Quantitative reagent-free detection of fibrinogen levels in human blood plasma using Raman spectroscopy

Abstract

Fibrinogen assays are commonly used as part of clinical screening tests to investigate haemorrhagic states, for detection of disseminated intravascular coagulation and as a predictor of a variety of cardiovascular events. The Clauss assay, which measures thrombin clotting time, is the most commonly used method for measuring fibrinogen levels. Nevertheless, inconsistencies are present in inter-manufacturer reagent sources, calibration standards and methodologies. Automated coagulation analysers, which measure changes in optical density during the prothrombin time (PT-Fg), have found use in many hospitals. However, the PT-Fg method is found to give falsely elevated values due to varying choices of calibrants, reagents and analysers. As an alternative, Raman spectroscopy has previously been applied to the analysis of blood and its various constituents to determine various analyte concentrations such as glucose, urea, triglycerides and cholesterol. In this study, Raman spectroscopy was investigated for its ability to accurately quantify fibrinogen concentration in blood plasma. Samples collected from 34 patients were analysed by Raman spectroscopy and the resultant spectra were fitted with a Partial Least Squares Regression model using target values obtained through a pre-calibrated Clauss fibrinogen assay. Various spectral pre-processing methods were utilised to prepare data to be entered into a calibration model. A root mean square error of prediction of 0.72 ± 0.05 g/L was achieved with as few as 25 spectra. In this pilot study, Raman spectroscopy has been demonstrated to be a robust technique providing rapid and reagent-free quantification of fibrinogen levels in blood plasma and a potential alternative to the Clauss assay.