A single subanesthetic IV ketamine infusion produced large, rapid reductions in depressive symptoms in people with MDD or PTSD at 2 hours and 1 day. Ketamine caused transient declines in attention, executive function and verbal memory at 2 hours that resolved by 1 day (attention impairment was greater in patients than controls), did not affect working memory, and cognitive changes were unrelated to antidepressant response.
Papers cited by this study that are also in Blossom
Serafini, G., Howland, R. H., Rovedi, F. et al. · Current Neuropharmacology (2014)
Intravenous (IV) subanesthetic doses of ketamine have been shown to reduce psychiatric distress in both major depressive (MDD) and posttraumatic stress disorder (PTSD). However, the effect of ketamine on cognitive function in these disorders is not well understood. To address this gap, we examined the effect of a single dose of IV ketamine on cognition in individuals with MDD and/or PTSD relative to healthy controls (HC). Psychiatric (n = 29; 15 PTSD, 14 MDD) and sex- age- and IQ matched HC (n = 29) groups were recruited from the community. A single subanesthetic dose of IV ketamine was administered. Mood and cognitive measures were collected prior to, 2 h and 1 day post-ketamine administration. MDD/PTSD individuals evidenced a large-magnitude improvement in severity of depressive symptoms at both 2-hours and 1 day post-ketamine administration (p’s < .001, Cohen d’s = 0.80–1.02). Controlling for baseline performance and years of education, IV ketamine induced declines in attention (ATTN), executive function (EF), and verbal memory (VM) 2 h post-administration, all of which had resolved by 1 day post-ketamine across groups. The magnitude of cognitive decline was significantly larger in MDD/PTSD relative to HC on attention only (p = .012, d = 0.56). Ketamine did not affect working memory (WM) performance. Cognitive function (baseline, change from baseline to post-ketamine) was not associated with antidepressant response to ketamine. Results suggest that while ketamine may have an acute deleterious effect on some cognitive domains in both MDD/PTSD and HC individuals, most notably attention, this reduction is transient and there is no evidence of ketamine-related cognitive dysfunction at 1 day post-administration.
Ketamine is an N-methyl-D-aspartate (NMDA) receptor antagonist with rapid antidepressant and anti‑suicidal effects when given intravenously (IV) at subanesthetic doses. Its growing off‑label clinical use in major depressive disorder (MDD) and posttraumatic stress disorder (PTSD), and the approval of intranasal esketamine for treatment‑resistant depression, have raised concerns about central nervous system risks, particularly cognitive dysfunction. Prior experimental and clinical literature indicates that high doses and chronic abuse of ketamine can impair attention, memory and psychomotor function, but evidence is limited regarding the cognitive effects of therapeutic, subanesthetic IV ketamine in people with MDD or PTSD; notably, no prior study had examined PTSD specifically, and existing MDD studies report inconsistent results that may reflect methodological differences. Davis and colleagues set out to characterise the acute cognitive effects of a single therapeutic (subanesthetic) IV ketamine dose in adults with MDD and/or PTSD compared with matched healthy controls (HC). The investigators assessed attention, working memory, executive function and verbal memory at baseline, 2 hours and 1 day after ketamine administration, and tested three main questions: whether ketamine produces rapid mood improvement in the clinical group, whether it causes an acute cognitive decline at 2 hours that resolves by 1 day, and whether baseline symptom level influences cognitive responses. Tests were chosen for sensitivity to ketamine effects and for low practice effects at short retest intervals.
Participants were community‑recruited adults aged 18–60 years, English speaking and without recent regular psychiatric medication use. The clinical group comprised people meeting DSM‑IV‑TR criteria for MDD (n = 14) and/or PTSD (n = 15) on the SCID, required to be in a current major depressive episode if diagnosed with MDD, free of other psychiatric disorders except anxiety disorders, and off psychotropic medication for at least 2 months. The healthy control group (n = 29) had no current or lifetime psychiatric diagnoses and no first‑degree relatives with psychiatric illness. Groups were matched on age, sex and smoking status; years of education differed between groups. Ketamine was administered intravenously in a neuroimaging context under medical supervision with monitoring of vital signs before and during the 3 h post‑injection period. Two closely similar dosing regimens were used across participants because of parallel imaging protocols: half the subjects received an initial bolus of 0.23 mg/kg over 1 min followed by 0.58 mg/kg per hour for 1 h, and the other half received a constant infusion of 0.5 mg/kg over 40 min. The investigators report no substantive differences in demographics, clinical variables or cognition as a function of dosing regimen. Mood was assessed using the Structured Clinical Interview for DSM (SCID) at screening, and symptom measures included the Hamilton Depression Rating Scale, Beck Depression Inventory‑II (BDI‑II) and Montgomery–Åsberg Depression Rating Scale (MADRS) at baseline, 2 h and 1 day post‑ketamine. Treatment response was defined using the minimum clinically important difference (MCID) on the BDI‑II (three points). Premorbid IQ was estimated with the Wechsler Test of Adult Reading. Cognition was measured with Cogstate tests selected for sensitivity to ketamine effects and minimal practice effects: Detection (DET) for psychomotor speed, Identification (IDN) for visual attention, One Back (ONB) for working memory speed, One Card Learning (OCL) for visual learning accuracy, Groton Maze Learning Test (GMLT) for executive function (errors), and International Shopping List Test (ISLT) for verbal memory. Main measures were standardised to age‑stratified normative data, signs were adjusted so that negative z‑scores indicated poorer performance, and outcomes were combined into an attention composite (DET+IDN), a working memory composite (OCL+ONB) and two individual domain scores (GMLT for executive function, ISLT for verbal memory). Statistical analysis proceeded in stages. Group matching was tested with chi‑squared tests and ANOVAs. Repeated‑measures ANOVA examined mood change across time and diagnostic group. For cognitive outcomes, change scores from baseline to each post‑ketamine timepoint were analysed using ANCOVA models with baseline score and years of education as covariates (one model per cognitive outcome). The clinical group was combined (MDD+PTSD) for primary analyses to preserve power and because prior literature suggested similar effects; exploratory analyses compared MDD and PTSD subgroups. Multivariate ANCOVA was used to examine relationships between baseline cognition, diagnosis and treatment response. Cohen's d was reported for effect sizes and the Benjamini–Hochberg procedure controlled for multiple comparisons.
Sample characteristics showed the MDD/PTSD and HC groups were similar in age, sex, IQ and BMI, although years of education differed (t = 8.78, p = 0.004). Baseline mood ratings indicated moderate depression in the clinical group (MADRS mean 22.7 ± 4.7; BDI‑II mean 22.0 ± 8.7). No cognitive scores or change scores met the pre‑specified outlier criterion, so no data were excluded. Mood outcomes: Repeated‑measures ANOVA revealed a significant interaction of diagnostic status and depressive symptom severity across timepoints (F = 11.39, p < .001, d = 0.86). Mean symptom severity in the MDD/PTSD group decreased from moderate at baseline to mild at both 2 h and 1 day post‑ketamine. Using the BDI‑II MCID, 9/14 (64.3%) MDD participants were classified as responders, improving on average 11.6 points (52.5%) at 2 h and 8 points (36.4%) at 1 day; 11/15 (73.3%) PTSD participants were responders, improving on average 11.8 points (53.3%) at 2 h and 9.3 points (42.2%) at 1 day. On the clinician MADRS at 2 h, PTSD participants had significantly lower scores than MDD participants (mean difference 7.1 points, p < .001, d = 0.99). Cognitive outcomes: At 2 h post‑ketamine, both groups showed declines in executive function (EF) and verbal memory (VM); healthy controls also evidenced declines in attention (ATTN), VM and EF at 2 h, but the pattern differed between groups. In the combined MDD/PTSD group, attention, EF and VM performance declined at 2 h and returned to baseline by 1 day. Working memory (WM) performance was not affected by ketamine in either group at either timepoint. When comparing groups, the magnitude of decline in attention was significantly larger in the MDD/PTSD group than in HC (F = 6.99, p = .012, d = 0.56); EF and VM declines did not differ significantly between groups. All ketamine‑related cognitive declines resolved by 1 day after dosing. Exploratory analyses: Direct comparisons between MDD and PTSD subgroups showed no significant differences on attention or verbal memory. Individuals with MDD performed worse than those with PTSD on executive function at 2 h and on working memory at 1 day (WM: F = 4.56, p = .031). No significant relationships were found between baseline cognitive functioning and treatment responder status, nor between change in cognitive functioning and responder status.
Davis and colleagues interpret the findings as showing robust, rapid antidepressant effects of a single subanesthetic IV ketamine dose in both MDD and PTSD, accompanied by transient, domain‑specific cognitive effects. Specifically, ketamine produced acute declines in attention, executive function and verbal memory at 2 h post‑administration in the combined clinical sample, with the attentional decrement larger in the MDD/PTSD group than in healthy controls; all observed cognitive impairments resolved by 1 day. Working memory was unaffected. The authors note that, from a neuropsychopharmacological standpoint, therapeutic IV ketamine induces short‑lived impairments in higher cognitive functions, and in their mixed clinical sample this effect extends to attention. The discussion situates these results within prior literature showing that NMDA antagonists can disrupt motor coordination and attention, and that higher ketamine doses impair psychomotor performance relevant to driving safety. The transient nature of the deficits observed here—restricted to the acute period and resolving after drug clearance—is consistent with human experimental models and with recovery seen after abstinence in chronic users. By including a PTSD subgroup, the study contributes new data suggesting generally comparable acute cognitive effects across MDD and PTSD, with a few domain‑specific differences in exploratory analyses. The investigators acknowledge several limitations that temper interpretation. The study lacked a randomised, placebo‑controlled design, so practice effects, fatigue or stress cannot be ruled out as contributors to cognitive changes. Sample size was modest, limiting representativeness and statistical power. Only short‑term effects (2 h and 1 day) were assessed, so longer‑term cognitive consequences and the exact timecourse of recovery remain unknown. The cognitive battery, while sensitive to ketamine effects, is not exhaustive and differences from other studies may reflect measure selection. Procedural variability in dosing regimens (two similar infusion protocols) was present, though the authors report no effect of dosing on outcomes. Finally, the clinical sample excluded participants taking psychiatric medication, limiting generalisability to broader patient populations. Given the potential impact of cognitive deficits on functional recovery in MDD and PTSD, the authors recommend longitudinal studies with larger, more representative samples, inclusion of cognitive batteries in clinical trials of ketamine, and careful attention to study design when evaluating ketamine's cognitive risks and benefits.
Prior to analyses, the main performance measure for each cognitive outcome measure was standardized using the mean and standard deviation of age stratified (18-34, 35-49, 50-59, and 60-69 years old) normative data. Because the numerical direction for abnormal scores was different for tests that used accuracy (i.e., higher values indicate less impairment) and speed (i.e., higher values indicate greater impairment), the signs of standardized scores were adjusted so that negative signs indicated performance worse than the relevant age matched mean and positive scores indicated performance better than the age matched mean. To reduce the number of outcome measures and also the potential for Type I error, the main outcome measures from the seven cognitive tests were compiled into two composite scoresand two individual scores. An attention composite was computed by averaging the age-standardized scores from the DET and IDN test. A WM composite was computed by averaging the age-standardized scores on the OCL and ONB tests. For each subject, the attention composite and the WM composite was re-standardized with reference to the mean and SD of the same composite scores derived from the normative data. The four age-standardized outcome measures (hereafter termed cognitive outcomes) were then organized into the main cognitive domain they measured with attention (ATTN) measured by the DET/ IDN composite, WM measured by the OCL/ONB composite, EF measured by the GMLT, and VM measured by the ISLT and submitted to statistical analyses.
Consistent with both study hypotheses and prior findings, significant improvement in depressive symptom severity was observed in individuals with both MDD and PTSD at both 2 h and 1 day post-ketamine administration. Our hypotheses regarding the effect of ketamine on cognition, however, were only partially supported. In the MDD/PTSD group, ATTN, EF, and VM performance declined 2 h post-ketamine administration, and then returned to baseline 1 day after dosing. In healthy controls this same pattern of change was observed for EF and VM, but not ATTN, which showed no significant decline. IV ketamine administration had no acute effects on working memory in either the MDD/PTSD or healthy adults. Comparison of the acute ketamine related change between the MDD/PTD group and healthy adults indicated the effects on EF and VM were equivalent in magnitude, whereas ketamine related decline in ATTN was significant greater in the MDD/PTSD group. From a neuropsychopharmacological perspective the current results show that IV ketamine, given at established therapeutic doses, induces an acute and substantial decline in higher cognitive functions such as EF and VM in both MDD/PTSD and in healthy adults. In MDD/PTSD, this effect extends to attentional functions. In healthy adults, IV ketamine was not associated with any acute decline in attention or working memory. Importantly, all negative effects on cognition resolved completely following clearance of ketamine (i.e., 1 day after dosing). The pattern of ketamine-related cognitive decline here is consistent with findings from previous studies which have examined the effects of the drug in MDD. This finding suggests that ketamine administration induces an acute reduction that was clinically meaningful and domain specific in MDD/PTSD. Examination of previous literature helps to put this finding in context; NMDA antagonists, including ketamine, have been observed to disrupt motor coordination in both human and preclinical research (e.g., induce symptoms including ataxia, catalepsy). Indeed, higher doses of ketamine, sufficient to induce an analgesic response, have been shown to induce acute deficits in attention and psychomotor functioning (reflected in steering variation and swerving on a driving simulator), which has been found to be sufficient to impair driving safety. Thus, both the observed impairment in attention and psychomotor speed, and restriction of this reduction to the acute period following ketamine administration, is consistent with previous research. Overall, these findings add to the growing literature on the relationship between subanesthetic ketamine and cognitive functioning, both acutely following administration and after allowing for clearance of the drug. Of further interest, we address this issue in a mixed sample of participants with MDD and PTSD for the first time. Interestingly, exploratory analyses suggested generally equivalent effects of ketamine on cognition (acute decline) in those with MDD and PTSD. With two exceptions (EF scores 2 h post-ketamine and VM scores 1 day post), results among individuals with MDD and PTSD were comparable. Cognition in those with MDD and PTSD on other measures did not differ significantly at any timepoint, or on EF and VM at any other timepoints. Further, a majority of participants in both diagnostic groups qualified as responders to ketamine based on both self-report and clinician-administered measures, supporting the utility of ketamine as a potential antidepressant treatment for both conditions. Nonetheless, understanding the relationship between ketamine and cognition remains essential to its safe use as a therapeutic agent in both MDD and PTSD. Some research has suggested that cognitive deficits may mediate the relationship between depression and functional impairment, and that lasting cognitive impairment can impede functional recovery even following remission of MDD. With the exception of one acute impairment specific to the domain of attention, we found that one IV administration of subanesthetic doses of ketamine-consistent with current clinical practice-did not substantively affect cognitive functioning. These findings contribute to a growing body of literature supporting the use of ketamine as a relatively safe and effective therapeutic agent for potentially treatment-resistant conditions including MDD and PTSD. A number of methodological limitations warrant consideration in evaluating these results. First, the primary limitation of this study is the lack of randomized/controlled design and absence of a placebo condition. It is possible that other variables, including practice effects, fatigue, and stress (known to have a deleterious effect on cognitive functioning) affected cognition during the course of the study. In the absence of a placebo condition, the potential impact of such variables cannot be quantified. Thus, the lack of a randomized, controlled design in this study limits conclusions. Second, the sample size was relatively small, limiting the variability and therefore possibly the representativeness of the sample. Notably, modest sample sizes are common among published evaluations of the relationship between ketamine and cognition to date, a fact which may contribute to variability in observed results (due to limited generalizability, possible confounds related to statistical power where not carefully managed). Third, the relationship between ketamine and cognitive functioning was only evaluated short-term over two timepoints. As such, conclusions concerning the likely effect of ketamine on cognition in the long term cannot be drawn. Likewise, conclusions concerning the immediate effects of ketamine on cognition and how rapidly they resolve cannot be drawn. Fourth, the cognitive battery used in our study was extensive, but not comprehensive. Observed variability in results between this and other comparable studies may be attributable in part to use of measures evaluating slightly different cognitive domains. Fifth, as noted one half of participants in each group received a slightly IV ketamine according to a slightly different dosing regimen (0.08 mg difference, 1 h vs. 40 min period), resulting in some procedural variability. Of note, careful examination of data prior to completion of primary analyses revealed no effects of dosing on clinical of cognitive variables. Similarly, no group differences in clinical or demographic variables were observed prior to ketamine administration. Thus, we are confident that procedural variability did not confound results. Further, conclusions are limited to a single acute IV administration of ketamine, and therefore provide limited insight concerning the effect of chronic administration on cognitive functioning. Finally, the population recruited was not currently taking psychiatric medication, and was thus not wholly representative of the psychiatric population in general. As researchers and clinicians continue to explore, and potentially expand the use of ketamine for the treatment of high-risk clinical populations, systematic evaluation of the treatment's effect-both single and more protracted dosing-on cognitive functioning should be undertaken. A thorough understanding of the relationship between ketamine and cognition will begin with longitudinal examination in a large, representative sample (within or across populations such as MDD and PTSD to facilitate examination of potentially meaningful differences between them) with consideration of other relevant clinical variables including stress exposure. In light of the potential impact of cognition on functional outcomes in both MDD and PTSD, inclusion of cognitive batteries into clinical trials involving ketamine should be considered. Further, it is essential that researchers consider the impact of research design and methodology on their ability to answer questions concerning ketamine's effect on cognition.
Create a free account to open full-text PDFs.
Abdallah, C. G., Charney, D. S., Duman, R. S. et al. · Neuron (2019)
Sanacora, G., Frye, M. A., McDonald, W. et al. · JAMA Psychiatry (2017)
Short, B., Fong, J., Galvez, V. et al. · Lancet Psychiatry (2017)
Wilkinson, S. T., Katz, R. B., Toprak, M. et al. · Journal of Clinical Psychiatry (2018)
Basso, L., Bönke, L., Aust, S. et al. · Journal of Psychiatric Research (2020)
Murrough, J. W., Burdick, K. E., Levitch, C. F. et al. · Neuropsychopharmacology (2014)