A comment on Fond and colleagues' systematic review and meta-analysis of ketamine in the treatment of depressive disorders (Psychopharmacology 2014; Jul 20 [Epub ahead of print]).

Abstract

Dear Dr. Price, We read with great interest the systematic review and metaanalysis of ketamine in the treatment of depressive disorders recently published in Psychopharmacology by Fond and colleagues (2014). It is a welcome addition to a rapidly growing area of research that is deserving of careful and rigorous synthesis in order to guide evidence-based practice. Having recently completed our own systematic review and metaanalysis of randomized controlled trials (RCT) of ketamine in the treatment of depressive episodes (McGirr et al. 2014), we were curious about some striking differences between our results and, by consequence, our interpretation of the principal findings, and those of Fond and colleagues (2014). An important difference to have emerged from our parallel systematic reviews and meta-analyses is conflicting results as to the relative efficacy of ketamine in unipolar as compared to bipolar major depression. Specifically, our meta-analysis reported superior efficacy of ketamine in unipolar as compared to bipolar major depression, representing a standardized mean difference (SMD) of 1.07 vs. 0.68, respectively. We also presented rates of clinical response and remission that paralleled this difference in favor of unipolar depression. Meanwhile, the meta-analysis by Fond and colleagues (2014) reported ketamine’s superiority in bipolar (SMD= 1.34) rather than unipolar major depression (SMD=0.91). A common explanation for such differences might have been the use of alternative statistical methods. Yet, our statistical methodologies were very similar overall, even employing the same statistical software package. Indeed, both metaanalyses employed random effect models, though the correlations between preand post-ketamine depression scores were not specified by Fond and colleagues nor was the final weighting detailed. Another possibility relates to differences in the clinical trials that were included. Our investigation was limited to randomized double-blind placebo-controlled trials using a clinician-administered depressive symptom scale as the main outcome measure, and therefore, we did not include the trials by Kudoh et al. (2002), Ghasemi et al. (2013), or Valentine et al. (2011), which were included by Fond and colleagues (2014). Further, we included data from Lapidus and colleagues’ (2014) recent intranasal ketamine RCT, which was not included by Fond and colleagues (2014). These four RCTs that differentiate our final data, however, were all limited to patients with unipolar major depression and, therefore, cannot explain the marked difference in effect size in the treatment of bipolar major depression. The difference in efficacy identified by our two studies with respect to bipolar major depression is quite striking, especially considering that it is derived from only two RCTs (Diazgranados et al. 2010; Zarate et al. 2012). After careful reexamination of our data and that presented by Fond and colleagues (2014), we believe that this disparity likely results from an erroneous recording of the data from Diazgranados and colleagues’ (2010) crossover trial. From this trial among patients with bipolar depression, Fond and colleagues reported a SMD of approximately 1.75 favoring ketamine, while the published RCT clearly stated a Cohen’s d of 0.67 (95 % CI 0.42–0.91) (Diazgranados et al. 2010). We also wish to highlight another inconsistency in the report by Fond and colleagues. Specifically, the inclusion criteria for their quantitative meta-analysis specified that only A. McGirr (*) Department of Psychiatry, University of British Columbia, 11th Floor, 2775 Laurel Street, Vancouver, BC V5Z 1M9, Canada e-mail: alexander.mcgirr@alumni.ubc.ca