Aims: Nuclear magnetic resonance (NMR) spectroscopy-based metabolome-wide association studies (MWAS) has been increasingly used in advanced laboratory medicine and biomarker discovery. Here, we will apply NMR-based MWAS to develop new diagnostic test for exudative and transudative pleural effusions (PE) with improved accuracy.
Methods: PE will be classified into exudates and transudates according to the etiology and analyzed using 1H-NMR spectroscopy (600 MHz). Biomarkers will be determined using MWAS using Receiver Operating Characteristic Curve Explorer and Tester (ROCCET) and filtered using a metabolome-wide significance level (MWSL) at p value < 4 x 10-6.
Results: There were 50 exudates and 17 transudates. Using MWAS, lipoprotein was determined to be the best biomarker that differentiated exudates from transudates (p value: 1.01 x 10-10) with an area-under-ROC curve of 0.96 (95% CI: 0.89-0.99), sensitivity of 98% and specificity of 88%. The diagnostic performance is superior to the current standard, Light's criteria, which showed a specificity of 65% despite the same sensitivity at 98%.
Discussion: The high level of lipoproteins in exudates is related to the larger capillary pore-size secondary to inflammation. In contrast, without inflammation, the capillary pore-size remains intact in transudates. We envisage NMR-based lipoprotein profiling will be a powerful tool for pleural capillary pore-size estimation and will become a new standard.
(C) 2015 Royal College of Pathologists of Australasia