Abstract
Postoperative intracranial infection represents significant complications closely associated with a poor prognosis in patients with brain tumors. We aimed to investigate the value of cerebrospinal fluid (CSF) parameters in detection and monitoring of postoperative intracranial infection. This study assessed the diagnostic values CSF parameters for postoperative intracranial infection through the ROC curve and Fisher discriminant model, as well as the prognostic values through Kaplan-Meier analysis and multivariate Cox regression in brain tumor patients. 7 statistically significant CSF variables were extracted and identified, leading to the development of the infection discrimination model with a diagnostic sensitivity of 86.79% and a specificity of 77.01%. Its performance was in excellent agreement with clinical confirmation, as indicated by a κ value of 0.609, which also was validated by the external validation, with 92.86% sensitivity and 64.00% specificity, as well as a κ value of 0.602. Furthermore, elevated levels of CSF total protein, leukocytes, multinucleated cells, and multinucleated cell ratio, along with reduced glucose levels, were found to be associated with prolonged infection control, underscoring their prognostic significance. Multivariate Cox regression analyses identified elevated CSF total protein, reduced glucose and increased leukocyte count as independent predictors of prolonged time to infection control. We established and validated a Fisher model to objectively and quantitatively discriminate true-positive intracranial infection, as well as predicted the infection control time using CSF parameters.