Evaluating the Diagnostic Accuracy of Point-of-Care CXCL13 Measurements for Lyme Neuroborreliosis

Background: Measurements of the B-cell attracting chemokine CXCL13 in cerebrospinal fluid (CSF) has become increasingly popular in the diagnostic work-up for Lyme neuroborreliosis (LNB). ReaScan is a point-of-care test for CXCL13 with a short turnaround-time. However, the diagnostic cut-off value for LNB has not been firmly established. Methods: We conducted a retrospective cohort study including all ReaScan CXCL13 measurements performed on CSF samples at Copenhagen University Hospital, Rigshospitalet, from patients of all ages suspected for LNB in 2020-2025. Patients were categorized in accordance with the European Federation of Neurological Societies (EFNS) diagnostic guidelines for LNB as definite, possible, or non-LNB based on a retrospective review of patient records. The manufacturer-informed cut-off value for positive test results was applied and the diagnostic accuracy was calculated based on the ability to identify patients with possible and definite LNB. In addition, a Youden-index optimized cut-off value was calculated and evaluated on the cohort. Results: A total of 1181 ReaScan CXCL13 measurements from children and adults were evaluated against final diagnoses. Using the manufacturer-informed cut-off value (14 AU/mL) ReaScan CXCL13 reached a sensitivity of 38% and a specificity of 98% for identifying cases of possible and definite LNB. A Youden-index optimized cut-off value (1.79 AU/mL) significantly improved the diagnostic accuracy reaching a sensitivity value of 71% and a specificity of 92%. Conclusions: These real-world clinical data suggest that the current cut-off value for the ReaScan CXCL13 analysis is not optimal for diagnosing LNB. Lowering the cut-off level would significantly increase the diagnostic accuracy.