Mood and neuropsychological effects of different doses of ketamine in electroconvulsive therapy for treatment-resistant depression

Zhong and colleagues situate the study within the clinical problem of treatment-resistant depression (TRD), noting that a substantial minority of patients fail to remit after adequate trials of antidepressant drugs. Electroconvulsive therapy (ECT) is presented as an effective option for TRD, but response rates when common anaesthetic agents (for example propofol, thiopental or etomidate) are used are only around 50–60%, motivating investigation of augmentation strategies. Ketamine, an NMDA receptor antagonist with rapid antidepressant effects at low intravenous doses, has been proposed as an anaesthetic agent for ECT to improve antidepressant outcomes and possibly protect cognition, but prior ECT studies have produced inconsistent results and suggested a possible dose–response relationship for ketamine's mood effects. This study set out to compare three anaesthetic approaches for ECT in TRD: an anaesthetic dose of ketamine, a subanaesthetic ketamine plus propofol combination, and propofol alone. The investigators aimed to evaluate effects on depressive symptoms, seizure parameters during ECT, neurocognitive outcomes, and safety, with particular interest in whether anaesthetic versus subanaesthetic ketamine produces different antidepressant and cognitive effects.

This randomised, double-blind clinical trial enrolled 90 inpatients with TRD at a single centre between April 2011 and April 2014. Inclusion required a current major depressive episode (major depressive disorder or bipolar disorder) meeting ICD-10 criteria and failure to respond to at least two antidepressants of different classes at adequate dose and duration. Major exclusions were other primary psychiatric diagnoses (for example schizophrenia or dementia), seizure history, substance abuse, pregnancy, neurological disorders, serious physical illness or contraindications to ECT or anaesthesia. Participants were randomised to one of three anaesthetic regimens (n = 30 per group): ketamine 0.8 mg/kg intravenous (IV) push (ketamine group); ketamine 0.5 mg/kg IV plus propofol 0.5 mg/kg IV (ketamine þ propofol group); or propofol 0.8 mg/kg IV (propofol group). Both raters and patients were blinded to allocation. All patients received atropine 1 mg and succinylcholine 1 mg/kg for muscle relaxation. Bilateral ECT was administered three times per week for three weeks (total eight treatments) using a Thymatron System IV; seizure threshold was estimated by the half-age method. No antidepressant or antipsychotic medications were given during the ECT course. Primary clinical assessments were the 17-item Hamilton Depression Rating Scale (HDRS-17) and the 18-item Brief Psychiatric Rating Scale (BPRS-18), administered at baseline, after ECT sessions 1, 2, 3, 4 and 6 (on the morning before the next scheduled ECT) and 48–72 hours after the eighth (final) ECT. Cognitive testing (Word Fluency, Digit Symbol, Digit Span, Wisconsin Card Sorting Test (WCST), Tower of Hanoi, Trail Making Test (TMT) Parts A and B, and Visual Reproduction) was done at baseline and 48–72 hours after the eighth treatment. Seizure parameters (seizure duration, electric quantity delivered, seizure energy index) and systolic/diastolic blood pressures were recorded. Response was defined as ≥50% reduction in HDRS-17 score; remission as HDRS-17 ≤7. Statistical analyses used SPSS 18.0. Repeated-measures analyses (General Linear Model) tested effects of treatment group and time on HDRS-17, BPRS-18, electric quantity, seizure duration, seizure energy index and blood pressure. Kruskal–Wallis and Mann–Whitney U tests (with Bonferroni correction) were used for skewed data and group comparisons of cognitive change scores (baseline minus post-ECT). One-way ANOVA with post hoc tests was applied for normally distributed variables. A two-tailed p < 0.05 was considered statistically significant.

Ninety patients (36 men, 54 women; mean age 30.6 years, SD 9.15, range 15–67) were randomised equally across the three groups. Ten, 13 and 11 patients in the ketamine, ketamine þ propofol and propofol groups, respectively, had bipolar disorder; baseline demographics and depression scores did not differ significantly between groups. Depressive symptoms improved across all groups over the ECT course (GLM main effect of time on HDRS; reported F = 3084.8, p < 0.001). A significant group-by-time interaction was reported for HDRS-17 (F = 9.736, p < 0.001), and overall group differences were observed (ketamine vs propofol p < 0.001; ketamine vs ketamine þ propofol p = 0.011; ketamine þ propofol vs propofol p = 0.033). Patients receiving ketamine had lower HDRS-17 scores than those receiving propofol after the first treatment (p = 0.025), and this advantage persisted from the second through the eighth treatment (all p < 0.001). The ketamine þ propofol group showed lower HDRS-17 scores than propofol after the second treatment (p = 0.025) and at several subsequent time points (third, fourth, sixth), with differences attenuating by the eighth session. Comparisons between ketamine and ketamine þ propofol indicated lower HDRS-17 scores in the ketamine group from the second treatment onward. No responders were identified before completion of the third treatment; both ketamine-containing regimens had higher response rates after the third and fourth ECT compared with propofol, but those differences were not maintained after the sixth and eighth sessions. Remission rates differed only after the eighth treatment: the ketamine group had significantly higher remission versus ketamine þ propofol (p < 0.001) and versus propofol (p < 0.017). A nominal difference between ketamine þ propofol and propofol (p = 0.018) did not remain significant after Bonferroni correction. General psychopathology (BPRS-18) improved over time in all groups (main effect of time), with a main effect of group reported; post hoc tests showed the ketamine group had better overall BPRS scores than the propofol group (p < 0.01). Subscale analyses indicated greater improvement in anxiety–depression symptoms for ketamine and ketamine þ propofol versus propofol. Seizure parameters differed by anaesthetic. Electric quantity required rose over sessions (main effect of time), but the ketamine group required a lower electric quantity than the ketamine þ propofol and propofol groups at most time points (group-by-time interaction F = 6.314, p < 0.001). Seizure duration showed a main effect of group (F = 22.4, p < 0.001): duration was longer in the ketamine group than in propofol and ketamine þ propofol; ketamine þ propofol also produced longer seizures than propofol. The seizure energy index had a main effect of group (F = 4.3, p < 0.05); the ketamine þ propofol group showed a higher seizure energy index than the propofol group. On cognitive testing, baseline performance did not differ between groups. After the eighth ECT, the propofol group showed greater decline in executive measures: larger reductions in WCST categories completed and a greater increase in steps needed for the Tower of Hanoi compared with the ketamine group (p < 0.05). The ketamine þ propofol group had a greater decline in WCST categories than the ketamine group (p < 0.05). Trail Making Test Parts A and B completion times increased (worsened) more in the propofol group than in the ketamine group (p < 0.05). No group differences were reported for Word Fluency, Digit Symbol, Digit Span or Visual Reproduction tests. Safety data showed no major adverse events; transient headaches and nausea occurred across groups and resolved within 0.5–1 hour without intervention. Systolic blood pressure showed a main effect of group (F = 40.962, p < 0.001) and a group-by-time interaction (F = 2.615, p = 0.02); both ketamine-containing regimens increased systolic pressure compared with propofol. Diastolic blood pressure showed a main effect of treatment (F = 39.939, p < 0.001), with higher diastolic pressures after each ECT in the ketamine and ketamine þ propofol groups; the ketamine group tended to have higher diastolic pressure than ketamine þ propofol at most sessions. The authors report these haemodynamic changes as temporary and without clinical significance.

Zhong and colleagues interpret their findings as evidence that ketamine, at anaesthetic concentrations, speeds and augments the antidepressant effects of ECT in TRD and confers relative protection against some executive cognitive impairments compared with propofol. Both anaesthetic (0.8 mg/kg) and subanaesthetic (0.5 mg/kg with adjunct propofol 0.5 mg/kg) ketamine regimens produced earlier improvements in HDRS-17 scores than propofol, but the larger ketamine dose yielded greater magnitude and a higher remission rate by the end of eight ECT sessions. The authors link these dose-related antidepressant differences to prior dose–response observations for intravenous ketamine and argue that propofol’s anticonvulsant properties may blunt seizure quality and thereby reduce ECT efficacy. Seizure-related findings are emphasised as mechanistically relevant: ketamine produced longer seizure durations and lower required electric quantity relative to propofol, and the ketamine þ propofol regimen increased seizure energy index compared with propofol. The investigators suggest these electrophysiological effects may underlie the improved clinical outcomes. Regarding cognition, they propose that ketamine’s NMDA antagonism could mitigate excitotoxic injury during ECT and that the lower electrical doses used in the ketamine group may also contribute to better executive function outcomes. Safety considerations are addressed: ketamine produced transient increases in systolic and diastolic blood pressure consistent with known sympathomimetic effects, but no clinically significant adverse events occurred during the eight treatments. The authors acknowledge several limitations: blinding was not formally tested and might have been compromised; cognitive testing occurred only at baseline and 48–72 hours post-treatment so practice effects cannot be excluded; and dissociative effects of ketamine were not assessed with a specific instrument (for example CADSS). They also note differences between their sample and prior meta-analyses—specifically inclusion of only TRD patients and a higher proportion with bipolar disorder—which could explain divergent findings. Finally, the authors call for larger, multicentre studies to confirm synergistic effects of ketamine and ECT and to examine dissociative safety outcomes more closely.