Comparison of 3D C-arm-based registration to conventional pair-point registration regarding navigation accuracy in ENT surgery.

Objective Navigation surgery on the skull base requires high navigation accuracy. The registration process is related to the main loss in accuracy. This study compared titanium screw registration with an inbuilt registration process of a 3-dimensional (3D) C-arm. Study Design Experimental phantom study. Setting Operating room. Subjects and Methods Four skull models were fabricated with a 3D printer based on the patient's computed tomography (CT) data sets and fitted with an individually customized silicone skin. A 3D-isocentric C-arm fluoroscopic image intensifier system combined with a flat panel detector performed scans of petrous bones (PB) and paranasal sinuses (PS). The navigation accuracy of pair-point registration (PPR) with titanium screws was compared with C-arm-based registration. Results Overall navigation accuracy was 1.53 0.51 mm after PPR and 1.26 +/- 0.12 mm after C-arm registration (P = .0259). PPR showed the best accuracy results on PS (1.28 +/- 0.69 mm), followed by right PB (1.43 +/- 0.52 mm) and left PB (1.74 +/- 0.69 mm). A significant difference was seen only between PS and left PB (P = .0206). In contrast, C-arm registration revealed significantly lower target registration errors (TREs) on PB (0.99 +/- 0.23 mm right PB, P < .0001; 1.2 +/- 0.35 mm left PB, P = .0412) compared with PS. When comparing both registration modalities, C-arm registration was significantly superior on PB. With respect to specific anatomic locations, C-arm-based registration showed significantly lower TREs on the frontal and lateral skull base than PPR. Conclusion C-arm-based navigation shows higher navigation accuracy on the skull base compared with PPR. As the 3D C-arm allows real-time imaging and real-time navigation, it will be a helpful tool for skull base surgeons.