(993) Serum and CSF Concentrations of Midazolam and 1-Hydroxy-Midazolam in Neurosurgical Patients With EVD

Introduction: The penetration of midazolam into the CNS is unknown. We hypothesized that CSF concentrations of midazolam and 1-hydroxy-midazolam were not correlated to serum concentrations or dosages.

Methods: Prospective study of adults in the neurosurgical ICU with an EVD receiving midazolam by continuous infusion. Serum and CSF samples were obtained 12-24 hours and 72-96 hours after infusion initiation and concentrations determined in duplicate by high-performance liquid chromatography with tandem mass spectrometry. Bivariate correlation analyses used Pearson coefficient.

Results: Thirty-one serum and CSF samples were obtained from 18 patients. Mean age was 53.2±14.5 years and the primary diagnoses were subarachnoid hemorrhage (n=10), intraparenchymal hemorrhage (n=3), and stroke (n=2). At sampling, the mean midazolam infusion rate was 3.9±4.4 mg/hr and the previous 12-hour cumulative dose was 51.4±78.2 mg. Midazolam serum concentration (126.2±89.3 ng/mL) showed moderate correlation (r2=0.68, p< 0.001) with serum 1-hydroxy-midazolam concentration (17.7±17.6 ng/mL) but not CSF midazolam concentration (3.9±2.5 ng/mL; r2=0.24, p=0.0051) or CSF 1-hydroxy-midazolam concentration (2.5±0.6 ng/mL; r2=0.47, p=0.3). CSF midazolam concentration did not correlate with CSF 1-hydroxy-midazolam concentration (r2=0.003, p< 0.001). The mean ratio of 1-hydroxy-midazolam concentration to midazolam concentrations in the serum (0.14±0.2 ng/mL) did not correlate with CSF ratio (1.06±0.83 ng/mL; r2=0.06, p=0.19). Concentrations and ratios were unrelated to midazolam infusion rate or 12-hour cumulative dose. Mean concentrations of midazolam and 1-hydroxy-midazolam in the serum and CSF were similar between timepoints. Similarly, the ratios of 1-hydroxy-midazolam to midazolam in the serum and CSF remained stable over time but showed no correlations.

Conclusions: Continuous infusions of midazolam results in measurable concentrations of midazolam and 1-hydroxy-midazolam in the CSF; however, concentrations of midazolam and the metabolite poorly represent serum concentrations or dosages. Accumulation of midazolam and the metabolite in serum or CSF over time were not evident. Pharmacodynamic assessments are warranted.