Urinary lipoarabinomannan: A novel diagnostic tool for distinguishing gastrointestinal tuberculosis from Crohn's disease.

Background: The differentiation between gastrointestinal tuberculosis (GITB) and Crohn's disease (CD) is challenging. Detection of urinary lipoarabinomannan (LAM), a glycolipid component of the Mycobacterium tuberculosis cell wall, has shown potential as a non-invasive diagnostic marker for tuberculosis.

Objective: We evaluated the diagnostic accuracy of urinary LAM in distinguishing GITB from CD.

Methods: This prospective study included patients diagnosed with GITB, CD or those with indeterminate conditions (January 2021 to April 2022). Comprehensive clinical evaluations, laboratory investigations, computed tomography (CT) enterography, colonoscopy and histopathological analyses were performed. First morning midstream urine samples were collected and analyzed using TB LAM antigen kit. The analytical team was blinded from the clinical data. Sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and overall diagnostic accuracy of urinary LAM were determined.

Results: Of 98 patients, 36 were diagnosed with GITB and 62 with CD. Urinary LAM was positive in nine out of 36 GITB patients, yielding a sensitivity of 25% (95% C.I. 12.12-42.20%) and a PPV of 100% (95% C.I. 66.37-100.00%). None of the CD patients tested positive for urinary LAM, resulting in a specificity of 100% (95% C.I. 94.22-100.00%) and NPV of 69.66% (95% C.I. 65.54-73.50%). Overall diagnostic accuracy of urinary LAM in differentiating GITB from CD was 72.45% (95% C.I. 62.54-80.99%). Notably, the addition of urinary LAM testing to the existing diagnostic criteria improved the accurate identification of GITB from 44% to 55.6%.

Conclusion: Urinary LAM testing exhibits high specificity and PPV, making it a significant adjunct in the diagnostic process for GITB.