Letter to the Editor: Reconsidering water based boosters for small bowel capsule endoscopy

Abstract

Ferreira et al’s in the World Journal of Gastrointestinal Endoscopy conducted a prospective, multicenter, quasi-randomized trial comparing three intraprocedural booster regimens for small bowel capsule endoscopy: Polyethylene glycol, polyethylene glycol with ascorbic acid, and still water. The study has several strengths including the prospective, multicenter design with efforts to minimize confounding, use of the validated Small Bowel Cleansing Assessment and Report scoring system with excellent interobserver agreement, and assessment of patient tolerability. Key limitations include quasi-randomization, unequal group allocation, a wide non-inferiority margin, and the lack of formal confidence interval-based analysis. In agreement with recent studies such as the PrepRICE trial, the CROHN PREP trial, and a recent umbrella meta-analysis, the findings suggest that booster timing may matter more than the agent used, though reduced cleansing in the distal small bowel with water remains a concern. Future non-inferiority trials are needed before water-based boosters can be adopted as standard practice.

Key Words: Small bowel capsule endoscopy; Bowel preparation; Polyethylene glycol; Booster regimen; Small bowel cleansing; Diagnostic yield

Core Tip: Ferreira et al reported that a still water booster for small bowel capsule endoscopy achieved comparable bowel cleansing, diagnostic yield, and completion rates to polyethylene glycol based boosters, while significantly improving tolerability. We note several methodological strengths and limitations that may warrant closer consideration when interpreting the findings. We also place these results in the context of recent randomized trials, meta analyses, and current society guidelines, which continue to offer mixed recommendations regarding optimal SBCE preparation and booster strategies.

TO THE EDITOR

We read with interest the article by Ferreira et al[1], “Bowel Preparation With Still Water for Small Bowel Capsule Endoscopy: Less Is More” published in the World Journal of Gastrointestinal Endoscopy. This prospective, multicenter, quasi-randomized study examines the optimal booster strategy for small bowel capsule endoscopy (SBCE), an important and unresolved issue in gastroenterology. The authors’ findings provide valuable insight into this ongoing debate.

CURRENT GUIDELINES FOR SBCE BOWEL PREPARATION

Current guidelines recommend bowel preparation before SBCE, with The European Society of Gastrointestinal Endoscopy providing the most specific recommendation of a low fiber diet, a 12-hour fast, and 2 L of polyethylene glycol (PEG)[2]. However, the optimal bowel preparation regimen for SBCE lacks consensus[3]. Figure 1 outlines the guidelines for SBCE from several major gastroenterology organizations[4,5]. Although several meta-analyses have demonstrated improved mucosal cleansing with PEG, these improvements have not consistently translated into higher diagnostic yield[6-8]. Purgative regimens are also associated with poorer patient tolerability[7]. Notably, none of these guidelines specifically address the use of intraprocedural “boosters”. Boosters are purgatives or fluids administered after the capsule has entered the small bowel, rather than before the procedure. Theoretically, boosters deliver the cleansing agent directly ahead of the advancing capsule, which can more effectively clear bile and debris from the mucosal surface at the time of image capture. Emerging evidence strongly supports this approach: Marmo et al[4] demonstrated that the time interval between the last dose of purgative and SBCE was the critical determinant of cleansing quality, and the PrepRICE trial[3] confirmed that intraprocedure regimens enhanced visualization, reduced transit time, and improved diagnostic yield compared with night-before preparation. The umbrella meta-analysis by Io Lei et al[8] further corroborated that same-day dosing showed the greatest cleansing benefit, concluding that timing of administration is more critical than the laxative’s dosage.

Figure 1 Current guidelines on bowel preparation for small bowel capsule endoscopy[4,5]. ESGE: European Society of Gastrointestinal Endoscopy; NASPGHAN: North American Society for Pediatric Gastroenterology, Hepatology and Nutrition; ACG: American College of Gastroenterology; AGA: American Gastroenterological Association; BSG: British Society of Gastroenterology; JGES: Japan Gastroenterological Endoscopy Society; ISG: Indian Society of Gastroenterology; KSGE: Korean Society of Gastrointestinal Endoscopy; PEG: Polyethylene glycol; SBCE: Small Bowel Capsule Endoscopy.

In this prospective, multicenter, quasi-randomized study, Ferreira et al[1] evaluated 261 patients undergoing SBCE at two Portuguese university affiliated hospitals between March 2023 and September 2024. The investigators adopted a booster based strategy rather than a traditional bowel preparation strategy, focusing on the effect of an intraprocedural booster administered after capsule entry into the small bowel. All participants received a standardized baseline preparation consisting of a clear liquid diet, a 12-hour fast, and 100 mg of simethicone with water at the time of capsule ingestion. Once small bowel entry was confirmed, patients were assigned to receive one of three booster regimens: 1 L of PEG, 1 L of PEG with ascorbic acid, or 1 L of still water. Figure 2 compares traditional bowel preparation steps with the booster-based strategy used in Ferreira et al[1]. Investigators assessing the studies were blinded to group allocation. The primary outcome was bowel preparation quality, assessed using the validated Small Bowel CLEansing Assessment and Report (SB-CLEAR) score. Secondary outcomes included complete examination rate, diagnostic yield, small bowel transit time, and patient-reported tolerability. No significant differences were observed among the groups in these outcome measures. Notably, however, patient tolerability was significantly better in the still water group[1].

Figure 2 Traditional bowel preparation steps vs booster based strategy. SBCE: Small Bowel Capsule Endoscopy; PEG: Polyethylene glycol.

STRENGTHS AND LIMITATIONS

This study is well designed and has several methodological strengths. The prospective, multicenter design reduces the risk of center-specific bias and increases generalizability. The SB-CLEAR scoring system was an appropriate tool, as it was developed specifically to address the heterogeneity introduced by the use of multiple grading systems. This is particularly important given that inconsistency in cleansing assessment methods has been a major limitation in prior studies[8]. Using SB-CLEAR, there was excellent interobserver agreement, strengthening the reliability of the scoring and the reported findings[1].

In addition, the study design effectively isolated the effect of the booster regimen itself. A real-time viewer confirmed that the capsule had reached the small bowel before the booster was administered, allowing for precise timing and minimizing confounding. All patients underwent the same baseline preparation and received an equal volume of booster (1 L), which isolated the effect of the solution composition and removed confounding effects of volume. The investigators further strengthened the reliability of the findings by blinding capsule readers to group allocation and using dual independent review of each examination.

The study also considers clinically meaningful and patient focused outcomes. The inclusion of patients with a broad range of SBCE indications, including suspected and established Crohn’s disease, celiac disease, and suspected small bowel bleeding, expands the applicability of the findings to patients encountered in routine clinical practice. The inclusion of patient-reported tolerability as a formal outcome is valuable. As SBCE is increasingly used for longitudinal disease monitoring, especially in Crohn’s disease, patient experience and the burden associated with bowel preparation are increasingly relevant considerations[9]. The improved tolerability observed with still water without compromise in diagnostic performance highlights the utility of a simplified preparation strategy.

Despite these strengths, there are important weaknesses that should be considered. The largest weakness lies in the study design of how patients were assigned to groups. Rather than an equal number of participants in each group, they based the groups on the last digit of the patient identification number, which resulted in a 4:4:2 ratio among groups. The smaller water group consisted of 72 participants, and the PEG groups 1 and 2 consisted of 96 and 92 participants respectively. The reduced statistical power to detect differences involving the water protocol. Assigning patients based on identification number is not true randomization as there is a lack of allocation concealment. Researchers can predict which group the patient will enter based on identification number, which in turn can introduce selection bias[10-13]. There is no justification provided by the authors for this unequal ratio, or whether the power calculations were adjusted to the reduced size of the water group.

This study satisfied the predefined non-inferiority criteria, however the margin is relatively wide at 15% non-inferiority. Given the clinical importance of bowel preparation, this margin may be considered overly permissive. The water group achieved an adequate preparation rate of 81.9%, compared with 91.4% in the PEG plus ascorbic acid group, representing a 9.5% absolute difference favoring PEG plus ascorbic acid. Technically this falls within the predefined non-inferiority threshold but could still hold as clinically meaningful in practice. However, only P values from χ² comparisons were reported, and a non-significant P value from a superiority-style χ² test does not, by itself, demonstrate non-inferiority. Establishing non-inferiority formally requires a one-sided confidence interval (or the confidence interval for the risk difference) to fall entirely within the predefined margin. Because the observed difference was 9.5 percentage points in favor of PEG plus ascorbic acid relative to the 15% non-inferiority margin, the confidence interval is necessary for determining whether the true difference remains within the acceptable range for non-inferiority. A sufficiently wide interval could cross the predefined margin and would make the non-inferiority conclusion less certain. This concern is compounded by the smaller size of the water arm (n = 72), combined with the absence of a reported power or sample-size calculation for the 4:4:2 allocation. Given the limited sample size in the comparator arm, the study may have been underpowered for the non-inferiority comparison, which further limits confidence in this conclusion.

The data from the study found that there was a significantly lower SB-CLEAR score observed in the third tertile with water preparation (2.3 vs 2.5-2.6; P = 0.021). This was acknowledged but not emphasized by the authors, although this finding could be important clinically, and it has a plausible mechanistic basis. If still water cleanses primarily through a volumetric mechanism, increasing luminal fluid and stimulating peristalsis rather than remaining osmotically active in the lumen, its effect would be expected to weaken as water is progressively absorbed across the proximal small bowel. This leaves comparatively less cleansing benefit by the time the capsule reaches the distal ileum, which is consistent with where the water arm had poorer bowel cleansing. Because the distal ileum is a common pathological site in Crohn’s disease and other small bowel disorders, bowel cleansing is important in this region to maintain optimal visualization and limit missed abnormalities. Although the average score remained above the accepted threshold of 2, it is important to account for individual variation and the possibility that the average can mask patients who experience inadequate visualization of the distal small bowel. The PrepRICE trial[3] emphasizes the importance of improved cleansing as it demonstrated that increased cleansing in the second and third tertiles increased diagnostic yield and that ability of physicians to detect angioectasia.

Additionally, only the PillCam SB3 system was used in this study, which limits application to other capsule endoscopy platforms. Patients receive capsule endoscopy for various clinical indications, but this study lacked sufficient power to evaluate subgroup differences. This is important as previous research suggests that bowel purgatives may be more beneficial in patients with Crohn’s disease[8]. There was also incomplete reporting of doperidone use and endoscopic capsule placement, which both affect capsule transit and completion rates, introducing a confounding variable. Lack of patient blinding could have affected subjective measures like comfort, treatment compliance, and tolerability. Most importantly, the authors acknowledge their results as hypothesis-generating with proper recognition that additional studies are needed to confirm these findings.

OTHER EMERGING RESEARCH ON SBCE BOWEL PREPARATION

The findings of Ferreira et al[1] are consistent with several prior studies that have questioned the benefit of purgatives before SBCE. Lamba et al[6] in 2022 found no statistically significant difference in P2 lesion detection or visualization quality between clear fluids, 2 L PEG, and 1 L PEG plus ascorbic acid in patients with suspected small bowel bleeding. However, patients were seen to tolerate clear fluids better, making preparation more comfortable. Similarly, Hookey et al[14] in 2017 reported no advantage of PEG or sodium picosulfate over clear fluids alone, and Rahmi et al[7] in 2022 showed that although PEG improved cleansing quality, it did not significantly increase diagnostic yield.

Alternatively, there are also several studies that support purgative preparation. An umbrella meta analysis, including 53 primary studies and 9636 patients, found that PEG based preparations significantly improved cleansing and increased diagnostic yield, with the greatest benefits in patients with Crohn’s disease[8]. Additionally, the greatest benefits were observed with same-day and post-ingestion preparation strategies, which was a similar timing strategy used by Ferreira et al[1]. This raises the possibility that the relatively favorable performance of the water group may be attributed to the timing of the booster regimen, not fully by the absence of purgative preparation.

The CROHN PREP trial[9] is relevant as it directly compared patients with Crohn’s disease that received simplified water based preparation to those that received 2 L PEG. The study found no statistically significant differences in bowel cleanliness, diagnostic yield, or completion rates. They also observed increased patient acceptability with water preparation, which is similar to Ferreira et al[1]. The proposed mechanism by both the investigators and current literature is that high volume water intake can increase gastric emptying and intestinal peristalsis, which promotes bowel cleansing through a volumetric rather than osmotic mechanism. Previous studies have also demonstrated 4 L clear liquids was non inferior to PEG for capsule endoscopy image quality[15].

CONCLUSION

In conclusion, the study by Ferreira et al[1] adds to the growing body of literature that the effectiveness of post-ingestion boosters in the scope of capsule transit and small bowel cleaning could be more volume dependent than on the purgative properties of the solution itself, if administered after the capsule reaches the small bowel. However, this study contains several methodological limitations including, including quasi randomization, unequal group allocation, wide non inferiority margin, and the absence of formal non inferiority analysis. Interpretation of the findings should be done so carefully until confirmed further. The results should also be considered in the context of the PrepRICE trial, which demonstrated improved diagnostic yield with intraprocedure purgatives, as well as the umbrella meta analysis supporting PEG based preparation for cleansing adequacy. Additional large scale, properly randomized non inferiority trials with formal statistical analyses and indication specific subgroup evaluation are needed before water based boosters can be considered a standard alternative to PEG based regimens.