[Exploratory research on diagnostic essentials of pediatric tuberculous pleurisy].

Objective: To summarize the clinical diagnostic essentials of pediatric tuberculous pleurisy so as to reduce the misdiagnosis rate. Method: A retrospective study was conducted on 113 cases of tuberculous pleurisy who were seen from August 2006 to September 2014 in the second Department of Respiratory Medicine, Beijing Children's Hospital Affiliated to Capital Medical University. Meanwhile, another 113 cases of children over 5 years of age with mycoplasma pneumoniae pneumonia complicated with pleural effusion were randomly selected as control group. The following five items were analyzed in relation to the diagnosis: fever, cough, lung lesions, pleural effusion(unilateral or bilateral), adenosine deaminase (ADA) level in pleural effusion, tuberculin skin test(PPD). The sensitivity and specificity of single item and combined items in diagnosis were studied. Result: (1) Among two groups, the proportion of patients with persisting fever was 92.0% (104 cases) and 98.2% (111 cases), while the proportion of patients with cough was 47.8% (54 cases) and 99.1% (112 cases) (χ(2)=76.3, P <0.01), respectively. There were 25(22.1%) cases and 113(100.0%) cases with lung lesions (except for atelectasis) in two groups (χ(2)=144.1, P <0.01), respectively. The proportion of patients with unilateral pleural effusion was 94.7%(107 cases) and 71.7%(81 cases)(χ(2)=21.4, P <0.01) while the patients with positive PPD results was 93.8%(106 cases) and 3.5%(4 cases)(χ(2)=184.3, P <0.01). The level of ADA in pleural effusion in both groups are comparable[46.4(22.0-70.9)U/L]and[39.5(26.3-78.5)U/L]. (2)The sensitivity of the lack of cough, lack of lung lesions, unilateral pleural effusion and PPD test positive in diagnosis of tuberculous pleurisy were 52.2%, 77.9%, 94.7%, 93.8%, while the specificity was 99.1%, 100.0%, 28.3%, 96.5%, respectively. Among the four items, the specificity of unilateral pleural effusion was only 28.3%, so it needed to be added with other indicators to improve the specificity. The sensitivity of unilateral pleural effusion combined with positive PPD, lack of cough, lack of lung lesions were 88.5%, 50.4% and 73.5%, while the specificity was 96.5%, 99.1%, and 100.0%, respectively, which significantly improved the specificity so as to reduce misdiagnosis. Conclusion: In children aged over 5 years who present with persistent fever and pleural effusion, tuberculous pleurisy should be considered even in the absence of cough or lung lesions(except for atelectasis), or with positive PPD result, any of the three features is specific for the diagnosis. Unilateral pleural effusion is more meaningful for the diagnosis when it is combined with any of the above three features.