If you don't remember your password, you can reset it by entering your email address and clicking the Reset Password button. You will then receive an email that contains a secure link for resetting your password
If the address matches a valid account an email will be sent to __email__ with instructions for resetting your password
Jane and Terry Semel Institute for Neuroscience and Human Behavior, Department of Psychiatry and Biobehavioral Sciences, Geffen School of Medicine at the University of California, Los Angeles, CA, USA
Major depressive disorder is a highly prevalent psychiatric disorder. Despite an extensive range of treatment options, about a third of patients still struggle to respond to available therapies. In the last 20 years, ketamine has gained considerable attention in the psychiatric field as a promising treatment of depression, particularly in patients who are treatment resistant or at high risk for suicide. At a sub-anesthetic dose, ketamine produces a rapid and pronounced reduction in depressive symptoms and suicidal ideation, where serial treatment appears to produce a greater and more sustained therapeutic response. However, the mechanism driving ketamine’s antidepressant effects is not yet well understood. Biomarker discovery may advance knowledge of ketamine’s antidepressant action, which could in turn translate to more personalized and effective treatment strategies. At the brain systems-level, neuroimaging can be used to identify functional pathways and networks contributing to ketamine’s therapeutic effects by studying how it alters brain structure, function, connectivity, and metabolism. In this review, we summarize and appraise recent work in this area, including 51 articles that use resting-state and task-based functional MRI, arterial spin labeling, positron emission tomography, structural MRI, diffusion MRI, or magnetic resonance spectroscopy to study brain and clinical changes 24 hours or longer after ketamine treatment in populations with unipolar or bipolar depression. Though individual studies have included relatively small samples, different methodological approaches and report disparate regional findings, converging evidence supports that ketamine leads to neuroplasticity in structural and functional brain networks that contribute to or are relevant to its antidepressant effects.
A systematic review and meta-analysis of randomized, double-blind, placebo-controlled trials of ketamine in the rapid treatment of major depressive episodes.
). Though symptom relief is typically transient (<1 week), larger and more sustained antidepressant response and remission occur with serial ketamine therapy(
In-vivo studies of brain structure, function, connectivity and metabolism can help discern the functional pathways and systems contributing to ketamine’s therapeutic effects at the macroscale(
Resting-state connectivity studies as a marker of the acute and delayed effects of subanaesthetic ketamine administration in healthy and depressed individuals: A systematic review.
) from neuroimaging studies in MDD published in English from 2011-2022 that investigate the effects of ketamine on functional and structural brain systems, or imaging biomarkers predictive of ketamine antidepressant response. Original research articles that collected brain imaging and clinical outcomes ≥24 hours post-ketamine administration in adult populations with unipolar or bipolar depression were included. Preclinical studies addressing the molecular and cellular mechanisms of ketamine(
Neural correlates of rapid antidepressant response to ketamine in treatment-resistant unipolar depression: a preliminary positron emission tomography study.
). Figure 1 elaborates search methods and inclusion criteria.
Figure 1Original research articles were identified and evaluated according to the PRISMA review method. Literature searches were performed in the PubMed and Google Scholar databases using key words (right) and the inclusion/exclusion criteria (bottom) provided. For more details regarding methods see Supplemental Material.
Extensive neuroimaging evidence supports the pathophysiology of MDD includes the dysregulation of multiple large-scale brain networks. For example, connectivity within or between the default mode (DMN), salience (SN), cingulo-opercular (CON), frontoparietal/central executive network (FPN/CEN), orbito and ventromedial-affective (OAN and VMN) networks are repeatedly implicated in depression, and appear to influence treatment outcomes(
). These networks and their nodes are pivotal for emotion processing, autonomic responses to emotion, memory and social cognition (hippocampal, amygdala, septal area, anterior thalamus and hypothalamus), higher cognition, motivation and mood regulation (dorsal and subgenual anterior cingulate cortex (dACC/sgACC) and dorsolateral (DLPFC) and parietal association regions), reward processing (ventral striatum, habenula) and vegetative states (midbrain/brainstem structures)(
). Accordingly, the majority of recent ketamine imaging studies have used blood-oxygen-level-dependent (BOLD) resting-state functional magnetic resonance imaging (rsfMRI) to measure the temporal coherence of intrinsic brain activity across different functional brain systems in the absence of a specific task(
). In this review, we report rsfMRI findings for large-scale cortical networks and nodes, and for whole brain and global brain connectivity (Table 1).
Table 1Resting state fMRI studies of ketamine treatment in depressed populations.
Study
Sample
Other Medications
Ketamine Treatment
Data Collection
Depression Assessment
Summary of Findings
Abdallah et al. 2017(49); NCT00768430
NA: 22 Unipolar TRD***, 47 HC§
NO Antidepressant and antipsychotics-free ≥1week prior to treatment.
Cohort A++: •Single IV ketamine, 0.5mg/kg over 40 mins •Placebo: 0.45mg/kg midazolam over 40 mins Cohort B++: •Single IV ketamine, 0.23mg/kg in 2 mins followed by 0.58mg/kg over 70 mins •Placebo: oral placebo •Saline IV
MADRS, QIDS, Brief psychiatric rating scale, Clinician-administered dissociative state scale
(+) GBCr post ketamine in bilateral dmPFC and right dlPFC for TRD, normalizing to HC No significant change in GBCr detected in placebo group
Abdallah et al., 2017(47); NCT00548964, NCT00768430, NCT01880593
N: 18 Unipolar TRD**, 25 HC
NO Antidepressant and antipsychotics-free ≥1week prior to treatment. Benzodiazepines witheld at day of each scan
Single IV infusion 0.5mg/kg ketamine over 40 mins+
Neuroimaging: •Baseline •24 hours post infusion Clinical Assessment: •Baseline •24 hours post infusion
MADRS Response was defined as ⩾50% reduction in MADRS
(+) GBCr in right lateral PFC after treatment for all TRD participants (-) GBCr in left cerebellum after treatment for all TRD participants (+) Change in GBCr (post-pre ketamine) in right lateral PFC and left anterior insula in Rs (+) GBCr postketamine in bilateral caudate, bilateral lateral PFC, and middle temporal of Rs (-) FC within PFC after ketamine (+) FC between PFC and other brain regions after ketamine
Abdallah et al., 2018(48); NCT01046630
N: 58 Unipolar MDD
NO
Single IV infusion 0.5mg/kg ketamine, 100mg of lanicemine, or saline placebo over 60 mins++
(+) PFC GBCr in ketamine during infusion and at 24hr post-treatment compared to baseline (+) GBCr in dlPFC, dmPFC, and fmPFC during ketamine infusion (+) GBCr in dorsolateral and dorsomedial PFC at 24 hrs post-treatment Vertex wise analysis showed depression improvement positively correlated with GBCr in dPFC during infusion and 24hrs post treatment but negative correlation with GBCr in ventral PFC during infusion
Evans et al., 2018(31); NCT00088699
N: 33 Unipolar TRD*, 25 HC§
NO Medication free ≥2 weeks prior to treatment
Single IV infusion 0.5mg/kg ketamine or saline placebo during first session followed by alternative treatment 2 weeks later+++
Neuroimaging: •Baseline •2 and 10 days post infusion Clinical Assessment: •Baseline •40, 80, 120 and 230 mins post infusion on days 1, 2, 3, 7, 10 and 11
MADRS
(+) RSFC between bilateral insula, middle frontal gyrus, post-central gyrus, and occipital gyrus and DMN 2 days after ketamine infusion compared to placebo infusion that was no longer evident at day 10 RSFC between insula (strongest in right posterior insula) and DMN normalized for MDD 2 days after ketamine- no significant difference between TRD and HC - that returned to baseline at day 10
Chen et al., 2019(33); R000019001-UMIN000016985
N: 48 Unipolar TRD**
YES Permitted to remain on antidepressants
Single IV infusion 0.5mg/kg ketamine, 0.2mg/kg ketamine, or saline placebo over 40 mins+++
Neuroimaging & Clinical Assessment: •Baseline •48 hours post infusion
MADRS
(-) RSFC between left and right dACC after ketamine, associated with reduction in suicidal ideation (-) in RSFC between dlPFC and right frontal pole in 0.5mg/kg group (+) RSFC between right dlPFC and left superior parietal lobe associated with reduced suicidal ideation in 0.2mg/kg group (+) RSFC between right dACC and right anterior middle temporal gyrus in 0.2mg/kg group
Gärtner et al., 2019(39); NCT02099630, NCT03609190
NB: 24 Unipolar TRD**
YES Prior antidepressant medication remained unchanged
Site A: Single IV infusion 0.5mg/kg racemic ketamine+ Site B: Singe IV infusion 0.25mg/kg S-ketamine+
Neuroimaging & Clinical Assessment: •Baseline •24 hours post infusion
MADRS, HDRS
(+) RSFC between sgACC and 4 regions (SMA, dlPFC, anterior PFC, and OFC) associated with symptom reduction ↓Whole-brain connectivity at baseline associated with (-) depressive symptoms 24 hrs after ketamine
Single IV infusion at 0.5mg/kg or 0.2mg/kg ketamine or saline placebo over 40 mins+++
Neuroimaging: •Baseline Clinical Assessment: •Baseline •40, 80, 120, and 240 mins post infusion •2, 3, 4, 5, 6, 7, and 14 days post infusion
HDRS
↓Baseline RSFC between bilateral superior frontal cortex and striatum associated with (-) depressive symptoms after 0.2mg/kg ketamine infusion
Kraus et al., 2020(50); NCT00088699
N: 28 Unipolar TRD*, 22 HC§
NO Medication free ≥2 weeks prior to treatment
Single IV infusion 0.5mg/kg ketamine or saline placebo during first session followed by alternative treatment 2 weeks later+++
Neuroimaging: •Baseline •2 and 3 days post infusion •Baseline Clinical Assessment: •Before, during, and after each infusion
MADRS, BID, Hamilton Anxiety Rating Scale
Could not replicate (+) connectivity observed in Abdallah et al. 2017 in Neuropsychopharmacology after ketamine treatment
Sahib et al., 2020(46); NCT02165449
N: 61 Unipolar TRD**, 40 HC
YES Monoaminergic antidepressants allowed if stable ≥6 weeks prior to treatment No Benzodiazapine ≤72 hours prior to treatment
Four serial IV infusions of 0.5mg/kg ketamine over 40 mins+
Neuroimaging & Clinical Assessment: •Baseline •24 hours post first infusion •24-72 hours post fourth infusion
HDRS-17 Rm: post-treatment HDRS ≤7
(+) RSFC within SMN after ketamine, normalizing towards HC (-) RSFC between ventral node and visual node after ketamine, normalizing towards HC. ↑RSFC between visual cortex and DMN at baseline that (-) with each infusion (-) RSFC between cerebellum and SN between first and fourth ketamine treatment, normalizing towards HC (-) RSFC between cerebellum and striatum for Rm, normalizing towards HC. Opposite effect in NRm pre- and post-treatment change in whole brain FC significantly different between Rs and NRs
Zhuo et al., 2020(51)
N: 38 Bipolar TRD**
YES Remained on mood stabilizer or antipsychotic medication if clinically stable
Serial IV infusion of 0.5mg/kg ketamine over 40-50 mins on days 2, 4, 6, 8, 10, 12, 15, 18, and 20 of study+
Neuroimaging: •Baseline •Day 2, 7, 14, and 21 after beginning treatment
HDRS-17
Changes in GBCd after first day of ketamine and peaked at day 7. Effects mostly gone by end of third week. (-) GBCd in bilateral insula, right caudate nucleus and bilateral inferior frontal gyrus (+) GBCd in bilateral postcentral gyrus, sgACC, bilateral thalamus, and cerebellum. No significant associations with clinical measure
Mkrtchian et al., 2020(42); NCT00088699
N: 33 Unipolar TRD*, 25 HC§
NO Medication free ≥2 weeks prior to treatment
Single IV infusion of 0.5mg/kg ketamine or saline placebo during first session followed by alternative treatment 2 weeks later+++
Neuroimaging: •2 days post infusions Clinical Assessments: •Baseline •40, 80, 120, and 230 mins on 1, 2, 3, 7, 10, and 11 days post infusions
MADRS, SHAPS
(+) RSFC between VS-left dlPFC, DC-right ventrolateral PFC, DC-pgACC, and VRP-OFC in TRD but (-) in HC post-ketamine. (+) RSFC in DC-right vlPFC correlated with improved day 2 SHAPS (+) RSFC in DC-pgACC correlated with improved day 10 SHAPS
Nakamura et al., 2021(40); UMIN-CTR No. UMIN000017529
N: 14 unipolar TRD***, 1 bipolar TRD*
NO Discontinuation of antidepressant medications
Serial IV infusions of ketamine over 40 mins twice a week over 2 weeks (4 total) with concurrent daily oral placebo or lithium carbonate (600-800mg) daily++
Neuroimaging: •Baseline •6-24 hours post infusions Clinical Assessments: •4 hours post final infusion
MADRS, Young Mania Rating Scale, Rs: Decrease in MADRS ≥50%
Significant cluster defining Rs and NRs between the amygdala and sgACC in the right AN (-) RSFC between amygdala and sc/sgACC in the right hemisphere at baseline associated with (+) depressive symptoms (MADRS) RSFC between amygdala and sc/sgACC in the right hemisphere Rs>NRs both at baseline and followup
Siegel et al., 2021(30); NCT01179009
N: 23 Unipolar TRD**, 27HC
YES SSRI and SNRI allowed if constant for ≥6 weeks prior to infusion
Continuous 96h IV infusion of ketamine started at 0.15mg/kg/h at 10am on day 1 and titrated to tolerance twice daily to target rate of 0.6mg/kg/h+
Neuroimaging: •Baseline •2 weeks post infusion Clinical Assessments: •2, 4, 6, and 8 weeks post infusion
MADRS
(-) RSFC within DMN 2 weeks after ketamine (-) RSFC between sgACC and DMN 2 weeks after ketamine across hemispheres (+) RSFC between sgACC and bilateral cACC and bilateral anterior insula (-) RSFC within limbic system especially with anterior thalamus (+) RSFC between limbic regions and cortical areas, especially cingulo-opercular network
Rivas-Grajales et al., 2021(44); NCT00768430, NCT01880593
N: 35 Unipolar TRD**
NO Antidepressant free ≥1 week prior to treatment
Single IV infusion of 0.5mg/kg ketamine over 40 mins+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post infusion
MADRS QIDS-SR
(+) RSFC between right habenula and right frontal pole of dlPFC associated with (-) MADRS scores (+) RSFC between right habenula and right occipital pole, right temporal pole, right parahippocampal gyrus, and left lateral occipital cortex associated with (-) QIDS-SR
Vasavada et al., 2021(34); NCT02165449
N: 44 Unipolar TRD**, 50 HC
YES Monoaminergic antidepressants allowed if stable ≥6 weeks prior to treatment No Benzodiazapine ≤72 hours prior to treatment
Four serial IV infusions of 0.5mg/kg ketamine over 40 mins+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post first infusion •24-72 hours post fourth infusion
HDRS-17 SHAPS DASS BIS
(+) RSFC between right amygdala and right CEN normalizing towards HC with each infusion (-) RSFC between left amygdala and SN with each infusion and correlated with improved BISS. Greater negative connectivity between right hippocampus and left CN correlated with improved SHAPS
Zhang et al. 2020(32)
N: 36 Unipolar TRD**
-
Six serial IV infusions of 0.5mg/kg over 40 mins followed by propofol-ECT treatment 24 hours later+
Neuroimaging: •Baseline •1st, 3rd, 7th, 10th, and 14th day after first treatment Clinical Assessment: •Baseline •7th and 14th day of treatment
HAMD
(-)GBCd in medial prefrontal lobe, sgACC, posterior cingulate, thalamus, hippocampus, and orbitofrontal lobe on day 7 (-) FC within DMN through third day
*≥1 failed treatment **≥2 failed treatment ***≥3 failed treatment TRD: Treatment resistant depression MDD: Major Depressive Disorder HC: Healthy Control §: HC received ketamine +: Open Label ++: Randomized Placebo +++: Randomized Crossover IV: Intravenous Hrs: Hours Mins: Minutes MADRS: Montgomery-Åsberg Depression Rating Scale QIDS: Quick Inventory of Depressive Symptomatology QIDS-SR: Quick Inventory of Depressive Symptomatology Self Report (+): Increase (-): Decrease ECT: Electroconvulsive therapy
HDRS: Hamilton Depression Rating Scale BDI: Beck Depression Inventory DASS: Depression Anxiety and Stress Scale BISS: Behavioral Inhibition System Scale SHAPS: Snaith-Hamilton Pleasure Scale Rs: Responders NRs: Non responders Rm: Remitters NRm: Non remitters GBCr: Global brain connectivity regression GBCd: Global functional connectivity density RSFC: Resting state functional connectivity PFC: Prefrontal cortex dmPFC: Dorsomedial Prefrontal cortex fmPFC: Frontomedial Prefrontal cortex ↑: Higher ↓: Lower
ATwo studies were pooled. Cohort A conducted at one site in Houston as an add on to a ketamine clinical trial. Cohort B was conducted at a different site in New Haven CT to investigate modulation of ketamine effects by lamotrigine and only included healthy controls
BTwo different scanners used to collect rsfMRI data. Scanning site used as a covariate in the analysis. Post-hoc analysis found similar results when testing sites individually
Several independent rsfMRI studies report changes in resting-state functional connectivity (RSFC) in MDD in the days following ketamine treatment. The DMN, engaged during introspective activities and consisting of anterior/dorsal (medial PFC, precuneus, and inferior parietal cortex) and ventral subsystems (hippocampus/parahippocampus), is perhaps the most widely studied large-scale network in depression(
). Several studies have reported ketamine-related changes in DMN RSFC. For example, 2 weeks following a continuous 96-hour intravenous (IV) ketamine infusion (0.6mg/kg/hour), RSFC decreased within ventral DMN limbic nodes, but increased between subcortical and cortical DMN/CON nodes(
). However, these patterns did not vary based on ketamine response. A separate study similarly observed increased RSFC between the DMN and the insula, a key node of the CON and SN, as well as with frontal, parietal and occipital cortices 2 days after single IV ketamine compared with placebo in MDD(31). Here, increased DMN-insula connectivity normalized towards control levels 2-days post-ketamine but reversed 10-days following treatment. Another study found when adding ketamine to propofol-electroconvulsive treatment, RSFC within the DMN decreased(
Ketamine plus propofol-electroconvulsive therapy (ECT) transiently improves the antidepressant effects and the associated brain functional alterations in patients with propofol-ECT-resistant depression.
The FPN/CEN, comprised of dorsal and ventral frontal-parietal subsystems, is a large resting-state network active during attention, cognitive states and emotion-regulation frequently linked with depression. Also implicated in ketamine MRI studies, decreased RSFC has been reported within the frontal component of the CEN 48 hours following single dose ketamine in PFC and dACC seed-based analysis(
Antidepressant and antisuicidal effects of ketamine on the functional connectivity of prefrontal cortex-related circuits in treatment-resistant depression: A double-blind, placebo-controlled, randomized, longitudinal resting fMRI study.
). Negative and positive correlations with suicidal ideation were observed for RSFC between left-right dACC, and DLPFC-left superior parietal cortex for each dose respectively. Explicitly examining RSFC between the hippocampus and amygdala with the DMN, FPN/CEN and SN, increased and normalized RSFC was observed between the right amygdala and the right CEN 24 hours after patients received four serial ketamine infusions(
). Further, decreased left amygdala-SN RSFC associated with improved behavioral inhibition, while negative connectivity between the right hippocampus and the left CEN correlated with improved anhedonia(
The sgACC, considered a node of the VMN and reciprocally connected with limbic, ventral striatum, habenula, and thalamic structures, is involved in modulating emotion and reward response(
), though RSFC between the sgACC and DMN decreased. At least two studies have linked sgACC connectivity with overall antidepressant response post ketamine. Specifically, a multi-site study reported associations between depression symptom improvement(
Functional connectivity between the amygdala and subgenual cingulate gyrus predicts the antidepressant effects of ketamine in patients with treatment-resistant depression.
Functional connectivity between the amygdala and subgenual cingulate gyrus predicts the antidepressant effects of ketamine in patients with treatment-resistant depression.
Prior data also suggest fronto-striatal circuitry influences ketamine response. Here, lower PFC-striatal RSFC at baseline has been found to associate with subsequent improvements in depressive symptoms(
The habenula, part of the epithalamus with direct connections to limbic structures and the basal ganglia, plays a role in emotion, reward and motivation(
). Further, increased RSFC between the right habenula and right and left occipital cortex, right temporal pole, and right parahippocampal gyrus associated with improved subjective mood ratings(
The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression.
The larger literature suggests that several interacting brain networks underlie the pathophysiology of depression, and that multiple functional networks contribute to antidepressant response(
). Prior ketamine rsfMRI studies have mostly focused on targeting major networks or components thereof, potentially missing other changes in the functional connectome. To address this limitation, one study employed a data-driven whole-brain approach before and 24 hours after serial ketamine therapy(
) to show RSFC within and between the somatomotor network (SMN), and FPN/CEN and visual networks (VN) normalized with treatment and distinguished controls from patients at baseline. This study also showed a normalization of circuitry between the cerebellum, SN and striatum, which correlated with antidepressant response(
Global brain connectivity (GBC), which measures the connectivity strength between each voxel to all other gray matter voxels, has been investigated in four ketamine studies. One study(
) found ketamine-related GBC increases in the PFC and reduced GBC in the cerebellum. Treatment responders showed greater GBC changes in the lateral PFC, caudate, and insula 24 hours post single-dose ketamine. Follow-up seed-based analysis suggested ketamine reduced hyper-connectivity within the PFC and enhanced hypo-connectivity between the PFC and other brain regions. In a larger sample, the same investigators found an increase in PFC GBCr 24 hours after ketamine when compared to placebo and baseline(
Evaluating global brain connectivity as an imaging marker for depression: influence of preprocessing strategies and placebo-controlled ketamine treatment.
) did not find significant changes to GBC when neuroimaging was performed 2-3 days post-ketamine. Finally, a study investigating GBC density after serial ketamine in treatment resistant bipolar depression(
Transient effects of multi-infusion ketamine augmentation on treatment-resistant depressive symptoms in patients with treatment-resistant bipolar depression - An open-label three-week pilot study.
) found decreased density in the bilateral insula, right caudate and bilateral VLPFC and increased density in the bilateral postcentral gyrus, sgACC, thalamus, and cerebellum. These changes appeared 24 hours post-ketamine, peaked at one week, and diminished by the third week; significant associations with clinical measures were not detected.
Summary
Subanesthetic ketamine leads to plasticity in multiple resting state networks and their components (Figure 2). Since existing investigations have used different analysis approaches (e.g., theory-driven seed-based vs data-driven whole brain analysis methods), sample sizes are typically small and have overlapped amongst reports, and study designs (open-label, randomized controlled trial, post-ketamine assessment time points) and clinical populations vary, convergence amongst findings remains relatively low. However, this work still provides important insight and future leads concerning the influence of subanesthetic ketamine on functional brain circuity and clinical outcomes. For example, existing data supports that ketamine leads to decreased activity within the DMN(20,34,35) and within the FPN/CEN(24), potentially affecting behaviors associated with negatively-biased self-referential processing and executive functions/emotional regulation in depression, respectively. When generalizing findings, RSFC instead appears to increase between different large-scale cortical networks such as the DMN and FPN/CEN with nodes including the sgACC, anterior insula, striatum, amygdala, habenula)(
The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression.
). Together, these patterns suggest ketamine modulates circuitry that may be both under- and over-reactive prior to treatment. Data-driven results also emphasize the contribution of sensory systems and cortico-striatal-cerebellar loops that encompass the SN, as a potential biomarker for ketamine response(
). Notably, individual studies have shown that changes in RSFC in nodes including the amygdala, sgACC and habenula to other subcortical or higher cortical association regions relate to clinical response(
Functional connectivity between the amygdala and subgenual cingulate gyrus predicts the antidepressant effects of ketamine in patients with treatment-resistant depression.
Antidepressant and antisuicidal effects of ketamine on the functional connectivity of prefrontal cortex-related circuits in treatment-resistant depression: A double-blind, placebo-controlled, randomized, longitudinal resting fMRI study.
Figure 2This illustration synthesizes results across functional MRI studies that have addressed the effects of subanesthetic ketamine on resting state functional connectivity (RSFC) in patients with major depression. Here, connectograms are used to describe increasing (top) and decreasing (bottom) changes in RSFC between specific brain regions and resting state networks observed across studies following ketamine treatment. Corresponding atlases that are color coordinated with the resting state networks (top) and brain regions (bottom) represented in the connectograms are shown on the right. A table detailing the connections and their corresponding references can be found in Supplemental Table 1. Acronyms: DMN=Default Mode Network, CEN=Central Executive Network, CON=Cingulo-opercular Network, SMN=Somatomotor Network, vlPFC= Ventrolateral Prefrontal Cortex, dlPFC=Dorsolateral Prefrontal Cortex, SMA=Sensorimotor Area, aPFC=Anterior Prefrontal Cortex, dACC=Dorsal Anterior Cingulate Cortex, pgACC= Pregenual Anterior Cingulate Cortex, sgACC=Subgenual Anterior Cingulate Cortex, SN=Salience Network, OFC=Orbitofrontal Cortex, Striatum= Caudate and Putamen
Disturbances in emotion regulation and processing and other cognitive functions linked with depression can be probed by examining fMRI BOLD signal as participants actively engage in functional tasks(
Palmer SM, Crewther SG, Carey LM, START Project Team (2014): A meta-analysis of changes in brain activity in clinical depression. Front Hum Neurosci 8: 1045.
). Here, we summarize findings from ketamine studies using task-based fMRI organized by brain activation task, including response inhibition, reward processing, emotional judgment, and emotional face recognition (Table 2).
Table 2Task-based fMRI studies of ketamine treatment in depressed populations.
Study
Demographics
Other Medications
Ketamine Treatment
Data Collection
Depression Assessment
Summary of Findings
Sahib et al., 2020(55); NCT02165449
N: 47 Unipolar TRD**, 32 HC
YES Monoaminergic antidepressants allowed if stable ≥6 weeks prior to treatment No Benzodiazapine ≤72 hours prior to treatment
Four serial IV infusions 0.5mg/kg ketamine over 40 mins+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post first infusion •24-72 hours post fourth infusion
HDRS-17 Rm: End of treatment HDRS≤ 7
Conditioned Approach Response Inhibition (CARIT): (-) activation between baseline and fourth infusion in inferior frontal cortex and dlPFC along superior and inferior parietal regions and right cerebellum (-) in activation in visual cortex and superior parietal regions of left hemisphere after fourth infusion (-) BOLD activity for DMN, FPN, DAN, and SN in right hemisphere following ketamine (-) NoGo>Go activity in bilateral precentral gyrus for NRm after serial ketamine (+) NoGo>Go activity in bilateral precentral gyrus for Rm after serial ketamine Average baseline contrast values in bilateral precentral gyrus show negative correlation with % change in HDRS score after serial treatment (-) in contrast values in bilateral precentral gyrus 24 hrs post ketamine significantly associated with serial HDRS improvement
Loureiro et al., 2021(56); NCT02165449
CN: 46 Unipolar TRD**, 32 HC
YES Monoaminergic antidepressants allowed if stable ≥6 weeks prior to treatment No Benzodiazapine ≤72 hours prior to treatment
Four serial IV infusions 0.5mg/kg ketamine over 40 mins+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post first infusion •24-72 hours post fourth infusion
HDRS-17 QIDS DASS
Conditioned Approach Response Inhibition (CARIT): (-) connectivity between cerebellum and FPN and the SMN in Rm only Baseline connectivity between cerebellum and FPN and cerebellum and SN significantly correlated with ketamine related %change QIDS for both Rm and NRm
Murrough et al., 2015(68); NCT00548964, NCT00768430, NCT01880593
N: 18 Unipolar TRD**, 20 HC
NO Antidepressant medication free ≥1 week prior to treatment
Single IV infusion of 0.5mg/kg ketamine over 40 mins+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post infusion
MADRS
Facial emotion perception task: (+) neural response to positive emotions (happy 100% > neutral) following ketamine centered in the right caudate but not correlated with pre-ketamine depressive symptoms (+) in post-ketamine right caudate connectivity associated with improved MADRS score
Sterpenich et al., 2019(60); NCT01135758
N: 10 Unipolar TRD**
YES Stable medications ≥6 weeks prior to treatment
Single bolus IV infusion 0.5mg/kg ketamine over 1 min+
Neuroimaging: •Baseline •1 day post infusion Clinical Assessments: •Baseline •40, 80, 110, and 230 post infusion
MADRS HDRS-21 BDI-II
Reward task adapted from Monetary Inventive Delay Task: (+) changes in insula and OFC during anticipatory phase of reward task one day after ketamine compared to baseline and in ventral striatum and OFC 7 days after ketamine compared to baseline (more activated for positively cued trials) (+) activation of VS and OFC in response to winning vs losing when comparing day 7 with baseline (+) more active substantia nigra/ventral tegmental area when winning than losing one day after ketamine and 7 days after ketamine compared to baseline Emotional Judgement Task: patient reaction time faster after ketamine (baseline vs day 1 and day 7) but not effect of emotion (-) activation of amygdala and insula response to negative pictures after 1 day after ketamine (-) activation of insula and dACC 7 days after ketamine in response to negative pictures (+) medial substantia nigra/ventral tegmental area more active for negative than positive pictures at baseline and then become (+) active for positive than negative pictures at day 7
NO Antidepressant and other medication free at time of scan
Single IV infusion 0.5mg/kg ketamine over 40 mins+
Neuroimaging: •Baseline •5 days post infusion
MADRS, TEPS, STICSA
Reward Incentive Flanker Task: (-) sgACC activation to positive feedback with ketamine but not negative feedback Higher pre-ketamine sgACC activation to positive feedback associated with better improvements in anhedonia after ketamine
Reed et al., 2018(66); NCT00088699
N: 33 Unipolar TRD*, 26 HC§
NO Medication free ≥2 weeks prior to treatment
Single IV infusion 0.5mg/kg ketamine over 50 mins or saline placebo followed by alternative treatment two weeks later+++
Neuroimaging: •Baseline •Immediately after infusion •1-3 days post infusion Clinical Assessments: •Baseline •40, 80, 120, 240 mins post infusion •1, 2, 3, 7, and 10 days post infusion
MADRS
Dot probe task with emotional face stimuli: (+) activation post-ketamine compared to post-placebo in left middle occipital gyrus across groups (-) activation post-ketamine compared to post-placebo in left temporal and inferior frontal cortices across groups (-) in activation post-ketamine vs post-placebo in MDD patients in right frontal cortex, dACC, and left inferior occipital gyrus Medial prefrontal and anterior cingulate cluster showed deactivation to angry trials and activation to happy trials in MDD participants post placebo and reversed after ketamine %change in MADRS significantly associated with magnitude of activation (positive post ketamine and negative post placebo) (-) MADRS score associated with (-) activation to angry trials and greater activation to happy trials in left parahippocampal gyrus and amygdala, bilateral cingulate gyri, precuneus, and left medial and middle frontal gyri
Reed et al., 2019(67); NCT00088699
N: 33 Unipolar TRD*, 24 HC§
NO Medication free ≥2 weeks prior to treatment
Single IV infusion of 0.5mg/kg over 50 mins and single IV of saline solution placebo two weeks apart+++
Neuroimaging: •Baseline •Immediately after infusion •1-3 days post infusion Clinical Assessments: •Baseline •40, 80, 120, 240 mins post infusion 1, 2, 3, 7, and 10 days post infusion
MADRS
Facial Recognition Task: (-) activation post ketamine in bilateral frontal, temporal, precuneus and posterior cingulate regions in MDD, normalizing towards HC when compared to placebo Greater difference in activity pattern in the left temporal gyri and bilateral precuneus/posterior cingulate between explicit and implicit processing conditions in MDD participants after ketamine
Loureiro et al., 2020(70); NCT02165449
N: 27 Unipolar TRD**, 31 HC
YES Monoaminergic antidepressants allowed if stable ≥6 weeks prior to treatment No Benzodiazapine ≤72 hours prior to treatment
Four serial IV infusions of 0.5mg/kg ketamine over 40 mins+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post first infusion •24-72 hours post fourth infusion
HDRS SHAPS DASS
Emotional faces task: Post-treatment change in fearful>objects contrast correlated with %DASS and %SHAPS change in right amygdala
Downey et al. 2016(71); NCT01046630
N: 56 Unipolar MDD
NO
Single IV 0.5mg/kg ketamine, 100mg lanicemine, or saline placebo over 60 mins++
Neuroimaging: •Baseline •During infusion Clinical Assessments: •Baseline •Immediately post infusion •4 hours and 24 hours post infusion
BDI MADRS
PHMRI
(+) BOLD response by both drugs in ACC predicted symptomatic improvement 24 hours and 1 week following infusion but no antidepressant effect when compared to placebo
McMillan et al. 2020(72); ACTRN12615000573550
N: 26 Unipolar MDD
YES Stable medications ≥4 weeks prior
Single IV 0.25mg/kg bolus ketamine followed by 0.25mg/kg over 45 minutes or placebo separated by 3 week period+++
Neuroimaging: •During infusion Clinical Assessments: •Baseline •3 hours, 1 day, 1 week, and 2 weeks post infusion
MADRS
PHMRI (+) BOLD in right insula and left post-central gyrus during ketamine infusion associated with antidepressant response
Stippl et al. 2021(69)
N: 16 unipolar MDD
YES Patients permitted to remain on psychopharmacological medication
Single IV 0.25mg/kg S-ketamine or 0.5mg/kg racemic ketamine+
Neuroimaging: Baseline Clinical Assessments: Baseline 24 hours post infusion
HDRS BDI
Emotional Working Memory Task: No significant association with change in depressive symptoms
*≥1 failed treatment **≥2 failed treatment ***≥3 failed treatment TRD: Treatment resistant depression MDD: Major Depressive Disorder HC: Healthy Control §: HC received ketamine +: Open Label ++: Randomized Placebo +++: Randomized Crossover IV: Intravenous Hrs: Hours Mins: Minutes BOLD: Blood oxygen level dependent HDRS: Hamilton Depression Rating Scale MADRS: Montgomery-Åsberg Depression Rating Scale QIDS: Quick Inventory of Depressive Symptomatology (+): Increase (-): Decrease
CLoureiro et al 2021 report different sample sizes for each time point. T1 represents baseline, T2 represents 24 hours after single infusion and T3 represents 24 hours after fourth infusion
). Decreased activation in the inferior/dorsolateral PFC and parietal regions, the right cerebellum, and visual cortex were observed during response inhibition pre-post serial ketamine treatment, which normalized towards healthy controls(
). Decreased activation in prefrontal-parietal inhibitory control networks and motor and insular regions 24 hours post-single and serial treatment, were associated with and predicted subsequent improvements in mood and rumination. In an overlapping sample, investigators used psychophysiological-interaction models (PPI) to interrogate relationships between the cerebellum, and FPN, SMN, and SN networks(
). Results revealed significant decreases in PPI-connectivity between the cerebellum, FPN and SMN in remitters only. Baseline values of cerebellar-FPN and cerebellar-SN PPI were Would associated with clinical outcomes(
The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression.
). A study using an Incentive Flanker Task showed reduced sgACC activation to positive and negative feedback 5 days following single-dose ketamine, normalizing towards healthy controls(
). Greater pre-ketamine sgACC hyperactivity to positive feedback also associated with larger post-treatment improvements in anhedonia. Using a modified Monetary Incentive Delay Task, an independent study found increased insula and orbitofrontal activation during the anticipatory phase of reward 24 hours post-ketamine(
) (often hypo-active in MDD(63)) was observed 7 days post-infusion.
Several studies have used variants of the emotion recognition task to probe ketamine’s effects on emotion networks. Using the Emotional Judgment Task, during which participants rate the emotional valence of pictures, decreased amygdala and insula response was observed during negatively-valanced picture ratings 24 hours post-ketamine; insula and dACC response decreased 7 days post-ketamine(
). Additionally, the substantia nigra/ventral tegmentum appeared more active for negative than positive picture ratings pre-treatment, a pattern that reversed 7 days following ketamine(
). Since the amygdala, insula, and dACC are often found overreactive when processing negatively valanced stimuli in MDD(64,65), these findings suggest ketamine may normalize this particular hyperactive signature of MDD.
Using an Attentional Bias Dot Probe task with emotional face stimuli to investigate attentional bias pre-to-post ketamine, MDD participants showed a reversal of PFC and dACC activity for happy and angry trials 1-3 days post-ketamine versus placebo(
) associated with decreased and increased activation for angry and happy trials respectively in the left parahippocampus and amygdala, bilateral ACC, precuneus, and left PFC. Using an implicit and explicit Facial Recognition task, the same group showed reduced activity post‐ketamine in frontal, temporal, and precuneus regions in patients that normalized towards healthy controls(
). Further, greater differences in left temporal and bilateral precuneus activation were observed between explicit and implicit conditions post-ketamine in MDD. Conversely, a separate study found a significant increase in neural response to positively-valanced faces (happy>neutral) in the right caudate that linked with improved depressive symptoms(
). Another study used an emotional N-back task with verbal stimuli found significant associations between lower baseline dorsomedial PFC activity and improvements in cognitive but not depressive symptoms(
). Here, both patients receiving ketamine or electroconvulsive therapy (ECT) showed a decrease in amygdala response while processing positive and negative face stimuli. Changes in inferior parietal activity correlated with overall symptom improvement, and BOLD change in frontal regions correlated with anxiety and anhedonia.
Two studies have correlated change in BOLD signal during ketamine infusion with longer term clinical response. Results showed significant associations with increased BOLD response in the ACC(71) and in the right insula and left post-central gyrus(
Summary: To date, fMRI studies of ketamine response have focused on probing functional systems of reward, emotion and cognitive control. Considering the differences in tasks and neural targets, only partial consensus exists among studies (Figure 3). However, all find that ketamine perturbs components of MDD-relevant neural circuits. Notably, regional changes in brain activation during response inhibition (PFC, parietal and cerebellum regions) associate with improved mood and rumination(
). Tasks assessing emotional response show that changes in neural signal in limbic regions, the caudate, PFC, and parietal cortex relate to improved clinical response(
Figure 3This illustration summarizes results from functional imaging studies using a) response inhibition, b) reward processing, c) emotional judgment, and d) emotional face recognition tasks to investigate neural changes in functional systems associated with ketamine treatment in MDD. Increases and decreases of brain activity observed are displayed by task. A table indicating corresponding references to changes in brain activity can be found in Supplemental Table 2.
). All identified PET publications meeting inclusion criteria studied single-dose ketamine infusion (Table 3).
Table 3PET studies of ketamine treatment in depressed populations.
Study
Demographics
Other Medications
Ketamine Treatment
Data Collection
Depression Assessment
Summary of Findings
Estrelis et al., 2018(115)
N: 14 Unipolar MDD, 13 HC§
NO No psychiatric medication in month prior to study
Single IV infusion initial bolus 0.23mg/kg over 1 min ketamine followed by constant infusion of 0.58mg/kg over 1 hr+
Neuroimaging: •Baseline •During infusion •24 hrs post infusion Clinical Assessments: •30 mins and 24 hrs post infusion
MADRS BDI-II
Tracer: [11C]ABP688 Widespread binding reductions from baseline for observed during ketamine and 24 hours after infusion for both HC and MDD. Not significantly different between diagnoses ↓mGluR5 in hippocampus associated with (-) depression symptoms (MADRS and BDI-II)
Tiger et al., 2020(78)
N: 30 Unipolar TRD*
NO Ongoing medication washed out corresponding to ≥5x half-life of drug
Four serial IV infusion 0.5mg/kg ketamine or placebo (isotonic NaCL) over 40 mins given twice a week over 2 weeks++
Neuroimaging: •Baseline •24-72 hours post-infusion Clinical Assessments: •Baseline •During PET •1, 2, 3, 18, and 24 hrs post infusion
MADRS, MINI, QIDS-SR, PHQ, CGIS, EuroQol5D
Tracer: [11C]AZ10419369 No significant differences in change in BPND over time between ketamine and placebo 16.7% ↑BPND in hippocampus in response to first ketamine infusion Inverse correlation between baseline BPND in VS and ΔMADRS after first treatment in ketamine group Baseline BPND in DBS correlated negatively with ΔMADRS with ketamine ΔBPND with treatment did not correlate with antidepressant effects
Lally et al., 2014(75)
N: 36 Bipolar (I or II) TRD*
YES Continued mood stabilizers, no psychotropic medication or psychotherapy ≥2 weeks prior to treatment
Single IV infusion 0.5mg/kg ketamine or placebo (0.9%saline solution) followed by alternative treatment two weeks later+++
Neuroimaging: •2 hours post infusion Clinical Assessments: •Baseline •40, 80, 120, 230 mins post infusion •1, 2, 3, 7, 10, and 14 days post infusion
MADRS, SHAPS
Tracer: 18-FDG ΔVS rCMRGlu after ketamine significantly related to %ΔSHAPS score 230 mins post infusion ΔMADRS significantly predicted ΔVS rCMRGlu but not SHAPS Whole brain analysis: Association between improved SHAPS and ↑dACC and ↑cerebellum rCMRGlu ΔAnhedonia levels following ketamine not related to general changes in depressive symptoms associated with ↑dACC, pregenual cingulate/callosal region, and right dPFC, fusiform gyrus, claustrum, and putamen metabolism
Lally et al., 2015(74)
N: 20 TRD**
NO Drug free for ≥2 prior to treatment
Single IV infusion ketamine over 40 mins+
Neuroimaging: •Baseline Clinical Assessments: •Baseline •40, 80, 120, 230 mins post-infusion •Daily for subsequent 28 days
MADRS SHAPS
Tracer: 18-FDG No association between rCMRGlu and SHAPS score at baseline Baseline metabolism did not correlate with change in anhedonia Significant negative association between ↑dACC rCMRGlu and (-) anhedonia and after ketamine ↑glucose metabolism in cluster in right hippocampus and entorhinal cortex associated with (-) anhedonia ↓rCMRGlu in right OFC and left inferior frontal gyrus associated with (-) anhedonia
Chen et al., 2018(76)
N: 24 Unipolar TRD***
YES Stable antidepressant treatment ≥2 weeks prior to treatment
Single IV infusion of 0.5mg/kg, 0.2mg/kg or placebo (saline) over 40 mins+++
Neuroimaging: •Baseline Clinical Assessments: •Baseline •40, 80, 120, 240 mins post infusion •1 day post infusion
HDRS-17
Tracer: 18-FDG High dose ketamine treatment showed (+)SUV in supplementary motor area and dorsal anterior cingulate cortex (dACC) than low dose ketamine treatment ΔSUV in dACC negatively associated with HDRS-17 symptoms at day 1 ((+)ΔSUV (-)depressive symptoms)
Ortiz et al. 2015(77)
N: 29 Bipolar TRD*
YES Stable mood stablizer (lithium or valproate) ≥4 weeks at therapeutic levels
Single IV infusion 0.5mg/kg over 40 mins+++
Neuroimaging: •Baseline Clinical Assessments: •Baseline •230 mins post infusion •1, 2, 3, 7, and 10 days post infusion
MADRS
↑Baseline Shank3 levels associated with (+)rMRGlu in hippocampus and amygdala ΔMADRS not significantly correlated with ΔrMRGlu
*≥1 failed treatment **≥2 failed treatment ***≥3 failed treatment TRD: Treatment resistant depression MDD: Major Depressive Disorder HC: Healthy Control §: HC received ketamine +: Open Label ++: Randomized Placebo +++: Randomized Crossover F: Female M: Male IV: Intravenous Hrs: Hours Mins: Minutes MADRS: Montgomery-Åsberg Depression Rating Scale SHAPS: Snaith-Hamilton Pleasure Scale HDRS: Hamilton Depression Rating Scale BDI: Beck Depression Inventory
Most PET studies evaluating ketamine response at least 24 hours post treatment in MDD used [18F]-fluorodeoxyglucose (FDG) PET to measure glucose metabolism, a proxy for estimating glutamatergic neurotransmission(
) in TRD(76). The dACC functions to integrate emotional and cognitive processes and is likewise implicated is rsfMRI and task fMRI ketamine studies (see previous sections). Post-ketamine improvements in anhedonia have also been associated with increased FDG in the hippocampus(
), also regions implicated in fMRI studies. Increased glucose metabolism in the hippocampus and amygdala have been specifically observed in bipolar patients with higher baseline levels of Shank3, a protein involved in glutamatergic neurotransmission, following single ketamine infusion. However, findings did not associate with changes in depressive symptoms >24 hours post-treatment(
). Using a serotonin (5-HT1B) receptor selective radioligand to investigate serotonin1B receptor binding in SSRI-resistant depression, one study found a 17% increase in non-displaceable binding to the 5-HT1B receptor (BPND) in the hippocampus 1-3 days following ketamine infusion(
), however changes in BPND were not correlated with antidepressant effects.
Summary
The PET literature suggest regional changes in glucose metabolism post-ketamine are associated with improvements in overall depressive symptoms (dACC)(
Persistent antidepressant effect of low-dose ketamine and activation in the supplementary motor area and anterior cingulate cortex in treatment-resistant depression: A randomized control study.
Persistent antidepressant effect of low-dose ketamine and activation in the supplementary motor area and anterior cingulate cortex in treatment-resistant depression: A randomized control study.
). These changes in neurofunction and circuitry suggest that ketamine’s therapeutic mechanism may be at least partially attributed to downstream effects on glutamatergic signaling that may be attributable to synaptic and dendritic remodeling reported in animal studies(
). Observed increases in hippocampal BPND(78) suggest changes in 5-HT1B receptor density may contribute to ketamine’s therapeutic effects, as has been observed in conjunction with anti-depressive properties of ketamine in preclinical models(
S-Ketamine Mediates Its Acute and Sustained Antidepressant-Like Activity through a 5-HT1B Receptor Dependent Mechanism in a Genetic Rat Model of Depression.
Arterial spin labeling (perfusion) MRI measures cerebral blood flow (CBF) using magnetically labeled arterial blood without the need for any radioactive ligand. ASL CBF measurements have been validated using O-water PET(81), and since perfusion is normally coupled with metabolism, yield information comparable to FDG-PET(82). Thus, by providing quantitative measures of brain hemodynamics, ASL provides information regarding neurofunction that is complementary to fMRI.
Increased CBF in the thalamus has been observed 24 hours after single IV ketamine (
). Furthermore, change in thalamic perfusion has been associated with greater reductions in depressive symptoms, and patients with lower baseline thalamic perfusion showed larger increases in perfusion after ketamine(
). One study leveraged multiband pseudo-continuous ASL (pCASL) MRI to compare global and regional CBF (rCBF) in TRD patients at baseline and 24 hours after receiving single, and four serial ketamine infusions(
Neural correlates of rapid antidepressant response to ketamine in treatment-resistant unipolar depression: a preliminary positron emission tomography study.
), post-single ketamine rCBF increased in the posterior cingulate and in visual association regions. Post-serial infusion therapy, rCBF decreased in the bilateral hippocampus and right insula, normalizing towards levels of healthy controls. Further, acute changes in CBF in visual areas predicted improvements in anhedonia, anxiety, and overall mood following treatment(
Neural correlates of rapid antidepressant response to ketamine in treatment-resistant unipolar depression: a preliminary positron emission tomography study.
), as well as improved anhedonia and anxiety (Table 4). Continued ketamine treatment includes the subsequent engagement of deeper limbic structures and the insula (affective networks and SN).
Table 4ASL studies of ketamine treatment in depressed populations.
Study
Demographics
Other Medications
Ketamine Treatment
Data Collection
Depression Assessment
Summary of Findings
Sahib et al., 2020(85) NCT02165449
N: 22 Unipolar TRD*, 18 HC
YES Monoaminergic antidepressants allowed if stable ≥6 weeks prior to treatment No Benzodiazapine ≤72 hours prior to treatment
Four serial IV infusions of 0.5mg/kg ketamine over 40min+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post first infusion •24-72 hours post fourth infusion
HDRS SHAPS
(+) mean CBF after first infusion that normalizes/decreases after fourth infusion (+) regional CBF after first infusion in mid and posterior cingulate and proximal association areas in paracentral lobule, cuneus, precuneus, and higher order visual association regions like fusiform (-) baseline CBF in fusiform associated with (+)ΔHDRS after first infusion ΔCBF in cuneus after first infusion positively correlated with change in overall mood, anhedonia, an apathy after serial treatment (-) CBF in bilateral hippocampus and right insula after serial infusion
Gärtner et al., 2022(83)
N: 21 Unipolar MDD
YES No restrictions to permitted medication
Single IV infusion 0.5mg/kg racemic ketamine or 0.25mg/kg S-ketamine over 45min+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post infusion
MADRS HDRS
(+) thalamic perfusion 24 hours after ketamine associated with reduced symptom severity, evident in both sites and when pooled patients with ↓baseline thalamus perfusion show larger ↑perfusion after ketamine ↓thalamic perfusion at baseline associated with more reduced depressive symptoms
Gonzalez et al., 2020(84)
N: 11 Unipolar TRD**
YES Stable antidepressant medication ≥1mo prior to treatment
Single IV infusion 0.5mg/kg ketamine over 40min+
Neuroimaging: •Baseline •1hr, 6hrs, and 24hrs post infusion
MADRS HDRS QIDS-SR
(+) CBF in thalamus (-) CBF lateral occipital cortex NRs showed decrease CBF in ventral basal ganglia after ketamine
*≥1 failed treatment **≥2 failed treatment ***≥3 failed treatment TRD: Treatment resistant depression MDD: Major Depressive Disorder HC: Healthy Control §: HC received ketamine +: Open Label ++: Randomized Placebo +++: Randomized Crossover Δ: Change IV: Intravenous Hrs: Hours
Fewer studies have addressed how subanesthetic ketamine impacts brain structure, likely due to the expectation that structural plasticity occurs slowly, and most ketamine MRI investigations only include short follow-ups (Table 5).
Table 5sMRI studies of ketamine treatment in depressed populations.
Study
Demographics
Other Medications
Ketamine Treatment
Data Collection
Depression Assessment
Summary of Findings
Gallay et al., 2021(99); ACTRN12618001412224
N: 30 MDD with chronic suicidality
YES Concurrent psychiatric medication reported in (
Once weekly dose of oral ketamine for 6-week, starting at 0.5mg/kg and titrated by 0.2-0.7mg/kg based on tolerance with maximum dose of 3.0mg/kg at 6th treatment+
Neuroimaging: •Baseline •End of 5-week treatment
-
VBM: (+) bilateral grey matter in putamen, thalamus, caudate, nucleus accumbens, and periaqueductal grey after ketamine No cortical findings
Herrera-Melendez et al., 2021(101)D NCT02099630, NCT03609190
N: 33 Unipolar TRD** (23 CHB, 10 UZH)
YES No constraints on antidepressant medication
CHB: Single IV of 0.5mg/kg racemic ketamine over 40 mins+ UZH: Single IV of 0.25mg/kg S-ketamine over 40 mins+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post infusion
VBM: ↑GMV of bilateral rostral anterior cingulate at baseline associated with greater change in depressive symptoms
Dai et al., 2020(100)
N: 21 Unipolar MDD (10 with comorbid PTSD), 29 HC§
NO No psychotropic or regular medication in past 2 months or history of psychiatric medication
36 participants received single IV bolus infusion of 0.23mg/kg over 1 minute followed by constant infusion of 0.58mg/kg over 1 hour+ 14 participants received single IV of 0.5mg/kg over 40 mins+
Neuroimaging: •Baseline •24 hours post infusion Clinical Assessments: •Baseline •24 and 48 hours post infusion.
HDRS-24 PTSD Checklist
TBM: (-) left lateral OFC volume in MDD after ketamine (+) left angular gyrus, left inferior parietal gyrus, left middle cingulate and paracingulate gyri, left middle occipital gyrus, left supramarginal gyrus, left precuneus 24 hours after ketamine in full sample including comorbid PTSD (+) right precentral gyrus, right opercular IFG, right rolandic operculum, right insula, and right postcentral gyrus 24 hours after ketamine in unipolar MDD group (-) in midbrain area volume in MDD group 24 hours after ketamine
Zhou et al., 2020(92); ChiCTR-OOC-17012239
N: 44 Unipolar TRD**
YES Stable medication for ≥4weeks prior to treatment
Six serial IV infusions of 0.5mg/kg over 40 mins thrice-weekly for two weeks+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post sixth infusion
MADRS Rs defined as ΔMADRS >50%
FreeSurfer: (+) right hippocampal volume 24 hours after last ketamine treatment compared to baseline No significant associations between hippocampal volume changes and inflammatory cytokine changes
Zhou et al., 2020(93)l; ChiCTR-OOC-17012239
N: 44 Unipolar TRD**, 45 HC
YES Stable medication for ≥4weeks prior to treatment
Six serial IV infusions of 0.5mg/kg over 40 mins thrice-weekly for two weeks+
Neuroimaging & Clinical Assessments: •Baseline •24 hours post sixth infusion
MADRS Rs defined as ΔMADRS >50%
FreeSurfer: Baseline Rs > NRs in left hippocampal subiculum body volume (+) left amygdala and right hippocampus volume after ketamine (+) left amygdala volume in Rs after ketamine (+) left hippocampal CA4 body, left GC-ML-DG body, right CA4 head and right ML head significantly increased after ketamine (+) left CA1 body, CA4 body, GC-ML-DG body, and right GC-ML-DG body, ML head of Rs after treatment (+) left subiculum body of NRs after treatment ↑pretreatment volumes in right thalamus and left subiculum head hippocampal subfield correlated with greater (-) in MADRS scores Δleft amygdala and left ΔCA4 body volume negatively correlated with (-)MADRS scores after treatment
Abdallah et al., 2015(97); NCT00768430
N: 13 Unipolar TRD***
NO Medication free ≥1 week prior to treatment
Single IV infusion of 0.5mg/kg of ketamine or 0.045mg/kg midazolam over 40 mins++
Neuroimaging: •Baseline Clinical Assessments: •Baseline •24 hours post infusion
MADRS
Positive association between ΔMADRS and baseline left hippocampal volume
Abdallah et al., 2017(94) NCT00768430
N: 16 Unipolar TRD***
NO Medication free ≥1 week prior to treatment
Single IV infusion of 0.5mg/kg of ketamine over 40 mins++
Neuroimaging & Clinical Assessments: •Baseline •24 hours post infusion
MADRS R defined as post-treatment MADRS<10
(-) left nucleus accumbens volume following treatment in Rs only (+) left hippocampal volume in Rs (-) bilateral nucleus accumbens volumes after treatment in participants with ↑left hippocampal volume, no change in participants with ↓total or left hippocampal volume
Niciu et al., 2017(98)E NCT00088699
N: 55 Unipolar TRD*
NO Psychotropic medication or ECT free ≥2 weeks prior to infusion
Single IV infusion of 0.5mg/kg ketamine over 40 mins+
Neuroimaging & Clinical Assessments: •Baseline •230 mins, 24 hours and 1 week post infusion
MADRS
No significant associations between baseline hippocampal, thalamic, or amygdalar volumes and antidepressant response
Siegel et al., 2021(30); NCT01179009
N: 23 Unipolar TRD**, 27 HC
YES SSRI and SNRI allowed if constant for ≥6 weeks prior to infusion
Continuous 96h IV infusion of ketamine started at 0.15mg/kg/h at 10am on day 1 and titrated to tolerance twice daily to target rate of 0.6mg/kg/h+
Neuroimaging: •Baseline •2 weeks post infusion Clinical Assessments: •2, 4, 6, and 8 weeks post infusion
MADRS
Smaller Pre-treatment right hippocampal volume associated with better MADRS score
*≥1 failed treatment **≥2 failed treatment ***≥3 failed treatment TRD: Treatment resistant depression MDD: Major Depressive Disorder HC: Healthy Control §: HC received ketamine +: Open Label ++: Randomized Placebo +++: Randomized Crossover IV: Intravenous HDRS: Hamilton Depression Rating Scale MADRS: Montgomery-Åsberg Depression Rating Scale PTSD: Post-traumatic stress disorder R: Responders NRs: Non-responders PRs: Partial responders (+): Increase (-): Decrease
FA: Fractional Anisotropy MD: Mean diffusivity RD: Radial Diffusivity SLF: Superior longitudinal fasciculus ILF: Inferior longitudinal fasciculus UF: Uncinate fasciculus CC: Corpus callosum VBM: Voxel-based morphometry TBM: Tensor-based morphometry GMV: Gray matter volume OFC: Orbitofrontal cortex GC-ML-DG: Granule layers and molecular layers of dentate gyrus CA4: Cornu Ammonis 4 hippocampal subfield ML: Molecular layer of the hippocampus Δ: Change ↑: Higher ↓: Lower
DTwo participant sites including Charité University Hospital Berlin (CHB) and University Hospital Zurich (UZH). CHB used a Siemens Tim Trio scanner, UZH used a Philips Achieva TX scanner.
ETo pool clinical assessments from the two sites HAM-D scores were converted to MADRS scores and calculated as a percentage of change from baseline to follow up
). Initial evidence suggests that single and serial ketamine treatment may also impact hippocampal volume. Increased right gross hippocampal volumes have been observed 24 hours post-serial infusion(
). At the hippocampal subfields level, ketamine-induced increases in the right CA4 head/molecular layer (ML) and left CA4 body, and dentate gyrus/granule cell/molecular layer (GC-ML-DG) are reported; increases in left CA1 body, CA4 body, bilateral GC-ML-DG and right ML head occurred in responders only(
Volumetric changes in subcortical structures following repeated ketamine treatment in patients with major depressive disorder: a longitudinal analysis.
). An independent investigation observed increased left gross hippocampal volumes in ketamine remitters but not in non-remitters after single-dose ketamine(
) following a 96-hour infusion. After standard single infusion treatments, one study reported larger left pretreatment hippocampal volume was associated greater antidepressant response(
The antidepressant efficacy of subanesthetic-dose ketamine does not correlate with baseline subcortical volumes in a replication sample with major depressive disorder.
Volumetric changes in subcortical structures following repeated ketamine treatment in patients with major depressive disorder: a longitudinal analysis.
The antidepressant efficacy of subanesthetic-dose ketamine does not correlate with baseline subcortical volumes in a replication sample with major depressive disorder.
Though evidence is limited, ketamine may also affect the structure of other subcortical structures. Following serial ketamine, increased left amygdalar volume was correlated with antidepressant response(
). However, increase nucleus accumbens, bilateral putamen, thalamus, caudate, and periaqueductal gray matter volume was detected following 6-week oral (racemic) ketamine treatment in voxel-based analyses(
Decreased volumes of the left lateral OFC and an increase in the right precentral gyrus, right opercular IFG, right operculum, right insula, and right postcentral gyrus have been reported 24 hours after ketamine(
Volumetric changes in subcortical structures following repeated ketamine treatment in patients with major depressive disorder: a longitudinal analysis.
), despite mixed laterality of reported effects, and may also relate to antidepressant response. Notably, preclinical models have shown that ketamine induces neurotrophic processes, such as increased expression of brain derived neurotrophic factor (BDNF), and increased hippocampal spine density(
R)-Ketamine Rapidly Ameliorates the Decreased Spine Density in the Medial Prefrontal Cortex and Hippocampus of Susceptible Mice After Chronic Social Defeat Stress.
Volumetric changes in subcortical structures following repeated ketamine treatment in patients with major depressive disorder: a longitudinal analysis.
Volumetric changes in subcortical structures following repeated ketamine treatment in patients with major depressive disorder: a longitudinal analysis.
Volumetric changes in subcortical structures following repeated ketamine treatment in patients with major depressive disorder: a longitudinal analysis.
Volumetric changes in subcortical structures following repeated ketamine treatment in patients with major depressive disorder: a longitudinal analysis.
) volumes might associate with subsequent clinical response.
Diffusion MRI
Diffusion-weighted MRI (dMRI) tracks and quantifies water diffusion in brain tissue to evaluate white matter microstructure and other tissue properties(
). In a preliminary study, patients who responded to single IV ketamine had significantly higher fractional anisotropy (FA) and lower radial diffusivity (RD) in the cingulum and forceps minor at baseline compared to non-responders, suggesting an individual’s predisposition for ketamine response may be influenced by regional white matter microstructure(
). When investigating changes in diffusion metrics four hours after single ketamine infusion, increases in FA in the forceps minor and bilateral unicate fasciculus negatively correlated with symptom improvement(
). Further, greater pre-single infusion FA in the cingulum (hippocampal portion) and left superior longitudinal fasciculus associated with improved depressive symptoms(
Together dMRI studies in MDD, though few, suggest ketamine may have a unique effect on white matter microstructure and connectivity (Table 6). However, higher resolution models of diffusion(
) have not yet been investigated within the context of ketamine treatment that may reveal other measurable changes in regional white matter tissue properties.
Table 6dMRI studies of ketamine treatment in depressed populations.
Study
Demographics
Other Medications
Ketamine Treatment
Data Collection
Depression Assessment
Summary of Findings
Sydnor et al., 2020(105); NCT02544607
N: 13 Unipolar TRD*
YES Stable antidepressants and psychotherapy for ≥28 days prior, ketamine naïve
Single 0.5mg/kg IV infusion over 40 mins+
Neuroimaging: •Baseline •4 hours post infusion Clinical Assessments: •24 hours post infusion
HDRS
↑pre-infusion FA in left CB-hippocampal portion and left SLF associated with improvements in HDRS (+) FA in bilateral ILF, left SLF, and right UF between pre and post infusion ↑FA in CC-forceps minor and bilateral UF negatively correlated with improvement in HDRS
Neuroimaging: •Baseline Clinical Assessments: •Baseline •24 hours post infusion
MADRS NRs defined as ΔMADRS <50%
Rs FA > NRs FA in cingulum and forceps minor (-) in RD in forceps
Only NRs showed significantly ↓FA and ↓MD in forceps minor and ↑RD in cingulum compared to HC
Nugent et al., 2019(106) 2; NCT00088699
N: 30 Unipolar TRD*, 26 HC§
NO Psychotropic medication free for ≥2 weeks
Single 0.5mg/kg IV infusion or saline placebo+++
Neuroimaging: •Baseline •Follow up Clinical Assessments: •Baseline •40, 80, 120, and 230 minutes post infusion •1, 2, 3, 7, 10, and 11 days post infusion
MADRS
(-) FA in tracts connecting right amygdala and sgACC associated with better clinical response (-) FC between left amygdala and sgACC associated with better response to ketamine
*≥1 failed treatment TRD: Treatment resistant depression MDD: Major Depressive Disorder HC: Healthy Control §: HC received ketamine +: Open Label ++: Randomized Placebo +++: Randomized Crossover F: female M: male IV: Intravenous HDRS: Hamilton Depression Rating Scale MADRS: Montgomery-Åsberg Depression Rating Scale Rs: Responders NRs: Non-responders
Magnetic resonance spectroscopy (MRS) may reveal changes in glutamate signaling or other aspects of cellular integrity and metabolism linked with or predictive of ketamine’s antidepressant effects. However, none of the four published studies using single voxel proton (1H) MRS to investigate downstream changes in brain metabolites ≥1-day post-single IV ketamine detected significant changes in brain metabolites or associations with ketamine response in MDD(109–112)(Table 7).
Table 7MRS studies of ketamine treatment in depressed populations.
Study
Demographics
Other Medications
Ketamine Treatment
Data Collection
Depression Assessment
Summary of Findings
Valentine et al., 2011(109)
N: 10 Unipolar MDD
NO Psychotropic medication free ≥2weeks prior to study
Single IV infusion of saline over 40 mins followed by single IV infusion 0.5mg/kg ketamine over 40 mins one week later+++
Neuroimaging: •Baseline •3 hours after infusion start •2 days post infusion Clinical Assessments: •Baseline •60 mins, 180 mins, 24 hours, 48 hours, 72 hours, 5 days, and 7 days post-infusion
HDRS-25 BDI HARS BPRS
No change in AANt content following ketamine No correlation with clinical response No cortical AANT
Milak et al. 2016(110)
N: 11 Unipolar MDD
NO Psychotropic medication free ≥2weeks prior to study
Single intravenous infusion 0.5mg/kg over 40 mins+
Neuroimaging: •Baseline •4 during infusion •1 immediately post infusion Clinical Assessments: •Baseline •230 mins and 24 hours post infusion
HDRS-24
Higher glutamate and GABA compared to baseline but not correlated with clinical response Norketamine levels at 90 mins post infusion correlated with improvement in depressive symptoms
Milak et al. 2020(111)
N: 38 Unipolar MDD
NO No psychotropic medication or medication likely to interact with GABA or glutamate ≥2weeks, no neuroleptics ≥2month, no fluoxetine ≥6 weeks prior to study
Single intravenous infusion of 0.1, 0.2, 0.3, 0.4, or 0.5mg/kg ketamine over 40 min+
Neuroimaging: •Baseline •4 during infusion •1 immediately post infusion Clinical Assessments: •Baseline •24 hours post infusion
HDRS-22
Glutamate correlated with clinical improvement when ketamine blood level not included in model GABA not correlated with clinical improvement
Evans et al., 2018(112); NCT00088699
N: 20 Unipolar TRD*, 17 HC§
NO Medication free ≥2weeks prior to study
Single IV infusion 0.5mg/kg ketamine or saline placebo with alternative treatment 2 weeks later+++
Neuroimaging: •Baseline •24 hours post infusion. Clinical Assessments: •Baseline •Same day as infusion
HDRS SHAPS
Non-significant increase in glutamate and tNAA post-ketamine compared to baseline and placebo
*≥1 failed treatment TRD: Treatment resistant depression MDD: Major Depressive Disorder HC: Healthy Control §: HC received ketamine +: Open Label ++: Randomized Placebo +++: Randomized Crossover F: female M: male IV: Intravenous HDRS: Hamilton Depression Rating Scale MADRS: Montgomery-Åsberg Depression Rating Scale SHAPS: Snaith-Hamilton Pleasure Scale HARS: Hamilton Anxiety Rating Scale BPRS: Brief Psychiatric Rating Scale AANt: Amino acid neurotransmitter tNAA: N-acetyl aspartate summed with N-acetylaspartylglutamate
A growing number of neuroimaging studies have investigated brain-systems level effects of subanesthetic ketamine treatment in MDD. Although results are difficult to reconcile given differing study design and analyses, recurring patterns of neurofunctional plasticity are evident. The fMRI literature consistently suggests ketamine impacts depression-relevant functional brain networks and brain activity, most frequently implicating networks encompassing prefrontal, limbic and striatal regions. Several studies reported increased fronto-striatal connectivity(
Transient effects of multi-infusion ketamine augmentation on treatment-resistant depressive symptoms in patients with treatment-resistant bipolar depression - An open-label three-week pilot study.
Functional connectivity between the amygdala and subgenual cingulate gyrus predicts the antidepressant effects of ketamine in patients with treatment-resistant depression.
), often associated with improved depressive symptoms and normalizing towards values in healthy controls. This suggests part of ketamine’s therapeutic mechanism may be driven by its ability to regain cognitive control over emotional activity. PET and perfusion ASL studies have also identified networks and regions that may serve as biomarkers of ketamine response, including the ACC(74–76), striatal(
Neural correlates of rapid antidepressant response to ketamine in treatment-resistant unipolar depression: a preliminary positron emission tomography study.
). These summarized changes in functional brain circuitry may be driven by rapidly occurring synapto- and dendrogenesis observed in signaling pathways downstream to NMDAR or AMPA mediated actions in preclinical models of ketamine treatment(
Given the small sample sizes, heterogeneity of depression and variability in imaging analysis methods, there is a need for systematic replication of prior results, which should be a goal of future studies. Though participants serve as their own controls in longitudinal studies, differences concerning whether concurrent antidepressant medication is allowed and stable could potentially influence results. Thus, more refined investigation of the effects of simultaneous antidepressant treatment may assist in deciphering underlying response mechanisms to benefit treatment approaches. While considering the limitations of the current literature (Table 8), sample demographics and neuroimaging parameters (Supplemental Tables 3-16), and study design (Table 1, Table 2, Table 3, Table 4, Table 5, Table 6, Table 7) are made available for readers to evaluate the potential impact of each study, contextualize findings, and inform the design of future studies.
Table 8Here, we outline the are several limitations in the field to consider, particularly when trying to summarize findings and for designing future studies of ketamine antidepressant treatment.
Limitations in the Field
Notes
Small Sample Sizes
Majority of current studies consist of small sample sizes, with the largest study included in this review included 61 participants with MDD. Testing the generalizability of these findings in larger samples is crucial improving our understanding of ketamine’s therapeutic mechanisms.
Lack of Consensus regarding TRD Definition
There is a lack of consensus regarding the number of failed treatments that the defining TRD, therefore the severity of treatment resistance within the population of each study is variable.
There were frequently differences in criteria for permitted concurrent antidepressant and antipsychotic medication, with some studies requiring participants to be medication free while others only requiring stable medications. It is possible that interactions between ketamine and concurrent antidepressant medication may determine treatment response(
Although majority of the studies generally administered ketamine intravenously at 0.5mg/kg over 40 minutes, there were varying treatment doses and administrative methods and studies investigating serial ketamine treatment varied greatly in the number of treatments administered, which likely influences treatment response.
Observations Made During Differing Windows Post-Ketamine
There was significant variability regarding when neuroimaging and clinical assessments were collected following ketamine treatment, many of which may miss the window of ketamine’s therapeutic effects.
Inconsistent Network Nomenclature
The still maturing field of network neuroscience lacks consistent nomenclature across studies, and is further complicated by fact that particular brain regions are also shared across specific networks(
These reviewed studies have provided important leads regarding how ketamine modulates structural and functional brain systems to elicit antidepressant effects and have pointed to neurobiological features that appear to influence subsequent ketamine response in MDD. However, lack of replication, and data-driven cross-validation that more directly link macrostructural properties or changes to ketamine’s underlying antidepressant mechanisms are required before this work can meaningfully impact clinical decision making and patient outcomes.
Abnormal Hippocampal Subfields May Be Potential Predictors of Worse Early Response to Antidepressant Treatment in Drug-Naïve Patients With Major Depressive Disorder.
A pilot in vivo proton magnetic resonance spectroscopy study of amino acid neurotransmitter response to ketamine treatment of major depressive disorder.
Assessment of Relationship of Ketamine Dose With Magnetic Resonance Spectroscopy of Glx and GABA Responses in Adults With Major Depression: A Randomized Clinical Trial.
Research reported in this publication was supported by the National Institute of Mental Health Grant No. U01MH110008 and the National Institute of Neurological Disorders And Stroke of the National Institutes of Health under Award Number T32NS048004. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Mental Health or the National Institutes of Health.
A systematic review and meta-analysis of randomized, double-blind, placebo-controlled trials of ketamine in the rapid treatment of major depressive episodes.
Resting-state connectivity studies as a marker of the acute and delayed effects of subanaesthetic ketamine administration in healthy and depressed individuals: A systematic review.
Neural correlates of rapid antidepressant response to ketamine in treatment-resistant unipolar depression: a preliminary positron emission tomography study.
Ketamine plus propofol-electroconvulsive therapy (ECT) transiently improves the antidepressant effects and the associated brain functional alterations in patients with propofol-ECT-resistant depression.
Antidepressant and antisuicidal effects of ketamine on the functional connectivity of prefrontal cortex-related circuits in treatment-resistant depression: A double-blind, placebo-controlled, randomized, longitudinal resting fMRI study.
Functional connectivity between the amygdala and subgenual cingulate gyrus predicts the antidepressant effects of ketamine in patients with treatment-resistant depression.
The 16-Item Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression.
Evaluating global brain connectivity as an imaging marker for depression: influence of preprocessing strategies and placebo-controlled ketamine treatment.
Transient effects of multi-infusion ketamine augmentation on treatment-resistant depressive symptoms in patients with treatment-resistant bipolar depression - An open-label three-week pilot study.
Palmer SM, Crewther SG, Carey LM, START Project Team (2014): A meta-analysis of changes in brain activity in clinical depression. Front Hum Neurosci 8: 1045.
Persistent antidepressant effect of low-dose ketamine and activation in the supplementary motor area and anterior cingulate cortex in treatment-resistant depression: A randomized control study.
S-Ketamine Mediates Its Acute and Sustained Antidepressant-Like Activity through a 5-HT1B Receptor Dependent Mechanism in a Genetic Rat Model of Depression.
Volumetric changes in subcortical structures following repeated ketamine treatment in patients with major depressive disorder: a longitudinal analysis.
Abnormal Hippocampal Subfields May Be Potential Predictors of Worse Early Response to Antidepressant Treatment in Drug-Naïve Patients With Major Depressive Disorder.
The antidepressant efficacy of subanesthetic-dose ketamine does not correlate with baseline subcortical volumes in a replication sample with major depressive disorder.
R)-Ketamine Rapidly Ameliorates the Decreased Spine Density in the Medial Prefrontal Cortex and Hippocampus of Susceptible Mice After Chronic Social Defeat Stress.
A pilot in vivo proton magnetic resonance spectroscopy study of amino acid neurotransmitter response to ketamine treatment of major depressive disorder.
Assessment of Relationship of Ketamine Dose With Magnetic Resonance Spectroscopy of Glx and GABA Responses in Adults With Major Depression: A Randomized Clinical Trial.
00321-4&id=gr1.jpg){kind=link}
00321-4&id=gr2.jpg){kind=link}
00321-4&id=gr3.jpg){kind=link}