Usefulness of a Three-Dimensional-Printed Model in the Treatment of Irreducible Atlantoaxial Dislocation with Transoral Atlantoaxial Reduction Plate.

Objective To evaluate the usefulness of a 3D-printed model for transoral atlantoaxial reduction plate (TARP) surgery in the treatment of irreducible atlantoaxial dislocation (IAAD). Methods A retrospective review was conducted of 23 patients (13 men, 10 women; mean age 58.17 ± 5.27 years) with IAAD who underwent TARP from January 2015 to July 2017. Patients were divided into a 3D group (12 patients) and a non-3D group (11 patients). A preoperative simulation process was undertaken for the patients in the 3D group, with preselection of the TARP system using a 3D-printed 1:1 scale model, while only imaging data was used for the non-3D group. Complications, clinical outcomes (Japanese Orthopaedic Association [JOA] and visual analogue score [VAS]), and image measurements (atlas-dens interval [ADI], cervicomedullary angle [CMA], and clivus-canal angle [CCA]) were noted preoperatively and at the last follow up. Results A total of 23 patients with a follow-up time of 16.26 ± 4.27 months were included in the present study. The surgery duration, intraoperative blood loss, and fluoroscopy times in the 3D group were found to be shorter than those in non-3D group, with statistical significance. The surgery duration was 3.29 ± 0.45 h in the 3D group and 4.68 ± 0.90 h in the non-3D group, and the estimated intraoperative blood loss was 131.67 ± 43.03 mL in the 3D group and 185.45 ± 42.28 mL in the non-3D group. No patients received blood transfusions. The intraoperative fluoroscopy times were 5.67 ± 0.89 in the 3D group and 7.91 ± 1.45 in the non-3D group. Preoperatively and at last follow up, JOA and VAS scores and ADI, CCA, and CMA were improved significantly within the two groups. However, no statistical difference was observed between the two groups. However, surgical site infection occurred in 1 patient in the 3D group, who underwent an emergency revision operation of the removal of TARP device and posterior occipitocervical fixation; the patient recovered 2 weeks after the surgery. In 2 patients in the traditional group, a mistake occurred in the placement of screws, with no neurological symptoms related to the misplacement. Conclusion Preoperative surgical simulation using a 3D-printed real-size model is an intuitive and effective aid for TARP surgery for treating IAAD. The 3D-printed biomodel precisely replicated patient-specific anatomy for use in complicated craniovertebral junction surgery. The information was more useful than that available with 3D reconstructed images.