Background and Aims: Propofol is increasingly being used for sedation in colonoscopy; however, its benefits over midazolam (± short-acting opioids) are not well quantified. The objective of this study was to compare safety, satisfaction, and efficiency outcomes of propofol versus midazolam (± short-acting opioids) in patients undergoing colonoscopy. Methods: We systematically searched Medline, Embase, and the Cochrane library (to July 30, 2018) for randomized controlled trials of colonoscopies performed with propofol versus midazolam (± short-acting opioids). We pooled odds ratios for cardiorespiratory outcomes using mixed-effects conditional logistic models. We pooled standardized mean differences (SMDs) for patient and endoscopist satisfaction and efficiency outcomes using random-effects models. Results: Nine studies of 1427 patients met the inclusion criteria. There were no significant differences in cardiorespiratory outcomes (hypotension, hypoxia, bradycardia) between sedative groups. Patient satisfaction was high in both groups, with most patients reporting willingness to undergo a future colonoscopy with the same sedative regimen. In the meta-analysis, patients sedated with propofol had greater satisfaction than those sedated with midazolam (± short-acting opioids) (SMD, .54; 95% confidence interval [CI], .30-.79); however, there was considerable heterogeneity. Procedure time was similar between groups (SMD, .15; 95% CI, .04-.27), but recovery time was shorter in the propofol group (SMD, .41; 95% CI, .08-.74). The median difference in recovery time was 3 minutes, 6 seconds shorter in patients sedated with propofol. Conclusions: Both propofol and midazolam (± short-acting opioids) result in high patient satisfaction and appear to be safe for use in colonoscopy. The marginal benefits to propofol are small improvements in satisfaction and recovery time.