Intravenous Ketamine for Late-Life Treatment-Resistant Depression: A Pilot Study of Tolerability, Safety, Clinical Benefits, and Effect on Cognition

Treatment-resistant depression (TRD) in older adults is associated with poor long-term outcomes, including disability, cognitive impairment, and elevated suicide risk, yet evidence-based treatment options are limited. Earlier randomized trials in late-life TRD support only a few augmentation strategies and non-pharmacological interventions, and remission rates with common pharmacotherapies have been under 30%. Ketamine, an NMDA receptor antagonist shown to produce rapid antidepressant effects in younger adults, is a promising candidate, but data on intravenous (IV) ketamine in older populations are scarce. Concerns also exist about ketamine's potential adverse cognitive effects based on observational studies of heavy users and age-related vulnerability to medication-related cognitive change. Oughli and colleagues therefore conducted a multisite pilot clinical trial to evaluate acute and continuation IV ketamine in adults aged 60 years and older with TRD. The primary aims were to assess acceptability, tolerability, and safety; secondary objectives included measuring clinical benefit on depressive symptoms and effects on cognition, with particular focus on executive function (EF) and overall fluid cognition. The investigators hypothesised that twice-weekly IV ketamine would be acceptable and well tolerated, and that it would improve depressive symptoms and cognitive performance, especially EF.

This open-label, multisite pilot study enrolled 25 community-dwelling adults aged 60 years and older with TRD between October 13, 2020 and November 6, 2021. Recruitment occurred at five academic sites; the trial was approved by local institutional review boards and registered (ClinicalTrials.gov: NCT04504175). Inclusion required a DSM-5 diagnosis of major depressive disorder without psychotic features confirmed by SCID-5, failure of ≥2 adequate antidepressant trials in the current episode as measured by the Antidepressant Treatment History Form (ATHF), and at least moderate depressive symptoms (PHQ-9 ≥15). Key exclusions included a clinical diagnosis of dementia or Short Blessed Test score ≥10, schizophrenia-spectrum or bipolar disorder, recent substance use disorder or lifetime recreational ketamine/PCP use, certain medications contra-indicated with ketamine, uncontrolled cardiovascular conditions, life expectancy <1 year, and baseline systolic blood pressure >150 mm Hg or diastolic >90 mm Hg. Participants continued a stable oral antidepressant dose for at least 4 weeks before and during the infusion phase. The acute treatment phase consisted of IV racemic ketamine 0.5 mg/kg infused over 40 minutes twice weekly for 4 weeks. Partial responders at the end of the acute phase (MADRS total score <10 or ≥30% reduction from baseline) were eligible for a continuation phase of once-weekly infusions for 4 additional weeks. Weight was confirmed before each infusion, and standard peri-infusion monitoring of heart rate, blood pressure, pulse oximetry and dissociation was performed (every 10 minutes during infusion, and for up to 2 hours afterwards). Dissociative symptoms were measured with the first six items of the Clinician-Administered Dissociative States Scale (CADSS) pre-infusion and at 40 and 90 minutes after infusion start. Craving was assessed by a 4-item self-rated Craving Scale. Clonidine (up to 0.6 mg pre-treatment and up to 0.3 mg during infusion) was available at investigators’ and participants’ discretion to mitigate dissociative or hypertensive effects; no other rescue medications were protocolised. Primary safety and tolerability outcomes included adverse events (AEs), serious adverse events (SAEs), treatment-emergent hypertension (SBP >160 or DBP >100), dissociation, and craving. Efficacy outcomes were changes in depressive symptoms measured by MADRS, with response defined as ≥50% reduction and remission as MADRS <10; a ≥30% reduction determined eligibility for continuation. Cognitive outcomes used the NIH Toolbox Cognition Battery at baseline, end of the acute phase, and end of the continuation phase, with specific tests for EF (DCCS, Flanker, List Sorting), episodic memory, processing speed, and a Fluid Cognition Composite Score. Analyses included descriptive summaries of baseline characteristics, calculation of response and remission rates, and paired t-tests (with 95% confidence intervals and Cohen's d) for changes in MADRS and cognitive measures between timepoints. The investigators computed a per-subject Spearman correlation between CADSS scores (40 minutes post-infusion) and infusion number, averaging these correlations across subjects with bootstrap confidence intervals. Linear models examined associations between changes in EF measures and responder/remitter status or change in MADRS, adjusting for baseline EF, education, and gender; NIH Toolbox scores were age-adjusted so age was not included in those models. Analyses used complete cases for each measure.

Twenty-five participants began the study. Mean (SD) age was 71.5 (4.9) years, 13/25 (52%) were female, all reported European descent, and mean (SD) education was 16.2 (1.9) years. Medical burden was moderate (CIRS-G mean 8.1, SD 5.4), baseline depression severity was moderately severe (MADRS mean 24.4, SD 7.9), and the mean ATHF score for prior adequate trials was 3.3 (SD 1.4). Completion rates were high: 22/25 (88%) completed the acute phase and 15/15 (100%) of eligible participants entered and completed the continuation phase. Three participants discontinued the acute phase (two due to perceived poor response; one gave no reason). There were no treatment-related SAEs. Two participants (8%) experienced mild, manageable nausea with vomiting and headache. Infusion-related transient hypertension occurred in 5/25 (25%), and clonidine was given to 8/25 (32%) as pre-treatment during every infusion for blood-pressure management; one participant reported dry mouth possibly related to clonidine. No infusions were discontinued due to elevated blood pressure or other AEs. Dissociative symptoms occurred in all participants at some point, peaking at ~40 minutes and resolving by 90 minutes post-infusion. CADSS scores tended to decrease across successive infusions: the average Spearman correlation between CADSS (40 minutes) and infusion number was −0.28 with a 95% bootstrap percentile confidence interval of (−0.43, −0.12). No participant discontinued due to inability to tolerate infusions. Regarding craving, 2/25 (8%) reported high craving scores (50–100) on some Craving Scale items, and 4/25 (16%) reported moderate preoccupation with the next dose (scores 10–50). Clinical outcomes showed that 15/25 (60%) achieved ≥30% MADRS reduction at the end of the acute phase and entered the continuation phase. Mean MADRS decreased by 9.4 points from baseline to end of acute phase (95% CI 6.46 to 12.32), with a large effect size (Cohen's d ≈ 1.19) and a paired t statistic of −6.66 (df = 22, p < 0.01). Using the study's response/remission definitions, 12/25 (48%) met criteria for response (≥50% reduction) and 6/25 (24%) for remission (MADRS <10) at the end of the acute phase. At the end of the continuation phase, 7/15 (47%) met response criteria and 4/15 (27%) met remission criteria. Among those in continuation, mean MADRS increased (worsened) by 3.5 points from start to end of continuation (95% CI 0.38 to 6.56), Cohen's d 0.95, paired t = 2.41 (df = 14), p = 0.03. Cognitive testing was completed for 23 participants. During the acute phase, the Fluid Cognition Composite Score improved with a medium-to-large effect size (Cohen's d ≈ 0.61). The three EF subtests (Flanker, Dimensional Change Card Sort, and List Sorting Working Memory) also showed improvements. Linear models adjusting for baseline EF, education, and gender found no significant association between changes in EF measures and responder or remitter status, nor between EF change and change in MADRS. During the continuation phase there were no statistically significant changes in EF measures; effect sizes were small (Cohen's d 0.07–0.33). The age-corrected Fluid Cognition Composite increased during the acute phase and showed no evidence of further change during continuation.

The investigators highlight four principal findings. First, IV ketamine was acceptable to older adults with TRD as evidenced by high completion rates in both the acute and continuation phases. Second, the regimen appeared generally well tolerated and safe in this sample: adverse events were uncommon, no treatment-related serious adverse events occurred, dissociation was transient and diminished over repeated infusions, and clonidine was useful for managing blood-pressure elevations without requiring infusion discontinuation. Third, IV ketamine demonstrated potential clinical benefit, with nearly half of participants meeting response criteria and about one quarter achieving remission after 4 weeks of twice-weekly infusions; these effects were largely preserved during continuation. Fourth, measures of executive function improved during the acute phase and were sustained among those who entered continuation, while overall fluid cognition did not worsen over the 8-week exposure. The authors situate these results amid earlier work showing rapid antidepressant effects of ketamine in younger adults and a paucity of IV ketamine data in older populations. They acknowledge concerns from animal and observational human studies about NMDA antagonist-related transient neuronal effects and the potential for cognitive harm with long-term or heavy use, noting that the present findings provide initial reassurance that repeated low-dose IV ketamine over 8 weeks did not produce detectable cognitive decline in this sample. Key limitations noted by the investigators include the small sample size, lack of racial and ethnic diversity, absence of randomisation and placebo or active control, no adjustment for multiple testing, and restriction to a relatively younger segment of older adults (mean age 71.5 years), which limits generalisability to older or more medically complex patients or those with more severe or acutely suicidal TRD. They also acknowledge the possibility that observed EF improvements reflect practice effects rather than ketamine-specific effects, although they argue the absence of similar practice-related gains in processing speed or immediate memory weakens that alternative explanation. On balance, the authors present these results as preliminary and hypothesis-generating. They recommend a larger, adequately powered randomized controlled trial to confirm tolerability, safety, and efficacy of IV ketamine in late-life TRD, to probe mechanisms underlying cognitive and mood changes, and to determine whether benefits persist beyond 8 weeks.