Serum C-reactive protein as an adjunct for identifying complicated parapneumonic effusions.

INTRODUCTION: Distinguishing non-purulent complicated parapneumonic pleural effusions (CPPE) from uncomplicated parapneumonic pleural effusions (UPPE) is challenging. We aimed to determine whether serum C-reactive protein (sCRP), alone or in combination with classical pleural fluid parameters, is useful in making such discrimination.

METHODS: The study was composed of a total of 104 consecutive patients, of whom 47 had UPPE and 57 had CPPE. Standard biochemical pleural fluid data along with sCRP were measured.

RESULTS: sCRP at the time of thoracentesis or chest tube insertion was significantly higher in CPPE (238 mg/L) than UPPE (147 mg/L). At the optimum cutoff value of 200 mg/L, sCRP had a sensitivity, specificity, likelihood ratio positive, likelihood ratio negative, and area under the receiver-operating characteristic curve for diagnosing CPPE of 58 %, 81 %, 3.1, 0.52, and 0.67, respectively. The combination of sCRP >200 mg/L with pleural fluid glucose <60 mg/dL using an "and" rule achieved a specificity of 98 %, whereas both parameters combined in an "or" rule had a sensitivity of 81 %, which was higher than that of pleural fluid pH (57 %) or glucose (54 %).

CONCLUSIONS: sCRP, when combined with classical pleural fluid biochemistries, improves the diagnostic accuracy in identifying those patients with non-purulent parapneumonic effusions who need chest drainage.