Comparative efficacy and safety of different surgical strategies for refractory obsessive-compulsive disorder:...

To conduct a network meta-analysis to compare the efficacy and safety of various surgical strategies for refractory obsessive-compulsive disorder (OCD), including ablative surgery (ABL) and deep brain stimulation (DBS), with the aim to guide clinical treatment.

We searched major electronic databases for different surgical interventions of OCD. The primary outcomes were changes in the Yale-Brown Obsessive Compulsive Scale (Y-BOCS) at 1 year and at the longest follow-up (LFU); the secondary outcomes included responder rates of Y-BOCS (≥35% reduction) and changes in global function, depression, and anxiety; and the safety outcomes included surgery-related adverse events (SRAEs) and serious adverse events (SAEs).

A total of 75 studies involving 1259 patients and 20 surgical strategies were enrolled. Most interventions resulted in significant improvements in Y-BOCS scores, with a reduction of around 10-15 points. Among them, radiofrequency capsulotomy (RF-Cap, mean difference [MD]: 17.251 at 1 year; MD: 17.458 at LFU) and inferior thalamic peduncle DBS (ITP-DBS, MD: 18.126 at 1 year; MD: 20.209 at LFU) were associated with the greatest improvements. Subthalamic nucleus + ventral capsule/ventral striatum DBS (STN + VC/VS-DBS) also exhibited good efficacy at the LFU (MD: 20.780), although data were lacking at 1 year. In terms of safety, ABL was associated with a higher rate of SRAEs than DBS (26 VS. 22%, p = 0.0325), with mechanical-Cap exhibiting the highest SRAE rate (47.5%). However, both DBS and ABL showed good acceptability, with no significant difference in SAEs.

Based on the current analysis, RF-Cap and ITP-DBS were associated with the largest improvements; however, the evidence for ITP-DBS is based on a small sample size, and should therefore be interpreted with caution. More head-to-head studies are needed to directly compare different surgical techniques and identify individual treatment options.