Corrected white blood cell count, cell index, and validation of a clinical model for the diagnosis of health care-associated ventriculitis and meningitis in adults with intracranial hemorrhage

Objectives: To evaluate the utility of the corrected CSF white blood cell (WBC) count, cell index, CSF lactate, CSF glucose and a newly developed diagnostic model for the diagnosis of healthcare-associated ventriculitis or meningitis (HCAVM) in the setting of intracranial hemorrhage (ICH). Patients and methods: A case-control study of 111 adult patients with ICH with HCAVM (cases) or without HCAVM (controls) matched 1:2 by age, Glasgow Coma Scale (GCS), and Acute Physiology and Chronic Health Evaluation II (APACHE II) score enrolled in a large tertiary care center from 2003 to 2016. Results: Subjects were appropriately matched by age, GCS, and APACHE II score (P > 0.2). Cases had significantly higher CSF WBC count (uncorrected and corrected), cell index, and CSF lactate, but lower CSF glucose levels than controls (P < 0.05). There were no differences between CSF protein, CSF neutrophilic pleocytosis, and serum C-reactive protein between cases and controls (P > 0.1). The diagnostic accuracy as analyzed by the area under the receiver operating characteristic curve (AUC of ROC) was found to be good for the cell index (0.825), fair for the corrected CSF WBC count (0.770), and poor for the diagnostic model and uncorrected CSF WBC count (0.652 and 0.653, respectively). Conclusion: The diagnosis of HCAVM in patients with ICH remains challenging and although no single parameter is sufficient for diagnosis the cell index proved to be an important indicator of infection in our study.