Ventricular and lumbar cerebrospinal fluid analysis in 77 HIV-negative patients with Cryptococcal meningitis who received a ventriculoperitoneal shunt.

Abstract

Lumbar cerebrospinal fluid (CSF) parameters are widely studied and have wide clinical applications, but ventricular CSF has rarely been studied since it is relatively difficult to obtain. To determine whether there are differences between ventricular and lumbar CSF parameters and whether the differences have clinical significance, we retrospectively reviewed 77 patients with Cryptococcal meningitis who received a ventriculoperitoneal shunt. We analyzed the following parameters: white blood cell count, total protein concentration, CSF/blood glucose ratio, chloride ion concentration, and Cryptococcal count. All parameters between lumbar and ventricular CSF were remarkably different (all p < 0.001). White blood cell count, total protein level and Cryptococcal count were lower in ventricular CSF than in lumbar CSF, while CSF/blood glucose ratio and chloride ion concentration were higher. Compared to patients without ventriculomegaly, patients with ventriculomegaly had a significantly higher total protein concentration in ventricular CSF (p = 0.047). Compared to patients without surgical complications, patients with complications had a significantly lower CSF/blood glucose ratio in ventricular CSF (p = 0.032). The lumbar CSF parameters had no significant differences between these groups. The changes in lumbar CSF indices over time after shunt placement were also analyzed. After shunt placement, total protein concentration was transiently increased, white blood cell count, CSF/blood glucose ratio and chloride ion concentration were continued at the preoperative level until two months after shunting surgery. These findings suggest that the composition of ventricular CSF differs from that of lumbar CSF, and different CSF parameters have disparate rostro-caudal gradients in patients with Cryptococcal meningitis. Furthermore, ventricular and lumbar CSF parameters may have different clinical implications. Transient deterioration of lumbar CSF parameters after ventriculoperitoneal shunt placement may not be due to disease progression, but to change in CSF flow rate by CSF shunts.