Objectives: Nurse-administered propofol has gained attention as a safe and effective means of sedation for patients undergoing endoscopic procedures. However, little is known about the effect of propofol on the technical performance of colonoscopy.
Methods: Three separate studies were conducted. In the first study, we reviewed procedure notes from consecutive colonoscopies performed by a single experienced endoscopist at our hospital endoscopy unit on patients sedated with either nurse-administered propofol (n = 162) or midazolam/narcotic (n = 164). In the second study, 100 eligible colonoscopy outpatients were randomized to receive either nurse-administered propofol (n = 50) or midazolam/fentanyl (n = 50). In both studies, the measured parameters included visualization of the cecum, time required to reach the cecum, repositioning of the patient, and the application of abdominal counterpressure. In a third study, we reviewed the rate of cecal intubation and colonic perforation in the first 2357 patients in our unit receiving nurse-administered propofol.
Results: In the retrospective comparative study, there was no difference in the cecal intubation rate in those receiving propofol (99.4%) compared to those receiving midazolam/narcotic (97%; p= 0.1), and three of five failed cecal intubations in the latter group resulted from obstructing masses. Patients sedated with propofol were repositioned less frequently compared to those receiving midazolam/narcotic (3.7%vs 26.2%) (p < 0.0001). Abdominal pressure was employed in 9.9% of patients sedated with propofol compared to 19.5% (p= 0.01) of those given midazolam/narcotic. The mean time to reach the cecum was lower in the propofol group than in the midazolam/narcotic group (4.6 min vs 6.0 min, p= 0.002). In the prospective randomized study, the endoscopist intubated the cecum in all 100 patients. Patients in the propofol group were repositioned less frequently than those in the midazolam/fentanyl group (2%vs 24%, respectively, p= 0.001). The number of cases requiring abdominal counterpressure was not significantly different between the propofol and midazolam/fentanyl groups (12%vs 24%, respectively, p= 0.1). The mean time to reach the cecum in the propofol group (3.2 min) was similar to that in the midazolam/fentanyl group (3.8 min, p= 0.08). Among the first 2357 patients in our unit undergoing colonoscopy with nurse-administered propofol, the rate of complete colonoscopy was 99.2% and there were no perforations.
Conclusion: Nurse-administered propofol sedation is safe and simplifies the technical performance of colonoscopy compared to midazolam/narcotic sedation.