Presentation and treatment of eosinophilic gastroenteritis in Busan and Gyeongnam, Korea: A multicenter study

Abstract

BACKGROUND

Eosinophilic gastroenteritis (EGE) is a relatively rare disease characterized by eosinophilic infiltration of the gastrointestinal wall. Limited data are available regarding the clinical characteristics, treatment patterns, and outcomes of patients with EGE in Korea.

AIM

To investigate the clinical features, endoscopic findings, treatments, outcomes, and treatment response and relapse predictors in the abovementioned population.

METHODS

We retrospectively reviewed the medical records of 73 patients with histologically confirmed EGE in Busan and Gyeongnam, Korea, from 2010 to 2023. Clinical, endoscopic, laboratory, and treatment-related data were collected.

RESULTS

The median patient age was 52 years, and 50.7% of the patients were female. Of the patients, 35 had enterocolitis, 34 had gastroduodenitis, and four had gastroenterocolitis. Endoscopic abnormalities were observed in 82.2% of the patients. Proton pump inhibitors were prescribed primarily for gastroduodenitis, whereas corticosteroids were prescribed more frequently for enterocolitis and gastroenterocolitis. The clinical response rates were high for all groups (75.0%-90.3%); however, relapse occurred more frequently in patients with enterocolitis (43.3%). Male sex was associated with clinical response, although this association should be interpreted with caution; no significant predictors of relapse were identified.

CONCLUSION

Patients with EGE exhibit diverse clinical characteristics and treatment patterns. While the clinical response to primary treatment is relatively high, relapse is more common in patients with enterocolitis, indicating the need for location-based strategies and close follow-up.

Key Words: Eosinophilic enteropathy; Gastritis; Enteritis; Colitis; Treatment outcome

Core Tip: This retrospective multicenter cohort study analyzed 73 Korean patients with eosinophilic gastroenteritis. The clinical features, endoscopic findings, treatment modalities, and outcomes were evaluated based on the disease location. Corticosteroids and proton pump inhibitors were the main treatments with high clinical response rates. Relapse was more frequent in patients with colonic involvement, and male sex was suggested to be associated with a favorable treatment response. These findings provide real-world insights into eosinophilic gastroenteritis management in Korea, emphasizing the need for tailored therapies and close follow-up in high-risk groups.

INTRODUCTION

Eosinophilic gastroenteritis (EGE) is a rare gastrointestinal (GI) disease characterized by eosinophilic infiltration of the intestinal wall, with clinical manifestations ranging from mild abdominal discomfort to severe GI dysfunction[1-3]. EGE primarily affects the stomach, small intestine, and colon, and its clinical features vary depending on the site and depth of invasion[2,4]. According to Klein’s classification, EGE can be classified into mucosal, muscular, and serosal types, with the mucosal type being the most common[5].

The pathophysiology of EGE is not yet fully understood; however, allergic mechanisms, including food and environmental factors, are thought to play an important role, closely related with Th2-mediated immune responses[1,2,4]. Additionally, a decline in Helicobacter pylori (H. pylori) infection has been suggested as a factor that may increase susceptibility to eosinophilic GI disorders[6,7].

Most studies on EGE have been limited to single-center experiences or small case series, predominantly conducted in Western populations, which may not fully reflect regional differences in the clinical presentation or management[8]. Although the prevalence of EGE seems to be increasing[8,9], accurate epidemiologic data remain uncertain owing to diverse clinical presentations and diagnostic difficulties[10,11]. In Asia, including Korea, a few studies have reported the epidemiology and clinical features of EGE, with the prevalence of eosinophilic GI disorders other than eosinophilic esophagitis among symptomatic patients ranging from 1.2% to 5.7%[9]. Although corticosteroids are commonly used as the primary treatment, the long-term outcomes, relapse rates, and optimal management strategies remain unclear. Moreover, differences in the prevalence of H. pylori infection and dietary habits between Western and Asian populations have been identified as potential factors influencing the clinical spectrum of EGE[8,12,13]. Notably, H. pylori prevalence is substantially higher in East Asia (approximately 47%-72%) than in Western populations, where a pooled H. pylori exposure rate of 8.9% has been reported[8]. H. pylori infection has been suggested to be inversely associated with Th2-mediated eosinophilic GI disorders[8].

A comprehensive understanding of the clinical and endoscopic features of EGE and factors influencing the treatment response and recurrence is essential for optimizing management and improving patient outcomes. Therefore, this multicenter study aimed to investigate the clinical characteristics, endoscopic findings, treatment modalities, and clinical outcomes of patients with EGE in Korea and to evaluate the potential predictors of clinical response and relapse.

MATERIALS AND METHODS

Study design and patients

This multicenter retrospective study included adult patients (aged ≥ 18 years) diagnosed with EGE at centers in Busan and Gyeongnam, Korea, between 2010 and 2023. Patients were eligible if they met the following inclusion criteria: (1) Histologic evidence of eosinophilic infiltration (≥ 20 eosinophils per high-power field) on biopsy or surgical specimens, or eosinophilic ascites; and (2) Presence or absence of GI symptoms, including patients who were incidentally diagnosed during endoscopic evaluation. Patients were excluded if they met any of the following criteria: (1) Coexisting GI diseases such as eosinophilic granuloma, inflammatory bowel disease, or GI malignancy; (2) Diagnosis limited to eosinophilic esophagitis only; (3) Drug-induced eosinophilia; or (4) Parasitic or other infectious causes. All included patients underwent EGE treatment during the study period.

Data collection

Clinical, endoscopic, laboratory, histopathological, and treatment data were reviewed by examining the electronic medical records. The following information was collected: (1) Demographic and baseline characteristics, such as age at diagnosis, sex, residence, and history of allergic diseases; (2) Clinical features, including symptoms at presentation and the involved GI segments; (3) Endoscopic findings, including erythema, edema, erosion, ulceration, bleeding, nodularity, pale mucosa, or normal appearance; (4) Laboratory and histopathologic results, such as the peripheral eosinophil count, immunoglobulin E level, tissue eosinophil count per high-power field, and depth of infiltration; and (5) Treatment details, including the type [e.g., proton pump inhibitors (PPIs), corticosteroids, antihistamines, elimination diet, immunomodulators, monoclonal antibodies, or supportive care] and duration of therapy, with combination therapies recorded when applicable. Conservative treatment was defined as observation with or without symptomatic management, without the use of corticosteroids or PPIs.

Outcomes

The primary outcomes were: (1) Clinical characteristics of patients with EGE, including symptoms at presentation, laboratory and endoscopic findings, and histological features; (2) Treatment modalities according to the disease location; and (3) Treatment outcomes, including the clinical response and relapse. The clinical response was defined as symptomatic improvement, regardless of histological improvement, and relapse was defined as symptom recurrence after treatment cessation with a tissue eosinophil count ≥ 20 per high-power field. The secondary outcome was the identification of predictors of clinical response and relapse.

Statistical analysis

Continuous variables are summarized as the median and interquartile range (IQR), whereas categorical variables are reported as the count and percentage. Descriptive statistics were used for the initial group comparisons. Categorical data were compared using the χ² test or Fisher’s exact test, and continuous data were analyzed using the Mann-Whitney U test. Multivariate logistic regression was performed to identify the predictors of clinical response and recurrence, with results expressed as odds ratios (ORs) and 95% confidence intervals (CIs). Statistical significance was defined as a P value < 0.05. All analyses were performed using MedCalc version 23.3.7 (MedCalc Software, Ostend, Belgium).

RESULTS

Baseline characteristics

A total of 73 patients with EGE were included in the study; their baseline characteristics are presented in Table 1. The median age was 52 years (IQR: 35-63 years), and 37 patients (50.7%) were female. The allergic comorbidities included asthma (10.5%), allergic rhinitis (7.9%), atopic dermatitis (3.9%), and food allergies (3.9%). The median serum immunoglobulin E level was 243 IU/mL (IQR: 51-531 IU/mL), and the median eosinophil count was 209 cells/μL (IQR: 111-644 cells/μL). Most patients lived in urban areas (94.5%), and most (89.0%) presented with GI symptoms at diagnosis. Endoscopic abnormalities were observed in 60 (82.2%) patients. The seasonal distribution at diagnosis was relatively balanced across spring (26.0%), summer (27.4%), fall (23.3%), and winter (23.3%). Of the 73 patients, 72 completed follow-up, with a median duration of 10.8 months (IQR: 2.5-45.3).

Table 1 Baseline characteristics of the patients with eosinophilic gastroenteritis, n (%).

Variable	Total (n = 73)
Age, median (IQR)	52 (35-63)
Female sex	37 (50.7)
Alcohol use	17 (23.3)
Smoker	19 (26.0)
Allergic disease
Allergic rhinitis	6 (7.9)
Atopic dermatitis	3 (3.9)
Asthma	8 (10.5)
Serum IgE, IU/mL, median (IQR)	243 (51-531)
Eosinophils, cells/μL, median (IQR)	209 (111-644)
Food allergy	3 (3.9)
Season at diagnosis
Spring	19 (26.0)
Summer	20 (27.4)
Autumn	17 (23.3)
Winter	17 (23.3)
Urban residence	69 (94.5)
Symptomatic at diagnosis	65 (89.0)
Endoscopic abnormality	60 (82.2)

IQR: Interquartile range; IgE: Immunoglobulin E.

Clinical features and endoscopic findings

The clinical features varied depending on the site of GI involvement. Abdominal pain was the most common symptom in patients with gastroduodenitis (33.3%) and enterocolitis (28.6%), whereas diarrhea was more common in patients with gastroenterocolitis (60.0%) and enterocolitis (46.4%). Other symptoms included nausea and vomiting, dyspepsia, abdominal bloating, bloody stools, and ascites (Table 2). Notably, 8 patients (11.0%) did not present with GI symptoms at diagnosis; in these cases, the diagnosis was made incidentally, with 7 patients identified during screening gastroscopy revealing erythema, edema, or erosion in the stomach, and 1 patient diagnosed following colonoscopy showing colonic erosion. Biopsies were performed to confirm the diagnosis in all cases. The endoscopic findings varied depending on the site of involvement. Erythema and edema were observed in all groups, whereas pale mucosa was observed only in patients with gastroduodenitis. Figure 1 shows the distribution of GI involvement at diagnosis.

Figure 1 Intestinal segments and organs involved at diagnosis. Multisegment involvement of ≥ 2 sites. Among the eight patients who experienced multisegment involvement, three experienced stomach/duodenum involvement, one experienced stomach/colon involvement, one experienced duodenum/colon involvement, one experienced small bowel/colon involvement, and two had involvement in all four sites.

Table 2 Symptoms and endoscopic findings according to gastrointestinal involvement, n (%).

Variable	Gastroduodenitis (n = 34)	Enterocolitis (n = 35)	Gastroenterocolitis (n = 4)
Symptoms
Abdominal pain	14 (33.3)	16 (28.6)	2 (40.0)
Nausea/vomiting	6 (14.3)	3 (5.4)	-
Dyspepsia	5 (11.9)	1 (1.8)	-
Diarrhea	4 (9.5)	26 (46.4)	3 (60.0)
Bloating	4 (9.5)	3 (5.4)	-
Bloody stool	-	5 (8.9)	-
Ascites	1 (2.4)	-	-
Endoscopic findings
Erythema	17 (27.0)	12 (23.5)	2 (25.0)
Edema	13 (20.6)	13 (25.5)	2 (25.0)
Bleeding	2 (3.2)	2 (3.9)	1 (12.5)
Erosion	14 (22.2)	7 (13.7)	1 (12.5)
Ulcer	3 (4.8)	2 (3.9)	1 (12.5)
Nodularity	8 (12.7)	3 (5.9)	1 (12.5)
Pale mucosa	5 (7.9)	-	-

Some patients presented with multiple endoscopic findings.

Initial treatment methods and clinical outcomes

The initial treatment for EGE was primarily corticosteroids and PPIs, either alone or in combination, depending on disease location. In patients with gastroduodenitis, corticosteroids were the most common treatment (35.3%, n = 12), followed by PPIs (26.5%, n = 9). In contrast, in patients with enterocolitis, corticosteroids were primarily used (65.7%, n = 23) and PPIs were less commonly used (5.7%, n = 2). A small group of patients with gastroenterocolitis was primarily treated with corticosteroids (75.0%, n = 3). The median treatment duration for all patients was 2.0 months (IQR: 1-4 months). When analyzed according to the location of EGE involvement, the median duration was 2.0 months for gastroduodenitis and enterocolitis, and 1.4 months for gastroenterocolitis.

All patient groups demonstrated high clinical response rates to treatment. Patients with enterocolitis showed the highest response rate at 90.3% (28/31), followed by patients with gastroduodenitis (84.4%, 27/32), and patients with gastroenterocolitis (75.0%, 3/4). In particular, most patients with gastroduodenitis responded to PPI monotherapy, with symptom improvement occurring in 8 of 9 patients (88.9%). Similarly, 10 of 12 patients with gastroduodenitis (83.3%) and 20 of 23 patients with enterocolitis (87.0%) responded to corticosteroid monotherapy.

Among patients treated with corticosteroids, four patients with severe symptoms initially received intravenous methylprednisolone at doses of 30-60 mg, which was subsequently switched to oral administration. The remaining patients received oral corticosteroids according to symptom severity, with doses ranging from 5 mg to 30 mg per day. Across all patients, the mean initial corticosteroid dose was approximately 26.2 mg (range: 5-62.5 mg).

When conservative treatment was analyzed separately, among the 10 patients whose responses could be evaluated, 8 (80.0%) achieved symptomatic improvement. When stratified by disease location, symptom improvement was observed in 3 of 4 patients (75.0%) with gastroduodenitis and in all 5 patients (100%) with enterocolitis, whereas no improvement was noted in a single gastroenterocolitis case managed conservatively.

However, the relapse rates significantly differed between groups. Relapse was most common in the enterocolitis group (43.3%, 13/30), whereas it occurred in only 13.8% of patients with gastroduodenitis (4/29). No relapses were observed in the gastroenterocolitis group. These findings are summarized in Table 3.

Table 3 Initial treatment modalities and clinical outcomes of patients with eosinophilic gastroenteritis, n (%).

Variable/outcome	Gastroduodenitis (n = 34)	Enterocolitis (n = 35)	Gastroenterocolitis (n = 4)
Initial treatment type
PPI	9 (26.5)	2 (5.7)	-
Steroid	12 (35.3)	23 (65.7)	3 (75.0)
PPI + steroid	5 (14.7)	1 (2.9)	-
PPI + antihistamines	1 (2.9)	-	-
PPI + steroid + antihistamines	1 (2.9)	-	-
Steroid + antihistamines	-	1 (2.9)	-
Antihistamines	-	1 (2.9)	-
Conservative treatment	6 (17.6)	7 (20.0)	1 (25.0)
Clinical outcomes
Clinical response	27/32 (84.4)	28/31 (90.3)	3/4 (75.0)
Relapse	4/29 (13.8)	13/30 (43.3)	0/4 (0)

The denominator indicates the number of patients with follow-up data after excluding those lost to follow-up. PPI: Proton pump inhibitor.

Factors associated with the treatment outcome

Multivariate logistic regression analysis identified male sex as being significantly associated with the clinical response (OR = 12.69; 95%CI: 1.25-128.49; P = 0.032). Age, the presence of allergic diseases, disease site, use of steroids or PPIs, and endoscopic abnormalities were not significant predictors of clinical response (Table 4).

Table 4 Factors associated with the clinical outcomes of patients with eosinophilic gastroenteritis.

Variable	OR	95%CI	P value
Age (years)	1.01	0.95-1.07	0.859
Male sex	12.69	1.25-128.49	0.032
Allergic disease (yes vs no)	2.58	0.24-27.82	0.436
Disease location (reference: Gastroduodenitis)
Enterocolitis	2.81	0.39-19.98	0.303
Gastroenterocolitis	2.53	0.15-42.11	0.518
Steroid use (yes vs no)	1.84	0.30-11.50	0.512
PPI use (yes vs no)	3.33	0.35-31.51	0.295
Endoscopic findings (abnormal vs normal)	0.40	0.03-4.70	0.467

Clinical response, defined as achievement of symptomatic improvement following initial treatment, was used as the dependent variable in the multivariate logistic regression. OR: Odds ratio; CI: Confidence interval; PPI: Proton pump inhibitor.

In the multivariate logistic regression analysis for recurrence, none of the evaluated variables, including age, sex, allergic disease, disease location, treatment modality, and endoscopic abnormalities, were independently associated with recurrence.

DISCUSSION

In this multicenter cohort study of patients with EGE in the Busan and Gyeongnam regions, the patients exhibited diverse clinical and endoscopic characteristics and had a relatively high clinical response rate to treatment. Most patients presented with GI symptoms including abdominal pain, diarrhea, nausea, and vomiting. These symptoms were nonspecific and tended to vary depending on the affected GI segment. Previous studies of Asian populations have reported similar findings: Abdominal pain was present in up to 75% of patients with eosinophilic GI disorders excluding eosinophilic esophagitis, while diarrhea was observed in nearly half[8]. A Chinese study reported abdominal pain in 89.1% and diarrhea in 61.8% of patients, with nausea and vomiting also frequently observed[1,14]. Endoscopic abnormalities were identified in 82.2% of patients in the present study, which contrasts with several Western studies reporting a higher prevalence of normal findings[5,15-17]. However, our findings are consistent with previous Asian reports describing erythema, erosion, and ulceration[18]. Together, these findings suggest that there are potential phenotypic differences between Western and Asian patients.

In this study, the treatment strategies varied according to the disease location. Patients with gastroduodenitis were primarily treated with PPIs or corticosteroids, whereas corticosteroids were more commonly used in patients with enterocolitis and gastroenterocolitis. Although EGE is generally thought to respond poorly to PPIs[19], our study revealed a high clinical response rate of patients with gastroduodenitis to PPI monotherapy, with 8 of 9 (88.9%) patients achieving symptom improvement. This is in line with recent Japanese studies reporting a PPI response rate > 70% in eosinophilic esophagitis[20]. Additionally, in a Singaporean study, pediatric patients with EGE showed high remission rates with PPI combined with dietary therapy[21]. These results support the hypothesis that PPIs may exert not only acid-suppressive but also anti-inflammatory effects[1], and suggest that PPIs may represent an effective treatment option for select Asian patients with EGE, particularly those with upper GI involvement. Corticosteroids remain the most widely used therapy for EGE[1,2,12]. The patients in our cohort responded well to corticosteroids; 10 of 12 patients with gastroduodenitis (83.3%) and 20 of 23 patients with enterocolitis (87.0%) treated with corticosteroid monotherapy experienced symptom improvement. However, the relapse rate was considerably higher in the enterocolitis group (43.3%), underscoring the chronic and relapsing nature of EGE.

Multivariate analysis revealed that male sex was a significant predictor of the clinical response (OR = 12.69; 95%CI: 1.25-128.49). However, the wide CI likely reflects the limited sample size; thus, this result should be interpreted with caution. Notably, EGE has generally been reported to show either a balanced sex distribution or female predominance[22,23], whereas male-dominant disease is often a characteristic of eosinophilic esophagitis[13,24]. These findings suggest that sex may influence the treatment response in Asian patients with EGE, indicating the need for further investigation. In contrast, age, history of allergic disease, disease location, use of corticosteroids or PPIs, and endoscopic abnormalities were not significantly associated with the treatment outcomes. No significant predictors of recurrence were identified, which is consistent with previous reports which identified inconsistent and heterogeneous risk factors[14,25].

This study has some strengths. To the best of our knowledge, this is the largest Korean cohort study on EGE to date, and the sample size is relatively large compared to previously published Asian cohort studies. Most published studies have been single case reports, small case series, or analyses primarily involving pediatric patients.

However, this study also has some limitations. First, its retrospective design may have led to selection and information biases. Second, the sample size of 73 patients, which is relatively large for a multicenter study in Korea, may be insufficient for drawing definitive conclusions. In particular, the small number of patients with gastroenteritis limited the statistical power to identify predictors of treatment response or recurrence. Third, the differences in treatment strategies between institutions may have influenced the results. Fourth, a uniform cutoff of ≥ 20 eosinophils per high-power field was applied across gastric, small bowel, and colonic biopsies to standardize case inclusion; however, the eosinophil density may differ by site, which could be considered a potential limitation.

CONCLUSION

In conclusion, this study provides real-world evidence of the clinical characteristics, treatment patterns, response, and relapse outcomes in patients with EGE. Clinicians should be aware that EGE can present with various symptoms and endoscopic findings, and that treatment should be tailored based on symptoms and the disease location. PPIs may be an alternative treatment option for certain patients, particularly those with gastroduodenal involvement. Close follow-up is required for patients with colonic involvement due to the risk of recurrence. In the future, larger standardized multicenter studies are required to determine the optimal treatment strategies.