OBJECTIVEAtlantoaxial instability is usually corrected by anterior and/or posterior C1–2 fusion. However, fusion can lead to considerable loss of movement at the C1–2 level, which can adversely impact a patient’s quality of life. In this study, the authors investigated the stability and function of a novel posterior artificial atlanto-odontoid joint (NPAAJ) by using cadaveric cervical spines.METHODSThe Oc–C7 regions from 10 cadaveric spines were used for anteroposterior (AP) translation and range of motion (ROM) tests while intact and after destabilization, NPAAJ implantation, and double-rod fixation.RESULTSThe mean AP C1–2 translational distances in the intact, destabilization, and double-rod groups were 6.53 ± 1.07 mm, 11.54 ± 1.59 mm, and 3.24 ± 0.99 mm, respectively, and the AP translational distance in the NPAAJ group was significantly different from that in the intact group (p < 0.05). The AP translational distance in the NPAAJ group was not significantly different from that in the double-rod group (p = 0.24). The mean flexion, extension, and axial rotation ROM values of the NPAAJ group were 9.87° ± 0.91°, 8.75° ± 0.99°, and 61.93° ± 2.93°, respectively, and these were lower than the corresponding values in the intact group (p < 0.05). The mean lateral bending ROM in the NPAAJ group (9.26° ± 0.86°) was not significantly different from that in the intact group (p = 0.23), and the flexion, extension, and rotation ranges in the NPAAJ group were 79.5%, 85.2%, and 82.3%, respectively, of those in the intact group.CONCLUSIONSUse of NPAAJ for correction of atlantoaxial instability disorders caused by congenital odontoid dysplasia, odontoid fracture nonunion, and C-1 transverse ligament disruption (IA, IB, and IIB) may restore the stability and preserve most of the ROM of C1–2. Additionally, the NPAAJ may prevent soft tissue from embedding within the joint. However, additional studies should be performed before the NPAAJ is used clinically.