Utility of cell index in the diagnosis of healthcare-associated ventriculitis and meningitis: an analytical cross-sectional study

Background The diagnosis of healthcare-associated ventriculitis and meningitis (HCAVM) can be complex because multiple factors confound the interpretation of cerebrospinal fluid (CSF) tests. The cell index (CI) may help in the diagnosis of HCAVM. It does not incur additional medical cost and it avoids delays from the turnaround time of CSF cultures. It is derived by calculating the ratio of CSF white blood cell (WBC) and red blood cell (RBC) divided by the ratio of peripheral WBC and RBC. This study aimed to evaluate the diagnostic utility of this parameter. Methods An analytic, observational, cross-sectional study was conducted at the University of the Philippines - Philippine General Hospital. All admitted pediatric and adult patients from 2015 to 2022 who underwent external ventricular drain (EVD) insertion for hydrocephalus secondary to intracranial hemorrhage (ICH), acute ischemic stroke, intracranial neoplasms, traumatic brain injury, or congenital hydrocephalus were screened. Records of patients fulfilling the inclusion criteria were then reviewed. Results A total of 363 patients underwent EVD insertion from 2015 to 2022. Of these, 161 were included in the study. Two-thirds (66.5%) were adults >= 19 years old whereas the remaining were pediatric patients 1 to < 19 years old. There were no patients < 12 months old as they fulfilled at least one exclusion criteria. Forty-nine of them were later confirmed to have HCAVM based on the CDC/NHSN criteria. A CI cut-off of >= 1.21 gave a maximum sensitivity of 30.6% and specificity of 86.4%. Receiver operating characteristic area under the curve (AUC-ROC) analysis was 0.585. Subgroup analysis by age showed sensitivity of 52.9% in the pediatric age group and 3.13% in adults. Subgroup analysis by neurologic indication showed sensitivity of 27.6% for ICH and 35.0% for neoplasms. Subsequent AUC-ROC analyses, however, showed that CI failed to adequately diagnose HCAVM in these subgroups. Conclusions In our population of neurologic patients who underwent EVD insertion, the cell index is not a reliable parameter in the diagnosis of HCAVM.