A Low-FODMAP Diet in the Management of Children With Functional Abdominal Pain Disorders: A Protocol of a Systematic Review

Supplemental Digital Content is available in the text.

T he worldwide pooled global prevalence of functional abdominal pain disorders (FAPD) in children aged 4-18 years was 13.5% in 2015, among which irritable bowel syndrome (IBS) was the most common disorder (8.8%) (1). The prevalence across studies varied from 1.6% to 41.2%. The available pharmacological and nutritional interventions for the management of FAPD in children are limited. A diet low in fermentable oligosaccharides, disaccharides monosaccharides, and polyols (FODMAPs) is widely used in adults and children with functional abdominal pain, despite limited available evidence (2). The rationale for use of a low-FODMAP diet is based on the assumption that a decrease in the short-chain fermentable carbohydrate load both prevents the osmotic effect of FODMAPs, resulting in a decrease in small intestine water volume, and limits the exaggerated fermentation of FODMAPs by colonic microbiota and associated gas production, which may alleviate recurrent abdominal pain (3). A typical low-FODMAP diet should consist of 3 stages: an elimination of most FODMAPs (oligosaccharides [fructans and galactooligosaccharides], polyols [sorbitol or mannitol], excess fructose, and lactose) to the established cutoff levels (4), a rechallenge to identify FODMAP triggers of symptoms with thresholds, and the development of an individual modified low-FODMAP diet based only on identified FODMAP triggers (5). However, in the pediatric population, a simplified approach (i.e., elimination of only some food groups with high FODMAP content or a few targeted FODMAPs-a bottom up approach) is commonly used to avoid over-restriction (5,6). For a summary of definitions of low-FODMAP diets reported in various publications in the literature, see Table 1.

What is Known
• A typical low-FODMAP diet is one of the well-documented and recommended treatments in the management of adults with irritable bowel syndrome. • Evidence for use of a low-FODMAP diet in the management of functional abdominal pain disorders (FAPD) in children is limited.

What is New
• This systematic review of rigorous methodological design will update current evidence on the efficacy and safety of using a low-FODMAP diet in children. • This review will assess the quality of recent randomized controlled trials performed in children with FAPD.

Translational Impact
The results of this systematic review may guide future clinical practice guideline development and treatment in regard to dietary management of children with FAPD.
also found to be no effect with use of a lactose-free diet (2 RCTs) (20,21). However, a few more studies have since been published in the last 2 years. We aim to systematically update evidence on the efficacy and safety of using a low-FODMAP diet (preferably a 3-step low-FOD-MAP diet but also only a strict low-FODMAP diet or restriction of individual FODMAPs) for the management of children with FAPD.
Each low-FODMAP diet or individual FODMAP restriction will be assessed separately.

METHODS
This protocol follows the Preferred Reporting Items for Systematic review and Meta-Analysis Protocols (PRISMA-P) guidelines A fructose malabsorption diet A dietary strategy with the avoidance of foods with high free fructose (>0.5 g/100 g fructose in excess of glucose) and short-chain fructans (>0.5 g per serving) content, limitation of the total fructose load (with or without glucose), encouragement of foods in which glucose was balanced with fructose, and coingestion of free glucose with high "free fructose" foods. Sometimes described as an initial ("primitive") version of a low-FODMAP diet.
Croagh et al, 2007 (8) FODMAP (fermentable oligo-, di-, and monosaccharides and polyols) A group of poorly absorbed short-chain carbohydrates and sugar alcohols (polyols), that include fructose and lactose (if fructose and lactose malabsorbtion is/are present), fructans (oligosaccharides of fructose), and polyols (such as sorbitol) (8) Shepherd et al, 2008 (9) A low FODMAP diet A diet which restricts quantities of all FODMAP and in particular fructans, fructose, and foods in which free fructose greatly exceeds free glucose. In a study: diet with exclusion of fruits containing fructose in excess of glucose, fructan-containing vegetables, wheat-based products, sorbitol-containing products, raffinose-containing foods, and lactose containing foods (in case of lactose malabsorption proven with breath hydrogen testing following a load of 50 g).
Tuck et al, 2014 (10) and Whelan et al, 2018 (11) "Individual modified low-FODMAP diet" (10) or "modified-FODMAP diet" (11) A stage of low-FODMAP diet in which patient reintroduces any foods well tolerated and only limit those foods resulting in symptoms (10); restriction is continued [keeping the type and amount of FODMAP intake below the tolerance threshold], but those FODMAP/foods that did not induce symptoms during the second stage are now included (11).

Bottom-up approach
An approach to implement a low-FODMAP diet with a reduction of foods with very large amount or specific FODMAP, then continued restriction to tolerance level (4-8 weeks).

Top-down approach
An approach to implement a low-FODMAP diet with an over restriction of all or most FODMAP to the cutoff levels set by Monash University, then liberalization of diet to tolerance level (4-8 weeks).

Strict low-FODMAP diet (2-6 weeks)
A diet with an exclusion of foods with high or moderate amounts of any of the FODMAP subgroups (excess fructose, lactose, galactooligosaccharides, fructans, mannitol, and sorbitol); followed by the FODMAP rechallenge (in which patient reintroduce individual FODMAP subgroups to determine which FODMAP they tolerate and which trigger symptoms) and adapted FODMAP diet (that only restricts foods and FODMAP sub-groups that are poorly tolerated and only to an extent that is absolutely necessary) Simplified low-FODMAP diet A low-FODMAP diet with a restriction of only frequently consumed very high FODMAP foods and suspected to be major triggers of their IBS symptoms.

An "expanding' low-FODMAP diet
A low-FODMAP diet aimed to broaden the patient's diet to include more low-FODMAP foods, knowing that these should be tolerated  (22) and PRISMA guidelines for undertaking and reporting the results of a systematic review and meta-analysis (23) will be followed at each stage of this review.

Type of Population
Studies involving children of any age diagnosed with any of the FAPD, according to established Rome Criteria (I-IV) (24)(25)(26)(27) or following the Apley Criteria (28) or other definition provided by the authors of the original study, will be included. If feasible, each FAPD will be assessed separately, as the signs and symptoms among them differ (29). Trials performed in children with comorbidities, as well as studies performed in adults, will be excluded.

Type of Intervention and Comparator
Studies using a low-FODMAP diet compared with any type of comparator (i.e., standardized [i.e., average national] or other diet or no intervention) will be eligible for inclusion. Trials that assessed the restriction of individual FODMAPs (i.e., lactose-free diet) also will be included, as in the pediatric population, such or similar approaches (i.e., elimination of only some food groups) are commonly used to avoid over-restriction (5,6). Each type of a low-FODMAP diet will be assessed separately to avoid bias associated with the mixing of dissimilar interventions (30). Studies assessing the effects of a gluten-free diet will not be considered for this review, as this intervention does not exclude any of the FODMAPs.

Type of Studies
Only RCTs, including those of crossover design (regardless of language, study duration, settings, and length of follow-up), will be included. As evidence for the employed population, intervention, comparison, outcomes, and studies (PICOS)' question is generally scarce (31), and to avoid bias associated with any exclusion of unpublished results, abstracts and letters to the editor, preliminary analyses will be included if available data are sufficient for the analyses. Ongoing studies will be excluded; however, they will be reported in the "Characteristics of ongoing studies" table.

Type of Outcomes Measures
The primary outcome will be the difference in abdominal pain intensity between the intervention and control groups, measured ideally with any validated measurement instrument (e.g., Visual Analogue Scale). However, if lacking, any method reported by the authors will be accepted.
Secondary outcomes will include change in pain frequency; stool consistency; other gastrointestinal symptoms (assessed as single symptoms or as group of gastrointestinal symptoms); school performance (i.e., school attendance) associated with FAPD symptoms; psychological functioning (i.e., anxiety) associated with FAPD symptoms; parent's work absenteeism associated with FAPD symptoms in child; health-related quality of life; adherence to the diet (compliance); growth (subjects' body weight and length); adverse events (measured as a proportion of the participants reporting an adverse event throughout the trial, extracted with characteristics). If feasible, the reported measurement instruments will be extracted.

Information Sources
The Cochrane Central Register of Controlled Trials (CEN-TRAL, the Cochrane Library), MEDLINE (PubMed), and EMBASE databases will be searched for relevant studies from inception to the date of completion of this review. A weekly e-mail update of new search results in PubMed will be set up. The search terms for this review will be based on the 3 previous similar systematic reviews (32)(33)(34). The search strategy will include use of the Cochrane Collaboration's validated filters (http://work.cochrane.org/rct-filters-different-databases) for identifying RCTs in PubMed and EMBASE, which will be combined with the main Medical Subject Headings (MeSH) and text key-words (see Supplemental Digital Content 2, http://links.lww.com/PG9/A27). The search will be carried out independently by 2 reviewers.

Other Sources
The ClinicalTrials.gov, National Institute for Health Research (http://www.ukctg.nihr.ac.uk/default.aspx), and Turning Research into Practice (https://www.tripdatabase.com/) websites also will be screened to identify unpublished and ongoing trials. Moreover, the International Clinical Trial Registry Platform (http://apps.who.int/ trialsearch/), which is a more comprehensive trial registry, will be considered, if available (at the time of the writing of this protocol, it is temporarily unavailable).

Study Selection Process
Two reviewers (AH, AS) will independently screen the title, abstract, and keywords of every study identified with the search strategy and other source records using the EndNote X9 Computer program [Version 9.3.3. Philadelphia, The Clarivate Analytics, 2020]. For eligible and unclear studies, relevant systematic reviews and reviews, the full texts will be retrieved and assessed against eligibility criteria (not for conference abstracts).

Data Collection and Management Process
Data extraction will be performed using data-extraction forms developed by the reviewers. Relevant data will be extracted by one reviewer (AS) and independently checked by the second reviewer (AH). Reviewers will attempt to contact the authors of the original study whenever reported data are insufficient for planned analysis. Any discrepancies with regard to the study selection and extraction will be resolved through discussion, and, if necessary, another review author (HS) will be consulted. Excluded RCTs with reasons for exclusion will be reported in a "Characteristics of excluded studies" table. The process of selection will be presented as a PRISMA flow diagram.

Data Items
Data will be extracted by one reviewer (AS) and independently checked by the second reviewer (AH). Any discrepancies will be resolved by discussion until a consensus is reached.
The reviewers will independently extract the following data from the included trials: Data items extracted from the included trials will be reported in a "Characteristics of included studies" table.

Risk of Bias in Individual Studies
The risk of bias of all included trials will be assessed independently by 2 reviewers (AS, AH) using the Cochrane Collaboration's tool for assessing risk of bias (including: adequacy of sequence generation, allocation concealment, and blinding of participants, personnel and outcome assessors; incomplete outcome data are addressed, free of selective outcome reporting, and free of other sources of bias). The risk will be classified as "high," "low," or "unclear," following the Cochrane Handbook for Systematic Review of Intervention Version 5.1 (35). Discrepancies between the reviewers will be resolved by discussion until a consensus is reached; in case of disagreement, the other reviewer (HS) will be consulted.

Data Synthesis
If feasible for analysis, data will be entered by one reviewer (AS) into Review Manager [Computer program. Version 5.2 Copenhagen, The Cochrane Collaboration, 2012] and checked for accuracy by the other reviewer (AH). Relative risk (RR) between the experimental and control groups with a 95% confidence interval (CI) will be used for dichotomous outcomes, and mean difference (MD) with a 95% CI will be used for continuous outcomes. All analyses will be based on the random-effects model. In the event of substantial unexplained heterogeneity of the analyzed population, intervention, comparators, and outcomes, the results of the trials will not be pooled in a meta-analysis but rather assessed and described narratively (30).

Heterogeneity
If feasible, heterogeneity will be quantified by χ 2 and I 2 , which can be interpreted as the percentage of the total variation between studies that is attributable to heterogeneity rather than to chance. A value of 0% to 40% indicates heterogeneity that may not be important, and larger values show increasing heterogeneity.

Subgroup Analysis
Whenever feasible, for the primary outcome, we will carry out subgroup analyses based on factors that potentially may influence it, such as age of participants; sex of participants; applied diagnostic criteria; duration of intervention; comparators; concomitant medications; and, in the case of the IBS population, IBS subtypes.

Meta-bias
The publication bias will be assessed by the generation of funnel plots as proposed by Egger et al (36), if at least 10 studies are included.