A 15-year-old boy with a history of allergic sinusitis, asthma, eczema, and multiple food allergies presented with a 1-month history of intermittent abdominal pain, and intractable vomiting. Initially clear, the patient’s emesis progressively became bilious and predominantly occurred postprandially, although the timing was not consistently related to meals. As a result, the patient has lost over 10 kg in 1 month. The abdominal pain was localized to both the periumbilical and epigastric regions, colicky in nature and severe in intensity, without radiation and identifiable triggers. The patient had 2 watery stools per day, without any blood or mucus, which subsequently progressed to complete cessation of bowel movements in the week prior to admission. The family history was unremarkable for chronic illnesses, autoimmune disorders, or malignancies. The patient reported no recent travel and no use of prescription medications or illicit substances.Physical examination revealed an uncomfortable-appearing young patient with a thin habitus. The anthropometric assessment demonstrated mild malnutrition with a body mass index of 17.0 kg/m2 (Z-score, −1.4). The abdominal examination revealed a soft consistency without signs of distension, organomegaly, or mass. Mild tenderness was noted in the epigastric area without any rebound phenomenon.Initial complete blood count (CBC) revealed leukocytosis (12.4 × 109/L) and marked eosinophilia (21%, 2604 cells/μL). A comprehensive metabolic panel (CMP) showed hypokalemia (2.7 mmol/L) associated with hypochloremic alkalosis (chloride 85 mmol/L, carbon dioxide 38 mmol/L), elevated bilirubin (2.0 mg/dL), aspartate aminotransferase (110 U/L) and alanine aminotransferase (106 U/L), approximately twice the upper limit of normal, elevated gamma-glutamyl transferase (94 U/L), and hypoproteinemia (5.3 g/dL) and hypoalbuminemia (2.8 g/dL) with a normal alkaline phosphatase (124 u/L) and glucose level (109 mg/dL). Coagulation studies showed a prolonged prothrombin time (16.4 seconds, international normalized ratio 1.3). Viral hepatitis serologies were negative. The immunoglobulin E (IgE) level was markedly elevated (1968 kU/L), whereas inflammatory markers (C-reactive protein CRP and erythrocyte sedimentation rate ESR) were within normal limits. A stool sample for ova and parasite examination was negative. Testing for celiac disease (anti-tissue transglutaminase IgA, total IgA) and inflammatory bowel disease (fecal calprotectin, stool heme occult test) was normal.The initial presentation suggested possible intestinal obstruction, but this was deemed unlikely following an unremarkable abdominal x-ray. Peripheral eosinophilia and gastrointestinal (GI) symptoms raised concern for parasitic infection, although this was considered less likely given a negative stool ova and parasite examination. Subsequent investigations for celiac disease and inflammatory bowel disease were unrevealing. Drug-induced eosinophilic gastroenteritis was unlikely given the absence of prior medication use. Hypereosinophilic syndrome was considered but remained a diagnosis of exclusion, requiring secondary causes to be ruled out.Previously documented hypoalbuminemia, elevated aminotransferases, hyperbilirubinemia, and an elevated INR prompted an abdominal ultrasonogram, which revealed ascites, common bile duct (CBD) dilation, and nonspecific mural thickening, most prominently involving the gastric wall (Figure 1A). A contrast-enhanced CT scan subsequently demonstrated large-volume ascites, marked CBD dilation, and diffuse wall thickening of the stomach and small bowel, with no evidence of masses, lymphadenopathy, or superior mesenteric artery compression (Figure 1B and C).The CT findings prompted further investigation, including esophagogastroduodenoscopy (EGD) and paracentesis. EGD demonstrated esophageal rings (Figure 2A), antral mucosal nodularity and thickening, and marked duodenal inflammation with diffuse mucosal swelling and erosions (Figure 2C). Esophageal biopsies revealed submucosal eosinophilic infiltrates with 25 to 30 eosinophils per high-power field (Figure 2B). Gastric biopsies showed Helicobacter pylori–associated gastritis (Figure 2E) with slight eosinophilic infiltration. Duodenal biopsies displayed mildly broadened villi and an intact brush border, with minimal eosinophils (Figure 2D). Ultrasonogram-guided paracentesis yielded straw-colored ascitic fluid exhibiting marked eosinophilia (80% of the cell population) (Figure 2F).Despite only mild eosinophilic infiltration in the gastric and duodenal biopsies, the combination of clinical presentation, imaging findings, marked peripheral eosinophilia, and eosinophilic predominance in the peritoneal fluid supported a diagnosis of a non-eosinophilic esophagitis eosinophilic gastrointestinal disease (non-EoE EGID), specifically eosinophilic gastritis and eosinophilic enteritis (serosal form) with esophageal involvement. Although esophageal biopsies showed histologic features suggestive of EoE, the absence of esophageal symptoms and the more prominent involvement of other GI regions make isolated EoE an unlikely primary diagnosis, based on current clinical criteria.1,2 Recent expert consensus guidelines recommend that when eosinophilic inflammation is identified in multiple segments of the GI tract—including the esophagus—but the principal clinical and pathological burden lies outside the esophagus, the diagnosis should reflect the predominant site of involvement, with esophageal findings characterized as “esophageal involvement” rather than standalone EoE.2EGIDs extend beyond EoE, and although awareness of these conditions has increased in recent years, diagnosis remains challenging due to their heterogeneous clinical presentation and low prevalence, estimated at 5.1 per 100 000 individuals.3 Non-EoE EGIDs, while sharing features such as eosinophilia, atopy, and predominantly Th2-driven inflammation with EoE, exhibit distinct inflammatory profiles for each clinicopathological entity, highlighting the need for further investigation into their pathogenesis.4,5Non-EoE EGID classification is based on the site and extent of eosinophilic infiltration, which can affect the mucosal, muscular, or subserosal layers of the GI tract.2 Mucosal involvement is the most common form, observed in 88% to 100% of cases.6 Isolated muscular and serosal forms are rare and often considered a progression of the mucosal type,6 making their incidence difficult to estimate.Mucosal non-EoE EGID presents in various forms: gastric involvement can lead to dyspepsia, vomiting, and occasional bleeding; enteritis is characterized by chronic diarrhea, abdominal pain, malabsorption, and potential failure to thrive; and colitis, although rare, presents with diarrhea and tenesmus.2 When mucosal disease is widespread, symptom overlap is common, or the disease may remain clinically silent despite histological evidence,2 as seen in our case with esophageal involvement.The muscular form shares overlapping symptoms with mucosal form, including abdominal pain, nausea, and bloating. However, it is more distinctly characterized by impaired motility due to thickened intestinal walls, resulting in constipation, distention, and bilious vomiting.2The serosal subtype, as exemplified in this case, is uniquely characterized by eosinophilic ascites.7 It involves eosinophilic infiltration of the serosal and subserosal layers of the GI tract, often presenting with overlapping symptoms seen in mucosal and muscular non-EoE EGID,2 making both diagnosis and management particularly challenging.In EGIDs, upper GI endoscopy typically reveals mucosal thickening, edema, erosions, or ulcerations.8 However, in the serosal variant, endoscopic findings may be inconclusive or appear normal. Therefore, ascitic fluid analysis—including cytology, culture, and acid-fast staining—becomes crucial for diagnosing and excluding other potential differentials, such as malignancies or tuberculosis.2Non-EoE EGIDs are clinicopathological entities, meaning histology alone is insufficient for a diagnosis.2 The diagnostic workup includes assessments for parasitic infection (travel history and stool ova and parasites examination) and atopy (history of asthma, dermatitis, and seasonal and food allergies). Initial laboratory evaluation typically includes a CBC with differential, CRP, ESR, and a CMP assessing electrolytes, liver function, albumin, and total protein.2 Although not diagnostic, imaging studies can assess the extent of intestinal involvement, bowel wall and mesenteric thickening, ascites, pyloric hypertrophy, and/or bowel dilation and stenosis (Figure 1).2Non-EoE EGIDs are chronic, often relapsing disorders, requiring individualized treatment approaches based on the affected GI segment, severity, and patient-specific factors.2 Currently, systemic corticosteroids are considered the primary treatment approach to induce remission.2 Biologic therapies have shown promise in reducing eosinophilic inflammation and achieving clinicohistologic remission in cases refractory to steroid therapy.9,10Dietary elimination therapy has proven effective in inducing remission; however, its routine use remains unsupported by sufficient evidence, and its efficacy in maintaining long-term remission is unknown.2 Accurate differentiation between IgE-mediated food allergies and food-related triggers of EGID relies on clinical evaluation, given the limited effectiveness of skin prick and serum IgE tests for this purpose.2The patient remained hemodynamically stable, with consistently unremarkable vital signs throughout the hospital stay. His persistent vomiting, unresponsive to antiemetic therapy, resulted in inadequate oral intake and necessitated the initiation of total parenteral nutrition (TPN). Upon admission, famotidine 20 mg twice daily was initiated to manage epigastric pain and dyspepsia and was continued following endoscopic confirmation of gastroduodenitis. Following the final diagnosis, intravenous (IV) methylprednisolone (1 mg/kg daily) was administered, leading to significant clinical improvement, including resolution of vomiting and abdominal pain. The patient transitioned to a full liquid diet with an amino acid-based formula, allowing for the discontinuation of TPN and gradual diet advancement. Treatment for H. pylori gastritis was deferred to the outpatient setting.Coagulopathy, attributed to malabsorption and successfully corrected with IV vitamin K supplementation, contrasts with the uncertain etiology of the hepatobiliary involvement, which may result from eosinophilic infiltration extending to the biliary tree, causing inflammation, biliary stasis, and potentially hepatitis.11After 7 days of IV methylprednisolone, the patient was transitioned to a tapering course of oral corticosteroids and discharged with follow-up appointments scheduled with a gastroenterologist, allergist, and primary care pediatrician.EGIDs encompass more than just EoE. Physicians should maintain a high index of suspicion for these conditions in patients presenting with unspecific GI symptoms like abdominal pain, vomiting, diarrhea, or constipation, especially when accompanied by eosinophilia and/or a history of atopy. Clinicians should familiarize themselves with updated consensus guidelines to navigate these challenges effectively.1,2The classification of non-EoE EGIDs is based on the site and extent of eosinophilic infiltration, which may involve the mucosal, muscular, or subserosal layers of the GI tract.2Although endoscopy and histological assessment are critical for a definitive diagnosis, non-EoE EGIDs are clinicopathological entities. A thorough evaluation of the patient’s medical history, clinical presentation, laboratory findings, and imaging studies is essential.Although steroids remain the mainstay for acute management, emerging therapies, including biologics, may offer steroid-sparing options for refractory or chronic cases.Collaboration between pediatric gastroenterologists, allergists, and dietitians is vital for optimizing outcomes, including identifying triggers, managing symptoms, and addressing nutritional deficiencies.Additionally, we would like to acknowledge Van H. Savell Jr, MD, Medical Director of Pediatric Pathology at Driscoll Children’s Hospital, for his assistance in obtaining the histopathology slides used in this case.
Aksenov et al. (Sun,) studied this question.