Aim: Benzodiazepines like midazolam are commonly used for long-term sedation of critically ill children requiring mechanical ventilation. Tolerance to midazolam may occur in these patients resulting in a ceiling effect with insufficient or missing sedative response to increases of midazolam infusion or bolus application. The aim of this study was to evaluate the feasibility of a drug rotation protocol replacing continuous infusion of midazolam with gamma-hydroxybutyrate (GHB) to counteract midazolam tolerance. Methods: This retrospective, observational study was conducted in a 14-bed pediatric intensive care unit of a tertiary referral center. Thirty-three mechanically ventilated children with tolerance to midazolam who received continuous infusion of GHB were included. Success of drug rotation from midazolam to GHB was defined as adequate sedation with GHB and subsequent reduction of required doses of midazolam. Results: In our cohort, drug rotation for at least 2 days could be successfully performed in 10 out of 34 children resulting in subsequent reduction of required doses of midazolam. Drug rotation to GHB failed in 24 patients due to insufficient sedation resulting in a premature termination of the protocol. In these children, dosing of midazolam could not be reduced following drug rotation. We could not identify factors which predict success or failure of drug rotation from midazolam to GHB. Conclusions: The data from our single-center study suggest that drug rotation from midazolam to GHB may be worth trying in children with midazolam tolerance during long-term sedation, but physicians should be aware of possible treatment failure.