Rapid neuroplasticity changes and response to intravenous ketamine: a randomized controlled trial in treatment-resistant depression

Major depressive disorder (MDD) is a leading cause of disability. Convergent animal and human evidence implicates deficits in neuroplasticity—such as altered synaptic protein expression, impaired BDNF/mTOR signaling, reduced synaptogenesis, and atrophy—in the pathophysiology of depression, affecting corticolimbic circuits and contributing to maladaptive behaviors. Post-mortem human studies show reduced neuroplasticity markers (e.g., BDNF, synapse-related gene expression), and neuroimaging reveals hypofunction and gray matter loss in PFC and hippocampus with decreased functional integration, consistent with rigid cognitive/affective processing. Ketamine has rapid, robust antidepressant effects even in treatment-resistant depression (TRD), but mechanisms remain under investigation. This study tests whether rapid microstructural changes consistent with neuroplasticity, measured via DTI mean diffusivity (MD) within depression-relevant regions, are associated with ketamine’s antidepressant response within 24 hours.

Background literature cited indicates: (1) depression is associated with structural and functional neuroplasticity deficits across corticolimbic circuits; (2) effective antidepressants can reverse such deficits; (3) human post-mortem and neuroimaging studies reveal reduced BDNF, synaptic markers, PFC/hippocampal volume loss, and altered network integration; (4) ketamine produces rapid antidepressant effects in TRD, with prior imaging work showing changes in amygdala and PFC activity and connectivity; (5) DTI mean diffusivity has been used as a putative in vivo marker of rapid microstructural plasticity in gray matter. The study aims to translate animal evidence of ketamine-induced synaptogenic mechanisms to a placebo-controlled human setting by linking rapid MD changes in key regions to clinical improvement.

Design: Randomized, placebo-controlled trial. Participants: 98 adults with unipolar major depression who had failed at least one antidepressant. Randomization: 2:1 to a single intravenous ketamine infusion (0.5 mg/kg) versus vehicle (saline). Imaging: Diffusion tensor imaging (DTI) conducted at baseline pre-infusion and 24 hours post-infusion. Primary imaging metric: Mean diffusivity (MD) in gray matter as a putative marker of microstructural neuroplasticity. Regions of interest (ROIs): left/right BA10, left/right amygdala, left/right hippocampus, and ventral anterior cingulate cortex (vACC). Clinical outcomes: Montgomery-Asberg Depression Rating Scale (MADRS) and Quick Inventory of Depressive Symptoms—Self-Report (QIDS-SR). Additional exploratory outcomes (remittance anxieties, positive affect, dissociative side effects) reported in supplement. Imaging acquisition (Siemens): Oblique-axial diffusion-weighted scans aligned to AC–PC; 72 slices; ascending interleaved; 2-mm isotropic voxels; matrix 104×104; TR = 2443 ms; TE = 88 ms; multiband factor = 4; 30 diffusion directions; b = 1000 s/mm2; 4 b0 images; bandwidth 2004 Hz/pixel; partial Fourier 7/8. Two sets with opposite phase encoding (A→P and P→A) per session; high-resolution structural MPRAGE at baseline (TR = 2400 ms; TE = 2.22 ms; 208 slices; 1 mm isotropic). Processing: FSL v5.0 standard DTI pipeline (corrections for motion, eddy currents, and field inhomogeneity), registration to template, voxel-wise MD maps. For each ROI, Δ-MD computed as MD pre-infusion minus MD at 24 h (positive Δ-MD indicates decreased diffusivity). Anatomical masks derived from MNI atlas; average ROI MD extracted. Extreme Δ-MD values Winsorized before analysis. Statistics: IBM SPSS v28.0.1. Group effects tested with independent-samples t-tests. Multiple linear regressions predicted change in MADRS and QIDS from Δ-MD, group (ketamine vs saline), and their interaction. Sensitivity analyses included covariates (concomitant psychotropic medication burden, level of treatment resistance, and baseline DTI-MD). Sample characteristics: In the DTI subgroup (n = 98; ketamine n = 67; saline n = 31), baseline demographics and clinical variables did not differ by group (p > 0.15). Baseline QIDS mean 15.36 (SD 3.99); baseline MADRS mean 32.81 (SD 5.39); 62.2% assigned female at birth; 80.6% taking psychotropic medications.

- In the full clinical trial sample: ketamine significantly improved depression at 24 h on MADRS (t(150) = −4.55, p < 0.001) and QIDS (t(148) = −2.60, p = 0.001). - In the DTI-imaged subset (n = 98): ketamine significantly improved MADRS (t(96) = −3.01, p = 0.003); QIDS change was not significant (t(95) = −1.10, p = 0.28). - Ketamine showed no significant main effect on raw or Winsorized Δ-MD in any ROI. - BA10: • QIDS: significant Δ-MD × group interaction; decreased MD (greater putative plasticity) predicted greater improvement specifically in ketamine group (β = 0.386, p = 0.035). • MADRS: significant main effect of group (β = 0.314, p = 0.002) and trend for Δ-MD × group interaction (β = 0.308, p = 0.082); greater MADRS improvement associated with MD reduction, especially in ketamine. • Right BA10 associations extended across both infusions (p = 0.007 for association with improvement). - Amygdala: • QIDS: main effect of Δ-MD (β = 0.275, p = 0.007); reduced MD predicted greater improvement irrespective of group. • MADRS: main effects of Δ-MD (β = 0.196, p = 0.046) and group (β = 0.268, p = 0.007); no Δ-MD × group interaction (p > 0.195). - Hippocampus: • MADRS only: significant main effect of group (β = 0.334, p = 0.001) and trend for Δ-MD × group interaction in right hippocampus (β = −0.303, p = 0.060); increased MD predicted greater MADRS improvement in the ketamine group (inverse direction to BA10/amygdala). • No significant effects for QIDS (p > 0.05). - No significant Δ-MD associations with depression change in vACC or right amygdala. - Sensitivity analyses including covariates (medication burden, TRD severity, baseline MD) did not materially change the findings.

The study links rapid microstructural changes, indexed by DTI mean diffusivity, to antidepressant response following ketamine. Decreases in MD (interpreted as increased microstructural plasticity) in left BA10 and left amygdala predicted greater clinical improvement, particularly in the ketamine group, aligning with literature implicating PFC/amygdala circuits in emotional processing biases in depression and their normalization by antidepressant treatments. Right BA10 effects appeared across both ketamine and saline groups, suggesting contributions to nonspecific therapeutic factors present in both arms. In contrast, hippocampal findings were inverse: increased MD predicted improved clinician-rated depression after ketamine. Possible interpretations include that increased MD may reflect processes other than reduced plasticity, such as transient physiological changes (e.g., cerebral blood flow increases) associated with recovery in specific symptom domains, or other tissue property changes; thus, MD increases might not straightforwardly map onto decreased neuroplasticity in hippocampus. No associations were observed in vACC or right amygdala, which may indicate a lesser role of the specific microstructural changes captured by MD in these regions for ketamine’s rapid effects (though measurement limitations cannot be excluded). Clinically, ketamine may induce a temporally-limited window of enhanced plasticity (at least up to 24 h), which could be leveraged through adjunctive interventions (e.g., cognitive training) and supportive environments to prolong or amplify antidepressant benefits.

Ketamine’s rapid antidepressant response is associated with acute microstructural changes consistent with neuroplasticity in select depression-relevant regions. Reductions in DTI mean diffusivity in BA10 and left amygdala relate to greater symptom improvement, while hippocampal MD changes show an opposite pattern for clinician-rated outcomes. These findings support neuroplasticity as a mechanism contributing to ketamine’s effects and motivate development of synergistic interventions and timing strategies to harness post-infusion plasticity. Future work should clarify regional mechanisms, longitudinal trajectories beyond 24 h, and causal pathways linking microstructural change to clinical outcomes.

- DTI-MD is an indirect proxy for neuroplasticity; other factors (e.g., inflammation, cerebral blood flow, tissue composition) can influence MD and confound interpretation. - Only acute (24 h) changes were measured; no longer-term neuroplasticity trajectories were assessed. - Imaging analyses were conducted in a subset of the full clinical trial, potentially underpowering interaction tests and limiting generalizability. - Heterogeneity of psychiatric responses complicates between-group comparisons; treatment protocols were not strictly controlled. - A constrained set of ROIs was selected a priori; multiple comparison adjustments were not applied. - Could not assess baseline group differences in Δ-MD or normalization effects due to design constraints. - Effect sizes were modest (R² ≈ 0.05–0.18). - No direct main effect of ketamine on Δ-MD was observed; mediating factors likely contribute to antidepressant response. - White matter microstructure and tractography were not assessed due to sequence timing and lack of priority hypotheses.