Does Sedation Type Affect Colonoscopy Perforation Rates?

BACKGROUND: Sedation with propofol is gaining popularity. It is unclear whether sedation with propofol is associated with colonoscopic perforation. OBJECTIVE: The purpose of this study was to compare perforation rates during colonoscopy using sedation with or without propofol. DESIGN: This was a retrospective case series study. SETTINGS: Data from a tertiary center were analyzed. Demographics, method of sedation, and type of endoscopic procedure performed were collected. PATIENTS: Patients who underwent a colonoscopy from January 2003 to October 2012 were analyzed. MAIN OUTCOME MEASURES: Perforation rate expressed per 10,000 colonoscopies was measured. RESULTS: A total of 118,004 colonoscopies were performed during the study period, with 48 perforations (0.041% or 4.1 per 10,000). Overall, the use of propofol was associated with a 2.5 times increased rate of perforation (6.9 vs 2.7 per 10,000; p = 0.0015). Similarly, in patients undergoing therapeutic colonoscopies, there was a 3.4-times increased risk of perforation associated with the use of propofol (8.7 vs 2.6 per 10,000; p = 0.0016). However, in patients undergoing diagnostic colonoscopies, there was no significantly increased risk of perforation with the use of propofol (4.2 vs 2.9 per 10,000; p = 0.64). In univariate and multivariate analyses, there were no differential perforation risks on the basis of sex, but each decade increase in age was associated with an increased risk of perforation. In those patients having a therapeutic colonoscopy, age (per decade) and propofol use were independently and significantly associated with an increased perforation risk, with adjusted ORs of 1.32 (p = 0.04) and 3.38 (p = 0.001). LIMITATIONS: This was a retrospective study with the potential for selection bias. CONCLUSIONS: This study shows that propofol administration is associated with an increased risk of colonoscopic perforation among patients undergoing a therapeutic colonoscopy; however, this association was not evident in patients undergoing a diagnostic colonoscopy. Further studies, such as a prospective, randomized clinical trial, should be done to further evaluate this association.