Correspondence: Masanobu Ito, MD, PhD, Department of Psychiatry, Juntendo University Faculty of Medicine, 2-1-1 Hongo, Bunkyo-ku, Tokyo 113-8431, Japan. Email: email@example.com
Change in catecholamine seems to be associated with not only effects of electroconvulsive therapy (ECT), but also adverse events associated with ECT. Our aim in this study was to investigate whether or not ECT influences the concentration of catecholamine over the long term. Patients with a major depressive episode or schizophrenia, diagnosed according to DSM-IV criteria, who were newly admitted to Juntendo University Hospital to receive ECT, were recruited.
Urine was collected during the 24 h before the first ECT treatment, during the 24 h after the first ECT treatment, during the 24 h after the final ECT treatment and during the 24 h 1 week after the final ECT treatment. Heart rate, the Hamilton Rating Scale for Depression and the Positive and Negative Syndrome Scale were assessed before and after ECT.
Twenty-four patients were included in the final sample, which consisted of 14 patients with major depressive episodes and 10 patients with schizophrenia. Abnormal electrocardiograms were indicated in four patients with depression during the ECT operation but all recovered naturally. There were no significant differences in the levels of dopamine, adrenaline or noradrenaline the day before the first ECT, a day after the first ECT, a day after the final ECT and a week after the final ECT.
These results suggest that ECT does not alter urine catecholamine levels after ECT over the long term. Further studies will be required to confirm these findings in a larger sample of patients.
WHILE ELECTROCONVULSIVE THERAPY (ECT) remains the most effective treatment for major depression, its mechanism of action is unclear. Both serotonergic and noradrenergic systems, as well as other neurotransmitter systems, have been implicated.[1, 2] Electrically induced convulsions produce marked cardiovascular changes indicative of strong vagal and sympathetic stimulation with concomitant changes in plasma and urinary catecholamine levels. With ECT, blood pressure and heart rate rise, and plasma levels of adrenaline and noradrenaline increase steeply to two to six times their normal levels. These values return to normal within 10 min.[3, 4]
Although the frequency of severe cardiac complications is not high with ECT, cardiovascular strain is largely manifested by marked paroxysmal rises in the arterial and venous pressures and by disturbances in the initiation and conduction of the cardiac impulse. One such condition is takotsubo cardiomyopathy, first described in Japan in the 1990s. It is a condition in which the shape of the left ventricle during systole resembles an octopus fishing trap, or takotsubo. There are several reports on takotsubo cardiomyopathy after ECT.[8-10] In addition, the coexistence of major depression with chronic heart failure (CHF) doubles the mortality rate of CHF.[11-14] CHF patients with asymptomatic left ventricular dysfunction coupled with high plasma noradrenaline levels also have a doubling in mortality.[15, 16] Even a modest (25%) increase in plasma noradrenaline levels over a 4-month period more than doubled mortality in CHF patients.
Thus, changes in catecholamine levels seem to be associated with adverse events resulting from ECT. The relationships between the levels of cerebrospinal fluid (CSF) noradrenaline or plasma noradrenaline and the effects of ECT remain controversial.[14, 18] However, levels of plasma catecholamine metabolize quickly, so it is difficult to determine the total change in catecholamine induced by ECT treatments over the long term. To detect it, we measured urine collected for 24 h. Our aim in this study was to determine whether or not ECT treatments influence the concentration of catecholamine over the long term.
This was a prospective observational study. All study protocols were approved by the institutional review board.
Ethnicity of all patients was Asian. Patients with a major depressive episode or schizophrenia, diagnosed according to DSM-IV criteria, who were newly admitted to Juntendo University Hospital to receive ECT, were recruited. Patients who were suffering from serious physical illness or having problems with alcohol or drug use were excluded. Each patient gave written, informed consent for participation in the study. If they took tricyclic antidepressant or Li, they were discontinued more than a week before the first ECT, as these drugs could cause arrhythmia at high risk.
All patients ate hospital diets during hospitalization.
Clinical evaluation of patients
The Hamilton Rating Scale for Depression (Ham-D) was used before the first ECT treatment and a week after the final ECT treatment in depressive patients.
The Positive and Negative Syndrome Scale (PANSS) was used before the first ECT treatment and 1 week after the final ECT treatment in patients with schizophrenia.
All subjects were induced with thiopental 1.5 mg/kg and paralyzed using suxamethonium 1.0 mg/kg. After being ventilated manually, all patients received standard bilateral brief-pulse ECT using the Thymatron System IV (Somatics, Lake Bluff, IL, USA) set at a current of 0.9 A and brief-pulse square wave with pulse width of 0.5 msec. The half-age method was used to determine the energy level.
The mean number of treatments received was 9.5 (range 5–20). Psychotropic medications were continued. Seizures were evaluated using the postictal suppression index % (PSI) in the Thymatron System IV, illustrating the degree of electroencephalography flattening immediately following the seizure, which has been reported to correlate with clinical efficacy.[21-25]
Collection and analysis of urine levels of catecholamine
Urine samples were taken for measurement of urine catecholamine before and after ECT treatment in a course of treatment according to the schedule in Figure 1. Subjects collected their urine for 24 h (from 09.00 hours to 09.00 hours) in a pot with 20 mL of 6N-hydrochloric acid to prevent catecholamine metabolism and they were stored at 4°C in a refrigerator. The pH of stored urine was kept from 1.0 to 3.0. A standard protocol was followed for the collection of all urine samples for the catecholamine assay. The total amount of catecholamine (dopamine, adrenaline and noradrenaline) was measured by high performance liquid chromatography (SRL, Tokyo, Japan). Reference ranges for each of the catecholamine measures are 365.0–961.5 μg/day (dopamine), 3.4–26.9 μg/day (adrenaline) and 48.6–168.4 μg/day (noradrenaline). Urine was collected during the 24 h before the first ECT treatment, during the 24 h after the first ECT treatment, during the 24 h after the final ECT treatment and during the 24 h 1 week after the final ECT treatment.
Heart rate and blood pressure
We measured heart rate and blood pressure before the first ECT and a week after the final ECT.
Electrocardiograms (ECG) were assessed before the first ECT and 1 week after the final ECT. In addition, ECG was monitored during each ECT treatment. ECG was double-checked by cardiologists.
Statistical processing and analysis of results were performed using spss for Windows version 11.0.1J (spss, Tokyo, Japan). All tests performed were two-tailed. A Shapiro–Wilk test was performed to test for normality of the sample distribution. A Friedman test was used to evaluate differences across multiple tests for each subject. Values of P < 0.05 were considered statistically significant. The significance of differences between groups in heart rate and blood pressure was assessed using the Mann–Whitney U-test.
A total of 43 patients were registered in the study. Of all the patients recruited, 11 patients were unable to collect urine correctly; and seven patients could not continue ECT treatment due to adverse events, such as prolonged delirium in three patients, changes from depressive to manic phase in two patients, prolonged tachycardia in one patient, and pulmonary edema one patient. In addition, one patient withdrew consent. Thus, 24 patients were included in the final analysis, including 14 with a major depressive episode and 10 with schizophrenia. Abnormal findings of ECG were not detected in all patients before and 1 week after the final ECT (Tables 1 and 2). Although abnormal findings, such as transient premature ventricular contraction (PVC) and transient atrial fibrillation (Af) during ECT treatments, were observed in four patients with depression, they were transient and of little clinical significance. In the first case, the QRS wave increased for a few minutes on the fifth ECT in five ECT sessions. In the second case, transient PVC was indicated for a few minutes on the fifth and sixth ECT of six ECT sessions. In the third case, transient Af was caused on the eighth ECT of eight ECT sessions. In the fourth case, arrhythmia was observed during the first ECT in a series of seven ECT sessions. ECT was continued after we consulted with a cardiologist. No schizophrenia patient experienced abnormal ECG during the ECT operation.
Table 1. Summary of the patients with depression (chronological catecholamine data were analyzed with Friedman's test)
Data of patients with depression
SD or range
***P < 0.001.
P: P-value of chronological data in depression.
ECG, electrocardiogram; ECT, electroconvulsive therapy; Ham-D, Hamilton Rating Scale for Depression.
Tricyclic antidepressants and Li were discontinued more than a week before the first ECT treatment.
†Number of patients with monotherapy. ‡Number of patients with the drug combined with other drugs. ECT, electroconvulsive therapy.
Mean Ham-D scores after ECT (9.88, SD = 6.49) were reduced in comparison to scores before ECT (mean = 40.94, SD = 7.89, P = 0.000976) (Table 1).
Patients with schizophrenia took some medicines during ECT (Table 3). There was no significant change in heart rate before and after ECT (mean = 86.2, SD = 18.40; mean = 82.9, SD = 18.63).
Mean total PANSS scores after ECT (30.42, SD = 20.20) were reduced in comparison to scores before ECT (mean = 124.1, SD = 63.20, P = 0.00506). Mean PANSS positive and negative scores after ECT (mean = 13.5, SD = 4.25; mean = 19.5, SD = 6.65) were reduced compared to scores before ECT (mean = 30.2, SD = 10.04; mean = 29.4, SD = 11.26) (Table 2).
The Shapiro–Wilk test indicated that chronological groups in catecholamine did not fit a normal distribution, so we performed a Friedman test.
Friedman values were 0.0565 in patients with depression and 0.145 in patients with schizophrenia. The results indicated that there were no significant differences in dopamine levels at 1 day before the first ECT, 1 day after the first ECT, 1 day after the final ECT and 1 week after the final ECT (Tables 1 and 2).
Friedman values were 0.112 in patients with depression and 0.421 in patients with schizophrenia. The results indicated that there were no significant differences in adrenaline levels at 1 day before the first ECT, 1 day after the first ECT, 1 day after the final ECT and 1 week after the final ECT (Tables 1 and 2).
Friedman values were 0.258 in patients with depression and 0.472 in patients with schizophrenia. The results indicated that there were no significant differences in noradrenaline levels at 1 day before the first ECT, 1 day after the first ECT, 1 day after the final ECT and 1 week after the final ECT (Tables 1 and 2).
Case with tachycardia after ECT
A 59-year-old woman (height 151.4 cm; weight 53 kg) was hospitalized for a severe depressive episode after failure of antidepressant pharmacotherapy. Physical examinations yielded unremarkable results, and findings from standard laboratory and imaging tests, including ECG, were normal. An ECT regimen was commenced after her admission. Blood pressure (BP) was 117/69 mmHg, and heart rate was 118 b.p.m. She received 0.5 mg of atropine sulfate intramuscularly 30 min before ECT treatment, then was induced with thiopental 65 mg and paralyzed using suxamethonium 50 mg. After being ventilated manually, the patient underwent bilateral ECT using the Thymatron System IV (Somatics) set at a current level of 0.9 A and a brief-pulse square wave with a pulse width of 0.5 msec. The half-age method was used to determine the energy level, which was 30%. Esmolol hydrochloride (20 mg) was given as poststimulus delivery. The patient experienced no adverse consequences. A fifth ECT treatment was started using the same procedures. The patient's heart rate increased to 140 b.p.m. for 20 min after ECT. An anesthetist injected verapamil hydrochloride (5 mg) intravenously. Her heart rate quickly decreased to 99 b.p.m. Catecholamine did not change very much after the fifth ECT. The patient was discharged from the hospital because her symptoms did not recur during the 10 days after the fifth ECT.
Patients with cardiovascular complication
Five of 24 patients (21%) experienced cardiovascular complications. ECT sessions were stopped in one patient because her tachycardia continued for a prolonged period. No patient with schizophrenia showed any complications. We divided patients into two groups according to their ECG. The abnormal ECG group included patients with depression and abnormal ECG during the ECT operation and the normal ECG group included patients with depression and normal ECG during the ECT operation (Fig. 2). We could not statistically analyze these data as there were only four patients with abnormal ECG. Ham-D after ECT in the abnormal ECG group seemed to be lower than in the normal ECG group.
Dopamine concentration in the urine 1 week after the final ECT in the abnormal ECG group seemed to be higher than in the normal ECG group (mean = 1.189 μg/mL, SD = 0.466; mean = 0.479 μg/mL, SD = 0.765).
Cardiac events are the leading cause of death in patients treated with ECT. Electrocardiographic changes during seizures and in the post-seizure period have gained attention. We assessed 43 patients who underwent ECT, including one patient who suffered a severe cardiac event, prolonged tachycardia, after ECT.
Variation in urine concentrations of catecholamine had no effect on maintenance of an acute response to ECT, and urine concentration of catecholamine did not change before or after ECT in patients with depression or schizophrenia. This is the first study on urine concentrations of catecholamine over a long period of time after ECT. Blood pressure and heart rate did not change before or after ECT.
Several previous studies have reported plasma levels of catecholamines after ECT.[14, 18, 28] In 2005, Gold indicated that noradrenaline in the cerebrospinal fluid fell to control values after ECT in depressive patients. In 1983, Linnoila showed that a series of 12 ECT reduced urinary dopamine and noradrenaline output significantly in depressed patients, but 3-methoxy-4-hydroxyhenylglycol (MHPG) and vanillylmandelic acid (VMA) did not change significantly before or after ECT.[29, 30] Linnoila collected 24-h urine samples during the second week after the last ECT. However, they did not describe the conditions of their ECT treatments. In 1988, Cooper described the effect of ECT on cerebrospinal fluid amine levels in schizophrenia and reported that no significant changes were observed in the concentration of MHPG. These results suggest that catecholamine levels increase immediately after ECT, but do not increase for the long term. The present finding that ECT does not alter urine catecholamine levels after ECT over the long term might make a contribution to understanding of physiological change among many factors for cardiovascular adverse events in ECT therapy.
Catecholamine levels and cardiovascular complications
There are several reports on takotsubo cardiomyopathy after ECT.[8-10, 32, 33] This condition is related to plasma catecholamine levels. One patient in the present study suffered from complications in the form of prolonged tachycardia immediately after ECT. In this case, urine adrenaline concentrations after the fifth ECT increased. This increase may be related to tachycardia because ECT produced a hyperdynamic cardiac state which was mediated by an ECT-induced increase in circulating catecholamine, particularly adrenaline.[4, 35] However, some patients in our study did not suffer the complication of prolonged tachycardia even when their urine adrenaline concentrations were higher than this patient. This may indicate that predicting cardiovascular complications after ECT is difficult with data based only on urine concentrations.
Does higher cardiovascular response to ECT predict early antidepressant effects?
It is very difficult to predict the effect of ECT, but some studies indicate that postictal cardiovascular responses, which included heart rate, blood pressure and rate pressure product (RPP), predicted the therapeutic efficacy of ECT for depression.[36, 37] In addition, five of 24 patients (21%) had cardiovascular complications. This percentage is similar to previous reports that showed rates of 7–27%.[38-40] We could not analyze these data statistically as the number of patients was very small. The mechanism of action of ECT might cause a short-term rise in catecholamine. However, an increase of the RPP with atropine does not change the effect of ECT. Controlling the catecholamine response might not attenuate the effect of ECT. To prevent cardiovascular complications, including takotsubo cardiomyopathy, we should control catecholamine levels. Further replication studies in larger samples are needed to draw definitive conclusions.
The first major limitation of the present study is that patients continued to take medicine, so we could not exclude the possibility that catecholamine concentration was altered by the medicine. The second major limitation is that we did not measure concentration of metabolites of catecholamine, although monoamines in plasma are rapidly metabolized into several metabolites, such as homovanillic acid, vanillylmandelic acid, and 3-methoxy-4-hydrophenylglycol. These levels in urine should be demonstrated in future studies, with ratios of metabolites and parent compounds.
There was no significant difference in urine catecholamine, dopamine, adrenaline or noradrenaline during ECT or a week after ECT. This result suggests that ECT does not alter urine catecholamine levels after ECT over the long term. Further studies will be required to confirm and extend these findings.
This study was funded in part by the Juntendo Institute of Mental Health and a High Technology Research Center Grant from the Ministry of Education, Culture, Sports, Science, and Technology of Japan and the Sportology Center at Juntendo University Graduate School of Medicine. We thank Tokiko Hatano MD, PhD, Emi Satomura, MD, Miwa Komatsu, MD, Yohei Kita, MD, Kyoko Nakamura, MD, Noriko Nakao, MD and nurses in the psychiatry ward for assistance in carrying out this study.