Midazolam with meperidine or fentanyl for colonoscopy: results of a randomized trial

A combination of midazolam and opioid is usually used to achieve sedation and analgesia during colonoscopy. Two commonly used opioids are meperidine and fentanyl, but few studies have compared their efficacy.

This randomized trial aimed to compare the efficacy and recovery time of 2 sedation regimens consisting of midazolam in combination with either meperidine or fentanyl.

A total of 300 consecutive, unselected adults attending outpatient colonoscopy at a District General Hospital were enrolled with informed consent and randomized to receive midazolam with meperidine or fentanyl. Data for procedure times, perceived discomfort (according to standard 100-mm visual analog scales [VAS]), and recovery time were collected. Patients and all endoscopy staff directly involved with the procedure were blinded to the regimen used.

Primary: patients' experience of pain (postrecovery VAS score); secondary: recovery time.

A total of 287 patients (150 female, mean [SD] age 54 [17] years) were studied. Recovery time (in minutes) was significantly shorter in patients receiving fentanyl (n = 138) than in those receiving meperidine (n = 149, mean +/- SE: 13.7 +/- 1.8 vs 18.7 +/- 1.7, P = .03), whereas there was no difference in the patients', endoscopists', or nurses' perception of pain during the procedure between the 2 groups. Both groups received a median dose of 3 mg of midazolam (range 2-5 mg). In patients receiving lower doses (2-2.5 and 3-3.5 mg), recovery times were significantly faster with fentanyl (P < .01 and <.05, respectively), whereas at higher doses of midazolam (> or =4 mg) there was no difference between the 2 groups.

The use of VAS scores and nurse assessment of recovery time were chosen in this study because, despite their subjectivity, these measures were felt to most closely reflect true clinical practice.

The use of fentanyl in combination with low-dose midazolam results in significantly faster recovery from sedation compared with meperidine, without any apparent loss of analgesic effect.