Understanding eosinophilic esophagitis: the cellular and molecular mechanisms of an emerging disease.

Eosinophilic esophagitis (EoE) has been increasingly recognized as a unique clinicopathological entity over the past two decades. In this short time, the mechanisms of a complex disease have begun to emerge. Patient studies suggest that EoE is an immunologic disease related to atopy. At the cellular level, eosinophils, mast cells, and B and T lymphocytes are increased in the esophageal mucosa in a patchy distribution throughout the length of the esophagus. Laboratory investigations have implicated aeroallergens, food allergens, and a unique T helper type 2 cytokine profile. EoE appears to be an antigen-driven hypersensitivity reaction characterized by a mixed IgE-dependent/delayed-type reaction and a distinct cascade of cytokines and growth factors. The causative events that lead to EoE in humans remain unknown.

Multiple squamous hyperplastic-fibrous inflammatory polyps of the oesophagus: a new feature of eosinophilic oesophagitis?

This report describes the unusual case of a 12-year-old boy with multiple polyps in the oesophagus and concurrent eosinophilic oesophagitis (EoE). Polyps were of a fibrous-inflammatory composition featuring eosinophils, mast cells, hyperplastic epithelium and fibrosis, which are all features described with EoE. EoE is an increasingly recognised clinicopathological disorder characterised by large numbers of eosinophils infiltrating the oesophageal mucosa. Polyps in the oesophagus are rare, have not previously been associated with EoE, and may represent a new feature of the disease.

FGF9-induced proliferative response to eosinophilic inflammation in oesophagitis.

BACKGROUND: Oesophagitis is characterised by basal cell hyperplasia and activated eosinophils, which release mediators including major basic protein (MBP). MBP and its mimetic polyarginine activate the calcium sensing receptor (CaSR) on oesophageal epithelium. Fibroblast growth factor 9 (FGF9) is implicated in epithelial homeostasis and proliferative response to injury, but has not been characterised in the oesophagus. OBJECTIVE: To characterise FGF9 in oesophageal epithelium and oesophagitis, as the result of MBP activation of the CaSR. METHODS: Human oesophageal epithelial cells (HET-1A) were used to compare affects of calcium, polyarginine and MBP-peptide on FGF9. HET-1A were transfected with interfering RNA (siRNA(CaSR)). FGF9, FGF receptors 2 and 3, bone morphogenetic protein (BMP)-2, BMP-4 and noggin mRNA expression were detected by reverse transcriptase polymerase chain reaction. FGF9 was measured from HET-1A and from normal, gastro-oesophageal reflux and eosinophilic oesophagitis (EoE) patient biopsies using ELISA and immunohistochemistry. HET-1A proliferation was studied using bromodeoxyuridine and MTT. RESULTS: FGF9 was secreted by HET-1A cells treated with polyarginine and MBP-peptide, but not calcium. This effect was abrogated by siRNA(CaSR). FGF9 receptor mRNA was present. HET-1A cells proliferated following rhFGF9, but not MBP-peptide treatment, and rhFGF9 altered transcription of downstream proliferation-related genes (noggin, BMP-2 and BMP-4). FGF9 was increased in biopsies from patients with eosinophilic oesophagitis, which correlated with basal hyperplasia. CONCLUSION: Eosinophil-released MBP acts on the CaSR to increase FGF9 in oesophageal epithelial cells, leading to proliferation. Increased FGF9 is found in biopsies of EoE patients and may play a role in the pathogenesis of oesophagitis.

Update in gastrointestinal allergic diseases.

Food allergy or hypersensitivity is defined as an adverse reaction to food protein which is immune mediated. Without standard definitions and reliable tests for many forms of allergic disease of the bowel, studies are difficult to interpret. The proceedings of a recent workshop on the classification of adverse immunologic reactions to foods provide the framework for this review. Recent studies have helped define the clinical spectrum and natural history of IgE and non-IgE-mediated food allergy, and provide insight into underlying pathophysiology and dietary management.

Allergic gastroenteropathy in preterm infants.

OBJECTIVES: To determine the clinical presentation, histopathologic features, and outcome of biopsy-proven allergic gastroenteropathy (AGE) in preterm infants. We hypothesized that AGE is a more frequent cause of gastrointestinal disease in this population than previously suspected. STUDY DESIGN: The retrospective portion of the study, from 1992 to 1997, included preterm infants <37 weeks' gestation who underwent biopsy because of suspected AGE. The prospective portion, from January to December 1998, included 20 infants undergoing endoscopy and biopsy because of suspected AGE. RESULTS: Twenty-five infants (12 retrospective/13 prospective) with mean gestational age of 29 weeks at birth and mean postnatal age at diagnosis of 78 days were diagnosed with AGE. Three clinical patterns of presentation were noted: group 1, gastroesophageal reflux disease (n = 5); group 2, non-specific feeding intolerance (n = 8); and group 3, lower gastrointestinal bleeding (n = 12). Ten patients had negative biopsy findings (3 retrospective/7 prospective) and had clinical features indistinguishable from those of groups 1 and 2. Patients in group 3 were most likely to have positive biopsy findings (12 of 12). Fifteen patients responded to a casein hydrolysate formula, and 10 patients required an amino acid-based formula. Patients with AGE who had eosinophilic infiltration and villous atrophy took longer to recover than those with eosinophilic infiltration alone (P <.03). Subsequently, most have tolerated formula challenges and are currently tolerating cow's milk. CONCLUSIONS: AGE may be an under-recognized cause of gastrointestinal symptoms in preterm infants. Confirmation with endoscopy and biopsy can be done safely and provides the basis for appropriate dietary management.

Dyspepsia in children and adolescents: a prospective study.

BACKGROUND: Dyspepsia is poorly characterized in the pediatric population. The goal of the current study was to describe the clinical constellation and natural history of dyspepsia in children and adolescents seen in a pediatric gastroenterology practice. METHODS: A standardized questionnaire was administered by a pediatric gastroenterologist to all subjects 5 or more years of age (and their parents or guardians) treated in a referral pediatric gastroenterology practice for 1 month or more of abdominal pain or discomfort, nausea, or vomiting. Subjects with dyspepsia and dyspepsia subtypes (ulcer-like, dysmotility-like) were identified by using previously defined adult criteria. Evaluation and treatment were performed at the discretion of the attending pediatric gastroenterologist. RESULTS: During a 1-year period, 257 patients were screened with 127 subjects fulfilling criteria for dyspepsia (59% girls, 85% white; median age, 11.7 years; median duration of symptoms, 8 months). Symptoms were ulcer-like in 26% and dysmotility-like (nausea predominance) in 15% of subjects. In those with dyspepsia, irritable bowel syndrome and gastroesophageal reflux were noted in 24% and 43%, respectively. Esophagogastroduodenoscopy and biopsy were performed in 56 subjects with 21 (38%) having mucosal inflammation (Helicobacter pylori in 5). The remaining 35 subjects (62%) were considered to have functional dyspepsia. Duration of symptoms less than 1 year and vomiting were risk factors for mucosal inflammation. Follow-up at 6 months to 2 years revealed 70% of subjects were either asymptomatic or much improved regardless of the cause of dyspepsia. CONCLUSION: Most children and adolescents with dyspepsia do not have serious disease. In our referral population H. pylori infection was unusual, and no peptic ulceration was found. Most subjects with functional dyspepsia have improvement of symptoms over time.

Activated eosinophils in esophagitis in children: a transmission electron microscopic study.

BACKGROUND: Esophagitis in infants and children is often characterized by eosinophilic inflammation. The underlying pathophysiologic mechanisms leading to this type of inflammation, and the role of eosinophils in the clinical expression of esophagitis, are unknown. The purpose of this study was to demonstrate the ultrastructural activation state of eosinophils in esophagitis in infants and children. METHODS: Standard transmission electron microscopy was used to examine endoscopic esophageal biopsy material from patients with and without esophagitis, as defined by standard histologic criteria. RESULTS: Twelve esophagitis and three control cases were studied. In patients with esophagitis, electron microscopy revealed numerous eosinophils throughout the mucosa and invariably demonstrated signs of activation, including inversion of core-to-matrix densities and lucency of granule core protein. Eosinophils in an activated state were seen in active diapedesis through vascular endothelium into the mucosa. Eosinophils were sometimes seen in proximity to lymphocytes. Biopsies of control patients did not demonstrate eosinophils. CONCLUSIONS: Eosinophils present in esophagitis are activated by electron microscopic criteria, and can been seen in an activated state entering into the mucosa. This suggests that eosinophils play an active role in the pathophysiology of this disorder, and that proinflammatory factors are present that selectively recruit and activate eosinophils in esophagitis in children.

Alterations in bone metabolism in children with inflammatory bowel disease: an in vitro study.

BACKGROUND: In patients with inflammatory bowel disease (IBD), accelerated bone loss and osteopenia have been found. Potential etiologies of these bone abnormalities have included malnutrition, poor calcium intake or absorption, and the use of corticosteroids. Recent studies have suggested that circulating pro-inflammatory cytokines, which are produced in inflamed bowel, can have a profound effect on bone metabolism, particularly bone resorption. Our aim was to characterize the effects of serum from subjects with IBD on bone metabolism in an in vitro bone culture system. METHODS: Organ cultures of fetal rat parietal bones were treated with sera from 9 subjects with Crohn's disease, 7 with ulcerative colitis, and 10 controls with functional bowel disease (age range of all subjects 7-16 years). Patients were also classified by disease activity, serum albumin level, erythrocyte sedimentation rate (ESR), and serum interleukin (IL) 6 levels. The effects of sera on bone formation and resorption were quantified. RESULTS: Compared with control serum, serum from patients with Crohn's disease significantly decreased bone dry weight (p < 0.01) and calcium content (p < 0.001) during 96 h of culture, while serum from ulcerative colitis patients had no effect. While no difference in collagen synthesis was noted between any of the three experimental groups, noncollagen protein synthesis was lower in the ulcerative colitis group than in the control group or those with Crohn's disease (p < 0.05). DNA content was similar in all groups. There was no significant effect of serum from any experimental group on bone resorption. There was no demonstrable relationship between clinical disease activity, ESR, or serum IL-6 levels and measures of bone metabolism. Histologic evaluation of cultured bone showed marked differences between control subjects and Crohn's disease patients, with the latter being characterized by disorganization of mineral and osteoid and morphologically abnormal osteoblasts. CONCLUSIONS: Serum from children with IBD has a significantly different effect than control serum on an in vitro model of bone metabolism. Our data suggest that circulating factors may affect osteoblasts and bone formation, leading to bone loss. Further work will be required to further characterize the nature of these factors and develop treatment strategies to minimize their effects.

Clinical outcome of ulcerative colitis in children.

OBJECTIVES: To characterize the response to current medical therapies in children with ulcerative colitis, and to identify those factors that may predict the need for colectomy. DESIGN: Retrospective chart review at two large pediatric inflammatory bowel disease centers. RESULTS: We identified 171 subjects ranging in age from 1.5 to 17.7 years at diagnosis (mean 11.2 years). Mean follow-up was 5.1 years. Of these subjects, 43% had mild disease at presentation and 57% had disease that was classified as moderate or severe. After treatment 90% of the former group and 81% of the latter group had resolution of symptoms by 6 months. During any subsequent yearly follow-up interval, approximately 55% of the entire study population was symptom free, 38% had chronic intermittent symptoms, and 7% had continuous symptoms. A significantly lower risk of colectomy was noted for those with initially mild disease compared with those with moderate/severe disease. At 1-year the risk of colectomy was 1% among those with mild disease versus 8% with moderate/severe disease; at 5 years, the risk of colectomy was 9% in the mild disease group versus 26% in the moderate/severe disease group (p <0.03). CONCLUSIONS: In the majority of pediatric subjects with ulcerative colitis remission is achieved in the first 6 months after therapy; thereafter disease is inactive in about 50% of patients during any given year of follow-up. Severity of disease at presentation is a significant risk factor for colectomy during the first 5 years of follow-up. Future management protocols with more aggressive initial therapy may be warranted in children with moderate/severe disease.

Relationship of interleukin-1 receptor antagonist to mucosal inflammation in inflammatory bowel disease.

Previous work has suggested that the interleukin-1 (IL-1) receptor antagonist, IL-1ra, may regulate mucosal inflammation in inflammatory bowel disease. The present study assessed the relationship of mucosal IL-1ra levels to histologic severity of inflammation and the related proinflammatory cytokines IL-1 beta and IL-6 in children with inflammatory bowel disease. Colonic biopsy specimens from 29 patients with ulcerative colitis, 27 with Crohn's disease, and 24 noninflammatory control subjects were assayed for IL-1ra, IL-1 beta, and IL-6 by enzyme-linked immunosorbent assay. Histologic activity was graded as none, mild, moderate, or severe. Mucosal IL-1 beta levels, but not IL-1ra levels, were significantly elevated in moderate/severely inflamed biopsies from patients with either ulcerative colitis (p < 0.01) or Crohn's disease (p < 0.001) compared with those with none/mild inflammation. The mucosal molar ratio of IL-1ra/IL-1 beta was significantly lower for moderate/severe inflammation compared with none/mild inflammation for patients with ulcerative colitis (p < 0.05) and Crohn's disease (p < 0.01). The mucosal IL-1ra/IL-1 beta ratio was similar in controls to none/mild inflamed biopsies from subjects with either ulcerative colitis or Crohn's disease. Our observations suggest that increasing mucosal inflammation in inflammatory bowel disease in children is associated with a decrease in the "normal" effective IL-1ra/IL-1 beta ratio in which IL-1ra predominates. The importance of this abnormality to the pathogenesis of inflammatory bowel disease awaits further study.

Cyclosporine for the treatment of fulminant ulcerative colitis in children. Immediate response, long-term results, and impact on surgery.

PURPOSE: Emergency surgery for fulminant colitis is often complicated by high-dose steroid therapy, poor nutrition, and psychologic maladjustment. Cyclosporine is effective for fulminant ulcerative colitis in adults, resulting in avoidance of immediate surgery in 75 percent of patients and a 55 percent long-term remission rate. Over the last five years, we studied the effectiveness of cyclosporine in children with fulminant colitis. METHODS: Fourteen patients with ulcerative colitis (age, 7-20 years) received cyclosporine after satisfying the following criteria: 1) greater than five bloody diarrheal stools per day; 2) severe abdominal pain; 3) no improvement after ten days of bowel rest, 4) intravenous methylprednisolone (1-2 mg/kg/day); and 5) parenteral nutrition. Treatment was begun with oral cyclosporine (4.6-9.6 mg/kg/day), and the dose was adjusted to achieve whole blood trough levels measured with a monoclonal radioimmunoassay between 150 and 300 ng/ml. If improved, patients were discharged on oral cyclosporine, prednisone, and a regular diet. RESULTS: Eleven of 14 patients (78 percent) responded within two to nine days and were able to consume a normal diet, had three or less soft stools per day, and had no pain. One did not respond after ten days and underwent an ileal pouch-anal anastomosis procedure. Two patients elected surgery after 20 days of therapy and a partial response. Of 11 patients who left the hospital, 4 had recurrent symptoms after 2 to 11 months of taking therapeutic doses of cyclosporine and 3 flare ups while weaning from cyclosporine after 4 to 8 months. Three patients have been weaned from cyclosporine after 8 to 13 months and have remained in remission from six months to five years. One patient is about to complete a six-month course of cyclosporine. Overall ten (72 percent) have undergone surgery, including 7 of 11 who responded initially to cyclosporine and left the hospital. Weight (P < 0.001), albumin (P < 0.01), erythrocyte sedimentation rate (P > 0.05), and prednisone dose (P < 0.001) improved significantly in the seven patients on cyclosporine who responded initially, left the hospital, and subsequently underwent surgery. CONCLUSIONS: Cyclosporine is effective in achieving clinical remission in 80 percent of children with refractory fulminant colitis; however, within one year, most initial responders will require colectomy because of a flare up of the disease. In a majority of patients, the role of cyclosporine therapy is to rapidly ameliorate symptoms and prevent precipitous colectomy, improve nutrition and psychologic adaptation, and reduce the steroid dose leading to surgery in a well-prepared patient.

Characterization of symptoms in children with recurrent abdominal pain: resemblance to irritable bowel syndrome.

We sought to prospectively characterize and compare the symptoms of children > or = 5 years of age with recurrent abdominal pain to previously established criteria for irritable bowel syndrome (IBS) in adults. For all eligible subjects, a detailed questionnaire concerning characteristics of abdominal pain and defecatory pattern was completed at presentation. In addition, a battery of screening tests was performed and additional evaluation was done at the discretion of their physician. In all, 227 subjects fulfilled the entrance criteria, but 56 were subsequently excluded because of diagnoses of inflammatory bowel disease (nine cases), lactose malabsorption (46 cases), or celiac disease (one case). Of the remaining 171 patients, 117 had IBS symptoms. In the IBS subjects, lower abdominal discomfort (p < 0.001), cramping pain (p < 0.0009), and increased flatus (p < 0.0003) were more common, whereas dyspeptic symptoms such as epigastric discomfort (p < 0.003), pain radiating to the chest (p < 0.009), and regurgitation (p < 0.02) were more common in the non-IBS subjects. Our study not only confirms the clinical heterogeneity of children with recurrent abdominal pain but also concomitantly demonstrates that most children with this disorder have symptoms that fulfill the standardized criteria for IBS in adults. The identification of subgroups of children with recurrent abdominal pain can provide a framework for the diagnosis of functional bowel disease as well as establish the need for invasive and expensive tests.

Characterization of circulating interleukin-1 receptor antagonist expression in children with inflammatory bowel disease.

The cytokines IL-1 beta and IL-6 appear to be important in the pathogenesis of inflammatory bowel disease (IBD). Recently, a naturally occurring interleukin-1 receptor antagonist, designated IL-1ra, which inhibits IL-1 beta activity in vitro and in vivo has been described. The purpose of the present study was to assess the circulating levels and relative relationships of IL-1ra, IL-1 beta, and IL-6 in children with IBD of varying severity. Serum/plasma samples were obtained from 32 children with ulcerative colitis, 45 with Crohn's disease, and 24 control patients. Cytokine assays were performed by enzyme-linked immunoassay. IL-1ra levels were significantly elevated in children with ulcerative colitis or Crohn's disease of moderate/severe activity compared to patients with inactive/mild IBD or control subjects (P < 0.001). IL-1 beta was only detectable in the circulation of two subjects with severe colitis (one ulcerative colitis, one Crohn's disease), and both had extremely elevated IL-1ra levels. IL-1ra levels were significantly related to IL-6 levels for patients with IBD (P < 0.00001). Our results suggest that circulating IL-1ra appears in increasing concentrations in children with mounting degrees of disease severity as determined by clinical scoring methods as well as by the level of IL-6. Future work will need to address the clinical and prognostic value of measuring circulating IL-1ra in individuals with inflammatory bowel disease.

Inflammatory bowel disease in children and adolescents.

Twenty-five percent of the patients with inflammatory bowel disease develop symptoms within the first two decades of life. The approach to children with these chronic illnesses involves an understanding of the medical, nutritional, and psychological issues of maturing children and adolescents. Children with Crohn's disease and ulcerative colitis have different clinical presentations than adult patients. The physician must decide when to perform diagnostic procedures and which management options will be successful.