The diffuse idiopathic skeletal hyperostosis classification based on CT and the correlation of ossification and sagittal spinal morphology / 中华骨科杂志

ABSTRACT

Objective:To investigate CT classification of diffuse idiopathic skeletal hyperostosis (DISH), and to analyze the correlation between the position of ossification in the anterolateral spine and the sagittal configuration of the spine.Methods:The medical records of 109 patients (70 male and 39 female) who underwent whole spine computerized tomography (CT) from October 2018 to October 2020 were retrospectively analyzed. The average age was 68.4±6.9 years old, ranging from 60 to 88 years old. High resolution CT volume rendering technique images were used to assess the degree of anterolateral spinal ossification in each vertebral space, and a CT grading system was established. Sagittal parameters such as thoracic kyphosis (TK), lumbar lordosis(LL), cervical lordosis (CL), sacral slope (SS), and thoracolumbar junction angle (TLJ) of the patients were measured. The sagittal morphology of the spine was divided into four types using the modified Abelin-Genevois (AG) sagittal classification. In AG type 1 patients, the kyphotic vertex was located in the middle of the thoracic spine (T 4-T 11). In AG type 2 patients, there was no significant kyphotic vertex. In AG type 3 patients, the kyphotic vertex was located in the thoracolumbar segment (T 12-L 2). In AG type 4 patients, the kyphotic vertex was located in the upper thoracic segment (T 1-T 3). Inter-observer and intra-observer reliability were calculated by intra-group correlation coefficient ( ICC). Statistical analysis was conducted to investigate the correlation between different AG types and ossification location and severity. Results:The new DISH grading system classifies the severity of anterolateral spinal ossification in each intervertebral space into grades 0 to 3 with an intra-observer ICC value of 0.871 and inter-observer ICC value of 0.874. Combined with Resnick's DISH diagnostic criteria, 97 patients (89.0%) in this study had four consecutive intervertebral spaces with ossification grade 1 or above. For these patients, in T 4-T 11, the standardized ossification grade of AG type 1 was 1.24±0.69, greater than that of AG type 2 (0.84±0.71) and AG type 3 (1.00±0.70), and the differences were statistically significant ( F=23.101, P<0.001). In T 12-L 2, the standardized ossification grade of AG type 3 was 1.44±0.87, which was higher than AG type 1 (1.06±0.84) and AG type 2 (0.72±0.63), the differences were statistically significant ( F=14.008, P<0.001). In this study, no patients with kyphosis apex in the cervicothoracic region (AG type4) were found. In T 1-T 3, there was no statistical difference between the three groups ( F=0.303, P=0.738); in the whole thoracic and lumbar spine (T 1-L 5), there was statistically significant difference in the total ossification grade ( F=14.374, P<0.001), there was no statistical difference between AG type 1 and AG type 3 ( P=0.254), both of which were higher than AG type 2 ( P<0.001). Conclusion:The new DISH ossification grading system proposed in this study has high credibility, which can be used in DISH's study. This study confirmed that the region where the apex of kyphosis is located is prone to anterolateral ossification of the spine.