Cervical Spine Deformity

Indications, Considerations, and Surgical Outcomes

Samuel K. Cho, MD; Scott Safir, MD; Joseph M. Lombardi, MD; Jun S. Kim, MD


J Am Acad Orthop Surg. 2019;27(12):e555-e567. 

In This Article

Radiographic Definition


It is important to consider the influence of CSD on global spinal alignment as compensatory changes occur to maintain horizontal gaze. Unlike the thoracolumbar spine (TLS), the CS can be divided into anterior (VB and disks) and posterior (facet joints) columns. The cervical load-bearing axis lies posterior to the VB of C2-C7, with posterior structures bearing about 64% of the axial load. This balance afforded by the stability of bony and ligamentous anatomy maintains the normal cervical curvature and maintains an upright head position[13]

PA and lateral radiographs should be used to assess coronal and sagittal alignment. Flexion-extension views allow for assessment of flexibility and stability. The most common method to evaluate cervical lordosis involves the use of Cobb angles. Similarly, the cervical curvature index (Ishihara) is an alternative method of assessing cervical spinal alignment numerically and is highly correlated with the C2-C7 Cobb angle.[14] Although the cervical Cobb angle is straightforward and has good interrater reliability, measurements such as the sagittal vertical axis (SVA) have gained traction in recent years (Figure 5).

Figure 5.

Cervical alignment, Ishihara index, and spinal sagittal alignment. C7S Beta = C7 tilt angle, c-SVA = cervical sagittal vertical axis, LS = lumbar lordosis, PI = pelvic incidence, PT = pelvic tilt, SS = sacral slope, SVA = sagittal vertical axis, TK = thoracic kyphosis. (Reproduced with permission from Endo K, Suzuki H, Sawaji Y, et al: Relationship among cervical, thoracic, and lumbopelvic sagittal alignment in healthy adults. J Orthop Surg (Hong Kong)2016;24[1]:92–96.)

GSA can be defined in multiple ways. Numerous studies have analyzed the health-related quality of life scores in relationship to adult TLD. These concepts have been echoed in the cervical deformity literature. Global alignment can be assessed by the C7 plumb line and the difference between the C2 and C7 SVA. The normal range of the C2–C7 SVA is reported to be 16.8 ± 11.2 mm.[15] Additionally, in a retrospective review of patients undergoing cervical laminoplasty, Oshima et al[16] found that postoperative functional outcome scores were markedly lower in patients with C2-C7 SVA of >+50 mm.

If full-standing spinal radiographs are unavailable, T1 sagittal angle or T1 slope is useful in predicting overall sagittal balance. The incidence of CK is likely to be twice as high if a patient has a higher T1 slope.[17] Although the high T1 slope demands a greater degree of cervical lordosis, patients may be unable to provide it, causing progressive kyphosis. Cervical lordosis is also strongly linked with T1 slope in maintaining horizontal gaze, and patients with higher T1 slope show more kyphotic changes after cervical laminoplasty at 2-year follow-up.[18] Lee et al[18] demonstrated that thoracic inlet alignment had notable correlation with craniocervical sagittal balance, similar to the effect that the pelvic incidence has on the lumbar spine. Thoracic inlet angle, along with neck tilt, affects the alignment of the CS in that the thoracic inlet angle must increase or decrease based on changes in T1 slope and cervical lordosis to maintain neck tilt at roughly 44° to minimize muscle energy expenditure.[19]

Measuring horizontal gaze is also a crucial clinical data point because it can occur with severe kyphosis and lead to a mechanical dysfunction of swallowing. Swallow dysfunction arises as a consequence of collapse of the pharyngeal space. The chin brow vertebral axis (CBVA) is used to assess horizontal gaze and is defined by the angle subtended between a line drawn from the patient's chin to brow and a vertical line. To obtain this distance, a photograph must be taken with the patient standing, with hips and knees extended and neck neutral or fixed. Lafage et al[20] reported on CBVA thresholds for disability and found that CBVA of <−4.8° or >+17.7° correlated with an Oswestry Disability Index of >40.

Ames Classification

Only one comprehensive CSD classification system exists, which was proposed by Ames et al;[21] it is an adaptation of the Scoliosis Research Society-Schwab classification for adult TLD. The system involves a deformity descriptor and five modifiers.

The classification system requires a full-length standing PA and lateral spine radiographs that include the CS and femoral heads, standing PA and lateral CS radiographs, modified Japanese Orthopaedic Association (mJOA) scores, and a clinical photograph or radiograph that includes the skull for measurement of CBVA.


Scoliosis, though infrequent, presents a distinct challenge for the spine surgeon. It is often found in association with congenital bony anomalies, Klippel-Feil syndrome, and NF type 1. Preoperative CT and MRI are both crucial, especially given the latter's ability to detect spinal dysraphism, which approaches a prevalence of 30% in patients with congenital spine deformity.[22] Surgical approach of this rare entity can be anterior, posterior, or combined (Figure 1). In a case series of 18 patients with isolated cervical scoliosis, the Cobb angle improved from 35.1° to 15.7°; however, a complication rate of 30.8% was related to surgery.[23]

Cervical-thoracic-lumbar Relationship

Given the fact that CK may represent normal alignment in some patients, often a close relationship exists between cervical and thoracolumbar alignment.[24] Smith et al[25] found that 53% of adult patients with TLD had a concomitant cervical deformity. Furthermore, in patients with normal horizontal gaze, thoracolumbar alignment and thoracic kyphosis (TK) directly affect the cervical alignment.[24] Patients with poor sagittal alignment often develop painful compensatory alignment changes to maintain upright posture including knee flexion, pelvic retroversion, thoracic hypokyphosis, and cervical hyperlordosis. Diebo et al[24] found that patients with SVA of >50 mm require cervical lordosis to maintain gaze.

In the presence of spondylotic degenerative changes, loss of compensatory mechanisms for positive sagittal alignment may occur, resulting in increased sagittal deformity.[26] Ames et al[27] analyzed spinal parameters in an asymptomatic population and found that pelvic incidence correlated with lumbar lordosis, lumbar lordosis correlated with TK, and TK correlated with cervical lordosis. The relationship between these parameters does not perfectly translate from one spinal segment to the next. For instance, the increase in cervical lordosis in response to TK may not be enough to maintain the head over the pelvis but does allow the patient to maintain horizontal gaze.

The relationship between cervical deformity and TLD has been analyzed before and after surgical treatment. In a study of 470 patients with TLD, a 53% prevalence of CSD and CSD was associated with C7-S1 SVA, pelvic tilt, and pelvic incidence-lumbar lordosis, suggesting that CSD should be investigated in patients presenting with other spine pathologies.[25]

After correction of the TLD, patients tend to be more misaligned 2 years from surgery with worse TK, T1 slope-cervical lordosis, cervical lordosis, cSVA, C2-T3 SVA, and global SVA compared with patients with TLD who were not operated on.[28] These patients additionally have worse Oswestry Disability Index and Scoliosis Research Society activity at 1 and 2 years. Even after controlling for magnitude of TLD, cervical alignment has a direct effect on health measures.[29]