EREDETIKÖZLEMÉNY

COMPARISON OF PROPOFOL AND ETOMIDATE

REGARDING IMPACT ON SEIZURE THRESHOLD
DURING ELECTROCONVULSIVE THERAPY IN
PATIENTS WITH SCHIZOPHRENIA
Gazdag Gábor,1 Tolna Judit,2 Iványi Zsolt3
1
Szent László Kórház, Addiktológiai és Pszichiátriai Ambulancia, Budapest,
2
Semmelweis Egyetem Pszichiátriai és Pszichoterápiás Klinika, Budapest,
3
Semmelweis Egyetem Aneszteziológiai és Intenzív Terápiás Tanszék, Budapest

APROPOFOLÉSAZETOMIDATE restimulációk számában nem találtunk a két szer

GÖRCSKÜSZÖBREKIFEJTETTHATÁSÁNAK között különbséget.
ÖSSZEHASONLÍTÁSASZKIZOFRÉNBETEGEK Megbeszélés. Vizsgálatunkban az etomidate-hoz
ELEKTROKONVULZÍVKEZELÉSESORÁN képest a propofol szignifikáns görcsrohamhossz
Háttér. Az elektrokonvulzív kezelés (ECT) an- csökkentõ hatását mutattuk ki szkizofrén bete-
esztéziájában használt szerek közül a propofol gek ECT kezelése során. Ugyanakkor sem a két
jelentõsebb mértékben csökkenti a görcsroham szer görcsküszöbre kifejtett hatásában, sem a
hosszát, mint az etomidate. Szkizofrén betege- szükséges restimulációk számában nem talál-
ken végzett korábbi vizsgálatunkban ugyanak- tunk szignifikáns különbséget. Eredményeink
kor a görcsroham megrövidülése miatt propofol alapján a propofol ismert görcsgátló hatása elle-
használata mellett sem volt szükség gyakoribb nére az ECT kezelés során a görcsküszöböt az
restimulációra. Ennek alapján azt feltételeztük, etomidate-hoz képest nem emeli meg, használata
hogy a propofol és az etomidate görcstevékeny- tehát nem okoz a betegek számára fokozott áram-
ségre kifejtett hatása 20 másodpercnél rövidebb terhelést.
görcsroham esetén azonos. Jelen vizsgálatunk- KULCSSZAVAK: propofol, etomidate, elektro-
ban az etomidate-ot és a propofolt hasonlítjuk konvulzív kezelés, szkizofrénia, görcsküszöb
össze, szkizofrén betegek elektrokonvulzív ke-
zelése során a görcsküszöbre és a görcsroham SUMMARY
hosszára kifejtett hatásuk alapján. Background. While propofol is known to shorten
Módszer. A randomizált, keresztezett elrendezé- seizures during electroconvulsive therapy, in our
sû vizsgálatban 30 szkizofrén beteg vett részt. Az previous study on patients with schizophrenia,
egymást követõ ülések során 1 mg/kg propofol there was no need for more frequent restimul-
és 0,2 mg/kg etomidate felváltva került alkalma- ations when using propofol compared with eto-
zásra. A görcsküszöböt mindkét altatószernél a midate. We hypothesized that etomidate and pro-
görcstevékenység kialakulásához szükséges sti- pofol have similar effects on seizure activity in
mulus dózisának titrálásával határoztuk meg. A cases where seizure duration is shorter than 20
görcsrohamok hosszát az EEG-vel és az EMG-vel seconds. In this study, etomidate and propofol
regisztrált értékek alapján hasonlítottuk össze. are compared regarding their impact on seizure
Eredmények. Etomidate-tal történt altatás esetén threshold and seizure duration.
az EEG-vel (49,6±23,1 s, 39,7±19 s p=0,026), és Method. 30 schizophrenic patients participated
az EMG-vel (41,4±22 s, 32,8±17,6 s p=0,016) in this prospective randomized cross-over study.
regisztrált görcsroham is hosszabbnak bizonyult, For anesthetic induction were 1 mg/kg of pro-
mint propofol alkalmazásakor. Ugyanakkor sem pofol or 0.2 mg/kg of etomidate used alternately.
a görcsroham kiváltásához szükséges minimális For both anesthetics, seizure threshold was de-
stimulus intenzitás tekintetében (41,58mC±13,6 termined by titrating the dose of the stimulus
mC, 41,58±11,1 mC, p=1,0), sem a szükséges necessary for eliciting a seizure. Seizure dura-

Neuropsychopharmacologia Hungarica 2007, IX/3; 125-130 125

EREDETIKÖZLEMÉNY GAZDAGGÁBOR,TOLNAJUDIT,IVÁNYIZSOLT

tions were also compared. Conclusion. During the ECT of patients with
Results. After etomidate induction, seizure dura- schizophrenia, propofol was shown to possess
tions registered either by EEG or by EMG were significant seizure-shortening properties, but it
longer than propofol treated cases (EEG: 49.6± does not elevate seizure threshold or drop seizure
23.1 s, versus 39.7±19 s, p=0.026; EMG: 41.4± duration under the minimal threshold more fre-
22 s, versus 32.8±17.6 s, p=0.016). However, no quently than etomidate does. Based on these
significant differences were found for minimum findings, we conclude that the use of propofol
seizure eliciting stimulation energy or the num- does not result in a greater electric load on the pa-
ber of restimulations between the two anesthetics tients than etomidate.
(41.58 mC±13.6 mC, versus 41.58 mC±11.1 mC, KEYWORDS: propofol, etomidate, electroconvul-
p=1.00). sive therapy, schizophrenia, seizure threshold

INTRODUCTION the minimal 20 s and may affects only seizure du-

ration and not threshold.
Reports published on the impact of propofol on
seizure activity have been controversial. Several
Aims
case reports describe seizures during anesthesia
with propofol (1-4) and epileptiform patterns in- To compare, during the ECT of patients with schi-
duced by propofol appearing on EEG (5). How- zophrenia, the effect of propofol on seizure dura-
ever, another publication reported that propofol tion and seizure threshold with that of etomidate.
did not significantly affect the EEG of those pa- Hypothesis: Although propofol reduces seizure
tients who have partial epileptic seizures (6). In duration during ECT treatment, does not elevate
the literature there is also a report showing that, seizure threshold comparing with etomidate.
under the effect of propofol, spikes on the EEG of
epileptic patients become scarcer (7). Tests on an- PATIENTS AND METHODS
imals proved propofol to have seizure-threshold- Patients
elevating properties dependent on dose (8-10).
Between October 2003 and February 2005, at the
The controversy can partially be resolved by the
Department of Psychiatry and Psychotherapy of
results of a study in which it was found that low
Semmelweis University, 68 patients were treated
doses of propofol tend to have epileptic, while
with ECT. The indication for treatment was in 16
high doses tend to have antiepileptic effects (5).
cases depression and in 52 cases schizophrenia.
Propofol, when compared to methohexital, short-
Patients were diagnosed on the basis of the opin-
ens seizure duration during ECT has been reported
ion of the treating physician, and that of an investi-
by numerous studies, but significant differences
gator (JT), who made her own diagnoses, after re-
have not been found in clinical efficacy (11-14),
viewing the clinical reports retrospectively, inde-
or various electrophysiological and vital parame-
pendently. (Interrater reliability was 95%). The
ters (15). During the ECT increasing doses of
three cases where the individual diagnoses dif-
propofol resulted in shortened seizure duration
fered were reevaluated by another investigator
compared to etomidate (16). In another study, the
(GG) and the diagnoses were made after identical
switch from propofol to etomidate resulted in a
opinions had been reached.
significant prolongation of the seizure duration
Involved in this prospective, randomized, cross-
(17). In a randomized cross-over study during the
over study were patients over 18 for whom the in-
ECT of patients with schizophrenia, seizure dura-
dication of ECT was a form of schizophrenia.
tions registered both by EEG and EMG were sig-
Each subject or guardian gave informed consent
nificantly shorter with propofol than with eto-
for ECT and study participation.
midate. Meanwhile, in the number of restimul-
Seven patients did not consent, in three cases,
ations necessitated by seizures shorter than 20 s
the treatment, on the basis of the decision of the
there was no difference between the two drugs
treating physician, was discontinued after the first
(18). The explanation for this result could be that
session and in three cases no epileptic activity
propofol drops the duration of the seizure under
were reached during one of the first two sessions.

126 Neuropsychopharmacologia Hungarica 2007, IX/3; 125-130

APROPOFOLÉSAZETOMIDATEGÖRCSKÜSZÖBREKIFEJTETTHATÁSÁNAKÖSSZEHASONLÍTÁSA EREDETIKÖZLEMÉNY

In one case, the organic origin of the psychosis stimulation was increased until the development
was revealed subsequently, and in two cases the of seizure activity was shown by clinical signs and
use of the anesthetics was not done in accordance on EEG-registration unambiguously. During the
with the study protocol. The seven schizoaffective seizure both EEG and the EMG of the isolated up-
patients were also excluded. Eventually, 30 pa- per limb were monitored and recorded. Only those
tients were involved in the examination. The study cases were enrolled in which the stimulation was
was approved by the Institutional Review Board successful during the first two sessions. When
of Semmelweis University. stimulation with one of the anesthetics resulted in
seizure activity, in the next session the other one
ECT Procedure was used for sleep induction and the titration pro-
ECT was performed twice a week (Tuesday and cedure was restarted at a dose of 25.2 mC.
Friday) early in the morning. The patients had 6
hour fastening period before the therapy. After en- Concomitant Medication
tering, a blood pressure cuff was applied to the left Concomitant antipsychotic medication was re-
upper limb and the blood pressure was measured. ceived by all except one of the patients. The drugs
EEG electrodes were then placed frontally and used were the following: haloperidol, zuclopen-
above the right mastoid. A pulsoxymetric sensor thixol, clozapine, risperidone, olanzapine, queti-
was fixed on the right. After registering the base- apine and amisulpiride. Concomitant sedative
line blood pressures and heart frequency, 0.5 mg therapy – involving clonazepam and diazepam –
atropine was administered intravenously, and the was received by 18 patients. Concomitant antide-
patient was preoxygenated using 6 l/min O2 via a pressant therapy using sertraline was applied in
facemask. Two minutes after atropine administra- one case. Mood stabilizers were applied in 4 cases,
tion, 10 mg lidocain at a concentration of 1% was with lithium used in 3 cases, and carbamazepine in
given, in order to prevent the venous irritation one case.
caused by either propofol or etomidate. As an in- During the two days preceding the ECT and
duction agent for anaesthesia, each patient re- during the treatment, there were no changes made
ceived 1 mg/kg propofol or 0,2 mg etomidate in a in the concomitant medication, either in doses or
randomised order, using a cross over study design. in the drugs applied.
The appropriate induction agent was injected over
5 sec. After loss of responsiveness to verbal com- Measurements and Calculations
mand, the blood pressure cuff was inflated 50 Seizure threshold was defined as the lowest dose
mmHg above the systolic blood pressure to isolate of stimulation which could elicit seizure activity
the circulation to the upper limb and ensure accu- proven both clinically and by EEG.
rate registration of the motor seizure. Succinyl- The lengths of EEG- and EMG-registered seizure
choline 50 mg was then given in order to avoid activities were determined and documented by
convulsion induced musculoskeletal injuries. the computer of the Thymatron DGx device.
Ventilation was assisted using face mask, AMBU These data were validated and corrected when
balloon and 4 l/min oxygen, until the electrical necessary by the investigators.
stimuli was performed. Just before the delivery of
stimulus the ventilation was stopped. An electrical Statistical Analyses
stimulus was delivered 1 min after succinyl-
Sample size was determined on the basis of our pi-
choline administration via bifrontotemporal elec-
lot study using the PASS statistical software (Hin-
trodes, using a Thymatron DGx (Somatics, Lake
tze J. 2001; NCSS and PASS. Number Cruncher
Bluff, IL, USA) instrument, providing bidirec-
Statistical Systems. Kaysville, Utah, USA). In or-
tional 1 ms square impulse by constant (0,9 A)
der to prove a difference in stimulus dose of 25.2
current . Dose titration was started with the small-
mC, with a standard deviation of 50,4 mC, to
est dose available (25.2 mC), and was increased in
reach a power of 0.8 at a significance level of 0.05,
each subsequent stimulation by 25.2 mC. In each
using a one-sample, two-sided nonparametric t-test
session, up to four stimulations were performed
were 29 patients required.
using gradually increased doses, with a 60-second
Data were analyzed using an SPSS 10.0 pack-
pause between the stimulations, without the ad-
age (Statistical Package for Social Sciences, Chi-
ministration of additional anesthetic. The dose of
cago, IL, USA). Descriptive data were presented

Neuropsychopharmacologia Hungarica 2007, IX/3; 125-130 127

EREDETIKÖZLEMÉNY GAZDAGGÁBOR,TOLNAJUDIT,IVÁNYIZSOLT

through their means, confidence intervals and able to achieve reliable anesthesia with 0.75 mg/kg
standard deviations. The normality of the distribu- propofol or 0.15 mg/kg etomidate, but we man-
tion was examined by calculating skewness and aged to trigger a satisfactory hypnosis using 1.0
kurtosis. If their distribution was normal, the mg/kg propofol and 0.2 mg/kg etomidate, which
paired data were compared with t-tests and in case supports the assumption of Avramov (16) that
of non-normal distribution with Mann-Whitney these are potentially equihypnotic doses. Ineffec-
U-tests. As the values in some cells were less than tiveness of low doses can be explained by the dif-
five, binomial data were evaluated with a Mc- ference between the two samples (Avramov had
Nemar test. depressed patients in his group, we had schizo-
phrenics in ours), or by the interaction in our study
RESULTS between concomitant psychopharmacological
medication and anesthetics.
In the study, there were 14 male and 16 female pa-
In this study in accordance with our previous
tients enrolled. Their mean age was 36.9 years
results (18), propofol, compared to etomidate, re-
(CI:32.0-41.9; SD:13.3). The underlying indica-
duced both the electric (p=0.026) and the motoric
tions for ECT were the following: schizophrenia
(p=0,016) seizure duration to a significant degree.
paranoid type (295.0): 3 patients, schizophrenia
In spite of this difference, during the analysis of
disorganized type (295.1): 8 patients, schizophre-
the data of stimulations exceeding the initial sei-
nia catatonic type (295.2): 19 patients.
zure threshold, neither the occurrence rate of sei-
In the sessions, that were analyzed in pairs, the
zures shorter than 20 sec, nor that of seizures
first anesthesia was accomplished using etomidate
shorter than 25 sec showed a significant difference
in 17 cases and propofol in 13 cases (p=0.58).
between the two drugs. Also, no difference could
Electroencephalographic (EEG) and motor
be demonstrated for seizure threshold for either
(EMG) mean seizure durations and seizure thresh-
anesthetic.
old values are shown in Table 1.
Based on all these findings, we conclude that
On the basis of the 20-second minimal seizure du-
the antiepileptic properties of propofol are mani-
ration recommended by Swartz (19), EEG seizure
fested only in a reduction in seizure duration,
duration, during the first stimulation to exceed sei-
without an elevation of seizure threshold. The fact
zure threshold, was shorter than necessary in case
that we were unable to prove any antiepileptic
of propofol in 3 instances and in case of etomidate
impact on seizure threshold can be explained with
in 2 instances (p=1.00). If minimal seizure dura-
the dose dependency of the antiepileptic proper-
tion is defined as 25 sec, EEG seizure duration
ties of propofol (5): in this study, we utilized low
was shorter in 3 cases with propofol and in 4 cases
doses of propofol (1 mg/kg). In another study,
with etomidate (p=1.00).
epileptic seizure during anesthesia with propofol
DISCUSSION was explained with the changing of the cerebral
concentration of the drug (4). This hypothesis can
In our study, propofol and etomidate were com- also partially explain our findings. In order to
pared regarding their impact on seizure threshold exactly describe the effects of propofol on seizure
and seizure duration in the ECT of patients with activity and the mechanism of this action are
schizophrenia. In our pilot study we tried to deter- further investigations required.
mine the lowest dose, which produces a satisfacto-
rily deep sleep for ECT treatment. We were not
Table 1
Seizure thresholds, EEG and EMG measured seizure durations after propofol and etomidate
induction (Data are presented with mean, confidence interval and standard deviation)
Seizure threshold (mC) EEG (sec) EMG (sec)
Propofol 41.58 (CI: 37.5-45.6; SD:13.6) 39.7 (CI:32.6-46.8; SD:19.0) 32.8 (CI:26.2-39.4; SD:17.6)
Etomidate 41.58 (CI: 36.46-46.7; SD:11.1) 49.6 (CI:41.0-58.2; SD:23.1) 41.4 (CI:33.2-49.6; SD:22.0)
Statistics Z=0; p=1.00 T=-2.23; p=0.026 T=-2.40; p=0.016

128 Neuropsychopharmacologia Hungarica 2007, IX/3; 125-130

APROPOFOLÉSAZETOMIDATEGÖRCSKÜSZÖBREKIFEJTETTHATÁSÁNAKÖSSZEHASONLÍTÁSA EREDETIKÖZLEMÉNY

The value of our study is reduced by the phar- and 0.2 mg/kg of etomidate can be considered
macotherapy concomitant with the seizure treat- equianesthetic doses. As it had earlier been pub-
ment. This was necessary because the patients lished (22), there is no reliable tool for measuring
subjected to ECT were almost exclusively show- the depth of anesthesia, especially when two dif-
ing severe symptoms, were resistant even to ferent hypnotics are compared.
combined psychopharmacotherapy; their medical
treatment was for ethical reasons indispensable. CONCLUSION
Therefore, the various concomitant medications Although propofol, reduced both the electric
might interact with propofol in a different way (p=0.026) and the motoric (p=0,016) seizure dura-
than with etomidate could not be ruled out. The tion to a significant degree in our study during the
antiepileptic properties of the low doses of lido- ECT of patients with schizophrenia did not elevate
caine (20), administered to prevent the local irrita- seizure threshold, and did not resulted more abor-
tion of the vein walls by the anesthetics, might tive seizures compared to etomidate. Based on
also have affected our results. Furthermore, lido- these findings, we conclude that the use of
caine might interact with the two anesthetics dif- propofol does not result in a greater electric load
ferently. on the patients than etomidate, and can be a safe
A fixed dose of induction agents may have alternative drug for the anesthesia of ECT.
different effects on people of different ages (21).
In our study, the age range of the participants was Corresponding author:
19-63 years. As no recommendation for the age- Dr. Gazdag Gábor
dependent dosing of the anesthetics was found, fix Szent László Hospital, Consultation.Liaison
doses were administered. Psychiatric Service
Although a dosage regime which had been pub- Hungary, 1097 Budapest, Gyáli út 5-7.
lished before was used, there is no hard evidence Telefon/fax: (1) 455-8125,
that in schizophrenic patients 1 mg/kg of propofol e-mail: gazdag@lamb.hu

REFERENCES tients with partial epilepsy. Anesthesi- propofol or methohexitone anesthesia.

ology. 1995; 82 (4): 843-51. Anesth Analg. 1992; 74: S192
1. Sutherland MJ, Burt P. Propofol and 7. Ebrahim ZY, Schubert A, Van Ness 13. Martensson B, Aniko B, Börje H, et
seizures. Anaesth Intensive Care. P, Wolgamuth B, Awad I. The effect al. A comparison of propofol and
1994; 22 (6):733-7. of propofol ont he electroencephalo- methohexital as anesthetic agents for
2. Harrigan PW, Browne SM, Quail gram of patients with epilepsy. Anesth ECT effects on seizure duration, ther-
AW. Multiple seizures following Analg. 1994; 78 (2): 275-9. apeutic outcome, and memory . Biol
re-exposure to propofol. Anaesth In- 8. Hasan MM, Hasan ZA, al-Hader AF, Psychiatry. 1994; 35: 179-189.
tensive Care. 1996; 24 (2):261-4. Takrouri MS. The anticonvulsant ef- 14. Fear CF, Littlejohns CS, Rouse E et
3. Manikandan S, Sinha PK, Neema fects of propofol, diazepam, and thio- al. Propofol anaesthesia in electro-
PK, Rathod RC. Severe seizures dur- penthal, against picrotoxin-induced convulsive therapy. Reduced seizure
ing propofol induction in a patient seizure int he rat. Middle East J duration may not be relevant. Br J
with syringomyelia receiving baclo- Anesthesiol. 1993; 12 (2): 113-21. Psychiatry. 1994; 165: 506-9.
fen. Anesth Analg. 2005; 100 (5): 9. Hartung J, Ying H, Weinberger J, 15. Geretsegger C, Rochowanski E,
1468-9. Cottrell JE. Propofol prevents or ele- Kartnig C, et al. Propofol and metho-
4. Walder B, Tramer MR, Seeck M. vates the treshold for lidocaine- hexital as anesthetic agents for electro-
Seizure-like phenomena and propofol: induced seizures in rats. J Neurosurg convulsive therapy (ECT): a compari-
a systematic review. Neurology. 2002; Anesthesiol. 1994; 6 (4): 254-9. son of seizure-quality measures and
58 (9):1327-32. 10. Lee VC, Moscicki JC, DiFazio CA. vital signs. J ECT. 1998; 14: 28-35.
5. Wang B, Bai Q, Jiao X, Wang E, Propofol sedation produces dose- 16. Avramov MN, Husain MM, White
White PF. Effect of sedative and hyp- dependent suppression of lidocaine- PF. The comparative effects of
notic doses of propofol ont he EEG induced seizures in rats. Anesth methohexital, propofol and etomidate
activity of patients with or without a Analg. 1998; 86 (3): 652-7. for electroconvulsive therapy. Anesth
history of seizure disorders. J Neuro- 11. Mitchell P, Torda T, Hickie I, Burke Analg 1995; 81: 596-602.
surg Anesthesiol. 1997; 9 (4): 335-40. C. Propofol as an anaesthetic agent for 17. Stadtland C, Erfurth A, Ruta U et
6. Samra SK, Sneyd JR, Ross DA, ECT: effect on outcome and length of al. A swich from propofol to etomi-
Henry TR. Effects of propofol seda- course. Aust N Z J Psychiatry. 1991; date during an ECT course increases
tion on seizures and intracranially re- 25: 255-61. EEG and motor seizure duration. J
corded epileptiform activity in pa- 12. Malsch E, Gratz I, Mani S. Efficacy ECT. 2002; 18 (1): 22-25.
of electroconvulsive therapy after

Neuropsychopharmacologia Hungarica 2007, IX/3; 125-130 129

EREDETIKÖZLEMÉNY GAZDAGGÁBOR,TOLNAJUDIT,IVÁNYIZSOLT

18. Gazdag G, Kocsis N, Tolna J, 20. Fu W, Stool LA, White PF et al. propofol anaesthesia. Acta Anaes-
Ivanyi Zs. Etomidate versus propofol Acute hemodinamic responses to thesiol. Scand. 2004; 48 (1): 27-34.
for electroconvulsive therapy in pa- electroconvulsive therapy are not re- 22. Lehmann A, Thaler E, Boldt J: [Is
tients with schizophrenia. J ECT lated to the duration of seizure activ- measuring the depth of anesthesia sen-
2004; 20 (4): 225-9. ity. J Clin Anesth. 1997; 9: 653-657. sible? An overview on the currently
19. Swartz CM, Abrams R. ECT 21. Schultz A, Grouven U, Zander I, available monitoring systems]. AINS
instuction manual. Somatics Inc. Beger FA, Siedenberg M, Schultz B. 2001; 36 (11): 683-92.
(1994) Age-related effects int he EEG during

130 Neuropsychopharmacologia Hungarica 2007, IX/3; 125-130