David Christopher Kieser, Derek Thomas Cawley, Takashi Fujishiro, Simon Mazas, Louis Boissière, Ibrahim Obeid, Vincent Pointillart, Jean-Marc Vital, and Olivier Gille
The objective of this study was to identify the risk factors of anterior bone loss (ABL) in cervical disc arthroplasty (CDA) and the subsequent effect of this phenomenon.
The authors performed a retrospective radiological review of 185 patients with a minimum 5-year follow-up after CDA (using Bryan, Discocerv, Mobi-C, or Baguera C). Postoperative radiographs were examined and compared to the initial postoperative films to determine the percentage of ABL. The relationship of ABL to potential risk factors was analyzed.
Complete radiological assessment was available in 145 patients with 193 CDRs and 383 endplates (average age 45 years, range 25–65 years, 54% women). ABL was identified in 63.7% of CDRs (48.7% mild, 11.9% moderate, 3.1% severe). Age (p = 0.770), sex (p = 0.200), postoperative alignment (p = 0.330), midflexion point (p = 0.509), maximal flexion (p = 0.080), and extension (p = 0.717) did not relate to ABL. There was no significant difference in the rate of severe ABL between implants. Multilevel surgery conferred an increased risk of any and severe ABL (p = 0.013 for both). The upper endplate, defined as superior to the CDA, was more commonly involved (p = 0.008), but there was no significant difference whether the endplate was between or not between implants (p = 0.226). The development of ABL did not affect the long-term range of movement (ROM) of the CDA, but did increase the overall risk of autofusion. ABL was not associated with pain or functional deficits. No patients required a reoperation or revision of their implant during the course of this study, and there were no cases of progressive ABL beyond the first year.
ABL is common in all implant types assessed, although most is mild. Age, sex, postoperative alignment, ROM, and midflexion point do not relate to this phenomenon. However, the greater the number of levels operated, the higher the risk of developing ABL. The development of ABL has no long-term effect on the mechanical functioning of the disc or necessity for revision surgery, although it may increase the rate of autofusion.
Takashi Fujishiro, Sachio Hayama, Takuya Obo, Yoshiharu Nakaya, Atsushi Nakano, Yoshitada Usami, Satoshi Nozawa, Ichiro Baba, and Masashi Neo
Kyphotic deformity resulting from the loss of cervical lordosis (CL) is a rare but serious complication after cervical laminoplasty (CLP), and it is essential to recognize the risk factors. Previous studies have demonstrated that a greater flexion range of motion (fROM) and smaller extension ROM (eROM) in the cervical spine are associated with the loss of CL after CLP. Considering these facts together, one can hypothesize that an indicator representing the gap between fROM and eROM (gROM) is highly useful in predicting postoperative CL loss. In the present study, the authors aimed to investigate the risk factors of marked CL loss after CLP for cervical spondylotic myelopathy (CSM), including the gROM as a potential predictor.
Patients who had undergone CLP for CSM were divided into those with and those without a loss of more than 10° in the sagittal Cobb angle between C2 and C7 at the final follow-up period compared to preoperative measurements (CL loss [CLL] group and no CLL [NCLL] group, respectively). Demographic characteristics, surgical information, preoperative radiographic measurements, and posterior paraspinal muscle morphology evaluated with MRI were compared between the two groups. fROM and eROM were examined on neutral and flexion-extension views of lateral radiography, and gROM was calculated using the following formula: gROM (°) = fROM − eROM. The performance of variables in discriminating between the CLL and NCLL groups was assessed using the receiver operating characteristic (ROC) curve.
This study included 111 patients (mean age at surgery 68.3 years, 61.3% male), with 10 and 101 patients in the CLL and NCLL groups, respectively. Univariate analyses showed that fROM and gROM were significantly greater in the CLL group than in the NCLL group (40.2° vs 26.6°, p < 0.001; 31.6° vs 14.3°, p < 0.001, respectively). ROC curve analyses revealed that both fROM and gROM had excellent discriminating capacities; gROM was likely to have a higher area under the ROC curve than fROM (0.906 vs 0.860, p = 0.094), with an optimal cutoff value of 27°.
The gROM is a highly useful indicator for predicting a marked loss of CL after CLP. For CSM patients with a preoperative gROM exceeding 30°, CLP should be carefully considered, since kyphotic changes can develop postoperatively.