Balanced Propofol Sedation Versus Fentanyl and Midazolam Titrated to Moderate Sedation for EGD: A Prospective, Randomized Clinical Trial

Abstract

Background: Balanced propofol sedation (BPS) combines propofol with low doses of an opiate and benzodiazepine, targets moderate sedation, and relies on the endoscopist for dosing decisions of sedative/hypnotics. While data has shown that BPS provides high patient satisfaction during colonoscopy, no randomized clinical trial has compared outcomes of BPS to standard sedation (SS) with an opiate and benzodiazepine alone during EGD. Methods: ASA Class 1-3 patients undergoing elective, outpatient EGD were randomized to moderate sedation with BPS using propofol, fentanyl, and midazolam or SS with fentanyl and midazolam alone. Two experienced endoscopists performed all procedures using capnography and 4 liters/minute of supplemental oxygen. The primary endpoint was patient satisfaction at discharge as measured by 100-millimeter visual analogue scale (VAS): 0=not at all satisfied, 100=completely satisfied. Secondary endpoints included (1) time to full recovery from sedation and (2) patient recall of intraprocedural pain, gagging, and anxiety as measured by VAS (0=none, 100=severe). Wilcoxon rank sum tests for continuous variables and Pearson's chi-square tests for categorical variables were used to assess differences between the two groups with p<0.05 considered significant. Linear regression was then performed to identify factors associated with patient satisfaction. We report results of an interim analysis after enrolling 81 patients (BPS 41, SS 40). Results: The groups were similar with respect to ASA and Mallampati classes, age, sex, and BMI. The median satisfaction score was significantly higher in the BPS arm than the SS arm (99 vs. 88, p<0.001). The BPS arm exhibited significantly less recall of intraprocedural gagging (median 1 vs. 14.5, p<0.001), pain (median 4 vs. 30, p=0.002), and anxiety (median 4 vs. 30, p=0.002) than the SS arm. The BPS arm had significantly shorter recovery times than the SS arm (median 11 vs. 26.5 minutes, p<0.001). Significant positive predictors of patient satisfaction in the multivariable regression model included use of BPS, increasing age, higher BMI, and higher fentanyl dose. No significant difference was observed in the number of patients experiencing transient deep sedation in the two groups (10 in BPS arm, 16 in SS arm, p=0.13). There were no episodes of major hypoxemia (SaO2≤85%) or significant adverse events (ventilatory assistance, use of reversal agents, hospitalization, or serious hemodynamic instability) in either group. Conclusions: BPS targeted to moderate sedation by adequately trained endoscopists yields superior patient satisfaction and shorter recovery times than SS alone during elective EGD.