Comparison of the Effects of Midazolam/Fentanyl, Midazolam/Propofol, and Midazolam/Fentanyl/Propofol on Cognitive Function After Gastrointestinal Endoscopy

BACKGROUND: Drugs used for sedation/analgesia during gastrointestinal (GI) endoscopy, including midazolam, fentanyl, and propofol, result in short-term, reversible decline in cognitive function. This prospective cohort trial aimed to identify the sedative/analgesic regimen associated with the least impairment of cognition at the time of discharge. METHODS: Patients undergoing elective GI endoscopy were included. Patients investigated at the Prince of Wales Hospital, Sydney, received midazolam/fentanyl (M/F), whereas patients investigated at the Prince of Wales Private Hospital, Sydney, received midazolam/fentanyl/propofol (M/F/P) or midazolam/propofol (M/P). Patients underwent a computerized neurocognitive test, the CogState Brief Battery, before sedation and at discharge. RESULTS: Patients in the M/F group who received gastroscopy (n=22) were administered midazolam 3.36 mg (±0.79 mg) and fentanyl 61.36 μg (±16.77 μg), those who received colonoscopy (n=50) were administered midazolam 3.98 mg (±1.06 mg) and fentanyl 74.50 μg (±24.48 μg), and those who received gastroscopy/colonoscopy (n=28) were administered midazolam 4.82 mg (±1.41 mg) and fentanyl 94.64 μg (±24.35 μg). Patients in the M/F/P group who received colonoscopy (n=45) were administered midazolam 2.77 mg (±0.55 mg), fentanyl 45.11 μg (±25.78 μg), and propofol 148.64 mg (±57.65 mg), and those who received gastroscopy/colonoscopy (n=36) were administered midazolam 2.64 mg (±0.472 mg), fentanyl 35.28 μg (±19.16 μg), and propofol 168.06 mg (±60.75 mg). Nineteen patients in the M/P group who received gastroscopy (n=19) were administered midazolam 2.37 mg (±0.04 mg) and propofol 13.68 mg (±37.74 mg). Neurocognitive scores were significantly lower in the postprocedure test compared with baseline scores for detection, identification, and one card learning (P<0.001). Postprocedure detection test scores were significantly impaired in the M/F group compared with the M/F/P and M/P groups. Predictors of poorer neurocognitive function were midazolam dosage >3 mg (P<0.006) and fentanyl dosage >50 μg (P<0.009). CONCLUSION: The use of propofol in GI endoscopy allows for less exposure to midazolam and fentanyl and is associated with improved cognition at the time of discharge.