Real-world effectiveness of repeated intravenous ketamine infusions for treatment-resistant depression in transitional age youth

Chisamore and colleagues frame the study within the problem of treatment-resistant depression (TRD), defined here as inadequate response to two or more adequate antidepressant trials, which affects roughly 30% of patients treated with monoaminergic antidepressants and accounts for a disproportionate share of depression-related costs and morbidity. Transitional age youth (TAY; 18–25 years) are highlighted as a distinct group: brain maturation continues into the mid-20s, onset of mood disorders and suicidality is common in this window, and conventional pharmacotherapies raise particular concerns about safety and tolerability in younger populations. The authors note a gap in the literature on intravenous (IV) ketamine for TRD specifically in TAY, despite established evidence for rapid antidepressant and anti-suicidal effects of ketamine in general adult (GA) samples and some adolescent data pooled across age ranges. This retrospective analysis therefore aims to determine whether repeated IV ketamine infusions produce comparable antidepressant, anti‑suicidal, anxiolytic, functional and tolerability effects in TAY patients with TRD compared to a matched GA cohort. The investigators hypothesised that four ketamine infusions over two weeks would show similar clinical benefits and safety in TAY as in GA patients, and set out to examine symptom trajectories and adverse effects in a real‑world outpatient clinic setting.

This study is a retrospective, matched cohort analysis of patients treated at the Canadian Rapid Treatment Center of Excellence (CRTCE), an outpatient clinic administering IV ketamine for TRD. Eligible patients were stage 2 TRD (inadequate response to at least two antidepressant trials), referred by psychiatrists or primary care providers, and had a primary mood disorder diagnosis confirmed by a staff psychiatrist using DSM‑5 criteria. Patients with a history of psychosis or active substance use disorder were excluded from treatment; those with prior substance use disorder were eligible if abstinent for at least 3 months. Medical screening by anaesthesiology reviewed cardiovascular stability prior to treatment. Treatment consisted of four IV ketamine infusions over 2 weeks, starting at 0.5 mg/kg infused over 40 minutes in saline; dose could be increased to 0.75 mg/kg after two infusions if response was incomplete and the lower dose was tolerated. Vital signs were recorded at baseline and monitored every 15 minutes during infusions until 20 minutes post‑infusion. Dissociative symptoms were prospectively assessed during and after infusions using a reduced six‑item version of the Clinician Administered Dissociative States Scale (CADSS‑6); a score of ≥3 on CADSS‑6 indicated presence of dissociation. Other adverse effects (nausea, headache, dizziness) were rated on a 1–4 severity scale. The retrospective cohort of TAY patients (initially 54, with 2 excluded for missing baseline QIDS‑SR16) was matched to a GA control sample aged 30–60 on sex, primary diagnosis, baseline depression severity (QIDS‑SR16) and degree of treatment resistance (number of past antidepressants), with tolerances of ±1 for baseline QIDS and ±3 for treatment resistance. Outcome measures included the Quick Inventory of Depressive Symptomatology Self‑Report 16‑item (QIDS‑SR16) as the primary outcome, the QIDS SI item for suicidal ideation, Generalized Anxiety Disorder 7‑item (GAD‑7) for anxiety, and functional measures (Sheehan Disability Scale and Work Productivity and Activity Impairment questionnaire) when available. Data were collected at baseline, after each infusion, and at a post‑treatment follow‑up interview. For analysis the investigators used linear mixed‑effects models to test main effects of infusion and group and their interaction for each outcome, with alpha set at 0.05. Pairwise comparisons across infusions were Bonferroni‑corrected. Response categories were predefined: partial response as ≥25% reduction in QIDS‑SR16, response as ≥50% reduction, and remission as total QIDS‑SR16 ≤5. Partial eta squared (η2) was reported as an effect size metric. The extracted text does not clearly report the exact sample size of the matched GA group, though paired figures are reported throughout.

Sample and baseline data: After excluding two TAY patients without baseline QIDS‑SR16, 52 TAY patients were analysed; 48 had major depressive disorder and 4 had bipolar depression. Completion of post‑treatment follow‑up scales was incomplete: 32 TAY patients (61.5%) completed follow‑up clinical scales. The clinic’s voluntary scale completion led to missing data; the text reports a decline from 52 to 37 TAY and 38 GA participants from baseline to post‑treatment follow‑up. Demographic details such as mean age and BMI are indicated to be presented in a table in the source but are not clearly reported in the extracted text. Depressive symptom severity (QIDS‑SR16): There was a significant main effect of infusion on total QIDS‑SR16 scores across both groups (p < 0.001), with a medium effect size for infusion (η2 = 0.322). No significant main effect of group (p = 0.402) and no significant group × infusion interaction (p = 0.242) were observed, indicating comparable reductions in depressive symptoms for TAY and GA patients. Mean reductions from baseline to follow‑up were similar: 5.71 points for TAY and 5.89 points for GA. Bonferroni‑adjusted pairwise comparisons showed significant decreases from baseline to each infusion; the only non‑significant interval was from post‑infusion 3 to follow‑up (p = 1.000). At approximately 1 week after the fourth infusion, remission rates (QIDS‑SR16 ≤5) were 9.4% in TAY and 8.6% in GA. Response rates (≥50% reduction) at follow‑up were 25.0% for TAY and 31.4% for GA; partial response rates (≥25% reduction) were 18.8% for TAY and 8.6% for GA. Suicidal ideation (QIDS SI item): Infusions produced a significant reduction in suicidality across both groups (main effect p < 0.001), with no significant group effect (p = 0.161) or interaction (p = 0.423). Mean decreases were 0.67 points for TAY and 0.58 points for GA. Pairwise comparisons indicated significant decreases from baseline to post‑infusion 2, post‑infusion 3 and follow‑up (p < 0.001). The effect size for infusion on suicidality was small (η2 = 0.125). The authors note a substantial floor or ceiling effect: 48 of 104 total patients had an SI score ≤1 at baseline, including 20 of 52 TAY patients, and 38% of TAY and 54% of GA patients had low or no SI at baseline. Generalized anxiety (GAD‑7): Anxiety scores decreased significantly with infusions in both groups (main effect p < 0.001). A small but significant main effect of age group on GAD‑7 reduction was observed (p = 0.036), while the group × infusion interaction narrowly missed significance (p = 0.053). Mean decreases from baseline to follow‑up were 3.85 points for TAY and 3.50 points for GA. Pairwise comparisons showed significant reductions from baseline to post‑infusion 3 and to follow‑up (p < 0.001); the infusion effect size was medium (η2 = 0.271). The extracted text notes limited power for some comparisons: only 26 GA patients had GAD‑7 scores at post‑infusion 3. Functioning (SDS/WPAI): Infusions were associated with significant improvement in work (p = 0.017), social (p < 0.001) and family (p < 0.001) functioning. A main effect of group was present for work functioning (p = 0.017) but not for social (p = 0.225) or family (p = 0.062) measures; no significant interactions were found. Mean score reductions from baseline to follow‑up in work functioning were 1.70 for TAY and 1.35 for GA. Social scores fell by 2.07 for TAY and 1.45 for GA; family scores fell by 1.55 for TAY and 1.61 for GA. Effect sizes for infusion on functional measures were small (η2 ≈ 0.15–0.17). Safety and tolerability (CADSS‑6 and adverse effects): Dissociative symptoms measured by CADSS‑6 showed a significant main effect of infusion (p = 0.012) but no group effect (p = 0.451) and no interaction (p = 0.947). Both groups had peak mean CADSS‑6 scores at infusion 1 and the lowest at infusion 2 (TAY = 2.71, GA = 3.06), with small mean changes from infusion 1 to 4 (TAY 0.30, GA 0.52). The study reports only mild, transient adverse effects such as dissociation, nausea and dizziness, and no serious adverse events during the observed treatment period.

The investigators interpret their findings as showing that repeated IV ketamine infusions over two weeks produce comparable reductions in depressive symptoms, suicidal ideation, anxiety and functional impairment in TAY patients and a matched GA cohort. Both groups experienced clinically meaningful reductions in QIDS‑SR16 (>5 point mean reduction) and significant declines in suicidality; the absence of significant group or interaction effects led the authors to conclude there was no evidence that TAY patients respond differently to this treatment protocol. Ketamine’s rapid onset of antidepressant and anti‑suicidal effects is contrasted with the slower action of serotonergic antidepressants, which the authors note may be particularly relevant in younger patients. A notable finding discussed is a small difference in anxiety severity between age groups: baseline anxiety appeared higher in TAY and group analyses showed a main effect of group on GAD‑7 scores. Although a trend suggested GA patients might have a more rapid anxiolytic response, the least squares analysis did not find a statistically significant difference at post‑infusion 3 and the authors attribute some uncertainty to limited sample size for GAD‑7 at that time point. They therefore recommend further study of ketamine’s anxiolytic effects in TAY. Safety observations are described as reassuring in the short term: side effects were generally mild and transient, tolerability improved over successive infusions, and no serious adverse events were recorded in the clinic’s typical two‑week treatment course. Nevertheless, the investigators emphasise that long‑term safety and maintenance strategies were not assessed here and merit further research. The authors acknowledge several important limitations of their analysis: its retrospective design, voluntary scale completion leading to missing data, lack of a contemporaneous placebo control, and limited power for some subgroup/timepoint analyses. Matching mitigated some confounding but did not substitute for randomisation. Despite these constraints, the study is presented as providing real‑world data on an understudied age group.

Chisamore and colleagues conclude that repeated IV ketamine infusions produced similar reductions in depression, anxiety, suicidality and functional impairment in TAY patients as in a matched general adult TRD sample, with comparable short‑term tolerability and safety. They recommend further investigation of IV ketamine in 18–25 year‑olds using prospective randomised controlled trials and longer follow‑up to better define efficacy, anxiolytic effects, and long‑term safety.