There are significant differences between paediatric and adult gastrointestinal (GI) disease, and even the normal GI histology shows some variation. Excess eosinophils in the gut can be present in food allergies (frequent in children) but also in eosinophilic gastroenteritis. Inflammatory bowel disease (IBD) has some distinctive features when it presents in children. Genetic factors may play a greater role in pathogenesis, and the combination of clinical and histological criteria required for diagnosing Crohn’s Disease or ulcerative colitis (UC) is different. Atypical UC patterns are more common and clinical presentation of Crohn’s disease can also be different; the frequency of granulomas in Crohn’s disease is significantly higher in children. Inflammatory Bowel Disease Unclassified (IBDU) has its highest frequency in younger patients, possibly because of atypical characteristics of paediatric UC. Monogenic forms of IBD-like colitis typically develop during infancy or early childhood, an can have features of Crohn’s disease, UC or IBD unclassified. Histology is often indistinguishable from conventional IBD. Necrotising enterocolitis (NEC) primarily affects the small intestine of premature infants, with haemorrhagic necrosis of the bowel wall. Fibrosing colonopathy has been reported in children exposed to high doses of pancreatic enzymes and is characterised by bowel wall thickening, submucosal fibrosis and chronic mucosal inflammation.
There are significant differences between paediatric and adult gastrointestinal (GI) disease, and even the normal histology shows some variation between adults and children. In addition, some non-neoplastic lesions of the GI tract are typical of, or exclusive to, paediatric subjects. Moreover, in some conditions such as inflammatory bowel disease (IBD), the clinical and pathological features can be significantly different because the gut mucosa seems to react somewhat differently in children. Thus, pathologists should be aware of the particular histopathological aspects of paediatric conditions in order to minimise the risk of diagnostic error and of delayed or inadequate therapy.
Normal Paediatric Mucosa
Although at first sight the paediatric GI mucosa may seem identical to adult mucosa in a routine haematoxylin and eosin (H&E) slide, there are some differences (Fact Sheet 7.1).
The extent of cardiac mucosa can vary.
The morphology of small bowel villi is slightly different.
Lymphoid aggregates in the small and large bowel are more numerous.
The density of the myenteric plexus and of ganglion cells in the myenteric plexus is higher in very young children.
In the upper GI mucosa, the main distinctive feature in children concerns the characteristics of the gastro-oesophageal transition zone. In the fetus, this zone has no deep glands before 21 weeks of gestation. Mixed mucous-oxyntic glands develop after 21 weeks and cardiac mucosa is present in about 50% of subjects after the first postnatal week.1, 2 Thus, cardiac mucosa is probably a gastric mucosal derivative, rather than a metaplastic condition representing an early manifestation of gastro-oesophageal reflux in the lower squamous oesophagus (as some authors claim).3 In a study of 30 autopsies of patients under 18 years of age, cardiac mucosa was always present but the segment was usually very short, with an average length of less than 2 mm.4 Therefore, this may be a ‘dynamic’ mucosa that alters during development and reacts variably to reflux and to other types of injury, particularly in children.
Gastric mucosa seems also to react differently in children. In a small series of six children with pancreatic metaplasia of the gastric mucosa, all biopsies were negative for Helicobacter pylori and none showed significant inflammation.5 The occurrence of pancreatic metaplasia in children who do not have atrophic gastritis raises the possibility that it may be a developmental phenomenon related to gastric mucosal differentiation. In contrast, pancreatic metaplasia of the adult gastric mucosa is strongly associated with chronic atrophic gastritis.
The intestinal mucosa shows some differences in morphology in children, particularly in the small bowel. The ultrastructure of the small intestine in healthy children is slightly different from the appearance in adults. Ridge-shaped villi predominate in infants, while the villi usually resemble fingers or leaves in children above 3 years of age and in adults.6 Villi tend to be shorter and the crypts longer in children.
The cellularity of the intestinal lamina propria also shows differences. An age-related decline in lymphocytes and mast cells may occur in normal large bowel mucosa. However, the gradient between sites persists: the lamina propria in the right colon is more cellular than in the left colon and rectum and includes lymphocytes and fairly numerous plasma cells and eosinophils, while in the left colon there are fewer eosinophils and only superficial plasma cells.7 Lymphoid aggregates are a normal component of colorectal mucosa and are frequently associated with branching of overlying crypts. The number of lymphoid aggregates in children is higher in both small and large intestines, probably reflecting differences in the innate immune system (Figure 7.1). Aggregates of lymphoid follicles are scattered throughout the small intestine but have their highest concentration within the ileum, where they are designated Peyer’s patches. Peyer’s patches are more prominent during infancy and childhood than they are in adulthood.8 Sometimes termed lymphoid nodular hyperplasia when more exuberant, these aggregates are especially numerous in children under 10 years of age, particularly in the terminal ileum but also in the rectum and colon. They are not usually associated with clinical symptoms.9 They may represent a state of enhanced non-specific immune response that may be ‘physiological’ in this age group.10.11 The exact incidence of lymphoid nodular hyperplasia in adults is unknown, and definitions vary, but it is probably rare if it does exist.12
Figure 7.1 Lymphoid aggregates in the terminal ileum. The number of lymphoid aggregates is higher in children, both in the small and large intestine.
There are also some age-related differences in the myenteric plexus of normal human bowel. The myenteric plexus meshwork may become less dense during the first years of life, and the density of ganglion cells within the myenteric plexus may decrease significantly during the first 3–4 years.13
Gut microbiota in children are different from those in adults. Bacterial colonisation influences the development of the infant intestine. Microbial colonisation runs in parallel with immune system maturation and plays a role in intestinal physiology and regulation.14 The gut microbial community is dynamic during the first years of life, before stabilising to an adult-like state by 3 or 4 years of age.14–18 This difference may influence the many ways in which the gut reacts to various stimuli and insults, as microbiota/host interactions have an important influence on human health and disease.
Allergic conditions of the GI tract include a spectrum of disorders that result from adverse immune responses to dietary antigens. These conditions include immediate GI hypersensitivity (anaphylaxis), oral allergy syndrome,19 allergic colitis/ proctocolitis/enterocolitis (adverse food reactions), allergic eosinophilic esophagitis, eosinophilic gastritis/gastroenterocolitis, and coeliac disease.20
Allergic Colitis and Enterocolitis
GI food allergies are quite common in infants and children21 and include allergic colitis (AC), allergic enterocolitis, and food protein enteropathies (Fact Sheet 7.2). The term ‘food allergy’ refers to an immune reaction (immunoglobulin [Ig] E-mediated, cell-mediated, or both21) that develops in response to the ingestion of a specific type of food. In IgE-mediated disorders, symptoms in other target organs are common. In cell-mediated disorders, the symptoms are usually localised to the gut. These disorders are usually transient and may have variable repercussions on the nutritional state of the patient.
GI food allergies are frequent in children and can cause allergic enterocolitis/allergic colitis
The rectum and sigmoid are affected most often
Eosinophil infiltrates (≥60 per 10 hpfs) occur in all mucosal compartments, particularly in the lamina propria
Other causes of eosinophilia require exclusion
Histological criteria proposed by Odze et al.: the diagnosis is likely if all four are present in the appropriate clinical setting:
Increase in lamina propria eosinophils (≥60 per 10 hpfs), often in association with lymphoid aggregates
Epithelial or muscularis mucosae eosinophil infiltration ( ≥1 eosinophil per hpf)
A mixture of eosinophils and neutrophils in crypt abscesses
Absence of specific features (parasites, ova, crypt architectural abnormalities) that might suggest an alternative diagnosis
Allergic colitis and enterocolitis typically affect infants and adolescents, and most infantile cases are probably the result of an allergy to cow’s milk protein. However, many foods have been associated with the development of AC.22, 23 The patients usually present with rectal bleeding (with or without diarrhoea) and generally recover promptly after elimination of the allergen from their diet.24, 25 Interestingly, eosinophilic colitis may occur more often in infants that are exclusively breast fed (probably because of the maternal diet). Although cow’s milk–induced allergy is relatively common, with an estimated prevalence between 2% and 7.5% in otherwise normal infants, the exact prevalence of AC overall is unknown.23, 26
The clinical diagnosis of AC can be difficult. Although a careful history detailing the relationship of symptom onset to feeding (as well as a personal or family history of allergies) may be helpful, the presence or absence of these features has no proven predictive value in identifying infants with AC.27 Symptoms may include abdominal pain, anorexia, and weight loss, but bleeding is the most common sign. Peripheral blood eosinophilia and stool eosinophils may be present. Most affected infants lack constitutional symptoms and are otherwise healthy. Sigmoidoscopy, in conjunction with the evaluation of multiple mucosal biopsy specimens, may be helpful in confirming the diagnosis, particularly in patients with severe disease and in patients whose condition does not improve after the introduction of an elimination diet.
Lesions in AC may involve any segment of the colon. They most often affect the rectum and sigmoid and the changes are usually most severe in this region. Typically, the abnormal areas are separated by intervening zones of mucosa that look entirely normal or show focal erythema. The abnormal mucosa may appear friable with increased nodularity suggestive of lymphoid hyperplasia. More severe cases may show decreased mucosal vascularity, multiple superficial erosions (aphthous type with erythematous margins), or, rarely, frank ulceration with surface exudate mimicking the endoscopic features of infectious colitis.24, 28
Microscopically, the most remarkable and characteristic histological feature is the presence of infiltrates of eosinophils in all mucosal compartments, but particularly in the lamina propria (≥60 eosinophils per 10 high-power fields [hpfs]), as well as numerous intact or degranulated eosinophils located at the base of the mucosa and interspersed among muscle fibres of the muscularis mucosae26 (Figure 7.2). Crypt abscesses also occur, but in AC are typically composed of both neutrophils and eosinophils. Crypt architecture is normal, without the features of chronicity typical of IBD (i.e. distorted, branched, or atrophic crypts, Paneth cell metaplasia, basally located lymphoid aggregates, and/or basal plasmacytosis). However, the features are not specific and AC is, in effect, a diagnosis of exclusion.
(A) Inflammatory infiltrate in the lamina propria, including numerous eosinophils (intact or degranulated).
(B) Crypt abscesses can also be present, typically including both neutrophils and eosinophils.
There are no consensus criteria on how many eosinophils are necessary to make a diagnosis of AC. Machida et al proposed a threshold of>20 eosinophils per hpf.29 Odze et al., on the basis of the findings in more than one study, proposed different histologic criteria: (1) an increase in the number of eosinophils in the lamina propria (≥60 per 10 hpfs), often seen in close association with lymphoid aggregates; (2) epithelial (crypt or surface) or muscularis mucosae eosinophil infiltration (≥1 eosinophil per hpf); (3) a mixture of eosinophils and neutrophils in crypt abscesses; and (4) the absence of other specific histological features (parasites, ova, crypt architectural abnormalities) that would suggest an alternative diagnosis.26, 30 According to the authors, the presence of all four histological criteria in a specimen strongly supports the diagnosis of AC if encountered in the appropriate clinical setting.
Involvement by eosinophilic infiltrates of other deeper layers of the GI wall in the absence of an identified cause should suggest a diagnosis of primary eosinophilic gastroenteritis, a different entity that is rarer in paediatric age groups.
Eosinophilic gastroenteritis (EGE) is an infrequent disease of unknown aetiology, is both IgE- and cell-mediated, and is characterised by focal or diffuse eosinophilic infiltration of the GI tract (Fact Sheet 7.3). The disease occurs over a wide age range – from infancy to the seventh decade – but usually presents between the second and fifth decades.31–33 The frequency of its diagnosis in children is increasing. There are no significant differences between childhood and adult forms. When the disease manifests in infancy, and if specific food sensitivities are identified, the likelihood of remission by late childhood is high. GI obstruction is the most common complication.
Stomach> small intestine> colon
Focal or diffuse eosinophilic infiltration of the GI tract
Increased numbers of eosinophils in the lamina propria (>50 per hpf)
Eosinophils often present in the muscularis and serosa
+/− Architectural mucosal changes
Eosinophilic infiltration of GI tract
Exclusion of parasitic disease
Absence of other systemic involvement
The clinical presentation is variable and includes abdominal pain, weight loss, vomiting, and diarrhoea. The traditional classification of EGE identifies three forms, depending on the depth of inflammation in the wall: mucosal, muscular, or serosal. This, together with the digestive tract segments involved, determines the clinical presentation. The stomach is the most common site, followed by small intestine and colon.34 Fatalities are rare.
There is limited understanding of the aetiology and pathogenesis. Peripheral eosinophilia, abundant eosinophils in the GI tract, and a dramatic response to steroids may constitute evidence that the disease is mediated by a hypersensitivity reaction.35 The damage to the GI tract wall seems to be a result of eosinophilic infiltration and degranulation.36 In EGE, food allergens may cross the intestinal mucosa and trigger an inflammatory response, with cytokines, interleukin (IL)-3, IL-5, and granulocyte macrophage colony stimulating factor facilitating the recruitment and activation of eosinophils.37 In fact, many patients have a history of food allergy and other atopic conditions such as eczema or asthma.38
Four criteria are required for the diagnosis of eosinophilic gastroenteritis: the presence of GI symptoms, eosinophilic infiltration of GI tract, exclusion of parasitic disease, and absence of other systemic involvement. The presence of peripheral eosinophilia is not a universal phenomenon.
The endoscopic appearance is not specific and includes erythematous, friable, nodular, and occasionally ulcerative changes. Patients with serosal disease present with ascites, and abdominal paracentesis demonstrates a sterile fluid with a high eosinophil count. Pleural effusion also may be present.33
The correct diagnosis depends mainly on the histological findings. Microscopy demonstrates increased numbers of eosinophils (often>50 per hpf) in the lamina propria,39 with large numbers in the muscularis and serosa in many cases (Figure 7.3). Localised eosinophilic infiltrates may cause crypt hyperplasia, epithelial cell necrosis, and small intestinal villous atrophy. There may be a diffuse enteritis with complete loss of villi, submucosal oedema, infiltration of the GI wall, and fibrosis. Mast cell infiltrates and hyperplastic mesenteric lymph nodes infiltrated with eosinophils may be present.40, 41 In superficial colonic biopsies, this condition can be very difficult to distinguish from allergic colitis. Moreover, it may be difficult to detect or completely absent because infiltration is often patchy. Laparoscopic full thickness biopsy may be necessary for histological diagnosis if there is no mucosal involvement.40
Figure 7.3 Eosinophilic gastroenteritis: an increase in numbers of eosinophils (often>50 per hpf) in the lamina propria but also in deeper layers (submucosa).
When eosinophilic gastroenteritis coexists with eosinophilic infiltration of other systems, the diagnosis of idiopathic hypereosinophilic syndrome is worth considering.39
Inflammatory Bowel Disease in Children
Inflammatory bowel disease is an important cause of GI pathology in children and adolescents. Its incidence in the paediatric population has apparently increased over the last few years in Western countries, and there has been a reduction in the median age of diagnosis of new IBD. Of those with new IBD, 25%–30% with Crohn’s disease and 20% with ulcerative colitis (UC) are less than 20 years old, and 10%–15% of IBD is diagnosed before the age of 18.42–47
Although there are many similarities between paediatric and adult-onset IBD, there are several differences as regards clinical and pathological features (Fact Sheet 7.4). The pathologist should be aware of the differences because this will help reduce the risk of diagnostic error, delayed or inadequate therapy, and other adverse clinical consequences. As in the adult population, a correct diagnosis relies on a multidisciplinary approach, taking into account the clinical, radiological, endoscopic, and histopathological data.48
Genetic factors may play a greater role in the pathogenesis of inflammatory bowel disease in children.
The clinical and pathological criteria for a diagnosis of Crohn’s disease or ulcerative colitis are not exactly the same.
The underlying pathogenesis of IBD in adults and in children/adolescents appears to be similar, representing the outcome of a complex and poorly understood interaction of environmental, genetic, and immune factors. However, genetics may play a greater role in disease onset and susceptibility in children, who may have had less exposure to environmental factors.49, 50 In fact, age at onset can provide information about the type of IBD and its associated genetic features. Early postnatal onset forms of IBD may reflect monogenetic causes, as suggested by the demonstration of IL10 signalling defects in a group of patients with very early onset IBD (IBD presenting within the first few months of life).51
Initial evaluation of children and adolescents with suspected IBD should aim to determine whether IBD is in fact present, and to differentiate between UC and Crohn’s disease.52 In young children with an aberrant presentation of colitis, IBD, and particularly UC, are worth considering even if the pathology is not typical.53
To establish a correct histological diagnosis of IBD, the pathologist should evaluate multiple microscopic features (including architectural changes, epithelial cell abnormalities, and inflammatory features) (Figure 7.4) and correlate the findings with clinical signs and endoscopic appearances. The combination of architectural changes and basal plasmacytosis (defined as an excess of plasma cells at the base of the mucosa, often between the base of the crypts and the muscularis mucosae) favours a diagnosis of chronicity and helps in the differential diagnosis between IBD and acute self-limiting colitis. Recent guidelines focusing on histological criteria for the diagnosis of IBD depend mainly on data from adult patients.54 However, the combination of clinical and histological criteria needed to formulate a diagnosis of Crohn’s disease or UC is not always the same in adults and children.
Figure 7.4 IBD in children: architectural changes, epithelial cell abnormalities and active inflammation. The chronic histological changes of ulcerative colitis may be less severe in children than in adults.
Colonic mucosal biopsies from children with UC, especially those under 10 years of age, may show a lesser degree of abnormality than biopsies from adults with fewer features of chronicity. According to a study by Washington et al. of initial rectal biopsies, follow-up biopsies and resections from children and adults with UC, initial biopsies are less likely in children than in adults to show diffuse architectural abnormalities (including villiform surface, crypt atrophy, and branching crypts). In contrast, lamina propria inflammation, crypt abscesses (defined as the presence of neutrophils within crypt lumens) and cryptitis (defined as the presence of neutrophils within crypt epithelium) are equally common in both groups.55 In a study by Glickman et al., biopsies from children under 10 years of age showed in some cases only minimal evidence of architectural distortion and only a slight increase in lamina propria plasma cells.56 These children had significantly lower scores than adults and older children in every category of histological abnormality except for basal plasmacytosis. In addition, features of activity, e.g. cryptitis and crypt abscesses, and evidence of epithelial injury, e.g. mucin depletion, cuboidal shape, nuclear enlargement, loss of surface cells, erosion, and ulceration, seemed in this study to be significantly less prevalent in children than in adults. These features of paediatric UC mucosa can make the distinction from acute self-limited colitis more difficult than in adults, especially if the pathologist is unaware of the differences.
As children approach adulthood, the degree of inflammation and architectural distortion is similar to that found in adults.57 Indeed, the distinction between younger and older children seems to be important, with the age of 10 perhaps representing a useful threshold.
Paediatric Ulcerative Colitis
Classically, UC is a continuous inflammatory process, beginning in the rectum and extending proximally.54 Atypical patterns of inflammation in UC such as rectal sparing, patchiness, periappendiceal inflammation, and upper GI inflammation (that are usually more strongly associated with Crohn’s disease) are probably more common in children than in adults (Fact Sheet 7.5 and Table 7.1). Although most patients with UC have convincing histological features of chronic colitis on mucosal biopsy of the rectum, several studies have shown that paediatric patients with new-onset untreated UC may sometimes show relative or complete rectal sparing or patchy colitis.55–58 In one report, initial rectal biopsies from 5 of 12 children with UC showed only mild patchy inflammation or normal mucosa.59 Another study of UC in children showed patchiness of microscopic features of chronicity (21% of patients) and relative (23%) or absolute (3%) rectal sparing,56 and in another series only 66% of rectal biopsies from paediatric patients with new-onset UC had architectural features of chronic colitis.55
Subtotal colitis or pancolitis
Rectal sparing (according to some reports)
Upper GI involvement
Biopsies may show
A lesser degree of abnormality
Fewer features of chronicity
Distinction of UC from acute self-limited colitis may be more difficult
|Adult UC||Childhood UC|
|Microscopic ileitis||Typically absent||Common (no granuloma)|
|Microscopic gastritis||Typically absent||Rare (no granuloma)|
|Relative rectal sparing||Usually only after treatment||May be present at diagnosis|
|Patchiness||Only after treatment||May be present at diagnosis|
|Crypt distortion||Often prominent||Often less pronounced|
The extent and severity of the disease is also apparently different among children and adults, a higher proportion of children with UC presenting initially with subtotal colitis or pancolitis,60 e.g., 42% of children compared to 11% adults in one report.56 In another study, there was left-sided colitis (involvement extending no further proximally than the splenic flexure) at presentation in 10% and pancolitis in 90%.60 The EUROKIDS Registry reported extensive colitis or pancolitis in 77% of children.60
Upper GI inflammation (see also Chapter 16) may occur in children with UC but is not common. Nevertheless, UC remains a possible cause of upper GI inflammation in the setting of known IBD. In the upper GI tract, few features discriminate reliably between UC and Crohn’s disease except for non-cryptolytic granulomas, which strongly favour Crohn’s disease. Recommended investigations in paediatric IBD include both upper GI endoscopy and ileocolonoscopy with multiple biopsies, while upper GI endoscopy is not routine in adults. Indeed, upper GI endoscopy with biopsy is standard practice in all children irrespective of the presence or absence of upper GI symptoms.61 Non-specific inflammation in the upper GI tract may be present in up to 75% of children with UC, although the mechanism(s) underlying these abnormalities are unclear and the true prevalence of upper GI involvement by UC is very difficult to determine. For example, oesophagitis, minimal to mild non-specific gastritis or focally enhanced gastritis (FEG) can be present.62
FEG is an inflammatory gastric lesion defined as the presence of one or more foci of pit or glandular inflammation containing lymphocytes, histiocytes, and granulocytes, with a relatively normal background mucosa63 (Figure 7.5). The inflammatory cells may infiltrate the epithelium. FEG in adults seems to be a non-specific finding, but in paediatric patients is associated with a higher likelihood of a future diagnosis of IBD. The number of FEG foci in a review of 119 gastric biopsies from paediatric IBD patients ranged from 1 to 5 (mean 2.2) in each biopsy specimen, while each FEG lesion involved from 1 to 12 pits/glands, and mostly up to 3 pits/glands (87%). The total number of glands involved as part of a focus of FEG was higher in patients with UC than in Crohn’s disease (6.4 ± 5.1) but the lesion was also present in patients with Crohn’s disease (4.0 ± 3.0).64
Figure 7.5 Focally enhanced gastritis (FEG) is characterised by one or more foci of inflammation including lymphocytes and histiocytes, with or without neutrophils, around pits or glands with a relatively normal background mucosa. Inflammatory cells may also extend into the epithelium.
Unlike FEG, granulomas in the upper GI tract are a specific feature of Crohn’s disease in all series65, 66 and upper GI granulomas in Crohn’s disease are more common in children than in adults (see Chapter 16). Histological abnormalities of the upper GI tract include other lesions besides FEG and granulomas. Some may be more common in children than in adults. For example, gastroduodenal ulceration, villous atrophy, increased intraepithelial lymphocytes, and cryptitis may be more common in UC paediatric patients.66–68
Backwash ileitis occurs in approximately 25% of adults with pancolonic UC, but in children the prevalence is unknown. In one report, 39% of children presenting with new UC (39%) showed terminal ileal erythema with associated nonspecific inflammation and no erosions or ulcers.69 A caecal patch occurred in 2% of patients in the EUROKIDS Registry, in contrast to the frequency of 12%–25% or higher described in adults, although frequencies vary widely.52, 70–72
Paediatric Crohn’s disease
Extensive inflammatory involvement is often already present during infancy or early childhood present in children and adolescents with Crohn’s disease at disease onset (CEDATA-GPGE registry data).73 Children with Crohn’s disease have a greater tendency than adults to present with isolated colonic disease (L2 Paris classification), whereas adolescents and adults are more likely to have ileal disease (L1).72, 73 Isolated ileocaecal involvement, which is typical in adulthood, is rare in children and adolescents, occurring in only 13.3%.73 Terminal ileal disease and stricturing disease are more common in children diagnosed after 10 years of age than in younger patients, approaching the frequency seen in adults.74
The frequency of upper GI tract involvement in Crohn’s disease (see also Chapter 16) seems also to be higher in children. Though difficult to determine precisely, the prevalence in adult Crohn’s disease is low in most reports (0.3%–16%) while oesophageal, gastric, or duodenal pathology may be identified in up to 30% of paediatric Crohn’s disease patients.75 Histology of the upper GI tract (showing specific lesions such as granulomas) may confirm a diagnosis of Crohn’s disease in 11%–29% of cases that was otherwise not obvious76, 77 (Figure 7.6). Focal cryptitis of the duodenum and focally enhanced gastritis occurred significantly more frequently in children with Crohn’s disease than in children with UC and non-IBD.78
Figure 7.6 Upper GI tract in Crohn’s disease with granuloma in duodenal mucosa.
The frequency of granulomas in Crohn’s colitis is higher in children than in adults, and this is not entirely attributable to the higher number of colonoscopic biopsies (Table 7.2).79 Compared with adults, the incidence of granulomas may be twice as high in children (Figure 7.7).79, 80 The difference could be a result of evolution and regression of granulomas at different time intervals during the course of the disease. This could in turn reflect an immunological reaction by immature tissue, i.e. in children, when challenged by the (currently elusive) aetiological agent(s) responsible for Crohn’s disease. In the EUROKIDS registry, there were granulomas in 43% of paediatric patients.74