Assessment of Novel MRI Tools in Tuberculosis of the Spine

Abstract

Background: Spinal tuberculosis remains an important cause of non-traumatic vertebral pathology in tuberculosis endemic regions. Conventional MRI is sensitive for marrow, disc, epidural and paraspinal disease, but overlap with non-tuberculous vertebral lesions can reduce diagnostic confidence. Quantitative MRI markers such as the T1 ratio of marrow (TROM), apparent diffusion coefficient (ADC) and chemical shift imaging-derived in-phase/opposed-phase signal behavior may improve characterization. Objective: To evaluate the role of TROM, diffusion-weighted imaging with ADC measurement and chemical shift imaging in multiparametric MRI assessment of spinal tuberculosis. Materials and methods: This comparative cross-sectional study was conducted at Teerthanker Mahaveer Medical College and Research Centre over 18 months using a 1.5 Tesla Siemens Magnetom Avanto MRI scanner. The analysis compared 86 patients with spinal tuberculosis with 86 controls with non-tuberculous vertebral lesions. Routine spine MRI was supplemented with sagittal DWI/ADC maps and sagittal T1 GRE Dixon in-phase and opposed-phase imaging. TROM, ADC and IP/OP ratio were measured using region-of-interest analysis. Diagnostic performance was assessed using sensitivity, specificity, accuracy and receiver-operating characteristic analysis. Results: Mean age and sex distribution were comparable between groups. TROM was significantly lower in spinal tuberculosis than controls. ADC was significantly higher in spinal tuberculosis and IP/OP ratio was also significantly higher (280.51 +/- 38.13 vs 197.08 +/- 53.56, p < 0.001). ADC showed the highest diagnostic performance at a cutoff >=1000 x10-6 mm2 /s, with sensitivity 98%, specificity 100% and accuracy 99%. TROM at a cutoff <=0.75 showed sensitivity 95%, specificity 90% and accuracy 92%, while IP/OP ratio at a cutoff >=250 showed sensitivity 93%, specificity 85% and accuracy 89%. AUC values were 0.999 for ADC, 0.998 for TROM and 0.891 for IP/OP ratio. TROM correlated negatively with ADC and IP/OP ratio, while ADC correlated positively with IP/OP ratio. Conclusion: Quantitative MRI parameters are useful adjuncts for differentiating spinal tuberculosis from non tuberculous vertebral lesions. ADC was the strongest individual marker, while TROM and IP/OP ratio provided complementary diagnostic value. Incorporating these parameters into routine multiparametric spine MRI may improve diagnostic confidence without contrast administration.