Original Articles: Gastroenterology
In our surgical unit, we observed that the presence of eosinophilia (EOS) and chronic inflammatory infiltrate (CINF) is incidentally reported in rectal biopsies of children investigated for suspected Hirschsprung disease (HSCR), although the relevance of these findings is not clear. In the present study, we aimed to correlate the clinical symptoms with the incidental findings of EOS and CINF. There has been an increasing interest in defining the role of inflammatory infiltrate in GI disorders, both in adults and children, and inflammatory infiltrates, consisting of eosinophils and plasma cells, have been most commonly described in association with malignancy, infection, allergic disease, and intestinal pseudo-obstruction (1), but there is scanty information in children. Eosinophilic gastroenteritis was first described by Kaijser in 1937 (2) and since then several studies have focused on the distribution and significance of the eosinophils in the paediatric gastrointestinal (GI) tract. Recent studies (3–5) have documented an increasing prevalence of eosinophil-associated GI disorders parallel to the increased prevalence of allergic disorders (6). Allergy can cause inflammation of the rectal wall and the resulting eosinophilic infiltration leads to changes in the mucus composition, dysmotility, and ultimately to constipation (7). Cow's-milk protein allergy particularly has been associated with inflammation of the rectal mucosa and chronic constipation in children (8,9). DeBrosse et al (10) noticed that patients with a history of atopy have a trend towards greater eosinophilic infiltrates in the colon compared with those without history of atopy. Eosinophils infiltrating the intestinal surface and crypt epithelium have also been observed in inflammatory bowel disease, although the real implication is still unclear (11,12). In our unit, EOS and CINF have been incidentally reported by the pathologists in rectal biopsies of children investigated for HSCR in the neonatal period or in older children with chronic constipation. The significance of these findings is unknown and there are no studies that have evaluated whether further management is required. Our aim was to evaluate whether the incidental findings of EOS and CINF in rectal biopsies of children require further management.
Following approval by the Cambridge University Hospital research ethics committee, a retrospective study of 364 children that underwent rectal biopsy for suspected HSCR (either for failure of passage of meconium in the neonatal period or for chronic constipation), between 2000 and 2010, was performed. Biopsies were either taken with a “suction” technique (rbi2 Aus Systems Pty Ltd, Allenby Gardens, Australia) in children younger than 3 months or via an open (full thickness) technique under general anaesthesia in older children. The suction biopsies were confirmed to have mucosa and submucosa to be comparable with the open (full thickness) biopsies. The biopsies were stained with haematoxylin and eosin and no specific neuronal or lymphocyte staining was performed. Histological findings were reviewed in all of the patients, who were then divided into 3 groups: patients with no ganglia (confirmed HSCR); patients with normal ganglia but with EOS and/or CINF (EOS/CINF group); patients with normal biopsy (normal ganglia and no EOS or CINF).
Children were followed up either in the surgical clinic or paediatric gastroenterology clinic. Patients’ characteristics and weight were reviewed from inpatient and outpatient records. All of the patients had weight measured by standard balance scales and recorded at the time of biopsy and at follow-up. To identify those children with failure to thrive, weight standard deviation scores (weight z score) were calculated for each individual patient and were derived using British 1990 standards (13). Incidence of GI symptoms (ie, constipation, diarrhoea, recurrent abdominal pain), and incidence of food/milk allergy were reviewed. Data, reported as median (range) and mean ± standard error mean, were compared by Fisher exact test, Mann-Whitney test, and paired t test.
A total of 364 children underwent rectal biopsy and 109 (41%) were confirmed to have HSCR (Fig. 1). The remaining 255 children were found to have normal ganglion cells.
Patients With EOS/CINF
Forty-four (17%) of the 255 had EOS or CINF (Fig. 1, Table 1); 3 patients had mixed eosinophilia and CINF. There were 24 boys (54%). The biopsy was carried out in the neonatal period for delayed passage of meconium and intestinal obstruction in 13 (29%) children. The remaining 31 children underwent rectal biopsy at a median of 21 months (2–145) for chronic constipation. In 18 (64%) children with EOS and 12 (75%) children with CINF, there was delayed passage of meconium for >24 hours (P = 0.5). The inflammatory infiltrate (Fig. 2A and B) was found in different sites (Table 2) including lamina propria, crypt epithelium, and submucosa. The median number of cells per high-power field (HPF) was 8 (4–15). Age at biopsy was significantly lower in children with EOS compared to children with CINF (P = 0.009). Follow-up was similar between the 2 groups (P = 0.6). Twenty-one (48%) were found to have food and/or milk allergy and 30 (68%) had persistent constipation and/or other GI symptoms at follow-up. Only 3 patients of the 13 that had the biopsy taken in the neonatal period continued to have constipation at follow-up (P = 0.0001 vs older children). At biopsy, patients with EOS or CINF had impaired growth, with a weight-for-age z score of −0.68 ± 0.25; 7 patients (18%) had failure to thrive (ie, a weight-for-age z score ≤−2.0) at biopsy. There was no significant catch-up growth (ie, increase in weight-for-age z score; Fig. 3) at follow-up (P = 0.8), and 8 (21%) had failure to thrive.
Patients With Normal Biopsy
From the 211 children with normal biopsies, 44 were selected, matching for age and weight at biopsy and for sex with the 44 patients with EOS and CINF. Ten of these 44 children had persisting constipation at follow-up (P = 0.0001 vs patients with EOS/CINF) and only 1 patient was found to have atopy (P < 0.0001 vs patients with EOS/CINF). Four patients with persisting constipation were referred for psychological input because of associated behavioural problems. Growth was impaired at the time of biopsy, with a weight-for-age z score of −0.70 ± 0.25; 7 patients (18%) had failure to thrive, similar to those with EOS of CINF. However, in patients with normal biopsy, was significant catch-up growth, demonstrated by an increase in the weight-for-age z score at follow-up (P = 0.003), with only 3 patients (7%) continuing to have failure to thrive (Fig. 4).
We have observed that in our unit the findings of EOS and CINF are incidentally reported by the pathologists in rectal biopsy of children investigated for HSCR, either in the neonatal period for failed passage of meconium or in infancy for chronic constipation. Nonetheless, the clinical relevance of these findings is not clear. Eosinophilic infiltrates have been found in association with different GI conditions, including eosinophilic gastroenteritis, coeliac disease, inflammatory bowel diseases, drug-induced colitis, acute GI radiation injury, eosinophilic cryptitis, and autoimmune disorders (14,15,17,18). Furthermore, the presence of an eosinophilic infiltrate is common in biopsies taken from caecum and ascending colon (14–16), but there has been no study to document whether these findings in rectal biopsies of children require further management. We found a specific distribution of inflammatory infiltrates with eosinophils predominantly localised in the mucosa and the CINF mainly localised in the lamina propria. The implication of this finding is not clear, but it has been well documented that in normal subjects, eosinophils are present only in the lamina propria of the stomach, small intestine, caecum, and colon (4) but not in the mucosa; however, eosinophilic infiltration of the colonic mucosa, lamina propria, and muscularis is demonstrated in allergic proctocolitis in children (19). Eosinophils are also known to extend into submucosa in cow's-milk protein allergy, whereas lymphocytes extend into submucosa in ulcerative colitis and Crohn disease; however, in less severe conditions they may be confined to the mucosa. The presence of eosinophils in the surface and crypt epithelium has also been observed in inflammatory bowel disease (12) and focal eosinophilic infiltration in Crohn disease is an important parameter in the histological differentiation between colonic Crohn disease and ulcerative colitis. None of our patients was diagnosed with inflammatory bowel disease and this could be due to the young age of our population (median of 5.3 months) and the relative short follow-up period (median of 4.6 months), but we suggest that these children are possibly at risk for developing inflammatory bowel disease later in life and should be closely monitored. Another controversial issue is the number of inflammatory cells that should be present in normal subjects and, although we found a median of 6 cells per HPF, there is lack of consensus regarding the number of inflammatory cells to define EOS, and geographic variations in the number of eosinophils in the GI tract have also been described (20). Large numbers of eosinophils located in the muscularis mucosae or in crypt abscesses have been found to be significantly associated with allergy-related disease (25); however, Behjati et al (21) found no correlation between the change in number of colonic eosinophils and clinical symptoms. Similarly, other authors (12) found no correlation between total blood eosinophilic count, positivity of allergy tests, allergic history, and number of eosinophils in the biopsies. These authors concluded that in children with irritable bowel syndrome and patients with a variety of other diagnoses, including allergies, the distribution is mostly superficial, with a lower total cellularity.
Our study documents a significant incidence of GI symptoms at follow-up (68%) when compared with those with a normal biopsy (P = 0.0001). Similarly, Behjati et al (21) also reported a high incidence of GI symptoms, such as recurrent abdominal pain (50%), diarrhoea (30%), and constipation (20%), in children with eosinophilic colitis. The precise mechanism that could lead to constipation in children with eosinophilia is not clear, but it has been suggested that eosinophils interfere with the GI motility (22) and possibly have negative effects on the myenteric ganglion cells (23). There is also evidence that mast cells and eosinophils are key cells in the development of dysmotility, by mediating their action through release of potent granule constituents (24). We also found a high incidence of food/milk allergy when compared with the group with normal biopsy. Kaijser (2) initially proposed that the eosinophilic colitis was an allergic phenomenon, and other authors (21) have found an abnormal serum IgE levels in these patients but failed to find a significant association with a history of atopy. Because of the retrospective nature of our study, we do not have serum IgE levels data. Interestingly, we found that there was no significant change in the weight-for-age z score at follow-up (P = 0.8), and 8 (20%) patients had persistent failure to thrive (ie, a weight-for-age z score ≤−2.0). In contrast, in the group of patients with normal biopsies, there was significant catch-up growth, demonstrated by a change in the weight-for-age z score at follow-up (P = 0.003) with only 3 patients (7%) with failure to thrive.
The significance of the CINF is less clear and to the best of our knowledge, there are no studies that have investigated this histological finding in rectal biopsy of children. Interestingly, we found that patients with CINF were significantly older than patients with eosinophilia. It may be that patients with eosinophilia could subsequently develop a chronic inflammatory status because we found 3 patients with a combination of these histological findings.
Our study demonstrates that EOS and/or CINF are found in 17% of children who undergo rectal biopsy to exclude HSCR. A significant number of these children (48%) will need further medical input for the presence of persisting GI symptoms, food/milk allergy, and failure to thrive and the possibility of developing inflammatory bowel disease later in life. Our findings indicate that children with EOS and CINF on rectal biopsy should be followed up by a specialist paediatrician for further management.
1. Lowichik A, Weinberg AG. Eosinophilic infiltration of the enteric neural plexuses in Hirschsprung's disease. Pediatr Pathol Lab Med
2. Kaijser R. Allergic diseases of the gut from the point of view of the surgeon. Arch Klin Chir
3. Noel RJ, Putnam PE, Rothenberg ME. Eosinophilic esophagitis. N Engl J Med
4. Rothenberg ME. Eosinophilic gastrointestinal disorders (EGID). J Allergy Clin Immunol
5. Straumann A, Simon HU. Eosinophilic esophagitis: escalating epidemiology? J Allergy Clin Immunol
6. Bach JF. The effect of infections on susceptibility to autoimmune and allergic diseases. N Engl J Med
7. Scaillon M, Cadranel S. Food allergy and constipation in childhood: how functional is it? Eur J Gastroenterol Hepatol
8. Iacono G, Cavataio F, Montalto G, et al. Intolerance of cow's milk and chronic constipation in children. N Engl J Med
9. Turunen S, Karttunen TJ, Kokkonen J. Lymphoid nodular hyperplasia and cow's milk hypersensitivity in children with chronic constipation. J Pediatr
10. DeBrosse CW, Collins MH, Buckmeier Butz BK, et al. Identification, epidemiology, and chronicity of pediatric esophageal eosinophilia, 1982–1999. J Allergy Clin Immunol
11. Jeziorska M, Haboubi N, Schofield P, et al. Distribution and activation of eosinophils in inflammatory bowel disease using an improved immunohistochemical technique. J Pathol
12. Pensabene L, Brundler MA, Bank JM, et al. Evaluation of mucosal eosinophils in the pediatric colon. Dig Dis Sci
13. Freeman JV, Cole TJ, Chinn S, et al. Cross sectional stature and weight reference curves for the UK, 1990. Arch Dis Child
14. Rothenberg ME, Mishra A, Brandt EB, et al. Gastrointestinal eosinophils. Immunol Rev
15. Rothenberg ME, Mishra A, Brandt EB, et al. Gastrointestinal eosinophils in health and disease. Adv Immunol
16. Yang GY, West AB. What do eosinophils tell us in biopsies of patients with inflammatory bowel disease? J Clin Gastroenterol
17. Desreumaux P, Nutten S, Colombel JF. Activated eosinophils in inflammatory bowel disease: do they matter? Am J Gastroenterol
18. Levy AM, Kita K. The eosinophil in gut inflammation: effector or director? Gastroenterology
19. Maloney J, Nowak-Wegrzyn A. Educational clinical case series for pediatric allergy and immunology: allergic proctocolitis, food protein-induced enterocolitis syndrome and allergic eosinophilic gastroenteritis with protein-losing gastroenteropathy as manifestations of non-IgE-mediated cow's milk allergy. Pediatr Allergy Immunol
20. Pascal RR, Gramlich TL, Parker KM, et al. Geographic variations in eosinophil concentration in normal colonic mucosa. Mod Pathol
21. Behjati S, Zilbauer M, Heuschkel R, et al. Defining eosinophilic colitis in children: insights from a retrospective case series. J Pediatr Gastroenterol Nutr
22. Murch S. Allergy and intestinal dysmotility—evidence of genuine causal linkage? Curr Opin Gastroenterol
23. Schappi MG, Smith VV, Milla PJ, et al. Eosinophilic myenteric ganglionitis is associated with functional intestinal obstruction. Gut
24. Bengtsson U, Knutson TW, Knutson L, et al. Eosinophil cationic protein and histamine after intestinal challenge in patients with cow's milk intolerance. J Allergy Clin Immunol
25. Winter HS, Antonioli DA, Fukagawa N, et al. Allergy-related proctocolitis in infants: diagnostic usefulness of rectal biopsy. Mod Pathol
Keywords:Copyright 2012 by ESPGHAN and NASPGHAN
constipation; eosinophilia; food allergy; rectal biopsy