Secondary Logo

Journal Logo

Original Articles: Gastroenterology

Clear Liquid Versus Low-fibre Diet in Bowel Cleansing for Colonoscopy in Children: A Randomized Trial

Mytyk, Aleksandra; Lazowska-Przeorek, Izabella; Karolewska-Bochenek, Katarzyna; Kakol, Dariusz; Banasiuk, Marcin; Walkowiak, Jaroslaw; Albrecht, Piotr; Banaszkiewicz, Aleksandra

Author Information
Journal of Pediatric Gastroenterology and Nutrition: May 2018 - Volume 66 - Issue 5 - p 720-724
doi: 10.1097/MPG.0000000000001832
  • Free


What Is Known

  • Bowel preparation is crucial for proper diagnostic and therapeutic value of colonoscopy.
  • There is no standardized bowel preparation schedule for children.

What Is New

  • Comparison of the effectiveness of a clear liquid and a low-fibre diet (in addition to macrogole solution) for bowel preparation for colonoscopy in paediatric patients.
  • Contribution towards determining whether the European Society of Gastrointestinal Endoscopy recommendations for adults might be effectively applicable for children, as this might improve colonoscopy preparation in paediatric population.

Since colonoscopy is a standard diagnostic and therapeutic procedure in paediatric gastroenterology, one of the hot topics in the colonoscopy field is bowel preparation. There are many reasons for this. Only a well-prepared colon enables proper diagnostic evaluation and therapeutic interventions during colonoscopy. This is why many observational and interventional studies have been conducted to find the ideal bowel preparation schedule that should be easy, safe, quick, and comfortable for patients. The safety of the procedures also depends on the quality of the bowel preparation. Moreover, it has been noted that an ill-prepared bowel is likely to lead to a repeated colonoscopy, which in the paediatric population means that not only bowel preparation and colonoscopy but also general anaesthesia must be repeated. It also doubles the cost of the procedure.

Because many thousands of colonoscopies are performed each year, there is no 1 standardized bowel preparation schedule that would guarantee an adequate and tolerable bowel cleanness in the paediatric population (1,2). To date, a number of medications have been studied, including macrogoles, senna, bisacodyl, citrates, and others, in 1, 2, 3, or 4-day regimens (1). A clear liquid diet for 1 or 2 days preceding colonoscopy is usually prescribed in most bowel preparation schedules (1). The effectiveness of these schedules varies from 57% to 100% (1). Tolerability, when assessed, is also highly variable.

In 2013, European Society of Gastrointestinal Endoscopy (ESGE) published new recommendations for bowel preparation for colonoscopy, that dedicated for adults (3). Briefly, ESGE recommends 4 L of polyethylene glycol (PEG) solution as the most safe for routine bowel preparation either in split regimen or same-day regimen (in the case of afternoon colonoscopy). Moreover, the delay between the last dose of the PEG solution and the colonoscopy should be minimized and should not exceed 4 hours. Additionally, ESGE states that a low-fibre diet is not inferior to a clear liquid diet on the day preceding colonoscopy.

Since a great majority of colonoscopies in children are performed under general anaesthesia, most of the ESGE recommendations do not apply to children, except for those concerning the diet in bowel preparation for colonoscopy. The aim of the study was to compare the effectiveness of a clear liquid diet and a low-fibre diet, in addition to macrogole solution, for bowel preparation for colonoscopy in paediatric patients, determine whether adult ESGE recommendation in this respect is equally applicable to children. Additionally, we asked children and parents to assess the overall tolerability of the two schemes of bowel preparation.


We conducted a prospective, randomised single-blind trial at the Department of Paediatric Gastroenterology and Nutrition in Warsaw, Poland, between September 2014 and September 2016. Eligible patients included those referred for colonoscopies between the ages of 6 and 18 years. Excluding criteria included chronic heart or kidney disease and a history of gastrointestinal tract surgery.

Using a computer-generated randomisation list, the patients were randomly divided into 2 groups: the first received a clear liquid diet and the second a low-fibre diet on the day before colonoscopy, both groups allowed to eat and drink ad libitum within their prescribed diets. Children allocated to the clear liquid diet were allowed to drink mineral water, tea, and clear juice without pulp, that is, apple juice or grape juice. Those children on a low-fibre diet were allowed to drink milk and eat milk products, some soups, bread and rolls, sandwiches, meat, fish, eggs, pasta, honey, and others.

In the afternoon, all participants were asked to drink a standard PEG solution (Fortrans Ipsen Pharma, France, containing PEG 4000 with electrolytes [PEG-ELS]: potassium chloride, sodium chloride, sodium bicarbonate, sodium sulphate anhydrous) at a dose of 66 mL/kg to a maximum of 4 L. The endoscopy was performed the next day in the morning under general anaesthesia after 6 hours of fasting. All of the children were asked not to consume red-coloured food or food containing little seeds for 3 days beforehand. All patients received written instructions on how to prepare the bowel for colonoscopy, including diet and macrogole regimen.

The main outcome was the effectiveness of the bowel cleansing, assessed according to the Boston Bowel Preparation Scale (BBPS) in the 2 study groups. We also assessed segmental BBBS. The study outcomes were measured immediately after the end of colonoscopy by both doctors and nurses. Additionally, the bowel preparation tolerance was assessed on a 10-point visual analogue scale (VAS) scale by parents and children. We also asked about the side effects of bowel preparation.

The BBPS is graded from 0 to 9. It is a 4-point (0–3 point) scoring system applied to each of 3 regions of the colon: right colon (cecum and ascending colon), transverse colon (with hepatic and splenic flexures), and left colon (descending colon, sigmoid colon and rectum). The BBPS has already been validated (4). All colonoscopies were performed by 3 staff endoscopists with more than 10 years’ experience. The doctors and nurses who performed the procedure were blind as to which group the patient was assigned.

The study was powered to detect a difference of 20% between groups with good BBPS scores (≥5). The α = 0.05 and the β = 0.20 were assumed. The sample size was 100 patients in each group.

The Student t test was used to compare continuous variables with normal distribution. Otherwise, Mann-Whitney U test was applied. z Test of proportion was used to compare proportions of patients with BBPS ≥ 5, as assessed by doctors and nurses. Categorical variables were evaluated by χ2. Spearman coefficient was used to test correlation between time and BBPS score. A P value of <0.05 was considered statistically significant. Data were analysed using Statistica 12 (Statsoft, Tulsa, OK).

All parents and children signed the written consent form before participating in the study. The study was approved by the Ethics Committee of Clinical Investigation of the Medical University of Warsaw, Poland (No 242/2013). The study protocol was registered at (NCT02102373).


In total, we enrolled 184 patients. Of those, 96 were allocated to the clear liquid group, and 88 received the low-fibre diet. The mean age of both groups was 15 years. Most of the children (124/184) were diagnosed with inflammatory bowel diseases (IBD). There were no differences between the 2 study groups regarding age, mean weight, and sex. The details of the characteristics of the study groups are shown in Table 1.

Baseline characteristics of study groups

The median total BBPS score was assessed as 7 points in the study groups, both by doctors and nurses (P = 0.09). More patients were prepared sufficiently well (BBPS ≥ 5 points) for colonoscopy when assessed by doctors compared with nurses: 165/173 (95.4%) and 139/158 (88%), respectively, P = 0.007. Table 2 presents total BBPS scores in the 2 study groups. There were no differences between study groups in segmental BBPS scores (Fig. 1). In the low-fibre diet group, we found a correlation between the PEG-ELS drinking time and the median BBPS score for the descending colon and rectum (r = 0.28, P = 0.01 in doctors’ assessment and r = 0.31, P = 0.01 in nurses’ assessment).

Total Boston Bowel Preparation Scale score in the study groups
Segmental Boston Bowel Preparation Scale (BBPS) scores for clear liquid diet and low-fibre diet groups.

In the low-fibre diet group, we found an inverse correlation between the time between PEG-ELS drinking and colonoscopy and the median total BBPS score when assessed by nurses (r = -0.223, P = 0.049) but not by doctors (r = −0.149, P = 0.19). Adverse effects are presented in Table 3. Tolerability of bowel preparation was assessed as good (6 points on VAS scale). There were no differences in VAS scores between the clear liquid diet and low-fibre diet groups in parents’ (P = 0.31) and children's (P = 0.66) assessments.

Adverse effects in the study groups


In this prospective study, we found no difference between a clear liquid diet and a low-fibre diet combined with PEG-ELS solution the day before colonoscopy in children aged 6 to 18 years. Our results are in line with the results of a recently published meta-analysis by Song et al, which indicated no difference between the 2 dietary regimes in terms of excellent or good bowel preparation (relative risk [RR] 1.01, 95% confidence interval [CI] 0.91–1.13, P = 0.39) in the adult population (5). Our study is the first to confirm this relatively new finding in the paediatric population. The fact that a low-fibre diet is not different from a clear liquid diet in colonoscopy preparation is even more important for children than for adults because children often protest against the procedure and do not want to agree to any unpleasant regimen or procedure even when offered an explanation for why they are performed. Therefore, a less-restrictive diet regimen has a higher chance of compliance.

Overall, we achieved good bowel cleanness (BBPS ≥ 5) in the majority (95.4%) of patients. Our results are similar to those achieved by Park et al, who assessed two diet regimens in addition to PEG-ELS (6). They found an overall good bowel preparation, as assessed by the Ottawa scale: 2.97 ± 2.0 for a clear liquid diet and 2.46 ± 1.78 for a low-fibre diet; P = 0.06. Our results are as good or better than those in most studies, which assess bowel preparation using PEG or PEG-ELS in children (1). A direct comparison is, however, difficult due to the variety of doses of PEGs used in these studies. Moreover, bowel preparation lasted from 1 to 4 days and additional medications were used in some cases. We believe that the very good bowel preparation we achieved in the majority of our patients is a result of the short period for drinking PEG-ELS (<5 hours in the study groups). Our findings confirm the results of 2 studies with a short bowel preparation period. Adamiak et al, in their retrospective study, noted effective bowel cleaning in 93% of patients who were drinking PEG for 2 hours in the afternoon the day before colonoscopy (7). In a prospective study by Abbas et al, patients took PEG for a few hours the day before colonoscopy, and 93.5% of them had good bowel cleaning (8).

We achieved good bowel preparation in the majority of our patients when we used a lower dose of PEG-ELS than suggested in the paediatric joint North American Society for Pediatric Gastroenterology, Hepatology and Nutrition (NASPGHAN) and American Society for Gastrointestinal Endoscopy recommendation (66 mL/kg compared with 80 mL/kg, respectively) (2). This is a very important finding because one of the well-known problems in bowel preparation in children, but also in adults, is difficulty ingesting a sufficient volume of preparation formula. It seems that a smaller amount of PEG-ELS drunk within shorter time works better than a larger volume, which can be drunk over a longer time. The results of our study showed no difference in proportion of patients with BBPS score of ≥5 in patients with and without diarrhoea when assessed both by doctors (P = 0.6 for clear liquid diet, P = 0.9 for low fibre diet) and nurses (P = 0.8 for clear liquid diet, P = 0.9 for low fibre diet). There is no paediatric study that evaluates whether diarrhoea has an impact on bowel preparation for colonoscopy. In Nett et al's study (9), the authors analysed the bowel preparation quality for colonoscopy in adults with IBD and found that advanced age, male gender, diabetes, obesity, multiple comorbidities, some drugs, immobility, inpatient status, and lower education level, but not the disease activity (including diarrhoea) were predictors of suboptimal bowel preparation. Their study showed no difference in bowel preparation between patients with and without IBD.

To objectify our results, we decided to ask both doctors and endoscopy nurses to assess the BBPS of all colonoscopies. The results we obtained from doctors and nurses are strongly internally consistent. When we started our study, we did not know the results of a very small Schindler et al study (published in May 2016) which demonstrated a high interobserver reliability of the BBPS scores assigned by the endoscopy nurses, gastroenterology faculty and fellows. The results of our study suggest that involvement of endoscopy nurses’ in the BBPS evaluation of bowel preparation could increase the efficiency of quality improvement efforts within an endoscopy unit.

In our study, we did not find any correlation between the delay between the last dose of PEG and the start of colonoscopy and BBPS, irrespective of study group. This finding is in opposition not only to the ESGE guidelines but also to the results of several studies on adult patients which underline that the delay between the last dose of bowel preparation and the start of colonoscopy should be minimized and should be no longer that 4 to 6 hours (3,10,11). In our study, this delay was much longer (mean 13 hours) because all the colonoscopies were performed in the morning and under general anaesthesia. No previous paediatric study evaluating preparation for colonoscopy assessed this outcome. We cannot give a reasonable explanation for why the results of the BBPS in the presented study are different than those performed in adults. We can, however, speculate that the delay between the last dose of bowel preparation and the start of colonoscopy may be less important than the short drinking time for bowel cleansing.

In our study, we found a similar tolerability of the two diet regimens, and they were assessed as 6 points on a 10-point VAS scale. Parents’ and children’ assessments were similar (P = 0.6). In a meta-analysis that aimed to assess randomized controlled trials comparing clear liquid and low-fibre diets in bowel preparation for colonoscopy in adults, 5 of 7 studies also assessed patient tolerance to recommended dietary regimes; in these studies, 1078 participants were enrolled (12–16). The heterogeneity of these studies was noted; however, the moderate level of evidence suggested that there was a near-significant difference, with more patients reporting tolerance in the low-residue diet group (RR 1.06; 95% CI 1.02–1.11).

The main advantage of our study is showing that there was no difference between a low-fibre diet and a clear liquid diet when given with PEG-ELS on the day preceding colonoscopy in the paediatric population. We hope our results will provide a basis to simplify and liberalize recommendations on bowel preparation for colonoscopy in children. Our study has some shortcomings. The median age of the study participants was 15 years. We did not assess the group of children below the age of 6. This is, however, the most challenging group of patients, and we are convinced that they require a separate study protocol. We used a nonvalidated scale to assess the quality of bowel preparation; however, no scale has yet been validated in the paediatric population. We decided to use BBPS, as it is the simplest and most recommended in Europe and has already been validated in the adult population.


In this randomised, single-blind study, we found no difference between a low-fibre diet and a clear liquid diet, in addition to PEG-ELS, in bowel preparation for colonoscopy in children.


The authors thank Beata Winnicka and Malgorzata Nowakowska, endoscopic nurses, for their kind help during the procedures.


1. Pall H, Zacur GM, Kramer RE, et al. Bowel preparation for pediatric colonoscopy: report of the NASPGHAN endoscopy and procedures committee. J Pediatr Gastroenterol Nutr 2014; 59:409–416.
2. Lightdale JR, Acosta R, Shergill AK, et al. ASGE Standards of Practice Committee. Modifications in endoscopic practice for pediatric patients. Gastrointest Endoscopy 2014; 5:699–710.
3. Hassan C, Bretthauer M, Kaminski MF. Bowel preparation for colonoscopy: European Society of Gastrointestinal Endoscopy (ESGE) guidelines. Endoscopy 2013; 45:142–150.
4. Lai EJ, Calderwood AH, Doros G, et al. The Boston Bowel Preparation Score: a valid and reliable instrument for colonoscopy-oriented research. Gastrointest Endosc 2009; 69 (3 pt 2):620–625.
5. Song GM, Tian X, Ma L, et al. Regime for bowel preparation in patients scheduled to colonoscopy: low-residue diet or clear liquid diet? Evidence from systematic review with power analysis. Medicine (Baltimore) 2016; 95:e2432.
6. Park DI, Park SH, Lee SK, et al. Efficacy of prepackaged, low residual test meals with 4L polyethylene glycol versus a clear liquid diet with 4L polyethylene glycol bowel preparation: a randomized trial. J Gastroenterol Hepatol 2009; 24:988–991.
7. Adamiak T, Altaf M, Jensen MK, et al. One-day bowel preparation with polyethylene glycol 3350: an effective regimen for colonoscopy in children. Gastrointest Endosc 2010; 71:573–577.
8. Abbas MI, Nylund CM, Bruch CJ, et al. Prospective evaluation of 1-day polyethylene glycol-3350 bowel preparation regimen in children. J Pediatr Gastroenterol Nutr 2013; 56:220–224.
9. Nett A, Velayos F, McQuaid K. Quality bowel preparation for surveillance colonoscopy in patients with Inflammatory Bowel Disease is a must. Gastrointest Endoscopy Clin N Am 2014; 24:379–392.
10. Siddiqui AA, Yang K, Spechler SJ, et al. Duration of the interval between the completion of bowel preparation and the start of colonoscopy predicts bowel-preparation quality. Gastrointest Endosc 2009; 69 (3 pt 2):700–776.
11. Seo EH, Kim TO, Park MJ, et al. Optimal preparation-to-colonoscopy interval in split-dose PEG bowel preparation determines satisfactory bowel preparation quality: an observational prospective study. Gastrointest Endosc 2012; 75:583–590.
12. Scott SR, Raymond PL, Thompson WO, et al. Efficacy and tolerance of sodium phosphates oral solution after diet liberalization. Gastroenterol Nurs 2005; 28:133–139.
13. Rapier R, Houston C. A prospective study to assess the efficacy and patient tolerance of three bowel preparations for colonoscopy. Gastroenterol Nurs 2006; 29:305–308.
14. Melicharkova A, Flemming J, Vanner S, et al. A low-residue breakfast improves patient tolerance without impacting quality of low-volume colon cleansing prior to colonoscopy: a randomized trial. Am J Gastroenterol 2013; 108:1551–1555.
15. Delegge M, Kaplan R. Efficacy of bowel preparation with the use of a prepackaged, low fibre diet with a low sodium, magnesium citrate cathartic vs. a clear liquid diet with a standard sodium phosphate cathartic. Aliment Pharmacol Ther 2005; 21:1491–1495.
16. Stolpman DR, Solem CA, Eastlick D, et al. A randomized controlled trial comparing a low-residue diet versus clear liquids for colonoscopy preparation: impact on tolerance, procedure time, and adenoma detection rate. J Clin Gastroenterol 2014; 48:851–855.

Boston Bowel Preparation Scale; bowel preparation; endoscopy

Copyright © 2017 by European Society for Pediatric Gastroenterology, Hepatology, and Nutrition and North American Society for Pediatric Gastroenterology, Hepatology, and Nutrition