Abstract

Objective: It was aimed to evaluate clinical and laboratory differences and characteristic aspects between cases that were classified in three subgroups as tuberculous spondylodiscitis (TS), pyogenic spondylodiscitis (PS) and brucellar spondylodiscitis (BS).

Methods: Adult patients with spondylodiscitis diagnosis were evaluated retrospectively. Patients who had compatible epidemiological data with brusellosis and had Brucella tube agglutination test ≥1/160 or 4-fold increased titer two weeks later were diagnosed as BS. Cases with M. tuberculosis growth in the tissue samples and/or chronic granulomatous inflammation were diagnosed as TS. Cases in which diagnostic evaluation could not be made and clinical response was elicited with antituberculous treatment were also defined as TS. Cases in which pyogenic bacteria were isolated in samples or culture-negative cases that cured with antibiotic treatment were diagnosed as PS.

Results: Among 103 patients there were 46 patients in TS, 37 in PS and 20 in BS groups. TS cases were younger (p=0.011) and had longer duration of complaints (p=0.044). Localized warmth and redness were observed more often and leukocyte counts, neutrophil counts, C-reactive protein levels and erythrocyte sedimentation rates were higher in PS (p<0.05). The thoracal involvement was most common in TS (41.3%) and lumbar in PS (67.6%). Sensitivity of determining causative organism was 56.7% for computerized tomography-guided percutaneous biopsy. Abscess formation was most frequent in TS and least in BS.

Conclusions: Cervical or thoracal involvement with abscess formation and paravertebral spread in magnetic resonance imaging should be primarily considered as TS. Patients with previous spine surgery, localized warmth and redness on spine, elevated acute phase reactants and lumbar involvement should be considered as PS.