Clinical Effectiveness and Tolerability of Electroconvulsive Therapy in Patients with Neuropsychiatric Symptoms of Dementia

Abstract

Background: Dementia frequently presents with aggression, agitation, and disorganized behavior for which current treatment is partially effective and is associated with significant adverse effects.

Objective: The aim of this study was to retrospectively assess the clinical effectiveness and tolerability of electroconvulsive therapy (ECT) in a sample of patients with neuropsychiatric symptoms of dementia (NPS) and to explore factors associated with response and with cognitive adverse effects.

Methods: We examined the clinical records of 25 patients with dementia and a pre-existing psychiatric disorder treated with ECT at an academic mental health hospital between April 1, 2010 and January 28, 2016. Twenty-nine acute ECT courses and fifteen maintenance courses were reviewed. We assessed treatment effectiveness and cognitive adverse effects as well as factors associated with response to treatment, including pre-existing psychiatric disorders, concomitant pharmacological treatment and types of dementia.

Results: ECT resulted in a clinically meaningful response in 72% of acute treatment courses. Cognitive adverse effects affecting functioning were reported in 7% of the acute treatment courses. Maintenance treatment was effective in sustaining the response in 87% of treatment courses with two reports of significant cognitive adverse effects. One patient fell and experienced a hip fracture a day after treatment. Use of antipsychotic or antidepressant medications, pre-existing psychiatric disorder, or gender were not associated with response.

Conclusion: This study shows meaningful clinical effectiveness and good tolerability of ECT in patients with severe NPS of dementia. Furthermore, maintenance ECT was effective in sustaining treatment response.

Keywords

Alzheimer’s disease dementia electroconvulsive therapy neuropsychiatric symptoms

INTRODUCTION

Neuropsychiatric symptoms (NPS) are commonly present in patients with dementia and are associated with negative outcomes including considerable distress to patients, heavy burden on caregivers, and institutional placement [1, 2]. In long-term care facilities, over 90% of residents with dementia exhibit at least one behavioral disturbance: 82% demonstrate aggression, 60% psychosis, and 42% depressed mood [3]. Psychopharmacology is broadly used to treat NPS despite the fact that no medication is US Food and Drug Administration (FDA) approved for this indication and many of the medications have only modest efficacy for some of the symptoms [4, 5]. In one study, Selbaek et al. found that 75% of nursing home residents with dementia are prescribed psychotropic medications [6]. Similar results were reported in a large, recent, prescription dispensing study [7]. This Canadian study also showed a trend of decrease in antipsychotic and an increase in antidepressant medication prescriptions over the last years. Atypical antipsychotics are the most commonly used medications to treat NPS. Ballard and Corbett summarized 18 randomized controlled trials evaluating the impact of 6–12-week treatment with atypical antipsychotics [4]. They found a significant benefit mainly in the treatment of aggression but rather a more limited benefit for other behavioral and psychological symptoms, while no proven benefit was found for treatment given over longer than a 12-week period. Moreover, significant adverse effects are associated with the use of atypical antipsychotics in this population including sedation, extrapyramidal symptoms, gait disturbances, falls, cerebrovascular events, and death [4, 8]. A significant increase in mortality risk has resulted in a formal FDA black box warning against the use of atypical and typical anti-psychotics for dementia-related psychosis or behavioral disorders [9]. Individually tailored non-pharmacological interventions such as dementia-care mapping and person-centered care can be effective, mainly in less severe cases of NPS [10, 11]. Electroconvulsive therapy (ECT) is the most effective treatment for treatment resistant depression and has been shown to be safe and effective in geriatric patients with and without dementia [12 –15]. Hypotheses on the potential of ECT to treat refractory NPS have been put forward [16], however, there is relative paucity of data on the use of ECT for patients with NPS associated withdementia.

One prospective open study and several retrospective chart reviews investigated the safety and effectiveness of ECT for the treatment of NPS with and without pre-existing psychiatric comorbidity. Acharya et al. studied 23 patients with dementia-related agitation and/or aggression treated with right unilateral brief-pulse ECT. Treatments were well tolerated and there was significant improvement in agitation as assessed by both the Cohen-Mansfield agitation inventory (short form) and the neuropsychiatric inventory. The clinical global impression average score improved form markedly agitated/aggressive to borderline agitated/aggressive. In ten patients who were able to complete the Mini-Mental State Examination at both baseline and discharge, the average score showed modest improvement from 8.2 to 10.1 [17]. A previous chart review by the same group (n = 16) in patients with NPS related to dementia treated with bilateral ECT showed similar results; significant reduction in agitation and aggression and a decrease in the clinical global impression scale from 6 to 2.1. Eight patients showed transient postictal confusion and two showed severe confusion that required modification of treatment [18]. An earlier case series of 11 patients with Alzheimer’s dementia (AD) related agitation, treated mostly by bilateral ECT, found that 9 out of 11 patients had an improvement in their symptoms (most of them reaching remission), with a significant decrease in hospitalizations over the long term [19]. Six of the 9 responders went on to receive maintenance ECT over the subsequent year; however, the details of treatment frequency and specific outcomes were not reported.

In the present study, we retrospectively assessed the clinical effectiveness and tolerability of both acute and maintenance ECT treatment in a sample of patients with dementia and associated NPS. We also explored factors that can potentially affect response or cognitive adverse effects.

METHODS

Subjects

The study was conducted at the Centre for Addiction and Mental Health (CAMH), a large academic mental health hospital in Toronto, Canada. A retrospective chart review was conducted for all patients with the diagnosis of dementia and NPS treated with ECT between April 1, 2010 and January 28, 2016. The study was approved by the CAMH research ethics board. Patient electronic health records were reviewed for the following: diagnoses, concomitant pharmacological treatment, response to ECT and adverse effects, ECT parameters for each treatment, clinician global improvement (CGI) rating at the end of each treatment course, and a clinician global rating for cognitive adverse effects.

ECT practice

The law in Ontario, Canada states that if a person is incapable with respect to a treatment, consent may be given or refused on his or her behalf by a valid substitute decision maker (i.e., spouse, child, or guardian). There are no specific restrictions regarding consent to ECT treatment. Since most of the patients with dementia were incapable of consent, consent of a substitute decision maker was sought prior to treatment. A MECTA spECTrum 5000Q ECT machine was used with amplitude fixed at 800 mA for all treatments. For bitemporal placement (BT), 1.0 msec pulse width was used, and for right unilateral ultra-brief pulse (RUL-UB), the pulse width was 0.3 msec. Stimulus titration method was used to find the individual seizure threshold. An adequate seizure was defined as a coherent and regular seizure lasting≥15 s (on 2 channel EEG and by motor response). The intensity was set to 1.5 times the seizure threshold for BT treatment and to 6 times the seizure threshold for RUL-UB treatment. Methohexital 0.75–1.0 mg/kg was the anesthetic agent and succinylcholine 0.4–0.8 mg/kg was used for muscle relaxation. For hypertension, IV labetalol was used as needed and IV granisetron or odansetron were used in cases of severe treatment relatednausea.

Acute ECT treatment

Acute treatments were administered 2–3 times a week until no additional response could be achieved, usually over the course of 4–5 weeks.

Maintenance ECT treatment

In general, after completing an acute course with a significant response, the approach was to initiate a maintenance treatment using a taper-down pattern as follows: weekly treatments for 4 weeks, bi-weekly treatments for 8 weeks, and then monthly treatments. In some cases, mainly due to inability to attend outpatient treatment or as a result of patients’ or family preference, patients did not proceed to maintenance treatment despite a robust response to the acute course.

Assessment of treatment response

Clinical Note Clinical Global Improvement (CN-CGI)

Response to ECT was evaluated from the electronic chart using a 4-points scale similar to the scale described by Kristensen et al. [20]. This scale corresponds well with the first 4 items from the original clinical global impressions-improvement scale (i.e., Very much improved, Much improved, Minimally improved, No change) [21] and we have referred to this as the clinical note CGI (CN-CGI). This methodology was applied to both the acute and the maintenance courses referring to the pre-acute state as the baseline for comparison. The CN-CGI was determined as follows:

Excellent: The patient’s chart clearly stated that the patient had profited dramatically from the ECT treatment. Examples include resolution or near resolution of target symptoms, rapid discharge after treatment, and referral to maintenance treatment.

Good: The patient’s chart indicated clear good response. Examples include significant reduction in severity of target symptoms, referral to maintenance treatment and statements like “good effect”, “improved mental status”, and “marked improvement”.

Moderate: The patient’s chart indicated that the treatment had some or partial beneficial effect on target symptoms. Examples include statements like “modest response” or “partial response”.

Poor/none: The patient’s chart indicated that the treatment had no or minimal effect on the patient, as assessed by the treating physician.

Patients with a CN-CGI score of 1 or 2 were defined treatment responders while patients scored with CN-CGI of 3 or 4 were defined as non-responders.

Validation of the CN-CGI scoring

At our institute, attending psychiatrists are encouraged to complete the CGI-I scale at the conclusion of ECT treatment course, but this information was only available for 19 of the 29 acute courses and 3 of the 13 maintenance courses. We used this subset of treatment courses to validate the CN-CGI method. The CN-CGI was evaluated by an experienced ECT psychiatrist (MI) and when there was a disagreement with the CGI-I scale, another experienced ECT psychiatrist and geriatric psychiatrist (DMB) reviewed the chart for treatment response and his score was used. When comparing classification of marked response (score of 1 or 2) versus non-response (score of 3 or 4), a good inter-rater agreement was found between the CN-CGI and the CGI-I (kappa of 0.65).

Cognitive impairment

ECT related cognitive impairment was also evaluated from the patients’ charts and rated on a 1 to 4 scale (1-none; 2-mild; 3-moderate; and 4-severe). For the same subset of treatment courses as with the CGI-I data presented above, there was a similar rating for the cognitive impairment by the ECT treating clinician. A rating of 1 or 2 was considered not to be clinically significant, while a rating of 3 or 4 was considered as clinically significant for the purpose of analyses. Here too we compared our chart based rating to the rating in the electronic health record and resolved disagreements as described above. We found a moderate inter-rater agreement between the two scales (kappa of 0.5).

Statistical analysis

We compared treatment response and cognitive impairment rates by various clinical and treatment features. The differences in these features between responders and non-responders were calculated to determine their potential association with treatment outcome. Barnard’s test was used to compare rates of categorical data and logistic regression was used for numerical data. All statistical tests were two-tailed with significance level set at 0.05 and were performed in R (version 3.2.2, R Development Core Team).

RESULTS

Demographics and ECT details

This study included 25 patients who received 29 acute ECT courses and 13 patients (12 of the 24 acute patients and 1 other patient) who received 15 maintenance ECT courses. In most courses, the patients were female (72% and 60% of the acute and the maintenance courses respectively), inpatients (79% and 67%), with a lack of capacity to provide consent for treatment (83% and 80%) and were treated with bitemporal placement (86% and 93%). All of the patients were diagnosed with dementia and severe and resistant NPS and all but two had a previous psychiatric disorder, in most cases, major depressive disorder (MDD). In the majority of the ECT courses, patients had concomitant antipsychotic medications (72% and 80%), many were on antidepressants (41% and 53%), and none were on benzodiazepines≥2 mg lorazepam or equivalent. The average acute course consisted of 12 treatments delivered over 36 days and the average maintenance course consisted of 25 treatments delivered over 429 days(Table 1).

Treatment response

Based on the CN-CGI, 72% of the acute treatment courses and 87% of the maintenance courses resulted in a marked response (i.e., good or excellent). Based on the CGI data (available for 19 of the acute treatment courses) the response rate was 63%. Table 2 summarizes the response frequencies to the acute treatment divided by various potential predicting factors. No significant differences were found between responder status and any of the following factors: gender, capacity to provide consent for treatment, use of antipsychotic or antidepressant medication, or diagnosis of MDD. Interestingly, 10 of the 11 patients diagnosed with AD showed marked response to the treatment versus 11 of the 17 patients without AD (see Table 1 for dementia types breakup), but the difference in proportion was only in the trend range (p = 0.09). In 8 of the 23 acute treatment courses administered to inpatients, patients were discharged within 10 days of the last treatment. In 6 out of 21 acute courses, where patients were on antipsychotic treatment, the scheduled or PRN dose was decreased during the course of the ECT.

Cognitive impairment

Based on the clinical note cognitive impairment scale, significant cognitive adverse effects of ECT were evident in 7% of the acute courses and in 13% of the maintenance courses. Based on the attending psychiatrist score (available for 19 of the acute treatment courses), significant cognitive adverse effects were reported in 21% of the courses. No significant differences in clinical or treatment factors were found to be associated with presence of cognitive adverse effects (defined by either the clinical note or the attending psychiatrist scale≥3). One patient fell and fractured her hip a day after her 2nd treatment. This patient resumed the acute treatment after a two-week break with good results.

DISCUSSION

To our knowledge this is the first study to report on the effectiveness of both acute and maintenance ECT in patients with dementia and NPS. Our findings suggest that ECT is an effective and well-tolerated treatment for patients with severe and treatment resistant NPS of dementia and pre-existing psychiatric illness. In over 70% of the courses, 12 acute ECT treatments administered 2 or 3 times per week, resulted in a significant improvement of aggression, disorganized behavior, agitation, and psychotic symptoms.

Most patients in this cohort had a comorbid diagnosis of major depression and the improvement in NPS also led to improvement in their affective symptoms. In 2 cases of comorbid mania and 4 cases of comorbid schizophrenia or schizoaffective disorder, the improvement was also apparent in the manic and primary psychotic symptoms. The results of this study support previous literature [13 , 17–19] demonstrating the effectiveness of acute course of ECT in treating NPS of dementia in patients with and without pre-existing psychiatric disorders. Further, this study adds that ECT can be a viable treatment option for maintenance of remission from NPS in patients with dementia. Our data shows that an average of a treatment every 2 weeks was helpful in sustaining remission for over a year and was well tolerated.

Treatment related cognitive adverse effects were reported in less than 10% of the treatment courses and there were no other reported adverse effects. However, it is challenging to precisely assess cognitive adverse effects of ECT in patients with dementia in the absence of a structured cognitive assessment. In this study, most treatment courses were administered with bilateral ECT. This was also the case in the Ujkaj et al. study. However, Acharya et al. used right unilateral electrode placement with a pulse width of 0.5–1 ms at 4–6 times the seizure threshold. In that study, 2 out of 19 patients were switched to BT placement due to lack of response and 4 additional patients started treatment with BT placement [17]. RUL electrode placement with ultra-brief pulse (0.3–0.37 ms) has been shown to have better cognitive side effect profile compared to bilateral treatment [22]. While the effectiveness of RUL-UB ECT was recently replicated in a large multi-center study of geriatric depression [23], it was not compared to bilateral ECT in patients with NPS of dementia. Future studies should look at effectiveness of RUL-UB in NPS related to dementia. Our review suggests that in general ECT is safe to combine with psychotropic medication in patients with dementia as most of the patients in our study were receiving some form of medication. It is, however, prudent to re-assess the need for these medications prior to and during the course of ECT.

There are several limitations to the present study. We had no comparison group with sham treatment or with treatment as usual. However, sham treatment (i.e., administering general anesthesia with no actual treatment) is unsafe and unethical. Finding a comparable usual care group from retrospective data would be challenging as ECT in this study was only offered to patients who failed other treatments. In the present study, most patients had a pre-existing psychiatric disorder which may hinder the ability to generalize the study results to patients with NPS without psychiatric comorbidity. Furthermore, in patients with dementia and history of mental illness it is also difficult to distinguish between primary affective or psychotic symptoms and NPS of dementia that can closely mimic them. Another limitation is the use of a global clinical impression to rate response and cognitive impairment. Specific rating scales for NPS like the Cohen-Mansfield agitation inventory [24] or the Neuropsychiatric Inventory [25] would have enabled characterization of change in specific symptom domains in response to ECT. Global rating, however, has the power to capture clinically significant changes which are often more relevant to clinical practice than statistically significant changes in a specific symptom domain or a specific rating scale. Moreover, the use of CGI and CN-CGI can be easily implemented in prospective and retrospective studies as well as in the clinical practice allowing pooling and comparing treatment responses between different sites and protocols. Lastly, our failure to find specific factors associated with response or cognitive side-effects might be related to a relatively small number of treatment courses in the present study. Future larger studies or pooling data from multiple sites can be used to study such potential factors and help in refining patient selection and treatment protocols.

In conclusion, based on the limited evidence, ECT seems to be a safe and effective treatment in patients with severe NPS of dementia. In cases that respond to acute ECT, maintenance ECT appears to sustain treatment response and is well-tolerated. Future studies are needed to study effectiveness of ECT in patients with isolated NPS of dementia and to study comparative effectiveness of different ECT techniques in larger randomized cohorts.

Footnotes

ACKNOWLEDGMENTS

This study was supported by CAMH and the Temerty Family Foundation. The data capture technology used in this study was developed with an Ontario Ministry of Health and Long Term Care Academic Health Sciences Centre Alternative Funding Plan Innovation Fund Grant.

Authors’ disclosures available online ().

References

Borsje

, Wetzels

, Lucassen

, Pot

, Koopmans

(2015) The course of neuropsychiatric symptoms in community-dwelling patients with dementia: A systematic review. Int Psychogeriatr 27, 385–405.

Feast

, Orrell

, Charlesworth

, Melunsky

, Poland

, Moniz-Cook

(2016) Behavioural and psychological symptoms in dementia and the challenges for family carers: Systematic review. Br J Psychiatry 208, 429–434.

Brodaty

, Draper

, Saab

, Low

, Richards

, Paton

, Lie

(2001) Psychosis, depression and behavioural disturbances in Sydney nursing home residents: Prevalence and predictors. Int J Geriatr Psychiatry 16, 504–512.

Ballard

, Corbett

(2013) Agitation and aggression in people with Alzheimer’s disease. Curr Opin Psychiatry 26, 252–259.

Sultzer

, Davis

, Tariot

, Dagerman

, Lebowitz

, Lyketsos

, Rosenheck

, Hsiao

, Lieberman

, Schneider

, CATIE-AD Study Group (2008) Clinical symptom responses to atypical antipsychotic medications in Alzheimer’s disease: Phase 1 outcomes from the CATIE-AD effectiveness trial. Am J Psychiatry 165, 844–854.

Selbaek

, Kirkevold

, Engedal

(2007) The prevalence of psychiatric symptoms and behavioural disturbances and the use of psychotropic drugs in Norwegian nursing homes. Int J Geriatr Psychiatry 22, 843–849.

Vasudev

, Shariff

, Liu

, Burhan

, Herrmann

, Leonard

, Mamdani

(2015) Trends in psychotropic dispensing among older adults with dementia living in long-term care facilities: 2004-2013. Am J Geriatr Psychiatry 23, 1259–1269.

Schneider

, Dagerman

, Insel

(2005) Risk of death with atypical antipsychotic drug treatment for dementia: Meta-analysis of randomized placebo-controlled trials. JAMA 294, 1934–1943.

U.S. Dept. of Health and Human Service, Public Health Advisory: Deaths with Antipsychotics in Elderly Patients with Behavioral Disturbances, http://www.fda.gov/drugs/drugsafety/postmarketdrugsafetyinformationforpatientsandproviders/ucm053171 Posted 11 April, 2005, Accessed 2 February, 2017.

10.

Ayalon

, Gum

, Feliciano

, Areán

(2006) Effectiveness of nonpharmacological interventions for the management of neuropsychiatric symptoms in patients with dementia: A systematic review. Arch Intern Med 166, 2182–2188.

11.

Chenoweth

, King

, Jeon

Y-H

, Brodaty

, Stein-Parbury

, Norman

, Haas

, Luscombe

(2009) Caring for Aged Dementia Care Resident Study (CADRES) of person-centred care, dementia-care mapping, and usual care in dementia: A cluster-randomised trial. Lancet Neurol 8, 317–325.

12.

Geduldig

, Kellner

(2016) Electroconvulsive therapy in the elderly: New findings in geriatric depression. Curr Psychiatry Rep 18, 40.

13.

Liang

, Lam

, Ancill

(1988) ECT in the treatment of mixed depression and dementia. Br J Psychiatry 152, 281–284.

14.

Price

TRP

, McAllister

(1989) Safety and efficacy of ECT in depressed patients with dementia: A review of clinical experience. Convuls Ther 5, 61–74.

15.

Rao

, Lyketsos

(2000) The benefits and risks of ECT for patients with primary dementia who also suffer from depression. Int J Geriatr Psychiatry 15, 729–735.

16.

Burgut

, Popeo

, Kellner

(2010) ECT for agitation in dementia: Is it appropriate?. Med Hypotheses 75, 5–6.

17.

Acharya

, Harper

, Achtyes

, Seiner

, Mahdasian

, Nykamp

, Adkison

, Van der Schuur White

, McClintock

, Ujkaj

, Davidoff

, Forester

(2015) Safety and utility of acute electroconvulsive therapy for agitation and aggression in dementia. Int J Geriatr Psychiatry 30, 265–273.

18.

Ujkaj

, Davidoff

, Seiner

, Ellison

, Harper

, Forester

(2012) Safety and efficacy of electroconvulsive therapy for the treatment of agitation and aggression in patients with dementia. Am J Geriatr Psychiatry 20, 61–72.

19.

Sutor

, Rasmussen

(2008) Electroconvulsive therapy for agitation in Alzheimer disease: A case series. J ECT 24, 239–241.

20.

Kristensen

, Bauer

, Hageman

, Jørgensen

(2011) Electroconvulsive therapy for treating schizophrenia: A chart review of patients from two catchment areas. Eur Arch Psychiatry Clin Neurosci 261, 425–432.

21.

Guy

(1976) ECDEU Assessment Manual for Psychopharmacology. US Department of Heath, Education, and Welfare Public Health Service Alcohol, Drug Abuse, and Mental Health Administration, Rockville, MD.

22.

Sackeim

, Prudic

, Nobler

, Fitzsimons

, Lisanby

, Payne

, Berman

, Brakemeier

E-L

, Perera

, Devanand

(2008) Effects of pulse width and electrode placement on the efficacy and cognitive effects of electroconvulsive therapy. Brain Stimulat 1, 71–83.

23.

Kellner

, Husain

, Knapp

, McCall

, Petrides

, Rudorfer

, Young

, Sampson

, McClintock

, Mueller

, Prudic

, Greenberg

, Weiner

, Bailine

, Rosenquist

, Raza

, Kaliora

, Latoussakis

, Tobias

, Briggs

, Liebman

, Geduldig

, Teklehaimanot

, Lisanby

, CORE/PRIDE Work Group (2016) Right unilateral ultrabrief pulse ECT in geriatric depression: Phase 1 of the PRIDE Study. Am J Psychiatry 173, 1101–1109.

24.

Cohen-Mansfield

, Marx

, Rosenthal

(1989) A description of agitation in a nursing home. J Gerontol 44, M77–M84.

25.

Cummings

(1997) The Neuropsychiatric Inventory: Assessing psychopathology in dementia patients. Neurology 48, S10–S16.