Abstract
To the Editor,
In the recent review of the British Thoracic Society guidelines for the investigation of unilateral pleural effusion in adults, 1 there was no mention of biochemical analysis of pleural fluid for the diagnosis of parapneumonic pleural effusions (PPE). PPE is an exudate associated with pneumonia, abscess or bronchiectasis, called empyema when it contains pus. 2 The biochemical analysis of pleural fluid is important for the diagnosis of PPE. The aim of our study was to determine the accuracy of biochemical parameters of pleural fluid for the diagnosis of PPE, using receiver-operating characteristic (ROC) techniques by analysing the area under the ROC curve (AUC) and determining the optimal cut-off value. We studied 207 pleural fluids obtained anaerobically by thoracocentesis using gasometry syringes in 110 men and 97 women aged between 1 and 90 y (average 58 y), from June 2005 to July 2006. The following variables were analysed: protein, glucose, lactate dehydrogenase (LDH), lactate and amylase by INTEGRA 400 (Roche Diagnostics S.L®, Barcelona, Spain), adenosine deaminase (ADA) by Modular P (Roche Diagnostics S.L®), pH and pCO2 by gasometer IL-1620 (Izasa S.A.®, Barcelona, Spain), and the aetiological diagnosis of pleural effusion discharged from hospital after evaluating clinical and radiological data and laboratory tests. Protein was analysed by Biuret reaction, glucose by enzymatic reference method with hexokinase, LDH, lactate, amylase and ADA by enzymatic colorimetric methods according to the International Federation of Clinical Chemistry. The serum LDH reference range was 135–225 U/L. The diagnosis of PPE required an acute febrile illness with purulent sputum and pulmonary infiltrates in association with a pleural effusion or positive pleural fluid culture or positive pleural fluid Gram stain or purulent pleural fluid. Pleural fluids were classified into two groups: PPE and non-PPE. Statistical analysis was performed using the software Medcalc®. Forty-six of 207 pleural fluids studied were PPE (14 uncomplicated PPE, eight complicated PPE and 24 empyemas) and 161 were not PPE (63 transudates, 80 malignant effusions, seven tuberculous effusions and 11 other aetiologies). The AUC and the cut-off values are shown in Table 1.
The area under the curve (AUC) and the cut-off values of biochemical parameters of pleural fluid for the diagnosis of parapneumonic effusions (CI, confidence interval)
ADA, adenosine deaminase; LDH, lactate dehydrogenase
Protein is an essential first step in the analysis of pleural fluid and will usually distinguish transudates from exudates: LDH measurement is only required when protein results are equivocal. 1 Amylase is not useful for the diagnosis of PPE. ADA is usually used for the diagnosis of tuberculous effusions. 3 Glucose and lactate are the most commonly used parameters for the diagnosis of PPE. 4,5 pH and pCO2 are used for the diagnosis of complicated PPE or empyemas that require endothoracic drainage tube. 6,7 In our study, ADA, glucose and lactate showed a high accuracy for the diagnosis of PPE, pH and pCO2 best identified complicated PPEs or empyemas, but LDH was the biochemical parameter of pleural fluid that best identified all PPEs. In conclusion, LDH, lactate, glucose and ADA are the biochemical parameters of pleural fluid that show a higher accuracy for the diagnosis of PPE.
Footnotes
Acknowledgements
This work was supported by grants of the Andalucian (P06-CTS-0160) and Spanish (PI 060091; RD06/00130001) governments.