Abstract
<h3>Aims</h3> In 1972, Light <i>et al</i>. described a biochemical criteria to classify exudative and transudative pleural effusion (PE) using the level of protein and lactate dehydrogenase in PE and in blood. We hypothesise the classification can be further streamlined by using a single marker. <h3>Methods</h3> PE samples from 155 patients were studied using<sup>1</sup> H-NMR spectroscopy. NMR spectra were analysed using multivariate analysis for data mining. Potential markers were evaluated using receiver operating characteristic curve (ROC) analysis. <h3>Results</h3> Using Light's criteria, 136 exudates and 19 transudates were classified. Principal component analysis showed a clear clustering between exudative from transudative with R<sup>2</sup> = 0.94 and Q<sup>2</sup> =0.87. A 2-fold increase in lipid signals were found in exudates with area under ROC=0.96 (95% CI 0.92-0.99), sensitivity=85.3% (95% CI: 78.2-90.8) and specificity=100% (95% CI 82.4-100%). <h3>Discussion</h3> For exudates, the permeability of pleura increases which allows the large-size lipids to get into the pleural space from the systemic circulation. Therefore, lipid could serve as an indirect marker for pleural permeability. Despite the unknown lipid identity, this method is simple which makes use of a single marker and requires no biochemical analysis from paired blood sample. In conclusion, lipid is a neglected but useful marker for the classification of PE.
Published Version
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call