RESUMO
Previously established criteria were used to classify 253 pleural effusions as transudates (65 cases), neoplastic exudates (67 cases), tuberculous exudates (65 cases), or miscellaneous exudate (56 cases). The parameters pleural LDH (PLDH), pleural LDH/serum LDH ratio (P/SLDH), and pleural protein/serum protein ratio (P/SPROT) were compared with pleural cholesterol (PCHOL) and the pleural cholesterol/serum cholesterol ratio (P/SCHOL) with regard to their usefulness for distinguishing between pleural exudates and transudates. The PCHOL values determined were 28.5 +/- 12.8 mg/dl for transudates, 88.1 +/- 30 mg/dl for neoplastic exudates, 96.5 +/- 28 mg/dl for tuberculous exudates, and 88 +/- 35.9 mg/dl for the miscellaneous group; the differences between the transudate group and the others are statistically significant (p less than 0.001). The sensitivity and specificity of P/SPROT for diagnosis of exudates were both 89 percent; the sensitivity of PLDH was 67 percent and its specificity was 95 percent; the sensitivity and specificity of P/SLDH were both 84.6 percent. Using Light's three criteria as a battery, the sensitivity was 94.6 percent and its specificity was 78.4 percent. All the transudates and 17 (9 percent) of the 188 exudates had PCHOL values below 55 mg/dl, so that with this threshold, PCHOL had a sensitivity of 91 percent and a specificity of 100 percent for diagnosis of exudates. With a threshold of 0.3, P/SCHOL had a sensitivity of 92.5 percent and a specificity of 87.6 percent. The number of misclassifications by PCHOL was less than with any other of the parameters, with statistically significant differences with respect to PLDH (p less than 0.001) and P/SLDH (p less than 0.01). We conclude that determination of PCHOL and P/SCHOL is of great value for distinguishing between pleural exudates and transudates, and should be included in routine laboratory analysis of pleural effusions.