Abstract
A pleural effusion is defined as eosinophilic when eosinophils represent ≥ 10% of the total nucleated cell count, and accounts for approximately 10% of all pleural effusions. The diagnostic significance of eosinophilic pleural effusion has yet to be determined.
Objective and Methods: A retrospective study was conducted on 65 patients with eosinophilic pleural effusion to evaluate the correlation between the percentage of eosinophils present in the pleural fluid and the benign or malignant nature of the effusion. An original aspect of current study was the evaluation of other variables in association with pleural eosinophilia, in particular pleural fluid lymphocytosis (≥ 50%), and the presence or absence of fever.
Results: Data showed the trend towards a decrease in neoplastic incidence with increasing percentages of eosinophilic counts, although this correlation was not statistically significant. The presence of fever correlated with low incidence of neoplasms (10% of neoplastic effusions in patients with fever) and was the most significant variable (p=0.001), with a Negative Predictive Value of neoplastic disease of 90%, with sensitivity 92.6% and specificity 47.4%.
When evaluated together with fever, eosinophils increased their discriminating sensitivity to the benign or malignant nature of the effusion but lost in specificity.
When evaluated as absence or presence of lymphocytosis (≥50% lymphocytes), associated with eosinophilia, lymphocytes were significantly associated with the neoplastic nature of the effusion.
Conclusions: the study showed that the finding of eosinophilic pleural effusion should not be considered an indicator of benignity of the effusion; the association of other parameters with eosinophilia, lymphocytosis of the pleural fluid and fever can provide more precise prognostic indications; a high percentage of eosinophils, the absence of lymphocytosis and the presence of fever would seem to be associated with a low probability of a neoplastic nature of the effusion.