Related Experiment Video
Updated: Jul 1, 2025
Diffusion Tensor Magnetic Resonance Imaging in Chronic Spinal Cord Compression
Published on: May 7, 2019
Does Magnetic Resonance Imaging Predict Neurological Deficit in Patients with Traumatic Lower Lumbar Fractures?
Karthik Ramachandran1, R Dinesh Iyer1, Prashasth Belludi Suresh1
1Department of Spine Surgery, Ganga Medical Center and Hospitals Pvt. Ltd., Coimbatore, India.
Study Design:
A retrospective cohort study.
Purpose:
This study aimed to understand the role of magnetic resonance imaging (MRI) in predicting neurological deficits in traumatic lower lumbar fractures (LLFs; L3-L5).
Overview Of Literature:
Despite studies on the radiological risk factors for neurological deficits in thoracolumbar fractures, very few have focused on LLFs. Moreover, the potential utility of MRI in LLFs has not been evaluated.
Methods:
In total, 108 patients who underwent surgery for traumatic LLFs between January 2010 and January 2020 were reviewed to obtain their demographic details, injury level, and neurology status at the time of presentation (American Spinal Injury Association [ASIA] grade). Preoperative computed tomography scans were used to measure parameters such as anterior vertebral body height, posterior vertebral body height, loss of vertebral body height, local kyphosis, retropulsion of fracture fragment, interpedicular distance, canal compromise, sagittal transverse ratio, and presence of vertical lamina fracture. MRI was used to measure the canal encroachment ratio (CER), cross-sectional area of the thecal sac (CSAT), and presence of an epidural hematoma.
Results:
Of the 108 patients, 9 (8.3%) had ASIA A, 4 (3.7%) had ASIA B, 17 (15.7%) had ASIA C, 21 (19.4%) had ASIA D, and 57 (52.9%) had ASIA E neurology upon admission. The Thoracolumbar Injury Classification and Severity score (p =0.000), CER (p =0.050), and CSAT (p =0.019) were found to be independently associated with neurological deficits on the multivariate analysis. The receiver operating characteristic curves showed that only CER (area under the curve [AUC], 0.926; 95% confidence interval [CI], 0.860-0.968) and CSAT (AUC, 0.963; 95% CI, 0.908-0.990) had good discriminatory ability, with the optimal cutoff of 50% and 65.3 mm2, respectively.
Conclusions:
Based on the results, the optimal cutoff values of CER >50% and CSAT >65.3 mm2 can predict the incidence of neurological deficits in LLFs.
