Clinical Research Of Differential Diagnosis Between Crohn’s Disease And Intestinal Tuberculosis

ObjectiveTo validate the diagnostic value of domestic and foreign mathematical models for differential diagnosis between Crohn’s disease(CD)and intestinal tuberculosis(ITB),further investigate the key clinical features for differentiation.Methods72 patients with definite diagnosis of CD or ITB who were at the gastroenterology outpatient clinic or hospitalized in Sir Run Run Shaw Hospital affiliated of Zhejiang University from March 2007 to November 2016 were enrolled.Based on the clinical data,validate the two differential diagnostic models created by He et al from China and Jung from South Korea respectively.The clinical,serological,radiological,endoscopic and pathological features of the 72 patients were retrospectively compared.Clinical features with significant difference were validated by binary logistic regression analysis to further investigate the key clinical features for differentiation.ResultsClinical and endoscopic scoring system by He et al had higher specificity(91.2%)while the seven-marker model by Jung had higher Sensitivity(97.4%)and accuracy(84.7%).The area under the receiver operating characteristic(ROC)curve of two models were the same and the agreement of both models were high.Further analyze the clinical data,the clinical features indicative of CD were younger age,longer duration of symptom,perianal disease,involvement of small intestine excluding terminal of ileum,longitudinal ulcers,features indicative of ITB were past history of pulmonary tuberculosis or active pulmonary tuberculosis,tuberculin test positive,T-SPOT.TB positive,crypt abscess,confluent granuloma,more granulomas per section,larger granuloma.The clinical,radiological and endoscopic features show no significant difference between CD and ITB by binary logistic regression analysis.ConclusionsDistinguishing CD from ITB should combine clinical,serological,radiological,endoscopic and pathological features.Domestic and foreign mathematical models can help to differentiate CD from ITB.The efficacy of differentiation can be improved by adding differential features including more detailed pathological features.