3 L split-dose polyethylene glycol is superior to 2 L polyethylene glycol in colonoscopic bowel preparation in relatively high-BMI (≥ 24 kg/m²) individuals: a multicenter randomized controlled trial

This study was a multicenter, endoscopist single-blinded RCT. This study was approved by the research ethics boards from all participating hospitals. In addition, we registered at the Chinese Clinical Trial Registry (ChiCTR2000039068).

Patients scheduled to undergo colonoscopy were selected from 7 tertiary hospitals in Sichuan Province from January to November 2021. The inclusion criteria were as follows: (1) patients who underwent colonoscopy for the first time within one month; (2) age of 18-65 years; (3) BMI ≥ 24 kg/m²; and (4) signed informed consent. Patients were excluded if any of the following conditions were satisfied: (1) pregnant or lactating women; (2) history of colon resection; (3) serious heart disease (acute heart failure and acute coronary syndrome, serious liver disease (acute liver failure and decompensate cirrhosis belonged to Child-Pugh class C, serious kidney disease (eGFR<15 ml/min), serious lung disease (respiratory failure with PaO2 less than 60 mmHg or PaCO2 higher than 50mmHg). (4) severe electrolyte disorders; (5) gastrointestinal bleeding with hemoglobin level less than 70 g/L; (6) intestinal obstruction, toxic megacolon, severe inflammatory bowel disease; or (7) refusal to use PEG.

Eligible participants were randomly assigned to the experimental group or the control group. The experimental group was given a 3 L split-dose regimen (1 L PEG was taken at 8:00 pm one day before the colonoscopy date, and 2 L PEG was taken 4-6 h before the colonoscopy). The control group was given a single dose regimen of 2 L PEG, taken 4-6 h before the colonoscopy.

The laxative was PEG electrolyte powder (specification: 68.56 g/bag or 137.15 g/bag, Shenzhen Wanhe Pharmaceutical Co., LTD.), whose main component was PEG 4000. Once enrolled, each participant received a uniform education both verbally and in writing. Participants were guided to consume a low-residue diet one day before the colonoscopy (Supplementary Table 1). During bowel preparation, the participants were instructed to take approximately 250 mL PEG every 15 minutes. Colonoscopy was performed by experienced endoscopists (defined as performing >1000 colonoscopies). In addition, all endoscopists were blinded to the randomization status. Both Narrow band Imaging Endoscopy (NBI) and white light endoscope were used to detect polyps. In our study, the colonoscopy withdrawal time was no less than 6 minutes. For those cases with poor bowel preparation, the withdrawal time could be up to 10 minutes.

The bowel preparation quality was assessed using the Boston Bowel Preparation Scale (BBPS) score [14], which was a 4-point scoring system used to assess 3 segments of the colon: the right colon, the transverse colon and the left colon. Score 0: Unprepared colonic segment with mucosa not seen because of solid stool that cannot be cleared. Score 1: Portion of mucosa of the colonic segment seen, but other areas of the colonic segment not well seen because of staining, residual stool, and/or opaque liquid. Score 2: Minor amount of residual staining, small fragments of stool, and/or opaque liquid, but mucosa of colonic segment seen well. Score 3: Entire mucosa of colonic segment seen well, with no residual staining, small fragments of stool, or opaque liquid. The BBPS score was evaluated independently by two endoscopists, and disagreements were resolved by discussion until a consensus was reached. Adequate bowel preparation was defined as a score of ≥2 on all colon segments. Bowel cleanliness was divided into excellent (total score: 8-9), good (total score: 6-7, each segment ≥ 2), fair (total score: 3-5, or total score: 6-7 but any segmental score < 2), and poor (total: score 0-2). The examples of original figures on the cleansing level of colonoscopy were presented in the supplementary Fig. 1.

The primary outcome was the rate of adequate bowel preparation (RABP). The secondary outcomes included BBPS score; polyp detection rate (PDR); cecal intubation rate (CIR); and adverse reactions during bowel preparation, including dizziness, weakness, nausea, vomiting, abdominal pain, abdominal distension, and anal pendant expansion. Reports of adverse events were collected by telephone within two weeks after colonoscopy.

Based on current research, the RABP in normal-weight people is approximately 80%, while that in people with high BMIs is less than 70%. Therefore, we assumed that the RABP for high-BMI individuals using 3 L PEG would be approximately 80%. The α value was set at 0.05, the β value was set at 0.2, and the lost to follow-up rate was set at 0.2. The sample size calculated by the professional sample size calculation tool (Medsci App 5.6.4) was 696 cases.

Participants from each center were randomly and equally assigned to the 2 L and 3 L groups according to the block randomization schedule, with a block size of 4. The random number tables were independently generated by the China Evidence-based Medicine Center, West China Hospital of Sichuan University, using SAS 9.4 as the generation tool.

IBM SPSS 26.0 was used for statistical analysis. Qualitative data were expressed as frequencies (percentages) and were compared using Person’s χ2 test or Fisher’s exact test, as appropriate. Quantitative data were reported as the mean with standard deviation (SD), or median with interquartile range (IQR). Normally distributed quantitative data were analyzed with a T test, while nonnormally distributed quantitative data were compared using the Mann–Whitney U test. A value of P<0.05 was used for all statistical analyses. Participants were divided into subgroups based on BMI classification s[15] and comorbidities. Moreover, a logistic regression model was used to evaluate risk factors for inadequate bowel preparation.

As shown in Table 1, males accounted for 69.0% in 3 L split-dose group and 72.9% in 2 L group, respectively. The average height and weight were 166.4 ± 8.1 cm and 74.4 ± 10.13 kg, respectively, and the BMI was mainly concentrated in the 25-29.9 kg/m² range. Regarding education levels, most of the patients had a high school education (226, 32.5%) or an undergraduate diploma (351, 50.5%). In addition, 164 (23.6%) and 130 (18.7%) patients had a history of smoking or drinking, respectively. The main comorbidities were hypertension and constipation, accounting for 127 (18.3%) and 117 cases (16.8%) among the total individuals. A total of eight patients with compensatory cirrhosis were included, with a total Child-Pugh score of 5, belonged to class A. As for operations, 16 people in the 3 L group (4 appendectomies, 3 cholecystectomies, 9 gynecological operations), and 14 people in the 2 L group (2 appendectomies, 7 cholecystectomies, 5 gynecological operations), respectively, had the history of abdominal operation. No significant difference was found in all comparison between 3L and 2L groups.

3 L split-dose PEG group achieved a higher RABP in total colon than 2 L PEG group (81.2% vs. 74.9%; P = 0.045). For BBPS score, individuals in 3 L PEG group got a higher score than those in 2 L group at the left colon2.33 ± 0.62 vs. 2.18 ± 0.64 (P = 0.003), transverse, right, and total colon in the 3 L versus 2 L group were, 2.41 ± 0.55 vs. 2.28 ± 0.58 (P = 0.002), 1.98 ± 0.51 vs. 1.91 ± 0.56 (P = 0.070) and 6.71 ± 1.15 vs. 6.37 ± 1.31 (P < 0.001), respectively. In addition, there was a significant difference in the composition of bowel-cleansing grades between the two groups (P = 0.006), and patients using 3 L spilt-dose regimen were more likely to obtain excellent bowel preparation.

Polyps were detected in 214 patients in 3 L split-dose group with a total number of 735 and 183 patients in 2 L PEG group, with a total of 696 polyps. The median number of polyps detected in the 3 L group was 1 (0-3) and that in the 2 L group was 1 (0-2); the difference was statistically significant (P = 0.030). The PDR of the 3 L group was higher than that of the 2 L group: 62.0% and 52.9% (P = 0.015). The CIRs were nearly 99.7% in both groups. As for other endoscopic findings, we found submucosal tumor in 6 patients of the 3 L PEG group, and in 5 patients of the 2 L PEG group. Besides, colon diverticulum was detected in 17 and 26 patients in the 3 L and 2 L PEG group, respectively (Table 2).

28 individuals in the 3 L group and 22 in the 2 L group were lost to follow up, without data on adverse reactions. No serious adverse events requiring medical intervention occurred in either group. As shown in Table 3, participants in the 3 L PEG group were more likely to feel nausea than those in the 2 L PEG group (30.8% vs. 19.3%; P = 0.001), but most of the cases were mild. The two groups were comparable regarding the incidence of other adverse events, including dizziness (8.5% vs. 9.0%, P = 0.876), vomiting (14.8% vs. 12.1%, P = 0.330), abdominal pain (16.4% vs. 11.5%, P = 0.078), abdominal distension (30.5% vs. 24.0%, P = 0.064), weakness (11.6% vs. 11.5%, P = 0.966), and anal pendant expansion (16.8% vs. 12.9%, P = 0.147).

Subgroup analysis based on BMI classifications by WHO criteria showed that overweight individuals (BMI 25-29.9 kg/m²) in the 3 L split-dose group had a higher RABP than those in the 2 L groups (82.9% vs. 72.2%, P = 0.006). However, in relatively normal (BMI 24-24.9 kg/m²) and obese individuals (BMI ≥ 30 kg/m²), the RABP was similar between the two regimens (Table 4). For individuals with constipation, the 3 L split-dose regimen was superior to the 2 L regimen in ARBP (P = 0.044). No significant differences were observed for subgroups based on hypertension (P = 0.704) and diabetes (P = 0.064).