Introduction

Electroconvulsive therapy (ECT) refers to the induction of seizures via the passage of electric current for therapeutic purposes. It is a well-established treatment modality for treating acute episodes of psychoses or affective disorders. It has demonstrated benefits in various other psychiatric as well as neurological conditions. ECT is proven to be efficacious in case of failure of pharmacotherapy maintenance [1]. While the efficacy is unchallenged, relapse after a successful course of treatment in the acute episode has been challenging. The rates of relapse after discontinuation of ECT are high [2]. This warrants some form of treatment for prophylaxis to prevent relapse or recurrence of episodes. Continuation ECT (C-ECT) and maintenance ECT (M-ECT) has become a common psychiatric practice worldwide [3].

The current review was done to synthesize the literature on M-ECT in the management of psychiatric disorders. Specifically, our objectives were (1) to describe the evidence for M-ECT in major psychiatric illnesses and (2) to explore the literature on M-ECT in special populations.

Definitions

It is common practice in psychiatry to continue using the form of treatment used in acute symptom control to prevent relapses and recurrences. After acute-phase ECT is completed, the continuation of ECT for the first 6 months is known as C-ECT. Beyond 6 months, it is called M-ECT [4]. Thus, while C-ECT intends to reduce the relapse risk, M-ECT is used to reduce the risk of recurrence of further episodes (Fig. 1). This 6-month dividing line is only by convention [5]. Often, the terms M-ECT and C-ECT are used interchangeably. In this paper, the term M-ECT will be used to refer to any use of ECT beyond the acute phase treatment.

Fig. 1
figure 1

Phases of electroconvulsive therapy

Full size image

Relapse After Index ECT: a Problem Statement

ECT is highly effective in treating acute psychiatric illnesses. However, patients often relapse after the acute phase ECT. Studies have reported that more than 50% of patients experience relapse following a successful short-term course of ECT treatment [6]. Almost half of the patients with depression on continuation pharmacotherapy relapse within a year, with maximum risk in the first 6 months [7]. A follow-up study in depressed patients found an overall relapse rate of 53% in 1 year regardless of parameters used for index ECT [8]. The relapse rate for patients with schizophrenia ranges from 42.7 to 63.6%, with most recurrences happening within 6 months after treatment [9, 10•]. These findings suggest that an illness with a natural course severe enough to warrant ECT for acute management tends to have a high relapse without adequate long-term treatment. The discontinuation of M-ECT has been shown to be associated with significant clinical deterioration and re-hospitalization [2, 11,12,13]. The COVID-19 pandemic was an exceptional situation that further buttressed this argument. Studies examining the clinical outcome of patients in whom M-ECTs and C-ECTs were discontinued due to the COVID-19 pandemic showed high relapse rates after discontinuation [6, 14,15,16,17].

Methods

Search Strategy

An electronic search of MEDLINE and Google Scholar databases was done (till June 2023) for articles on M-ECT and C-ECT. The searches were done by two independent reviewers who were qualified psychiatrists. The search strategy used for MEDLINE was (((continuation[Title/Abstract]) OR (maintenance[Title/Abstract])) AND (((ECT[Title/Abstract]) OR (Electroconvulsive therapy[Title/Abstract])) OR (electroconvulsive therapy[MeSH Terms]))) AND ((((((((((((((((psychiatr*[Title/Abstract]) OR (mental disorder[Title/Abstract])) OR (schizophrenia[Title/Abstract])) OR (psychosis[Title/Abstract])) OR (psychotic disorder[Title/Abstract])) OR (schizo*[Title/Abstract])) OR (bipolar[Title/Abstract])) OR (mania[Title/Abstract])) OR (depressi*[Title/Abstract])) OR (obsessive compulsive disorder[Title/Abstract])) OR (mental disorder[MeSH Terms])) OR (disorder, psychotic[MeSH Terms])) OR (affective psychosis, bipolar[MeSH Terms])) OR (depressive disorders[MeSH Terms])) OR (mania[MeSH Terms])) OR (disorder, obsessive compulsive[MeSH Terms])). Manual searches from the references of the generated articles were also done to include any articles that were missed in the electronic search.

Study Selection and Data Extraction

There were 725 results in the initial search. Only articles in English were included. We included original research such as controlled trials and observational studies in psychotic and affective disorders and case reports in areas where evidence of M-ECT is otherwise lacking such as Obsessive Compulsive Disorder (OCD). We focused on research pertaining only to clinical efficacy and outcomes and excluded articles focusing on other parameters such as cost-effectiveness, cognitive outcomes, and side effect profile. Review articles, viewpoints, and commentary papers were not included as they did not contain any original data. With these criteria, we included 175 articles for initial title/abstract screening. Following this screen, 96 articles were selected for full-text review. After excluding articles that were not relevant to the objectives, 45 articles remained and were included for the final review. Two independent reviewers were involved in the study selection process, and both agreed regarding the included papers.

This review aimed to synthesize the existing literature on the efficacy of M-ECT on various psychiatric disorders. Although there are prior systematic reviews on ECT for disorders such as depression and psychosis, we favored a narrative review for two reasons: first and more importantly, it did not constrain us from including evidence in the form of case reports in areas where original articles were lacking (such as OCD and special populations such as pregnant women), and second, it permitted us to provide a broader perspective on a topic where controlled studies are limited.

The articles included in the review were categorized into the following themes: trials that evaluated the efficacy of M-ECT in major psychiatric disorders (depression, bipolar illness, obsessive–compulsive disorder). Information regarding the effect of M-ECT on cognition and cost-effectiveness of M-ECT and the effectiveness of ECT in special populations such as children, pregnant women, and older adults is also discussed.

Results

Among the 45 articles included in the review, 37 were original articles and 8 were case reports. Of the 37 original articles, there were only 8 randomised Controlled Trials (RCTs), out of which only one was published in the last 5 years. This indicates that controlled trials are lacking in this area.

Evidence for Maintenance ECT

Since the inception of ECT, the use of C-ECT and M-ECT has been well-documented [18]. However, it is reserved only for the most resistant cases in clinical practice. It is pertinent to note that the American Psychiatric Association (APA) Task Force in 1974 reported a role for ECT in treating depression, intractable mania, and treatment-resistant schizophrenia; however, they did not mention C-ECT in the report [19]. Later, the 2001 task force report by APA established indications for C-ECT and M-ECT in patients responding to an acute ECT course when (1) pharmacotherapy has been ineffective in treating acute episodes or preventing relapse or recurrence.; (2) administering pharmacotherapy safely is not possible; or (3) the patient or their designated representative has expressed a preference for ECT [4]. Table 1 summarizes the evidence for M-ECT in psychotic and affective disorders.

Table 1 Characteristics of included studies on psychotic and affective disorders
Full size table

Maintenance ECT in Major Depressive Disorder

There is extensive literature covering the use of M-ECT following index ECT for depression. A systematic review in 2014 concluded that the combination of C/M-ECT and pharmacotherapy outperforms monotherapy in preventing the recurrence of unipolar depressive episodes [20•]. Three RCTs [21,22,23] and one quasi-experimental study [24] were included in the review. Elias et al. (2017) conducted a meta-analysis of RCTs on the efficacy of M-ECT in unipolar and bipolar depression in 436 subjects [21, 22, 25, 26]. They found that M-ECT was associated with significantly less relapses when compared with pharmacotherapy alone at 6 months and 1 year [27•]. One of the latest systematic reviews on M-ECT in depression conducted by Rowland et al. (2023) also suggested reduced relapse rates in patients receiving C/M-ECT with no effect on global cognitive function [28••]. Introduction of M-ECT also reduces mean hospital days and illness days/year along with overall relapse rates and increases chances of clinical remission [29,30,31,32,33,34,35].

Several other reviews provide an exhaustive discussion of this area [20•, 36,37,38]. An RCT randomly assigned remitted depressed patients after ECT to maintenance pharmacotherapy (M-Pharm) consisting of lithium plus venlafaxine versus this combination plus four-weekly ECT treatments followed by “as needed” supplemental treatments for the return of depressive symptoms. This method of using M-ECT was termed “STABLE” (“symptom-titrated, algorithm-based longitudinal ECT”). The follow-up period was 6 months, at the end of which the combined modality group had a lower mean Hamilton Rating Scale for Depression (HAM-D) scores [25]. Combined M-ECT and pharmacotherapy has been shown to be associated with significantly lesser relapses/recurrences when compared with maintenance pharmacotherapy alone [39,40,41,42,43].

In conclusion, patients with intractable, recurrent mood disorders and associated psychotic symptoms often benefit more from M-ECT. Specifically, it can be considered in patients who had multiple hospitalizations in the past, multiple medication trials and who needed ECT to remit from the acute episode. However, patients with comorbid personality disorders may do less well. Clinicians must use their own judgment to make decisions on a case-by-case basis.

Maintenance ECT in Bipolar Disorder

Not only depression but M-ECT has also been reported to reduce occurrences of mania/hypomania in bipolar patients [44, 45]. Most research has consisted of naturalistic, retrospective, and prospective follow-up studies and case reports. The patient numbers are generally very small, with heterogeneity in outcome measures and use of concomitant medications. Observational studies suggest that M-ECT is useful in reducing the number of psychiatric hospitalizations and hospitalization days in bipolar disorder [13, 31]. Several case reports exist of patients with refractory bipolar disorder whose illness was stabilized with M-ECT [46,47,48]. Positive evidence of the effectiveness of M-ECT in decreasing morbidity in rapid cyclers was found in a pre-post study design [49]. Data also supports the prophylactic role of M-ECT in bipolar disorders [33, 50]. We could not find any RCTs assessing M-ECT specifically in bipolar disorder. However, studies on resistant depression have included bipolar depression as well.

Despite the lack of robust evidence in the form of controlled trials, M-ECT seems useful in treating refractory bipolar disorder and rapid cyclers. It can also be used as a maintenance treatment to prevent relapse in bipolar patients. Continuation of mood stabilizers and other psychotropics along with M-ECT can be left to the clinician’s discretion as deemed necessary.

Maintenance ECT in Psychotic Disorders

Patients with schizophrenia also experience a rapid relapse of symptoms after abruptly discontinuing ECT, similar to patients with affective disorders [25, 51]. M-ECT is recommended for treating patients with treatment-resistant schizophrenia (TRS) since additional ECT sessions after acute ECT can maintain mood improvement or prevent mood disorders from relapsing [52]. The first RCT of continuation or maintenance ECT for schizophrenia observed combining M-ECT with high-dose neuroleptic treatment is more effective than either treatment alone in patients with ECT-responsive TRS [53]. The improvement observed was sustained and even decreased to the level of the symptoms in the clozapine alone group at the end of 2 years in a retrospective study of TRS patients [54]. Several other studies have replicated the same findings [55,56,57,58]

Ward et al. (2018) systematically reviewed three RCTs, five retrospective studies, nine open-label trials, and 18 case reports, suggesting that M-ECT is effective for preventing relapse in schizophrenia along with continued antipsychotic treatment [10•]. Not only in schizophrenia but also in schizoaffective disorder, the time to re-hospitalization in patients receiving M-ECT and drugs was observed to be longer when compared with patients receiving maintenance pharmacotherapy alone [43, 59, 60].

Evidence suggests that M-ECT efficiently maintains the improvement in psychotic symptoms [61,62,63]. Some poor predictors of M-ECT response in schizophrenia include higher brief psychiatric rating scale (BPRS) scores, lower percentage reduction in BPRS scores, lower Global Assessment of Functioning (GAF) scores at baseline, earlier onset of illness, more hospitalization, and lower education levels [64]. Patients with good prognostic features may benefit even with more spacing between ECT sessions.

Maintenance ECT in Obsessive–Compulsive Disorder (OCD)

M-ECT is not an approved treatment for OCD but has been reported effective in treatment-refractory cases of OCD. A total of 8 case reports on the effect of M-ECT in OCD have been reported till date (Table 2). The cases described are mostly treatment-refractory OCD [65,66,67,68] or comorbid OCD and schizophrenia/schizoaffective/mood disorder [69, 70]. M-ECT in a patient of OCD with Asperger syndrome has been reported [71]. Apart from these, it has also been used in antipsychotic-induced OCD [72]. No controlled trials evaluating the efficacy of M-ECT in OCD are available.

Table 2 Characteristics of included studies on obsessive–compulsive disorder (OCD)
Full size table

Cost-Effectiveness

Studies on cost analyses of M-ECT are unavailable yet. Regarding direct costs, M-ECT might be costlier than maintenance pharmacotherapy, depending on the setting. However, hospitalization rates, which can be considered a proxy marker of cost-effectiveness, are reduced consistently in the follow-up period during M-ECT [12]. M-ECT has been shown to reduce time spent in depression from 50 to 34% of life-years in a western study; the authors found that M-ECT may be a cost-effective treatment option for drug-resistant depression [74]. It reduces the indirect costs related to illness-related unemployment, absenteeism, lost wages, and other disabilities [75].

Impact of Maintenance ECT on Cognition

Cognitive problems have been demonstrated with acute ECT but are limited to the first few days post-treatment. Interestingly, information processing speed, working memory, anterograde memory, and some aspects of executive function are improved beyond baseline levels after 15 days [76]. In phase 2 of the Consortium for Research in ECT (CORE) trial, no significant differences in Mini Mental State Examination (MMSE) scores were observed between M-ECT and maintenance pharmacotherapy groups. Interestingly, in this study, the relapse rate in the C-ECT group was higher than the maintenance pharmacotherapy group, although the difference was not statistically significant [21]. Similar findings of the superiority of M-ECT over maintenance pharmacotherapy were noted in prospective studies of other affective and psychotic disorders [22, 43, 53]. At 6 months, extensive neuropsychological assessment did not find evidence of cognitive decline with M-ECT and no significant differences in cognitive functioning between M-pharm and M-ECT groups [77]. One of the latest systematic reviews including nine studies suggested no detrimental effect of M-ECT on cognitive functioning [78]. In individual cases where concerns of cognitive side effects arise, high-dose unilateral ECT can be preferred over bilateral ECT due to its superior cognitive effect profile [8].

Clinical Considerations

Dosing

There is a lack of established guidelines for the duration and frequency of M-ECT. In clinical practice, M-ECT is tailored to individual patient’s needs, usually given for the duration of the natural course of illness. The prolonging remission in depressed elderly (PRIDE) study followed a symptom-titrated algorithm-based longitudinal ECT (STABLE) approach that included an initial fixed treatment schedule of four ECT treatments in the first month followed by a symptom-driven, individualized treatment algorithm [51]. Clinicians also follow a fixed-interval schedule with treatments every 1–4 weeks and an “as-needed” approach with 1–2 additional treatments each time there are signs of relapse. The APA recommends reassessing the need for M-ECT at least every 3 to 6 months [36]. Clinicians are encouraged to assess the benefits and risks of the procedure at each session, as well as the need for greater or lesser spacing of treatment.

Stopping of Maintenance ECT

If remission has been sustained long enough, M-ECT can be stopped and symptoms monitored. Rescue ECT can be given if relapse signs appear [20•, 64]. A tapering schedule of weekly sessions of 2 to 4 weeks followed by a gradual decrease to once per month has been described [79].

Maintenance ECT in Special Population

Older Adults

Continuation-maintenance ECT is also an effective option in geriatric patients, in whom pharmacotherapy has its own risks [41]. ECT is generally safe in geriatric population with medical comorbidities, intolerability, or poor response to pharmacotherapy [80]. Second phase of the PRIDE study found that remitted older patients randomized to receive continuation ECT plus pharmacotherapy (venlafaxine and lithium) showed significantly lower depressive symptoms during a 6-month period when compared to patients treated with pharmacotherapy alone [81••].

A large systematic review of M-ECT in depressed elderly patients by van Schaik et al. (2012) concluded that M-ECT is as effective as continuing pharmacotherapy in geriatric depression after a successful course of ECT in the acute stage [82]. Acute ECT, followed by M-ECT, also reduced readmission rates and duration of hospital stay in elderly patients with schizophrenia [60]. Although the risk of cognitive side effects has been raised, data suggests using newer ECT techniques, like right unilateral ultra-brief pulse ECT can minimize adverse ECT-related cognitive effects [83]. Thus, M-ECT can be considered a safe and effective treatment in geriatric patients with medical comorbidities, where pharmacotherapy is considered risky.

Children and Adolescents

Data regarding the efficacy of M-ECT in this age group is scarce. Case reports and case series of M-ECT for patients with autism presenting with catatonia and self-injurious behaviors not responding to medications have been described; notably, a reduction in the frequency of M-ECT sessions led to relapse [84, 85]. Unspecified catatonia has been reported in a cohort of patients with Down’s syndrome maintaining well on M-ECT [86]. A retrospective study on the effectiveness of M-ECT in children with schizophrenia spectrum disorder showed that a combination of M-ECT and pharmacotherapy resulted in significantly lower positive and negative symptoms of schizophrenia (PANSS) scores at 6 months following the acute ECT course. Furthermore, when the frequency of ECT was reduced, two out of seven patients experienced a relapse [87]. Section 95 of the Mental Health Care Act in India mandates the consent of guardians and the Mental Health Review Board for the use of ECT in minors [88]. Thus, M-ECT in children should be considered based on a careful risk–benefit assessment.

Pregnant Women

Only case reports are available for M-ECT in pregnant women. In one of the case reports, a patient with depression received 18 sessions in the second and third trimesters of pregnancy, and 13 sessions were administered post-elective cesarean delivery as maintenance treatment. The patient was shifted successfully to pharmacotherapy 6 months postpartum [89]. A similar case has also been reported in another pregnant female with psychotic depression [90]. No controlled trials of M-ECT are available in this population.

Discussion

The primary purpose of this review was to examine the evidence of M-ECT in various psychiatric disorders and psychiatric subpopulations. Our findings point to M-ECT being a safe and effective option for prolonging remission in certain patients such as those non-responsive/intolerant to pharmacotherapy, elderly, and patients with comorbidities in whom potential for drug interactions limit pharmacological options. Reduced relapse rates and recurrences have been observed in mood and psychotic disorders [10•, 20•, 27•].

M-ECT has considerable evidence for preventing relapses and recurrences in mood disorders, with greater evidence for unipolar depression. No study has found M-ECT to be inferior to pharmacotherapy. The majority of the RCTs on depression are conducted in geriatric population and depression with psychotic symptoms. Consistent evidence suggests that combined pharmacotherapy and M-ECT outperform either of these treatments administered alone [20•]. M-ECT in bipolar disorder has been shown to stabilize the illness course in rapid cyclers, with a significantly lesser number of episodes and more illness-free days [91]. Similar to major depressive disorder (MDD), evidence on bipolar disorder suggests using concomitant medications, often mood stabilizers, along with M-ECT for relapse prevention [42].

Studies on psychosis also have found M-ECT to be beneficial in maintaining remission. Further, it also prevents rehospitalization and improves the quality of life in patients with psychosis [10•, 92]. Most studies have allowed the use of concomitant antipsychotics, which aligns with the real-world practice. Electrode placement may have a significant effect on cognitive dysfunctions; however, studies comparing the effects of electrode placements are rare.

However, the evidence should be taken with a pinch of salt as most of the available studies are either observational or non-randomized in nature. Various factors may limit the overall interpretation of the results such as sample heterogeneity, differences in the type of wave and electrode placement, and variations in concomitant psychotropics and periods of follow-up. In addition, there is a clear lack of evidence for M-ECT in other psychiatric disorders, such as OCD, and in vulnerable populations, such as children and pregnant women. There is a dearth of data to formulate recommendations for M-ECT administration. A tapered approach tailored to individual patient needs is usually practiced. Decisions regarding when to taper and when to stop the sessions may be taken based on symptom improvement, functional recovery, and adverse effects. More research is needed on the ideal schedules of M-ECT in various psychiatric disorders. It is important to address issues related to dosing of stimulus, electrode placement, and simultaneous pharmacotherapy. There is a need to expand the use of ECT clinical registries because RCTs in this area are difficult to accomplish.

There are a few limitations to the present review. Only two databases were searched, and only English-language articles were included in the review. Due to a lack of studies with robust evidence, case reports and observational studies were included. Further, given the wide breadth of articles included, concerns remain regarding external validity. Future reviews focusing on mechanism, differences in dosing, frequency of administration, electrode placement, and adverse events can add to the current understanding.

Conclusion

Psychotic and affective disorders represent conditions with the most evidence for M-ECT in maintaining clinical remissions and preventing hospitalizations. However, more randomized clinical trials are required to build clear evidence and inform recommendations for M-ECT in these disorders. Although evidence has extended to other psychiatric disorders such as OCD, data is not robust. Further, studies exploring the best strategies in terms of dosing schedule and frequency for delivering M-ECT are also required to guide clinicians in everyday practice.