Download PDF
Pharmacopsychiatry 2024; 57(01): 30-34
DOI: 10.1055/a-2197-9635
Original Paper

Augmentation of Electroconvulsive Therapy with Oral Caffeine: A Retrospective Analysis of 40 Patients with Major Depression

Peter Nyhuis
1   Klinik für Psychiatrie, Psychotherapie und Psychosomatik, St. Marien Hospital Eickel, Herne, Germany
,
Dorothea Mücke
1   Klinik für Psychiatrie, Psychotherapie und Psychosomatik, St. Marien Hospital Eickel, Herne, Germany
,
Michael Specka
2   LVR-University Hospital Essen, Department of Psychiatry and Psychotherapy, Medical Faculty, University of Duisburg-Essen, Essen, Germany
,
Norbert Scherbaum
2   LVR-University Hospital Essen, Department of Psychiatry and Psychotherapy, Medical Faculty, University of Duisburg-Essen, Essen, Germany
› Author Affiliations
› Further Information
Also available at
 
 PDF Download Permissions and Reprints

Abstract

Objective Studies have demonstrated the efficacy of injectable caffeine as an augmentation method in electroconvulsive therapy (ECT). This study investigated whether orally administered caffeine increases seizure duration during ECT.

Methods Medical records of 40 patients treated with a series of ECT were retrospectively analyzed. Patients whose electroencephalogram (EEG) seizure duration had dropped<30 s, or motor seizure duration<15 s were included. They subsequently received oral caffeine (0.2 g) before ECT sessions. Primary outcomes were EEG seizure duration and motor seizure duration, compared with those from the last pre-caffeine session (baseline) and the first five caffeine-augmented sessions. The mental state was assessed with the Global Assessment of Functioning (GAF). In addition, data on maximum heart rate, maximal arterial pressure, and adverse effects were collected.

Results The EEG seizure duration increased by 14.9 s (52%) on average between baseline and the first caffeine-augmented session. The increased length remained widely stable over the subsequent ECT sessions. EEG seizure duration was>30 s in more than 80% of sessions. A statistically significant increase in motor seizure duration appeared only in the 2nd and 3rd of five sessions with augmentation. Oral caffeine pretreatment was overall well tolerated. The percentage of patients with at least serious mental impairment (GAF score≤50) dropped from 77.5 to 15%.

Conclusions Results suggest the utility of oral caffeine (0.2 g) to increase ECT-induced seizures in patients with durations below clinically significant thresholds.


#

Key words

caffeine - electroconvulsive therapy (ECT) - augmentation - seizure duration

Introduction

Electroconvulsive therapy (ECT) is an effective treatment for severe pharmaco-resistant depression, bipolar disorder, and psychosis [1] [2], with overall response rates up to 80% [3] [4]. While the mechanism of action of ECT is still the subject of research [4], the induction of a sufficiently long epileptic seizure is considered central to its effectiveness. Although the literature is still not clear about whether determined seizure durations are essential for clinical effectiveness, a duration of at least 30 s, measured in the electroencephalogram (EEG), is widely assumed as minimal duration [5]. In addition, the duration of a motor seizure of at least 20 s is assumed to be necessary for the therapeutic effect of ECT [6].

ECT is usually carried out as a series of sessions. For the individual patient, the probability of reaching the 30 s threshold decreases with an increasing number of sessions [7]. Therefore, the efficacy of several augmentation methods for the prolongation of seizure duration has been investigated. In this regard, methods such as hyperventilation, application of etomidate, ketamine, or flumazenil, and premedication with theophylline, aminophylline, or caffeine [8] [9] were investigated.

Caffeine is a xanthine alkaloid that produces central nervous system (CNS) stimulation via adenosine receptor antagonism [10]. After oral intake, it is completely resorbed within 20 min and reaches the internal organs and the CNS within a few minutes. Its half-life depends on age and sex, but can be estimated as roughly 4 h. Due to its lipophilic nature, it crosses the blood-brain barrier well and reaches a bioavailability of 90 to 100% [11]. Caffeine mediates a stimulating effect in the CNS based on its competitive inhibition at the subtype A2 adenosine receptors. Moreover, caffeine inhibits the decomposition of cyclic adenosine-3’,5’-monophosphate (cAMP), thus enhancing the release of adrenaline through cAMP as a second messenger. A decrease in the seizure threshold, as intended by ECT augmentation, is primarily achieved through the inhibition of adenosine effects [12].

The ability of injectable caffeine sodium benzoate to lengthen seizure duration in ECT was repeatedly demonstrated [5] [13]. Case studies demonstrated an increase of up to 127% in seizure duration after pretreatment with injectable caffeine, compared with treatment without caffeine [14]. A case series of ECT with pretreatment of intravenous caffeine-sodium-benzoate solution demonstrated increased seizure duration without relevant side effects [15]. Likewise, other studies using caffeine citrate solutions showed prolongation of seizure in ECT [16].

Regarding the administration of oral caffeine and its effect on ECT seizure duration, to our knowledge, only one study has been carried out so far. Doses between 300 and 1000 mg, administered 1 h before ECT, successfully lengthened seizure duration that had decreased below 25 s, in patients with a diagnosis of depression. However, this study was carried out in a sample of geriatric patients [17]. We therefore investigated the administration of a tablet with oral caffeine (200 mg) as an augmentation strategy for ECT in patients with mental illness, whose seizure duration decreased below the required threshold.


#

Methods

Procedures

The Department of Psychiatry, Psychotherapy, and Psychosomatics, St. Marien Hospital Eickel in Herne, Germany, has been carrying out ECTs since 2009. The frequency of ECT sessions is about 400 per year. The indication for ECT is made by a consultant in psychiatry. Patients receive information about ECT therapy and the required anesthesia, and give their written consent. Patients undergo a thorough medical examination, including electroencephalogram (EEG) and blood laboratory analysis. A psychiatrist, an anesthetist, and a nurse are present at each session. ECTs are carried out twice weekly for each patient. Usually, a series consists of 6 up to 12 treatments and sometimes more, if necessary, depending on clinical symptoms and tolerance of the patient.

The device used is Thymatron System IV, manufactured by Somatics Inc. A series starts with unilateral stimulation on the right side. The Thymatron System IV uses a brief pulse wave, delivering a constant 900 mA stimulus, with a maximum duration of 8 s. The maximum charge is 1008 mC (Millicoulomb) with a corresponding maximum energy of 198.8 Joules. The seizure threshold is determined by age-based dosing [18]. All patients started with the right unilateral (RUL) ECT stimulus. Doses are increased during the course of treatment according to the characteristics of ictal activity, and as usual in ECT-units in Germany [19]. After achieving 504 mC in RUL stimulation, all patients receive this charge during bitemporal ECT stimulation. General anesthesia is induced with methohexital (Brevimytal) 1.5 mg/kg. Intravenous succinylcholine (1 mg/kg) is used for muscle relaxation. Ventilation is assisted with a face mask and 100% oxygen. Patients are monitored with regard to heart rate, oxygen saturation, and blood pressure during ECT and during post-anesthetic recovery.

Two prefrontal-mastoid channels are used to monitor the ictal EEG. The seizure duration is measured by EEG during treatment and also by measurement of the motor seizure. In addition, the postictal suppression index (PSI) is determined and registered. It is based on a quantitative comparison between the amplitude of the EEG during a seizure and the flattening after a seizure. A PSI>75% is considered as indicating a sufficient convulsion [20].

Patients with bilateral stimulation at 100% of the machine’s performance who show a seizure duration (EEG)<30 s or a motor seizure duration<15 s during the course of treatment, receive premedication with 0.2 g oral caffeine in tablet form, 30 min before electroconvulsive therapy. The decision for augmentation is, therefore, independent of a patient’s number of sessions. The standard dosage of caffeine tablets available in Germany is 200 mg.


#

Sample

The present retrospective study includes all patients who received caffeine premedication between January 2011 and July 2018 during their ECT treatment (including maintenance treatment) in the St. Marien Hospital Eickel in Herne. There were no exclusion criteria with respect to age, diagnosis, or sex ([Table 1]). Caffeine was given pre-ECT if a patient had shown a seizure duration, measured by EEG, of less than 30 s or if a motor seizure duration of less than 15 s had been measured (on the arm where the impact of succinylcholine is prohibited by a cuff). PSI had no influence on the decision for caffeine augmentation.

Table 1 Patient characteristics.

Age (years)

Mean (SD).

61.5 (15.7)

Median (Min-Max)

61.5 (29–88)

Sex

Male

18 (45%)

Female

22 (55%)

Primary Diagnosis

Schizophrenia

2 (5%)

Major depression

23 (57.5%)

Psychotic depression

10 (25.0%)

Bipolar disorder

5 (12.5%)

Duration of illness (years)

Mean (SD).

13.1 (10.9)

Median (Min-Max)

10.0 (1–49)

Antipsychotic medications

None

14 (35%)

One

24 (60%)

Two

2 (5%)

Antidepressants

None

4 (10%)

One

28 (70%)

Two

8 (20%)

According to the protocol, the administration of caffeine was usually started with the 7th session of ECT in a series usually consisting of 12 sessions. Then, patients were followed up until their 5th consecutive session with caffeine. It was expected that the augmenting effect of caffeine would decrease with the course of the ECT sessions. Anonymized data was collected from patient files, ECT records, anaesthesia records, and EEG recordings during ECTs. If the same patient received several ECT series with caffeine augmentation during the inclusion period, only his or her first treatment series was included. The study was accepted by the ethics committee of the Medical Faculty of the University of Duisburg-Essen (18–8505-BO).


#

Data collection

The major aim of this study was to investigate whether seizure duration (EEG) increases in ECT sessions with caffeine premedication compared to previous sessions without administration of caffeine. The study also collected data on demographics, duration of motor seizures, postictal suppression index, maximum heart rate and maximum arterial pressure and adverse effects. Global Assessment of Functioning (GAF) Scale was applied at treatment entry and at discharge by treating physicians to evaluate patients’ mental, social, and occupational functioning [21].


#

Statistical analysis

The primary outcome criterion was seizure duration as measured by EEG. Patients received different numbers of ECT sessions with caffeine, therefore number of patients decreased with increasing number of sessions. We compared the caffeine-augmented sessions (“follow-up”) with the baseline (last session without caffeine) using Wilcoxon signed-rank tests. Group comparison was performed using independent samples Welch-t- test. Because of a multitude of statistical tests carried out here, the Bonferroni-Holm correction was used to achieve a study-wise alpha level of 0.05 [22].


#
#

Results

Forty patients were included, and 174 ECT sessions with caffeine premedication were recorded. 23 out of 40 (57,5%) patients fulfilled the inclusion criterion of a seizure duration in EEG recordings of less than 30 s, 17 out of 40 patients (42.5%) fulfilled the criterion of a motor seizure duration of less than 15 s. The number of patients decreased from 40 to 15 from the first to the 5th session with caffeine augmentation, as ECT application was stopped in case of sufficient clinical improvement. Patients received an average number of 3.5 ECT sessions with caffeine administration.

From the last ECT session without caffeine (“baseline”) to the first caffeine-augmented session, the mean ECT seizure duration in the EEG increased from mean 28.4 (SD 10.5 s) to 43.3 (16.4) s (p<0.001, [Table 2]). Seizure duration was not significantly correlated between these two sessions (r=0.06; p=0.74). Changes between baseline and following ECT sessions had a similar magnitude for sessions 2 to 5 with a statistically significant increase of seizure duration for sessions 2 and 3 (sessions 4 and 5 comprised of n=22 and n=15, respectively, remaining patients only). An EEG seizure duration of 30 s or more was measured in 80.4% of all caffeine-augmented sessions. In particular, of those 23 patients with seizure duration below 30 s at baseline, 15 (65.2%) achieved a sufficient duration in the first session with caffeine augmentation and 6 (26.1%) in the second or third session. Only 3 patients (7.5%) did not achieve a seizure duration of 30 s or more: two of them had only 2 sessions with caffeine augmentation, and 1 patient had only one session with caffeine augmentation.

Table 2 Comparison of ECT duration (sec.) during the last session without caffeine (“baseline“) and during sessions with caffeine (“follow-up“). Due to decreasing number of patients during the course of the study, comparisons refer to different sample sizes (n).

Nr. of caffeine augmented ECT

n

Mean (SD) sec. seizure duration in the EEG

Proportion with seizure duration above 30 seconds

Mean (SD) sec. motoric seizure duration

Mean (SD) sec. motoric seizure duration

Mean (SD) sec. motoric seizure duration

Baseline

Follow-up

p1

Baseline

Follow-up

Baseline

Follow-up

p1

1

40

28.4 (10.5)

43.3 (16.4)

<0.001* 

42.5%

72.5%

13.8 (7.8)

17.7 (10.9)

0.121

2

34

28.5 (10.9)

42.2 (14.8)

<0.001* 

44.1%

79.4%

13.7 (8.1)

21.7 (9.7)

<0.001* 

3

27

28.9 (10.8)

45.3 (10.8)

<0.001* 

44.4%

92.6%

14.1 (8.1)

21.3 (8.0)

<0.001* 

4

22

29.0 (11.5)

38.7 (10.9)

0.009

40.9%

77.3%

14.8 (8.2)

18.5 (7.0)

0.123

5

15

30.1 (13.3)

41.2 (9.8)

0.029

46.7%

86.7%

15.2 (8.8)

17.7 (6.9)

0.363

1 Wilcoxon signed rank test.  *= statistically significant on a Bonferroni-Holm-corrected level of significance. ECT: electroconvulsive therapy; EEG: electroencephalogram; SD: standard deviation.

Regarding motor seizure duration, the mean increase from baseline (13.8, SD 7.8 s) to the first caffeine session (17.7 s, SD 10.9) was statistically not significant, while the increase from baseline to follow-up 2 and 3, respectively, was statistically significant ([Table 2]).

The last observation carried forward-analysis

In order to assess the effects of discontinuation of ECT treatment on the primary outcome (possible positive selection through drop-out of patients with insufficient seizure duration), missing data were substituted by the last available EEG seizure duration. The mean seizure duration remained stable above 40 s ([Table 3]).

Table 3 EEG seizure duration with last observation carried forward (n=40 for all time points).

Baseline

Follow-up 1

Follow-up 2

Follow-up 3

Follow-up 4

Follow-up 5

Mean

28,4

43,3

42,9

45,1

40,6

42,2

Median

26,5

40,5

41,5

44,0

39,0

42,0

SD

10,5

16,4

15,8

13,4

13,4

12,6

Minimum

3

15

15

23

22

24

Maximum

52

88

88

88

88

88

EEG: electroencephalogram; SD: standard deviation.


#

Clinical course

At treatment entry, 77.5% of patients showed GAF scores of 50 points or less, which indicated serious symptoms and impairment of everyday functioning (45% with a GAF score of 50 points, 25% with a score of 40 points, and 7.5% with a score of 30 points). At the end of treatment, 15% were still in that range (10% with a score of 50 points, 2.5% with a score of 40 points, and 2.5% score of 30 points). In 39 out of 40 patients (97.5%), there was an increase in GAF scores at discharge compared to admission (42.5% with 10 points increase, 45% with 20 points, and 10% with 30 points). For those patients who did not achieve a seizure duration>30 s (n=3), the increase in GAF scores (mean 13.3, SD 5.8) was not significantly different from the remaining patients (n=37; mean 16.9, SD 6.7; p=0.39, Welch t-test).


#

Safety issues

At baseline, ECT-related side effects were noted in 17 out of 40 patients (42.5%). These were headache (n=9), high blood pressure (n=4) during ECT sessions and the following 5 min (>140/90 mmHg), nausea (n=2), and cognitive disturbance (n=2). This remained widely stable in the first follow-up session: no side effects 40% (n=24), headache (n=5), high blood pressure (n=5), nausea (n=2), cognitive disturbance (n=2). Regarding the remaining ECT follow-up sessions, 48 out of 134 (35.8%) were associated with side effects: 20 with headache, 19 with high blood pressure, 7 with cognitive disturbance, and 2 with nausea. Whether a patient had shown side effects or not at follow-up session 1 was not associated with the number of further follow-up sessions (mean 2.5 for both subgroups). No clinically severe or lasting side effect was documented.

The mean of the maximum heart rate during a seizure was 91.6/min (SD 9.9) at baseline and 92.8/min (9.5) during the first caffeine session (p=0.74). Means for the remaining follow-up sessions were between 89.8/min and 92.6/min, and comparisons with the respective baseline samples resulted in p-values between 0.15 and 0.49. At baseline, 10.5% of patients had a maximum heart rate>100/min; during follow-up, 6.7% of patients. Maximum systolic blood pressure was between 120 and 220 mmHg at baseline (mean 184.1, SD 18.5). During the first caffeine-augmented session, it was a bit lower, on average (mean 179.7, SD 18.6; p=0.013 not statistically significant on the Bonferroni-Holm corrected level of significance), and ranged between 140 and 220 mmHg. During the remaining sessions, it was between 184.1 and 185.3, on average, and comparisons with the respective baseline values resulted in p-values between 0.057 and 0.743 (not statistically significant).


#
#

Discussion

In the present uncontrolled study, unblended gathered data were retrospectively analyzed regarding seizure duration after administration of oral caffeine (0.2 g) 30 min before an ECT session compared to seizure duration of previous sessions without caffeine augmentation. Patients included had previously shown insufficient EEG- and/or motor seizure duration. In this sample, patients showed longer seizure duration in the case of caffeine administration. After augmentation with caffeine, more than 80% of all analyzed ECT sessions achieved a seizure duration above 30 s. In parallel, increasing GAF scores indicate a substantial clinical improvement. The augmentation with caffeine was well tolerated by the patients and was not associated with an increase in side effects, especially regarding a critical increase in blood pressure and heart rate.

These results are in line with findings from other uncontrolled studies, in which caffeine administered intravenously was associated with increased EEG or motor seizure duration in patients who had not or had shown a decline of their seizure durations under a critical threshold [5] [13] [14] [15] [20] [23] [24]. In addition, two controlled studies showed that the efficacy of ECT may be enhanced with intravenously administered caffeine pretreatment, with either lower ECT dosage or fewer treatments required to attain the same level of efficacy as the control group [24] [25]. Although the previous studies confirmed the safety of the use of intravenously administered caffeine, cases of cardiac arrhythmias associated with caffeine and ECT have also been described [9].

Regarding oral caffeine, to our knowledge, only one published report exists on its use as augmentation in ECT. In 30 geriatric patients whose seizure duration had fallen below 25 s, titrated doses of 300–1000 mg caffeine successfully increased seizure duration, without evidence of ECT-related complaints or serious cardiovascular or other complications [17]. The present study documents that oral caffeine achieved sufficient seizure lengths in 80% of sessions with caffeine augmentation as recorded in the EEG in a sample of patients with different mental disorders, including non-geriatric patients. This effect was already achieved by a dosage of 200 mg caffeine, which is the dosage of commercially available caffeine tablets in Germany. Taking into account possible side effects of the administration of caffeine, it is of importance that in this study, oral caffeine already in a dose of 200 mg sufficiently augmented seizure duration compared to 300–1000 mg in the cited study. However, from the reported data of that study, it also could be concluded that 300–1000 mg caffeine is more effective, but it is not clear if it is due to the intravenous route or due to the higher dose. Furthermore, it must be pointed out that – especially in elderly patients – seizure duration during the ECT course does not seem to be essential for effectiveness [26] [27].

Our study did not investigate possible effects of gender or additionally administered medication on seizure duration, for this could only be analyzed in a larger sample. In addition, this study is a not controlled, not randomized, and only retrospective observational investigation, and the data must be interpreted cautiously. And lastly, the exact mechanism of action of caffeine in increasing seizure duration in ECT is not known until now and was not examined in our investigation.

In sum, our results confirm prior studies using caffeine as an augmentation agent of ECT. Here, oral caffeine was administered. Based on our data, augmentation with oral caffeine may effectively increase seizure duration and can, therefore, contribute to optimizing the response to ECT. Augmentation with oral caffeine was well tolerated in this sample.


#
#

Conflicts of Interest

The authors declare that they have no conflict of interest.


Correspondence

Dr. med Peter Nyhuis
St. Marien Hospital Eickel
Marienstraße 2
44651 Herne
Germany    

Publication History

Received: 28 July 2023
Received: 06 October 2023

Accepted: 12 October 2023

Article published online:
23 November 2023

© 2023. Thieme. All rights reserved.

Georg Thieme Verlag KG
Rüdigerstraße 14, 70469 Stuttgart, Germany


   Permissions and Reprints