Low dose oral ketamine treatment in chronic suicidality: An open-label pilot study

Suicide remains a major global public health problem despite an overall decline in rates, and most suicides are associated with psychiatric disorders—particularly mood disorders. Conventional antidepressant treatments can take weeks to exert benefit, do not directly target glutamate or GABA neurotransmission, and leave a substantial proportion of patients with persistent symptoms; around 30% of people with depression meet criteria for treatment-resistant depression after failing two or more antidepressant trials. Ketamine, a non-competitive NMDA receptor antagonist, produces rapid glutamatergic effects that upregulate neurotrophic signalling and synaptic connectivity and has shown fast-acting reductions in depressive symptoms and suicidal ideation in trials that have mainly used the intravenous (IV) route. IV administration, however, is resource intensive and not readily scalable to many clinical settings, prompting interest in oral ketamine as a more accessible alternative, although evidence about its safety, dosing and feasibility in people with chronic suicidal ideation is limited. Can and colleagues therefore designed an open-label, flexible-dose pilot trial to assess whether once-weekly oral ketamine over 6 weeks, with a 4-week no-ketamine follow-up, is feasible, safe and tolerable and whether it reduces suicidality in adults with chronic suicidal thoughts. The primary objective was change in suicidal ideation measured by the Beck Scale for Suicide Ideation (BSS), with secondary assessments of suicidal attributes, depressive symptoms, functioning and well-being. The authors hypothesised that oral ketamine would produce a significant reduction in suicidality from baseline to the post-treatment visit.

This was an open-label, single-arm pilot trial conducted at the Thompson Institute between August 2018 and November 2019. Adults (≥18 years) presenting with chronic suicidality were recruited via referrals from local general practitioners, psychiatrists and advertising; chronic suicidality was operationally defined as suicidal ideation occurring continuously or intermittently over months to years and was confirmed by the principal investigator and a consultant psychiatrist. Inclusion required a BSS score of ≥6 at screening. Key exclusions included a history of psychosis or mania/hypomania, acute suicidality requiring urgent intervention, significant uncontrolled cardiovascular disease or other specified medical contraindications, abnormal liver function, pregnancy or breastfeeding. Participants remained on their usual psychotropic medications during the trial. The intervention comprised six once-weekly oral doses of racemic ketamine, administered as a liquid in fruit juice under clinical supervision. The initial dose was 0.5 mg/kg and doses were adjusted at each visit by +0.2 to +0.5 mg/kg or −0.2 to −0.7 mg/kg according to tolerability, with a maximum permitted dose of 3.0 mg/kg at the sixth treatment. Participants were observed for at least 90 minutes after dosing, advised not to undertake activities requiring alertness for 12 hours, and had twice-weekly phone safety checks between treatments. Baseline assessments within 14 days prior to treatment included physical examination, laboratory tests, urinalysis and standardised rating scales; MRI and EEG were performed but those data are not reported in this paper. Primary and secondary outcomes were assessed using validated instruments. The primary outcome was change in BSS across three timepoints: pre-ketamine (baseline), post-ketamine (one to seven days after the final dose; week 6) and follow-up (28–32 days after the final dose; week 10). A response on the BSS was defined as either ≥50% improvement from baseline to post-ketamine or a post-ketamine BSS score ≤6; a prolonged response applied the same criteria at follow-up. Secondary measures (pre-ketamine to follow-up) included the Suicidal Ideation Attributes Scale (SIDAS), Montgomery–Åsberg Depression Rating Scale (MADRS), Social and Occupational Functioning Assessment Scale (SOFAS) and the WHO-5 wellbeing index, with prolonged antidepressant response defined as ≥50% improvement in MADRS. Safety and tolerability measures completed 60 minutes post-dose included the Patient Rated Inventory of Side Effects (PRISE), FIBSER, Clinician Administered Dissociative States Scale (CADSS), Young Mania Rating Scale (YMRS) and Brief Psychiatric Rating Scale (BPRS); vital signs and urine/blood monitoring were performed at prespecified timepoints. For inferential analyses the investigators used repeated measures ANOVA for the primary outcome across three timepoints with Bonferroni-corrected pairwise comparisons and reported Cohen's d effect sizes. Secondary outcomes were analysed with repeated measures ANOVAs across two timepoints (pre-ketamine and follow-up). Analyses were conducted in SPSS version 24.0 with significance set at p = 0.05.

Of 64 people screened, 40 met inclusion criteria; five were withdrawn for clinical concerns and three declined to continue, leaving 32 participants who received the six-week treatment. Thirty participants completed the follow-up assessment (two were lost to follow-up). The sample comprised 17 females and 15 males with mean age 45.7 ± 14.2 years and mean weight 91.1 ± 24.8 kg at first treatment. All participants met DSM-5 criteria for major depressive disorder (MDD); MDD was the primary diagnosis for 28 participants, while smaller numbers had primary OCD, PTSD or GAD with comorbid MDD. Two-thirds of the sample were unemployed or not in the labour force. Primary outcome: Mean BSS scores fell markedly from a pre-ketamine mean of 20.0 ± 4.7 (n = 32; range 12–30) to 5.6 ± 7.6 at post-ketamine (n = 32; range 0–22) and were 9.1 ± 8.2 at follow-up (n = 30; range 0–25). A repeated measures ANOVA showed a significant main effect of time (p < 0.001). Pairwise comparisons indicated large effects for pre-ketamine versus post-ketamine (p < 0.001; Cohen's d = 2.04) and pre-ketamine versus follow-up (p < 0.001; d = 1.54). There was a statistically significant increase in BSS from post-ketamine to follow-up (p < 0.05; d = −0.42), though the mean follow-up score remained at less than 50% of the pre-ketamine mean. Using the predefined responder criteria, 69% of participants were BSS responders at week 6 and 50% met prolonged responder criteria at week 10. The categorical criterion of BSS ≤6 was met by 95% at post-ketamine (all of whom also met the ≥50% improvement criterion) and by 80% at follow-up. Thirteen participants were responders at both post-ketamine and follow-up. Secondary outcomes: All secondary measures showed statistically significant improvement from pre-ketamine to follow-up. SIDAS mean scores decreased from 26.7 to 17.2 (p < 0.001), shifting the group mean from a ‘high’ to a ‘low’ suicidal ideation category. MADRS scores fell from 38.6 (severe depression range) to 15.9 (mild), SOFAS clinician-rated functioning improved from 57.3 to 69.3, and WHO-5 wellbeing increased from 17.6% to 33.6% though remained below the 50% threshold for ‘good’ wellbeing. Sixty-three per cent of participants were prolonged antidepressant responders by MADRS criteria; of these, 13 were also prolonged BSS responders while six were MADRS responders without prolonged BSS response. Safety and tolerability: No serious adverse events were reported. Vital sign changes were transient; mean systolic blood pressure rose by 3.6 mmHg and mean diastolic blood pressure by 4.7 mmHg from pre-ketamine to 30 minutes post-dose. No significant changes in heart rate, respiratory rate or oxygen saturation were observed. The most frequent side effects across the six weeks were decreased energy and fatigue, anxiety, poor concentration, restlessness, general malaise, dry mouth, dizziness and tremor. Symptom frequency and reported distress were greatest in week 1 and generally declined across weeks 2–6, with distressing fatigue reported by about 22% at week 6 and other symptoms at <10% by week 6. Approximately 94% of participants (30/32) rated overall symptom intensity as mild, and 31/32 rated overall burden as mild. Dissociative symptoms measured by CADSS did not exceed a rating of 2 (moderate) across the trial; a small number of participants endorsed particular CADSS items at moderate level on no more than one or two occasions. No BPRS positive-symptom item exceeded mild (rating 3) and no YMRS total score exceeded 6 (well below the threshold for mania). No participants withdrew because of ketamine side effects.

Can and colleagues interpret their findings as supporting the feasibility, safety and tolerability of once-weekly, low-dose oral ketamine for people with chronic suicidal ideation. They report a very large reduction in suicidal ideation at the group level from baseline to post-treatment, with a substantial effect size (Cohen's d > 2.0) that remained large at four-week follow-up (d > 1.5) despite a modest rebound in scores after treatment cessation. Secondary measures of suicidal attributes, depressive symptoms, functioning and wellbeing shifted in the same direction, with MADRS moving into the mild range and SIDAS into a low-risk category at follow-up; improvements in functioning and wellbeing were smaller and the group mean wellbeing score remained in the ‘poor’ range. The investigators note that response rates (69% antisuicidal response at week 6; 50% prolonged antisuicidal response at week 10) are broadly consistent with those reported in IV ketamine studies, suggesting oral administration could be a viable alternative when administered under supervision. They highlight that their dosing strategy was conservative—once-weekly administration with a maximum of 3.0 mg/kg—contrasting with some earlier oral protocols that used higher doses or more frequent dosing. Tolerability was acceptable in this sample, with mainly transient, mild adverse effects and no serious adverse events. Key limitations are acknowledged by the authors: the open-label, uncontrolled design introduces the possibility that non-specific factors including expectancy and frequent clinical contact contributed to observed improvements; concurrent psychopharmacological treatments were permitted, so the independent contribution of ketamine cannot be isolated; and the trial did not prospectively define a minimum duration for ‘chronic’ suicidality (although all participants had ideation for at least six months). The sample size was small and the authors caution that efficacy cannot be definitively established without randomised controlled trials. They recommend larger, randomised studies to examine whether ketamine's antisuicidal effects are separable from its antidepressant effects and to evaluate impacts on suicidal behaviour and attempts.