Moderate versus deep sedation in adults undergoing colonoscopy: Systematic review and meta-analysis.

Objective To perform a systematic review and meta-analysis comparing effectiveness and safety of moderate and deep sedation during colonoscopy. Research design and methods We searched MEDLINE, EMBASE, CENTRAL and google scholar in May 2017 and updated in March 2018 to identify all randomized controlled trials that compared the effectiveness and safety of moderate and deep sedation during colonoscopy. The quality of studies was assessed using the "Risk of bias" tool. The primary endpoints were defined as patient satisfaction, physician satisfaction, incidence of recall, and incidence of desaturation. Recovery time was also evaluated. Review Manager and Comprehensive Meta-Analysis software were used for statistical analysis. Results A total of 919 patients from three studies were included in the final analysis. The combined analysis did not reveal any differences in patient satisfaction between moderate and deep sedation (RR =0.94; 95%CI: 0.86 to 1.04; Pchi 2 = 0.06; I2 = 65%; number needed to treat to harm [NNTH] = 15.6; 95%CI: NNTH 7.8 to ∞ to number needed to treat to benefit [NNTB] = 3078.0), physician satisfaction (RR =0.35; 95%CI: 0.02 to 6.95; Pchi 2 < 0.001; I2 = 100%; NNTB =1.6; 95%CI: NNTB 1.5 to NNTB 1.8), incidence of recall (RR =5.82; 95%CI: 0.51 to 66.48; Pchi 2 = 0.11; I2 = 60%; NNTH =11.0; 95%CI: NNTH 7.5 to NNTH 20.5), or recovery time (mean difference = -6.77; 95%CI: -16.21 to 2.67; Pchi 2 < 0.001; I2 = 99%). However, incidence of desaturation was higher in the deep group than in the moderate group (RR =0.18; 95%CI: 0.01 to 0.99; Pchi 2 = 0.48; I2 = 0%; NNTB =56.7; 95%CI: NNTB 31.6 to NNTB 273.1). Conclusions Moderate sedation showed comparable safety and effectiveness to deep sedation with respect to patient satisfaction, physician satisfaction, incidence of recall, and recovery time.