A Systematic Review and Meta-Analysis of Robot-Assisted Deep Brain Stimulation: Comparative Insights With Conventional Techniques Publisher Pubmed



Dashtkoohi M¹ ; Ashtari S¹ ; Delbari P¹ ; Mehrabi MM¹ ; Karimi A³ ; Kakoienejad MM² ; Khaboushan AS¹ ; Ohadi MAD¹
Source: Neurosurgical Review Published:2025

Abstract

Introduction: Recent robotic-assisted surgical systems have shown promising efficiency and accuracy in brain surgeries. However, their application in deep brain stimulation (DBS) surgery remains limited. Methods: Studies from the Embase, Scopus, and Pubmed databases were included using a modified search string, following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines. We extracted technical aspects of robotic-assisted DBS (RA-DBS) surgery, patient characteristics, accuracy, safety, and overall clinical results. We applied a random effects model for analysis. Heterogeneity was evaluated using Cochran's Q test and the I2 statistic. Additionally, subgroup analysis was conducted for the type of anesthesia, and frame-based versus frameless procedures and using microelectrode recordings (MER). Quality appraisal was conducted using the Newcastle–Ottawa Scale. Results: The pooled mean radial error (RE) and Euclidean distance (3D-EE) in robotic surgeries were 1.04 mm (95% CI 0.93 to 1.15 mm) and 1.19 mm (95% CI: 0.94 – 1.45 mm). RA-DBS outcomes regarding RE were comparable to conventional stereotactic DBS (C-DBS) surgery (SMD: -0.56, 95% CI: -1.35 – 0.22, P-value: 0.16). Also, the difference in the rates of reported complications was insignificant (OR: 0.26, 95% CI: -0.64 – 1.15, P-value: 0.58). None of the frame-based or frameless (P-value: 0.36), the use of MER (P-value: 0.11), or the type of anesthesia (P-value: 0.27) showed significant differences. However, there was an insignificant lower RE trend in the studies that employed frame-based techniques, used general anesthesia, and did not utilize MER. Conclusion: Our study shows an acceptable level of error associated with RA-DBS. We found that the accuracy and complication rates were comparable to C-DBS. Due to substantial heterogeneity in the pooled mean RE and 3D-EE, more studies with higher sample sizes are required to evaluate RA-DBS. © The Author(s), under exclusive licence to Springer-Verlag GmbH Germany, part of Springer Nature 2025.