OBJECTIVETo explore the value of spinal canal and dural sac dimensions for the treatment of lumbar disc herniation in MRI.

METHODSThe clinical data of 144 patients with single-level lumbar disc herniation underwent nonsurgical or surgical treatment from January 2010 to December 2012 were retrospectively analyzed. There were 91 patients in the nonsurgical group, including 55 males and 36 females, ranging in age from 20 to 68 years old with an average of (43.37±12.48) years; and there were 53 patients in the surgical group, including 28 males and 25 females, ranging in age from 20 to 64 years old with an average of (42.98±12.95) years. JOA scores (29 scores) were used to evaluate clinical manifestation (including subjective symptoms, objective findings, limitation of daily activities and bladder function) and outcomes. The parameters related to spinal canal and dural sac dimensions (including spinal canal midsagittal diameter and available diameter, lateral recess width, spinal canal and dural sac cross-sectional area) in the initial axial T2-weighted MRI were measured, and odds ratio of available diameter to midsagittal diameter, odds ratio of lateral recess width to midsagittal diameter and area ratio of dural sac to spinal canal were calculated. Then, the differences of all parameters between two groups, and the correlations with initial JOA scores were analyzed.

RESULTS(1) All patients were followed up from 1 to 3 years with an average of 2.1 years. JOA scores before treatment were 16.27±2.96 in nonsurgical group and 12.64±3.30 in surgical group, there was statistically significant difference (t=6.319, P<0.01). At final follow-up time, there was no statistically significant difference in JOA scores (25.41±2.22 vs 25.76±2.29), improvement rate [(72.95±12.54)% vs (76.80±9.45)%], and the excellent and good rate (84.91% vs 78.02%) between two groups (P>0.05). But, the relapse rate of nonsurgical group was higher than surgical group (14.29% vs 5.67%). (2) Spinal canal midsagittal diameter and available diameter, lateral recess width, spinal canal and dural sac area, the ratio of available diameter to midsagittal diameter, and the ratio of lateral recess width to midsagittal diameter in surgical group were smaller than that of nonsurgical group, but the area ratio of dural sac to spinal canal was larger, and there were statistically significant differences between two groups (P<0.01). (3) The initial JOA scores showed significantly positive correlation with spinal canal midsagittal diameter and available diameter, lateral recess width, and canal and dural sac area (P<0.01); also presented positive correlation with the ratio of available diameter to midsagittal diameter and the ratio of lateral recess width to midsagittal diameter (P<0.05); but there was a significantly negative correlation between initial JOA scores and the area ratio of dural sac to spinal canal.

CONCLUSIONBoth nonsurgical and surgical treatment of lumbar disc herniation can obtain good effect, but the recurrence rate of non-surgical treatment is higher. Preoperative MRI measurement parameters of spinal canal and dural sac dimensions has certain value for the treatment selection of lumbar disc herniation, but further refinement and validation is still required.