Erythrophages do not develop when lumbar CSF and blood samples are mixed in vitro

Background Cerebrospinal fluid (CSF) analysis is a crucial method in the diagnostic process for suspected subarachnoid hemorrhage (SAH), especially when cerebral imaging is negative or inconclusive. CSF cytology (detection of erythrophages or siderophages) is used to determine whether a bloodstained CSF resembles a genuine SAH. Whether erythrophages may develop in vitro after a traumatic puncture in case of delayed CSF analysis is unclear. An in vitro development of erythrophages after traumatic puncture would diminish the diagnostic properties of CSF analysis. We assessed whether erythrophagocytosis is detectable in CSF after an imitated traumatic lumbar puncture. Methods We mimicked a traumatic lumbar puncture by mixing surplus CSF with whole blood from the same patient. From this mixture, cytological specimens were obtained immediately and repeatedly at time intervals of 1 h, until 7 h after mixing, or until the mixture was exhausted. Each cytological specimen was microscopically examined independently by four experienced CSF cytologists for the presence of erythrophages. Results We studied 401 CSF cytological specimens of 96 punctures in 90 patients. We could not identify any erythrophages in all cytological specimens. Fleiss’ Kappa for interrater-reliability was 1.0. Conclusions We did not find evidence for an in vitro erythrophagocytosis after a mimicked traumatic lumbar puncture. Therefore, the occurrence of erythrophages in CSF cytology can be regarded as a reliable sign of an autochthonous bleeding in the subarachnoid space. Our results support the crucial role of CSF analysis in clinical practice in case of a suspected SAH but negative cerebral imaging. Electronic supplementary material The online version of this article (10.1186/s12987-018-0116-3) contains supplementary material, which is available to authorized users.