A study by Meyer et al. in 437 children with food allergy diagnosed by food elimination and rechallenge found that symptoms consistent with neuro-enteric disturbances were present in the majority, namely vomiting (57.8 %), back-arching and screaming (50 %), constipation (44.6 %), diarrhea (81 %), abdominal pain (89.9 %), and abdominal bloating (73.9 %). Rectal bleeding was seen in 38.5 % of patients [10]. The majority of patients were initially managed with a milk, soy, egg, and wheat-free diet (41.7 %) and at a median age of 8 years, 24.7 % of children still required to eliminate some of the food allergens suggesting that a proportion do not outgrow this allergic tendency.

Data in animal models of hypersensitivity reactions have shown that antigen challenge in vivo results in neurally mediated gastrointestinal dysmotility, including effects on the stomach (gastric emptying and secretion) [11] and small intestine (abolition of the migrating motor complex, increase in aborally propagating clustered contractions, and disruption of fasting and fed patterns) [12–15]. In sensitized animals with delayed gastric emptying, food antigen challenge showed histological evidence of mast cell degranulation in the gastric mucosa, increased intraluminal release of histamine, and increased markers specific for mucosal mast cell degranulation [16]. Furthermore, hypersensitivity to food proteins induced an increase in number and activation of mast cells and chronic motor alterations, such as intestinal hypermotility that seems to persist long after antigen challenge [17]. In sensitized rats, mucosal mast cells appear to mediate the motor responses induced by chronic oral exposure to ovalbumin [18]. Proteases from degranulated mast cells in close proximity to autonomic and enteric nerves cause acute and long-term hyperexcitability of ileal neurons in animal models by activating proteinase-activated receptor 2 (PAR2) on these neurons [19]. PAR2 has been suggested to play a role in nociceptive signaling, by sensitizing the vanilloid receptor 1 (TRVP1) to induce visceral hyperalgesia [20]. Exposure of mice to enteric-coated antigen promotes a T helper 2-associated eosinophilic inflammatory response that involves the esophagus, the stomach, and the small intestine and Peyer’s patches and leads to the development of gastric dysmotility [21]. Furthermore, after oral ovalbumin challenge, allergic mice present higher levels of anxiety with increased activity in brain areas associated with emotional and affective behavior [22].

Although all the mechanisms by which food allergy induces disturbances in gut motility and sensation in patients are still poorly understood it is well known that allergic reactions to food evoke immune inflammatory cell infiltration and activation at various gastrointestinal mucosal sites. Food allergy, either due to IgE- or non-IgE-mediated mechanisms , is commonly thought to elicit gut mucosa inflammation, where different types of immune cells (i.e., MCs, eosinophils, and T and B lymphocytes) are present and scattered along different sites of the gut [23]. Mast cells (MCs ) are regarded as key effector cells of both immediate and delayed-type hypersensitivity reactions. Gastrointestinal MCs usually act either as effector cells secreting autocrine factors or facilitate the recruitment of other immunocompetent and inflammatory cells (i.e., eosinophils, lymphocytes, and neutrophils), which may in turn contribute to the persistence of allergic reactions [24]. On activation, MCs release a variety of bioactive substances, including vasoactive, nociceptive, and pro-inflammatory mediators, as well as neurotransmitters. Given their close proximity to enteric neurons, MC degranulation is capable of activating neural reflexes and muscle contractility leading to changes in gut motility [25, 26].

In both children and adults with irritable bowel syndrome (IBS), the average numbers of mucosal MCs have been shown to be increased at different intestinal sites, and to be in closer proximity to mucosal nerve endings compared to controls [27–29]. Not only were MCs found to release larger amounts of mediators such as histamine and tryptase, their spatial proximity to nerve fibers correlated with the severity of perceived abdominal pain in patients with IBS [30]. Eosinophils also have a recognized role in GI dysmotility [31].

It is increasingly recognized that early life events including inflammation, trauma, and stress may influence neuromuscular function and result in functional gastrointestinal disorders later in life [32]. For allergy this process is mainly implicated in abdominal pain and discussed further in the section of functional abdominal pain disorders.

Eosinophilic Esophagitis (EoE) is often considered an allergic condition given the nature of the inflammation seen with a predominance of allergy type cells, increased prevalence of a personal or family history of atopy, seasonal variation in symptoms in a proportion of cases suggesting a role for environmental allergens and response to allergen avoidance. Dysphagia and food impaction are common symptoms in older children with eosinophilic esophagitis (EoE) and represent the initial presenting symptoms in up to 20 % of cases [33]. Although both focal and diffuse esophageal anatomical abnormalities, including strictures, rings, or reduced organ caliber, might be responsible for generation of the symptoms, abnormalities in esophageal motility and distensibility are the most common causes [34–36].

By using the endo-FLIP (Functional Luminal Imaging Probe) system, which allows the assessment of the esophageal wall distensibility and hence of tissue remodeling and fibrosis, Nicodème et al. correlated in 70 symptomatic adult patients with EoE the esophageal distensibility with both susceptibility to food impaction and esophageal mucosal eosinophilia [37]. Reduced esophageal distensibility, which correlated with food impaction, was found in both patients with EoE and those with proton pump inhibitors-responsive EoE as compared to controls, whereas no correlation was found between esophageal distensibility and eosinophil mucosal infiltration. The authors suggested that reduced esophageal distensibility (<225 mm²) in patients with EoE is predictive for food impaction and need for dilation. No such studies are available in pediatric age.

Wide ranges of esophageal motor abnormalities have been described in patients with EoE. Using conventional manometry , the esophageal motor pattern described includes normal peristalsis, complete aperistalsis, ineffective peristalsis secondary to simultaneous contractions, nutcracker esophagus, diffuse esophageal spasms, and achalasia. Nurko et al. found abnormalities in esophageal peristalsis in 41 % of children with EoE, including isolated and high-amplitude contractions and ineffective peristalsis both during fasting and during meals [38]. Notably, during the study all episodes of dysphagia correlated with abnormal esophageal peristaltic events. Similar motor abnormalities have been described also using high resolution manometry, although pan-esophageal pressurization seems to be the most consistent findings in EoE patients [39, 40].

In healthy individuals, both inner circular and outer longitudinal esophageal muscle layers are perfectly synchronous during peristalsis . Contraction of the longitudinal layer during deglutition is responsible for shortening of the esophagus, returning to its normal length when deglutition ends. By using simultaneously esophageal US and manometry, Korsapati et al. evaluated the interaction of circular and longitudinal muscle layers in patients with EoE and showed that the presence of asynchronicity between contractions of two muscle layers during peristalsis at the expense of longitudinal represents an abnormality that may contribute to the development of dysphagia [41].

The pathogenesis of esophageal motor abnormalities in EoE patients is still unknown. It is well documented that there is an eosinophilic infiltration in all esophageal layers. Studies in vitro have shown that eosinophils are capable to increase the contraction of the fibroblasts, and their degranulation is associated with axonal necrosis [21, 42, 43].

Moreover, eosinophil-derived major basic protein binds muscarinic acetylcholine receptors, which in turn can lead to smooth muscle contraction and subsequent dysmotility [44]. Mast cells may also play a role in the generation of esophageal dysmotility. An increased number of MC in the esophageal mucosa as well as an upregulation of MC genes have been described in patients with EoE [45]. It is well known the effect of MC mediators on fibrogenesis as well as the effect on enteric neuromuscular function.

Gastro-esophageal reflux disease (GERD) and cow’s milk allergy (CMA) are both extremely common in infancy . Several studies emphasize a causal relationship between GERD and CMA at least in a subgroup of infants with GERD [46].

The prevalence of GERD attributable to CMA ranges between 16 and 56 %. Iacono et al. reported in 42 % of infants with GERD symptoms and histologic esophagitis the disappearance of reflux symptoms after they were put on a cow’s milk free diet and the reappearance of symptoms on subsequent formula challenge [47]. Nielsen et al. showed that 56 % of children with severe GERD were found to have CMA on double-bind or open challenge [48]. Recently, Yukselen et al. identified food allergy in 65 of 151 (43 %) children with GERD refractory to medical therapy, and the majority of them (58/65, 89 %) were allergic to cow’s milk, whilst only a small number (7/65, 11 %) to egg. Interestingly, the authors reported that only half of patients with GERD and food allergy had both positive oral challenge and skin prick test and/or specific IgE, whereas in the remaining half only the oral challenge confirmed the diagnosis of food allergy [49].

With the exception of those patients with mild typical CMA manifestations , such as atopic dermatitis, rhinitis, and diarrhea, it is challenging to discriminate between the symptoms observed in primary GERD and those of GERD associated with CMA. Early studies have advocated the role of pH monitoring to discriminate between primary GERD and that secondary to CMA. A particular phasic pH pattern characterized by a slow and progressive decrease in esophageal pH between two feeds was suggested as a sensitive and specific index for identifying patients with CMA-induced GERD [50, 51]. However, this finding has not been confirmed by subsequent studies [48, 52]. Nielsen and colleagues [48] performed 48-h pH monitoring in ten children with severe GERD and CMA, with cow’s milk elimination diet at day 1 and cow’s milk challenge at day 2. Although the authors showed that children with GERD and CMA had more abnormal pH monitoring than those without the association, they failed to find any differences in the reflux index between the two recording days. By using impedance-pH monitoring, which allows detection of acid, weakly acidic and weakly alkaline reflux, Borrelli et al. found in a subgroup of infants and children with CMA and GERD that cow’s milk challenge increases the number of weakly acidic reflux episodes suggesting that a challenge during 48 h impedance-pH monitoring might increases the yield in identifying this subgroup of patients [53].

The mechanisms by which food allergy induces GER are still poorly understood. Data in animal models of immediate-type hypersensitivity reactions have shown that antigen challenge in vivo results in neurally mediated foregut dysmotility, such as delayed gastric emptying . Ravelli et al. showed that milk challenge induces gastric electrical dysrhythmias and delayed gastric emptying in infants with vomiting due to CMA [54]. Schäppi et al. showed that early-onset neuroimmune interactions induced by cow’s milk challenge in the gastric mucosa of atopic children are associated with rapid derangement of gastric myoelectrical activity [55]. Notably, in this study, cow’s milk challenge induced rapid degranulation of mast cells and eosinophils. Activated mast cells were closely associated with mucosal nerve fibers, and released mast cell tryptase was co-localized with proteinase activated receptors 2 (PAR-2) on mucosal nerve fibers. In the same timeframe as these morphological changes occurred, there was a rapid (within 2 min) induction of electrogastrographic myoelectrical abnormalities. Intriguingly, in experimental animals, gastric activation of PAR-2 induces neurally mediated motor and secretory responses represented by a fundic biphasic contractile response, which involves relaxation followed by contraction, and suppression of acid production [56, 57]. Furthermore, it has been shown that a reduced prevalence of normal electrical rhythm and an increased rate of episodes of dysrhythmia are associated with antral hypo-contractility, which in turn leads to a gastric emptying delay [58, 59].

Delayed gastric emptying may increase GER by increasing the availability of material to reflux or by inducing prolonged gastric distention and more transient lower esophageal sphincter relaxations (TLESR) . Simultaneous esophageal manometry and gastric emptying breath tests in healthy adults and premature infants showed a moderate but significant correlation between gastric emptying delay and rate of postprandial TLESRs [60, 61]. Moreover, using simultaneous gastric emptying and pH-MII, Sifrim and colleagues showed that the slower the gastric emptying, the higher the pH and proximal extent of the reflux episodes [62]. Furthermore, it is well known that nonacid reflux episodes are more likely to occur during feeding and during the first postprandial hours, with greater frequency in infants compared with older children. Thus, it could be hypothesized that neuroimmune interactions induced during cow’s milk challenge by activating gastric PAR2 suppress the acid gastric production and deranges the gastric motor activity, which in turn delay the gastric emptying and increase the rate of TLESRs, resulting in an increase in the number of nonacid reflux episodes.

The high prevalence of the GERD -CMA association might be explained by different diagnostic modalities used for GERD, including endoscopy, histology, and esophageal pH monitoring, as well as different diagnostic criteria for cow milk protein allergy. However, being this association far beyond what can be expected from pure coexistence of the two entities, it should induce pediatricians to screen for possible concomitant CMA mainly in those infants and children with GERD unresponsive to medical treatment [63].

Infantile colic (IC ) describes a symptom complex of excessive and inconsolable crying in babies that otherwise appear to be healthy and thriving. Although implied in its name, the exact focus or nature of colic is not known. It classically develops in the first 2–4 weeks of life and persists through to the third or fourth month of age, affecting between 15 and 40 % of infants [6]. The impact of IC can be considerable and therefore, although the natural history is of gradual resolution without harmful consequences, many parents and physicians have sought causative factors and interventions to try and alleviate the symptoms (see Chap. 34). Allergy is one such factor.

The association between IC and allergic disorders has long been suggested mainly through three bodies of evidence, namely, response of IC to dietary exclusion, predisposition of IC to atopic conditions, and potential similarities in disturbances of neuro-enteric function and microbiome between IC and food allergy [64]. A large number of studies have investigated the response of IC to dietary exclusion either through the use of maternal exclusion of dietary antigens whilst breastfeeding or use of specialized milk formulae. Although some studies have shown no differences in prevalence of IC between breast-fed and bottle-fed infants [6, 65–67], or any significant improvement in cry duration in breast-fed infants with IC in whom mothers have eliminated cow’s milk protein [68], others have suggested that maternal exclusion of cow’s milk protein or indeed a broader exclusion is beneficial [69, 70].

Hill et al. explored an extensive maternal exclusion of allergens (excluding cows milk protein, eggs, wheat, soya, peanuts, tree nuts, and fish) in breast-fed babies with IC and reported a significant response rate (defined as ≥ 25 % reduction in cry/fuss duration; 74 % vs. 37 %) and reduction in infant crying duration in the intervention group compared to controls [70]. In formula fed infants, Iacono et al. showed improvement or complete resolution in 50 of 70 infants with diagnosed severe IC and put onto cow’s milk protein exclusion (Soya milk formula). Upon rechallenge with cow’s milk protein-containing formula all 50 infants relapsed and then again showed remission with exclusion [71]. More recently, the use of soy-based milks in infants under 6 months of age has fallen out of favor given concerns over their content of phytoestrogens with potential adverse effects [72]. A number of studies have looked at excluding cow’s milk protein with the use of extensively hydrolysed formulae. In a randomized, double-blind, parallel trial Lucassen studied 43 healthy, thriving formula-fed infants with IC (<6 months old, crying >3 h per day on at least 3 days per week). The infants were randomized to receive either a whey hydrolysate formula (n = 23) or standard formula (n = 20) and the difference in the duration of crying (minutes per day) compared between the qualification week and intervention week. Analysis (intention to treat) found that the whey hydrolysate group showed a mean decrease in crying duration of 63 min per day. The study characteristics, however, did not allow significant differences to be determined [73]. Variable improvements have also been reported in other studies using dietary modification in including the use of extensively hydrolysed cow’s milk formulae [74–76].

Further association between IC and food allergy is perhaps supported by longitudinal studies on infantile colic [65, 71, 77]. Savino et al. carried out a prospective study on 103 infants aged 31–87 days who were then recalled after 10 years and evaluated. Not only did colic appear to predispose to recurrent abdominal pain (p = 0.001) but there was an association with the development of allergic disorders including atopic eczema and food allergy (p < 0.05). Furthermore, a family history of gastrointestinal diseases and atopic diseases was significantly more prevalent in infants with colic than in controls (p < 0.05) [78]. It is, however, unclear whether there were any elements of recall bias in these studies. Further, more robust studies are needed to understand this association or indeed whether IC captures a population of allergic children that could be identified earlier in life.

It is possible that allergy, through its immune-mediated pathways, initiates changes within the function of the enteric nerves (dysmotility and visceral hypersensitivity) or is associated with alterations in the microbiome. The putative mechanisms are well described within this chapter and later in the book. There are, of course, considerable ethical issues with advocating or justifying research or indeed routine interventional assessment (e.g., endoscopy) in children with IC in the absence of any red flags for organic disease. Alterations in the microbiome and putative effects of these on the enteric nerve network and function are increasingly reported in recurrent abdominal pain. Although dysbiosis is similarly described in both food allergy and infantile colic in children [79, 80], the exact mechanisms and effects of these are unclear. There is evidence, however, that altering the microbiota population with the use of probiotics may have an impact on IC [81].

In conclusion, there remains considerable debate on whether the literature to date conclusively confirms an association between IC and food allergy given that the quality of the published studies has been poor with issues around study design, disease definition, and consistency of interventions with dietary modification. Furthermore, it is unclear whether any reported effects of dietary manipulation relate to true immune-mediated allergy or merely reflect issues around tolerance of ingested foods (osmolality, protein quality) or effects on gastrointestinal motility [82]. Overall, current estimates suggest that cow’s milk allergy is present in only 2–5 % of babies with colic [83] (see Chap. 34). The new Rome IV guidelines state that there is inadequate evidence to support the routine elimination of allergens such as cow’s milk protein for the treatment of IC [84]. It remains possible that a small subset of children with IC, especially those with red flags of allergy (Table 19.1), may derive potential benefit from dietary manipulation, however this must be balanced against the natural history of improvement, nutritional adequacy in mothers and infants as well as the risk, conversely, of compounding the development of food allergy arising from loss of tolerance due to allergen avoidance [85].

Recurrent abdominal pain (RAP) is a common complaint in children and subclassified in the recently published Rome IV criteria , into four functional abdominal pain disorders (FAPD) , including functional dyspepsia (FD), irritable bowel syndrome (IBS), abdominal migraine, and FAPD—not otherwise specified [84, 86] (see Chaps. 35, 36, 38, and 42). Although there may be variability in predominant regions of the GI tract involved FAPDs seem to share common pathogenic mechanisms of visceral hypersensitivity and central hypervigilance , which appear to result from disruption of the microbiota-gut-brain axis and abnormal enteric neuro-immune interactions.

As we broach the question of whether there is an association between FAPDs and allergy, a number of similarities between the two conditions become apparent. Not only are the symptom complexes similar between the two disorders with a predominance of abnormalities in sensation and motility, the key immune cells implicated in both disorders show considerable overlap, namely T cells, mast cells, and eosinophils [10, 87, 88]. The pathophysiolog y of FAPDs such as IBS has been extensively studied in adults and, more recently, in children [29, 89] with the most consistent finding of an increased numbers of mast cells sitting in close apposition to nerve fibers in the gut mucosa as part of a low-grade inflammation. In adults, jejunal biopsies of IBS-D patients exhibit increased mast cell counts and evidence of their activation, which correlates with clinical symptoms [90]. Mast cells have also been implicated in functional dyspepsia [31]. Faure et al. analyzed the inflammatory cells in the colonic and gastric mucosa of children with functional dyspepsia or IBS. Eleven of 12 patients with IBS and 9 of 17 patients with FD had evidence of mostly low-grade inflammation of the intestinal mucosa [91]. Another study noted that 71 % of children evaluated for suspected functional dyspepsia had duodenal eosinophilia (>10 eosinophils per high-power field of view) [92]. The real clinical significance and link to allergy, however, remains unclear.

In one of the earliest studies to explore the potential association between recurrent abdominal pain in children and allergy, Kokkonen et al. examined a consecutive series of 84 children (43 males, 41 females, mean age 7.9 years) referred with recurrent abdominal pain over a period of 1 year. All patients underwent gastro-duodenoscopy and biopsy. Based on an open elimination-challenge test , a considerable proportion (33 %) of subjects was diagnosed with food allergy, showing a close relationship with duodenal lesions namely mild elevations in the numbers of eosinophils [93]. The findings of this study, however, were probably reflective of a degree of referral bias given that in a subsequent community-based study in 404 children the same researchers found evidence of milk intolerance in only 9 subjects out of 64 with recurrent abdominal pain [94].

In an open label study of gastric mucosal cow’s milk challenge in 10 atopic (personal and/or family history) and 6 nonatopic children (ages 2–12 years) investigated consecutively with symptoms of functional dyspepsia. Simultaneous endoscopy, surface electrogastrography, and milk challenge were undertaken and laser scanning fluorescence microscopy used to examine the association of mast cell tryptase with mucosal nerves in the gastric mucosa before and after challenge. Eosinophils and mast cells within the lamina propria were increased in number in children with atopic functional dyspepsia and degranulated rapidly after cow’s milk challenge in the atopic group. Mast cells were closely associated with mucosal nerve fibers and released tryptase, which colocalised with proteinase-activated receptors on mucosal nerve fibers. The gastric antral slow wave became abnormal within 2 min of antigen challenge in atopic children showing a decrease in the normal myoelectrical rhythm paralleling an increase in bradygastria (P < 0.01). The study, however, was small and timing of the reaction unclear as how it relates to the broader group of FAPDs [55].

Part of the current challenge is that many patients with FAPD report and consider their symptoms to be related to meals [95] and similarly not all responses to diet are likely to represent true allergy given reported “reactions” by parents in almost 15 %. In adult studies almost a third of patients with IBS report the onset or worsening of symptoms after meals (28 % within 15 min and 93 % within 3 h) [96, 97]. Even though in some studies adult and pediatric patients with IBS have been reported to have a higher incidence of atopy [98], it has been difficult to confirm that this food-related association is due to allergy especially given the timing of the response appears to favor an immediate hypersensitivity mechanism, which is not the predominant form seen in GI allergy. Furthermore, a placebo response of up to 50 % is recognized in children and adults with FAPDs [99]. However, a recent review of the IBS literature found a significant number of studies, including in children, reporting an association between atopy (including food allergy and asthma) and IBS. The authors suggest that the “concept of food allergy should be included as a possible cause of IBS, and a dietary approach may have a place in the routine clinical management of IBS” [100]. Certainly there is emerging evidence of the value of specialized diets in the management of children with FAPDs, including low FODMAP diets [101]. The exact mechanisms, nutritional sufficiency and safety of such restrictive diets in children have not been confirmed and should be approached with caution until better clarification is achieved.

Some efficacy of treatments used in allergy such as anti-histamines and mast cell stabilizers has been reported although again it is unclear whether this is a true effect on decreasing allergic inflammation or indeed exerting a placebo response [88, 102]. Perhaps most interestingly and a possible explanation for the absence of overt inflammation in many children with FAPDs and suspected food allergy despite clear changes in nerve function is the concept of programming of such function earlier in life perhaps at a time of a more significant inflammatory response. It is increasingly recognized that early life events including gastrointestinal inflammation, trauma, and stress may result in maladaptive responses that could lead to the development of chronic pain conditions such as FGIDs [32]. In a study of 52 subjects diagnosed with CMA in the first year of life (mean age 8.1 ± 4.48 years, 62 % girls) and 53 controls (mean age 9.7 ± 4.20 years, 55 % girls), Saps et al. found a much higher proportion (44.2 %) of subjects who reported GI symptoms which included abdominal pain, constipation, or diarrhea compared with only 20.7 % of controls (odds ratio 3.03, P = 0.01). Abdominal pain was significantly more common in cases (30.8 %) versus controls (9.4 %) (odds ratio 4.27 [1.43–12.7]). They concluded that CMA in the first year of life constituted a risk factor for the development of FGIDs in children many years later. Saps et al. [103] and Olen et al. [104] confirmed this association, finding that children with a personal history of allergy-related diseases (asthma, allergic rhinitis, eczema, and food hypersensitivity) earlier in life were more prone to have abdominal pain at 12 year of age. An association with abdominal pain was also present when considering food allergy alone, but only for children who presented it at the age of 8 years. The risk of having IBS appeared to be increased amongst subjects reporting intolerance to a higher number of foods [105]. In the study from Lillestøl et al. [106], atopic patients had increased intestinal permeability and density of IgE-bearing cells compared with non-atopic patients, but gastrointestinal symptoms did not differ between groups. These partially conflicting data may suggest that further studies are needed to assess the long-term role of early allergy in developing functional abdominal pain. More recently, Tan et al. evaluated 11,242 children (age range: 7–18 years) with IBS and 44,968 age- and sex-matched control subjects who had been examined between 2000 and 2008 showing that children with antecedent allergic diseases had a greater risk of IBS than controls (p < 0.001) [107]. Such early life programming has now been implicated in a number of scenarios including post-infectious irritable bowel syndrome [32].