PURPOSE: To evaluate the various radiological abnormalities in patients with proven esophageal tuberculosis. MATERIAL AND METHODS: The case records of 23 patients with proven esophageal tuberculosis were evaluated retrospectively for various radiological abnormalities. Twenty-two patients had secondary involvement of esophagus in the form of direct extension of mediastinal and pulmonary tuberculosis or spinal tuberculosis. Only 1 patient had primary involvement of the esophagus with no evidence of disease elsewhere. The diagnosis was confirmed by endoscopic and CT-guided biopsy/aspiration cytology in 7 and 6 cases, respectively. Diagnosis was made on the basis of surgical biopsy of lymph node and autopsy in 1 patient each. In the remaining 8 patients the diagnosis was based on radiological and endoscopic findings and the response to antituberculous treatment. RESULTS: Chest radiography (CXR) was abnormal in 65% patients. While the findings were non-conclusive for esophageal tuberculosis, characteristic lesions of tuberculosis in lungs or spine were suggestive of tuberculous etiology. In 15 patients, CT of the chest confirmed the corresponding CXR findings and also showed additional findings of mediastinal lymphadenopathy when CXR was normal. Fourteen patients showed mediastinal lymphadenopathy on CT of the chest. In all these patients, more than one group of lymph nodes was involved. The characteristic hypodense center of lymph nodes suggestive of tuberculosis was seen in 12 patients. Radiological abnormalities seen in barium swallow examination were extrinsic compression, traction diverticula, strictures, sinus/fistulous tracts, kinking and pseudotumor mass of esophagus in decreasing order of frequency. The middle third of the esophagus was found to be the most frequent site of involvement.
PURPOSE: To evaluate the various radiological abnormalities in patients with proven esophageal tuberculosis. MATERIAL AND METHODS: The case records of 23 patients with proven esophageal tuberculosis were evaluated retrospectively for various radiological abnormalities. Twenty-two patients had secondary involvement of esophagus in the form of direct extension of mediastinal and pulmonary tuberculosis or spinal tuberculosis. Only 1 patient had primary involvement of the esophagus with no evidence of disease elsewhere. The diagnosis was confirmed by endoscopic and CT-guided biopsy/aspiration cytology in 7 and 6 cases, respectively. Diagnosis was made on the basis of surgical biopsy of lymph node and autopsy in 1 patient each. In the remaining 8 patients the diagnosis was based on radiological and endoscopic findings and the response to antituberculous treatment. RESULTS: Chest radiography (CXR) was abnormal in 65% patients. While the findings were non-conclusive for esophageal tuberculosis, characteristic lesions of tuberculosis in lungs or spine were suggestive of tuberculous etiology. In 15 patients, CT of the chest confirmed the corresponding CXR findings and also showed additional findings of mediastinal lymphadenopathy when CXR was normal. Fourteen patients showed mediastinal lymphadenopathy on CT of the chest. In all these patients, more than one group of lymph nodes was involved. The characteristic hypodense center of lymph nodes suggestive of tuberculosis was seen in 12 patients. Radiological abnormalities seen in barium swallow examination were extrinsic compression, traction diverticula, strictures, sinus/fistulous tracts, kinking and pseudotumor mass of esophagus in decreasing order of frequency. The middle third of the esophagus was found to be the most frequent site of involvement.