The nerve root sedimentation sign for differential diagnosis of lumbar spinal stenosis: a retrospective, consecutive cohort study

Purpose Using MR imaging, nerve root sedimentation sign (SedSign) was demonstrated to have a high sensitivity and specificity for diagnosis of symptomatic lumbar spinal stenosis (LSS) in selected patients. This study was to evaluate the diagnostic value of SedSign in differential diagnosis of LSS and non-specific low back pain (LBP) in consecutive patients. Methods A series of consecutive patients with lumbar spinal MRI examination for back/leg pain in orthopeadic clinic were included. These patients were followed up and divided into two groups, symptomatic LSS and nonspecific LBP, according to symptoms and radiological findings. Using MR images, SedSign was assessed by two spine surgeons and one radiologist independently. Then sensitivity and specificity of SedSign was calculated. Result A total of 320 patients (105 LSS and 215 nonspecific LBP) were included. The SedSign had a sensitivity of 77.1 % and specificity of 47.0 % in the whole cohort. When these patients were stratified by dural sac cross-sectional areas (CSA), the SedSign had a sensitivity of 95.0 % and specificity of 4.7 % in patients with CSA <= 80 mm(2) (severe radiologic stenosis), sensitivity of 74.2 % and specificity of 22.6 % in patients with CSA 80-100 mm(2) (moderate radiologic stenosis), and sensitivity of 58.8 % and specificity of 61.0 % in patients with CSA 100-120 mm(2) (mild radiologic stenosis). In selected cases composed by LSS patients with CSA <= 80 mm(2) and non-specific LBP patients with CSA > 120 mm(2), however, the SedSign had a sensitivity of 95.0 % and specificity of 80.0 %. Conclusion The present data demonstrated that the SedSign was not able to discriminate symptomatic LSS from non-specific LBP after adjusting by dural sac CSA. The diagnostic value of the SedSign was still uncertain.