In the present study, we retrospectively investigated various factors under the hypothesis that the ADI and T1 slope may be associated with sagittal alignment after atlantoaxial fusion. We found an association between the preoperative ADI and difference in the T1 slope after atlantoaxial fusion in the patients with RA. A preoperative ADI higher than 7.92 mm was an independent predictor for T1 slope increase after atlantoaxial fusion. We found a nearly 4.6-fold higher T1 slope increase in the higher preoperative ADI group (>7.92 mm) than in the lower preoperative ADI group.
Previous studies on subaxial cervical spinal changes after atlantoaxial fusion have reported limited findings with regard to predicting these changes in patients with RA because they enrolled patients with various etiologies of C1-C2 instability. In the present study, we evaluated patients with AAI due to RA who underwent TAF or C1LM-C2P screw fixation. Therefore, our study represents a more homogeneous group than do previous studies. Changes in the subaxial cervical spine can develop not only as a natural course of RA but also as a consequence of upper cervical fusion or disruption of the extensor muscles involved in posterior cervical surgery.[24–26] For these reasons, we focused on atlantoaxial fusion for AAI in patients with RA. To our knowledge, this study is the first to demonstrate the relationship between the preoperative ADI and postoperative changes in the T1 slope after atlantoaxial fusion in patients with RA.
C1-C2 arthrodesis has been widely used in the treatment of AAI in patients with long-standing RA. The C1LM-C2P screw fixation technique and TAF technique have been introduced to stabilize AAI.[6, 7] However, atlantoaxial fusion limits the motion of the atlantoaxial spine. In addition, we often encounter postoperative subaxial alignment changes in some cases, which can be a cause of neck pain or neurologic impairment.[13–18] Kyphotic changes in the subaxial cervical spine are one of the adverse events following atlantoaxial fusion. Approximately 33%-48% of all patients who undergo atlantoaxial fusion develop postoperative kyphosis or swan-neck deformity of the lower cervical spine. Yoshimoto et al. reported that 42% of their patients who underwent atlantoaxial fusion showed progression of kyphosis in the subaxial cervical spine, which is attributable to atlantoaxial fusion in a hyperextended position.[18, 27–29] However, these results are limited in significance because these previous studies included a variety of surgeries and diseases.
The C1-C2 fixation angle has been emphasized as a key factor to regulate cervical subaxial alignment in atlantoaxial fusion in previous reports.[14, 17, 18, 30, 31] In asymptomatic individuals, there was a negative linear correlation between the angles of C1-C2 and C2-C7.[30, 31] For patients with AAI, C1-C2 arthrodesis in a hyperlordotic position could cause sagittal kyphosis of the lower cervical spine. The subaxial kyphosis is more frequently developed as the C1-C2 fixation angle increases from surgery. Therefore, atlantoaxial fixation in excessive lordotic alignment in a hyperextended position should be avoided to prevent subaxial malalignment postoperatively.[14, 18]
Previous studies have reported correlations among several cervical alignment parameters, including the T1 slope, C2-C7 angle, and Oc-C2 angle.[32] Various types of cervical surgery were associated with changes in postoperative sagittal balance and postoperative symptoms. Kwon et al. reported that the C2-C7 SVA after two-level ACDF was affected more significantly by the SVA and C2-C7 angle than by the T1 slope.[22] Knott et al. suggested that the T1 slope is the most important predictor of the C2-C7 SVA and recommended to perform cervical radiography at an upright position when the T1 slope is below 13° or above 25°.[19] Kim et al. measured cervical sagittal alignment after laminoplasty and reported that cervical kyphotic deformity at the 2-year postoperative follow-up increased with increasing preoperative T1 slope.[21] Hyun et al. suggested that the T1 slope and C2-C7 lordosis mismatch is a cervical analog for cervical lordosis and thoracic lumbar pelvic incidence.[33] These studies revealed that a deformity of the upper cervical spine is compensated by the subaxial cervical spine, including the T1 slope.
In addition, several studies have reported associations between clinical outcomes and alignment of the cervical spine after cervical spinal surgery. Naderi et al. suggested that an abnormal cervical curvature was associated with less improvement in neurological symptoms after surgery.[34] In a double-blinded randomized trial, improvement in cervical sagittal alignment was not correlated with clinical outcomes after anterior cervical fusion; however, an improved segmental angle was associated with an improvement in clinical outcomes.[35] Guérin et al. also reported that improvement in the segmental angles, as opposed to that in the cervical lordotic angles, is correlated with improvement in clinical symptoms after cervical disc replacement.[36] Improvement in cervical sagittal alignments was associated with better clinical outcomes after cervical spinal surgery.
In our study, we demonstrated a significant positive correlation between the preoperative ADI and T1 slope difference after atlantoaxial fusion. A higher preoperative ADI (>7.92 mm) was related to more increases in the T1 slope after surgery. The T1 slope is well known parameter that may be easily used in evaluating sagittal balance in particular situations full column radiographs are not available.[19] It is constant morphological values within an individual and positively correlated with subaxial lordosis to maintain sagittal balance of the cervical spine.[20, 21] Moreover, an increasing T1 slope has been shown to correlate with greater sagittal malalignment of the dens significantly.[19, 32]
The T1 slope significantly increased more in the patients with a higher preoperative ADI, which adversely affected the sagittal alignment of the cervical spine after atlantoaxial fusion. Moreover, the increasing T1 slope significantly correlated with sagittal malalignment of the cervical spine. For these reasons, surgeons should consider cervical sagittal balance when fixating C1-C2 screws during surgery in patients with higher preoperative ADIs.
Our study aimed to investigate various factors under the hypothesis that the ADI and T1 slope may be associated with sagittal alignment after atlantoaxial fusion for AAI among patients with RA. When the preoperative ADI was ≥7.92 mm, the patients who underwent surgical treatment showed significantly more increases in the T1 slope. This suggests that AAI due to RA causes degenerative changes in the subaxial cervical spine as well as atlantoaxial lesions, which might affect cervical sagittal alignment. Therefore, performing surgical treatment in patients with lower ADIs would lead to better sagittal alignment and less pain. The ADI in patients with AAI will increase over time. Surgeons need to monitor ADI changes closely and administer appropriate treatments in response to such changes. Nuchal pain caused by cervical spinal instability can be resolved after cervical fusion. However, further improvement in symptoms can be expected when sagittal alignment is carefully considered. For this reason, the appropriate cervical angle should be considered after fixating the head in the prone position before surgery; the appropriate angle should also be considered for rod fixation after screw fixation of the C1 and C2 during surgery.
This study has several limitations. First, it was a retrospective study of a relatively small sample of patients. Second, the retrospective nature of the analysis in this study might have introduced some patient selection bias. Third, the only symptom that was investigated before and after surgery was neck pain, as measured by the VAS score. Thus, the relationship between the T1 slope and clinical outcomes might need further assessment. We also did not evaluate the complications after surgery. These limitations indicate the need for further research and clinical studies in this field. Prospective studies on a larger sample are needed. Subsequent prospective studies should not only investigate the cervical sagittal angle but also analyze the overall spinal angle.