Constipation, defined as difficulty or pain on defecation, is a common problem in children which can be challenging to manage. Constipation most commonly first appears amongst toddlers (ages 2 to 4 years) [1] but can also occur in babies often when weaning on to solids [2] and affects up to a third of children [3]. In children, constipation usually occurs without evidence of an underlying disease (functional constipation) and is often the result of stool retention associated with toilet training, changes in diet or routine, stressful events, illness or withholding [4]. Constipation can be a chronic problem in children, sometimes resulting in significant physical morbidity and psychosocial distress [5].

The management of constipation in children involves education, establishing a regular toileting routine, behavioural modification, the treatment of faecal impaction and laxative maintenance therapy [2,6]. There are multiple laxatives that are safe to use in children, however polyethylene glycol based laxatives have become one of the mainstay of therapy as they are effective, well tolerated and easy to administer [7].

Polyethylene glycol is available as either 3,350 or 4,000 g/mol, with or without electrolytes [8]. Polyethylene glycol softens the stool and relieves constipation by essentially holding the water it is ingested with and not via drawing water into the lumen from the body. This retained water softens hardened stools and increases the stool volume which encourages natural peristaltic reflexes [9].

Several meta-analysis [10,11] and a Cochrane systematic review [12] have assessed the efficacy of polyethylene glycols for constipation in children. The Cochrane review suggested that polyethylene glycols are more effective than placebo, lactulose, liquid paraffin and magnesium hydroxide, however the authors recommended caution due to issues associated with heterogeneity and risk of bias [12]. The most recent meta-analysis concluded that polyethylene glycols are an effective therapy for constipation, but were not significantly more effective than other laxatives [11]. These reviews did not distinguish between the different polyethylene glycol formulations and the clinical utility off added electrolytes was not assessed.

A key to the successful management of constipation in children is for the child to be able to take an adequate dose and comply with the regimen. The inclusion of electrolytes can affect the palatability of polyethylene glycol, which on its own is essentially tasteless. What is not known is whether the added electrolytes augments, reduces or has little impact on the clinical effectiveness of polyethylene glycol in the management of constipation in children.

The objectives of this systematic review and network meta-analysis (NMA) were to assess the relative efficacy, safety and tolerability of polyethylene glycol with or without electrolytes in the management of constipation in children. The primary efficacy end point was the mean number of bowel movements per week.

Search methodology

Text word searches were carried out using EMBASE, MEDLINE, MEDLINE in Progress and the Cochrane Database and Systematic Reviews databases covering inception to April 2015. Search terms were (constipation) AND (PEG OR polyethylene OR macrogol OR movicol OR idrolax OR miralax OR transipeg OR forlax OR colyte OR golytely OR isocolan OR nulytely) NOT colonoscopy. Only published randomised controlled trials comparing oral polyethylene glycol with placebo or a comparator laxative in children with constipation were included in this analysis. A separate analysis of data in adults using the same methodology was conducted in parallel and has been published elsewhere [13]. The diagnosis of constipation was based on the Rome I, II or III diagnostic criteria, clinical symptoms or the clinician’s opinion. Bibliographies of all identified relevant studies and reviews were used to perform a recursive search. Only papers published in English were included in the analysis.

Data extraction

Two independent reviewers followed a four-step process for data collection. First, titles and abstracts of the identified citations were assessed according to the research objectives. Next, potentially relevant articles were screened as full texts. For the studies that met the selection criteria one reviewer extracted relevant data using a standardised database and the accuracy of the extracted data was confirmed by a second reviewer. Both reviewers then determined if the study met the inclusion criteria and was to be included in the analysis. Disagreements were resolved by consensus involving two additional reviewers.

The following data were extracted: author; publication year, title; journal; patient population (adult or paediatric); study design; definition of constipation; patient age; characteristics; exclusion/ inclusion criteria; mean duration of constipation prior to study intervention; intent to treat population (ITT); comparability of treatment groups; subanalysis based on age; study duration; study medications, dose and duration of treatment. To standardise the assessment of efficacy, a single objective outcome, the mean number of bowel movements per week (mean plus standard deviation [SD] or standard error [SE]) was used. Where available the assessment after a 2-week treatment period was used, otherwise the value at the end of treatment was used. If means were not reported, medians and interquartile ranges were used. Data was normalised to mean number of bowel movements per week. To assess secondary endpoints: Laboratory data, vital signs, tolerability data and measures of compliance were collected.

The risk of bias was assessed as per the recommendations published in the Cochrane Handbook for Systematic Reviews of Interventions [14].

Data analysis

Studies were grouped according to whether the polyethylene glycol had added electrolytes (PEG+E) or not (PEG) and then according to the comparator (placebo or different active controls).

The statistical methods used for the NMA have been previously published [13]. For all comparisons where there was more than one study available, standard meta-analysis was performed using a Bayesian fixed effects and a Bayesian random effects models. All the available data were combined using a NMA with a Bayesian random effects model fitted to assess the relative effectiveness of PEG and PEG +E [15]. In this analysis, for each simulation the result for active treatment was compared with control. If the mean stool frequency for active treatment was greater than the mean stool frequency for control, this was regarded as active being better than control, for that simulation. Across all simulations the percentage of simulations where active is greater than control was obtained and is reported as a percentage, ‘probability best’. If the two treatments are equivalent, the ‘probability best’ is 50%, while if the active treatment is always superior to the control the probability best is 100%.

Sensitivity analysis was performed by repeating the analysis using only high quality studies with a low risk of bias.

Literature search

Literature searches identified 1,612 potentially relevant abstracts which were reduced to 1,484 after removal of duplicates. 29 publications underwent full-text review and 15 studies were included in the final analysis (Figure 1) [16-30].

Figure 1: Flow diagram for study selection.

Study characteristics

Of the 15 qualifying studies, 9 studies were with PEG; 1 versus placebo and 6 versus other osmotic laxatives and 2 versus stool softeners; 5 studies were with PEG+E, 1 versus placebo, 3 versus other osmotic laxatives and 1 versus bulk forming laxatives; and 1 study made a direct comparison between PEG and PEG+E. Table 1 provides an overview of the individual studies.

Table 1: Summary of the fifteen included studies.

1,384 patients were randomised to receive polyethylene glycol with or without electrolytes or control (placebo or active controls). Five of the 15 studies were assessed as having a low risk of bias [14] (Table 2 and Figure 2).

Table 2: Individual study results included in the network meta-analysis.

Figure 2: Risk of bias. Figure

Patients were predominantly experiencing chronic functional constipation [16-23,25-30] with the exception of one study being amongst patients with spina bifida with chronic neurogenic constipation [24]. Overall there was approximately equal proportion of male and female patients with the average patient age ranging from 2.0 to 11.3 years.

Network meta-analysis

Figure 3 shows the network formed by the different treatments used to manage constipation in this NMA. Across the 15 studies there were 19 connections between the different treatments. Table 2 summarizes the mean number of bowel movements per week for all individual studies.

Figure 3: Network diagram, constipation treatments and direct comparisons included in the analyses (PEG=Polyethylene glycol alone, without electrolytes; PEG+E=Polyethylene glycol with electrolytes).

Table 3 summarizes the results for the standard meta-analysis for direct evidence (Bayesian random effects pairwise meta-analysis). None of these direct comparisons yielded statistically significant differences between the either of the polyethylene glycols and their comparators.

With the inclusion of more data, as accommodated by the NMA, both PEG and PEG+E were found to be significantly more effective than placebo, increasing the mean number of bowel movements per week by 2.3 and 2.2 respectively. Comparisons with the different active controls favoured the polyethylene glycols, but were not statistically significant. The addition of electrolytes did not enhance the efficacy of polyethylene glycol as illustrated by the direct comparison between PEG and PEG+E (Table 3 and Figure 4).

Table 3: Results for the direct head-to-head random effects meta-analysis and network meta-

Figure 4: Pairwise comparisons for PEG and PEG+E from the network meta-analysis (The control group for the direct comparison between PEG+E and PEG is PEG without electrolytes. PEG=Polyethylene glycol alone, without electrolytes; PEG+E=Polyethylene glycol with electrolytes; MD=Difference in mean stool frequency; LCrL=Lower credible limit).

Safety and tolerability

Fourteen of the 15 studies provided evidence on tolerability of polyethylene glycol and five studies provided safety data (laboratory results and/or vital signs), 2 for PEG+E [16,24] and 3 for PEG [17,20,30] (Table 4). No clinically relevant changes in laboratory measures were reported for either PEG or PEG+E. Polyethylene glycol with or without added electrolytes were well tolerated and generally better tolerated than other active constipation therapies (lactulose [16,17,25,29] liquid paraffin [22] and magnesium hydroxide [23,25]). Treatment-related side effects were predominantly gastrointestinal, including abdominal pain, diarrhoea and nausea. The only head-tohead clinical trial of PEG and PEG+E found that PEG was better tolerated than PEG+E with a lower incidence of nausea [26]. Serious treatment-related adverse events were uncommon as were discontinuations due to adverse events.

Table 4: Safety and tolerability of polyethylene glycol.

Compliance, willingness to continue therapy

Only four studies assessed patient compliance or willingness to continue laxative therapy. Of most relevance is the direct comparison of PEG and PEG+E. In this study more children had no difficulties taking PEG than PEG+E (96% vs. 52%, p<0.001). This ease of use benefit was associated with a trend towards improved compliance with PEG (98% vs. 88%, p=0.062) [26]. Two studies demonstrated better patient acceptance and compliance of PEG versus magnesium hydroxide [18,23]. One study demonstrated better compliance with PEG+E than a bulk forming agent (p=0.002) [21].

This NMA provides no evidence for the addition of electrolytes to polyethylene glycol in the management of constipation in children. The added electrolytes does not improve the efficacy, safety or tolerability of polyethylene glycol, but it may reduce compliance.

The benefits of a NMA is that it allows the use of all the available data to address the clinical questions at hand, where standard metaanalysis may have insufficient direct comparisons to interrogate the issue. This is the case with determining the efficacy of polyethylene glycol versus placebo. Two previous meta-analyses have failed to address this important clinical issue [10,11]. However, this NMA demonstrated that both PEG and PEG+E are more efficacious than placebo increasing the number of bowel movements per week by 2.3 and 2.2 respectively. These results are consistent with the Cochrane review where polyethylene glycol with or without electrolytes increased the mean number of bowel movements per week by 2.61 relative to placebo [12].

Both PEG and PEG+E were well tolerated and overall better tolerated than other laxatives such as lactulose. The most common treatment-related side effects were gastrointestinal, which were generally mild and transient in nature. The only head-to-head comparison of PEG and PEG+E, found PEG to be better tolerated with a lower incidence of nausea [26]. No safety signals were identified that would suggest a safety benefit for the added electrolytes in the management of constipation.

Polyethylene glycol alone is tasteless and, when dissolved, colourless. In paediatric practice, it can be masked by dissolving in a drink of the child’s choice. The addition of electrolytes alters the palatability of polyethylene glycol for some children making it more difficult to administer. This was demonstrated in the only direct clinical comparison of PEG and PEG+E. The taste of PEG was rated as superior than PEG+E (p<0.001) with almost all (96%) children having no difficulties with taking PEG compared to approximately half taking PEG+E (p<0.001). This study also indicated that improved palatability has the potential to enhance compliance [26]. Better patient acceptability was also demonstrated for PEG versus magnesium hydroxide [18,23] and in a long-term trial versus previous laxative therapy [31].

Study heterogeneity is a potential limitation for this NMA. There was a wide range of paediatric populations, from infants to adolescents and the assessment period varied between 2 weeks and 6 months. To reduce the impact of heterogeneity stool frequency was assessed at 2 weeks where the data was available, which was the case in 40% (six) of the studies, Even though the dose of polyethylene glycol varied between studies, over half (eight) of the studies used a dose 0.4 to 0.7 mg/kg/day, with the most common mean daily dose being 0.5 mg/kg/ day. Similarly, the age of enrolled patients varied across the studies, however 60% (nine) of the studies were conducted amongst patients with a mean age ranging between 4 and 9 years, with a further 4 studies being conducted amongst patients aged less than 4 years old. Ten of the studies had a high risk of bias. When these studies were excluded in the sensitivity analysis, the results remained essentially unchanged, but they were no longer statistically significant.

This NMA demonstrated that polyethylene glycol with and without electrolytes are effective, well tolerated treatments for constipation in children from the age of 6 months. The addition of electrolytes does not appear to provide any clinical benefits over polyethylene glycol alone. There is no enhancement of efficacy, safety or tolerability. As polyethylene glycol alone is tasteless, it is easier to administer to children and its enhanced patient acceptability has the potential to improve adherence. This analysis supports the ongoing use of polyethylene glycol as a first-line treatment of constipation in children. Formulations without electrolytes should be considered first to aid administration and compliance.

Statistical analysis was provided by Dr. Ann Solterbeck from Statistical Revelations Pty Ltd, whose contributions have been funded by Key Pharmaceuticals Pty Ltd.