Ruud Schreurs

Predictability in orbital reconstruction. A human cadaver study, part III: Implant-oriented navigation for optimized reconstruction

Navigation-assisted orbital reconstruction remains a challenge, because the surgeon focuses on a two-dimensional multiplanar view in relation to the preoperative planning. This study explored the addition of navigation markers in the implant design for three-dimensional (3D) orientation of the actual implant position relative to the preoperative planning for more fail-safe and consistent results. Pre-injury computed tomography (CT) was performed for 10 orbits in human cadavers, and complex orbital fractures (Class III/IV) were created. The orbits were reconstructed using preformed orbital mesh through a transconjunctival approach under image-guided navigation and navigation by referencing orientating markers in the implant design. Ideal implant positions were planned using preoperative CT scans. Implant placement accuracy was evaluated by comparing the planned and realized implant positions. Significantly better translation (3.53 mm vs. 1.44 mm, p = 0.001) and rotation (pitch: -1.7° vs. -2.2°, P = 0.52; yaw: 10.9° vs. 5.9°, P = 0.02; roll: -2.2° vs. -0.5°, P = 0.16) of the placed implant relative to the planned position were obtained by implant-oriented navigation. Navigation-assisted surgery can be improved by using navigational markers on the orbital implant for orientation, resulting in fail-safe reconstruction of complex orbital defects and consistent implant positioning.

Quantitative Assessment of Orbital Implant Position – A Proof of Concept

Introduction In orbital reconstruction, the optimal location of a predefined implant can be planned preoperatively. Surgical results can be assessed intraoperatively or postoperatively. A novel method for quantifying orbital implant position is introduced. The method measures predictability of implant placement: transformation parameters between planned and resulting implant position are quantified. Methods The method was tested on 3 human specimen heads. Computed Tomography scans were acquired at baseline with intact orbits (t0), after creation of the defect (t1) and postoperatively after reconstruction of the defect using a preformed implant (t2). Prior to reconstruction, the optimal implant position was planned on the t0 and t1 scans. Postoperatively, the planned and realized implant position were compared. The t0 and t2 scans were fused using iPlan software and the resulting implant was segmented in the fused t2 scan. An implant reference frame was created (Orbital Implant Positioning Frame); the planned implant was transformed to the reference position using an Iterative Closest Point approach. The segmentation of the resulting implant was also registered on the reference position, yielding rotational (pitch, yaw, roll) as well as translational parameters of implant position. Results Measurement with the Orbital Implant Positioning Frame proved feasible on all three specimen. The positional outcome provided more thorough and accurate insight in resulting implant position than could be gathered from distance measurements alone. Observer-related errors were abolished from the process, since the method is largely automatic. Conclusion A novel method of quantifying surgical outcome in orbital reconstructive surgery was presented. The presented Orbital Implant Positioning Frame assessed all parameters involved in implant displacement. The method proved to be viable on three human specimen heads. Clinically, the method could provide direct feedback intraoperatively and could improve postoperative evaluation of orbital reconstructive surgery.

Implant-oriented navigation in orbital reconstruction. Part 1 : technique and accuracy study

Intraoperative navigation is frequently used to assess the position of the implant in orbital reconstruction. Interpretation of the feedback from the navigation system to a three-dimensional position of the implant needs to be done by the surgeon, and feedback is only gathered after the implant has been positioned. An implant-oriented navigation approach is proposed, with real-time intuitive feedback during insertion. A technical framework was set up for implant-oriented navigation, with requirements for planning, implant tracking, and feedback. A dedicated navigation instrument was designed and a software tool was developed in order to meet the technical requirements. An accuracy study was performed to investigate the accuracy of the method in comparison to the regular navigation pointer. A proof of concept was provided. The results showed a translation error of 1.12-1.15mm for implant-oriented navigation with regular registration (pointer 0.71-0.98mm) and 0.81mm with accurate registration (pointer 0.54mm). Rotational error was found to be small (<3°). Quantitative and intuitive qualitative feedback could be provided to the surgeon in real-time during insertion of an orbital implant. Following this proof of concept and accuracy study, the implications for the clinical workflow should be evaluated. An implant-oriented approach may form the foundation for augmented reality or robotic-aided implant insertion.