Randomized Trial of D-Cycloserine Enhancement of Cognitive-Behavioral Therapy for Panic Disorder

Abstract

Background Initial studies have provided a mixed perspective of the efficacy of d-cycloserine (DCS) for augmenting the efficacy of exposure-based cognitive behavioral therapy (CBT) for panic disorder. In this multicenter trial, we examine the magnitude of DCS augmentation effects for an ultra-brief program of CBT. Methods We conducted a double-blind, controlled trial at three treatment sites, randomizing 180 adults with a primary diagnosis of panic disorder to 5 sessions of treatment, with study pill (50 mg DCS or matching placebo) administered 1 hour prior to the final 3 sessions. Two booster sessions were subsequently provided, and outcome was assessed at post-treatment and 1-month, 2-month, and 6-month follow-up assessments. The primary outcome was the degree of reduction in the Panic Disorder Severity Scale. Additional analyses examined the role of severity and current antidepressant or benzodiazepine use as moderators of DCS augmentation effects. Results DCS augmentation resulted in significant benefit only early in the trial, with no beneficial effects of DCS augmentation evident at follow-up evaluations. We did not find that baseline severity or antidepressant or benzodiazepine use moderated DCS efficacy, but benzodiazepine use was associated with lower efficacy of CBT regardless of augmentation condition. Conclusions Consistent with other recent multicenter trials, the benefit of DCS was less than indicated by pilot study and reflected an acceleration of treatment response evident at treatment endpoint, but no advantage in response over follow-up evaluation. Our results did not support severity or concomitant medication moderators observed in previous trials of DCS augmentation. Clinical Trials Registry Name Exposure, D-Cycloserine Enhancement, and Genetic Modulators in Panic Disorder (DCSPanic) Clinical Trials URL https://clinicaltrials.gov/ct2/show/NCT00790868 Clinical Trials Registry Number NCT00790868

The Bergen 4-Day Treatment for Panic Disorder: A Pilot Study

“…In order to obtain a perspective on the outcome of this pilot study we searched the PsycINFO database for CBT studies of PD with or without agoraphobia, which had used PDSS as the primary outcome measure. We found nine such studies ( Barlow et al, 2000 ; Nakano et al, 2008 ; Bergström et al, 2010 ; Nordin et al, 2010 ; Wims et al, 2010 ; Nations et al, 2012 ; Beutel et al, 2013 ; Meyerbroeker et al, 2013 ; Otto et al, 2016 ) with 517 patients in total. There were six on standard face-to-face CBT with weekly sessions and three on Internet-based CBT with weekly modules.…”

D-cycloserine augmentation of behavior therapy for anxiety and obsessive-compulsive disorders: A meta-analysis

“…In October 2016, the study basis was updated resulting in three new studies [40,44,45]. Journal articles were thoroughly checked in order to determine their value for the present meta-analysis.…”

“…First, there are three new high quality studies [40,44,45] that were not included in the analysis of Mataix-Cols et al [43]. Second, quite a few primary studies used multiple outcomes to assess symptom severity, but meta-analyses appear to have only used one outcome per study.…”

“…Later findings, however, found no clear advantage of DCS at the end of treatment (i.e., 8 sessions or higher) but a faster treatment response earlier in treatment in psychological disorders such as social anxiety disorder [ 36 ], agoraphobia and panic disorder [ 23 ], or OCD [ 29 ]. Nonetheless, a faster symptom reduction in the beginning of treatment can have additional benefits, e.g., with respect to limited CBT therapist time and lower treatment costs [ 40 ].…”

“…Even though the most recent meta-analysis [ 43 ] was published not long ago, we believe that the current meta-analysis is important for the following two reasons. First, there are three new high quality studies [ 40 , 44 , 45 ] that were not included in the analysis of Mataix-Cols et al [ 43 ]. Second, quite a few primary studies used multiple outcomes to assess symptom severity, but meta-analyses appear to have only used one outcome per study.…”