Midazolam in upper gastrointestinal endoscopy: a single-blind dose-finding study.

Midazolam is a water soluble benzodiazepine with potent sedative and amnestic properties. Sixty unpremedicated patients who were to undergo upper endoscopy for diagnostic purposes were enrolled in an open, non-randomized study to assess the efficacy of 4 increasing dosages of midazolam (0.05, 0.10, 0.15 and 0.20 mg/kg). Efficacy measures included Trieger test (psychomotor drawing), presence of anterograde amnesia, patient rating (PR) of the sedation, evidence of inflammation at the site of the injection and physician's global assessment (PGA). The 4 patient groups were similar in age, sex and ASA class. Eight of the 15 patients who received the 0.05 mg/kg and 1 patient each in the 0.10 mg/kg and 0.15 mg/kg group were considered to be treatment failures and required supplemental diazepam. PR (excellent-good) was constant for all doses throughout. Nearly all patients had anterograde amnesia. Deterioration in the Trieger test results was associated with increasing dose (ANOVA, p less than 0.05). There were no significant changes in vital signs. Signs of phlebitis were noted in two patients who received both diazepam and midazolam. PGA rating rose from 26.7% (0.05 mg/kg) to 80.0% (0.20 mg/kg) (p less than 0.01). Few adverse effects (unrelated to dosage) were noted. Midazolam is a well tolerated benzodiazepine which provides satisfactory sedation for endoscopy. If no premedication is given, there are no benefits to be gained from using a dose greater than 0.10 mg/kg.