Comparison of Tuberculous and Brucellar Spondylitis on Magnetic Resonance Images

STUDY DESIGN: Retrospective cross-sectional study. OBJECTIVE: To investigate the differences between tuberculous spondylitis (TS) and brucellar spondylitis (BS) on magnetic resonance images (MRI). SUMMARY OF BACKGROUND DATA: Both TS and BS are common spinal affections and can cause deformities and permanent neurologic deficiencies without prompt diagnosis and treatment. However, differentiating the types of spondylitis clinically and radiographically is challenging. Despite the diagnosis studies on spine infection throughout the literature, the outcome has not been established. Our study about the different characteristics between TS and BS on MRI may be helpful. METHODS: Eighteen patients with TS and 26 with BS were included in this study. MRI images were obtained and analyzed. For statistical analysis, the χ test was performed. Statistical significance was defined as P < 0.05. RESULTS: There were statistically significant differences between TS and BS on the MRI images: (i) subligamentous spread to three or more vertebral levels [TS 54% (7/13) vs. BS 8% (1/12)]; (ii) subligamentous spread to fewer than three vertebral levels [23% (3/13) vs. 58% (7/12)]; (iii) thoracic spine involvement [50% (9/18) vs. 4% (1/26)]; and lumbar spine involvement [22% (4/18) vs. 77% (20/26)]; (iv) abnormal signal from the vertebral disc on T2-weighted images [33% (7/21) vs. 85% (30/35)]; (v) focal or fan-shaped hyperintense signals on middle sagittal fat-suppressed weighted images [14% (7/50) vs. 50% (31/62)] and especially fan-shaped hyperintense signal [0% (0/50) vs. 23% (14/62)]; (vi) spinal kyphosis [22% (11/50) vs. 3% (2/62)]; (vii) vertebral collapse [42% (21/50) vs. 2% (1/62)]; (viii) peridiscal bone destruction [22% (11/50) vs. 44% (27/62)]; and (ix) psoas abscesses [6% (3/50) vs. 0% (0/62)]. CONCLUSION: MRI is a sensitive imaging technique and can clearly define the different features between TS and BS, so it should be the first choice for imaging in the diagnosis of spondylodiscitis. LEVEL OF EVIDENCE: 4.