Neuroimaging markers of the delayed effects of ketamine

Student thesis: Doctoral ThesisDoctor of Philosophy


Major Depressive Disorder is a chronic mood disorder, affecting approximately 320 million people worldwide. Rumination associated with biased AM recall and anhedonia linked to altered reward processing, are two of the most debilitating symptoms of depression that could persist in remission and are not successfully targeted by commonly prescribed antidepressants. Ketamine, a glutamate receptor antagonist, has emerged in the last two decades as a potent antidepressant with robust effects. The drug’s fast acting (2h post drug administration) and relatively long-lived antidepressant action (24h-48h after a single ketamine infusion) also targets anhedonia and rumination. (Lehman et.,al 2016, Lally et al., 2014) However, the mechanism by which the drug exerts its antidepressant effects in the brain are not well understood.

The aim of the present study was to identify brain areas that present with significantly altered activations 2h after the ketamine administration, when the antidepressant effects of the drug become detectable, and are important for reward processing and emotionally valenced AM (Autobiographical Memory) recall. For that purpose, we recruited remitted depressed volunteers who were scanned while performing an AM task (VAMP) and a reward processing task (MID). We hypothesized that ketamine, during the MID task would significantly increase striatal activations while during the VAMP task, the drug would modulate activations in the PCC, the sgACC and the amygdala, compared to placebo. These brain areas could serve as neuroimaging markers that could be linked to ketamine’s antidepressant action.

For this study we recruited 37 remitted-depressed, drug-naïve, male and female volunteers who took part in a double-bind, placebo-controlled, cross-over design. Ketamine (0.5mg/kg) and placebo ware administered intravenously over a 45min continuous infusion. Two hours after the drug administration, participants were scanned while performing the MID and the VAMP task. The MID is a reaction time task during which participants need to perform fast button presses, in response to a target stimulus, to win different monetary rewards (Knutson et al., 2001). A cue that is presented before the target signifies the value of the reward that they could win in each trail. The reward anticipation and feedback phases of this task were analysed at a whole brain level and activations were also examined at predefined ROIs associated with reward processing, namely the striatal regions, the VTA, the amygdala and the insula.

The VAMP task is a personalised AM task that was developed specifically for this study. During the task statements from positive, negative and neutral events are presented to participants who are asked to recall those events and rate their emotional valence at the time of the scan. The fMRI data from this task were analysed at a whole brain level and a task connectivity analysis was performed for predefined ROIs that are associated with emotional memory recall and present with altered activations in depression, namely the amygdala, the PCC and the sg ACC.

Scales and questionnaires that measure rumination, anhedonia, subjective well-being and the dissociative/psychotomimetic effects of ketamine were administered at appropriate time points during the study.

Ketamine 2h after its administration did not produce any significant changes to the anhedonia and the subjective well-being of our participants. Whole brain analysis of the MID task did not reveal any significant changes during the anticipation and feedback phase of the task, 2h post ketamine, compared to placebo. The ROI analysis of the task revealed significant increases in the activation of the VS (Ventral Striatum) and the caudate. These increases were identified during the feedback phase of successful and unsuccessful MID trials that were associated with low monetary rewards. Also activity in the VTA (Ventral Tegmental Area) increased during the feedback phase of unsuccessful trials, irrespective of the reward magnitude.

Whole brain analysis of the VAMP placebo data revealed that the precuneus, the middle frontal gyrus and the thalamus presented with significantly altered activation during active AM recall. Ketamine, 2h after its administration and at the whole brain level, did not produce any significant changes to the task, compared to placebo. When the task connectivity was examined between ketamine and placebo, ketamine decreased connectivity between the amygdala and the visual cortex for positive and negative memory recall, compared to neutral. Decreased connectivity between the amygdala and the hippocampus was also identified for positive memory recall compared to neutral, when ketamine was compared to placebo.
Date of Award1 Apr 2020
Original languageEnglish
Awarding Institution
  • King's College London
SupervisorMitul Mehta (Supervisor)

Cite this