Quantitative Pleural Fluid Echogenicity for Differentiating Transudative From Exudative Pleural Effusions.

Introduction Thoracic ultrasonography (TUS) is more sensitive than chest radiography for identifying pleural fluid. It aids in differentiating between transudates and exudates by observing fluid echogenicity and septations. This study aimed to assess whether quantitative estimation of pleural effusion echogenicity, using a novel index termed pleural fluid relative echogenicity (PFRE), can non-invasively differentiate exudates from transudates. Methods The study recruited 140 patients with pleural effusion. TUS stills of pleural effusion were stored as grayscale images for correlation with biochemical results. Two distinct methods were employed to quantify echogenicity. The first involved selecting two regions of interest (ROI), one within the liver and another within the effusion. The second method involved measuring the quantitative echogenicity of the overall area of pleural effusion and comparing it with that of the liver in the same image. PFRE was calculated as the ratio of pleural fluid echogenicity to liver echogenicity. The predictive ability of PFRE in distinguishing transudates from exudates was evaluated, taking Light's criteria as the gold standard. Results A total of 140 patients were included. Per Light's criteria, 44 patients had transudative effusion and 96 had exudative effusion. Using the ROI method, the median PFRE was 0.26 (0.15-0.44) in transudates and 0.55 (0.35-0.72) in exudates (p<0.001). Positive correlation was found between the effusion protein levels and PFRE (r=0.303, P<0.001). There was a negative correlation between glucose levels and PFRE (r= 0.176, P= 0.039). The area under the curve (AUC) for PFRE to differentiate transudates from exudates was 0.77 (95% confidence interval (CI) 0.69 to 0.84). A cut-off of ≤0.236 for PFRE showed a sensitivity of 47.73%, a specificity of 90.62%, a positive predictive value of 70%, and a negative predictive value of 79.1% to predict transudative effusion. The overall area method showed less accurate results. Conclusion PFRE can differentiate between transudative and exudative effusion non-invasively with a moderate degree of accuracy.