German Journal of Psychiatry

ISSN 1433-1055


Seizure Threshold Changes During Acute and Maintenance ECT in Schizophrenic Patients

Worrawat Chanpattana1, M.L. Somchai Chakrabhand2

1Department of Psychiatry, Srinakharinwirot University

2Department of Mental Health

 Corresponding author: Worrawat Chanpattana, M.D., Department of Psychiatry, Srinakharinwirot University, 681 Samsen, Dusit, Bangkok 10300, Thailand, E-mail: worch@loxinfo.co.th


Background: Determination of the seizure threshold can help guide selection of stimulus dosage in electroconvulsive therapy (ECT). We lack this information in patients with schizophrenia as well as in maintenance ECT (M-ECT). Objective: To determine seizure-threshold changes during acute and M-ECT in 41 patients with schizophrenia. Method: In acute ECT (Phase I), initial threshold was estimated at the first two treatment sessions, then the thresholds were quantified at the seventh, fourteenth, and twentieth ECT. During M-ECT (Phase II), seizure thresholds were estimated at the first treatment, then every 3 months for 1 year. Results: There was a 213 + 179% rise in seizure threshold at the end of Phase I. At the end of Phase II, fifteen patients had no change in thresholds compared to the first estimates of Phase II; eighteen others showed a further increase, and thresholds of the last eight patients decreased gradually. The magnitude of threshold increase of Phase II was 17 + 43%. The overall increase in thresholds at the end of Phase II was 243 + 178%. Conclusions:  Increases in seizure threshold seen during acute ECT are generally sustained during maintenance ECT in remitted patients with schizophrenia (German J Psychiatry 2000;3: 25-36).

Key words: Seizure threshold, acute and maintenance ECT, empirical titration technique, anticonvulsant effect, schizophrenia

Supported by the Thailand Research Fund, grant BRG 3980009  


Since its inception in 1938, optimization of electroconvulsive therapy (ECT) has been a focus of interest (Potter, 1994; Weiner, 1994). The most fundamental view of the mechanism of action of ECT probably came from the classic research conducted by Ottosson (1960 a&b, 1962). This work led to the widely adopted conclusions that 1) eliciting an adequate generalized seizure is both necessary and sufficient for the efficacy of ECT; and 2) increasing the stimulus intensity above that necessary to elicit an adequate seizure does not enhance either the response rate or the speed of clinical response, but results in increased cognitive side effects. The consensus that optimization of ECT is to insure that each patient has an adequate seizure at each treatment by using a minimum dosage of the stimulus charge (Fraser 1982) has had a great impact on clinical practice. The National Institute of Health Consensus Conference of ECT (1985) also recommended that the lowest amount of electrical energy to induce an adequate seizure should be used.

A number of studies conducted over the last decade have demonstrated that each of these central beliefs is wrong. Electrically induced seizures are not Ďall-or-noneí phenomena, and are subject to a wide variety of influences that may affect both their therapeutic and adverse effects (Weiner et al., 1991; Sackeim, 1994; Sackeim et al., 1994). In concept, seizure threshold is the smallest dose of electrical charge that can induce a seizure (Small et al., 1978). Several lines of evidence indicate that both the efficacy and the cognitive side effects of ECT may depend on the extent to which the stimulus intensity exceeds the patientís seizure threshold (Sackeim et al., 1987a-c, 1991, 1993; Chanpattana et al., in press a). The results suggest that optimizing electrical stimulus intensity during ECT requires a determination of seizure threshold, and administering proper stimulus intensity is necessary to insure the therapeutic efficacy of ECT.

The American Psychiatric Association (APA) Task Force on ECT (2000) concludes that the empirical titration technique provides the most precise method for quantifying seizure threshold. They recommend the use of moderately suprathreshold stimulation (50-150% above seizure threshold, or 1.5-2.5 times threshold) in patients treated with bilateral ECT, and moderately-to-markedly suprathreshold stimulation (150-450% above threshold, or 2.5-5.5 times threshold) in patients treated with right unilateral ECT.

Unfortunately, there has been a dearth of studies in assessing seizure threshold during maintenance ECT as well as in patients with schizophrenia. At the present time, a number of studies of depressed patients report the threshold estimates over 6-8 sessions (Sackeim et al., 1987 a&b; Malitz et al., 1986; Coffey et al., 1995; Shapira et al., 1996), and only one study of schizophrenic patients examines up to 20 sessions (Chanpattana et al., in press b). There has never been a long-term study using the structured dose-titration method in assessing the threshold change during an ECT course. We report here seizure threshold changes assessed by the empirical titration technique in patients with schizophrenia during the courses of acute and maintenance ECT.



The study was approved by the local Ethics Committee of the Faculty of Medicine of Srinakharinwirot University and the National Review Board of Research Studies in Humans in Thailand.

Forty-one patients with acute psychotic exacerbations and with a history of prior responsiveness to ECT, who met the DSM-IV criteria for schizophrenia (APA, 1994) were referred for ECT because of failure to respond to neuroleptic treatment. Psychiatric diagnosis was based on the consensus of three psychiatrists and also had to concur with the patientsí medical records. Diagnosis in the medical records had to be consistent throughout the episodes of illness. Other inclusion criteria were a minimum pretreatment score of 37 on the Brief Psychiatric Rating Scale (BPRS, Overall & Gorham, 1962; 18 items, rated 0-6), and age 16-50 years. Patients were excluded if they had received treatment with depot neuroleptics or ECT during the previous 6 months, had psychotic disorders due to a general medical condition, alcohol or other substance abuse, serious medical illness, or were receiving medicines with known effects on seizure threshold (e.g., antiepileptics, benzodiazepines, b-blocker, theophylline). Consent was obtained from the patients and/or their guardians after complete description of the study.

The study consisted of two phases: Phase I, acute treatment, and Phase II, maintenance treatment (M-ECT) for one year. 


Psychotropic medicines prescribed prior to the study were discontinued at least 5 days before the first ECT treatment. Flupenthixol 12 mg/day was prescribed to each patient during the first week and increased up to 24 mg/day depending on tolerability, and was continued throughout the study. Benzhexol (4-15 mg/day) was used to control extrapyramidal symptoms, with dosage titrated on a clinical basis. The dosages of both medicines were kept constant after the eighth week. No other medicines were used.

ECT was administered three times per week. After atropine 0.4 mg intravenously, anesthesia was given with a minimal dosage of thiopental (2-4 mg/kg) and 0.5-1 mg/kg of succinylcholine. Patients received positive pressure ventilation from the administration of anesthetic agent until resumption of spontaneous respiration. The ECT instruments were a MECTA SR1 and Thymatron DGx; each patient was treated with the same instrument throughout the treatment course. Bitemporal bilateral electrode placement was used exclusively. The tourniquet method and two channels of prefrontal electroencephalogram (EEG) were used to assess seizure duration.

Operationally for study purposes we defined seizure threshold as the lowest stimulus charge that produced an adequate seizure, i.e., bilateral tonic-clonic motor activity that lasted at least 30 seconds together with EEG evidence of seizure. Initial seizure threshold was estimated by our dose-titration schedule (Table 1) at the first two treatments. The first stimulus at the first session was 10% of total charge. If this failed to elicit an adequate seizure, the stimulus charge was increased in increments of 10% steps. A maximum of four stimulations per session was allowed, with an interval of at least 40 seconds between each without giving additional thiopental. At the second treatment session for each patient, a stimulus dose 5% lower than that at the first session was given, as listed in Table 1. If an adequate seizure occurred, that dose was taken as the initial threshold; if not, the first sessionís stimulus dose was taken. The stimulus charge 10% above threshold was given at the subsequent treatment; thereafter, the stimulus dose was increased by a 10% step for a short seizure.

Seizure threshold was quantified at the seventh, fourteenth, and twentieth treatment sessions. Starting with the patientís prior threshold dose, if it resulted in a short seizure, a 50% increment from prior threshold to the present stimulus dose was used. If an adequate seizure was not obtained, a 75% increase in stimulus dose was administered. Should this increase still not produce an adequate seizure, the last stimulus dose was used and adapted as patientís threshold. Increments of stimulus charge were adjusted close to this protocol in patients who had modest threshold increases.



Table1. Dose-titration schedule for MECTA SR1 and Thymatron DGx



Thymatron DGx 

































































Extra level**









































































* Increase by one level (10% step) is recommended for use in dose titration at the first or subsequent treatments.


** The extra level is used at the second treatment session only.


Response Criteria for Entering Phase II

A 3-week stabilization period was used as a response criterion (Chanpattana, 1998, 1999; Chanpattana et al., 1999 a&b), the patients who passed this criterion were eligible to enter Phase II. Briefly, patients who showed clinical improvement (BPRS scores < 25), went on to pass a 3-week stabilization period in which these effects had to be sustained. The stabilization period had the following treatment schedule: 3 regular ECTs (3 treatments/week) in the first week, then once a week for the second and third weeks, during which BPRS scores were always < 25. The total number of ECTs was limited to 20 treatments. All patients in the study could pass this response criterion, and acute ECT treatments were terminated.

Maintenance ECT

All patients received a combination treatment with M-ECT and the same dosage of flupenthixol. The ECT treatment procedures were the same as in Phase I. The ECT schedule was fixed during the first six months, starting with 4 weekly, followed by 10 biweekly treatments. Then M-ECT was given every 2-4 weeks depending on the patientís clinical status over the previous 6 months. No additional ECT treatment was given outside of this schedule. Relapse was defined as a BPRS score of > 37 that persisted for 2 consecutive ratings, 3 days apart.

Seizure threshold was estimated at the first weekly treatment, then at the third, sixth, and ninth months, and after one year. The stimulus dose given was 10% lower than a prior threshold. If it resulted in a short seizure, a prior threshold dose was administered. If a short seizure was elicited, a 50% increase from prior threshold to the present stimulus dose was used. Then, the last stimulus dose was taken if an adequate seizure was not obtained. The stimulus charge 10% above threshold was given at the subsequent treatment; thereafter, the stimulus dose was increased by a 10% step for a short seizure.

Statistical Analyses

Seizure thresholds were analyzed after logarithmic transformation to improve the normality of the data distribution. For discontinuous data, c2 tests were used to test for significant differences among groups. When sample size was small, the Fisherís exact test was used. Differences between groups on single, continuous variables were evaluated with t tests. Paired t tests were used to assess the differences of thresholds between two estimations. Relations between continuous variables were examined with the Pearsonís product-moment correlation. The degree to which variables could predict seizure threshold was examined by a stepwise multiple regression analysis. Values are given as mean + SD. All significances are two-tailed. SPSS 9.05 (1996 SPSS Inc.) was used for all analyses.


Table 2 shows clinical characteristics of 41 patients who participated in this study. Twenty-seven patients received ECT with MECTA SR1 and 14 with Thymatron DGx. Table 3 presents seizure thresholds as a function of gender and ECT instruments, of both Phases I and II.


Table 2. Subject Characteristics


Mean + SD


Age (yr)

32.2 + 6.7



30 women, 11 men



 32 paranoid, 5 disorganized, 2 catatonic, 2 undifferentiated


Onset (yr)

21 + 4.8


Duration of illness (yr)

11.3 + 6


Duration of current episode (yr)

1.2 + 1.4


Prior failure of adequate neuroleptic trials

3.8 + 1.2


Mean CPZ equivalent dose (mg)

1,162 + 311


Prior failure of flupenthixol treatment



No. of psychiatric admissions

5.5 + 4.3


No. of ECT treatments

12.5 + 5


Dosage of flupenthixol (mg)

23 + 2.3


BPRS scores at entry

50.3 + 9.1


GAF scores at entry

32.2 + 5.1


MMSE scores at entry

27 + 3.2


Thiopental (mg)

136.7 + 25.8


Succinylcholine (mg)

23.7 + 5.5



Abbreviations: BPRS = Brief Psychiatric Rating Scale, GAF = Global Assessment of Functioning, MMSE = Mini-Mental State Exam, CPZ = chlorpromazine.        


Phase I

Initial seizure threshold was 82.7 + 36.6 millicoulombs (mC). There was a substantial variability in thresholds, ranging from 25.2 to 180 mC (7-fold). There was no difference between gender, t (39) = 1.02, N.S. (Table 3). Initial threshold estimated with the MECTA was higher than the Thymatron, t (39) = 3.02, p = 0.004. All patients had seizures at the first session, with an average of 1.7 + 0.7 stimulations (range: 1-3). Initial threshold was positively related with ECT instrument (Spearmanís r = 0.43, p = 0.005; Thymatron = 1, MECTA = 2) and thiopental dosage (r = 0.43, p = 0.005). Stepwise multiple regression analysis revealed that both the instrument (t = 3.39, p = 0.002) and thiopental [t = 3.34, p = 0.002; F (2,40) = 11.31, p < 0.0001] accounted for 37.3% of the variance, 62.7% remaining unexplained.

The average number of stimulations at the seventh, fourteenth, and twentieth sessions was 2.4 + 0.7 (1-4), 2.4 + 1.2 (1-4), and 1.4 + 1.1 (1-4), respectively. Seizure threshold quantified, at each patientís last estimation, with MECTA was higher than Thymatron (t = 3.43, df = 39, p = 0.001). All patients had a rise in seizure threshold at the end of Phase I, the magnitude of increase was 213 + 179% [range: 40-860%; t (1,40) = 13.4, p < 0.0001]. There were no differences in the threshold increase either between gender [t (39) = 1.05, N.S.] or instrument [t (39) = 1.7, N.S.]. Seizure-threshold increase was positively related to number of ECT treatments (r = 0.46, p = 0.003) and onset of illness (r = 0.43, p = 0.005), and negatively related to succinylcholine dosage (r = 0.35, p = 0.027). Stepwise multiple regression analysis revealed that number of treatments (t = 2.43, p = 0.02) and onset of illness [t = 2.12, p = 0.04; F (2,40) = 7.95, p = 0.001] explained 29.5% of the variance. There was a substantial reduction in seizure duration over an ECT course [motor: t (1,40) = 4.23, p < 0.0001; EEG: t (1,40) = 4.26, p < 0.0001].



Table 3. Seizure Threshold as a Function of Gender and ECT Devices*




ECT devices






Thymatron DGx


(n = 30)

 (n = 11)

(n = 27)

(n = 14)

Phase I





Initial threshold

80.7 + 40

88.1 + 25.6

92 + 34.3

 64.8 + 35.2 a

Last estimates

254.5 + 165.5

227.1 + 104.3

291.3 + 150.2

162 + 113.9 b

% Threshold increase

263.5 + 170.3

168.9 + 119.8

275.2 + 181.4

185.4 + 114.6

Phase II





First treatment

231.2 + 193.9

165 + 127.2

241.4 + 204

159.5 + 104.9 c

Third month

291 + 171.4

216.7 + 88.4

314.5 + 160

187.2 + 110.8 d

Sixth month

305.2 + 166.3

216.7 + 88.4

322.8 + 156.9

201.6 + 113.6 e

Ninth month

300.4 + 158.5

216.7 + 88.4

315.6 + 150.1

205.2 + 112.6 f

One year

284.6 + 163.2

212.1 + 90.1

303.6 + 153.8

190.8 + 112.1 g

% Threshold increase

22.9 + 48.1

1.3 + 18.7 h

6.8 + 30.4

37 + 56.8

% Overall increases

281.4 + 191.9

138.6 + 62.7 I

244.5 + 175.5

240.5 + 190.1

* Values are given in mean + SD, in millicoulombs.

a t = 3.02, df = 39, p = 0.004; b t = 3.43, df = 39, p = 0.001; c t = 3.74, df = 39, p = 0.001, d t = 2.79, df = 39, p = 0.008; e t = 2.63, df = 39, p = 0.012; f t = 2.59, df = 39, p = 0.013  g t = 2.71, df = 39, p = 0.01; h t = 2.07, df = 39, p = 0.045;  i t = 3.59, df = 39, p = 0.001


Phase II

All patients received ECT combined with flupenthixol, using a fixed treatment schedule during the first 6 months. Thereafter, 30 patients continued to receive biweekly treatment, 7 had ECT every 3 weeks, and monthly ECT was scheduled to 4 patients over the last 6 months. No patients suffered relapse at the end of Phase II.

Seizure thresholds of patients receiving MECTA were consistently higher than Thymatron, in all estimations (Table 3). The average number of stimulations was 1.9 + 0.4 (1-3), 2.2 + 0.6 (1-3), 2.2 + 0.5 (1-3), 1.9 + 0.5 (1-3), and 1.8 + 0.5 (1-3), respectively. There was a trend towards a difference in the threshold increase between instruments, t (39) = 1.86, p = 0.08. Women had larger increments in threshold than men, t (39) = 2.07, p = 0.045. Fifteen patients had no change in thresholds, 4 of whom had their thresholds at the maximum charge of the instruments; 18 had a modest threshold increase, and 8 patients had a gradual decrease in seizure thresholds. The magnitude of threshold increase of Phase II was 17 + 43% [range: 60% decrease-150% increase, t (1,40) = 1.52, p = 0.14], which only had an inverse correlation with thiopental dosage (r = 0.46, p = 0.004). Threshold increase during the first six months (21 + 41%, range: 50% decrease-150% increase) was larger than that of the previous six months [-0.7 + 31%, range: 67% decrease-100% increase; t (1,40) = 2.43, p = 0.02]. There was a further reduction in seizure duration over Phase II [motor: t (1,40) = 3.85, p < 0.0001; EEG: t (1,40) = 4.3, p < 0.0001].

In the total sample, one-way analyses of variance (ANOVAs) were conducted on the demographic, clinical, and treatment variables, with threshold-change group (i.e., threshold-increased, threshold-stable, and threshold-decreased groups) as a between-subject factor. Significant main effects of the threshold-change group were followed by Scheffe post comparisons of the three groups to identify pair-wise differences. Table 4 presents all variables among the three threshold-change groups that had statistically significant differences. There were significant main effects of the three groups for illness duration [F (2,40) = 3.48, p = 0.041], episode duration [F (2,40) = 3.69, p = 0.034], number of ECTs in Phase I [F (2,40) = 5.85, p = 0.006], and threshold increase of Phase I [F (2,40) = 4.93, p = 0.013]. Post hoc comparisons indicated that the threshold-increased group received fewer ECTs than the two others (pís = 0.02 and 0.035), and had less increment of thresholds in Phase I than the threshold-decreased group (p = 0.014).



Table 4. Clinical characteristics of patients as a function of the threshold-change group*



Threshold-change Groups






     (n = 18)

    (n = 15)

     (n = 8)

Illness duration (yr)

  13.9 + 6.9 (4-25)

 9.3 + 4.4 (3-19)

  9.1 + 4.2 (3-15)

Episode duration (yr)

   1.9 + 1.5 (.25-5)

 0.8 + 1 (.08-4)

  0.7 + 1 (.08-3)

Number of ECTs in Phase I


   9.8 + 2.8 (7-16)


14.5 + 5.6 (7-23)


   15 + 5.2 (9-23)

Threshold increase of Phase I (%)


156 + 108 (40-380)


197 + 175 (50-586)


373 + 238 (100-860)


* Values are expressed as mean + SD.

  Only clinical variables having statistical significances are presented




The overall threshold increase from the beginning of Phase I to the end of Phase II was 243 + 178% [range: 33.3% decrease-700% increase; t (1,40) = 13.75, p < 0.0001]. Interestingly, there was one patient who had a threshold estimate at the end of Phase II (50.4 mC) lower than her initial threshold at Phase I entry (75.6 mC). There was no difference in threshold increase between instruments, t (39) = 0.07, N.S. Women had more threshold increase compared to men, t (39) = 3.59, p = 0.001. An overall threshold increase was negatively related to gender (Spearmanís r = 0.39, p = 0.013; women = 0, men = 1) and initial threshold (r = 0.35, p = 0.027). Stepwise multiple regression analysis revealed only gender [t = 2.4, F (1,39) = 5.78, p = 0.021) represented 12.9% of the variance. There was a marked reduction in seizure duration over the ECT course [motor: t (1,40) = 7.02, p < 0.0001; EEG: t (1,40) = 7.46, p < 0.0001].


In Phase I, all patients had a rise in seizure threshold; the magnitude of increase

was 213 + 179%. During Phase II, 15 patients had no change in thresholds compared to the first estimates of Phase II; 18 showed a further increase, and the thresholds of the last 8 patients decreased gradually. The magnitude of threshold increase of Phase II was 17 + 43%. The overall increase of thresholds at the end of Phase II was 243 + 178%. The present study demonstrates a substantive increase in threshold over acute ECT, which appears to reach a plateau during maintenance ECT. Therefore, regular estimation of seizure threshold of each patient during ECT is necessary to ensure that the proper stimulus dose is used. This is the first study examining seizure-threshold changes during both acute and maintenance ECT treatments in remitted patients with schizophrenia.

Overestimation of seizure threshold is of critical concern. In order to avoid using too weak a stimulus intensity in treating patients with treatment-refractory schizophrenia, a criterion for seizure adequacy longer than the usual recommendations was set [20-25s (APA, 1990); 15s of motor, and/or 25s of EEG (Royal College of Psychiatrists, 1995)]. Our criterion might affect the results of both Phases. Nonetheless, initial threshold was quantified at the first two treatments instead of once at the first session as used in other studies, in order to obtain a more accurate estimate. In addition, a conservative dose-titration schedule was used in all subsequent threshold estimations. Therefore, these particular dose-titration strategies might be a methodological strength of our study.

Restriction of concomitant pharmacotherapy represents a further strength of our study. Nonetheless, the question remains as to whether flupenthixol might have any effects on seizure threshold. Unfortunately, there has been only one study in the literature pertaining to this issue. Chanpattana et al (in press b) estimated initial seizure threshold and its changes by means of the empirical titration technique in 93 patients with schizophrenia receiving ECT combined with flupenthixol. They could not find such effects, since there were no correlations between the thresholds and flupenthixol dosages in all 4 assessments.

Seizure thresholds of patients quantified with MECTA were always higher than with Thymatron. There are three likely reasons for this. First, there was a different gender ratio of patients receiving ECT with each instrument. There were more men with MECTA (9 men, 18 women) than Thymatron (2 men, 12 women), F = 0.013. Seizure threshold is known to be higher in men than women (Sackeim et al., 1991; Coffey et al., 1995; Shapira et al., 1996). Second, this dose-titration schedule provided a uniform increment of stimulus dose (10% step, Table 1), which referred to the maximum charges of each instrument (576 mC of MECTA and 504 mC of Thymatron). Thus, the stimulus charge of MECTA was always higher than Thymatron in all levels. And, third, Chanpattana et al (in press c) conducted a prospective, randomized controlled study on 88 patients with schizophrenia and schizoaffective disorders, comparing initial seizure threshold estimated by the empirical titration technique with MECTA SR1 and Thymatron DGx instruments. The measured seizure thresholds were found to be higher with the MECTA than the Thymatron instrument, 61% on average. Underlying the differences between the two instruments are systematic differences in stimulus characteristics, and the greater efficiency associated with stimuli of lower charge rate (Swartz, 1994), lower pulse width (Swartz et al., 1997), lower pulse frequency (Devanand et al., 1998), and longer train duration (Swartz & Larson, 1989).                                        

Women had higher thresholds than men over both phases (Table 3). This might be an artifact from a small number of patients in the study, since there was no significant difference of age between women and men [32 + 7 vs. 32.9 + 5.8 years, respectively; t (39) = 0.4, N.S.]. The relationship between onset of illness and threshold increase presumably follows the correlation between onset of illness and age (r = 0.48, p = 0.001). 

The number of ECT treatments was the most significant predictor of threshold increase of Phase I. This finding is similar to our prior study (Chanpattana et al, in press b). The result may be explained on the basis of a decrease in neural metabolic activity that reflects potentiation of the endogenous inhibitory processes following ECT-induced seizure (Prohovnik et al., 1986; Sackeim et al., 1986). The findings from these studies might also explain the results of a progressive decrease in seizure duration over both acute and maintenance treatments.      

There was one patient who had a threshold estimate at the end of Phase II (50.4 mC) lower than her initial seizure threshold (75.6 mC). This patient received prior treatment with psychotropic agents possessing anticonvulsant properties (i.e., diazepam 10 mg hs and propanolol 30 mg/day for agitation), which might explain this finding (Kellner et al., 1997).

Interestingly, the threshold-increased group received a fewer number of ECTs than the two others, despite having longer durations of illness and current episode. The longer durations of illness and current episode are known to indicate poor responsiveness to both ECT and pharmacotherapy (Chanpattana et al., 1999 a&b; Kalinowsky & Worthing, 1943; WHO, 1979). The results may be explained by the quality of responsiveness to prior ECT of the threshold-increased patients. Twelve of 18 patients in the threshold-increased group had good responses to prior M-ECT, compared to 6 of 15 and 3 of 8 patients in the threshold-stable and threshold-decreased groups (Fís = 0.04), respectively.

This study provides some of the first information on seizure threshold and its change with ECT among patients with schizophrenia over both acute and maintenance treatments. The magnitude of threshold increase was large during an index course, then appeared to reach a plateau over maintenance treatment. Our findings emphasize the recommendation that seizure threshold should be estimated regularly in each patient during the treatment course, to justify the proper stimulus intensity and optimize the ECT efficacy. In addition, this study also suggests another important question for future research: How long will it take for a seizure threshold to return to its baseline?



This study was supported by the Thailand Research Fund, grant BRG 3980009. The author thanks Wiwat Yatapootanon, M.D., Yaowalak Prasertsuk, B.Sc, M.S., and W.V. McCall, M.D., M.S. for their technical support.



American Psychiatric Association. Diagnostic and Statistical Manual of Mental Disorders, 4th ed. Washington, DC: American Psychiatric Press. 1994, 10.

American Psychiatric Association Task Force Report on ECT: The practice of ECT: Recommendations for treatment, training, and privileging. Washington, DC: American Psychiatric Press, 1990.

American Psychiatric Association Task Force Report on ECT: The practice of ECT: Recommendations for treatment, training, and privileging. Washington, DC: American Psychiatric Press, 2000 (in press).

Chanpattana W. Maintenance ECT in schizophrenia: A pilot study. J Med Assoc Thai 1998; 81: 17-24.

Chanpattana W. The use of stabilization period in ECT research in schizophrenia: I. A pilot study. J Med Assoc Thai 1999; 82: 1193-9.

Chanpattana W, Chakrabhand S, Kongsakon R, Techakasem P, Buppanharun W . Short-term effect of combined ECT and neuroleptic therapy in treatment-resistant schizophrenia. J ECT 1999; 15: 129-139.

Chanpattana W, Chakrabhand S, Sackeim HA et al. Continuation ECT in treatment-resistant schizophrenia: A controlled study. J ECT 1999; 15: 178-192.

Chanpattana W, Chakrabhand S, Buppanharun W, Sackeim HA. Effects of stimulus intensity on the efficacy of bilateral ECT in schizophrenia: A preliminary study. Biol Psychiatry (in press a).

Chanpattana W, Chakrabhand S, Buppanharun W, Raksakietisak S. Seizure threshold rise during ECT in schizophrenic patients. Psychiatry Res (in press b).

Chanpattana W, Buppanharun W, Chakrabhand S. Seizure threshold in electroconvulsive therapy: Differences between instruments. J ECT (in press c).

Coffey CE, Lucke J, Weiner RD et al. Seizure threshold in electroconvulsive therapy: II. The anticonvulsant effect of ECT. Biol Psychiatry 1995; 37: 777-88.

Devanand DP, Lisanby SH, Nobler MS, Sackeim HA. The relative efficiency of altering pulse frequency or train duration when determining seizure threshold. J ECT 1998; 14: 227-35.

Fraser RM. ECT: A clinical guide. New York: John Wiley & Sons, 1982: 57.

Kalinowsky LB, Worthing HJ. Results with electroconvulsive therapy in 200 cases of schizophrenia. Psychiatric Quarterly 1943; 17: 144-53.

Kellner CH, Pritchett JT, Beale MD, Coffey CE. Handbook of ECT. Washington, DC: American Psychiatric Press, 1997: 26-33.

Malitz S, Sackeim HA, Decina P et al. The efficacy of electroconvulsive therapy: Dose-response interactions with modality. Ann NY Acad Sci 1986; 462: 56-64.

NIH Consensus Conference: Electroconvulsive therapy. JAMA 1985; 254: 2103-8.

Ottosson JO. Experimental studies of the mode of action of electroconvulsive therapy. Acta Psychiatr Scand 1960; 1-141 (suppl).

Ottosson JO. Effect of lidocaine on the seizure discharge in electroconvulsive therapy. Acta Psychiatr Scand 1960; 145: 7-32.

Ottosson JO. Electroconvulsive therapy of endogenous depression: An analysis of the influence of various factors on the efficacy of the therapy. J Ment Sci 1962; 108: 694-703.

Overall JF, Gorham DR. The Brief Psychiatric Rating Scale. Psychol Rep 1962; 10: 799-812.

Potter WZ. ECT methodologic issues. Psychopharm Bull 1994; 30: 455-9.

Prohovnik I, Sackeim HA, Decina P, Malitz S. Acute reductions of regional cerebral blood flow following electroconvulsive therapy: Interactions with modality and time. Ann NY Acad Sci 1986; 462: 249-262.

Royal College of Psychiatrists. The ECT handbook: The second report of The Royal College of Psychiatristsí Special Committee on ECT. London: Gaskell, 1995: 62.

Sackeim HA, Decina P, Prohovhik I, Malitz S, Kanzler M. Dosage, seizure threshold, and the antidepressant efficacy of electroconvulsive therapy. Ann NY Acad Sci 1986; 462: 398-410.

Sackeim HA, Decina P, Kanzler M et al. Effects of electrode placement on the efficacy of titrated, low-dose ECT. Am J Psychiatry 1987a; 144: 1449-55.

Sackeim HA, Decina P, Prohovhik I, Malitz S. Seizure threshold in electroconvulsive therapy: Effects of sex, age, electrode placement, and number of treatments. Arch Gen Psychiatry 1987b; 44: 355-60.

Sackeim HA, Decina P, Portnoy S et al. Studies of dosage, seizure threshold, and seizure duration in ECT. Biol Psychiatry 1987c; 22: 249-68.

Sackeim HA, Devanand DP, Prudic J: Stimulus intensity, seizure threshold, and seizure duration: Impact on the efficacy and safety of electroconvulsive therapy. Psychiatr Clin North Am 1991; 14: 803-43.

Sackeim HA, Prudic J, Devanand DP et al. Effects of stimulus intensity and electrode placement on the efficacy and cognitive side effects of electroconvulsive therapy. N Engl J Med 1993; 328: 839-46.

Sackeim HA, Long J, Luber B et al. Physical properties and quantification of the ECT stimulus: I. Basic principles. Convulsive Ther 1994; 10: 93-123.

Sackeim HA. Physical properties of the ECT stimulus. Convulsive Ther 1994; 10: 140-152.

Shapira B, Lidsky D, Gorfine M, Lerer B. Electroconvulsive therapy and resistant depression: Clinical implications of seizure threshold. J Clin Psychiatry 1996; 57: 32-8.

Small JG, Small IF, Milstein V. Electrophysiology in ECT. In: Lipton MA, DiMascio A, Killam KF (eds.). Psychopharmacology: A generation of progress. New York: Raven Press, 1978: 759-769.

Swartz CM, Larson G. ECT stimulus duration and its efficacy. Ann Clin Psychiatry 1989; 1: 147-52.

Swartz CM. Optimizing the ECT stimulus. Convulsive Ther 1994; 10: 132-4.

Swartz CM, Manly DT. ECT pulsewidth 0.5 millisecond is more efficient than 1.0 milliseconds stimuli [abstract]. In: Proceedings of the 149th APA Annual Meeting, San Diego, 1997; New Research Abstract No. 237, p. 132.

Weiner RD, Coffey CE, Krystal AD. The monitoring and management of electrically induced seizures. Psychiatr Clin North Am 1991; 14: 845-869.

Weiner RD. Treatment optimization with ECT. Psychopharm Bull 1994; 30: 313-20.

World Health Organization. Schizophrenia: an international follow-up study. Chicester: John Wiley & Sons, 1979.