Jean-David Albert1, Valérie Monbet2, Anne Jolivet-Gougeon3, Nadia Fatih4, Maëna Le Corvec5, Malik Seck6, Frédéric Charpentier4, Guillaume Coiffier7, Catherine Boussard-Pledel8, Bruno Bureau8, Pascal Guggenbuhl9, Olivier Loréal10. 1. Department of Rheumatology, CHU of Rennes, 2, rue Henri-le-Guilloux, 35000 Rennes, France; INSERM UMR 991, 35033 Rennes, France. Electronic address: jean-david.albert@chu-rennes.fr. 2. University of Rennes 1, 35043 Rennes, France; UMR-CNRS 6625, IRMAR Mathematics Research Institute of Rennes, 35033 Rennes, France; INRIA/ASPI, 35033 Rennes, France. 3. Pole of Biology, EA 1254 Microbiologie, CHU of Rennes, University of Rennes 1, 35033 Rennes, France. 4. Diafir™, avenue Chardonnet, Parc Lorans, 26J, 35700 Rennes, France. 5. Diafir™, avenue Chardonnet, Parc Lorans, 26J, 35700 Rennes, France; Laboratoire d'Ingénierie des Matériaux de Bretagne (LIMATB), University of Bretagne-Sud, CER Yves-Coppens, BP 573, 56017 Vannes cedex, France. 6. University of Rennes 1, 35043 Rennes, France. 7. Department of Rheumatology, CHU of Rennes, 2, rue Henri-le-Guilloux, 35000 Rennes, France; INSERM UMR 991, 35033 Rennes, France. 8. ISCR UMR-CNRS 6226, Verres et Céramiques, Université de Rennes 1, 35042 Rennes, France. 9. Department of Rheumatology, CHU of Rennes, 2, rue Henri-le-Guilloux, 35000 Rennes, France; INSERM UMR 991, 35033 Rennes, France; University of Rennes 1, 35043 Rennes, France. 10. INSERM UMR 991, 35033 Rennes, France; University of Rennes 1, 35043 Rennes, France.
Abstract
OBJECTIVE: To assess the ability of mid infrared deported spectroscopy to discriminate synovial fluids samples of septic arthritis patients from other causes of joint effusion. METHODS: Synovial fluids obtained from patients with clinically suspected arthritis were collected, analysed and classified according to standard diagnostic procedures as septic or non-septic. A spectroscopic analysis on synovial fluid samples was then performed using a coiled optical fiber made with chalcogenide glass. After a factorial analysis of the normalized spectra and the computation of a Fisher test used to select the most relevant components, a logistic regression model was fitted, allowing to attribute a score between 0 - non-septic -, and 1 - septic. RESULTS: In a first phase, we analysed the synovial fluids from 122 different synovial fluids including 6 septic arthritis among arthritis of various origins. Septic synovial fluids were identified with a sensitivity of 95.8% and a specificity of 93.9% and an AUROC of 0.977. The analysis of an independent set of samples (n=42, including two septic arthritis) gave similar values. CONCLUSIONS: Our data strongly supports the interest of mid infrared deported spectroscopy, which could be used potentially at point of care, for a rapid and easy diagnosis of septic arthritis. Now, the precision of the diagnosis must be evaluated through a multicentric study on a larger panel of patients.
OBJECTIVE: To assess the ability of mid infrared deported spectroscopy to discriminate synovial fluids samples of septic arthritispatients from other causes of joint effusion. METHODS: Synovial fluids obtained from patients with clinically suspected arthritis were collected, analysed and classified according to standard diagnostic procedures as septic or non-septic. A spectroscopic analysis on synovial fluid samples was then performed using a coiled optical fiber made with chalcogenide glass. After a factorial analysis of the normalized spectra and the computation of a Fisher test used to select the most relevant components, a logistic regression model was fitted, allowing to attribute a score between 0 - non-septic -, and 1 - septic. RESULTS: In a first phase, we analysed the synovial fluids from 122 different synovial fluids including 6 septic arthritis among arthritis of various origins. Septic synovial fluids were identified with a sensitivity of 95.8% and a specificity of 93.9% and an AUROC of 0.977. The analysis of an independent set of samples (n=42, including two septic arthritis) gave similar values. CONCLUSIONS: Our data strongly supports the interest of mid infrared deported spectroscopy, which could be used potentially at point of care, for a rapid and easy diagnosis of septic arthritis. Now, the precision of the diagnosis must be evaluated through a multicentric study on a larger panel of patients.