A Systematic Review of the Utility of Electroconvulsive Therapy in Broadly Defined Obsessive-Compulsive-Related Disorders

ABSTRACT

Obsessive-compulsive disorder (OCD) is characterized by recurrent and persistent thoughts, urges, or images that are intrusive, unwanted, and distressing (obsessions) and repetitive behaviors or mental acts that the individual performs to ignore, suppress, or neutralize the resulting anxiety or according to rigid rules.1 OCD affects 2.3% of the general population at some point in their lives and is often chronic and disabling.2 First-line treatments include high-dose serotonin reuptake inhibitors (SRIs) and exposure and response prevention (ERP).3 Individuals who show partial response to SRIs are often treated with antipsychotics or, preferentially, outpatient ERP.4 Up to 10% of patients are generally treatment refractory and are potential candidates for ERP administered in intensive residential treatments5 or more invasive treatment measures including deep brain stimulation6 or psychiatric neurosurgery (eg, capsulotomy).7

One issue that has often been raised in the literature is whether OCD patients, particularly those who are treatment refractory, will benefit from a course of electroconvulsive therapy (ECT).8 A systematic review9 on this topic failed to find evidence supporting the use of ECT in OCD, as the literature was restricted to case reports and series, nonrandomized trials, and cohort studies. Specifically, ECT responders were characterized by a more severe form of OCD with a later onset of OCD symptoms, less depression, and a fewer number of ECT sessions.9 Finally, the finding that ECT responders were less likely to be treated with conventional anti-OCD treatments (ie, adequate trials of SRIs and/or ERP) suggests that ECT could have been avoided if patients were treated more conservatively.9

In addition to insufficient evidence supporting the use of ECT in OCD, it is still unclear whether ECT may be helpful in those affected by obsessive-compulsive-related disorders (OCRDs) other than OCD. In DSM-5, OCRDs include body dysmorphic disorder, hoarding disorder, trichotillomania (hair-pulling disorder), and excoriation disorder (skin-picking disorder) among others.1 Further, despite not being officially recognized as OCRDs in DSM-5, some neurodevelopmental disorders have been considered part of an “extended” OCD spectrum on the basis of clinical and biological commonalities with OCD behaviors, including tic disorders and Tourette’s disorder, autism spectrum disorders, and stereotypic movement disorders such as self-injurious behaviors.10 These neurodevelopmental conditions have been noted to co-occur with each other and to be overrepresented among early onset OCD cases.11-13

In contrast to OCD patients, for whom the usefulness of ECT remains elusive, there is good reason to suspect that OCRDs may be responsive to ECT. First, body dysmorphic disorder and hoarding disorder are frequently delusional14,15 and associated with greater severity of depression.15,16 Second, symptoms of catatonia (the quintessential “ECT-responsive” neuropsychiatric disorder) overlap substantially with tic disorders and Tourette’s disorder17,18 (eg, echopraxia and echolalia, staring, and grimacing), autism19-21 (eg, negativism, mutism/verbigeration, and stereotypies/mannerisms), and self-injurious behaviors22 (eg, impulsivity and aggressiveness). Also, both cognitive23,24 and motor OCRDs25-28 can reach extreme levels of severity, occasionally involving life-threatening behaviors that can result in severe tegumentary lesions,23 trichobezoars (mass of hairs blocking the bowel),25 or even sepsis, thus prompting clinicians to consider ECT for some forms of OCRDs.

Despite the reasons for predicting positive responses of OCRD cases to ECT, there also have been negative experiences. For instance, Kalinowsky and Hippius29 noted that tic disorder patients who “learned” to suppress tics might be unable to do so and deteriorate as a consequence of the detrimental effects of ECT on their ability to focus attention. Worsening of OCD after ECT has been reported in a few cases,8 but it is unclear whether a similar phenomenon occurs in other OCRDs. To shed light on this poorly understood phenomenon and its treatment, we systematically reviewed studies reporting the effect of ECT in individuals with an official DSM-5 diagnosis of an OCRD, including body dysmorphic disorder, hoarding disorder, trichotillomania, and excoriation disorder, as well as conditions under an “extended” OCD spectrum including tic disorders, autism spectrum disorders, and self-injurious behaviors. This review was based partly on the patient, intervention, comparator, and outcome model30 and aimed to answer the following questions:

1. Has the effectiveness of ECT in OCRD patients been compared to any sham or active treatment?

a. If yes, what is the effectiveness of ECT compared to sham or active treatment?

b. If not, what is the highest level of evidence regarding the effectiveness of ECT in OCRD patients?

2. How do baseline features of OCRD patients who respond to ECT differ from OCRD patients who do not respond to ECT?

• Despite the lack of randomized controlled trials supporting the efficacy of electroconvulsive therapy in obsessive-compulsive-related disorders, it is still being prescribed to patients with severe forms of repetitive behaviors.
• Maintenance electroconvulsive therapy in obsessive-compulsive-related disorders may be warranted, particularly in cases showing more severe tic disorders and self-injurious behaviors, due to the high relapse rate.

METHODS

Study Selection Criteria

Type of studies. We searched for randomized controlled trials (RCTs), observational studies, retrospective investigations, case series, and single case reports describing ECT effects on the core repetitive behaviors across the full range of DSM-5 OCRDs (ie, body dysmorphic disorder, hoarding disorder, trichotillomania, and excoriation disorder), as well as tic disorders, autism spectrum disorders, self-injurious behaviors, kleptomania, and pyromania. Studies that, despite including the targeted populations, reported just the effects of ECT on other syndromes or behaviors (ie, depression in autism spectrum disorders) were preliminarily excluded from our research protocol. Editorials, letters, and reviews were included if sufficiently detailed original data were provided.

Types of ECT interventions. We selected all studies reporting the use of modified, unmodified, multiple-monitored, bilateral, and unilateral ECT with any dose, frequency, or level of stimulus. Studies that reported individuals receiving ECT who were also under current or previous pharmacologic treatments (eg, SRIs and antipsychotics) or different forms of psychotherapy (including cognitive-behavioral therapy [CBT] such as ERP, habit reversal, or behavior modification) were also included. Studies describing individuals treated with other brain stimulation techniques such as transcranial magnetic stimulation (TMS), transcranial direct-current stimulation (tDCS), deep brain stimulation (BDS), and vagus nerve stimulation were also included if the patients’ response to ECT was also reported.

Type of measures. Given the dearth of instruments employed across and within the target disorders of our review, the primary outcome of interest was the authors’ categorical description of clinically meaningful benefits of ECT on repetitive behaviors that characterize DSM-5 OCRDs and the extended OCD spectrum. However, we were also interested in evaluating how frequently reports of ECT use in broadly defined OCRDs described valid instruments for the assessment of severity of repetitive behaviors in each disorder of interest (including, but not restricted to, the Yale-Brown Obsessive-Compulsive Scale [YBOCS] modified for body dysmorphic disorder,31 the YBOCS modified for neurotic excoriation,32 the Saving Inventory-Revised,33 the Massachusetts General Hospital Hairpulling Scale,34,35 the Yale Global Tic Severity Scale,36 and the Children’s Yale-Brown Obsessive Compulsive Scale modified for autism spectrum disorders37 among others38).

Search Methods for Identification of Studies

Electronic searching. A systematic search was performed between June 6 and July 2, 2017, in the following databases: MEDLINE, Web of Science, Scopus, and LILACS. The keywords used for the search are listed in Table 1.

Other sources. The reference lists of studies that fulfilled inclusion criteria and textbooks deemed relevant for our research question29,39-45 were also examined for possible new studies. Further, complete and in-progress clinical trials still not published in clinicaltrials.gov and WHO clinical trials registry platforms were also considered for inclusion. No search restrictions (eg, date, language, or document type) were used.

Data Collection and Analysis

Selection of studies. Repeated titles across databases were removed, and the remaining articles were checked and analyzed individually by our team, which comprised 2 psychiatrists (L.F.F. and J.N.Q.), 2 biomedical scientists (S.S.R. and J.B.S.A.), and 1 psychology student (K.B.B.). During weekly meetings that took place between June 2017 and February 2018, titles and abstracts were analyzed to exclude articles that clearly did not meet the inclusion criteria; then, the full text of selected articles was read for extraction of the data of interest. Eligibility was defined according to inclusion and exclusion criteria for each diagnostic group. When doubts or disagreement between authors regarding the characteristics of the collected variables occurred, a consensus was reached. If the doubts persisted, then corresponding authors of the articles in question were contacted for clarification. If no response was obtained, than the study was excluded from the analysis.

Data management. The authors of this review conducted the entire process. If a study fulfilled all of the criteria for inclusion, then its information was entered into a SPSS file (SPSS Inc, Chicago, Illinois). We evaluated experimental and group studies and individual cases. In terms of experimental and group studies, we specifically selected RCTs, the quality of which we planned to assess according to Cochrane criteria46 including (1) random sequence generation; (2) allocation concealment; (3) blinding of participants, personnel, and outcome assessors; (4) selective reporting; and (5) incomplete outcome data. In terms of individual cases, whenever possible, demographic and clinical data, alleged indications for ECT use, treatment history before ECT administration, information related to ECT, and clinical efficacy (including follow-up) were collected.

Relevant demographic and clinical data included age, sex, age at OCRD symptom onset, target diagnosis, patterns of comorbidities (eg, OCD/obsessive-compulsive symptoms, other OCRDs, major depressive disorder, bipolar disorder, or schizophrenia), and whether patients were examined with valid assessment tools. Possible indications for ECT were categorized into OCRDs secondary to other conditions known to be treatable with ECT (eg, depression); treatment-resistant OCRDs; severe OCRDs (ie, conditions resulting in self-injury or risks for patients’ physical health); severe major depressive disorder with suicidality or psychosis; catatonia; severe mania, psychosis, or agitation; previous response to ECT or seizures; and the presence of drug side effects or other indications.

We also collected data on individuals’ treatment history including previous use of SRIs, antipsychotics, any form of psychotherapy, and concomitant use of antidepressants and psychotherapy during ECT. Information on ECT parameters such as type of electrode placement (unilateral or bilateral), total number of ECT sessions, and weekly frequency of administration was also registered. Finally, data related to patterns of ECT response were collected, including positive versus negative response of OCRDs or associated depression to ECT, whether this response was based on a valid assessment tool or simply the authors’ subjective clinical impression, whether patients have been followed up (reassessed) in the absence of ECT use, whether there was a relapse during follow-up, whether a second course of ECT was required, and, if a second course of ECT was required, whether the patient responded.

Statistical analysis. As previously reported,9 a priori analyses of ECT responders versus nonresponders were planned on the basis of the following variables: age, sex, age at OCRD symptom onset, main OCRD, course of illness, major comorbidities, previous exposure to SRIs or antipsychotics, previous psychotherapy (including CBT), indication for ECT, and number of different indications for ECT. Categorical variables were described in frequencies and compared using χ2 square and Fisher exact test, and continuous variables were compared using the Mann-Whitney U test. Correction for multiple comparisons was performed using the Benjamini-Hochberg test47 with a false discovery rate set at 0.25.

RESULTS

The search resulted in 833 articles, of which 384 represented repeated references. After initial selection, the titles and abstracts of the remaining 449 articles were further screened, leaving 95 articles for eligibility assessment according to the selection criteria. Eventually, 5718,45,48-102 articles included OCRD individuals treated with ECT allowing qualitative analysis, and 5218,45,48-97 articles included information allowing quantitative analysis. The studies by Ghaziuddin et al,71 Yero et al,98 Larkin,99 Bailine and Petraviciute,100 and Thuppal and Fink101 were excluded from the quantitative analysis for lacking report of the effects of ECT on repetitive behaviors. We also included studies that did not focus primarily on ECT response but that, nevertheless, allowed collection of data related to response of OCRDs to ECT.48-53 Figure 1 shows the PRISMA diagram of the research steps.

We were unable to find any RCTs, controlled before and after studies, interrupted time series studies, historically controlled studies, cohort studies, case control studies, or cross-sectional studies. All identified articles were case reports or series. Supplementary Table 1 lists the included studies, their design and number of enrolled participants, and the primary indication for ECT administration. A total of 69 OCRD individuals were treated with ECT, of whom 69.2% presented self-injurious behaviors of different etiologies (autism spectrum disorders, tic disorders or Tourette’s disorder, and delusions), 35.4% Tourette’s disorder or tics (not explicitly involving self-injurious behaviors), 13% body dysmorphic disorder, and 2% trichotillomania (see patient features in Table 2).

The final sample included 64 patients for whom it was possible to establish a positive response to ECT, which was observed in 73.4% of the total cases, including 44.0% of the body dysmorphic disorder, 74.1% of the tic disorder, and 85.7% of the self-injurious behavior patients. Nevertheless, only 12 (17.4%) cases had formal assessments of the severity of their OCRDs before and after the treatment. Further, follow-up was described in only 27 individuals (39.1%), ranging from 1 week to 48 months. Although relapse was noted in more than half (55.6%) of individuals who were followed up, positive responses were reported in all individuals (n = 12) who received a second course of ECT.

We could establish the stimulus electrode placement in 45 OCRD cases: ECT was bilateral in 36 (80.0%), unilateral in 6 (13.3%), and unilateral followed by bilateral or vice versa in 3 (6.7%) patients. In 24 cases, mode of ECT placement was unclear. The mean (SD) number of ECT sessions administered in the first course of ECT, based on a total of 51 available cases, was 11.27 (5.73). Frequency of ECT administration was established in 21 patients. The mean (SD) number of ECT sessions per week was initially 2.95 (0.50). There was a reduction in the frequency of ECT in the first treatment course in 20 (37.0%) cases.

Individuals who did and did not respond to ECT were compared and contrasted (Table 3). Groups did not differ in terms of age, sex, age at OCRD onset, course and pattern of comorbidity, indication for ECT, electrode placement, total number of sessions, and frequency of administration. However, there was a trend for higher frequency of self-injurious behaviors between responders (P = .08). Articles describing ECT responders were more likely to report previous unsuccessful treatment with antipsychotics (P < .001), SRIs (P = .007), and CBT (P = .05). However, after the Benjamini-Hochberg procedure was applied, only differences in antipsychotic and SRI exposure remained significant. Publication bias was investigated by comparing the positive responses described in case reports (68.1%) to those reported in case series (n > 1) (31.9%), but no significant differences emerged (χ21 = 2.3, P = .12).

DISCUSSION

In this systematic review, we were unable to find any RCT on the use of ECT in OCRD patients. As such, it is not possible to provide unequivocal evidence of the efficacy of ECT in this group of individuals. Although a positive response was reported in 73.4% of published patients, all records were case reports and series, whose ability to provide reliable evidence of treatment benefits are generally undermined by a high probability of selection, information, and, frequently, publication biases.102,103 Accordingly, case reports and series lie at the penultimate level in the hierarchy of evidence of treatment benefits, being only superior to mechanistic reasoning.104 However, reviews of such cases have been reported as alternative “clinician-friendly methods” to test hypothetical therapeutic models.103,105 This theory is consistent with the finding that up to 22% of cases published in a high-impact journal have been followed by trials after a period of sometimes only 5 years.106 Thus, the available evidence does suggest that an RCT of ECT in OCRDs may be warranted, particularly in more severe tic disorder and self-injurious behavior cases.

We also found that less than 40% of all OCRD patients treated with ECT were evaluated after administration of the index course of ECT (ie, followed up). However, when these follow-up assessments were reported, relapses were noted in more than half of individuals. Similarly high rates of post-ECT relapse in the first year have been described in depression (51% in 1 year)107 and schizophrenia (42.7 to 63.6%).108,109 Fortunately, all OCRD individuals who relapsed responded favorably to a repeated course of ECT. High relapse rates coupled with prompt remission after treatment resumption are consistent with an ability of ECT to keep symptoms suppressed until spontaneous remission occurs.110 As many OCRDs follow a chronic course, future studies should examine the role of maintenance ECT in this specific population. Maintenance ECT has been used safely for up to 12 years111 and proved effective in depression,112,113 bipolar disorder,114 and schizophrenia.115 However, the cost of sessions and concerns of patients and families about the side effects of prolonged ECT may contribute to withdrawal and discontinuation of maintenance ECT.116

Although ECT responders and nonresponders did not differ in most parameters, there was a trend for higher frequency of self-injurious behaviors and greater exposure to unsuccessful trials with antipsychotics and SRIs among the former group. These findings are in sharp contrast to our previous review,8 in which ECT responders were less likely to have been treated with SRIs or CBT in the past. It is difficult to speculate on the significance of these findings, but they suggest that either (1) ECT is more effective as an augmentation strategy for patients under existing anti-OCRD treatments compared to patients who are not using antipsychotics or SRIs or (2) authors who describe response of OCRD patients to ECT are more likely to emphasize or report previous resistance to antipsychotics or SRIs. We tend to favor the second interpretation, as it may be difficult to support the use of ECT in an atypical population that is not usually seen as responsive to ECT.

Our study has a number of limitations. First, it partly consisted of an analysis of individual cases of OCRDs treated with ECT as reported in the literature, which are subject to missing data. Second, reviewed cases comprised a heterogeneous sample with self-injurious behaviors (69.2%), tic disorders and Tourette’s disorder (35.4%), body dysmorphic disorder (13%), and trichotillomania (2%). However, our study should be considered consistent with contemporary models (eg, Research Domain Criteria) suggesting that OCRDs may respond to treatments that cut across traditional diagnostic boundaries. Third, only 12 (17.4%) cases had formal assessments of severity, and no transdiagnostic measure of efficacy was available, so we had to rely on authors’ descriptions of treatment response to report efficacy and compare responders to nonresponders. Finally, although it would be interesting to investigate whether risk factors for relapse reflect features related to the underlying OCRD or the initial ECT course,117 low numbers precluded us from comparing patients that did versus did not relapse.

CONCLUSIONS

The finding that more than 70% of the reviewed cases (including body dysmorphic disorder, tic disorders, autism, and self-injurious behaviors) have shown some response to ECT should not be considered unequivocal evidence of its efficacy in OCRDs. The primary limitation to forming any firm conclusions is that there is currently a total lack of RCTs on the efficacy of ECT in different OCRDs. An additional limitation is that there is an absence of valid transdiagnostic tools for the assessment of treatment response across different OCRDs. Nonetheless, the available evidence does suggest that an RCT of ECT in OCRDs may be warranted, particularly in more severe tic disorders and self-injurious behavior cases in which maintenance ECT may be advisable due to the high relapse rate. The reported association between OCRD response to ECT and previous treatment with antipsychotics and antidepressants suggests that either (1) ECT is more effective as an augmentation strategy for existing anti-OCRD treatments or (2) authors who describe response to ECT are more likely to emphasize or report previous resistance to antipsychotics or antidepressants.

Submitted: June 16, 2018; accepted August 7, 2018.

Published online: October 18, 2018.

Potential conflicts of interest: None.

Funding/support: This work was supported by Conselho Nacional de Desenvolvimento Cient×­fico e Tecnológico (Dr Fontenelle, grant no. 308237/2014-5), Fundaç×£o Carlos Chagas Filho de Amparo ×  Pesquisa do Estado do Rio de Janeiro (Dr Fontenelle), the D’ Or Institute of Research and Education (Dr Fontenelle), the David Winston Turner Endowment Fund (Drs Fontenelle and Yücel); and the National Health and Medical Research Council of Australia (Dr Yücel, grant no. APP1117188).

Role of the sponsor: The supporters had no role in the design, analysis, interpretation, or publication of this study.

Previous presentation: This study was presented as an abstract at the 31st European Congress of Neuropsychopharmacology; October 6-9 2018; Barcelona, Spain.

Supplementary material: See accompanying pages.

REFERENCES

100. Bailine SH, Petraviciute S. Catatonia in autistic twins: role of electroconvulsive therapy. J ECT. 2007;23(1):21-22. PubMed CrossRef

101. Thuppal M, Fink M. Electroconvulsive therapy and mental retardation. J ECT. 1999;15(2):140-149. PubMed CrossRef

102. Garc×­a-Doval I, Albrecht J, Flohr C, et al; European Dermato-Epidemiology Network (EDEN). Optimizing case reports and case series: guidance on how to improve quality. Br J Dermatol. 2018;178(6):1257-1262. PubMed CrossRef

103. Vandenbroucke JP. Case reports in an evidence-based world. J R Soc Med. 1999;92(4):159-163. PubMed CrossRef

104. Howick J, Chalmers I, Glasziou P, et al. The 2011 Oxford CEBM Evidence Levels of Evidence (Introductory Document). Center for Evidence-Based Medicine website. http://www.cebm.net/index.aspx?o=5653. 2011. Accessed May 18, 2018.

105. Charlton BG, Walston F. Individual case studies in clinical research. J Eval Clin Pract. 1998;4(2):147-155. PubMed CrossRef

106. Albrecht J, Meves A, Bigby M. Case reports and case series from Lancet had significant impact on medical literature. J Clin Epidemiol. 2005;58(12):1227-1232. PubMed CrossRef

107. Kellner CH. Relapse after electroconvulsive therapy (ECT). J ECT. 2013;29(1):1-2. PubMed

108. Shibasaki C, Takebayashi M, Fujita Y, et al. Factors associated with the risk of relapse in schizophrenic patients after a response to electroconvulsive therapy: a retrospective study. Neuropsychiatr Dis Treat. 2015;11:67-73. PubMed CrossRef

109. Suzuki K, Awata S, Matsuoka H. One-year outcome after response to ECT in middle-aged and elderly patients with intractable catatonic schizophrenia. J ECT. 2004;20(2):99-106. PubMed CrossRef

110. Fox HA. Continuation and maintenance electroconvulsive therapy-a conceptual framework? J ECT. 2015;31(2):e27-e28. PubMed CrossRef

111. Elias A, Chathanchirayil SJ, Bhat R, et al. Maintenance electroconvulsive therapy up to 12 years. J Affect Disord. 2014;156:228-231. PubMed CrossRef

112. Elias A, Phutane VH, Clarke S, et al. Electroconvulsive therapy in the continuation and maintenance treatment of depression: Systematic review and meta-analyses. Aust N Z J Psychiatry. 2018;52(5):415-424. PubMed CrossRef

113. Gill SP, Kellner CH. Clinical practice recommendations for continuation and maintenance electroconvulsive therapy for depression: outcomes from a review of the evidence and a consensus workshop held in Australia in May 2017 [published online ahead of print February 7, 2018]. J ECT. PubMed CrossRef

114. Santos Pina L, Bouckaert F, Obbels J, et al. Maintenance electroconvulsive therapy in severe bipolar disorder: a retrospective chart review. J ECT. 2016;32(1):23-28. PubMed CrossRef

115. Ward HB, Szabo ST, Rakesh G. Maintenance ECT in schizophrenia: a systematic review. Psychiatry Res. 2018;264:131-142. PubMed CrossRef

116. Rodriguez-Jimenez R, Bagney A, Torio I, et al. Maintenance electroconvulsive therapy: cost-effectiveness and patient/family Satisfaction. J ECT. 2015;31(4):279. PubMed CrossRef

117. Petrides G, Tobias KG, Kellner CH, et al. Continuation and maintenance electroconvulsive therapy for mood disorders: review of the literature. Neuropsychobiology. 2011;64(3):129-140. PubMed CrossRef