Comparative Analysis of Efficacy and Safety of Frame-Based, Frameless, and Robot-Assisted Stereotactic Brain Biopsies: A Systematic Review and Meta-Analysis.

Abstract

Background and objectives: For 50 years, frame-based stereotactic brain biopsy has been the "gold standard" for its high diagnostic yield and safety, especially for complex or deep-seated lesions. Over the past decade, frameless and robotic alternatives have emerged. This report evaluates and compares the outcomes, diagnostic yield, and safety of these methods.

Methods: Major databases were screened for studies reporting data on diagnostic yield, postoperative hemorrhage, neurological deficits, and mortality after frame-based, robot-assisted or neuronavigation-assisted frameless biopsies. Meta-analysis with random-effect modeling was performed to compare diagnostic yield, operative duration, length of stay, complications, and mortality.

Results: A total of 92 studies were included with 9801 patients in the frame-based group, 2665 in the robot-assisted group, and 1862 in the frameless group. Pooled diagnostic yield rates were 97% (96%-98%, I 2 = 49%) in robot-assisted, 95% (94%-96%, I 2 = 74%) in frame-based, and 94% (91%-96%, I 2 = 55%) frameless groups with a statistically significant difference ( P < .01, I 2 = 71%). The mean total operative duration including anesthesia, system setup, patient registration, trajectory planning, and skin incision to closure was significantly shorter in the robot-assisted group (76.6 vs 132.7 vs 97.3 minutes, P < .01). The duration from skin incision to closure was comparable between the groups (robot-assisted: 37.8 mins, frame-based: 42.6 minutes, frameless: 58.2 minutes; P = .23). Pooled rates of symptomatic hemorrhage (0.005% vs 0.009% vs 0.007, P = .71, I 2 = 34%), asymptomatic hemorrhage (4% vs 3% vs 3%, P = .64, I 2 = 93%), transient neurological deficit (3% vs 2% vs 2%, P = .5, I 2 = 72%), permanent neurological deficit (0.001% vs 0.001% vs 0.0002, P = .78, I 2 = 47%), and mortality (0% vs 0.001% vs 0.006%, P < .01, I 2 = 10%) were similar between groups. Deaths were mainly due to postprocedural hemorrhage (robotic: 46%, frame-based: 48%, frameless: 72%).

Conclusion: Robot-assisted biopsy is not inferior in diagnostic yield and safety to the gold standard frame-based and neuronavigation-assisted frameless biopsy methods.