Eosinophilic esophagitis in Colombian children: relationship between food allergens, symptoms and endoscopic findings

Abstract

Background: Eosinophilic esophagitis (EoE) is an immune-mediated disease characterised by oesophageal eosinophilic infiltration, leading to chronic inflammation and tissue remodelling.

Objectives: To describe the clinical, epidemiological and endoscopic characteristics of children with EoE in Colombia, and explore associations between symptoms, allergen sensitisation and endoscopic findings.

Methods: Observational, multicentre retrospective study with an analytical component, conducted on paediatric patients diagnosed with EoE across 10 tertiary referral centres in Colombia between 2015 and 2023. All cases had histological confirmation with ≥15 eosinophils per high-power field. Data were extracted from standardised medical records by trained investigators.

Results: A total of 143 cases were included, with a male predominance (66.4%) and median age at diagnosis of 92 months (IQR 56-144). Relevant early exposures included prematurity (11.2%), low birth weight (10.5%), caesarean delivery (50.3%), infant formula use (62.2%) and antibiotics in infancy (27.9%). Family history of atopy was present in 54.6%, and EoE in 4.9%. Food specific IgE was present, particularly to cow's milk (56.6%) and egg (38.1%). Most frequent symptoms included abdominal pain (67.1%), nausea (60.1%) and vomiting (46.2%). Endoscopic findings included longitudinal furrows (70.9%) and white exudates (67.4%). Statistically significant associations were found between dysphagia and oesophageal rings (p=0.003), vomiting and white exudates (p=0.018), abdominal pain and longitudinal furrows (p=0.025) and regurgitation and oedema (p=0.031).

Conclusions: Paediatric EoE in Colombia shows symptom and endoscopic heterogeneity. Identifying potential allergen triggers may contribute to understanding symptoms and endoscopic patterns in EoE.

Keywords: Gastroenterology.

Figure 1. Relationship between allergen sensitisation and clinical symptoms in paediatric eosinophilic esophagitis. Bars represent the proportion of patients with each symptom among those sensitised and not sensitised to the corresponding allergen group (cow’s milk proteins, egg proteins, cereals and grains, legumes and tree nuts, and marine proteins). Percentages are accompanied by absolute numbers (n/N), where N reflects the number of children tested for each specific allergen. P values correspond to χ² or Fisher’s exact tests, depending on cell counts. Detailed numerical values and denominators are provided in table 3.

Figure 2. Relationship between allergen sensitisation and endoscopic findings (EREFS domains). Bars display the proportion of patients with endoscopic abnormalities—including oedema, rings, exudates, furrows and strictures—among those sensitised and not sensitised to each allergen group. Percentages include absolute frequencies (n/N), using the specific number of children tested for each allergen as the denominator. P values were derived from χ² or Fisher’s exact tests as appropriate. Full numerical data are provided in table 4. EREFS, Eosinophilic Esophagitis Endoscopic Reference Score.