Abstract: BACKGROUND: The domestic and overseas scholars have conducted a large number of studies on the measurement of spinal canal area in patients with lumbar spinal stenosis by CT and MRI. However, due to differences of individuals, spinal canal morphologys, measurement methods and measurement planes, there is no recognized measurement standard and value for the measurement of lumbar spinal canal area at present. There are few reports to evaluate the consistency and repeatability of CT and MRI in measuring lumbar spinal canal area.
OBJECTIVE: To analyze the consistency and repeatability of three-dimensional reconstruction CT and MRI in measuring the cross-sectional area of lumbar spinal stenosis, and to explore the best imaging measurement method for the cross-sectional area of lumbar spinal stenosis. 
METHODS: The preoperative imaging data of 102 patients with lumbar spinal stenosis who underwent surgical treatment in Department of Spinal surgery, the Affiliated Hospital of Southwest Medical University from January 2013 to January 2018 with three-dimensional reconstruction CT and lumbar MRI were collected. The corresponding spinal canal area of each narrow intervertebral disc on three-dimensional reconstruction CT and lumbar MRI images was measured by two spinal surgeons at three different time points. The spinal canal area corresponding to the midline plane of the narrow intervertebral disc parallel to the lower endplate of the upper vertebral body was selected for measurement. Paired t-test was used to analyze the difference in spinal canal area between the results of the two methods. Pearson correlation analysis was used to evaluate the correlation between the results of spinal canal area between the two methods. Intraclass correlation coefficient and Bland-Altman plot were used to analyze the consistency and repeatability of the two methods in measuring the area of narrow lumbar spinal canal. Z-test was used to compare the ICC values of interobserver and intraobserver in measurement of narrow lumbar spinal canal area by the two methods. The protocols were approved by the Affiliated Hospital of Southwest Medical University Ethics Committee (approval No. KY2020176). 
RESULTS AND CONCLUSION: (1) The values of narrow lumbar spinal canal measured by three-dimensional reconstruction CT and MRI were (136.28±2.38) mm2 and (139.98±2.30) mm2; there were significant differences between them (t=-3.96, P < 0.001). Pearson correlation analysis showed that there was a positive correlation between three-dimensional reconstruction CT and MRI measurement of narrow lumbar spinal canal area (r=0.950, P < 0.001). (2) The values of interobserver ICC and intraobserver ICC were 0.908-0.937 and 0.942-0.971. The values of interobserver ICC and intraobserver ICC measured by lumbar MRI were higher than those measured by three-dimensional reconstruction CT (P < 0.05). (3) Bland-Altman plot showed that the 95% distribution range of the difference of spinal canal area between the two methods was -8.0-5.5 mm2. Six points were outside the range, accounting for 3.66%. (4) The results showed that there was a significant difference in spinal canal area between three-dimensional reconstruction CT and MRI, but there was a strong positive correlation. The consistency and reproducibility of measurement of narrow lumber spinal canal area by two imaging examinations were good, but the consistency and repeatability of lumbar MRI in measuring narrow lumber spinal canal area were better than that of three-dimensional reconstruction CT.

Key words: lumbar spinal stenosis, spinal canal area, measurement, CT, MRI, consistency, repeatability