Robot-Assisted Placement of Convection-Enhanced Delivery Infusion Cannulas

BACKGROUND AND OBJECTIVES: Indications for convective delivery of putative therapeutics are rapidly expanding. Cerebellar targets, such as the dentate nucleus, will be crucial to target in certain conditions. Current skull-mounted, magnetic resonance (MR) imaging-compatible platforms preclude infratentorial approach to cerebellar targets. We assessed the feasibility, effectiveness, and accuracy of computed tomography (CT) and MR imaging guided, robot-assisted placement of convection enhanced delivery (CED) cannulas. METHODS: Cadaveric (n = 5) specimens underwent preoperative CT and MR-imaging for trajectory and target (bilateral dentate nuclei) planning followed by robot-assisted placement of suboccipital cannulas (n = 10). Placement accuracy was assessed using CT and MR-imaging. RESULTS: All cadaveric (3 female, 2 male) specimens underwent successful suboccipital placement of bilateral CED cannulas to the dentate nuclei along defined trajectories. Mean incision to cannula placement time was 22 ± 0.6 minutes (range, 13-35 minutes). Medial-lateral error was 1.2 ± 1.1 mm (range, 0.1-3.6 mm), rostral-caudal error was 1.0 ± 0.6 mm (range, 0.1-2.1 mm), and depth error was 0.7 ± 0.4 mm (range, 0.2-1.5 mm) on CT. Mean medial-lateral error was 1.4 ± 0.9 mm (range, 0.1-2.8 mm), rostral-caudal error was 0.9 ± 0.5 mm (range, 0.3-1.8 mm), and depth error was 0.9 ± 0.6 mm (range, 0.1-2.0 mm) on MR imaging. CT accuracy errors were analogous to MR imaging errors (t-test, P ≥ .3). CONCLUSION: Image-guided, robot-assisted placement of CED cannulas is feasible, accurate, and efficient. This paradigm can provide advantages in placing convective infusion cannulas compared with existing methods.