The effects of deep-brain non-stimulation in severe obsessive-compulsive disorder: an individual patient data meta-analysis

Non-intervention-related effects have long been recognized in an array of medical interventions, to which surgical procedures like deep-brain stimulation are no exception. While the existence of placebo and micro-lesion effects has been convincingly demonstrated in DBS for major depression and Parkinson's disease, systematic investigations for obsessive-compulsive disorder (OCD) are currently lacking. We therefore undertook an individual patient data meta-analysis with the aim of quantifying the effect of DBS for severe, treatment-resistant OCD that is not due to the electrical stimulation of brain tissue. The MEDLINE/PubMed database was searched for double-blind, sham-controlled randomized clinical trials published in English between 1998 and 2018. Individual patient data was obtained from the original authors and combined in a meta-analysis. We assessed differences from baseline in obsessive-compulsive symptoms following sham treatment, as measured by the Yale-Brown Obsessive-Compulsive Scale (Y-BOCS). Four studies met the inclusion criteria, randomizing 49 patients to two periods of active or sham stimulation. To preclude confounding by period effects, our estimate was based only on data from those patients who underwent sham stimulation first (n = 24). We found that sham stimulation induced a significant change in the Y-BOCS score (t = -3.15, P <0.005), lowering it by 4.9 +/- 1.6 points [95% CI = (-8.0, -1.8)]. We conclude that non-stimulation-related effects of DBS exist also in OCD. The identification of the factors determining the magnitude and occurrence of these effects will help to design strategies that will ultimately lead to a betterment of future randomized clinical trials.