Radiologic and laboratory differences in patients with tuberculous and parapneumonic pleural effusions showing non-lymphocytic predominance and high adenosine deaminase levels
التفاصيل البيبلوغرافية
العنوان:
Radiologic and laboratory differences in patients with tuberculous and parapneumonic pleural effusions showing non-lymphocytic predominance and high adenosine deaminase levels
Tuberculous pleural effusion (TPE) is characterized by lymphocytic predominance and high adenosine deaminase (ADA) levels. However, TPEs sometimes present non-lymphocytic predominance, and parapneumonic effusion (PPE) often exceeds the cutoff value of ADA for TPE. Thus, the differential diagnosis of cases with pleural fluid (PF) showing non-lymphocytic predominance and high ADA levels is challenging. However, limited data concerning the clinical differences in these patients are available. A retrospective study was conducted on TPE and PPE patients with PF showing non-lymphocytic predominance and ADA levels ≥40 U/L in 2009–2013 in a South Korean tertiary referral hospital. The clinical, laboratory, and computed tomography (CT) findings between the groups were analyzed using multivariate logistic regression to develop a prediction model with independent factors for TPE. Among 353 patients with TPE, 24 (6.8 %) showed PF with non-lymphocytic predominance and ADA levels of ≥40 U/L. Twenty-eight PPE patients who presented PF findings comparable with those of TPE patients were included in the control group. In the final analysis, PF ADA levels >58 U/L and nodular lung lesions on CT were independent positive predictors, while loculated effusion was an independent negative predictor for TPE. Using the prediction model, a score ≥ +3 provided a sensitivity of 88 %, specificity of 93 %, positive predictive value of 91 %, and negative predictive value of 90 % for TPE. PF ADA levels, nodular lung lesions, and loculated pleural effusion may help differentiate TPE from PPE in patients with PF showing non-lymphocytic predominance and ADA levels ≥40 U/L.
