Safety of Propofol Used as a Rescue Agent During Colonoscopy

Abstract

Purpose: Many patients who fail to get adequate sedation from narcotics and benzodiazepines during colonoscopy may not get a complete procedure. The administration of propofol by anesthesiologists during colonoscopy is increasing in the United States. However, its use by gastroenterologists has been controversial. We wish to report the safety of propofol used by gastroenterologists as a rescue agent for patients who fail to achieve adequate response to standard sedation. Methods: In patients who failed to get adequate sedation from narcotics and benzodiazepines, it has been our practice for the gastroenterologist to administer propofol (20-30 mg every 3 minutes) as needed, while closely monitoring patients' vital signs, and oxygen saturation every 3-5minutes. We performed a retrospective review of our hospital's electronic database for all colonoscopies from January 2006 to December 2009. We identified 403 subjects who required propofol as a rescue sedative. We also selected 403 controls, matched by their procedure date, undergoing colonoscopies using only benzodiazepines and narcotics. We then compared the rates of adverse effects (major - prolonged hypotension, ventilation, cardiac arrest and minor - transient hypotension, rash, transient hypoxia, over-sedation) in both groups. We also attempted to identify the factors which predicted propofol use. Results for quantitative variables are expressed as mean ± standard error of the mean. Results: There were no major adverse events in either group. The rates of minor adverse events in the propofol and control groups were 0.02 and 0.01, respectively (p=0.56). The following adverse effects were seen in the propofol group: transient hypotension (n=1), nausea/vomiting (n=3), agitation (n=2) and rash (n=1). The following adverse effects were seen with standard sedation: transient hypotension (n=2), nausea/vomiting (n=1), and oversedation (n=2). We also found that patients who received propofol were more likely to be younger (mean 61.0 ± 0.5 vs. 65.1 ± 0.5 years, p=<0.001), history of illicit drug use (OR 2.04, p <0.001), a longer procedure time (mean 41.5 ± 1.1 vs. 31.9 ± 0.9 minutes, p=<0.001). Logistic regression analysis also confirmed that these factors were independent predictors for propofol use (p=0.01). Conclusion: In conclusion, adjunctive propofol administered by gastroenterologists for conscious sedation does not increase the risk of the sedation and appears to be safe. It may be of value in patients who do not respond to narcotics and benzodiazepines such as those with difficult anatomy, substance abuse, and younger patients.