drug. This study was designed to evaluate the effects of therapeutic doses of lithium on cardiac arrhythmias, exercise performance and the exercise electrocardiogram.

November 23, 1976

The American Journal of CARDIOLOGY

Volume 39





methods Patient Population Twelve patients, 11 men and 1 woman, with various psychiatric diagnoses were studied (Table I). Only adult patients who were able to give informed consent and whose attending psychiatrist considered them a candidate for lithium therapy were included in the study. Severely agitated patients or those in whom a 48 hour delay in therapy was deemed deleterious were excluded. Patients receiving long-term treatment with other medications were accepted if no change in dose was made during the study period. No attempt was made to exclude patients with known preexisting heart disease or arrhythmias. Studies All patients were examined by one of us (A.T.) and gave informed consent. The study was approved by the Stanford University School of Medicine and the Palo Alto Veterans Administration Hospital Committees on the Use of Human Subjects in Research. All patients underwent the following studies before lithium administration: complete blood count, urinalysis, determinations of serum creatinine, sodium, potassium, chloride and carbon dioxide, evaluation of thyroid function, chest X-ray and electrocardiogram. Exercise electrocardiography in 10 of the 12 patients and 42 to 48 hour ambulatory electrocardiographic monitoring in all patients immediately preceded administration of lithium. Lithium carbonate was given orally, as prescribed by the attending psychiatrist, in daily divided doses ranging from 900 to 1,500 mg (24 to 40 mEq lithium/day). Serum levels of lithium were measured three times weekly. When these reached therapeugic levels (0.6 to 1.5 mEq/liter) in 6 to 14 days, 24 to 48 hour ambulatory electrocardiographic monitoring, exercise electrocardiogram and laboratory testing were repeated. Blood for serum lithium and other electrolyte determinations was drawn in the fasting state. Lithium levels were determined by atomic absorption spectrophotometry using a Perkin-Elmer no. 303 analyzer. Arrhythmia detection: Ambulatory electrocardiographic monitoring was continued in four patients for 8 to 10 days without interruption, in two of these patients (Cases 11 and

12) because of preexisting severe coronary artery disease with arrhythmias and in the other two (Cases 9 and lo), because frequent premature ventricular contractions were noted on the initial examination. Except for Patients 11 and 12, who were restricted to a cardiac surveillance unit for added protection, all patients continued their usual level of unrestricted activily. The 12 lead electrocardiogram was obtained with the patients supine in the postabsorptive state with use of a single channel Hewlett-Packard no. 1511 electrocardiograph. The ambulatory tracings were recorded using a single channel Avionics electrocardiographic tape recorder and a modified V1 lead system. The patients recorded a detailed diary of symptoms and activities, including sleep, as they followed their normal daily routine. The tapes were processed at 60 times real-time, using a digital computer system developed at Stanford Medical Center? This system provides plots of R-R interval, QRS duration and mean QRS vector for each heartbeat during the entire recording. These data are plotted simultaneously as a function of time, allowing visual detection of all arrhythmias. Verification of arrhythmias is made by sampling a digitized electrocardiographic signal from unlimited numbers of 1 minute segments of the recording. These segments are selected after visual inspection of the plots of R-R int,erval, QRS duration and mean QRS vector. Frequent or continuous sampling is made when there is a noisy signal or when there are complex arrhythmias. The entire record was divided into 1 hour segments and the total number of premature atria1 or ventricular contractions per hour was determined by visual inspection of the computer-generated plots. The presence of multiform premature ventricular contracGons, early (R on T) contractions, couplets, bigeminy and ventricular or supraventricular tachycardias was noted for each 24 hour recording. From the patient’s diary the recording was divided into waking and sleeping periods, and each period was analyzed separately. In each patient the data obtained before treatment with lithium were compared with those obtained when lithium blood levels were in the therapeutic range. Exercise electrocardiogram: The test was performed on a motorized treadmill and followed the protocol of Bruce and Hornstens (Patients 2 and 3 refused this test). Leads II and

TABLE I Clinical Status of Patients Before Lithium Theraw Case no.


History of Heart Disease

Abnormal Cardiac Findings

Chest X-Ray


Other Medications

ECG Data


Normal Normal ST&


None None Bilateral rales; s, gallop; pedal edema None None None None Mild Al

Normal Normal Cardiomegaly

51M 42M 6OM 54M 57M

None None HBP, CHF PAT None HBP PAT None None

Normal Normal Normal Normal Normal

Normal Normal Normal Normal Normal

9 10 11 12

51M 25M 64M 50M

None None Ml, VT Ml, Angina

PCS PCS PCs, S, gallop PCs, S, gallop

Normal Normal Cardiomegaly Cardiomegaly

PVCS PVCS PVCs, Ant MI PVCs,Ant Ml Tolbutamide

: 3

4 5 6


Age fyr) Sex 34F 56M 73M

Haloperrdol, 20 mg/day Digoxin, 0.25 mg/day; furosemide, 40 mg/day; haloperrdol, 6 mg/day Hydrochlorothia~ide, 50 mgiday Digoxin, 0.25 mgiday Diphenhydramine, 200 mg/day Diphenyihydantoin, 300 mglday; chlorpromazine, 300 mgiday .

Al = aortic insufficiency; Ant = anterior; CHF = congestive heart failure; ECG = electrocardiographic; HBP = hrgh blood pressure; MI = myocardisl infarction; PAT = paroxysmal atrial tachycardia; PCs = premature contractions; PVCs = premature ventricular contracttons; ST1 = S-T segment depression; VT = ventricular tachycardia.


November 23, 1976

The American Journal of CA~DIOL~Y

Volume 38



though three were on a salt-restricted diet and two were receiving long-term diuretic therapy.

V:, were monitored. On the initial test (before administration of lithium), patients were urged to perform maximally, and the tests were stopped at the request of the patient. On the second test (during lithium therapy) patients were encouraged to achieve the same level of exercise (in seconds) but were not permitted to exceed this level. No test was discontinued because of arrhythmia. Blood pressure was measured by a sphygmomanometer, with t.he patient standing, before and immediately after the exercise. The electrocardiogram was monitored continuously, and recordings of 20 seconds’ duration were obtained every minute of exercise and for 8 minutes of the recovery period. All premature beats occurring during exercise or the first 8 minutes of recovery were counted and recorded for every 100 sinus beats. Mathematical and statistical analyses were performed on a Hewlett-Packard model 9100A calculator, using standard

Electrocardiogram Resting electrocardiogram: Changes in the 12 lead tracing during lithium therapy were minor and limited to T wave changes. Seven patients had 1 mm or more of T wave depression or flattening, and five showed no change. No T wave changes were seen in the patients with severe cardiac disease (Cases 3, 11 and 12), all of whom began the study period with an abnormal electrocardiogram. Ambulatory monitoring: Table II summarizes the number of hours of ambulatory electrocardiographic monitoring during waking hours and sleep, and serum electrolytes and lithium levels during the ambulatory monitoring period for each patient. The duration of monitoring ranged from 27 to 74 hours (mean 42 hours) before lithium therapy and from 16 to 193 hours (mean 62 hours) during therapy. Serum electrolytes were normal and showed no significant change during lithium therapy. The number of days of lithium therapy before repeat ambulatory electrocardiographic monitoring ranged from 6 to 14. Serum lithium levels during the period of monitoring were in the therapeutic range of 0.8 to 1.4 mEq/liter (mean 1.2 mEq/liter) for all patients. In Patients 9 through 12, who were monitored continuously during the 6 to 8 days of lithium administration, peak serum lithium levels ranged from 1.1 to 1.4 mEq/liter.

programs. The mean number of premature ventricular and atria1 contractions before and during lithium therapy was compared using a standard two-tailed t test. Results Clinical Course Relevant clinical information is summarized in Table I. Before lithium therapy, four patients (Cases 1, 24 and 7) were completely free of detectable heart disease. Two patients (Cases 9 and 10) had frequent premature ventricular contractions but no other detectable abnormalities noted in extensive noninvasive studies, including a maximal treadmill electrocardiographic test and echocardiograms, carefully searching for mitral valve prolapse. Three patients (Cases 5,6, and 8) had mild abnormalities detected by history or-physical examination but remained free of symptoms. Three patients f Cases 3,ll and 12) had advanced coronary artery disease with a history of myocardial infarction, stable angina, congestive heart failure and short bursts of ventricular tachycardia. During lithium therapy: Cardiac effects such as symptomatic arrhythmias, worsening angina or heart failure were not demonstrated by any patient during lithium therapy. Because serum lithium levels were monitored closely, no patient had lithium toxicity, al-



Ventr~cuiar Arrhythmias During the entire study period, all patients except one (Case 5) had some ventricular arrhythmia. In five patients (Cases 1,2,4,6 and 8) the arrhythmia was limited to isolated rare premature ventricular contractions ranging in frequency from 1 to 7/day, with no change during lithium administration. These patients were considered to be free of significant arrhythmia and were excluded from further analysis. Five of the remaining six patients had frequent premature ventricular con-


Duration of Ambulatory Lithium Therapy



Case “0.

: 3 4 :

: 9 10 11 12 A = while



and Serum


Serum (mEq/liter)

32A: 20A; 32A; 29A; 28A; 21A;

14s 8s 14s 18s 9s 13s

25A; 14A; 50A; 23A; 20A; 23A;

185 13s 24s 19s 185 17s



4.0 3.9 4.8 4.4 4.4 4.8 4.5 4.3 4.3 4.4 4.7 4.3

139 141 137 139 143 140 137 137 140 140 142 143

awake; ECG = electrocardiogram;

and Lithium


Hours of




co, 27 25 33 22 25 31 20 18 21 30 23 24

S = while

Hours of ECG

Davs of Therapy 14 10 8 14 14 10 6 10 s” 6 6

30A; IOA; 9A; 21A; 27A; 22A; 2pA; 25A; 67A; 135A; 45A; 62A;

14s 6s 8s 22s 1 QS 125 18s 16s 335 5% 335 28s


Levels Before

and During

Therapy Serum (mEq/liter)




4.3 :,:: 4:6

141 140 139 143



4:8 4.7 3.8 4.7 4.7 3.8 4.2

ii1 140 136 740 143 140 141

104 99 99 107 102 9% 102 98 99 102 99 100


Li+ 1.3 14 019 0.9 1.3 14

0.3-l 0.3-l

.4 .3


November 23,1976

The American Journal of CARDIOLOGY

Volume 38










P c1 ’ %



60 -

FIGURE 1. Effect of lithium on premature ventricular contractions (PVC’s) during waking hours (A) and sleep (6). The level of statistical significance of the change in mean number of premature ventricular contractions per hour is given for each patient. During the waking state, the frequency of these contractions increased in three patients, lessened in one and showed no change in two. During sleep, the frequency increased in two patients, lessened in one and showed no change in three. NS = not significant; P = probability.


; 2






0 2 s

80 -


60 -




\ 4









(12) 0


(3)l - -* (8).

(7) e







1,000 F

FIGURE 2. Effect of sleep on premature ventricular contractions in patients with more than 100 such contractions in 24 hours, before (A) and during (B) lithium therapy. The level of statistical significance of the change in mean number of premature ventricular contractions per hour is given for each patient. All patients showed a decrease in frequency during sleep both before and during lithium therapy.





P< 0.001 ~

0 peo.05



Pc 0.001 P< 8.05






November 23, 1976

The American Journal of CARDIOLOGY

Volume 39

FIGURE 3. Effect of lithium on premature atrial contractions (PAC’s) during waking hours (A) and sleep (B). The level of statistical significance of the change in the mean number of premature atrial contractions/hour is given for each patient. In three of our patients who had premature atrial contractions before lithium therapy the contractions decreased in frequency during both waking hours and steep after lithium therapy. The effect of sleep on frequency was inconsistent.


tractions during the control monitoring period. While awake during lithium therapy, two patients had more frequent premature ventricular contractions, two showed no change and one had fewer premature ventricular contractions. Only one patient (Case 7) had frequent premature ventricular contractions during lithium therapy but none before therapy (Fig. IA). The effect of lithium on these six patients during periods of sleep was similar (Fig. IB). Effect of sleep: Before lithium therapy, all patients with frequent premature ventricular contractions had a highly significant decrease in these arrhythmias during sleep, This apparent antiarrhythmic effect of sleep remained unchanged during lithium therapy (Fig. 2). Complex ventricular arrhythmias: Episodes of multiform premature ventricular contractions, couplets, bigeminy or ventricular tachycardia were seen in four patients both before and during lithium therapy. It was not possible to count all such episodes accurately, but no clear increase after lithium was noted (Table III). Symptomatic arrhythmia did not develop in any patient. Atrial Arrhythmias

Atria1 arrhythmias were rare (Table III). Five patients (Cases 1,5,9, 10 and 11) had none, three (Cases 2,4 and 6) had rare premature atria1 contractions before lithium therapy, with no change during therapy. Four patients had a rate of 1.6 to 284.0 such contractions per hour before lithium therapy. Three of these four patients had a decrease in frequency of these contractions during lithium therapy, both while awake and while asleep (Fig. 3). Supraventricular tachycardia of 5 to 10 beats was recorded seven times in two patients before-and once during-lithium therapy.

TABLE 111 Summary of Arrhythmias Observed During Ambulatory ----


Before Lithium


Table III provides a summary of all arrhythmias observed during ambulatory electrocardiographic monitoring. Three patients (Cases 7, 10 and 11) had more frequent or more severe ventricular arrhythmias during lithium therapy, one patient (Case 9) showed improvement and eight showed no change. In all patients with premature atria1 contractions or supraventricular tachycardia, these arrh~hmias decreased during lithium therapy, and neither arrh~hmia developed during lithium therapy. Exercise


Exercise performance: The results of the exercise electrocardiogram are summarized in Table IV. Duration of exercise before and during lithium therapy was identical for all 10 patients, indicating that no deterioration of exercise capacity occurred during therapy. At the same work load for the two exercise tests, the maximal heart rate before lithium therapy was slightly higher than that observed during therapy (P

Effect of lithium on cardiovascular performance: report on extended ambulatory monitoring and exercise testing before and during lithium therapy.

Effect of Lithium on Cardiovascular Extended Ambulatory Monitoring Performance: Report on and Exercise Testing Before and During Lithium Therapy...
1MB Sizes 0 Downloads 0 Views