Endoscopic Monitoring of Treatment of Indeterminate Intestinal Lesions in a Prospective "Real-Life" Cohort in Eastern India Where Tuberculosis Remains Endemic: Distinguishing Intestinal Tuberculosis From Crohn's Disease.

Introduction It is sometimes difficult to differentiate between intestinal tuberculosis (ITB) and Crohn's disease (CD) in India, as both conditions may mimic each other. The aim was to differentiate ITB from CD in indeterminate intestinal lesions with a therapeutic trial of anti-tubercular therapy (ATT) and follow-up to find out the clinical, endoscopic, radiological, and histological predictors for differentiation between ITB and CD. Methods A prospective observational cohort study of patients diagnosed with ITB and CD according to the Asia-Pacific Guidelines in a "real-life" clinical setting was conducted. ITB was diagnosed by Paustian criteria with Logan's modification. CD was diagnosed according to European Crohn's and Colitis Organization (ECCO) guidelines. We put the patients with a definite diagnosis of ITB and those with an indeterminate diagnosis on ATT and followed them up clinically, endoscopically, and radiologically. Patients were reassessed clinically, endoscopically, and histologically eight weeks after the start of therapy. They were again evaluated endoscopically and radiologically after completion of six months of ATT. The CD patients continued anti-inflammatory, immunomodulator, biological, and/or steroid treatments. Results We conducted this prospective study on consecutive Indian patients who had 21 definite diagnoses of ITB, 26 definite diagnoses of CD, and 42 indeterminate diagnoses. We diagnosed 49 with ITB and 28 (57%) after a therapeutic trial. Ultimately, 40 patients received a CD diagnosis, with 14 (35%) not responding to the ATT therapeutic trial. In patients with ITB, symptomatic improvement after eight weeks of ATT is correlated with endoscopic healing, especially for ulcers but not necessarily for nodularity or strictures. In 50% of these patients, minimal nodularity/pseudopolypii as well as residual scarring was seen on endoscopy even after completion of therapy. Strictures in ITB patients persisted on endoscopy in 40% despite six months of ATT. GI bleeding (64% vs. 10%; p < 0.0001), chronic diarrhea (71% vs. 35%; = 0.02), fistula or sinuses (21% vs. 0%; < 0.01), and multiple site involvement of the intestine (73% vs. 6%; p < 0.0001) were significantly more common in CD than in patients with ITB. Fever (82% vs. 50%; < 0.01) and positive tuberculin tests were more common in ITB patients. PCR positivity and the presence of AFB in smear and culture could be demonstrated in only a small percentage of ITB patients. Conclusion Therapeutic trials in indeterminate intestinal lesions can distinguish ITB from CD without significant adverse effects. Strictures in patients with ITB do not resolve in all patients. GI bleeding, chronic diarrhea, fistulas or sinuses, multiple sites of involvement, and fever have the highest accuracy in differentiating ITB from CD.