Eosinophilic oesophagitis

This is the most common and best described of the EGID and is seen in both children and adults. It is a chronic immune/antigen mediated disease characterised clinically with symptoms related to oesophageal dysfunction and histologically by eosinophil predominant inflammation. There are high rates of concurrent asthma, allergic rhinitis, eczema and food allergy/anaphylaxis. It is more common in boys. Young children typically present with symptoms of gastro-oesophageal disease (GORD), food refusal or growth faltering whilst adolescents tend to present with dysphagia [21]. Complications include food impaction, oesophageal stricture formation and perforation. Serum IgE and skin prick tests for immediate type food allergy are warranted to identify comorbid food induced allergic disease, but these tests alone are not sufficient to make the diagnosis of food allergy driven eosinophilic oesophagitis. First line therapies must include the use of protein pump inhibitors, as oesophageal eosinophilia can be secondary to GORD. If symptoms fail to improve then dietary therapy should be tried as sole treatment or in conjunction with topical swallowed corticosteroids (steroids that are usually inhaled but are taken orally) [22].

Dietary therapy has been shown to be an effective treatment when assessing the histopathology of patients and three different regimens have been used. Complete amino acid based feeds have been shown to induce remission in 96% of patients, the majority of whom needed a nasogastric tube to complete the treatment; the trial of feeds is followed by an extended period of food reintroductions to identify specific allergens [23]. Dietary restrictions based on allergy testing or the exclusion of the most likely food antigens have also been used. The former was successful in 69% of patients [24].The latter (excluding cow’s milk, soy, wheat, egg, peanut and seafood) has demonstrated a 74% success rate [25]. A recent retrospective case series of patients treated with one of these three elimination diets showed the greatest efficacy with complete amino acid based (elemental) feeds with 96% remission. Empirical removal of the most common allergens from the diet was as successful as a directed diet based on skin prick tests and patch testing alone (remission rates of 81% and 65%); however, the treatment chosen was not randomised and depended on negotiations with the family [26].

Consensus recommendations suggest that the specific therapy prescribed should take into account the family lifestyle, adherence to therapy and resources [22, 27]. Multiple food exclusions are difficult for the family to manage and expose the child to the dangers of dietary inadequacy, so careful monitoring is required with amino acid based supplemental feeds or vitamin and mineral supplements as needed. Once the disease is in remission foods should be reintroduced sequentially with careful monitoring so that only the foods the child is allergic to continue to be excluded. In one study cow’s milk was found to be the most common allergen followed by wheat, egg and soy [28].

Eosinophilic gastroenteritis

Eosinophilic gastroenteritis is a chronic inflammatory disorder of the GI tract characterised by eosinophil accumulation in the mucosa or in deeper layers of the GI wall. Symptoms often mimic those of inflammatory bowel disease or irritable bowel syndrome. Under healthy conditions the stomach and the small intestine show detectable baseline numbers of eosinophils but their number is increased in eosinophilic gastroenteritis. Symptoms vary according to the site affected and include pain, dysmotility, nausea, vomiting, diarrhoea, blood loss, malabsorption, protein losing enteropathy and faltering growth [29].

There are no consensus guidelines on which to base therapies due to the rarity of this diagnosis and depends on whether associated food allergy and food allergens can be confirmed. Four food exclusion diets (milk, egg, wheat and soya free), few foods diets or exclusive elemental (amino acid based) enteral feeds can be tried [30]. If there is a limited response to these therapies then medical treatment is unavoidable [31].

Eosinophilic colitis

Primary eosinophilic colitis (EC), also known as allergic colitis of infancy or dietary protein induced proctocolitis, is thought to be non-IgE based (with skin prick tests usually negative).The exact immunological mechanism is not fully understood. This disease entity is predominantly a disorder of infants younger than 3 years. The classical presentation is of blood streaked stools in an otherwise healthy infant; however, it can also be seen in older children and adolescents where it may present with abdominal pain, anorexia and weight loss. Endoscopic findings include increased eosinophils and lymphonodular hyperplasia. Elimination of the causal protein is usually followed by resolution of symptoms [32]. In EC the most common causal protein is cow’s milk and, as a large number of infants with this disorder are breast fed, this necessitates a maternal exclusion diet [30]. In one series of 95 breast fed infants with EC dietary exclusions were successful in 89% of cases. Of the remainder, seven responded to an extensively hydrolysed protein formula (EHF) and the remaining 4 (all of whom had eczema) needed an amino acid based formula (AAF) [33]. Prognosis of EC that develops in infancy is good; however, for young adults the natural history is that of a more chronic disease [34].

Mammalian milks

Mammalian milk is not suitable to be used as an infant feed without modification due to its high renal solute load and inadequate vitamin and mineral content. The proteins in goat’s and sheep’s milks share antigenic cross-reactivity with cow’s milk proteins [38, 41]. Infant formulas based on goat’s milk are not recommended for use in GI food intolerances [42].

Soy protein based formulas

A soy protein based formula was used for the first time in 1929 to feed infants with cow’s milk protein allergy. Today these feeds are based on soy protein isolate supplemented with L-methionine to give a suitable amino acid profile for infants. They are lactose free with glucose polymer as the most common source of carbohydrate. The fat is a mixture of vegetable oils that provide long chain fatty acids, including essential fatty acids. Feeding modern soy formulas to infants is associated with normal growth, protein status and bone mineralisation [43].

However, in the UK the use of soy formulas in infants under the age of 6 months has been advised against by the Chief Medical Officer, unless there is a specific medical indication, due to their high phytoestrogen content [44]. There appears to be less risk to the infant after 6 months of age as the amount of isoflavones per kilogram body weight is reduced as dependence on formula as a source of nutrition decreases. The infant’s potentially vulnerable organ systems are likely to have matured by that age. The Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment is reviewing more recent research in this area.

Soy protein has a very large molecular weight and after digestion can generate a large number of potential allergens. IgE mediated soy allergy is common, affecting 0.4% children in the USA [45]. Severe GI reactions to soy protein formula have been described for more than 30 years and include enteropathy, enterocolitis and proctitis. In FPIES caused by cow’s milk protein 30%–64% had concomitant soy induced enterocolitis which responded to the use of EHF. It is suggested that an intestinal mucosa damaged by cow’s milk protein allows increased uptake and increased immunological reaction to soy protein. A reported 60% of infants with cow’s milk protein induced enterocolitis are equally sensitive to soy [46]. The American Academy of Pediatrics state that soy formulas are not recommended in the management of cow’s milk protein enteropathy or enterocolitis; however, most children can resume soy protein consumption safely after 5 years of age [43]. EPSGHAN concludes that the use of EHF (or AAF if the hydrolysate is not tolerated) should be preferred to soy protein feeds and the latter should not be used before the age of 6 months. If soy formulas are used for therapeutic use after the age of 6 months due to their lower cost and better acceptance, tolerance to soy protein should first be established by clinical challenge [46].

Older infants with documented IgE mediated allergy to cow’s milk protein can do well on soy protein based formula [47–49]. In other GI manifestations of possible cow’s milk allergy, such as constipation, soy feeds can be tried. In one of the first papers looking at cow’s milk allergy and chronic constipation 68% of the children responded to soy milk. Soy formula has the benefit of being at least half the cost of EHF and is much more palatable. Prospective studies looking at the incidence of soy allergy in patients with GI conditions are required. Soy infant formulas available in the UK are summarised in Table 7.5.

Feeds based on protein hydrolysates

Infants with cow’s milk allergy and proven or suspected soy intolerance need an alternative formula. The allergenicity or antigenicity of a particular protein is a function of the amino acid sequences present and the configuration of the molecule. An epitope is the area of a peptide chain capable of stimulating antibody production. During the manufacture of a hydrolysate the protein is denatured by heat treatment and hydrolysed by proteolytic enzymes leaving small peptides and free amino acids. The enzymes are then inactivated by heat and, along with residual large peptides, are removed by filtration [50].

The proteins used to make a hydrolysate vary and production methods also differ between manufacturers. The profile of peptide chain lengths between different feeds will not be identical, even when the initial protein is the same.

Potential problems with hydrolysate formulas

Despite the rigorous conditions employed in the manufacture of these feeds there are still potential sequential epitopes present that can be ‘recognised’ by sensitive infants. Extensively hydrolysed protein based feeds vary considerably in their molecular weight profile and hence in their residual allergenic activity (Table 14.7). Feeds with peptides >1500 Da have been demonstrated to have residual allergenic activity [51, 52]. The degree of hydrolysis does not predict the immunogenic or the allergenic effects in the recipient infant. It has been recommended that dietary products for treatment of cow’s milk protein allergy in infants should be tolerated by at least 90% of infants with documented cow’s milk allergy [42]. In instances where an infant is not malnourished and fails to tolerate one EHF, a second EHF from a different protein source can be tried.

Table 7.6 shows the composition of EHF available in the UK. Feed choice may be influenced by

• palatability, which is affected by the presence of bitter peptides. This is particularly important in infants older than 3 months of age
  
• coexisting fat malabsorption, where a feed with some of the fat as medium chain triglycerides (MCT) may be indicated
  
• cost, some hydrolysates being twice as expensive as others
  
• religion and culture, where parents do not wish their children to be given products derived from pork

SHS, Scientific Hospital Supplies.

^* Carbohydrate is present as glucose polymers unless otherwise stated.

^† Figures in parentheses indicate the percentage of fat present as medium chain triglycerides (MCT).

^‡ Suitable from 6 months of age.

^§ Liquid feed, designed for infants <9 kg body weight requiring a nutrient dense feed. Not suitable for cow’s milk allergy.

Introduction of hydrolysate formulas

Hydrolysed protein formulas should be introduced slowly to infants with severe GI symptoms as they have a higher osmolality than normal infant formulas. Feeds containing a high percentage of MCT should also be introduced gradually to ensure tolerance. If the diarrhoea is very severe then it may be necessary to introduce quarter strength feeds, grading up to full strength feeds over 4 days. If severe diarrhoea is present in an older infant it is preferable to stop all solids while a new feed is being introduced to assess tolerance.

In an outpatient setting, where symptoms may be less severe, full strength formula can usually be introduced from the outset. In infants older than 6 months there may be an advantage in initially mixing 1 part EHF to 3 parts their usual formula to slowly introduce the new taste and encourage acceptance, increasing the proportion of EHF over the next 3 days. Milk shake flavourings at 2%–4% concentration (2–4 g per 100 mL) can also be used in this age group if sucrose is not contraindicated.

If an infant refuses to drink the EHF a nasogastric tube needs to be passed to ensure adequate volumes are taken. Where faltering growth coexists, feeds can be fortified in the usual manner by increasing formula concentration or the judicial use of carbohydrate and fat supplements [53] (see Table 1.18). All changes should be made slowly to ensure they are tolerated.

Pepti-Junior, Aptamil Pepti 1 and Althera have sodium contents similar to standard infant formula which may be insufficient for an infant with increased stool losses. Low urinary sodium concentation (<50 mmol/L) alongside a normal plasma sodium concentration may indicate sodium depletion and supplementation with sodium chloride may be required.

Amino acid based infant formulas

Only pure amino acid mixtures are considered to be non-allergenic as there are no peptide chains present to act as epitopes. In infants who fail to tolerate an EHF this is the next logical step, so long as there is not a coexisting fat or carbohydrate intolerance. In these situations a modular feeding approach should be used (see Table 7.24). There are three infant AAFs available in the UK: Alfamino, Neocate LCP and Nutramigen AA and Alfamino (Table 7.7). Neocate Spoon, which contains highly refined rice starch as a thickener, can be offered as a nutrient dense hypoallergenic weaning food if required.

First line treatment with, EHF or AAF

At present there is a paucity of evidence on which to base this decision in GI food allergy. The World Allergy Organisation (DRACMA) report reviewed the few randomised studies comparing AAF with EHF (both whey and casein based) in IgE mediated allergy and could not identify a net benefit of substituting cow’s milk with AAF over EHF. They recognise that the response to feeds in non-IgE mediated cow’s milk allergy may be different and plan to address this in future updates [38].The DRACMA guidelines have been summarised in Table 14.8. ESPGHAN continues to recommend EHF as first line treatment over AAF; however, AAF may be considered as first line therapy in infants with severe enteropathy indicated by hypoproteinaemia and faltering growth. The incidence of infants reacting to EHF is estimated to be <10%; however, this may be higher in infants with enteropathy or multiple food allergies [40]. One retrospective study found that up to 30% of food allergic infants, many with delayed reactions to foods, were intolerant of EHF [17]. Intolerances to hydrolysate formulas resulting in GI disturbances have been described [52].

There is a difference in cost between EHF and AAF with the latter being more than twice as expensive. The Health Improvement Network has shown that EHF was the most effective option in the treatment of cow’s milk allergy, using data from primary care organisations. There were no significant differences in clinical outcomes between the two types of feed [54]. Differences in prescribing patterns can be seen in a single geographical area which shows that current guidelines are not being followed [55].There is an urgent need for a prospective randomised controlled trial looking at the risk of intolerance of EHF in GI allergy.