Abstract:

Objective To evaluate the clinical value of the combination of routine pleural effusion and serum assay in differentiating tuberculous pleural effusion from malignant pleural effusion. Methods A total of 160 patients with tuberculous pleural effusion and 267 patients with malignant pleural effusion were selected as the study group. Pleural effusion and serum levels of lactate dehydrogenase (LDH), adenosine deaminase (ADA) and carcinoembryonic antigen (CEA) were collected. Receiver operating characteristic (ROC) curve was used to determine the diagnostic value of pleural effusion ADA, serum LDH/ pleural effusion ADA and pleural effusion CEA/ serum CEA in tuberculous pleural effusion, and to determine the optimal cut-off value. The initial diagnosis procedure was designed as follows: the group 1 was defined as meeting the three conditions of pleural effusion ADA≥29.700 U/L, serum LDH/ pleural effusion ADA < 8.523 and pleural effusion CEA/ serum CEA < 1.096 (considering tuberculous pleural effusion), patients meeting the three conditions of pleural effusion ADA < 29.700 U/L, serum LDH/ pleural effusion ADA≥8.523 and pleural effusion CEA/ serum CEA≥1.096 were defined as the group 2 (malignant pleural effusion considered), and the remaining ones that did not meet all of the above conditions were defined as the group 3. For patients in the group 3, if multinucleated cells ≥ 50% in cell classification of pleural effusion, malignant pleural effusion was considered. For patients with monocytes > 50% in cell classification of pleural effusion, the optimal critical value of ADA in pleural effusion was determined as 34.450 U/L by ROC curve. If ADA≥34.450 U/L, tuberculous effusion was considered. If ADA < 34.450 U/L, malignant pleural effusion was considered. According to the initial diagnosis process, 151 patients with exudative pleural effusion were selected as the verification group. Results Through the initial diagnosis process, 5 patients in the study group did not meet the final diagnostic criteria, including 3 patients with tuberculous pleural effusion (the pleural effusion disappeared after standard anti-tuberculosis treatment), 1 patient was found to have lung cancer 3 years later, 1 patient was found to have colon cancer 2 years later, and 1 patient was complicated with ulcerative colitis. There were 2 patients with malignant pleural effusion, among whom 1 was pleural mesothelioma and the other was lung adenocarcinoma. The sensitivity and specificity of the study group diagnosed by this initial diagnostic procedure for tuberculous pleural effusion were 98.13% and 99.25%, respectively, which were higher than those using ADA alone for pleural effusion. After excluding 49 patients who did not meet the criteria of this study, 102 patients were finally included in the verification group, including 20 cases of tuberculous pleural effusion and 82 cases of malignant pleural effusion. Through the initial diagnosis process, only 2 patients with tuberculous pleural effusion were preliminarily diagnosed and did not meet the final diagnostic criteria, including 1 case of leukemia and 1 case of lymphoma. Conclusion The initial diagnosis process designed by combining conventional pleural effusion and serum tests has improved the accuracy of differentiating tuberculous and malignant pleural effusion.

Key words: tuberculosis, pleural effusion, pleural effusion, malignant, adenosine deaminase, carcinoembryonic antigen, lactate dehydrogenase