Prevalence and Factors Affecting Cervical Deformity in Adolescent Idiopathic Scoliosis Patients: A Single-Center Retrospective Radiological Study

INTRODUCTION

In humans, an optimal spinal balance helps to maintain the gravity line in an economic range that is also known as the cone of economy.¹ An adequate trunk balance in erect position also helps to align the head over the femoral heads and to sustain a horizontal visual gaze.² Adolescent idiopathic scoliosis (AIS) is recognized as a complex 3-dimensional deformation of the spine with lateral deviation in the coronal plane, alteration of the kyphosis/lordosis in the sagittal plane, and rotation of the vertebrae in the axial plane. It has been reported³ that there is a significant correlation between the loss of thoracic kyphosis and the development of cervical kyphosis (CK) in AIS patients. The development of cervical kyphosis in AIS patients may represent a compensatory measure from the thoracolumbar deformity.

The need to understand regional as well as global spinal alignment is increasingly recognized by spine surgeons. In adult deformity patients, sagittal malalignment has been demonstrated as a significant source of pain and disability. Recently, Smith et al⁴ reported that cervical deformity (cervical kyphosis [CK] and cervical positive sagittal malalignment [CPSM]) are highly prevalent in adult patients with thoracolumbar deformity. However, the prevalence of CK and CPSM and possible factors affecting CK and CPSM in AIS patients are less clear. Our objective in this study is to assess the prevalence and possible factors affecting cervical deformity in a series of AIS patients.

MATERIALS AND METHODS

RESULTS

A total of 99 patients were included in this study. The patients' mean age was 14 years (range, 10-18 years). There were 76 girls and 23 boys (Table 1). Forty-seven patients (47%) had cervical lordosis and their mean cervical lordosis angle was −10.1° (range, −1° to −30°). Three patients (3%) had a neutral cervical angle (0°). Forty-nine patients (49%) had cervical kyphosis and their mean cervical kyphosis angle was 12.8° (range, 1° to 50°). The patients' mean C2-C7SVA was 30.4 mm. Eighty-three patients (84%) had a C2-C7SVA less than 40 mm. Sixteen patients (16%) had a C2-C7SVA equal to or more than 40 mm (CPSM). The prevalence of CK and/or CPSM was 59% (Table 2).

CK was present in 32% of the patients who had >20° thoracic kyphosis; it was present in 75% of the patients who had ≤20° thoracic kyphosis (P = .003). No association between CPSM and thoracic kyphosis was found (20% versus 14%, P > .05; Table 3).

CK was present in 64% of the patients who had >20° coronal thoracic curve, and it was present in 37% of the patients who had <20° or no coronal thoracic curve (P = .05). Meanwhile, no association between CPSM and coronal thoracic curve was found (19% versus 13%, P > .05; Table 4).

DISCUSSION

Normal cervical lordosis is reported to be between −6.6° and −22.2° in typically developed adults.⁵ The range of cervical lordosis in typically developing adolescents has not been reported, but it has been found that there is a trend of increasing cervical lordosis with age.⁶ In 1995, Hilibrand et al³ studied the sagittal alignment of the cervical spine in 38 AIS patients and found that 89% of them had a straightened or kyphotic cervical spine alignment. They also noticed that there was a significant correlation between the loss of thoracic kyphosis and the development of cervical kyphosis for the entire group. Subsequently, Canavese et al ⁷ confirmed the high incidence of reduced cervical lordosis in AIS patients, with 34.4% of their patients having an average cervical kyphosis of 11°. In the current study we found that 49% of our patients had CK, which is consistent with these previous reports. We also found that the incidence of CK is significantly higher in those patients whose thoracic kyphosis curves are less than 20°. This is also consistent with the previous finding by Hilibrand et al.³

Other than the sagittal cervical angle, another cervical radiographic parameter that has been reported is C2-C7 sagittal vertical alignment (C2-C7SVA). Tang et al⁸ found that in adults, there is a significant correlation between C2-C7SVA and neck disability index scores, with a threshold C2-C7SVA value of approximately 40 mm. Smith et al⁴ studied the prevalence of cervical deformity in 470 adults with thoracolumbar deformity and found that CPSM was present in 29% of the patients. The threshold value for abnormal C2-C7SVA in adolescents has not been reported. Park et al⁶ assessed the effect of age on cervical sagittal alignment in 100 asymptomatic participants and found that C2-C7SVA does not differ significantly on the basis of age. Thus, in this study we choose to define CPSM in these AIS patients as C2-C7SVA equal to or more than 4 cm. In this series, 16% of the patients were found to have CPSM. The lower prevalence of CPSM and higher prevalence of CK in young patients with such a deformity may due to the fact that they have more flexible spines that are able to better compensate the thoracolumbar deformity.⁴ Meanwhile, we found that unlike CK, there is no association between CPSM and thoracic kyphosis. Together, our data may imply that the compensatory maneuver for loss of thoracic kyphosis is loss of cervical lordosis in an effort to maintain a neutral and balanced sagittal vertical axis.

We further studied the relationship between CK and/or CPSM prevalence and the patients' coronal thoracic curves. We found that the patients who had >20° coronal thoracic curve had higher CK prevalence but no CPSM association was found. It has been reported⁴ that adult scoliosis patients, especially those who have thoracic scoliosis, have a high prevalence of CK. Although these cervical deformities are not necessarily symptomatic in AIS patients, our findings do suggest that a cervical deformity evaluation should be included in the assessment of thoracolumbar deformity. However, further studies will be needed to define the clinical significance of the high prevalence of cervical deformity in these young patients.

There are also limitations in our study. We did not collect the data regarding the patients' symptoms in this series, and it is not clear if there is a relationship between cervical deformity and the patients' clinical symptoms.

In conclusion, our study shows that cervical deformity is highly prevalent in AIS patients. There is a significant correlation between the loss of thoracic kyphosis and the development of CK but not the development of CPSM. A high coronal thoracic curve is associated with CK prevalence, but it is not associated with CPSM prevalence. Our finding suggests that adolescents have more ability to compensate and maintain a neutral sagittal vertical axis compared with adults. Further study is needed to define the potential clinical significance of cervical deformity in AIS patients.