Current markers of Lyme neuroborreliosis (LNB) in children have insufficient sensitivity in the early stage of disease. The B-lymphocyte chemoattractant CXCL13 in the cerebrospinal fluid (CSF) may be useful in diagnosing LNB, but its specificity has not been evaluated in studies including children with clinically relevant differential diagnoses. The aim of this study was to elucidate the diagnostic value of CSF CXCL13 in children with symptoms suggestive of LNB.
Children with symptoms suggestive of LNB were included prospectively into predefined groups with a high or low likelihood of LNB based on CSF pleocytosis and the detection of Borrelia antibodies or other causative agents. CSF CXCL13 levels were compared between the groups, and receiver-operating characteristic analyses were performed to indicate optimal cutoff levels to discriminate LNB from non-LNB conditions.
Two hundred and ten children were included. Children with confirmed LNB (n=59) and probable LNB (n=18) had higher CSF CXCL13 levels than children with possible LNB (n=7), possible peripheral LNB (n=7), non-Lyme aseptic meningitis (n=12), non-meningitis (n=91) and negative controls (n=16). Using 18 pg/mL as a cutoff level, both the sensitivity and specificity of CSF CXCL13 for LNB (confirmed and probable) were 97%. Comparing only children with LNB and non-Lyme aseptic meningitis, the sensitivity and specificity with the same cutoff level were 97% and 83%, respectively.
CSF CXCL13 is a sensitive marker of LNB in children. The specificity to discriminate LNB from non-Lyme aseptic meningitis may be more moderate, suggesting that CSF CXCL13 should be used together with other variables in diagnosing LNB in children.
From the *Department of Pediatrics, Stavanger University Hospital, Stavanger, Norway; †Department of Medical Microbiology and ‡Department of Pediatrics, Hospital of Southern Norway Trust, Kristiansand, Norway; §Department of Pediatrics, Haukeland University Hospital, Bergen, Norway; ¶Department of Pediatrics, Haugesund Hospital, Haugesund, Norway; ‖Department of Pediatrics, Hospital of Southern Norway Trust, Arendal, Norway; and ††Department of Clinical Science, University of Bergen, Bergen, Norway.
Accepted for publication April 3, 2017.
Supported by grant from Stavanger University Hospital (laboratory analyses) and the Western Norway Health Authority (scholarship). The authors have no conflicts of interest or funding to disclose.
Address for correspondence: Bjørn Barstad, MD, Department of Pediatrics, Stavanger University Hospital, Gerd Ragna Bloch Thorsens gate 8, 4011 Stavanger, Norway. E-mail: email@example.com.