This triple-blind randomised controlled trial (n=15) in adolescents with suicidal ideation in a paediatric emergency department tested whether intravenous ketamine, midazolam or saline could be studied feasibly and safely. Recruitment and follow-up were good, blinding seemed effective, and no serious or unexpected adverse events occurred.
Papers cited by this study that are also in Blossom
Ionescu, D. F., Vande Voort, J. L., Niciu, M. J. et al. · Journal of Psychiatric Research (2014)
Objectives
Emergency department (ED) visits by adolescents with suicidal ideation have increased, with no rapid-acting treatments identified for ED use. This pilot study explored the feasibility of a randomized controlled trial of ketamine in the ED for adolescents with suicidal ideation.
Methods
Three-arm, triple-blinded randomized controlled trial comparing intravenous ketamine (0.5 mg/kg), midazolam (0.03 mg/kg), and saline (0.9%) in adolescents requiring hospitalization for suicidal ideation. We primarily reported recruitment proportion as a measure of feasibility. Secondary outcomes included: additional aspects of feasibility (barriers to enrolment, retention, blinding, and adverse events) and distribution of suicidal ideation scores 90 min post-infusion on questions 3-5 of the Columbia Suicide Severity Rating Scale, item 10 of the Montgomery-Åsberg Depression Rating Scale, and a pragmatic 0-10 Likert scale to estimate future randomized controlled trial sample size.
Results
Out of 21 eligible patients and guardians, 15 (71%, 95% CI 48, 87%) agreed to participate when offered the opportunity. Key barriers to enrolment included limited ED resources and absence of a consenting parent at time of assessment. Retention at 7, 14, 21, and 28-days was 93, 86, 73, and 73%. No serious or unexpected adverse events occurred. Blinding appeared reasonably effective. The median [IQR] Columbia Suicide Severity Rating Scale, Montgomery-Åsberg Depression Rating Scale, and pragmatic Likert scale scores at 90 min were 0 [0, 2], 4 [2, 4], 5 [5, 6] for ketamine; 0 [0, 3], 3 [3, 6], 6 [3, 7] for midazolam; and 3 [2, 3], 4 [4, 5], 6 [6, 7] for saline.
Conclusions
Despite a small sample, our participation rate was high and trial design appeared feasible and safe. Dedicated research nurses and a mature-minor consent process could address key barriers to enrolment in future trials to determine if ketamine could provide ED physicians with the first rapid-acting pharmacological treatment for acute pediatric suicidal ideation.
Pediatric emergency department visits for mental health concerns, including suicidal ideation, have risen markedly, and adolescents with suicidality are often admitted or transferred for psychiatric care. The authors note that current treatments, including pharmacological options and cognitive behavioural therapy, usually take weeks to months to reduce suicidal symptoms, leaving a gap for a rapid-acting, emergency-department-feasible intervention. They also point to adult evidence suggesting that sub-anaesthetic ketamine can reduce suicidal ideation quickly, while emphasising that little is known about its efficacy, tolerability, or feasibility in adolescents. Onikashvili and colleagues therefore set out to test whether a randomized trial of intravenous ketamine for suicidal ideation could be run in a paediatric emergency department, and to gather feasibility data to inform a future efficacy study. The paper is explicitly framed as a pilot randomized controlled trial designed to assess recruitment, retention, blinding, safety, and the behaviour of several suicide-related outcome measures over a short follow-up period.
The researchers conducted a three-arm, triple-blinded pilot randomized controlled trial at the British Columbia Children’s Hospital emergency department from June 2022 to January 2024. The study followed the CONSORT 2010 extension for randomised pilot trials and was regulated by Health Canada, with ethics approval obtained and the trial registered before recruitment. Adolescents aged 10-17 years were eligible if they endorsed suicidal ideation on the Columbia Suicide Severity Rating Scale, were judged by a paediatric psychiatrist to require admission, could complete a capacity-to-assent questionnaire, and had age-appropriate vital signs and a normal neurological examination. Exclusions included recent ketamine or benzodiazepine use, no guardian present, and medical or psychiatric contraindications. The extracted text does not clearly report the full list of contraindications in the main body, though they are mentioned as being in an online supplement. Participants were randomly assigned in a 1:1:1 ratio to intravenous ketamine 0.5 mg/kg, midazolam 0.03 mg/kg, or saline placebo, each infused over 40 min using pharmacy-prepared identical kits. Midazolam served as an active placebo to help maintain blinding. Research assistants, ED staff, and participants were intended to remain blinded to allocation. The primary feasibility outcome was the proportion of eligible adolescents, when resources were available, who were randomized. Secondary and exploratory outcomes included barriers to enrolment, retention at 7, 14, 21, and 28 days, blinding success, adverse events, return visits to the emergency department for suicidal concerns, and suicidal ideation scores at baseline, 90 min, and 24 h. The outcome measures were the Columbia Suicide Severity Rating Scale suicide items, the suicide item from the Montgomery-Åsberg Depression Rating Scale, and a pragmatic 0-10 Likert rating of current suicidality. Randomisation used variable block sizes prepared by a third-party statistician, and data were collected in REDCap. The authors used descriptive statistics for feasibility outcomes and proportional odds models, adjusted for baseline score, to estimate treatment effects and inform sample size calculations for a future definitive trial.
Grunebaum, M. F., Galfalvy, H. C., Choo, T. H. et al. · American Journal of Psychiatry (2018)
Wilkinson, S. T., Ballard, E. D., Bloch, M. H. et al. · American Journal of Psychiatry (2017)
Kosik-Gonzalez, C., Chen, L. N., Lane, R. et al. · Journal of the American Academy of Child & Adolescent Psychiatry (2026)
Ionescu, D. F., Qiu, X., Lane, R. et al. · International Journal of Neuropsychopharmacology (2020)
Zhou, Y., Lan, X-F., Wang, C. et al. · Child and Adolescent Psychiatry (2024)
Cullen, K. R., Amatya, P., Roback, M. G. et al. · Journal of Child and Adolescent Psychopharmacology (2018)
Ballard, E. D., Luckenbaugh, D. A., Walls, T. S. et al. · Journal of Psychiatric Research (2015)
Recruitment was stopped after 15 randomized participants because the authors wanted to analyse feasibility data promptly and because the available pool of eligible patients was smaller than expected. In total, 254 adolescents were screened, 25 met eligibility criteria, and 21 were eligible when clinical resources were available to support full enrolment. Of those 21, 15 agreed to participate, giving a recruitment proportion of 71.4% (95% CI 48, 87%). The main barriers to enrolment were limited emergency department staffing and resources, especially after post-pandemic viral surges, and the absence of a guardian to provide consent. The enrolled sample had a median age of 14 years, and most participants were female (13/15, 87%). Baseline characteristics and suicidal ideation scores were broadly similar across groups. Retention was 93% at 7 days, 87% at 14 days, and 73% at both 21 and 28 days. Two participants missed 24 h follow-up because they had already been discharged from the inpatient psychiatric unit, but later completed follow-up surveys by email. Blinding appeared reasonably effective overall. Successful blinding was reported for 10/15 nurses (66%), 13/15 participants (87%), and 8/15 research assistants (53%). No serious or unexpected adverse events were identified. The ketamine group showed some transient behavioural changes such as increased sociability, positivity, emotional lability, and dizziness; one participant described their “head felt funny”. In the midazolam arm, increased drowsiness was noted. Return visits to the emergency department for suicidal concerns were reported by two participants in the midazolam arm and one in the saline arm. At 90 min post-infusion, the ketamine arm showed median changes of -1 on the Columbia Suicide Severity Rating Scale suicide items, -2 on the Montgomery-Åsberg suicide item, and -2 on the pragmatic Likert scale. The authors report odds ratios with very wide confidence intervals because of the small sample, and therefore describe the estimates as highly uncertain. Using the observed effects, they derived future sample size estimates that varied substantially by outcome and comparator. For ketamine versus saline, the estimated sample sizes were 30 for the Columbia scale, 45 for the Montgomery-Åsberg suicide item, and 39 for the pragmatic Likert scale. For ketamine versus midazolam, the corresponding estimates were much larger and more variable, including 556 for the Columbia scale, 66 for the Montgomery-Åsberg item, and 90 for the Likert scale.
The authors interpret the findings as indicating that a randomized controlled trial of ketamine for acute suicidality in a paediatric emergency department is feasible, with acceptable recruitment, retention, blinding, and safety in this small pilot. They emphasise that the study provides practical information for planning a larger efficacy trial, particularly about how to manage consent, staffing, and outcome collection in a busy emergency setting. They place the results in the context of earlier work showing that system constraints can limit paediatric emergency research, especially during and after the COVID-19 pandemic. They also note that blinding is inherently challenging in ketamine studies because ketamine and active placebo can produce noticeable behavioural effects, yet overall blinding in this trial was reasonably effective. The authors compare their retention rate with that of other studies and suggest that differences may reflect the novel features of their trial, including a single-dose intervention in the emergency department and online follow-up. The paper also discusses the pattern of outcome effects in relation to earlier research. The authors say the variability across suicidal ideation measures is consistent with recent adolescent esketamine trials and that the direction of the ketamine effect is broadly in line with adult literature. However, they explicitly caution that the pilot was not powered to estimate efficacy and that the apparent pattern should not be taken as evidence of benefit. They instead argue that the difference between ketamine versus saline and ketamine versus midazolam reinforces the methodological value of an active placebo in a future efficacy study. Key limitations acknowledged by the authors include the very small sample, the pilot nature of the study, the limited ability to make inference about efficacy, and uncertainty about which suicide measures are best suited to the rapid time course of ketamine trials. They also note that the observed score patterns over time are descriptive only and do not establish measure validity. The authors suggest practical implications for future research and implementation, including dedicated research nurses, virtual consent processes, mature-minor consent approaches, separating monitoring from outcome assessment to preserve blinding, and collecting preferred names and pronouns in a gender-diverse population.
The authors conclude that this pilot demonstrated the feasibility of a ketamine randomized controlled trial for acute suicidality in a paediatric emergency department. They state that future trials should address enrolment barriers with dedicated research nurses and consent strategies such as virtual and mature-minor processes, and that a multicentre efficacy trial is now needed to determine whether ketamine could become the first rapid-acting pharmacological treatment for acute paediatric suicidal ideation in the emergency department.