Relapse prevention after index electroconvulsive therapy in treatment-resistant depression

Abstract

Background

One-third of patients who suffer from depression are resistant to conventional treatments. An acute course of electroconvulsive therapy (ECT) can lead to remission of depressive symptoms in a substantial portion of the treatment-resistant patients. However, prevention of relapse with depressive symptoms after the index course of ECT can be challenging. We review pertinent studies on the topic and analyze the best strategies to avoid relapse and recurrence of depressive symptoms.

Methods

We performed a systematic literature review of PubMed through April 2014 for clinical trials published in English to determine if continuation ECT, continuation medication, continuation psychotherapy, or combinations of these are the best strategy to avoid relapse and recurrence of depressive symptoms after an acute course of ECT. Clinical trials comparing ≥2 of the above strategies were included in the review.

Results

Although there are few rigorous randomized clinical trials in this area, most studies suggest that combined continuation ECT (C-ECT) and continuation pharmacotherapy are the most effective strategy in relapse prevention.

Conclusions

C-ECT and continuation pharmacotherapy may be more effective than either alone for preventing relapse. However, more definitive randomized clinical trials are needed.

INTRODUCTION

Approximately 20.9 million American adults a year suffer from a mood disorder.¹ Major depressive disorder (MDD) is the most prevalent mood disorder, and the leading cause of disability among Americans age 15 to 44.² It also is a risk factor for mortality including death due to suicide and other medical conditions. One-third of patients do not respond to pharmacotherapy (including medication combinations) and psychotherapy.³ Among treatment-resistant patients, >100,000 patients per year in the United States are treated with electroconvulsive therapy (ECT)⁴; relapse after ECT is common,⁵ however, and strategies for reducing the risk of relapse after ECT are under-studied. This review will focus on prevention of relapse or recurrence of depressive episodes of mood disorders after a successful acute course of ECT (depressive episodes being the most common indication for ECT).

Psychotherapeutic interventions such as cognitive-behavioral therapy can help patients with mild to moderate depressive disorders and should be tried early in treatment. Psychopharmacological agents are indicated for severe depression.^6,7 Unfortunately many patients do not respond or experience remission while taking psychopharmacological agents. For example, the response rate to a first antidepressant is 50%, and remission is only 37%.

Definition of response, remission, and relapse in ECT trials

In antidepressant trials, treatment response often is defined as a reduction in Hamilton Rating Scale for Depression (HRSD) Score of ≥50%.⁸ In psychotherapy trials, treatment response often is defined similarly to pharmacotherapy trials.⁹ However, in ECT clinical trials, response often is defined as reduction in HRSD₂₄ of ≥60% after an ECT course. The definition of remission in many ECT trials is symptom reduction of ≥60% on HRSD scores and a final HRSD score ≤10.^10–12

Some studies¹³ had 2 definitions of remission: “moderate and strict criteria.” In Prudic et al, the “moderate criteria (remitter10)” is a reduction of ≥60% on HRSD scores and a final HRSD score of ≤10. The “strict criteria (remitter7)” is a reduction of ≥60% on HRSD scores and a final HRSD score of ≤7.¹³

Relapse in clinical trials often is defined as the return of full syndromal symptomatic criteria. Alternatively, relapse is defined as a HRSD score of ≥16 and an absolute increase of 10 points for at least 1 week (over 2 consecutive visits).¹¹

As mentioned, the response rate to the first antidepressant is approximately 50% and the remission rate is 37%.³ This rate falls with each successive antidepressant trial, reaching 14% after the third trial and 13% after the fourth trial according to the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial.³ A total of 33% of patients do not achieve symptomatic remission (let alone functional recovery) despite multiple trials of medication (including augmentations with multiple medications).³ These represent millions of Americans each year who do not experience remission of depression,^14,15 and are deemed treatment-resistant. Of the patients that are resistant to conventional therapies, approximately 60% responded to ECT.^16,17

After a successful acute course of ECT, maintaining remission from depressive symptoms is a major challenge for clinicians and patients, because the relapse rate has been reported to be as high as 84% within the first 6 months after ECT.¹¹ To address this challenge, we performed a systematic review of the literature to best inform the course of continuation treatment.

METHODS

A systematic literature review of PubMed was performed through April 2014 for clinical trials published in English to determine whether continuation ECT, continuation medication, continuation psychotherapy, or combinations of these are the best strategy to avoid relapse and recurrence of depressive symptoms after an acute course of ECT. Clinical trials comparing ≥2 of the above strategies were included in this systematic review. Keywords used were (ECT OR electroconvulsive therapy OR convulsive therapy) AND (depression OR depressive disorder OR depressive episode) AND (maintenance OR continuation).

RESULTS

The options for continuation treatment are: (1) pharmacotherapy, (2) psychotherapy, (3) continuation ECT, or (4) a combination of the above interventions.

Continuation pharmacotherapy with or without continuation psychotherapy

A common practice after an index course ECT is to prescribe antidepressants for continuation therapy.¹⁸ Although earlier controlled trials of continuation medication alone (C-MED) reported encouraging results,^19,20 C-MED alone is not effective for all patients, with some studies showing high relapse rates.^13,21,22

Combinations of psychotropics might be better for preventing relapse than a single drug. When a combination of ≥2 drugs is used, lithium augmentation seems to lead to better results than antidepressants alone. This has been shown in a well-designed randomized, double-blinded, placebo-controlled clinical trial that enrolled 290 participants. Of these participants, 159 met remission criteria, and 84 (who remitted and agreed to participate) were randomized to 3 arms after remission after acute ECT.¹¹ Participants were randomized to the continuation arms for 24 weeks after depressive remission to receive nortriptyline alone, a combination of nortriptyline and lithium, or placebo. The study found that a combination of nortriptyline and lithium reduced the relapse rate (39%) compared with nortriptyline monotherapy (60%) or with placebo (84%) over 6 months.¹¹ Of note in this study, every relapse using nortriptyline-lithium occurred before the 5-week period except for 1 patient.

In another controlled trial after successful index ECT, C-MED resulted in a relapse rate of 51% within 6 months for treatment as usual, 60% for protocolized care with nortriptyline, and 39% for protocolized care with nortriptyline plus lithium (39%), vs a relapse rate of 84% for placebo.²³ This study included 56 randomized patients and was single-blinded.

In clinical practice, personalized treatment decisions are based on clinical, demographic, and logistic characteristics, with consideration of predictors of response, predictors of relapse, patient preferences, and long-term adverse effects. Of note, although cognitive adverse effects can be cause for concern in some cases during an acute course of ECT (especially bilateral ECT), during C-ECT cognitive side effects are not as much of a concern because the sessions are not timed as close together.

Logistic concerns include the availability of a family or friend who can transport the patient to continuation ECT (C-ECT) sessions and the ability to take time-off (if working) for C-ECT. Factors that predict higher likelihood of relapse after an acute course of ECT are in Table 1. Relapse prevention strategies are in Table 2.

Table 1.

Common relapse predictors

Medication resistance Female sex Residual depressive symptoms Comorbid untreated psychiatric disorders such as posttraumatic stress disorder, substance use disorder, or personality disorder History of unsuccessful treatment with ECT if the treatment parameter and electrode placement were optimized Highly recurrent major depressive disorder Depressive symptoms as part of a rapid cycling bipolar disorder Uncontrolled medical condition that is inducing or exacerbating the mood disorder such as untreated hypothyroidism or chronic severe pain disorder

Table 2.

Relapse prevention strategies

Starting continuation pharmacotherapy during the index course Aggressively treating residual symptoms after an index course of ECT Tailoring C-ECT session to the level of symptoms and clinical presentation without over or under treatment, rather than using a fixed schedule for continuation ECT Combined ECT and pharmacotherapy including possibility of lithium augmentation Gradually discontinuing (or minimizing if absolutely necessary) benzodiazepine use and other CNS depressants to both avoid the depressogenic effect of these substances as well as to avoid interference with seizure induction during ECT Treating other comorbid psychiatric conditions (formerly Axis I disorders) that might be hindering treatment such as posttraumatic stress disorder, substance use comorbidity etc. Treating comorbid personality disorder that might be hindering treatment such as treating borderline personality disorder with dialectical behavioral therapy Treating comorbid medical condition that could be exacerbating (or inducing) mood disorder such as treating hypothyroidism

C-ECT: continuation electroconvulsive therapy; ECT: electroconvulsive therapy.

Continuation ECT (C-ECT)

Because the difference between maintenance ECT and C-ECT is arbitrary, for the purpose of this article we will refer to both as C-ECT. C-ECT indicates any treatment session after the index course administered with an intention to prevent relapse or recurrence of depressive symptoms rather than to induce response/remission of an episode.

Relapse rates with placebo continuation therapy after index ECT are high.¹¹ Several studies verify the benefits of C-ECT in sustaining remission.^21,24 This might be helpful, especially in cases when patients did not respond to several pharmacological agents,²¹ are sensitive to medication side effects, or prefer ECT over medication.

A controlled, well-designed trial in this area led by Kellner and colleagues (Consortium for Research in ECT, CORE) evaluated the comparative efficacy of C-ECT vs combination C-MED (lithium carbonate plus nortriptyline) in preventing depressive relapse over a 6-month period after successful bilateral (BL) ECT for acute depression. This randomized, controlled clinical trial enrolled 201 participants in the phase II comparison of C-MED vs C-ECT. The study was single blinded for neuropsychological tests. This study found equal sustained remission rate of 46% throughout the study period for both the C-MED and the C-ECT arm of the study.²⁵

In a recent meta-analysis and systematic review of literature on post-ECT relapse that included 32 studies, the authors reported that in modern studies of C-MED, 51% of patients relapsed by 12 months after a successful acute course of ECT. A total of 37.7% of patients who relapsed did so within the first 6 months. The relapse rate at 6 months with C-ECT was similar to C-MED (37.2%).²⁶

Another recent literature review on C-ECT concluded that ECT is “an underused treatment option that can substantially reduce risks of relapse.”²⁷ The factors favoring consideration of C-ECT (with or without pharmacotherapy) are listed in Table 3.

Table 3.

Factors favoring consideration of continuation ECT

Positive response to index ECT Patient preference Failure of pharmacotherapy and psychotherapy to achieve remission in the past Intolerance to effective pharmacological agents or side effects unacceptable to the patient Limited medical illness that could increase risk of further ECT or anesthesia Trials to discontinue ECT result in rapid reemergence of depressive symptoms despite optimization of pharmacological interventions and/or pharmacotherapy Potential for high stakes complications is there were a relapse, ie, return of intense suicidality or catatonia No barriers to the logistic issues in complying with outpatient ECT (such as transportation and availability of a friend or family to be involved in care; ability to take a day off for the ECT procedure, if applicable)

ECT: electroconvulsive therapy.

Combination of C-ECT and pharmacotherapy or psychotherapy

A general principle with antidepressant therapy is that treatment for the acute phase also is effective as continuation treatment. However, after acute response to ECT, the most common clinical practice is to stop ECT and to use only medications as continuation therapy²⁸ despite the fact that some patients who undergo ECT have not responded to medications or psychotherapy before ECT.

Long-term remission may be preserved with combined C-ECT and pharmacotherapy, especially when symptoms did not respond to medication before ECT. Retrospective studies of hospitalization rates have reported lower rates of psychiatric hospitalization in patients receiving C-ECT with medications compared with admission rates before receiving ECT alone.^29,30

Another retrospective study comparing C-ECT plus medication (C-ECT group) vs C-MED alone (C-MED group) demonstrated significant advantage for the C-ECT group.³¹ The 2-year cumulative probability of surviving without relapse or recurrence for the ECT group was 93% while the 2-year cumulative probability of surviving without relapse or recurrence was only 52% for the C-MED group. At 5 years, the C-ECT group had relapse/recurrence free survival of 73%, while the C-MED group had relapse/recurrence free survival of 18%. The weaknesses of this study included its retrospective nature. The sample size was 59 (29 cases and 29 controls). Although on the high end for ECT studies, sample size was modest to allow for adjustments for some of the potential confounding variables. Because of the retrospective nature, there was no random assignment of the groups or blinding. The mean survival time for staying well for the C-ECT group was 6.9 years, while for the C-MED group it was 2.7 years.³¹ These results support the superior efficacy of C-ECT plus medications compared with C-MED alone in preventing relapse and recurrence in patients with depression who have responded to acute ECT.

A randomized trial by Brakemeir et al³² examined the most efficacious continuation intervention for patients who responded to acute ECT. After response—not remission—to acute right unilateral (RUL) ultra-brief ECT, 60 patients were randomized to cognitive-behavioral therapy (C-CBT) plus “guideline-based” C-MED, C-ECT (RUL ultra-brief) plus C-MED, or C-MED alone. The study found superiority in the C-CBT group on the proportion of patients who “remained well” after 12 months (primary outcome). In this study, 70% of the initial 90 patients responded to acute ECT, and 47% remitted. After 6 months, sustained response rates were 77% in the C-CBT arm, 40% in the C-ECT arm, and 44% in the C-MED alone arm. After 12 months, sustained response rates were 65% in the C-CBT group, 28% in the C-ECT group, and 33% in the C-MED alone group. The authors concluded that “ultra-brief pulse ECT as a continuation treatment correlates with low sustained response rates,” and that C-CBT plus C-MED “might be an effective continuation treatment to sustain response after successful ECT in MDD patients.”

During the initial acute ECT phase, participants were switched to BL ECT if they did not respond to RUL ECT. Nonetheless, during the randomization continuation phase, all participants in the C-ECT were given RUL ECT. It is not clear how many of the participants who had BL ECT were treated later with RUL ECT during the continuation phase, but this could make ECT appear to be ineffective during the continuation phase, because the BL ECT that was previously effective was discontinued. For the above reasons, it is hard to draw conclusions of the superior efficacy of C-CBT compared with the ultra-brief unilateral ECT used in this study without more rigorously replication of this study.

Another small prospective study examined 21 participants who received C-ECT (12 depressed and 8 schizoaffective), and 21 matched controls (received C-MED) who refused C-ECT. This study compared patients who received C-ECT plus pharmacotherapy with those who received C-MED alone. Limitations of this study include small sample size, no random assignment, no blinding, and the C-ECT group was not homogenous including both RUL and BL ECT and mixed diagnoses. However, the study was prospective and had a matched control group. The study found a significantly better outcome for patients receiving C-ECT plus pharmacotherapy compared with C-MED alone in terms of time to rehospitalization.³³

Another prospective, randomized, single-blind trial assessed tolerability and efficacy of continuation treatments for geriatric patients with psychotic depression after acute ECT remission.³⁴ The researchers compared the 2-year period outcome of 2 groups of geriatric patients with psychotic major depression who remitted after acute treatment of ECT plus nortriptyline. During the continuation phase, one group was randomized to nortriptyline alone (C-MED) (n = 17), and the other group was randomized to C-ECT plus nortriptyline (n = 16) (C-ECT). Over the 2 years of treatment, the mean survival time of remission was significantly longer for the combined C-ECT plus C-MED group compared with the C-MED group. There was no difference in tolerability between the 2 groups. The researchers concluded that the study supported the use of combined C-ECT plus antidepressant treatment in geriatric patients with psychotic major depression who remitted using acute ECT.

Another prospective, multicenter, randomized, controlled trial tested the “hypothesis that relapse prevention with continuation ECT plus pharmacotherapy is more effective than pharmacotherapy alone” after a course of acute ECT for major depressive episode (unipolar or bipolar depression).³⁵ The study randomized 2 parallel groups (after response to acute ECT) in 4 hospitals in Sweden. Participants (N = 56) were randomly assigned to either 29 C-ECT sessions with C-MED or C-MED alone for 1 year. The C-MED consisted of antidepressants (98%), lithium (56%), and antipsychotics (30%). The primary outcome was relapse within 1 year. (Relapse was defined as ≥20 points on the Montgomery-Åsberg Depression Rating Scale or inpatient psychiatric care or suicide or suspected suicide). Sixty-one percent of participants relapsed within 1 year in the C-MED alone arm vs 32% of those in the C-ECT plus C-MED arm (P = .036). (The Cox proportional hazard ratio was 2.32 (95% confidence interval, 1.03–5.22).) In the C-MED arm, there was 1 suspected suicide by intoxication (died from intoxication, the suicidal intent was not proven, and was considered in the study as a suspected suicide) and 3 suicide attempts but none in the combined C-ECT arm. The authors concluded that there was a “statistically significant advantage for combined treatment with pharmacotherapy and continuation ECT.”

The Prolonged Remission in Depressed Elderly (PRIDE) study is an ongoing multicenter, randomized clinical trial by the CORE group mentioned above. The PRIDE study examines whether combined pharmacotherapy (lithium plus venlafaxine) plus C-ECT, or combined lithium plus venlafaxine C-MED alone, is more effective in maintaining remission in depressed older patients with MDD after acute ECT. Data from this study will be valuable in informing clinical design making of combining ECT with these medications.

In summary, most studies have shown that combined C-ECT plus continuation pharmacotherapy (C-MED) have significantly better results in relapse prevention than either alone.^33–35 One study showed that C-CBT combined with C-MED is better that C-ECT combined with C-MED. However, it is hard to draw conclusions or recommend C-CBT as superior to ECT based on a single study because of design limitation, and results that are different from most of the literature.

Other considerations in continuation ECT and pharmacotherapy in clinical practice

There are other important points that need to be considered in the routine clinical aspects of continuation/maintenance therapy. First, it is helpful to have a rating scale to consistently and systematically assess improvement or relapse in depressive symptoms during acute ECT and continuation treatment. Although many clinical trials have used HRSD, clinicians can use simpler, self-administered scales, including the 9-item Patient Health Questionnaire.³⁶

Second, clinicians should individualize treatment decisions by weighing risks and benefits of psychotropics, especially lithium, among other alternatives for treatment-resistant depressive episodes of unipolar and bipolar disorders. Lithium can have several side effects including nausea, thyroid dysfunction, and renal dysfunction, and blood level monitoring (for lithium level, renal, and thyroid functions) is required. However, if used with caution and blood tests are checked regularly, most patients tolerate the drug well.³⁷ The risk from chronic lithium use includes renal toxicity.³⁸ If kidney function is monitored closely, however, lithium generally can be used safely in most cases, even for long periods of time.^38,39 Also, a systematic review and meta-analysis of 23 studies concluded that “any lithium-associated increase in serum creatinine is quantitatively small and of questionable clinical significance.”⁴⁰ However, routine monitoring of renal function is essential. Lithium augmentation plays an important role for managing treatment-resistant depression.⁴¹ It has potential benefits not only in treatment-resistant depression,⁴² but also in reducing suicide ⁴³ in bipolar and unipolar depression.^44,45 The drug also has a neuroprotective effect.⁴⁶

Third, although cognitive side effects—which can be of concern during acute ECT—are a possibility, they usually improve with time. Acute side effects of ECT have been discussed extensively elsewhere⁶ and are not the focus of this article. For the purposes of this article, C-ECT has less cognitive side effects than acute ECT, simply because of the timing of treatment sessions. Among C-ECT patients there likely is additional cognitive sparing as the treatment sessions are moved farther and farther apart. However, although this represents our clinical experience and the general consensus in the field, it has not been verified by empirical research. There are no studies comparing cognitive effects of ECT with session spacing during C-ECT, for example between patients getting ECT once per week vs once per month.

Fourth, one of the barriers to pharmacotherapy is the high rates of medication non-adherence.⁴⁷ Adherence to antidepressant treatment may range from 30% to 70%.⁴⁸ A study by Cramer and Rosenheck⁴⁹ examining pooled data reports that adherence rates for antidepressants rarely exceed 65%. Others found 37% of patients discontinued antidepressants by 12 weeks in the counselled group compared with 61% in the non-counselled group.⁵⁰ Another study found that only 20% to 34% of individuals filled ≥4 prescriptions.⁵¹

Physicians cannot predict medication adherence in many patients.⁴⁷ Non-adherence commonly is encountered in outpatient care not only with older medications such as tricyclic antidepressants, but also with newer medications such as selective serotonin reuptake inhibitors.⁴⁸ However, studies indicate that adherence for C-MED might be less than outpatient C-ECT. Dew and McCall found an outpatient C-ECT adherence rate of 91%.⁵² This is could be attributed to ECT being a more potent and rapidly acting form of treatment for treatment-resistant depressive episodes and/or suicidality.

As directions for further research, we recommend that further investigations in refinements of C-ECT and continuation pharmacotherapeutics combinations—and psychotherapy especially in patients who were non-treatment resistant to psychotherapy—with ECT are needed to further improve relapse prevention strategies after index ECT, especially for treatment-resistant mood disorders.

CONCLUSIONS

Although there are few rigorous randomized clinical trials in this area, the current literature suggests that combined C-ECT and pharmacotherapy may be more effective than either alone in preventing relapse. However, more definitive randomized clinical trials to guide the best strategies to avoid relapse after an acute course of index ECT are needed.

Table 4.

Prospective clinical trials and meta-analysis involving combining C-ECT with C-MED (or C-CBT)

Study	Design, N	Interventions	Treatment details/ECT/anesthesia	Primary outcome measure(s)	Baseline	Endpoint/outcomes	Comment
Nordenskjöld 201335	Multicenter, randomized (2 parallel groups), nonblinded, controlled trial	After response to acute ECT for MDE (unipolar or bipolar depression) participants were randomized (N = 56) to 29 C-ECT sessions with C-MED (n = 28) or C-MED alone (n = 28) for 1 year. The C-MED consisted of antidepressants (98%), lithium (56%), and antipsychotics (30%)	ECT Weekly ECT for 6 weeks, then every 2 weeks for 46 additional weeks (total of 29 ECT sessions). UL UB ECT (6 times ST) was used. Mecta Spectrum 5000Q device at 3 hospitals and a Thymatron System IV (Somatics, Inc., Lake Bluff, Ill) was used at one hospital. Propofol or thiopental was used as anesthetic, and succinylcholine as muscle relaxant Pharmacotherapy Patients received individualized medication. Venlafaxine was first choice, with lithium augmentation offered to all patients. Antipsychotics were allowed. Pharmacotherapy at randomization consisted of antidepressants (98%), lithium (56%), and antipsychotics (30%). If the patient experienced a relapse, any treatment change was allowed, including ECT for all patients regardless of initial allocation	Relapse within 1 year. (Relapse was defined as ≥20 points on the MADRS, psychiatric hospitalization, suicide, or suspected suicide)	Responders (not remitters) to acute ECT randomized (N = 56)	61% relapsed within 1 year in the C-MED alone arm vs 32% of those in the C-ECT plus CMED arm (P = .036). (The Cox proportional hazard ratio was 2.32 (1.03 to 5.22). In the C-MED arm there was 1 suspected suicide and 3 suicide attempts but not in the combined C-ECT arm	The authors concluded that there was a “statistically significant advantage for combined treatment with pharmacotherapy and continuation ECT”
Navarro et al 200834	Prospective, randomized, single-blind trial (N = 33) randomized Patients age ≥60 with psychotic major depression (in a MDE) were recruited	After remission with acute ECT, one group was randomized to nortriptyline alone C-MED (n = 17), and the other group was randomized to C-ECT plus nortriptyline (n = 16) (C-ECT)	Acute ECT (N = 38) Bilateral ECT 3 times per week with MECTA-SR2 plus nortriptyline (to obtain blood level of 80 and 120 ng/mL [maximum dose: 100 mg/d]) until remission Anesthesia: Succinylcholine (40 to 100 mg), atropine (0.5 to 1 mg), and thiopental (200 to 300 mg) Continuation randomization phase (N = 33): C-ECT group (n = 16). C-ECT plus nortriptyline was given. After 1 week of the acute phase, weekly ECT session for 1 month was administered, then every 2 weeks for 1 month, and then monthly ECT stimulus and the anesthetics were similar to acute ECT. C-MED group (n = 17). Nortriptyline combined with risperidone (up to 2 mg/d) for 6 weeks, then withdrawn over a 4-week period	Time to relapse/recurrence at 2-years (secondary: HDRS-17	Remitted (N = 33) after acute ECT randomized in phase II HDRS (considered remitters when the HDRS-17 score was < 8 at 3 consecutive ECT session)	Over 2 years of treatment in geriatric, psychotic, unipolar depressed ECT (plus nortriptyline) remitters, the mean survival time of remission was significantly longer for the combined C-ECT plus C-MED group (n = 12 completers; 11 Without relapse or recurrence) compared with the C-MED group (n = 13 completers; 5 Without relapse or recurrence). There was no difference in tolerability between the 2 groups	The researchers concluded the study supports combined C-ECT plus antidepressant treatment in geriatric patients with psychotic major depression who remitted with acute ECT
Brakenmeier et al 201332	Single site, 3-arm parallel-group randomized study after acute ECT, 90 inpatients with MDD, including 60 who responded to acute ECT	Randomized (N = 60) C-MED+C-CBT (n = 17) (n = 16 completers) C-MED+C-ECT (n = 25) (n = 18 completers) C-MED alone (n = 18) (n = 14 completers)	C-ECT arm: Acute ECT: 3 times/week, with Mecta 5000Q device RUL UB ECT. If no improvement switched to BL ECT Anesthesia: propofol C-ECT: with RUL UB ECT Anesthesia: propofol C-CBT: Group CBT C-MED alone: In all groups, C-MED included a guideline medication algorithm. There was no standard medication for all patients	Proportion of patients who remained well at 12 months	Responders (not remitters) after acute ECT randomized (N = 60) in phase II	C-MED only group 14 completed 12-month response rates: 33%. C-ECT group, 18 completed 12-month response rates: 28%. C-CBT group 17 completed 12-month response rates: 65%	The authors concluded: “this study suggests, for the first time, that combining CBT with individualized antidepressant therapy is an intervention that increases sustained response and decreases dropout rates in depressive patients who have previously responded to ECT. The study provides, in addition, evidence that ECT with ultra-brief pulses in conjunction with propofol anesthesia as used here in a therapy-resistant population, significantly reduces continuation ECT efficacy. In sum, the study might be a useful initial step in a long-term psychotherapeutic treatment regime for patients after ECT”

BL: bilateral; ECT: electroconvulsive therapy; HDRS: Hamilton Depression Rating Scale; MADRS: Montgomery-Åsberg Depression

Rating Scale; MDE: major depressive episode; ST: seizure threshold; UB: ultrabrief pulse ECT; UL: unilateral.