Assessment of Residual Sedation in Patients with Traumatic Brain Injury after Cessation of Midazolam Infusion: Is There a Threshold Plasma Concentration?

Abstract

In traumatic brain injury (TBI) patients, therapeutic drug monitoring of midazolam and its metabolites can help to assess residual sedation. Based on previous literature in non-TBI patients, it is assumed that concentrations of <100 ug/L of midazolam are non-sedative. The objective of this study was to assess the threshold sum concentration of midazolam and metabolites for residual sedation after cessation of midazolam infusion in critically ill TBI patients. We performed a retrospective study at the adult ICU of Erasmus University Medical Center between 07-2017 and 03-2024. TBI patients were included for analysis if sequential Glasgow Coma Scale (GCS) scores and at least one midazolam plasma concentration <100 ug/L was available. Residual sedation was evaluated based on improvement in the sum of the Eye and Motor scores from the GCS according to a decision rule. Among 51 patients, 75% (n = 38) had residual sedation. Median sum midazolam concentrations in the residual sedation versus no residual sedation group were 14 ug/L (3-32) and 7 ug/L (3-31), respectively. In the patients with residual sedation an improved EM score of 3 (2-5) was observed within 3 days (2-4) and 22 patients (58%) scored M6 within 4 (2-5) days. Finally, in the residual sedation and no residual sedation group we observed in hospital mortality rates of 5% and 46%, respectively. In TBI patients the threshold concentration for a residual sedative effect of midazolam appears to be much lower than previously described in non-TBI populations. Therefore, clinicians should be reluctant to make withdrawal or limitation of care decisions too early based on low GCS in TBI patients, even in the presence of very low plasma concentrations of midazolam and metabolites.