Clinical Effectiveness of Intravenous Racemic Ketamine Infusions in a Large Community Sample of Patients With Treatment-Resistant Depression, Suicidal Ideation, and Generalized Anxiety Symptoms: A Retrospective Chart Review

Treatment-resistant depression (TRD), concurrent suicidal ideation (SI), and co-occurring generalized anxiety disorder (GAD) symptoms represent large and growing public‑health problems, with suicide and depressive disorders causing substantial morbidity and mortality worldwide. Earlier research and recent meta-analyses have established that ketamine and intranasal esketamine can produce rapid antidepressant and anti‑suicidal effects in controlled trials, but most racemic ketamine studies have been small and the effectiveness of ketamine in routine community clinical settings remains uncertain. Oliver and colleagues set out to evaluate the clinical effectiveness and time course of benefit of intravenous racemic ketamine infusions in a large, real‑world cohort of patients treated in a multisite private infusion practice. The primary aim was to examine changes in depressive symptoms (PHQ‑9), suicidal/self‑harm ideation (PHQ‑9 item 9), and anxiety symptoms (GAD‑7) over the course of treatment, and to describe response and remission rates in this community sample treated with a typical acute course of infusions followed by maintenance as clinically indicated.

This study was a retrospective chart review of deidentified electronic health record data from three MindPeace Clinics locations in Virginia, covering patients seen between 9 November 2017 and 4 May 2021. The IRB at Virginia Commonwealth University reviewed the analysis plan and determined it was exempt from formal review because it did not meet criteria for human‑subjects research. Inclusion required a clinician diagnosis of a mood disorder and receipt of at least one ketamine infusion; patients treated primarily for non‑mood indications (for example, chronic pain) were excluded. Diagnostic information and treatment histories (including prior medication trials used to define TRD) were obtained from clinical interviews and self‑report recorded in the chart. Treatment protocols reflected clinical practice at the clinics: infusions were typically initiated at 0.5 mg/kg delivered over 40 minutes and titrated to achieve partial dissociative effects as measured by the Clinician‑Administered Dissociative States Scale (CADSS); the median dose after titration was 0.93 ± 0.24 mg/kg/40 minutes. The usual acute course comprised approximately six infusions over 2–3 weeks, with booster or maintenance infusions offered thereafter at the clinician's discretion. Outcome measurement relied on electronic self‑report: the Patient Health Questionnaire‑9 (PHQ‑9) was the primary outcome, with PHQ‑9 item 9 used to index self‑harm/SI; the Generalized Anxiety Disorder‑7 (GAD‑7) instrument was used for anxiety symptoms. Scales were completed at the initial evaluation, before each infusion, and via MoodMonitor follow‑up links sent every two weeks. TRD was operationalised as a history of two failed medication trials in the prior decade, based on patient report. Statistical analyses were conducted on an intention‑to‑treat basis, including all patients who received an infusion and provided at least one week of mood data. The authors applied longitudinal multilevel modelling: an exponential growth model with random effects to map PHQ‑9 trajectories, followed by linear mixed models grouping weeks from treatment initiation into seven time periods (week 1; weeks 2–3; weeks 4–6; weeks 7–12; weeks 13–24; weeks 25–51; 52+). Pretreatment scores were included as covariates, and comparisons across time periods used Bonferroni correction. Kaplan‑Meier estimates produced median and mean times to response (>50% PHQ‑9 improvement) and remission (PHQ‑9 < 5), with censoring at each patient's last observation. Missing data were not imputed, and univariate logistic regression tested demographic predictors of response and remission.

After excluding nine patients treated primarily for chronic pain, 424 patients were analysed. The cohort was 46% male and 54% female, aged 15 to 65+ years (mean 41.7 ± 15.4 years). Race was not collected. Seventy percent completed the main acute course of at least six infusions while 30% discontinued earlier. The median post‑titration dose was 0.93 ± 0.24 mg/kg/40 minutes. Depressive symptoms (PHQ‑9) decreased significantly over time. Pairwise comparisons showed that scores at week 1 and weeks 2–3 were each significantly different from later time periods (all P < .001 for week 1 comparisons; P < .001 or < .01 for weeks 2–3 comparisons), with no significant differences among the later five periods. Anxiety (GAD‑7) scores fell by about 30% over the treatment course, with most improvement evident by weeks 4–6; week 1 was significantly higher than all later periods (all P < .05) but later periods did not differ significantly from each other. Suicidal/self‑harm ideation (PHQ‑9 item 9) showed rapid reductions. Kaplan‑Meier estimates indicated that 5% of patients had achieved response by 7 days and 17% by 14 days; the median time to response was 36 days (mean 149 days). For remission, 3.5% were in remission at 7 days and 20% at 30 days; the median time to remission was 392 days (mean 489 days). When outcomes were expressed by number of infusions rather than calendar time: after three infusions, 14% had responded and 7% were in remission; after ten infusions, 72% had responded and 38% were in remission. Among responders, the median number of infusions to both response and remission was six, and 25% achieved response and remission after four infusions. Focusing on the subgroup reporting SI at baseline, 22% no longer reported SI after three infusions, 50% after six infusions, 75% after ten infusions, and 85% after 15 infusions. Logistic regression found no demographic variables that reached statistical significance as predictors of response or remission; history of psychosis (OR = 2.75, P = .086) and presence of suicide planning at initiation (OR = 1.82, P = .083) trended toward higher response rates but did not meet conventional significance thresholds. Adverse‑event reporting was limited in the dataset: nurses made routine follow‑up calls but the study did not systematically capture side effects. Two non‑fatal suicide attempts were reported during or shortly after ketamine treatment.

Oliver and colleagues interpret these findings as evidence that intravenous racemic ketamine infusions can produce substantial and relatively rapid improvements in depressive symptoms, suicidal ideation, and anxiety in a large, real‑world cohort of patients with TRD. The authors note that much of the clinical benefit occurred during the acute 6‑infusion phase, with many patients achieving response and remission within 3–6 weeks; additional gains were observed up to around ten infusions, after which further improvements were uncommon. Comparisons with earlier research are cautious: observed response and remission rates in this community sample approximate rates reported for rTMS and some community ECT samples, although the authors state that direct comparison is limited by differences in patient severity, setting, and study design. The rapid reduction in SI—more than 50% of patients with baseline SI no longer reporting it after the acute series—is highlighted as a clinically important finding that aligns with prior reports of ketamine's anti‑suicidal effects. The authors acknowledge several key limitations. The retrospective, uncontrolled design and absence of blinding prevent causal inference and introduce the possibility of selection and reporting biases. Reliance on the self‑report PHQ‑9 (without clinician‑rated assessments) and incomplete adverse‑event ascertainment constrain interpretation of benefit and safety. Important demographic variables (for example, race, socioeconomic status, education) were not systematically recorded, limiting generalisability. Attrition over time reduces confidence in longer‑term outcome estimates, and the sample—drawn from a referral‑based, mostly self‑pay clinic—may not represent more severely ill or publicly funded populations. Despite these caveats, the authors conclude that racemic ketamine infusion therapy was associated with clinically meaningful improvements in TRD, SI, and GAD symptoms in this large community cohort. They recommend further prospective, controlled comparator trials (including head‑to‑head studies versus rTMS and ECT) and randomised trials focused on ketamine as a primary treatment for suicidal ideation to validate and clarify the role of ketamine in routine clinical care.

Ketamine infusions in this large real‑world cohort were associated with substantial reductions in depressive symptoms, suicidal/self‑harm ideation, and anxiety, with most gains occurring during the standard six‑infusion acute course and additional benefit evident up to about ten infusions. Given the study's limitations, the authors call for prospective comparator and randomised trials to confirm effectiveness and inform placement of racemic ketamine relative to other procedural options for TRD.