Pleural Fluid Cholesterol in Differentiating Exudative and Transudative Pleural Effusion

Objectives. To study the diagnostic value of pleural fluid cholesterol in differentiating transudative and exudative pleural effusion. To compare pleural fluid cholesterol level for exudates with Light's criteria. Design. Cross sectional descriptive study. Settings. Medical wards of Tribhuvan University Teaching Hospital. Methods. Sixty two cases of pleural effusion with definite clinical diagnosis admitted in TUTH were taken and classified as transudates (19) and exudates (43). The parameters pleural fluid protein/serum protein ratio (pfP/sP), pleural fluid LDH/ serum LDH ratio, pleural fluid LDH (pfLDH) and pleural fluid cholesterol (pCHOL) were compared with clinical diagnosis with regard to their usefulness for distinguishing between pleural exudates and transudates. Results. The pCHOL values determined were 1.92 ± 0.75 for exudates, 0.53 ± 0.28 for transudates, the differences between the transudates and others are statistically significant (P < 0.0001). It is seen that pfP/sP ratio has a sensitivity of 81.4% and specificity of 82.6%; pfLDH/sLDH ratio has a sensitivity of 86% and specificity of 94.7% and pCHOL with sensitivity of 97.7% and specificity of 100% for differentiating exudative and transudative PE. Conclusion. The determination of pCHOL is of great value for distinguishing between pleural exudates and transudates and should be included in routine laboratory analysis of pleural effusion.


Introduction
Light et al. in 1972 found criteria to have sensitivity and speci�city of 99% and 98%, respectively, for differentiating transudative and exudative PEs (ratio of protein in pleural �uid and serum >0.5; ratio of LDH in pleural �uid and serum >0.6; pleural �uid LDH >2/3rd of upper limit of serum LDH) [1].
But the other investigators could only reproduce speci�cities of 70-86% using Light's criteria. Also it is found that 25% of patients with transudates pleural effusion are mistakenly identi�ed as having exudative effusion by Light's criteria. In cases of heart failure on diuretic therapy, the transudative PE has high protein [2].
Pleural �uid cholesterol can be used to classify exudates and transudates as it misclassi�es fewer cases than any other Light's parameters [3]. From meta-analysis, Heffner et al. 2002 have identi�ed pleural effusion of exudative type with at least one of the following conditions [4].
Pleural cholesterol is thought to be derived from degenerating cells and vascular leakage from increased permeability. ough the cause of the rise in cholesterol levels in pleural exudates is unknown, two possible explanations have been put forward.
According to the �rst, the cholesterol is synthesi�ed by pleural cells themselves for their own needs [5] (extrahepatic synthesis of cholesterol is now known to be much greater than was once thought, depends on the metabolic needs of cells, and is in dynamic equilibrium with cholesterol supply by LDL and cholesterol removal by HDL) [6], and the concentration of cholesterol in pleural cavity is increased by the degeneration of leukocytes and erythrocytes, which contain large quantities.
e second possible explanation is that pleural cholesterol derives from plasma; some 70 percent of plasma cholesterol is bound to low density, high molecular weight lipoproteins (LDL), and the rest to HDL or very low density lipoproteins (VLDL), and the increased permeability of pleural capillaries in pleural exudate patients would allow plasma cholesterol to enter the pleural cavity.
e reason to select the cutoff value of pleural �uid cholesterol as 45 (1.16 mmol/L) is that this cutoff value eliminates the possibility of being equivocal to transudates and exudates, and measurement of pleural cholesterol >45 mg/dL (1.16 mmol/L) has been used to improve the accuracy of differentiating transudative and exudative effusions [7].

Methods
Sample size of 62 consecutive pleural effusion cases that ful�lled the inclusion criteria and were admitted in the Department of Internal Medicine, TUTH were taken. e study period was conducted for one year from July 2010 to August 2011.

Exclusion Criteria
(1) Patients not willing to participate in the study, (2) age < 16 years, (3) patients without de�nite clinical diagnosis, (4) patients with pulmonary embolism or renal insufficiency with pleural effusion, (5) patients previously diagnosed and already on treatment.

Study Procedure.
Aer a detailed history and clinical examination, chest X-ray was done to localize pleural effusion. Diagnostic tapping of the pleural �uid was done in every case, and the help of ultrasonography of chest to localize the �uid was taken in some cases. All pleural �uid samples were tested for cell count, protein, glucose, LDH, pCHOL, Gram stain, bacterial culture, acid fast stain, and cytology. A concomitant blood sample was taken and tested for counts and biochemical parameters such as protein and LDH. Further investigations, such as computed tomography scan of chest, bronchoscopy, and �ne needle aspiration cytology (FNAC), were also done to determine etiology of pleural effusion when needed. e �rst sample of pleural �uid obtained in each patient was considered for analysis. Protein was measured by the biuret method, LDH by UV spectrophotometry at 37 ∘ C and 340 nm [8], and cholesterol with the Boehringer-Mannheim enzymatic method CHOD PAP (cholesterol oxidase peroxidise) [9].
Clinical diagnosis (i.e., etiological diagnosis) was made, and the pleural �uid parameters were analyzed with it. e following evidences were used to include or exclude the cases [10].
(1) Congestive heart failure: presence of clinical features (increased jugular venous pulse, tachycardia, and ventricular gallop) with cardiomegaly or echocardiac evidence of cardiac dysfunction.
(2) Renal diseases: elevated urea (>20 mmol/L) or creatinine > 167 micomol/L with or without signs or symptoms of �uid overload. Quiroga et al. [11], using 45 mg/dL of cholesterol as the cutoff in 80 patients, also reported a sensitivity of 83% and a speci�city of 100%. e statistical signi�cance of the parameters for etiological diagnosis was measured to �nd their usefulness.

Observations and Results
A total of 62 patients with de�nite clinical diagnosis, eligible for the study, were included in which 30.6% (19) cases were transudates, and 69.4% (43) cases were exudates (Figure 1).
It is seen that out of 62 cases (exudates 43 and transudates 19), protein ratio, as Light's parameter, identi�ed 39 cases as   exudates and 23 cases as transudates; LDH ratio identi�ed 38 cases as exudates and 24 cases as transudates, while pCHOL identi�ed 42 cases as exudate, and 20 cases as transudates (see Figure 5). It is seen that pfP/sP ratio has a sensitivity of 81.4% and speci�city of 82.6%; pfLDH/sLDH ratio has a sensitivity of 86% and speci�city of 94.7%, and pCHOL with sensitivity of 97.7% and speci�city of 100% for differentiating exudative and transudative P�s. All these parameters have a signi�cant value that is, <0.0001 (see Table 1). Also on Pearson correlation test, pCHOL correlation is 0.963 and protein ratio (pfP/sP) is 0.591 which suggests that pCHOL is highly correlated than protein ratio with clinical diagnosis for exudate which is signi�cant at the 0.01 level.

Discussion
In this study, a total of 62 patients, 19 with transudates and 43 with exudates, were considered according to the clinical diagnosis. e most frequent cause of pleural exudates is tuberculosis followed by lung cancer which is similar to the result of a study done in Malaysia where there is high incidence of tuberculosis [12�. Protein ratio identi�ed exudates with a sensitivity of 81.4% and speci�city of 82.6%. e pleural �uid to serum LDH ratio has a sensitivity and speci�city of 86% and 94.7%, respectively. Also on Pearson correlation test, pCHOL correlation, and protein ratio (pfP/sP) are 0.963 and 0.591, respectively. It suggests that pCHOL is highly correlated than protein ratio with clinical diagnosis for exudate which is signi�cant at the 0.01 level.
It is found that in transudates, parapneumonic, tubercular, and neoplastic pleural effusions, pCHOL levels were 0.53 ± 0.28 mmol/L, 1.81 ± 0.59 mmol/L, 2.08 ± 0.58 mmol/L, It was found that pCHOL criterion misclassi�ed only one case of malignant effusion as transudate and that happened with the protein ratio of Light's criteria too. Similar �ndings have been reported by others, who suggested that the mis-classi�ed exudates had low cell component concentrations because the pleura had only recently been affected by the tumor [3,13].
Other authors [14,15] believe that a more likely explanation is that the pathogenesis of neoplastic exudates involves more than one mechanism more frequently than that of other kinds.

Conclusion
It is concluded that pCHOL has a better sensitivity, speci-�city, and PPV in differentiating transudates and exudates than the parameters of Light's criteria. is also avoids the plasma protein, sLDH and pleural �uid protein, and LDH. erefore, it is, a more efficient, easier, and a more cost effective method to differentiate exudates from transudates. is study also suggests that determination of pCHOL should be in routine practice in cases of pleural effusion.