Our patient denied experiencing trauma or participating in sports activity that might have caused notable axial impact loading. We believe that some degree of repetitive shear force is likely to be applied on the ceramic ball of the artificial disc even without trauma, since there is normal anteroposterior motion coupled with flexion and extension. The fact that the part of the ceramic insert within the cranial metal plate remained intact while the convex-ball portion was fractured supports the hypothesis that shear force was the cause of fracture. Had it been axial impact loading, one might expect that both parts of the cranial ceramic insert would have been fractured. Since this semiconstrained implant permits only spherical motion and does not allow anteroposterior motion, shear forces to the convex portion of the ceramic implant could have been present, predisposing the ceramic implant to fracture, especially since ceramics are known to be brittle with shear forces1.
Surprisingly, fracture of the ceramic component was not diagnosed prior to the second surgical procedure. Prior to that surgery, we assumed that the disc height elevation along with partial decompression achieved during the index operation alleviated the patient's symptoms transiently but that this effect did not last long as the foraminal height decreased with slow settling of the prosthesis into the end plates. Since a wide decompression is needed when disc arthroplasty is performed and the decompression in our case appeared inadequate, the patient could have had recurrence of symptoms even without device failure, as inadequate decompression is reported to be the leading cause of arthroplasty failure9. Nevertheless, we believe that ceramic fracture with consequent disc and foraminal height reduction led to or contributed to the recurrence of the symptoms. The effect of the fractured ceramic pieces also could have contributed, although the pieces were surrounded by scar tissue and did not appear to cause direct neural injury or compression.
An important consideration in ceramic fracture following cervical disc arthroplasty is whether the implant fracture can be diagnosed without surgical exploration. Since our patient did not have a history of abnormal impact loading, did not feel a crunching, cracking, or squeaking sensation or sound, and had no acute onset of symptoms, there were no clinical reasons to suspect fracture of the prosthesis. Most alarmingly, the imaging studies yielded no clues to suggest that the ceramic component had fractured. Even in retrospect, with full knowledge of the events, the ceramic fracture was not seen on the imaging studies. Evaluation of the prosthesis on the CT scan (Fig. 2) and MRI was limited by the metal artifact. Even though the distance between the metal plates at C6-C7 on the lateral radiograph may appear slightly decreased, the decrease is minimal and was considered an indeterminate finding (Fig. 1), possibly because the fragments were contained in the concavity of the lower liner, preventing total collapse.
Although this report is limited by the lack of explant analysis and histological evaluation, several lessons can be learned from this case. First and foremost, spine surgeons need to be aware that ceramic components can fracture following cervical disc arthroplasty. These fractures can be difficult to diagnose since patients may not have symptoms suggestive of a fracture and a fracture may not be appreciated or visible on imaging studies. A ceramic fracture should, therefore, be suspected when a patient presents with unexplained neck pain and/or radiating pain after cervical disc arthroplasty with a ceramic prosthesis. Finally, we believe that when a fractured component is found intraoperatively, the prosthesis should be analyzed to determine the mode of failure and histological analysis should be performed.
Note: The authors thank Yoon Ju Kwon and Seung Min You, the spine research coordinators at our institution, for their contributions to the project.