Effects of Serial Ketamine Infusions on Corticolimbic Functional Connectivity in Major Depression

Abstract

Background: Ketamine is a highly effective antidepressant for patients with treatment-resistant major depressive disorder (MDD). Resting-state functional magnetic resonance imaging studies show disruptions of functional connectivity (FC) between limbic regions and resting-state networks (RSNs) in MDD, including the default mode network, central executive network (CEN), and salience network (SN). Here, we investigated whether serial ketamine treatments change FC between limbic structures and RSNs.

Methods: Patients with MDD (n = 44) were scanned at baseline (time 1 [T1]) and 24 hours after the first (T2) and fourth (T3) infusions of ketamine. Healthy control subjects (n = 50) were scanned at baseline, with a subgroup (n = 17) being rescanned at 2 weeks. Limbic regions included the amygdala and hippocampus, and RSNs included the default mode network, CEN, and SN.

Results: Ketamine increased right amygdala FC to the right CEN (p = .05), decreased amygdala FC to the left CEN (p = .005) at T2 versus T1 (p = .015), which then increased at T3 versus T2 (p = .002), and decreased left amygdala FC to the SN (p = .016). Decreased left amygdala to SN FC at T2 predicted improvements in anxiety at T3 (p = .006). Ketamine increased right hippocampus FC to the left CEN (p = .001), and this change at T2 predicted decreased anhedonia at T3 (p = .005).

Conclusions: Ketamine modulates FC between limbic regions and RSNs implicated in MDD. Increases in FC between limbic regions and the CEN suggest that ketamine may be involved in restoring top-down control of emotion processing. FC decreases between the left amygdala and SN suggest that ketamine may ameliorate MDD-related dysconnectivity in these circuits. Early FC changes between limbic regions and RSNs may be predictive of clinical improvements.

Trial registration: ClinicalTrials.gov NCT02165449.

Keywords: Antidepressant treatment; Functional connectivity; Ketamine; Major depression; Mood disorders; Personalized medicine.

Conflict of interest statement

Disclosures

The authors report no biomedical financial interests or potential conflicts of interest.

Copyright © 2020 Society of Biological Psychiatry. Published by Elsevier Inc. All rights reserved.

Figures

Figure 1.. Study design.

Ketamine group was administered clinical scales and received an MRI at three time points (T1: pre-treatment/baseline, T2: 24 hours post-first infusion of ketamine, and T3: 24–72 hours post fourth infusion of ketamine). The study length varied from 2–2.5 weeks depending on which day the first infusion started on. Fifty non-depressed healthy controls (HC) received an MRI at baseline (T1) and a subset of the fifty (n=17) HCs received a repeat assessment 2 weeks after T1.

Figure 2.. Amygdala connectivity to resting-state networks (RSNs).

A) Functional connectivity between the bilateral amygdala and RSNs (default mode network (DMN), right central executive network (RCEN), left central executive network (LCEN), and salience network (SN)). Connectivity between the amygdala and the RCEN and SN showed a time by hemisphere effect, therefore, the right and left amygdala were looked at separately in B and C. B) Functional connectivity between the left and right amygdala and RCEN. C) Functional connectivity between the left and right amygdala and SN. *p

Figure 3.. Hippocampal connectivity to resting-state networks (RSNs).

A) Functional connectivity between the bilateral hippocampus and RSNs (default mode network (DMN), right central executive network (RCEN), left central executive network (LCEN), and salience network (SN)). Connectivity between the hippocampus and the LCEN showed a time by hemisphere effect therefore the right and left hippocampus were looked at separately in B. B) Functional connectivity between the left and right hippocampus and LCEN. *p

Figure 4.. Correlations between measures of clinical improvement and reductions in FC between limbic regions and resting-state networks.

(A) Change in FC between the left amygdala (AmygL) and salience network (SN) after a single infusion of ketamine correlated with change in BIS at the end of treatment. (B) Change in FC between the AmygL and SN at the end of treatment correlated with change in BIS at the end of treatment. (C) Change in FC between the hippocampus (Hip) and right central executive network (RCEN) after a single infusion of ketamine correlated with change in SHAPs at the end of treatment. BIS, behavioral inhibition system scale; SHAPs, Snaith-Hamilton Pleasure Scale; T1, time 1 (baseline); T2, time 2 (24 hours after first infusion of ketamine); T3, time 3 (24 hours after fourth infusion of ketamine).