Abstract
Background: Benzodiazepine poisoning is a frequent cause of emergency department (ED) visits, often related to suicide attempts. Flumazenil is the only specific antidote, but its continuous infusion protocol remains controversial because of its uncertain outcome benefits and increased risk of adverse events. This study aimed to evaluate the effect of continuous flumazenil infusion on the time to recovery of consciousness and secondary outcomes in patients with benzodiazepine poisoning stratified by hospitalization status. Methods: A retrospective cohort study was conducted at a tertiary hospital in Seoul, Korea, including adults treated for benzodiazepine poisoning in the ED between April 2019 and March 2024. The primary outcome being the time from arrival at the ED to regaining consciousness. Multivariate regression identified independent predictors of delayed recovery. Results: Among the 370 patients, 52.4% were hospitalized. Flumazenil infusion was administered in 46.8% of the patients, more often in hospitalized patients. In this group, flumazenil infusion significantly reduced the median time to regain consciousness (13.7 vs. 19.4 h, p = 0.006) but did not affect the overall hospital stay. In nonhospitalized patients, flumazenil infusion did not shorten the awakening time or prolong the ED stay. Adverse events, mainly agitation, were more frequent with flumazenil infusion. Conclusions: Continuous infusion of flumazenil accelerates the recovery of consciousness only in hospitalized patients who are severely affected by benzodiazepine poisoning but with increased adverse events and no reduction in hospital stay. Individualized patient selection and evidence-based protocols are needed for optimal and safe antidote use.