In response to the comments made by Green et al (IJCP 1997; 51(7): 480), the discussion in my article in the June issue of this journal was restricted to the diagnostic dilemma between tuberculosis and Crohn's disease. Green et al are absolutely right to point out that tuberculosis is a great mimic; it should, in fact, be considered in the differential diagnosis of patients not only with ileocaecal disease but with any unexplained, undiagnosed and unresolved abdominal problem. Statistically speaking, most patients – at least, in the UK – with a clinical picture and radiology that are compatible with Crohn's disease will have Crohn's disease, but some of them, and not necessarily only those who are obviously immunosuppressed or who are first generation immigrants, may have tuberculosis. This small minority of patients with tuberculosis, when put on steroids with a presumptive diagnosis of Crohn's disease, will have a flare-up of the disease which may be fatal. As for the lymph nodes, while most calcified lymph nodes are tuberculous, not many patients with abdominal tuberculosis have calcified lymph nodes. To obtain the best results, various diagnostic investigations should be used judiciously, depending upon the type of suspected abdominal lesion, e.g. image-guided percutaneous fine needle aspiration cytology (FNAC) for abdominal masses, paracentesis and laparoscopic biopsy for ascitic peritoneal tuberculosis, and colonoscopic biopsy for colonic and ileocaecal lesions. Various techniques, such as multiple biopsies, biopsy-on-biopsy and endoscopic FNAC, have been described to increase the diagnostic yield of colonoscopy. While in some desperate situations it may be justified to give the patient a trial of empirical therapy when there is a high index of suspicion, it is usually preferable to make every effort to arrive at the diagnosis before treatment is started. That is exactly what the article attempted to do when the dilemma is between Koch's and Crohn's.
Read full abstract