DIAGNOSTIC ACCURACY OF MAGNETIC RESONANCE IMAGING IN PATIENTS WITH TUBERCULOUS SPONDYLITIS, USING HISTOPATHOLOGY AS THE GOLD STANDARD

Abstract

Objective: To determine diagnostic accuracy of magnetic resonance imaging (MRI) in patients with tuberculosis (TB) spondylitis, using histopathology as gold standard.

Study design: Cross-sectional diagnostic accuracy study.

Place and Duration: Islamabad Diagnostic Centre, F-8 Markaz, Islamabad, from October 2024 to March 2025 (6 months).

Methodology: Patients (n = 90) with clinical suspicion of TB spondylitis were included through consecutive non-probability sampling. All patients underwent MRI of affected spine. Histopathological examination of biopsy or surgical tissue specimens was used as reference standard for confirmation of disease. Data were analyzed using SPSS. MRI findings were compared with histopathology through 2×2 contingency table. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), diagnostic accuracy and area under the receiver operating characteristic (ROC) curve were calculated.

Results: Among 90 patients, 57 were males and 33 were females. Mean age was 41.35±7.03 years in males and 43.15±8.11 years in females. Mean disease duration was 9.51±4.89 months. Sensitivity of MRI was 98.59%, specificity of 84.21%, PPV of 95.89%, NPV of 94.12% and overall diagnostic accuracy of 95.56%. Area under ROC curve was 0.914 (95% CI: 0.815–1.00), indicating excellent diagnostic performance.

Conclusion: MRI showed better diagnostic accuracy for TB spondylitis and is valuable non-invasive tool for early diagnosis. However, histopathology still remains important for final confirmation in doubtful cases.