Considering the ample evidence of involvement of the glutamate system in the pathophysiology of depression, pre-clinical and clinical studies have been conducted to assess the antidepressant efficacy of glutamate inhibition, and glutamate receptor modulators in particular. This review focuses on the use of glutamate receptor modulators in unipolar depression.
To assess the effects - and review the acceptability - of ketamine and other glutamate receptor modulators in comparison to placebo (or saline placebo), other pharmacologically active agents, or electroconvulsive therapy (ECT) in alleviating the acute symptoms of depression in people with unipolar major depressive disorder.
We searched the Cochrane Depression, Anxiety and Neurosis Review Group's Specialised Register (CCDANCTR, to 9 January 2015). This register includes relevant randomised controlled trials (RCTs) from: the Cochrane Library (all years), MEDLINE (1950 to date), EMBASE (1974 to date), and PsycINFO (1967 to date). We did not apply any restrictions to date, language or publication status.
Double- or single-blind RCTs comparing ketamine, memantine, or other glutamate receptor modulators with placebo (or saline placebo), other active psychotropic drugs, or electroconvulsive therapy (ECT) in adults with unipolar major depression.
Data collection and analysis
Three review authors independently identified studies, assessed trial quality and extracted data. The primary outcomes for this review were response rate and adverse events.
We included 25 studies (1242 participants) on ketamine (9 trials), memantine (3), AZD6765 (3), d-cycloserine (2), Org26576 (2), atomoxetine (1), CP-101,606 (1), MK-0657 (1), N-acetylcysteine (1), riluzole (1) and sarcosine (1). Twenty-one studies were placebo-controlled and the majority were two-arm studies (23 out of 25). Twenty-two studies defined an inclusion criteria specifying the severity of depression; 11 specified at least moderate depression; eight, severe depression; and the remaining three, mild-moderate depression. Nine studies recruited only treatment-resistant patients. We rated the risk of bias as low or unclear for most domains, though lack of detail regarding masking of treatment in the studies reduced our certainty in the effect for all outcomes. We rated three studies as having high risk for selective outcome reporting. Many trials did not provide information on all the prespecified outcomes and we found no data, or very limited data, on very important issues like suicidality, cognition, quality of life, costs to healthcare services and dropouts due to lack of efficacy. Among all glutamate receptor modulators, only ketamine (administered intravenously) proved to be more efficacious than placebo, though the quality of evidence was limited by risk of bias and small sample sizes. There was low quality evidence that treatment with ketamine increased the likelihood of response after 24 hours (odds ratio (OR) 10.77, 95% confidence interval (CI) 2.00 to 58.00; 3 RCTs, 56 participants), 72 hours (OR 12.59, 95% CI 2.38 to 66.73; 3 RCTs, 56 participants), and one week (OR 2.58, 95% CI 1.08 to 6.16; 4 RCTs, 131 participants). The effect of ketamine was even less certain at two weeks, as data were available from only one trial (OR 0.93, 95% CI 0.31 to 2.83; 51 participants, low quality evidence). This was consistent across all efficacy outcomes. Ketamine caused more confusion and emotional blunting compared to placebo. There was insufficient evidence to determine if this increased the likelihood of leaving the study early (OR 1.90, 95% CI 0.43 to 8.47; 5 RCTs, 139 participants, low quality evidence). One RCT with 72 participants reported higher numbers of responders on ketamine than midazolam at 24 hours (OR 0.36, 95% CI 0.14 to 0.58), 72 hours (OR 0.37, 95% CI 0.16 to 0.59), and one week (OR 0.29, 95% CI 0.08 to 0.49). However, midazolam was better tolerated than ketamine in terms of blurred vision, dizziness, general malaise and nausea/vomiting at 24 hours post-infusion. The evidence contributing to these outcomes was of low quality. We found better efficacy of sarcosine over citalopram at four weeks (OR 6.93, 95% CI 1.53 to 31.38; 1 study, 40 participants), but not at two weeks (OR 8.14, 95% CI 0.88 to 75.48); fewer participants in the sarcosine group experienced adverse events (OR 0.04, 95% CI 0.00 to 0.68; P = 0.03, 1 study, 40 participants). This was based on low quality evidence. No significant results were found for the remaining glutamate receptor modulators. In one study with 18 participants, ketamine was more effective than ECT at 24 hours (OR 28.00, 95% CI 2.07 to 379.25) and 72 hours (OR 12.25, 95% CI 1.33 to 113.06), but not at one week (OR 3.35, 95% CI 0.12 to 93.83), or two weeks (OR 3.35, 95% CI 0.12 to 93.83). No differences in terms of adverse events were found between ketamine and ECT, however the only adverse events reported were blood pressure and heart rate. This study was rated as very low quality.
We found limited evidence for ketamine's efficacy over placebo at time points up to one week in terms of the primary outcome, response rate. The effects were less certain at two weeks post-treatment. No significant results were found for the remaining ten glutamate receptor modulators, except for sarcosine being more effective than citalopram at four weeks. In terms of adverse events, the only significant differences in favour of placebo over ketamine were in regards to confusion and emotional blunting. Despite the promising nature of these preliminary results, our confidence in the evidence was limited by risk of bias and the small number of participants. Many trials did not provide information on all the prespecified outcomes and we found no data, or very limited data, on very important issues like suicidality, cognition, quality of life, costs to healthcare services and dropouts due to lack of efficacy. All included studies administered ketamine intravenously, which can pose practical problems in clinical practice. Very few trials were included in the meta-analyses for each comparison; the majority of comparisons contained only one study. Further RCTs (with adequate blinding) are needed to explore different modes of administration of ketamine with longer follow-up, which test the comparative efficacy of ketamine and the efficacy of repeated administrations.