Clinical Features Associated With Response to Ketamine vs Electroconvulsive Therapy for Treatment-Resistant Depression

Allen,Justin;Lalangi Marasinghe, MD;Haley Schuster, MD;Mullen,Mark;Rajesh R. Tampi, MD, MS, DFAPA, DFAAGP;

A monthly column dedicated to reviewing the literature and sharing clinical implications.

Treatment-resistant depression (TRD) is a serious psychiatric illness for which treatment options are limited. This study attempts to help guide clinicians in deciding between electroconvulsive therapy (ECT) and ketamine in patients with TRD.

To evaluate whether selected clinical features of patients with TRD were associated with differential improvement with ketamine vs ECT.

This study was a secondary analysis of a phase 2/3 open-label, noninferiority randomized clinical trial (ELEKT-D; NCT03113968) conducted at 5 academic medical centers. The patient population included outpatients and inpatients with nonpsychotic TRD who were referred by their clinical provider and were eligible for treatment with ECT. The primary outcome for the trial was a 50% or more reduction in the Quick Inventory of Depressive Symptomatology Self-Report (QIDS-SR16) score from baseline. The clinician-rated Montgomery-Åsberg Depression Rating Scale (MADRS) was an additional measure of depression severity.

Participants were randomly assigned 1:1 to receive either 9 treatments of ECT at 6 times seizure threshold or 6 infusions of ketamine at 0.5 mg/kg of body weight. Treatments were given over 3 weeks. In the ELEKT-D trial, ketamine was found to be noninferior to ECT for the treatment of TRD without psychosis.

For the secondary analysis of ELEKT-D, baseline features evaluated included the following: premorbid intelligence (the North American Adult Reading Test-35 [NAART-35]), baseline depression severity (QIDS-SR16 or MADRS), cognitive functioning, concurrent use of benzodiazepine or atypical antipsychotic, body mass index, history of suicide attempt, inpatient vs outpatient status, anxious features present, and the presence of comorbid posttraumatic stress disorder (PTSD). Descriptive statistics were used to evaluate baseline features for associations with differential improvement between ketamine and ECT. Separate models analyzed QIDS-SR16 and MADRS depression severity, with repeated measures mixed-effects models for continuous outcomes and logistic regression for response and remission. Interaction terms tested whether baseline features predicted differential improvement with ketamine vs ECT, with post hoc stratification for significant findings. This was further examined with exploratory analyses stratified by treatment groups. Benjamini-Hochberg procedure controlled for false discovery rates (FDR) that occurred due to multiple comparisons, but unadjusted 2-sided P values (< .05) were also reported.

Study Strengths

1. There were similar clinical and demographic features between treatment groups.

2. The study used 2 scales to measure depression severity, 1 self-reported and 1 clinician rated.

3. The list of baseline factors evaluated was comprehensive.

Study Weaknesses

1. The number and duration of treatments given were short, ending after 3 weeks.

2. The study was open-label and not placebo-controlled.

3. Findings were not consistent between patient-rated QIDS-SR16 and clinician-rated MADRS.

4. The study classified TRD as “moderately severe or severe” vs “very severe” based on the QIDS-SR16 scale. Such classification may not be routinely used in clinical practice, limiting the clinical utility of the identified characteristics.

A total of 365 participants were included in the study. There were 195 participants randomly assigned to the ketamine group and 170 to the ECT group. Of the 203 participants originally allocated to the ECT group, 31 did not receive treatment intervention and 2 were lost to follow-up. Of the 200 participants originally allocated to the ketamine group, 4 did not receive treatment intervention and 1 was lost to follow-up. Baseline demographic and clinical characteristics were similar between the treatment groups. The mean age of participants was 46.0 years.

The study found that baseline QIDS-SR16 (P = .004) and inpatient status at first treatment (P < .001) were significantly associated with treatment response with ketamine vs ECT after controlling for the FDR. Participants with moderately severe or severe depression (defined as a baseline QIDS-SR16 score ≤ 20) had a 7.7-point decrease in the QIDS-SR16 score in the ketamine group compared with a 5.6-point decrease in the ECT group. Participants with very severe depression (defined as a baseline QIDS-SR16 score > 20) had an 8.4-point decrease in the QIDS-SR16 score in the ECT group compared with a 6.7-point decrease in the ketamine group by week 2; however, scores were similar at the end of week 3 (–9.0 vs –9.9). Participants who initiated treatment as outpatients had an 8.4-point decrease in the QIDS-SR16 score in the ketamine group compared with a 6.2-point decrease in the ECT group. Participants who initiated treatment as inpatients had a 10.9-point decrease in the QIDS-SR16 score in the ECT group compared with an 8.0-point decrease in the ketamine group. When using the MADRS score as the response measure, no significant differences were found in the treatment response with ketamine vs ECT after controlling for the FDR.

The study also examined whether different baseline features were associated with improved treatment response within the ECT and ketamine groups separately. After controlling for the FDR, when the MADRS was used as the response measure, several baseline features were associated with improvement in the ECT group. Those with higher premorbid intelligence (ie, NAART scores ≥ 85) receiving ECT had a 14.0-point decrease in the MADRS compared with an 11.2-point decrease in the participants with lower premorbid intelligence. Participants with comorbid PTSD receiving ECT had a 16.6-point decrease in the MADRS score compared with a 12.0-point decrease in those without PTSD. Participants with impaired memory recall receiving ECT had a 13.4-point decrease in the MADRS score compared with a 9.6-point decrease in the participants with unimpaired memory recall by week 2; however, MADRS scores were similar at the end of treatment (–14.3 vs –12.2). When using the QIDS-SR16, no baseline features were associated with greater improvement for ECT. When using either the MADRS or the QIDS-SR16, no baseline features were associated with greater improvement for ketamine.

Evidence from this secondary analysis of the ELEKT-D trial of ketamine vs ECT for nonpsychotic TRD indicated that greater treatment response was observed with the use of ketamine among outpatients, and that greater treatment response was observed with the use of ECT among inpatients. Ketamine led to greater treatment response in participants with “moderately severe or severe” TRD, whereas ECT led to greater treatment response in participants with “very severe” TRD. In this open-label trial, these findings were observed when the participant-rated QIDS-SR16 was used as the response measure; however, when the clinician-rated MADRS was used as the response measure, no baseline features were associated with differential improvement with ECT vs ketamine.

When clinicians and patients are faced with the task of deciding between ECT and ketamine for nonpsychotic TRD, this study can inform shared decision-making.

Among patients with TRD, both ECT and ketamine are efficacious treatment options to consider. Based on the data from this study, ketamine may be preferred among outpatients and for those with “moderately severe or severe” depression, whereas ECT may be preferred among inpatients and those with “very severe” depression. Future research is indicated to replicate and further support the findings in this study.

Dr Allen is a third-year psychiatry resident at Creighton University in Omaha, Nebraska. Dr Marasinghe is a third-year psychiatry resident at Creighton University. Dr Schuster is a fourth-year psychiatry resident at Creighton University. Dr Mullen is an assistant professor of psychiatry at Saint Louis University School of Medicine in Missouri. Dr Tampi is professor and chair of the Department of Psychiatry at Creighton University School of Medicine and Catholic Health Initiatives Health Behavioral Health Services. He is also an adjunct professor of psychiatry at Yale School of Medicine in New Haven, Connecticut, and a member of the Psychiatric Times editorial board.