The predisposing factors for the heterotopic ossification after cervical artificial disc replacement

Background context: Heterotopic ossification (HO) is defined as a formation of bone outside the skeletal system. The reported HO occurrence rate in cervical artificial disc replacement (ADR) is unexpectedly high and is known to vary. However, the predisposing factors for HO in cervical ADR have not yet been elucidated. Purpose: Investigation of the predisposing factors of HO in cervical arthroplasty and the relationship between degeneration of the cervical spine and HO occurrence. Study design: Retrospective study to discover predisposing factors of HO in cervical arthroplasty. Patient sample: A total of 170 patients who underwent cervical ADR were enrolled including full follow-up clinical and radiologic data. Outcome measures: Radiologic outcomes were assessed by identification of HOs according to McAfee's classifications. Methods: This study enrolled a total of 170 patients who underwent cervical ADR. Pre-existing degenerative change included anterior or posterior osteophytes, ossification of the anterior longitudinal ligament, posterior longitudinal ligament, or ligamentum nuchae. The relationships between basic patient data, pre-existing degenerative change, and HO were investigated using linear logistic regression analysis. Results: Among all 170 patients, HO was found in 69 patients (40.6%). Among the postulated predisposing factors, only male gender and artificial disc device type were shown to be statistically significant. Unexpectedly, preoperative degenerative changes in the cervical spine exerted no significant influence on the occurrence of HOs. The odds ratio of male gender compared with female gender was 2.117. With regard to device type, the odds ratios of Mobi-C (LDR medical, Troyes, France) and ProDisc-C (Synthes, Inc., West Chester, PA, USA) were 5.262 and 7.449, respectively, compared with the Bryan disc. Conclusions: Definite differences in occurrence rate according to the gender of patients and the prosthesis type were identified in this study. Moreover, factors indefinably expected to influence HO in the past were not shown to be risk factors thereof, the results of which may be meaningful to future studies.