Background: MRIs are a valuable inpatient diagnostic tool but the requirement to remain calm and still can be difficult to achieve due to clinical factors, such as claustrophobia, anxiety, and agitation (1). For most hospitalist patients in non-ICU settings, benzodiazepines (BZDs) are used to achieve minimal sedation for MRI, wherein patients are calm but do not have any respiratory or cardiovascular compromise (2). Lack of clear protocols and standard dosing may contribute to suboptimal dosing, leading to failure to complete the scan from undersedation or leading to adverse events from oversedation (3). This quality improvement project was implemented in response to a safety event in which a patient undergoing MRI was over-sedated from multiple BZD doses, leading to respiratory compromise. It aimed to improve the efficacy of BZD dosing and the safety of MRI scans requiring minimal sedation through the implementation of an electronic medical record (EMR)-based protocol.
Purpose: A clinical pathway was created to guide BZD-based minimal sedation for inpatient MRIs at our large academic, quaternary care hospital by an interdisciplinary workgroup consisting of clinical experts in Hospital Medicine, Anesthesiology, Radiology, Pharmacy and Nursing. Safety measures were incorporated into the pathway including guidance on BZD dosing based on age, limiting BZD administration to a maximum of two consecutive doses, requiring a clinical assessment and vital sign check with BZD administration, and providing indications to consult with an Anesthesiologist for monitored sedation (fig 1). Retrospective chart abstraction was performed on all patients in which the order set was activated.
Description: From March to October 2025, the MRI sedation order set was activated 45 times for 43 patients. 15 cases were excluded from analysis because of order set usage for non-MRI indications, patient refusal before BZD administration, or MRI no longer needed or aborted due to clinical instability. Of the 30 cases in which BZDs were administered and the MRI was attempted, the MRI was completed in 27 (90%) cases. Of the three incomplete cases, two patients required monitored anesthesia for MRI completion and one patient refused MRI attempt after first BZD administration. The average age of patients was 60 years with a range of 21 years to 92 years, indicating a wide age span of efficacy in dosing. Additionally, the order set provided minimal sedation for MRI scans lasting an average of 68 minutes and a maximum of 216 minutes. Most patients (80%) required only 1 dose of BZD to achieve adequate sedation. At our institution, diazepam was used most frequently (60%), followed by midazolam (23%), and lorazepam (17%). In terms of the primary outcome, there were zero instances of safety events, such as oversedation, hemodynamic instability, ICU transfer, rapid response team calls, and death (fig 2).
Conclusions: The implementation of an EMR-based protocol for MRI minimal sedation is effective in achieving adequate sedation for scan completion, while maintaining patient safety. Emphasis on safe BZD dosing, patient monitoring, and appropriate escalation to Anesthesiology promoted safe sedation. Its usage for a wide range of patient ages, scans of various durations, and other non-MRI indications likely indicates its adaptability for many patients and clinical contexts. System-wide implementation of clear protocols and similar EMR-based order sets should be considered to reduce adverse events associated with minimal sedation during routine procedures.
.png)
.png)