Abstract: Objective To analyze the imaging features of cervical tuberculosis and provide references for its diagnosis and treatment.  Methods A retrospective analysis of 84 cases of cervical tuberculosis patients’ X-ray, CT and MRI, admitted from January 2002 to December 2012 was conducted. The number of vertebral body involvement by tuberculosis, distribution, the changes of vertebra body and of intervertebral space of the involved vertebra body were analyzed, as well as the changes of spinal canal and paraspinal soft tissue. The value of Cobb angle was measured if the kyphosis angular deformity existed. Pearson correlation analysis was used to analyze the correlation between Cobb angle and the vertebral bony defect height and the course. P<0.05 was set as the criteria of statistical significance.  Results Of 84 cases of cervical tuberculosis, there was 1 case (1.2%, 1/84) involving single vertebral body, 63 cases (75.0%, 63/84)involving two vertebral bodies, and 20 cases (23.8%, 20/84) involving more than three vertebral bodies. There were 13 cases (15.5%, 13/84) accompanied involving the attachments, and no attachments involvement alone. The cervical physiological curvature of all cases was changed. There were 64 cases (76.2%, 64/84) with varying degrees reduction of cervical physiological curvature, and 20 cases (23.8%, 20/84) with varying degrees of kyphosis of cervical or cervicothoracic vertebra. The Cobb’s angle ranged from 12° to 82°, with an average of (29.0±19.6)°. The Cobb’s angle was significantly correlated with the vertebral bony defect height and the course (correlation coefficient r was 0.752 to 0.962, P<0.01, with statistical significance). There were 76 cases (90.5%, 76/84) with varying degrees vertebral bone destruction, and 8 cases (9.5%, 8/84) showed no bone destruction with only MRI signal changed. There were 72 cases (85.7%, 72/84) with varying degrees stenosis of the vertebral gap, and 12 cases (14.3%, 12/84) showed no stenosis. There were 39 cases (57.4%, 39/68) with spinal canal deformation with dural sac and nerve root compression, and 23 cases (27.4%, 23/84) with paralysis symptoms. There were 12 cases of cervicothoracic vertebra lesion (C₇-T₁), which accounted for 52.2% (12/23). There were 76 cases (90.5%, 76/84) with paraspinal soft tissue swelling, in which there were 28 cases (36.8%, 28/76) with only paraspinal soft tissue swelling and 48 cases (63.2%, 48/76) with paraspinal soft tissue swelling and abscess.  Conclusion Imaging examination is significantly important for the diagnosis of cervical vertebra tuberculosis. It contributes to define the range and degree of disease invasion and to understand the configuration change of cervical vertebra, and to guide the clinical treatment planning.

Key words: Tuberculosis, spinal/radiography, Cervical vertebrae, Paralysis, Kyphosis