H. pylori infection

There is a quiet pathophysiology revolution going on in the upper gastrointestinal tract that might have consequences for human health more broadly. The natural state of the human stomach is colonisation with H. pylori, with genetic evidence showing how closely linked the evolution of our two species is.1 In the majority of H. pylori-infected stomachs, the organism acts to reduce gastric acidity through production of urease. In recent years, H. pylori has generally been eradicated whenever found with little thought to the consequences. H. pylori carriage appears to be negatively associated with conditions as broad as allergy, asthma, obesity and acid reflux, all conditions on the rise, suggesting that the health consequences of population-level eradication of H. pylori may be significant.1

Current consensus guidelines on H. pylori in childhood are ambiguous about eradication of the organism when identified incidentally; however, eradication is supported by UK paediatric gastroenterologists when the organism is found, including in an otherwise normal-appearing stomach.2 ,3 The diagnosis and management of H. pylori in children are well described in the joint consensus guideline of European Society of Paediatric Gastroenterology, Hepatology and Nutrition (ESPGHAN) and North American Society of Paediatric Gastroenterology, Hepatology and Nutrition (summarised in box 1),3 with diagnosis having been summarised recently for Archives sister journal.4 Further discussion on H. pylori diagnosis and management therefore is out with the scope of this review.

The emergence of a ‘new’ clinical atopic disease in the oesophagus, EO, around 20 years ago has not yet been fully explained.5 Data are however emerging supporting a negative association between H. pylori and EO, including in children.6 ,7 One challenge of linking other conditions to H. pylori is the confounder of the hygiene hypothesis as atopic conditions have long been linked to increasing societal cleanliness, with H. pylori infection being increased in the opposite circumstance. The impact of a continued reduction in H. pylori prevalence on the health of children warrants significant consideration, particularly in the aetiological study of emerging diseases or those with changing incidence. It is unlikely however that the current decline in H. pylori prevalence can now be halted, and since there is unlikely to be uptake for deliberate inoculation of healthy people with a known carcinogen, there is likely little that can be done to address the unintended pathophysiological sequelae of H. pylori eradication.

Eosinophilic oesophagitis

EO is a primary oesophageal disorder characterised by an eosinophilic inflammatory infiltrate, driven by airborne or food-based allergens and resulting in oesophageal dysfunction. Symptoms vary with age of presentation, becoming more localised and specific with age.8 ,9 Younger children often present with vague symptoms such as feeding disorders, vomiting and abdominal pain, while the more specific symptoms of dysphagia and food impaction are more associated with teenage patients.8 The natural history is not fully understood but may lead to oesophageal stricture in untreated patients (figure 1). Affected patients are often atopic, particularly to foods, with an unexplained almost 3/4 male preponderance.10 ,11 Diagnosis of EO relies on oesophageal biopsy in children symptomatic of oesophageal dysfunction and so vigilance is required in identifying appropriate patients for endoscopic investigation, particularly in general or atopic paediatric practice (box 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1

Natural history of eosinophilic oesophagitis symptoms in childhood.

Confirmation of >15 eosinophils per high-powered field is the current standard diagnostic criterion.12 The oesophagus may look normal on endoscopy and, although there are no pathognomic features, some characteristic findings may be seen macroscopically such as trachealisation. Recent recommendations support taking 2–4 biopsies from the upper and lower oesophagus with important caveats being that gastroesophageal reflux disease is the main differential diagnosis, and endoscopic oesophageal pinch biopsies sample only the superficial mucosa.12

Initial confusion surrounding eosinophilic infiltration in gastroesophageal reflux disease and the early dogma that EO was non-responsive to acid-suppressant therapy have since been addressed with the recognition that a subset of patients with apparent EO are responsive to proton pump inhibitors (PPI).12 ,13 This is pragmatically an important point for the general paediatrician, with symptoms unresponsive to PPI acting as a good marker for tertiary gastroenterology referral (box 2). The concept of PPI responsiveness has since been incorporated into ESPGHAN consensus guidelines with the, perhaps confusing but pragmatic, description of PPI-responsive oesophageal eosinophilia (PPI-ROE).12 It remains unclear whether PPI-ROE and EO are distinct entities; nevertheless, the first step in the management of EO within the recent guidelines is to test for PPI responsiveness, presumably as this then becomes the mainstay therapy for this group of (∼4/10) patients. Confirmation of response should be histological, with one of the major challenges of caring for EO children being a reliance on endoscopy for objective evaluation of disease progression, although change in symptoms can be a useful proxy in some patients. No surrogate biomarker of oesophageal eosinophilic infiltration is in current routine practice.

For children not responsive to PPI treatment, there are two competing options, each with similar efficacy, namely dietary restriction and use of topical steroids. Taking these in turn, dietary therapy is most effective when a restrictive elemental diet is used (96% remission), though this is socially inconvenient and difficult to maintain in the medium to long term.14 Efficacy is reduced but similarly high (81% remission) with a six-food elimination diet excluding milk, egg, soya, wheat, nuts (peanuts and tree nuts) and fish (including shellfish).14 About a third of children will respond to a single food exclusion, and remission rates of 77% can be achieved by milk exclusion in addition to targeted exclusion based on positive skin-prick and allergen patch testing, though the sensitivity/specificity of tests vary considerably with different foods.15 There is currently no randomised controlled trial (RCT) evidence supporting nutritional therapy in EO. Steroid intervention is however supported by specific paediatric RCTs with both fluticasone and budesonide having proven efficacy against placebo when swallowed.16 ,17 ‘Asthma-type’ steroids are used in order to take advantage of topical effect in the oesophagus but high first-pass metabolism and a subsequent reduction in systemic side effects. Swallowed asthma-type steroids that are extensively metabolised by first–pass metabolism (eg, beclomethasone, fluticasone) are particularly useful where aeroallergens are implicated in the development of EO.

The current EO guidelines do not advocate dietary therapy over topical steroids or vice versa, but present a choice after failed response to PPI treatment.12 Systemic steroids are not recommended out with severe dysphagia, dehydration, weight loss, oesophageal strictures and failure of other treatments. Although it is recognised that stricturing can occur in EO, the natural history of this complication has not been fully described, hence the desired outcome in EO management at present remains histological remission, which is not always linked to clinical response. Again, a dependence on endoscopic reassessment is currently required in these children, at least until an adequate non-invasive surrogate marker is widely available.

For the general paediatrician, an awareness of EO and its clinical manifestation and investigation will help in identifying such patients from within cohorts of other atopic disease. The requirement for endoscopy will mean paediatric gastroenterologists are likely to manage the majority of these patients in the long term at present.

Coeliac disease

Coeliac disease continues to remain as a major childhood healthcare problem in the UK. A recent report from Scotland showed that the incidence of coeliac disease has increased from 1.8 to 11.7/100 000 between the period of 1990 to 1994 and 2005 to 2009, respectively.18 This paper showed an actual increase in the number of children presenting with classic symptoms of coeliac disease and also the children diagnosed with coeliac disease by active screening. The increased incidence of coeliac disease over the last three decades has been reported in South Wales as well.19 In this Welsh study, it was noted that >50% of patients exhibited few or no symptoms of coeliac disease and that the median age of diagnosis of coeliac disease had increased.

The British Society of Paediatric Gastroenterology, Hepatology and Nutrition and ESPGHAN have published guidelines recently on the diagnosis of coeliac disease in children.20 ,21 These guidelines allow the diagnosis of coeliac disease in symptomatic children without endoscopy and duodenal biopsy if they have significantly elevated coeliac antibody titres (ie, tissue transglutaminase >10 times upper limit of normal) on two occasions and high-risk human leucocyte antigen (HLA) genotype (HLA DR3-DQ2 or DR4-DQ8). This diagnostic approach without endoscopy and biopsy contrasts with the recently published guideline for diagnosis of coeliac disease in adults, where the recommendation is for biopsy confirmation of coeliac disease.22 It is important to highlight that children should not be started on gluten-free diet for coeliac disease without the diagnosis being confirmed by a paediatric gastroenterologist or paediatrician with gastroenterology interest.

Genetics and dietary exposure of gluten are important factors in the development of coeliac disease. HLA DR3-DQ2 and DR4-DQ8 are the strongest susceptibility loci in the development of coeliac disease. However, not everyone with these HLA genotypes develop coeliac disease. The chance of developing coeliac disease in children with genetic risk depends on the gene dose effect.23 A total of 6403 children with HLA haplotype DR3-DQ2 or DR4-DQ8 were followed up prospectively from birth. The risks of coeliac disease-associated antibody positivity and coeliac disease by the age of 5 years were 11% and 3%, respectively, among children with a single DR3-DQ2 haplotype and 26% and 11%, respectively, among those with two copies (DR3-DQ2 homozygosity). These results show that the DR3-DQ2 haplotype has a gene dose effect.

Strict and lifelong gluten-free diet is the treatment of coeliac disease. Gluten-free diet normalises the small bowel mucosal changes and also reduces the symptoms of coeliac disease. Gluten-free diet is also recommended for asymptomatic patients with coeliac disease. Many asymptomatic patients challenge this advice based on the practical difficulties of adhering to a restricted diet when they cannot perceive the symptomatic improvement. Recent reports from Finland showed that gastrointestinal symptoms, coeliac disease-associated antibodies and mucosal changes improved in asymptomatic adults with coeliac serology positivity when randomised to gluten-free diet compared with the group randomised to gluten-containing diet.24 Only social function scores improved more in the gluten-containing diet group than in the gluten-free diet group. No study subjects considered their experience to be negative and most expected to remain on gluten-free diet.

Research is progressing in the field of non-dietary treatment of coeliac disease. Recently, recombinant gluten-specific proteases have been found to be effective in adult patients with coeliac disease.25 ALV003 contains a prolyl endopeptidase from Sphingomonas capsulate in combination with another endopeptidase from germinating barley. Adults with biopsy-proven coeliac disease were randomised to receive ALV003 or placebo drug along with daily gluten challenge. Gastrointestinal symptoms secondary to gluten ingestion and mucosal injury were greater in the placebo group. There are currently no effective strategies for the prevention of coeliac disease. The Norwegian Mother and Child Cohort Study found an increased risk of coeliac disease in children introduced to gluten after 6 months.26 However, this finding is not replicated in the recently published research by Italian Society of Pediatric Gastroenterology Hepatology and Nutrition, which showed that neither breast feeding nor delayed introduction of gluten prevented the development of coeliac disease among infants at high risk.27 A total of 832 newborn infants who had a first-degree relative with coeliac disease were randomised to introduction of gluten at 6  or 12 months. The study subjects were screened for coeliac disease by serology testing at regular intervals. Even though at 2 years of age coeliac disease incidence was lower in children with delayed gluten introduction, at 5 years of age, coeliac serology positivity or overt coeliac disease was not significantly different between the two groups. A high-risk HLA genotype (HLA DR3-DQ2 or DR4-DQ8) was the most important factor in the development of coeliac disease (box 3).

Inflammatory bowel disease

The diagnosis of IBD has become more commonplace during paediatric years.28 Recent studies have particularly noted a rapid rise in the number of children diagnosed with Crohn's disease making an up-to-date knowledge of the presentation together with the initial investigation and management all the more important. Although the median age at which children are diagnosed with IBD has dropped slightly, clearly the presenting features and examination findings of IBD have not (table 1). Particular attention on examination should be paid to perioral and perianal inspection in children suspected of having Crohn's disease, as well as measurement and plotting of growth parameters in all children.

In children suspected to have IBD, after exclusion of enteric infection (including Clostridium difficile), a panel of simple blood tests (full blood count, erythrocyte sedimentation rate, C-reactive protein, liver function test) will show at least one abnormality in around 80% of children.29 In children where the diagnosis still remains uncertain, the measurement of a faecal inflammatory marker (most commonly faecal calprotectin (FC)) is helpful. Confirmation of a diagnosis of suspected IBD should then be made by a specialist in paediatric gastroenterology using upper and lower gastrointestinal endoscopy with biopsy together with small bowel imaging.30 Recently, the traditional examination of a barium follow-through to assess the small bowel has been replaced by small bowel MRI imaging, usually enterography.

Role of FC in the diagnosis of IBD

FC is a neutrophil marker that can be easily and relatively inexpensively measured in serum and stool. Its great clinical utility is being able to differentiate between non-infectious diarrhoea and IBD (box 4). Studies have clearly demonstrated that the median value at paediatric IBD diagnosis is higher than that of healthy paediatric controls who have been shown conclusively not to have IBD by endoscopy and biopsy.31 In this study, a median value of 1265 µg/g was found in treatment naïve patients with IBD at diagnosis compared with controls who had a value of 65 µg/g. The exact value that signifies when a patient should have further investigation for IBD is not universally agreed, but in working practice the authors would use a value of >250 µg/g to signify the need for follow-up or further investigations.32 The reported normal range of FC is <50 µg/g, but many other conditions in paediatrics can cause a raised value and it is extremely important that the test is only used when clinically appropriate and in the correct clinical context to avoid over-investigating or inappropriate investigations. Common causes of significantly raised calprotectin that can be confused symptomatically with IBD are enteric infection (diarrhoea±blood) and polyps (blood). Young children under the age of 4 normally have a calprotectin level of several hundred in health so clinicians should be aware of this; it would be unusual to need to do a FC in a patient of this age anyway as IBD is very unusual in the very young so the need for further investigations like endoscopy is usually based more on symptoms and other results, not FC.

Scientific advances in the understanding of IBD

The well-known familial risk of IBD with an approximate lifetime risk of 5–10% in first-degree relatives was the initial inspiration behind the IBD genetics studies of the past two decades. Most recently, these studies have confirmed genetic susceptibility in genes important in controlling several key biological pathways including autophagy, innate immunity and lymphocyte regulation pathways.33 The most recent studies have involved international consortia analysing samples from tens of thousands of patients and controls. There are now potentially >160 genes implicated in IBD pathogenesis.33 There is no major difference between the genetics of IBD in children and adults despite the fact that children have a more extensive disease phenotype. This similarity excludes patients with very early onset disease under age 6 and especially under age 2 who have often been found to have a monogenic cause for their IBD rather than the polygenic disease of older children and adults.34 These scientific advances have not been widely translated into clinical practice in contrast to targeted pharmacogenomic studies that look to represent an area of exciting clinical application of these recent genetic advances.35

The genetic work has been paralleled and perhaps overtaken by studies examining changes in the microbiome prior to and during IBD disease course. Indeed, the largest genetic IBD study undertaken to date in IBD has placed genetic response to the gut microbiome at the core of IBD pathogenesis.36 The altered gut microbiome at IBD diagnosis with an imbalance of the microbiota compared with healthy controls is generally described by the term ‘dysbiosis’.37 Crohn's disease is likely to result from a reduced bacterial diversity within the gut microbiome but an increase in the total number of bacteria compared with healthy controls. Scientific evidence exists for specific species in the aetiology of these changes, for example, pathogenic strains of Escherichia coli, but in truth the key species driving these changes vary between study populations and methodologies so the search for the ‘gatekeeper’ continues. There are active research programmes looking at how these bacterial populations can be changed clinically using, for example, exclusive enteral nutrition in Crohn's disease and faecal transplantation in ulcerative colitis. Microbial therapeutics looks likely to become the next revolution in IBD therapy.

The management of IBD has been extensively updated recently with several joint publications from ESPGHAN and European Crohn's and Colitis Organisation in the past few years and hence not covered in this review.