Multicenter testing of the rapid quantification of radical oxygen species in cerebrospinal fluid to diagnose bacterial meningitis

• Published in: PloS one 2015-05, Vol.10 (5), p.e0128286-e0128286
• Main Authors: Lukaszewicz, Anne-Claire, Faivre, Valérie, Bout, Hélène, Gayat, Etienne, Lagergren, Tina, Damoisel, Charles, Bresson, Damien, Paugam, Catherine, Mantz, Jean, Payen, Didier
• Format: Article
• Language: English
• Subjects: Acetic acid; Anesthesiology; Antibiotics; Bacteria; Bacterial infections; Bacterial meningitis; Brain; Brain Injuries - cerebrospinal fluid; Brain Injuries - complications; Brain Injuries - diagnosis; Brain Injuries - surgery; Cerebrospinal fluid; Confidence intervals; Critical care; Cytokines; Diagnosis; Diagnostic systems; Female; Gene expression; Genetic aspects; Head injuries; Humans; Infections; Inflammation; Intensive care; Intubation; Laboratories; Male; Meningitis; Meningitis, Bacterial - cerebrospinal fluid; Meningitis, Bacterial - diagnosis; Meningitis, Bacterial - etiology; Neurosurgery; Neutrophils; Oxidative stress; Oxygen; Patients; Phorbol 12-myristate 13-acetate; Physiological aspects; Reactive oxygen species; Reactive Oxygen Species - cerebrospinal fluid; Stimulation; Surgery; Trauma; Traumatic brain injury
• ISSN: 1932-6203; 1932-6203
• DOI: 10.1371/journal.pone.0128286

Meningitis is a serious concern after traumatic brain injury (TBI) or neurosurgery. This study tested the level of reactive oxygen species (ROS) in cerebrospinal fluid (CSF) to diagnose meningitis in febrile patients several days after trauma or surgery. Febrile patients (temperature > 38°C) after TBI or neurosurgery were included prospectively. ROS were measured in CSF within 4 hours after sampling using luminescence in the basal state and after cell stimulation with phorbol 12-myristate 13-acetate (PMA). The study was conducted in a single-center cohort 1 (n = 54, training cohort) and then in a multicenter cohort 2 (n = 136, testing cohort) in the Intensive Care and Neurosurgery departments of two teaching hospitals. The performance of the ROS test was compared with classical CSF criteria, and a diagnostic decision for meningitis was made by two blinded experts. The production of ROS was higher in the CSF of meningitis patients than in non-infected CSF, both in the basal state and after PMA stimulation. In cohort 1, ROS production was associated with a diagnosis of meningitis with an AUC of 0.814 (95% confidence interval (CI) [0.684-0.820]) for steady-state and 0.818 (95% CI [0.655-0.821]) for PMA-activated conditions. The best threshold value obtained in cohort 1 was tested in cohort 2 and showed high negative predictive values and low negative likelihood ratios of 0.94 and 0.36 in the basal state, respectively, and 0.96 and 0.24 after PMA stimulation, respectively. The ROS test in CSF appeared suitable for eliminating a diagnosis of bacterial meningitis.