Adequate bowel preparation is critical for lesion detection during colonoscopy. This study compared the efficacy and tolerance of sodium picosulfate with magnesium citrate (SP+MC) versus 1 L polyethylene glucol and ascorbate (1 L PEG+Asc) in water-assisted colonoscopy.

In this prospective study, we scored bowel preparation quality during colonoscopies in adults for all indications from June 2023 to April 2024. In this time period we converted from SP+MC to 1 L PEG+Asc as the standard purgative at our department. We intubated by water-assisted technique and withdrew with gas insufflation.  Bowel preparation quality was scored using a modified Boston Bowel Preparation Scale (m-BBPS), scored 0–3 in three colon segments during intubation with water-assisted technique. Scores >2 indicated adequate preparation, and 3 indicated excellent preparation. Additionally, tolerance to the purgative and cecum intubation time was recorded.  A subgroup was scored for BBPS (during withdrawal phase in gas insufflation), polyp detection rate (PDR), adenoma detection rate (ADR), sessile serrated lesion detection rate (SSLDR), patient-reported strong pain during colonoscopy and use of sedoanalgesia. Withdrawal time was measured in diagnostic procedures without biopsy. In the multivariate analysis, we corrected for gender and age above 50 years, and in subgroup analysis additionally for indication. 

We scored 1988 colonoscopies, mean age 54.9 years (SD 16.7), 52.3% women, 48.2% used 1 L PEG+Asc. The OR for adequate m-BBPS in all segments with 1 L PEG+Asc compared to SP+MC was 2.46 (CI [1.98, 3.07], p<0.001). OR for excellent m-BBPS right colon segment was 3.49 (CI [2.86, 4.26], p<0.001). 1 L PEG+Asc was less tolerated (OR 0.76, CI [0,61, 0,95], p=0.014). Cecum intubation times were similar between groups (15.4 (SD 10.5) vs. 16.0 minutes (SD 12.9), p=0.29). 

A subgroup of 1120 patients, mean age 55.0 years (SD 15.9), 52.9% women, 56.6% used 1 L PEG+Asc. The OR for adequate BBPS in all segments with 1 L PEG+Asc trended towards significance (OR 1.88, p=0.052). OR for excellent BBPS right colon segment was 1.56, p<0.001.  Patients reported less strong pain during colonoscopy with 1 L PEG+Asc (OR 0.57, p=0.021), but use of sedoanalgesia did not differ (OR 1.02, p=0.88). Mean withdrawal times for diagnostic procedures without biopsy were 9.20 (SD 4.7) and 8.8 (SD 3.2) minutes, respectively (p=0.81). OR for PDR was 1.23, p= 0.13. OR for ADR was 1.15, p=0.34. OR for SSLDR was 1.63, p = 0,068. 

The newly developed m-BBPS (scored during intubation in water-assisted colonoscopy) revealed large and significant differences in rates of adequate bowel preparation in favor of 1 L PEG+Asc, but differences in BBPS (scored in the withdrawal phase in gas insufflation) did not reach significance in the subgroup analysis. The difference in favor of 1 L PEG+Asc was especially large for excellent right colon m-BBPS, also showing significance for BBPS in the subgroup analysis. While 1 L PEG+Asc was less tolerated, it was associated with significantly less strong pain during colonoscopy. These findings suggest that 1 L PEG+Asc may enhance both procedural quality and patient comfort, supporting its adoption as a standard purgative for water-assisted colonoscopy.