Safety and Tolerability of the Acute Ketamine Treatment in Treatment-Resistant Depression: Focus on Comorbidities Interplay with Dissociation and Psychomimetic Symptoms

Recent developments in rapid-acting antidepressants, including esketamine nasal spray as an add-on to antidepressants, provide evidence for ketamine use in Major Depressive Disorder (MDD) and Bipolar Disorder (BP) for rapid remission of depressive symptoms, with concerns about safety and tolerability. A major issue is the risk of adverse events associated with dissociative symptomatology. There is limited and mixed evidence regarding whether dissociative symptoms predict response in TRD (TRD-MDD and TRD-BP), and overall little is known about the course of dissociation with ketamine in affective disorders. Dissociative states involve gaps in memory, depersonalization, derealization, and identity disturbance and are associated with ketamine administration. Standard measures for dissociation and potential treatment-emergent psychosis include the Clinician-Administered Dissociative States Scale (CADSS) and the Brief Psychiatric Rating Scale (BPRS, including BPRS+). Data on ketamine in TRD patients with somatic comorbidities are scarce. Consistent with depression measures (MADRS), safety measures (CADSS, BPRS) can assess dissociative and psychotic phenomena. This study aimed to assess the safety and tolerability profile of intravenous ketamine across eight administrations in inpatients with TRD (MDD and BP) in relation to clinical characteristics and comorbidities.

The paper highlights limited data on ketamine’s safety in TRD populations with somatic comorbidities and references prior esketamine trials showing acceptable safety without persistent psychotomimetic effects. Real-world pharmacovigilance has identified potential post-marketing safety signals warranting further study. Preclinical literature reports neurotoxicity concerns (e.g., Olney’s lesions, dendritic changes in GABAergic neurons) after ketamine or related NMDA antagonist exposure, though human relevance is uncertain. In epilepsy, depressive disorders are common and may present atypically; evidence on antidepressant efficacy is limited, and data on ketamine in epilepsy, particularly at subanesthetic doses, are sparse. Ketamine has both pro- and anti-convulsant properties and has been used in refractory status epilepticus. Psychotic disorders occur in a notable minority of patients with epilepsy. Overall, the literature underscores the need to examine ketamine’s CNS safety in TRD with comorbidities, especially epilepsy.

Design: Naturalistic observational registry (NCT04226963), single-site, unblinded, acute treatment phase with interim analysis. Participants: 49 adult inpatients (18–90 years) with TRD diagnosed with a depressive episode in MDD (including recurrent) or BP per DSM-5 using the MINI. TRD defined as inadequate response to ≥2 antidepressants of different categories for the current episode (ATRQ). Bipolar TRD defined as inadequate response to ≥2 distinct treatments in adequate dose/duration for the phase of illness. Inclusion: Medically stable, able to consent; continued baseline psychotropic and chronic disease medications unchanged during ketamine treatment. Exclusion: Uncontrolled medical conditions, prior adverse reaction to ketamine, active substance use, pregnancy, breastfeeding. Comorbidities considered: arterial hypertension, diabetes mellitus, hyperlipidemia, post-stroke, epilepsy, other. Intervention: Eight intravenous ketamine infusions over 4 weeks (0.5 mg/kg over 40 min) as add-on to standard-of-care treatments. Safety monitoring: Vital signs (heart rate, temperature, respiratory rate, blood pressure, oxygen saturation) every 15 min before, during, and up to 90 min after infusion; mental status with BPRS and CADSS to detect psychotic/dissociative symptoms; ECG before every second infusion and one week after last infusion; laboratory tests and scales at follow-up one week post-last infusion. Assessments: CADSS and BPRS collected 30 min post-infusion; MADRS and YMRS before the 1st, 3rd, 5th, and 7th infusions and at one-week follow-up; C-SSRS at screening. Outcomes: Primary outcome defined via MADRS (responders ≥50% improvement without remission; remitters MADRS ≤10 at follow-up). Safety outcomes focused on CADSS (dissociation) and BPRS/BPRS+ (psychotic symptoms), with particular attention to comorbidities. Statistics: IBM SPSS 25.0. Fisher’s exact test for categorical sociodemographic/clinical variables; mixed-model ANOVA with within-subject factor (repeated CADSS or BPRS measures over time) and between-subject factor (comorbidity group), including interaction effects; Kruskal-Wallis for quantitative variables; Cramer's V for associations; chain index method to estimate medium-term rate of change across administrations; α=0.05 with Bonferroni correction for multiple comparisons. Single-patient/single-rater design.

- Sample: 49 inpatients; 21 had somatic comorbidities. All were medically stable and continued baseline medications during ketamine. - Time course of acute effects: Post-dose maximum CADSS and BPRS scores rose by 30 min after ketamine, then declined to absent by 60 min post-infusion in all except the epilepsy subgroup. At one-week follow-up, neither CADSS nor BPRS scores were present. - CADSS (dissociation): Main effect over time significant only in the context of hyperlipidemia (F(4.58)=5.04; p<0.001; η2p=0.10), but Bonferroni-corrected simple effects were not significant for those without hyperlipidemia (n=40: F(7.40)=1.89; p=0.09; η2p=0.25) nor with hyperlipidemia (n=9: F(7.40)=1.92; p=0.09; η2p=0.25). Interaction table showed hyperlipidemia effect on CADSS: F=2.35; df=4.58; p=0.04. - BPRS (psychotic symptoms): Epilepsy was the only comorbidity significantly associated with changes over time. Main effect: F(3.96)=8.53; p<0.001; η2p=0.20; interaction significant. After Bonferroni correction, simple effects: without epilepsy (n=43): F(7.28)=1.53; p=0.198; η2p=0.28 (ns); with epilepsy (n=6): F(7.28)=3.54; p=0.008; η2p=0.47 (significant). In epilepsy subgroup, significant increases after the 1st (F(1.34)=10.41; p=0.003; η2p=0.23), 6th (F(1.34)=12.35; p=0.001; η2p=0.27), and 8th infusions (F(1.34)=18.05; p<0.001; η2p=0.35). Interaction table showed epilepsy effect on BPRS: F=7.37; df=3.96; p<0.001. - Overall safety: Short-term IV ketamine exhibited a favorable safety and tolerability profile regarding dissociative and psychomimetic symptoms in TRD-MDD and TRD-BP, with transient, rapidly resolving symptoms and no persistence at follow-up. Increased psychomimetic fluctuations were limited to patients with epilepsy.

Short-term ketamine infusions in a real-world inpatient TRD cohort (MDD and BP) showed favorable CNS safety and tolerability, with dissociative and psychomimetic symptoms peaking at 30 min and resolving by 60 min post-infusion, and absent at one-week follow-up. The notable exception was patients with epilepsy, in whom BPRS scores showed significant fluctuations over time, including spikes after the 1st, 6th, and 8th infusions. This suggests comorbid epilepsy, or potentially antiepileptic medications, may modulate ketamine’s psychomimetic effects. The findings align with esketamine trials reporting no sustained psychotomimetic harm, yet pharmacovigilance data indicate potential safety signals, underscoring the need for continued monitoring and research. Preclinical work raises theoretical concerns about neurotoxicity and dendritic changes from NMDA antagonism, but translational implications remain uncertain. Given the high psychiatric comorbidity burden in epilepsy and atypical mood disorder presentations, careful individualized treatment planning is warranted. Overall, the results support ketamine’s short-term CNS safety in TRD with comorbidities under close clinical supervision, with heightened vigilance in epilepsy.

Short-term intravenous ketamine, as add-on to standard-of-care psychotropic medication, demonstrated a favorable CNS safety and tolerability profile in TRD-MDD and TRD-BP inpatients with somatic comorbidities. Dissociative and psychotomimetic effects were transient and resolved rapidly post-infusion, with no persistence at one-week follow-up. Comorbid epilepsy was associated with significant fluctuations in psychotic symptom measures across treatments. Clinical practice should incorporate careful assessment of comorbidities and concomitant medications and ensure close supervision at each visit. Larger, long-term studies are needed to confirm the absence of sustained psychomimetic side effects and to clarify causal mechanisms.

- Small sample size (N=49) with potential underpowering. - Single-site, unblinded, observational design limits generalizability and causal inference; interim, post hoc analysis from a registry. - Acute treatment phase only; no long-term follow-up beyond one week. - CADSS measured only once post-dose (30 min), limiting precise characterization of peak and resolution time course. - Mixed diagnostic cohort (MDD and BP) and heterogeneous comorbidities; concomitant medications continued throughout, introducing confounding. - Increases in BPRS within epilepsy subgroup may reflect effects of antiepileptic medications rather than epilepsy per se. - Multiple comparisons addressed with Bonferroni correction; residual Type I/II error risks remain.