PICO 1: Are the clinical history and symptoms required for IBS diagnosis in children?

Statement: We recommend the assessment of patient’s symptoms and clinical history for diagnosis and management of children with IBS.

Statement endorsed, overall agreement: 100%: A + 100%, A 0%, A- 0%, D- 0%, D 0%, D + 0%.

LoE: unable to assess using GRADE methodology; SOR: consensus recommendation.

Summary of evidence: IBS is a clinical diagnosis. The Rome IV criteria currently provide symptom-based guidelines that can be used to diagnose children and adolescents with IBS [3]. They are focused on a careful history and physical examination that can help to identify this disorder and to establish a positive diagnosis of IBS in children, in the absence of alarm signs and symptoms. Alarm symptoms include pain disrupting sleep or localized in right upper or right lower quadrant, rectal bleeding, fever, weight loss, family history of inflammatory bowel disease or celiac disease, low hemoglobin level.

The clinical history should focus on the characteristics of abdominal pain, on changes in pain intensity related to bowel motions and on details about stool patterns.

Symptoms must be recurrent and should occur at least four times a month for a minimum of 2 months. Symptoms such as abdominal bloating and defecation urgency can also occur [22].

Symptoms can often appear after a gastrointestinal infection [23, 24] or after emotional distress [25] and patients may also experience increased sadness, interpersonal sensitivity and sleep disturbances [26]. Somatic symptoms and psychological problems, including anxiety and depression, are also commonly found in children with IBS [27] and their identification may aid in formulating the diagnosis and starting correct treatment of these children.

The Rome IV criteria identify different subtypes of IBS according to the stool consistency on the days with abnormal bowel movements and include IBS-C (constipation), IBS-D (diarrhea), IBS-M (mixed) and IBS-U (unsubtyped) [14, 28]. Patients with IBS-C have > 25% of their bowel movements associated with stool types 1 or 2 according to Bristol stool form scale (BSFS), while those with IBS-D have > 25% of their bowel movements associated with stool types 6 or 7. Patients with IBS-M have > 25% of their bowel movements associated with stool types 1 or 2 and > 25% of their bowel movements associated with stool types 6 or 7 [29]. The clinical subtypes should not be considered a rigid classification, as they could change over time, but the classification could help physicians to costumize from time to time management in children with IBS.

PICO 2: Should children with IBS diagnosis be regularly evaluated for psychological comorbidities?

Statement: We recommend psychological comorbidities assessment in children with IBS.

Statement endorsed, overall agreement: 100%: A + 94.4%, A 5.6%, A- 0%, D- 0%, D 0%, D + 0%.

LoE: unable to assess using GRADE methodology; SOR: consensus recommendation.

Summary of evidence: Several studies have reported increased anxiety and depression in children with IBS compared to healthy children [30,31,32,33,34,35]. Similarly to adults, whether abdominal pain or mental symptoms come first remains to be elucidated also in children [36,37,38,39]. Anxiety and depression appear to be associated with increased severity of abdominal pain and disability [40,41,42,43], although other studies did not confirm these data [44]. Additionally, several studies have shown that the association between psychological comorbidity and childhood abdominal pain increases the risk of having an IBS disorder in adulthood [36, 45,46,47].

Although anxiety and depression are widely studied, they may not be the most important factors in children with IBS. Holler and colleagues observed that somatization and pain catastrophizing mediate the association between anxiety/depression and the severity of IBS abdominal pain in children [27]. Previously, other studies have shown that children with functional abdominal pain had higher somatization scores compared to healthy children [48,49,50,51,52] and that this is more relevant for children with IBS than other functional abdominal disorders [50, 53]. In addition, Song and colleagues found that Korean children with IBS presented higher scores of stress compared to healthy children [54], a finding confirmed also by other authors [55]. Several studies have reported that functional gastrointestinal symptoms are significantly more common in children with a history of physical, emotional, and sexual abuse [56,57,58].

PICO 3: Is it more appropriate to approach children with suspected IBS using a positive diagnostic approach as opposed to one of exclusion?

Statement: We recommend a positive diagnostic strategy in children with symptoms suggestive of IBS.

Statement endorsed, overall agreement: 100%: A + 72.2%, A 27.8%, A- 0%, D- 0%, D 0%, D + 0%.

LoE: unable to assess using GRADE methodology; SOR: consensus recommendation.

Summary of evidence: The diagnosis of IBS is known to have a major impact on national healthcare systems and economies [17]. The estimated annual cost of caring for children with IBS in Europe is over 15 billion euros [18].

As stated in the Rome IV criteria [3], IBS is a functional bowel disorder in which recurrent abdominal pain is related to defecation or associated with change in bowel habits (frequency or appearance of stool). Symptom onset should occur at least 6 months prior to diagnosis and symptoms should be present on average at least 1 day per week during the last 3 months. Thus, IBS should be a positive diagnosis, based on medical interview, physical examination, and limited blood tests, and not the result of an exclusion algorithm aimed at ruling out all possible organic diseases. The decision to perform additional diagnostic procedures should be assessed individually. Velasco-Benítez and colleagues [59] demonstrated a sensitivity of 75% and a specificity of 90% with positive and negative predictive value of 85.8% and 79.9%, respectively for IBS diagnosis in children. Previously, also Miele and colleagues [60] had shown that the application of the Rome II criteria for functional gastrointestinal disease has a significant positive impact in reducing unnecessary gastrointestinal endoscopy in children.

Taken together, these considerations suggest that a positive approach is effective to reduce the cost of assessing IBS in children.

PICO 4: Should all children with a diagnosis of IBS be evaluated for occult constipation?

Statement: We recommend to rule out occult constipation in children with symptoms suggestive of IBS when therapeutic strategies have failed.

Statement endorsed, overall agreement: 100%: A + 72.2%, A 27.8%, A- 0%, D- 0%, D 0%, D + 0%.

LoE: unable to assess using GRADE methodology; SOR: consensus recommendation.

Summary of evidence: Constipation is a common condition in the pediatric population. Occult constipation is defined as a clinical condition with no obvious complaints of constipation and no symptoms suggestive of constipation, but at least one of the following: 1) hard stool consistency (stone or pellet-like) on rectal examination and 2) evidence of colon distended by feces on plain abdominal x-ray. Stool retention has been suggested as the cause of recurrent abdominal pain in children [61]. However, there are only two studies on this topic in children with IBS.

Pelvic floor function using anorectal manometry and balloon ejection test was tested in 67 adolescents with functional constipation (n = 16), fecal incontinence (n = 18), and IBS-C (n = 33) [62]. Patients classified as IBS-C were more likely to report weight loss (p = 0.03), bloating (p = 0.04), and incomplete rectal evacuation (p = 0.03), as compared to the other two groups. Furthermore, the test showed the presence of dyssynergy defecation in children with IBS-C.

Tosto and colleagues prospectively enrolled 26 consecutive children who meet Rome IV criteria for a diagnosis of IBS-D and IBS -M [63]. Patients who fulfilled criteria for suspect “occult constipation” received a bowel cleaning regimen with polyethylene glycol 3350 and were followed up for at least 6 months. 16 additional patients with IBS-C referred in the same period were enrolled as control. The endpoints were a decrease of more than 50% in abdominal pain intensity and frequency scores and resolution of diarrhea for patients with IBS-D and IBS-M. The endpoints were met by 8 (80%) and 14 (87%) of the patients with IBS-D and IBS-M, respectively, with decrease of abdominal pain and resolution of “diarrhea” (pseudo-diarrhea). The response was not significantly different from that observed in 15 (93%) of the IBS-C control group. Despite the small number of patients and the uncontrolled nature of the study, it suggests that a number of patients labeled as IBS-D or IBS-M may present functional constipation and should be managed as such.

PICO 5: Should children with IBS symptoms be tested for celiac disease (CD)?

Statement: We recommend serologic testing for CD in all children with IBS symptoms.

Statement endorsed, overall agreement: 100%: A + 88.9%, A 11.1%, A- 0%, D- 0%, D 0%, D + 0%.

LoE: Moderate; SOR: Strong.

Summary of evidence: IBS and Celiac Disease (CD) could present with similar manifestations in children, thus resulting in misdiagnosis. Hence, Rome IV criteria suggest an evaluation of CD in case of IBS-D [3]. Moreover, in central Europe, abdominal pain has been recently reported as the leading symptom in children with CD (in 33.3% of symptomatic children, and among those, in 66.4% of polysymptomatic children) followed by abdominal distension and diarrhea (56.7% and 54.2%, respectively) [64].

However, it is still unclear if children with IBS are more likely to have CD [65]. Cristofori and colleagues [66] in a 6-years prospective cohort study observed that 12 of 270 patients with IBS (4.45%) were positive for CD testing. Conversely, other authors failed in finding an association between recurrent abdominal pain and the prevalence of anti-endomysial antibody when compared to asymptomatic controls [67]. Falcon and colleagues reported that only 1 of 181 children with functional abdominal pain disorders (FAPDs) (0/84 with IBS) had positive CD serological testing, questioning the need for CD testing in all children with IBS [68].

However, the prevalence of CD in Europe is higher compared to other countries [69] and in Italy it has been recently reported as one of the highest in world in school-age children [70].

Therefore, taking into account the significant potential consequences of missing the diagnosis of CD, we recommend serologic testing for CD with quantitative IgA levels and IgA anti-tissue transglutaminase (tTG) in all children with IBS symptoms if CD prevalence in the population is > 1% (as it is in Italy).

PICO 6: Can fecal calprotectin, and/or CRP be used to rule out IBD in children with IBS symptoms?

Statement: We recommend the use of fecal calprotectin1 and C-reactive protein2 to exclude inflammatory bowel disease in patients with IBS symptoms and diarrhea without alarm features.

Statement endorsed, overall agreement: 94.4%: A + 72.2%, A 22.2%, A- 5.6%, D- 0%, D 0%, D + 0%.

1 LoE: Very low; SOR: Strong. 2 LoE: Very low; SOR: Conditional.

Summary of evidence: In the last decade, the use of fecal calprotectin (FC) as a non-invasive screening method to screen for intestinal mucosal inflammation has increased both in children and adults [71]. A recent systematic review, analyzing 8 pediatric studies, concluded that fecal calprotectin is a valuable test to exclude IBD and to avoid invasive investigations, with particular reference to colonoscopy [72]. In keeping with these data, Heida and colleagues demonstrated that children should not undergo endoscopy when FC levels are < 50 mg/g [73]. A flowchart providing a guideline on how to proceed with a child presenting with gastrointestinal (GI) symptoms according to FC levels has been suggested [74]: in absence of “red flag” symptoms and FC < 250 μg/g in two separate samples, IBD will be unlikely and further investigations should not be performed [74]; instead, FC levels of > 250 μg/g in two separate samples in children with GI symptoms suggestive of IBD support the need for further invasive procedures [74]. Recently, a study in 853 children showed that fecal calprotectin had a fair accuracy, superior to C-reactive protein (CRP), hemoglobin levels and erythrocyte sedimentation rate (ERS) to discriminate between organic and functional causes of abdominal pain [75].

Thus, the ESPGHAN expert group recommended using the fecal calprotectin as a tool to differentiate functional abdominal pain disorders from organic diseases [76].

If low values of FC can exclude IBD with sufficient accuracy, high values do not exclude IBS, since increased levels of fecal calprotectin can also be found in children with IBS compared to healthy controls [77,78,79]. These data indirectly confirm the presence of a low-grade inflammation also in children with IBS. Similarly, it has been reported that median high-sensitive CRP levels in the IBS group were significantly higher than in healthy controls (1.80, IQR 0.7–4.04 mg/l vs 1.20, IQR 0.5–2.97 mg/l respectively, p < 0.006,) with the highest levels in IBS-D patients showing greater symptoms severity [80].

Therefore, in patients with high levels of FC, IBS cannot be excluded and further examination could be required, whereas in patients with low levels of FC, IBD can be reasonably excluded.

PICO 7: Should IBS patients be routinely checked for stool pathogens?

Statement: We recommend against routine stool testing for enteric pathogens in children with IBS.

Statement endorsed, overall agreement: 100%: A + 88.9%, A 11.1%, A- 0%, D- 0%, D 0%, D + 0%.

LoE: Low; SOR: Conditional.

Summary of evidence: Zeevenhooven and colleagues [81] demonstrated that testing for Giardia lamblia in association to fecal calprotectin and celiac disease serology could have a high sensitivity and specificity in discriminating between organic and functional causes of chronic abdominal pain. Moreover, it has been suggested that in some countries such as Egypt, Pakistan, Turkey, and Poland, the prevalence of parasitic infections may justify stool testing in children with recurrent abdominal pain [82,83,84,85,86]. However, in developed countries, the prevalence of parasitic infections did not differ significantly in children with and without chronic abdominal pain [87, 88]. Even if there is the possibility to have post-infection IBS in children [23, 24, 89,90,91,92], routine testing for enteric pathogens is not recommended in children with suspected IBS.

PICO 8: When is colonoscopy indicated in patients with IBS symptoms?

Statement: We recommend colonoscopy only in patients with IBS symptoms and alarm features.

Statement endorsed, overall agreement: 100%: A + 83.3%, A 16.7%, A- 0%, D- 0%, D 0%, D + 0%.

LoE: unable to assess using GRADE methodology; SOR: consensus recommendation.

Summary of evidence: The Rome IV criteria and the ESPGHAN and ESGE guidelines suggest to perform colonoscopy only if alarm signs and symptoms for organic intestinal diseases exist [93, 94]. Among alarm symptoms and signs for organic diseases in children with chronic abdominal pain [3, 95] the classic “triad” of bloody diarrhea, weight loss, and/or positive serum inflammatory markers (CRP and/or ESR) [96], and/or high levels of fecal calprotectin were found to present possible benefits from endoscopy [97]. The clinical value of colonoscopy in children with symptoms suggestive of IBS cannot be determined since the studies on this field are generically related to chronic abdominal pain [98,99,100].

Some studies focusing on the presence of intestinal low-grade inflammation in patients with FGIDs have found increased inflammatory cells (e.g. eosinophils in the duodenum for functional dyspepsia and mast cells in the ileum or colon for IBS) [78, 79, 101,102,103,104]. However, the clinical value of those findings is limited because their detection did not influence patients’ management.

In clinical practice, a common justification for performing endoscopies in children with functional abdominal pain disorders is the need for parents reassurance by demonstrating normal findings on the procedure [105], even if it carries risks related to the invasive nature of the procedure and anesthesia. However, Bonilla and colleagues [106] showed that the outcome of children with abdominal pain who underwent a negative endoscopy was similar to children with abdominal pain who did not undergo endoscopy.

PICO 9: Should patients be tested for food allergy/intolerance?

Statement: We recommend against testing for food allergy/intolerance in children with IBS.

Statement endorsed, overall agreement: 100%: A + 72.2%, A 27.8%, A- 0%, D- 0%, D 0%, D + 0%.

LoE: unable to assess using GRADE methodology; SOR: Conditional.

Summary of evidence: According to Rome IV criteria, the diagnosis of IBS is symptom-based, and patients who fulfill criteria for IBS without other signs and symptoms suspicious for food allergy should not be tested for food allergies or intolerances.

IBS shares pathophysiologic features with allergies, including both immune and psychological patterns, even though the mechanism through which allergens might play a role in the determinism of IBS remains unclear. In the absence of systemic allergic symptoms or oral allergy syndrome, allergic triggers for IBS are unlikely to be identified by standard testing [107].

Indeed, evidences regarding a possible link between IBS and allergic diseases are controversial [108]. Some authors observed that children with asthma have a significantly higher IBS prevalence [109,