Do ictal EEG characteristics predict treatment outcomes in schizophrenic patients undergoing electroconvulsive therapy?
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
GULNIHAL GOKCE SIMSEK, SELMA ZINCIR, HUSEYIN GULEC, SEVGIN EKSIOGLU, UMIT BASAR SEMIZ, YASEMIN SIPKA KURTULMUS
Simsek GG, Zincir S, Gulec H, Eksioglu S, Semiz UB, Kurtulmus YS. Do ictal EEG characteristics predict treatment outcomes in schizophrenic patients undergoing electroconvulsive therapy? Nord J Psychiatry 2015;69:466–471. Objective: The aim of this study is to investigate the relationship between features of electroencephalography (EEG), including seizure time, energy threshold level and post-ictal suppression time, and clinical variables, including treatment outcomes and side-effects, among schizophrenia inpatients undergoing electroconvulsive therapy (ECT). Method: This is a naturalistic follow-up study on schizophrenia patients, diagnosed using DSM-IV-TR criteria, treated by a psychosis inpatient service. All participants completed the Brief Psychiatric Rating Scale (BPRS), the Global Assessment of Functioning (GAF) scale, the Frontal Assessment Battery (FAB) and a Data Collection Form. Assessments were made before treatment, during ECT and after treatment. Findings: Statistically significant improvements in both clinical and cognitive outcome were noted after ECT in all patients. Predictors of improvement were sought by evaluating electrophysiological variables measured at three time points (after the third, fifth and seventh ECT sessions). Logistic regression analysis showed that clinical outcome/ improvement did not differ by seizure duration, threshold energy level or post-ictal suppression time. Discussion and conclusion: We found that ictal EEG parameters measured at several ECT sessions did not predict clinical recovery/outcomes. This may be because our centre defensively engages in “very specific patient selection” when ECT is contemplated. ECT does not cause short-term cognitive functional impairment and indeed improves cognition, because symptoms of the schizophrenic episode are alleviated. • Electroconvulsive therapy, Ictal, Post-ictal suppression time, Schizophrenia, Side-effect, Treatment outcome. Gulnihal Gokce Simsek, M.D., Psychiatry Department, Yozgat State Hospital, Yozgat, Turkey, E-mail: [email protected]; Accepted 16 December 2014.
A
ntipsychotic drugs are considered optimal first-line treatments for schizophrenia. Even when appropriate treatments protocols are applied, 25% of patients demonstrate inadequate or no improvement in illness progression (1). Patients may have severe symptoms, experience long periods of hospitalization and high rates of lifelong dysfunction, including over the working years. Treatment drop-out creates a vicious circle in the schizophrenic population over time. Although electroconvulsive therapy (ECT) use features in some treatment algorithms for schizophrenia, the technique is not often applied except in Europe and in many developing countries (2). Affective symptoms, acute initiation and a short duration of schizophrenia are indicators that ECT may be useful. Some authors have stated
© 2015 Informa Healthcare
that ECT may be used to effectively treat acute schizophrenia of Type 1, but not Type 2 (chronic schizophrenia), unless depression is also present (3). A recent meta-analysis showed that a combination of ECT with antipsychotic medication was superior to antipsychotic monotherapy alone, rapidly decreasing global symptoms. Even when no further maintenance ECT sessions were given, relapse prevention was more effective when combination therapy was applied (4). Recent studies have highlighted the benefits afforded by ECT as a component of schizophrenia treatment. Useful long-term effects were evident, especially in combination with antipsychotic drugs (15–17). Only a few reports have addressed ECT therapeutic efficacy and variables affecting treatment outcomes in DOI: 10.3109/08039488.2014.1003403
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
EEG PREDICTION IN SCHIZOPHRENIC PATIENTS UNDERGOING ECT
schizophrenia patients. Such studies play an important role in identifying preventative steps reducing cognitive side-effects and markers of clinical progression. The advantage of using electroencephalography (EEG) during ECT sessions is not limited to simple determination of seizure duration. Many ictal EEG markers can be used to assess and predict treatment outcomes. Some such markers are post-ictal suppression time, the extent of slowing in EEG waves, the time of initiation of slowing, global EEG power, the threshold energy level and seizure time. Recent studies have emphasized the significance of seizure time, threshold energy level and post-ictal suppression time; cognitive side-effects of ECT have been noted in recent works (5–7). Post-ictal suppression time has received most attention (5). We investigated the efficacy and side-effects of daily applications of ECT. Our hypotheses were, first, that treatment outcomes in schizophrenia patients receiving ECT could be predicted by peri-ictal properties, including motor seizure time, threshold electrical energy and post-ictal suppression time; and, second, that cognitive functions would be preserved short-term after ECT sessions concluded. Although resistance to treatment remains a serious clinical problem, we wish to highlight the fact that ECT can appropriately contribute to a solution of this problem. Also, we were concerned to ensure the more effective use of ECT in daily practice and to create a basis allowing clinicians of the future to effectively employ ECT.
Materials and methods Subjects This naturalistic clinical follow-up study was performed in the Erenköy Mental Health Research and Training Hospital, on schizophrenia inpatients, in the interval August 2011– January 2012. All patients scheduled for ECT by a study-blind treatment team were reviewed. A total of 61 volunteer patients diagnosed using DSM-IV criteria and aged between 18 and 65 years participated. Patients with comorbidities affecting cognitive assessment, including another psychiatric/medical disorder, mental retardation, a history of head trauma or a history of drug abuse in the prior 12 months, were excluded. All patients were assessed at the beginning of the study; at the third, fifth and seventh ECT sessions; and 2–7 days after conclusion of ECT treatment. All participants signed informed consent forms before study commencement. Ethical approval was given by the Erenköy Mental Health Research and Training Hospital Institutional Review Board.
Concurrent medications All patients took psychopharmacological medications during the study. Of the 61 patients, 44 were taking a daily NORD J PSYCHIATRY·VOL 69 NO 6·2015
average of 19 mg haloperidol, 11 were taking 4.5 mg/day risperidone, 10 were taking 20 mg/day olanzapin, 10 were taking 43 mg/2 weeks risperidon LAI, nine were taking 82 mg/day clozapine, six were taking 350 mg/day quetiapine, three were taking 666 mg/day amisulpiride and one was taking 9 mg/day paliperidone.
Assessment tools Our primary measurement tool was the Brief Psychiatric Rating Scale (BPRS). The Global Assessment of Functioning (GAF) test was also used. Patients were given the Frontal Assessment Battery (FAB) test before and 2–7 days after ECT sessions concluded, to monitor cognitive functions and alterations therein. Interviews were conducted face-to-face with both patients and caregivers (who gave informed consent to be interviewed). We collected information on patient age, sex, educational level, marital status, family medical history, history of physical disease, age at onset of illness, illness duration, attempted suicide history, number of acute exacerbations of illness, treatment history, number of hospitalizations, recent ECT history and efficacy thereof, total number of ECT sessions, ECT session durations, ictal EEG parameters, anaesthetic dosage, average energy threshold, and scores on all assessment scales. The BPRS was developed by Overall & Gorham in 1962 (8) and modified later by Bech et al., with a reduction in the number of questions to 18. Each item is scored from 0 to 6 or from 1 to 7. The maximum total score is 72. We translated the scale into Turkish. No studies on validity or reliability were performed because intercultural usage of the scale is compromised, in epidemiological research, in terms of validity and reliability. Psychiatric symptoms were assessed by simple interview. Scale items explore symptoms of psychosis, depression and anxiety. The instrument records somatic complaints, mental anxiety, emotional withdrawal, thought disorganization, feelings of guilt and self-vilification, somatic anxiety, the presence of specific motor disorders, exaggerated self-esteem, depressive mood, hostility, scepticism, hallucinations, psychomotor retardation, refusal to co-operate, bizarre thought content, blunted and inappropriate affect, psychomotor agitation and disorientation/confusion. Dubois et al. (10) developed the FAB to assess six domains associated with frontal lobe-linked functions, striatofrontal functional disorders, movement disorders and frontal lobe damage. The short battery can be applied at the bedside to examine executive functioning. Answers are scored from 0 to 3 and the range of scores is from 0 to 16. Interview time is ∼10 min. The validity and reliability of the Turkish version have been demonstrated by Güleç et al. (11). The GAF scale was developed by Spitzer et al. to assess individual global functioning from a healthy stage to manifestation of a psychiatric disorder. The test is
467
GG SIMSEK ET AL.
relatively easy to administer, specific and sensitive, and finds applications in both clinical practice and research.
EEG device Galileo, Neoro SpA, Firenze, Italy (2004, model no. SN20325).
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
ECT technique All patients were given ECT three times a week. After preparation for ECT, bitemporal leads were applied and frequency and energy dosage were measured on the ECT machine using a constant current and a brief pulse stimulus (Thymatron System IV, Somatics Inc.; two-way square-pulse wave with four monitoring points [EEG1, EEG2, EKG and EMG]). The anaesthetic used is 0.5 mg/ kg propofol and for adequate muscular relaxation 0.5 mg/ kg succinylcholine. The initial energy dosage was calculated using the “half-age” formula. Seizure periods of 20 s and above were considered adequate and, if such times were not achieved, the dosage was titrated upwards and charge was increased by 50% (14). The cut-off point was set at 279 s, that of the median value of the mean total seizure time on ECT. Values below the cut-off point were termed “short ECT” and higher values “long ECT”. We used these criteria because of a lack of consensus on appropriate measures in the literature. Use of the median value seemed appropriate. The median ECT energy threshold was 37% and this was set as the cut-off point. Values below this level were termed “low threshold” and higher values “high threshold”. We used these criteria because of a lack of consensus on appropriate measures in the literature. Use of the median value seemed appropriate. Basal and peri-ictal EEG records were assessed by an experienced neurologist (YSK), who measured EEG seizure time, seizure frequency and post-ictal suppression time (the inactive period after termination of a seizure that was not an artifact) (7). All data from all ECT sessions undergone by all patients were examined and all information was included in the analysis.
Statistical analysis A statistical software package was used to evaluate data. Descriptive statistical methods (frequency, percentage, median and standard deviation) are used to present demographic and clinical information. The Mann–Whitney U test was used for correlation analysis of quantitative data from two subgroups because data distribution was not normal. If more than two groups were compared, we used the Kruskal–Wallis test, or the Mann–Whitney U test with Bonferroni correction, to seek significant correlations. Wilcoxon’s signed-rank
468
test was used when correlating data within groups. Spearman’s correlation analysis was used to investigate relationships between quantitative data. The 95% confidence intervals were calculated; P ⬍ 0.05 was regarded as significant and P ⬍ 0.01 highly significant.
Results Of all patients, 26.2% were female and 73.8% male; 18.0% worked; 4.9% were retired; 4.9% were pensions because of illness; 72.1% unemployed; 75.4% single; 11.5% married; 9.8% divorced; and 3.3% widowed. Mean patient age was 33.3 ⫾ 10.9 years; mean BMI 24.2 ⫾ 3.5 kg/m2; and mean years of education 9.4 ⫾ 3.5. Demographic data are shown in Table 1. The mean age onset of illness was 23.6 ⫾ 6.8 years, mean illness duration 9.6 ⫾ 7.4 years, mean number of exacerbations 5.6 ⫾ 5.7 and mean number of hospitalizations 4.4 ⫾ 5.1. The mean total number of ECT sessions was 6.8 ⫾ 1.2, the mean total duration of ECT 309.0 ⫾ 124.1 s and the mean doses of succinylcholine and propophole 36.0 ⫾ 9.5 and 44.3 ⫾ 15.8 mg. The mean energy threshold was 37.4 ⫾ 9.8%. Of all patients, 29.5% had treatment-resistant schizophrenia and the remaining 70.5% were inpatients with acute exacerbations of schizophrenia. Of all patients, 44.3% had histories of suicide attempts, 11.5% physical illnesses and 41.0% family histories of psychiatric disorders. Correlations between clinical features and EEG characteristics (age at commencement of illness and mean energy threshold; illness duration and mean energy threshold, post-ictal suppression time) were positive (Table 2). The cut-off point was set at 279 s, that of the median value of the mean total seizure time on ECT. Values below the cut-off point were termed “short ECT” and higher values “long ECT”. Comparisons of CGI, GAF and FAB scores before and after treatment revealed no significant differences in patients who underwent short and long ECT. The median ECT energy threshold was 37%; this was set as the cut-off point. Values below this level were termed “low threshold” and higher values “high threshold”. Comparisons of CGI, GAF and FAB scores before and after treatment revealed no significant differences in patients who underwent “low threshold” and “high threshold ECT. Logistic regression analysis of data from the third, fifth and seventh ECT sessions was performed to predict treatment outcomes (a decrease of 50% or more in the BPRS score) of ictal EEG characteristics including seizure duration, seizure energy threshold and post-ictal suppression time (Table 3). We also analysed clinical features. NORD J PSYCHIATRY·VOL 69 NO 6·2015
EEG PREDICTION IN SCHIZOPHRENIC PATIENTS UNDERGOING ECT
Table 1. Demographic features (n=61). Mean⫾ standard deviation 33.3 ⫾ 10.9 24.2 ⫾ 3.5 9.4 ⫾ 3.5
Age (years) BMI (kg/m2) Educational status (years)
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
n Sex Female Male Work status Working Retired Invalid Not working Marital status Single Married Divorced Widowed
%
16 45
26.2 73.8
11 3 3 44
18.0 4.9 4.9 72.1
46 7 6 2
75.4 11.5 9.8 3.3
Features of illness (n ⫽ 61)
Mean⫾ standard deviation
Age at commencement of illness (year) Duration of illness (years) Number of exacerbations Number of hospitalizations Total number of ECT sessions Total ECT session duration (seconds) Mean succinylcholine dosage (mg) Mean propofol dosage (mg) ECT threshold (%)
23.6 ⫾ 6.8 9.6 ⫾ 7.4 5.6 ⫾ 5.7 4.4 ⫾ 5.1 6.8 ⫾ 1.2 309.0 ⫾ 124.1 36.0 ⫾ 9.5 44.3 ⫾ 15.8 37.4 ⫾ 9.8
Attempted suicide history Yes History of physical illness Yes Diagnosis Treatment-resistant Acute exacerbation Family history of the disease Yes Treatment history Medical treatment Medical treatment⫹ ECT Recent ECT response Yes
Age at commencement of illness R P Duration of illness R P Number of hospitalizations R P
n
%
27
44.3
7
11.5
18 43
29.5 70.5
25
41.0
41 20
67.2 32.8
17
94.4
Discussion We investigated the relationships between EEG features, including seizure time, energy threshold level and post-ictal suppression time and clinical variables including treatment outcomes and side-effects in schizophrenia inpatients given ECT. We explored whether ECT outcomes could be predicted by peri-ictal EEG
Seizure duration
Threshold
Post-ictal
⫺ 0.80 0.541
0.274 0.032
0.025 0.846
0.195 0.132
0.289 0.024
0.311 0.015
0.128 0.327
0.228 0.078
0.108 0.408
characteristics including post-ictal suppression time, duration of motor seizure and electrical threshold and also searched for relationships between cognitive sideeffects of ECT and EEG characteristics. Table 3. Predictors of outcomes after three (n ⫽ 10), five (n ⫽ 47) and seven (n ⫽ 57) electroconvulsive therapy (ECT) treatments [demographic and clinical features and electroencephalography (EEG) characteristics]. B
BMI, body mass index; ECT, electroconvulsive therapy.
NORD J PSYCHIATRY·VOL 69 NO 6·2015
Table 2. Relationship between electroencephalography (EEG) characteristics and clinical features (n=61).
3 ECT Sex (1) Age Duration of illness Hospitalizations Axis 1 diagnosis (1) Threshold Mean seizure duration Mean post-ictal Constant 5 ECT Sex (1) Age Duration of illness Hospitalizations Axis 1 diagnosis (1) Threshold Mean seizure duration Mean post-ictal Constant 7 ECT Sex (1) Age Duration of illness Hospitalizations Axis 1 diagnosis (1) Threshold Mean seizure duration Mean post-ictal Constant
Sig.
Exp(B)
1 1 1 1 1
0.984 0.473 0.254 0.377 0.585
0.977 1.073 0.863 1.104 1.695
2.872 3.049
1 1
0.090 0.081
0.862 0.956
⫺ 0.543 0.554 0.961 5.763 3.548 2.638
1 1
0.327 0.104
0.581 318.145
1.018 0.040 ⫺ 0.099 ⫺ 0.063 0.258
1 1 1 1 1
0.197 0.529 0.231 0.388 0.749
2.767 1.041 0.905 0.939 1.295
0.200 0.710
1 1
0.654 0.399
0.977 0.984
⫺ 0.595 0.367 2.632 5.351 2.281 5.505
1 1
0.105 0.019
0.551 210.816
15.41 1.716 ⫺ 1.56 ⫺ 0.513 0.545
1 1 1 1 1
0.100 4,899,397.5 0.163 5.561 0.132 0.210 0.292 0.599 0.853 1.724
2.102 2.314
1 1
0.147 0.128
0.445 0.789
0.712 0.669
1 1
0.399 0.413
2.531 2,464.485
⫺ 0.023 0.071 ⫺ 0.148 0.099 0.528
S.E.
1.175 0.000 0.098 0.516 0.130 1.299 0.112 0.781 0.967 0.298
⫺ 0.149 0.088 ⫺ 0.045 0.026
0.789 1.663 0.064 0.396 0.083 1.434 0.073 0.744 0.809 0.102
⫺ 0.023 0.051 ⫺ 0.016 0.019
9.377 2.699 1.231 1.942 1.037 2.264 0.487 1.110 2.945 0.034
⫺ 0.810 0.559 ⫺ 0.237 0.156 0.928 7.810
Wald df
1.100 9.547
469
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
GG SIMSEK ET AL.
Of all patients, 29.5% had treatment-resistant schizophrenia and the remaining 70.5% were inpatients with acute exacerbations of schizophrenia or similar. We noted statistically significant improvements and decreases in the severity of illness, after ECT, as reported recently (13). Not only did cognitive functioning not deteriorate but it actually improved from the initial state. Statistically significant improvements were noted on all three scales (BPRS, GAF and FAB). Correlations between clinical features and EEG characteristics were sought and the work revealed that age at initiation of illness and energy threshold; duration of illness and mean energy threshold level; and duration of illness and post-ictal suppression time were all positively correlated. However, no statistically significant difference was evident between pre- and post-treatment FAB scores, indicating that no cognitive side-effects developed. No statistically significant relationships were evident between pre- and post-treatment FAB scores and ictal EEG findings. Similar findings have been reported in studies on patients with major depression and schizophrenia (13). Predictor EEG characteristics for changes in cognitive status after ECT are not discussed in this article. Cognitive status improvement was indeed statistically significant, but this was the reverse of was expected in literature when pre- and post-treatment data were compared. Therefore, a decrease in cognitive status, and development of an attention deficit, are part of the natural progression of illness, especially during severely symptomatic periods. When the duration of such periods was reduced, and patients improved, cognitive status also improved significantly after conclusion of ECT treatment. Further follow-up treatment with assessment of cognition would be valuable. The ECT and ictal EEG phases are, respectively: rapid low-frequency low-amplitude initiation, spikepolyspike and high-amplitude tonic phases, spike and slow-wave high- amplitude clonic phases, a phase of slowing of wave amplitude and frequency and the postictal suppression phase. It has been suggested that seizures exhibiting intensive inhibition of these processes are associated with more useful ECT sessions and better outcomes (2). Recent studies have shown that seizure length is not a useful marker of seizure adequacy. Of the various methods used to measure seizure effects, the length of the post-ictal suppression period is the best predictor of treatment outcomes after ECT. The post-ictal suppression time is also a useful prognostic marker (5). High-level EEG suppression was found to be associated with clinical improvement when quantitative post-ictal suppression time was measured in a study on patients with depression (2). In our present study, the ictal EEG parameters (seizure duration, seizure threshold and postictal suppression time) did not differ among ECT sessions, and were not associated with the decreases in
470
BPRS scores noted after each session. In addition, we explored how all measured EEG parameters predicted treatment outcomes. No EEG parameter measured at any of the third, fifth or seventh ECT treatments was a predictor of outcomes in our study. Of all patients, 57 improved after the seventh ECT, 47 after the fifth ECT and 10 after the third ECT session. None of the variables on sex, age, duration of illness, number of hospitalizations, inpatient diagnosis, seizure duration, seizure threshold, post-ictal suppression time or ECT treatment outcome was predicted by any variable recorded. This outcome seems surprising but we thought that may be related to the shyness of the clinicians who indicated ECT for patients. One of our limitations of the study is the emergence of uncontrollable properties for inclusion criteria because of the naturalistic pattern of research. Not controlling these variables may caused the improvement of the clinical and cognitive parameters to vanish. Our study should be confirmed by the data coming from further controlled studies; also, our study has features to lead and shed light to future studies. We suggest that our results were influenced by the fact that our study group was a “good treatment response group”. Patients who were predicted to be high-level responders were selected and, indeed, good responses were noted after ECT. This reflects defensive selection of patients for ECT because of the criticisms levelled against application of such invasive treatment. As all centres effected by the negative propaganda of ECT, our centre has an attitude for this treatment as well. Thus, it may be that more specified, selected patients were picked up for our study. This high treatment response may be under a clinical bias that the clinicians were blind to the process. Few studies include large sample sizes; this renders it difficult to gauge the true effect of daily ECT use. Also, we suggest that maintenance ECT and combination therapy featuring ECT are less preferable applications of ECT. Further valuable findings will be derived in future when large patient samples are examined and new applications of ECT will become apparent.
Conflict of interest None.
References 1. Goswami U, Kumar U, Singh B. Efficacy of electroconvulsive therapy in treatment resistant schizophrenia: A double-blind study. Indian J Psychiatry 2003;45:26–9. 2. Perera TD, Luber B, Nobler MS, Prudic J, Anderson C, Sacheim AH. Seizure expression during electroconvulsive therapy: Relationships with clinical outcome and cognitive side effects. Neuropsychopharmacology 2004;29:813–25. 3. Altınyazar V. Yüksel N. Combination therapies in schizophrenia. Clin Psychopharmachology Bull 2011;21:368–80 (in Turkish). NORD J PSYCHIATRY·VOL 69 NO 6·2015
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
EEG PREDICTION IN SCHIZOPHRENIC PATIENTS UNDERGOING ECT 4. Griskova I, Dapsys K, Andruskevicius S, Ruksenas O. Does electroconvulsive therapy affect cognitive components of auditory evoked P300? Acta Neurobiol Exp 2005;65:73–7. 5. Mayur P. Ictal electroencephalographic characteristics during electroconvulsive therapy: A review of determination and clinical relevance. J ECT 2006;22:213–17. 6. Azuma H, Fujita A, Sato K, Arahata K, Otsuki K, Hori M et. al. Postictal suppression correlates with therapeutic efficacy for depression in bilateral sine and pulse wave electroconvulsive therapy. Psychiatry Clin Neurosci 2007;61:168–73. 7. Krystal AD, Coffey CE, Weiner RD, Holsinger T. Changes in seizure threshold over the course of electroconvulsive therapeutic response and are detected by ictal EEG ratings. J Neuropsychiatry Clin Neurosci 1998;10:178–86. 8. Overall JE, Gorham DR. The Brief Psychiatric Rating Scale. Psychol Rep 1962;10:790–812. 9. Soykan C. Institutional differences and case typicality as related to diagnosis system severity, prognosis and treatment. Master’s thesis, Ankara: Middle East Technical University, Psychology Department; 1989 (in Turkish). 10. Dubois B, Slachevsky A, Litvan I, Pillon B. The FAB: A Frontal Assessment Battery at bedside. Neurology 2000;55:1621–6. 11. Güleç H, Kavakçı Ö, Güleç MY. Psychometrical properties of adaptation of frontal assessment battery for schizophrenia patients. Düşünen Adam 2007;20:151–7 (in Turkish). 12. Turvey CL, Coryell WH, Solomon DA, Leon AC, Long term prognosis of bipolar-I disorder. Acta Psychiatr Scand 1999;99:110–9. 13. Scott AF. Electroconvulsive therapy, practice and evidence. Brit J Psychiat 2010;196:171–2.
NORD J PSYCHIATRY·VOL 69 NO 6·2015
14. Krystal AD, Holsinger T, Weiner RD et al. Prediction of the utility of a switch from unilateral to bilateral ECT in the elderly using treatment 2 ictal EEG measures. J ECT 2000;16: 327–37. 15. Chanpattana W, Chakrabhand S. Combined ECT and neuroleptic therapy in treatment-refractory schizophrenia: Prediction of outcome. Psychiatry Res 2001;105:107–15. 16. Goswami U, Kumar U, Singh B. Efficacy of electroconvulsive therapy in treatment resistant schizophrenia: A double-blind study. Indian J Psychiatry 2003;45:26–9. 17. Chanpattana W, Kramer BA. Acute and maintenance ECT with flupenthixol in refractory schizophrenia: Sustained improvements in psychopathology, quality of life, and social outcomes. Schizophr Res 2003;63:189–93. Gulnihal Gokce Simsek, M.D., Psychiatry Department, Yozgat State Hospital, Yozgat, Turkey. Selma Zincir, M.D., Psychiatry Department, Erenkoy Mental Health Research and Training Hospital, Istanbul, Turkey. Huseyin Gulec, M.D., Psychiatry Department, Erenkoy Mental Health Research and Training Hospital, Istanbul, Turkey. Sevgin Eksioglu, M.D., Psychiatry Department, Siverek State Hospital, Siverek, Turkey. Umit Basar Semiz, M.D., Psychiatry Department, Mugla University, Mugla, Turkey. Yasemin Sipka Kurtulmus, M.D., Neurology Department, Erenkoy Mental Health Research and Training Hospital, Istanbul, Turkey.
471
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
GULNIHAL GOKCE SIMSEK, SELMA ZINCIR, HUSEYIN GULEC, SEVGIN EKSIOGLU, UMIT BASAR SEMIZ, YASEMIN SIPKA KURTULMUS
Simsek GG, Zincir S, Gulec H, Eksioglu S, Semiz UB, Kurtulmus YS. Do ictal EEG characteristics predict treatment outcomes in schizophrenic patients undergoing electroconvulsive therapy? Nord J Psychiatry 2015;69:466–471. Objective: The aim of this study is to investigate the relationship between features of electroencephalography (EEG), including seizure time, energy threshold level and post-ictal suppression time, and clinical variables, including treatment outcomes and side-effects, among schizophrenia inpatients undergoing electroconvulsive therapy (ECT). Method: This is a naturalistic follow-up study on schizophrenia patients, diagnosed using DSM-IV-TR criteria, treated by a psychosis inpatient service. All participants completed the Brief Psychiatric Rating Scale (BPRS), the Global Assessment of Functioning (GAF) scale, the Frontal Assessment Battery (FAB) and a Data Collection Form. Assessments were made before treatment, during ECT and after treatment. Findings: Statistically significant improvements in both clinical and cognitive outcome were noted after ECT in all patients. Predictors of improvement were sought by evaluating electrophysiological variables measured at three time points (after the third, fifth and seventh ECT sessions). Logistic regression analysis showed that clinical outcome/ improvement did not differ by seizure duration, threshold energy level or post-ictal suppression time. Discussion and conclusion: We found that ictal EEG parameters measured at several ECT sessions did not predict clinical recovery/outcomes. This may be because our centre defensively engages in “very specific patient selection” when ECT is contemplated. ECT does not cause short-term cognitive functional impairment and indeed improves cognition, because symptoms of the schizophrenic episode are alleviated. • Electroconvulsive therapy, Ictal, Post-ictal suppression time, Schizophrenia, Side-effect, Treatment outcome. Gulnihal Gokce Simsek, M.D., Psychiatry Department, Yozgat State Hospital, Yozgat, Turkey, E-mail: [email protected]; Accepted 16 December 2014.
A
ntipsychotic drugs are considered optimal first-line treatments for schizophrenia. Even when appropriate treatments protocols are applied, 25% of patients demonstrate inadequate or no improvement in illness progression (1). Patients may have severe symptoms, experience long periods of hospitalization and high rates of lifelong dysfunction, including over the working years. Treatment drop-out creates a vicious circle in the schizophrenic population over time. Although electroconvulsive therapy (ECT) use features in some treatment algorithms for schizophrenia, the technique is not often applied except in Europe and in many developing countries (2). Affective symptoms, acute initiation and a short duration of schizophrenia are indicators that ECT may be useful. Some authors have stated
© 2015 Informa Healthcare
that ECT may be used to effectively treat acute schizophrenia of Type 1, but not Type 2 (chronic schizophrenia), unless depression is also present (3). A recent meta-analysis showed that a combination of ECT with antipsychotic medication was superior to antipsychotic monotherapy alone, rapidly decreasing global symptoms. Even when no further maintenance ECT sessions were given, relapse prevention was more effective when combination therapy was applied (4). Recent studies have highlighted the benefits afforded by ECT as a component of schizophrenia treatment. Useful long-term effects were evident, especially in combination with antipsychotic drugs (15–17). Only a few reports have addressed ECT therapeutic efficacy and variables affecting treatment outcomes in DOI: 10.3109/08039488.2014.1003403
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
EEG PREDICTION IN SCHIZOPHRENIC PATIENTS UNDERGOING ECT
schizophrenia patients. Such studies play an important role in identifying preventative steps reducing cognitive side-effects and markers of clinical progression. The advantage of using electroencephalography (EEG) during ECT sessions is not limited to simple determination of seizure duration. Many ictal EEG markers can be used to assess and predict treatment outcomes. Some such markers are post-ictal suppression time, the extent of slowing in EEG waves, the time of initiation of slowing, global EEG power, the threshold energy level and seizure time. Recent studies have emphasized the significance of seizure time, threshold energy level and post-ictal suppression time; cognitive side-effects of ECT have been noted in recent works (5–7). Post-ictal suppression time has received most attention (5). We investigated the efficacy and side-effects of daily applications of ECT. Our hypotheses were, first, that treatment outcomes in schizophrenia patients receiving ECT could be predicted by peri-ictal properties, including motor seizure time, threshold electrical energy and post-ictal suppression time; and, second, that cognitive functions would be preserved short-term after ECT sessions concluded. Although resistance to treatment remains a serious clinical problem, we wish to highlight the fact that ECT can appropriately contribute to a solution of this problem. Also, we were concerned to ensure the more effective use of ECT in daily practice and to create a basis allowing clinicians of the future to effectively employ ECT.
Materials and methods Subjects This naturalistic clinical follow-up study was performed in the Erenköy Mental Health Research and Training Hospital, on schizophrenia inpatients, in the interval August 2011– January 2012. All patients scheduled for ECT by a study-blind treatment team were reviewed. A total of 61 volunteer patients diagnosed using DSM-IV criteria and aged between 18 and 65 years participated. Patients with comorbidities affecting cognitive assessment, including another psychiatric/medical disorder, mental retardation, a history of head trauma or a history of drug abuse in the prior 12 months, were excluded. All patients were assessed at the beginning of the study; at the third, fifth and seventh ECT sessions; and 2–7 days after conclusion of ECT treatment. All participants signed informed consent forms before study commencement. Ethical approval was given by the Erenköy Mental Health Research and Training Hospital Institutional Review Board.
Concurrent medications All patients took psychopharmacological medications during the study. Of the 61 patients, 44 were taking a daily NORD J PSYCHIATRY·VOL 69 NO 6·2015
average of 19 mg haloperidol, 11 were taking 4.5 mg/day risperidone, 10 were taking 20 mg/day olanzapin, 10 were taking 43 mg/2 weeks risperidon LAI, nine were taking 82 mg/day clozapine, six were taking 350 mg/day quetiapine, three were taking 666 mg/day amisulpiride and one was taking 9 mg/day paliperidone.
Assessment tools Our primary measurement tool was the Brief Psychiatric Rating Scale (BPRS). The Global Assessment of Functioning (GAF) test was also used. Patients were given the Frontal Assessment Battery (FAB) test before and 2–7 days after ECT sessions concluded, to monitor cognitive functions and alterations therein. Interviews were conducted face-to-face with both patients and caregivers (who gave informed consent to be interviewed). We collected information on patient age, sex, educational level, marital status, family medical history, history of physical disease, age at onset of illness, illness duration, attempted suicide history, number of acute exacerbations of illness, treatment history, number of hospitalizations, recent ECT history and efficacy thereof, total number of ECT sessions, ECT session durations, ictal EEG parameters, anaesthetic dosage, average energy threshold, and scores on all assessment scales. The BPRS was developed by Overall & Gorham in 1962 (8) and modified later by Bech et al., with a reduction in the number of questions to 18. Each item is scored from 0 to 6 or from 1 to 7. The maximum total score is 72. We translated the scale into Turkish. No studies on validity or reliability were performed because intercultural usage of the scale is compromised, in epidemiological research, in terms of validity and reliability. Psychiatric symptoms were assessed by simple interview. Scale items explore symptoms of psychosis, depression and anxiety. The instrument records somatic complaints, mental anxiety, emotional withdrawal, thought disorganization, feelings of guilt and self-vilification, somatic anxiety, the presence of specific motor disorders, exaggerated self-esteem, depressive mood, hostility, scepticism, hallucinations, psychomotor retardation, refusal to co-operate, bizarre thought content, blunted and inappropriate affect, psychomotor agitation and disorientation/confusion. Dubois et al. (10) developed the FAB to assess six domains associated with frontal lobe-linked functions, striatofrontal functional disorders, movement disorders and frontal lobe damage. The short battery can be applied at the bedside to examine executive functioning. Answers are scored from 0 to 3 and the range of scores is from 0 to 16. Interview time is ∼10 min. The validity and reliability of the Turkish version have been demonstrated by Güleç et al. (11). The GAF scale was developed by Spitzer et al. to assess individual global functioning from a healthy stage to manifestation of a psychiatric disorder. The test is
467
GG SIMSEK ET AL.
relatively easy to administer, specific and sensitive, and finds applications in both clinical practice and research.
EEG device Galileo, Neoro SpA, Firenze, Italy (2004, model no. SN20325).
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
ECT technique All patients were given ECT three times a week. After preparation for ECT, bitemporal leads were applied and frequency and energy dosage were measured on the ECT machine using a constant current and a brief pulse stimulus (Thymatron System IV, Somatics Inc.; two-way square-pulse wave with four monitoring points [EEG1, EEG2, EKG and EMG]). The anaesthetic used is 0.5 mg/ kg propofol and for adequate muscular relaxation 0.5 mg/ kg succinylcholine. The initial energy dosage was calculated using the “half-age” formula. Seizure periods of 20 s and above were considered adequate and, if such times were not achieved, the dosage was titrated upwards and charge was increased by 50% (14). The cut-off point was set at 279 s, that of the median value of the mean total seizure time on ECT. Values below the cut-off point were termed “short ECT” and higher values “long ECT”. We used these criteria because of a lack of consensus on appropriate measures in the literature. Use of the median value seemed appropriate. The median ECT energy threshold was 37% and this was set as the cut-off point. Values below this level were termed “low threshold” and higher values “high threshold”. We used these criteria because of a lack of consensus on appropriate measures in the literature. Use of the median value seemed appropriate. Basal and peri-ictal EEG records were assessed by an experienced neurologist (YSK), who measured EEG seizure time, seizure frequency and post-ictal suppression time (the inactive period after termination of a seizure that was not an artifact) (7). All data from all ECT sessions undergone by all patients were examined and all information was included in the analysis.
Statistical analysis A statistical software package was used to evaluate data. Descriptive statistical methods (frequency, percentage, median and standard deviation) are used to present demographic and clinical information. The Mann–Whitney U test was used for correlation analysis of quantitative data from two subgroups because data distribution was not normal. If more than two groups were compared, we used the Kruskal–Wallis test, or the Mann–Whitney U test with Bonferroni correction, to seek significant correlations. Wilcoxon’s signed-rank
468
test was used when correlating data within groups. Spearman’s correlation analysis was used to investigate relationships between quantitative data. The 95% confidence intervals were calculated; P ⬍ 0.05 was regarded as significant and P ⬍ 0.01 highly significant.
Results Of all patients, 26.2% were female and 73.8% male; 18.0% worked; 4.9% were retired; 4.9% were pensions because of illness; 72.1% unemployed; 75.4% single; 11.5% married; 9.8% divorced; and 3.3% widowed. Mean patient age was 33.3 ⫾ 10.9 years; mean BMI 24.2 ⫾ 3.5 kg/m2; and mean years of education 9.4 ⫾ 3.5. Demographic data are shown in Table 1. The mean age onset of illness was 23.6 ⫾ 6.8 years, mean illness duration 9.6 ⫾ 7.4 years, mean number of exacerbations 5.6 ⫾ 5.7 and mean number of hospitalizations 4.4 ⫾ 5.1. The mean total number of ECT sessions was 6.8 ⫾ 1.2, the mean total duration of ECT 309.0 ⫾ 124.1 s and the mean doses of succinylcholine and propophole 36.0 ⫾ 9.5 and 44.3 ⫾ 15.8 mg. The mean energy threshold was 37.4 ⫾ 9.8%. Of all patients, 29.5% had treatment-resistant schizophrenia and the remaining 70.5% were inpatients with acute exacerbations of schizophrenia. Of all patients, 44.3% had histories of suicide attempts, 11.5% physical illnesses and 41.0% family histories of psychiatric disorders. Correlations between clinical features and EEG characteristics (age at commencement of illness and mean energy threshold; illness duration and mean energy threshold, post-ictal suppression time) were positive (Table 2). The cut-off point was set at 279 s, that of the median value of the mean total seizure time on ECT. Values below the cut-off point were termed “short ECT” and higher values “long ECT”. Comparisons of CGI, GAF and FAB scores before and after treatment revealed no significant differences in patients who underwent short and long ECT. The median ECT energy threshold was 37%; this was set as the cut-off point. Values below this level were termed “low threshold” and higher values “high threshold”. Comparisons of CGI, GAF and FAB scores before and after treatment revealed no significant differences in patients who underwent “low threshold” and “high threshold ECT. Logistic regression analysis of data from the third, fifth and seventh ECT sessions was performed to predict treatment outcomes (a decrease of 50% or more in the BPRS score) of ictal EEG characteristics including seizure duration, seizure energy threshold and post-ictal suppression time (Table 3). We also analysed clinical features. NORD J PSYCHIATRY·VOL 69 NO 6·2015
EEG PREDICTION IN SCHIZOPHRENIC PATIENTS UNDERGOING ECT
Table 1. Demographic features (n=61). Mean⫾ standard deviation 33.3 ⫾ 10.9 24.2 ⫾ 3.5 9.4 ⫾ 3.5
Age (years) BMI (kg/m2) Educational status (years)
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
n Sex Female Male Work status Working Retired Invalid Not working Marital status Single Married Divorced Widowed
%
16 45
26.2 73.8
11 3 3 44
18.0 4.9 4.9 72.1
46 7 6 2
75.4 11.5 9.8 3.3
Features of illness (n ⫽ 61)
Mean⫾ standard deviation
Age at commencement of illness (year) Duration of illness (years) Number of exacerbations Number of hospitalizations Total number of ECT sessions Total ECT session duration (seconds) Mean succinylcholine dosage (mg) Mean propofol dosage (mg) ECT threshold (%)
23.6 ⫾ 6.8 9.6 ⫾ 7.4 5.6 ⫾ 5.7 4.4 ⫾ 5.1 6.8 ⫾ 1.2 309.0 ⫾ 124.1 36.0 ⫾ 9.5 44.3 ⫾ 15.8 37.4 ⫾ 9.8
Attempted suicide history Yes History of physical illness Yes Diagnosis Treatment-resistant Acute exacerbation Family history of the disease Yes Treatment history Medical treatment Medical treatment⫹ ECT Recent ECT response Yes
Age at commencement of illness R P Duration of illness R P Number of hospitalizations R P
n
%
27
44.3
7
11.5
18 43
29.5 70.5
25
41.0
41 20
67.2 32.8
17
94.4
Discussion We investigated the relationships between EEG features, including seizure time, energy threshold level and post-ictal suppression time and clinical variables including treatment outcomes and side-effects in schizophrenia inpatients given ECT. We explored whether ECT outcomes could be predicted by peri-ictal EEG
Seizure duration
Threshold
Post-ictal
⫺ 0.80 0.541
0.274 0.032
0.025 0.846
0.195 0.132
0.289 0.024
0.311 0.015
0.128 0.327
0.228 0.078
0.108 0.408
characteristics including post-ictal suppression time, duration of motor seizure and electrical threshold and also searched for relationships between cognitive sideeffects of ECT and EEG characteristics. Table 3. Predictors of outcomes after three (n ⫽ 10), five (n ⫽ 47) and seven (n ⫽ 57) electroconvulsive therapy (ECT) treatments [demographic and clinical features and electroencephalography (EEG) characteristics]. B
BMI, body mass index; ECT, electroconvulsive therapy.
NORD J PSYCHIATRY·VOL 69 NO 6·2015
Table 2. Relationship between electroencephalography (EEG) characteristics and clinical features (n=61).
3 ECT Sex (1) Age Duration of illness Hospitalizations Axis 1 diagnosis (1) Threshold Mean seizure duration Mean post-ictal Constant 5 ECT Sex (1) Age Duration of illness Hospitalizations Axis 1 diagnosis (1) Threshold Mean seizure duration Mean post-ictal Constant 7 ECT Sex (1) Age Duration of illness Hospitalizations Axis 1 diagnosis (1) Threshold Mean seizure duration Mean post-ictal Constant
Sig.
Exp(B)
1 1 1 1 1
0.984 0.473 0.254 0.377 0.585
0.977 1.073 0.863 1.104 1.695
2.872 3.049
1 1
0.090 0.081
0.862 0.956
⫺ 0.543 0.554 0.961 5.763 3.548 2.638
1 1
0.327 0.104
0.581 318.145
1.018 0.040 ⫺ 0.099 ⫺ 0.063 0.258
1 1 1 1 1
0.197 0.529 0.231 0.388 0.749
2.767 1.041 0.905 0.939 1.295
0.200 0.710
1 1
0.654 0.399
0.977 0.984
⫺ 0.595 0.367 2.632 5.351 2.281 5.505
1 1
0.105 0.019
0.551 210.816
15.41 1.716 ⫺ 1.56 ⫺ 0.513 0.545
1 1 1 1 1
0.100 4,899,397.5 0.163 5.561 0.132 0.210 0.292 0.599 0.853 1.724
2.102 2.314
1 1
0.147 0.128
0.445 0.789
0.712 0.669
1 1
0.399 0.413
2.531 2,464.485
⫺ 0.023 0.071 ⫺ 0.148 0.099 0.528
S.E.
1.175 0.000 0.098 0.516 0.130 1.299 0.112 0.781 0.967 0.298
⫺ 0.149 0.088 ⫺ 0.045 0.026
0.789 1.663 0.064 0.396 0.083 1.434 0.073 0.744 0.809 0.102
⫺ 0.023 0.051 ⫺ 0.016 0.019
9.377 2.699 1.231 1.942 1.037 2.264 0.487 1.110 2.945 0.034
⫺ 0.810 0.559 ⫺ 0.237 0.156 0.928 7.810
Wald df
1.100 9.547
469
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
GG SIMSEK ET AL.
Of all patients, 29.5% had treatment-resistant schizophrenia and the remaining 70.5% were inpatients with acute exacerbations of schizophrenia or similar. We noted statistically significant improvements and decreases in the severity of illness, after ECT, as reported recently (13). Not only did cognitive functioning not deteriorate but it actually improved from the initial state. Statistically significant improvements were noted on all three scales (BPRS, GAF and FAB). Correlations between clinical features and EEG characteristics were sought and the work revealed that age at initiation of illness and energy threshold; duration of illness and mean energy threshold level; and duration of illness and post-ictal suppression time were all positively correlated. However, no statistically significant difference was evident between pre- and post-treatment FAB scores, indicating that no cognitive side-effects developed. No statistically significant relationships were evident between pre- and post-treatment FAB scores and ictal EEG findings. Similar findings have been reported in studies on patients with major depression and schizophrenia (13). Predictor EEG characteristics for changes in cognitive status after ECT are not discussed in this article. Cognitive status improvement was indeed statistically significant, but this was the reverse of was expected in literature when pre- and post-treatment data were compared. Therefore, a decrease in cognitive status, and development of an attention deficit, are part of the natural progression of illness, especially during severely symptomatic periods. When the duration of such periods was reduced, and patients improved, cognitive status also improved significantly after conclusion of ECT treatment. Further follow-up treatment with assessment of cognition would be valuable. The ECT and ictal EEG phases are, respectively: rapid low-frequency low-amplitude initiation, spikepolyspike and high-amplitude tonic phases, spike and slow-wave high- amplitude clonic phases, a phase of slowing of wave amplitude and frequency and the postictal suppression phase. It has been suggested that seizures exhibiting intensive inhibition of these processes are associated with more useful ECT sessions and better outcomes (2). Recent studies have shown that seizure length is not a useful marker of seizure adequacy. Of the various methods used to measure seizure effects, the length of the post-ictal suppression period is the best predictor of treatment outcomes after ECT. The post-ictal suppression time is also a useful prognostic marker (5). High-level EEG suppression was found to be associated with clinical improvement when quantitative post-ictal suppression time was measured in a study on patients with depression (2). In our present study, the ictal EEG parameters (seizure duration, seizure threshold and postictal suppression time) did not differ among ECT sessions, and were not associated with the decreases in
470
BPRS scores noted after each session. In addition, we explored how all measured EEG parameters predicted treatment outcomes. No EEG parameter measured at any of the third, fifth or seventh ECT treatments was a predictor of outcomes in our study. Of all patients, 57 improved after the seventh ECT, 47 after the fifth ECT and 10 after the third ECT session. None of the variables on sex, age, duration of illness, number of hospitalizations, inpatient diagnosis, seizure duration, seizure threshold, post-ictal suppression time or ECT treatment outcome was predicted by any variable recorded. This outcome seems surprising but we thought that may be related to the shyness of the clinicians who indicated ECT for patients. One of our limitations of the study is the emergence of uncontrollable properties for inclusion criteria because of the naturalistic pattern of research. Not controlling these variables may caused the improvement of the clinical and cognitive parameters to vanish. Our study should be confirmed by the data coming from further controlled studies; also, our study has features to lead and shed light to future studies. We suggest that our results were influenced by the fact that our study group was a “good treatment response group”. Patients who were predicted to be high-level responders were selected and, indeed, good responses were noted after ECT. This reflects defensive selection of patients for ECT because of the criticisms levelled against application of such invasive treatment. As all centres effected by the negative propaganda of ECT, our centre has an attitude for this treatment as well. Thus, it may be that more specified, selected patients were picked up for our study. This high treatment response may be under a clinical bias that the clinicians were blind to the process. Few studies include large sample sizes; this renders it difficult to gauge the true effect of daily ECT use. Also, we suggest that maintenance ECT and combination therapy featuring ECT are less preferable applications of ECT. Further valuable findings will be derived in future when large patient samples are examined and new applications of ECT will become apparent.
Conflict of interest None.
References 1. Goswami U, Kumar U, Singh B. Efficacy of electroconvulsive therapy in treatment resistant schizophrenia: A double-blind study. Indian J Psychiatry 2003;45:26–9. 2. Perera TD, Luber B, Nobler MS, Prudic J, Anderson C, Sacheim AH. Seizure expression during electroconvulsive therapy: Relationships with clinical outcome and cognitive side effects. Neuropsychopharmacology 2004;29:813–25. 3. Altınyazar V. Yüksel N. Combination therapies in schizophrenia. Clin Psychopharmachology Bull 2011;21:368–80 (in Turkish). NORD J PSYCHIATRY·VOL 69 NO 6·2015
Nord J Psychiatry Downloaded from informahealthcare.com by Nyu Medical Center on 06/20/15 For personal use only.
EEG PREDICTION IN SCHIZOPHRENIC PATIENTS UNDERGOING ECT 4. Griskova I, Dapsys K, Andruskevicius S, Ruksenas O. Does electroconvulsive therapy affect cognitive components of auditory evoked P300? Acta Neurobiol Exp 2005;65:73–7. 5. Mayur P. Ictal electroencephalographic characteristics during electroconvulsive therapy: A review of determination and clinical relevance. J ECT 2006;22:213–17. 6. Azuma H, Fujita A, Sato K, Arahata K, Otsuki K, Hori M et. al. Postictal suppression correlates with therapeutic efficacy for depression in bilateral sine and pulse wave electroconvulsive therapy. Psychiatry Clin Neurosci 2007;61:168–73. 7. Krystal AD, Coffey CE, Weiner RD, Holsinger T. Changes in seizure threshold over the course of electroconvulsive therapeutic response and are detected by ictal EEG ratings. J Neuropsychiatry Clin Neurosci 1998;10:178–86. 8. Overall JE, Gorham DR. The Brief Psychiatric Rating Scale. Psychol Rep 1962;10:790–812. 9. Soykan C. Institutional differences and case typicality as related to diagnosis system severity, prognosis and treatment. Master’s thesis, Ankara: Middle East Technical University, Psychology Department; 1989 (in Turkish). 10. Dubois B, Slachevsky A, Litvan I, Pillon B. The FAB: A Frontal Assessment Battery at bedside. Neurology 2000;55:1621–6. 11. Güleç H, Kavakçı Ö, Güleç MY. Psychometrical properties of adaptation of frontal assessment battery for schizophrenia patients. Düşünen Adam 2007;20:151–7 (in Turkish). 12. Turvey CL, Coryell WH, Solomon DA, Leon AC, Long term prognosis of bipolar-I disorder. Acta Psychiatr Scand 1999;99:110–9. 13. Scott AF. Electroconvulsive therapy, practice and evidence. Brit J Psychiat 2010;196:171–2.
NORD J PSYCHIATRY·VOL 69 NO 6·2015
14. Krystal AD, Holsinger T, Weiner RD et al. Prediction of the utility of a switch from unilateral to bilateral ECT in the elderly using treatment 2 ictal EEG measures. J ECT 2000;16: 327–37. 15. Chanpattana W, Chakrabhand S. Combined ECT and neuroleptic therapy in treatment-refractory schizophrenia: Prediction of outcome. Psychiatry Res 2001;105:107–15. 16. Goswami U, Kumar U, Singh B. Efficacy of electroconvulsive therapy in treatment resistant schizophrenia: A double-blind study. Indian J Psychiatry 2003;45:26–9. 17. Chanpattana W, Kramer BA. Acute and maintenance ECT with flupenthixol in refractory schizophrenia: Sustained improvements in psychopathology, quality of life, and social outcomes. Schizophr Res 2003;63:189–93. Gulnihal Gokce Simsek, M.D., Psychiatry Department, Yozgat State Hospital, Yozgat, Turkey. Selma Zincir, M.D., Psychiatry Department, Erenkoy Mental Health Research and Training Hospital, Istanbul, Turkey. Huseyin Gulec, M.D., Psychiatry Department, Erenkoy Mental Health Research and Training Hospital, Istanbul, Turkey. Sevgin Eksioglu, M.D., Psychiatry Department, Siverek State Hospital, Siverek, Turkey. Umit Basar Semiz, M.D., Psychiatry Department, Mugla University, Mugla, Turkey. Yasemin Sipka Kurtulmus, M.D., Neurology Department, Erenkoy Mental Health Research and Training Hospital, Istanbul, Turkey.
471
