Epidemiological and Clinical Profile of Tuberculous Spondylodiscitis: A Retrospective Study from Somalia

Abstract Background: Spondylodiscitis remains a major cause of spinal morbidity in tuberculosis (TB)-endemic regions. This study aimed to evaluate the demographic profile, clinical characteristics, microbiological patterns, and predictors of clinical outcome among patients diagnosed with spondylodiscitis. Objective: To evaluate the prevalence of TB among patients with spondylodiscitis, describe their clinical and radiological characteristics, and identify predictors of clinical outcome. Methods: This retrospective observational study included 121 patients diagnosed with spondylodiscitis. Data were analyzed using IBM SPSS version 23.0. Descriptive statistics were used for demographic, clinical, radiological, and treatment variables. Associations were assessed using Chi-square or Fisher’s exact test for categorical variables and t-test or Mann–Whitney U test for continuous variables. Logistic regression analysis was performed to determine independent predictors of favorable outcome. Statistical significance was set at p ≤ 0.05. Results: The mean age of patients was 48.8 ± 18.1 years, with the highest proportion in the 46–60 age group (31.4%). Males constituted 53.7% of cases. Most patients were married (76.9%), urban residents (76.9%), and belonged to middle (50.4%) or low (45.5%) socioeconomic status. Clinically, back pain (92.6%) and radicular pain (82.6%) were the predominant presenting symptoms. Neurological deficit was present in 46.3% of patients, and epidural abscess was detected in 69.4%. The lumbar spine was the most commonly involved region (48.8%), followed by the thoracic spine (25.6%). The anterior spinal column was affected in 97.5% of cases. TB accounted for 24.8% of all cases , although among tested patients the positivity rate was 65.2% .). Additionally, 85 patients (70.2%) received empirical anti-TB therapy, representing the presumptive TB burden. Drug resistance was low, with monoresistance identified in 13.8% of confirmed TB cases. Among the 65 patients with documented follow-up, 70.8% achieved favorable outcomes (complete or partial recovery). Surgical intervention did not show a significant association with outcome in unadjusted comparisons (p=0.619); however, multivariate logistic regression revealed that surgery independently increased the odds of favorable outcome more than sixfold (OR=6.253, p=0.035). The model demonstrated good discrimination (AUC=0.849). TB exposure history emerged as the strongest independent predictor of unfavorable outcome (OR=0.055, p=0.003), followed by trauma history (OR=0.176, p=0.030). The final regression model demonstrated strong discriminative ability (AUC=0.849; McFadden’s pseudo-R²=0.286). Correlation analysis further demonstrated strong associations between epidural abscess and middle column involvement (r=0.762, p<0.001), as well as neurological deficit (r=0.508, p<0.001). TB exposure history showed a significant negative correlation with favorable outcome (r= −0.392, p=0.001). Conclusion: TB represents a substantial proportion of spondylodiscitis cases, with a confirmed prevalence of 65.2% among tested patients. Although only 24.8% of the total cohort had microbiologically confirmed TB, empirical anti-TB therapy was administered in over two-thirds of patients. Spondylodiscitis in this cohort was predominantly TB-related and frequently associated with neurological deficits and epidural abscess. TB exposure history and trauma were associated with poorer outcomes, whereas surgical intervention independently improved the likelihood of favorable recovery. These findings highlight the importance of microbiological confirmation and risk stratification in managing spondylodiscitis and after adjusting for confounding factors, surgery may significantly improve recovery in appropriately selected patients.