Recurrent Eosinophilic Pancreatitis With Eosinophilic Gastroenteritis: A Case Report

Introduction

Eosinophilic pancreatitis (EP) is a rare form of pancreatitis, with only 28 documented cases worldwide in the past 4 decades [1]. Its clinical manifestations are often non-specific, ranging from abdominal pain to obstructive symptoms such as jaundice and fever. Therefore, EP is primarily diagnosed through pathological examination. The hallmark histological criteria of EP include: (1) the pancreas has inflammatory pathologic changes that may be pseudocysts, parenchymal necrosis, atrophy, or fibrosis; (2) there is no sign of tumor cells; and (3) idiopathic inflammation with eosinophilia can be found on microscopic examination of the pancreas [2]. However, obtaining pancreatic tissue is invasive and technically challenging, contributing to frequent misdiagnosis.

The clinical significance of EP extends beyond the pancreas, as it can be a localized phenomenon or part of a systemic eosinophilic disorder, such as hypereosinophilic syndrome or eosinophilic gastrointestinal diseases (EGIDs). The co-occurrence of EP and eosinophilic gastroenteritis (EGE) is exceptionally rare [3]. This report describes a patient with recurrent acute pancreatitis complicated by concurrent, subclinical EGE. Through this case report, we advocate for considering upper endoscopy with biopsies in the diagnostic workup for pancreatitis of unknown etiology, regardless of gastrointestinal symptoms, to facilitate an accurate diagnosis and guide appropriate therapy.

Case Report

FIRST ADMISSION (ONE YEAR PRIOR):

He presented to our hospital 1 year ago with severe epigastric pain and bloating, but no nausea, vomiting, diarrhea, jaundice, fever, rash, edema, or weight loss. Serum amylase level (1141 U/L) and computed tomography (CT) findings (peripancreatic exudative) led to a diagnosis of acute pancreatitis. Peripheral eosinophilia (EO% 11%, absolute count 1.23×10⁹/L) was noted.

We implemented standard pancreatitis treatment while investigating the etiology. The patient denied recent greasy food intake, alcohol use, or binge eating. Initial evaluation excluded biliary obstruction, malignancy, and IgG4-related disease. Despite the absence of typical upper gastrointestinal symptoms like nausea and vomiting, we performed a gastrointestinal endoscopy due to the patient’s abdominal pain and peripheral eosinophilia. Biopsies unexpectedly revealed significant eosinophilic gastroenteritis (duodenum: 30 eosinophils/HPF, 15/HPF in the antrum), raising a strong suspicion for concurrent EP. A definitive pancreatic biopsy was advised but deferred at the patient’s request due to clinical improvement. He received a short course of intravenous methylprednisolone (40 mg/day for 3 days), which resulted in significant symptomatic relief. However, a prolonged steroid tapering regimen was not initiated at this stage due to the diagnostic uncertainty regarding specific pancreatic involvement (as a biopsy was deferred) and safety considerations regarding the risks of long-term steroid therapy without a definitive histologic diagnosis.

CURRENT ADMISSION (ONE YEAR LATER):

The patient was readmitted with recurrent severe abdominal pain. Laboratory test results showed elevated amylase (491 U/L) and persistent eosinophilia (12.2%, 0.82×109/L). Abdominal CT revealed pancreatic head inflammation with duct dilation (Figure 1), while MRCP excluded stones/tumors (Figure 2), confirming acute pancreatitis recurrence.

EXCLUSION OF ALTERNATIVE ETIOLOGIES:

A systematic evaluation was conducted to rigorously rule out common and uncommon etiologies of recurrent acute pancreatitis. Biliary causes were excluded based on the absence of gallstones or sludge on MRCP and normal liver function tests (total bilirubin 13.0 umol/L, direct bilirubin 4.2 umol/L, indirect bilirubin 8.8 umol/L, and gamma-glutamyl transferase 26 U/L). Alcohol-induced pancreatitis was ruled out as the patient strongly denied alcohol consumption. A drug-induced etiology was considered unlikely as the patient was not taking any medications known to be pancreatotoxic prior to the onset of symptoms. Infectious causes, particularly parasitic infections that can cause eosinophilia, were excluded by a negative stool examination result. To rule out malignancy, tumor markers were confirmed to be within normal ranges (CEA 1.7 ng/ml, CA19-9 7.03 U/mL, CA125 8.32 U/mL, AFP 2.49 ng/ml), and imaging showed no evidence of a pancreatic mass. Crucially, distinguishing EP from type 1 autoimmune pancreatitis (AIP) was a priority; serum IgG4 levels were within the normal range (66.3 mg/dL; reference < 201 mg/dL), and antinuclear antibodies (ANA) and anti-neutrophil cytoplasmic antibodies (ANCA) were negative. Notably, a seafood allergy was diagnosed just 2 days before onset of the pancreatitis. Despite the patient denying the consumption of these foods prior to each episode, the ingestion of related products could not be ruled out.

Given this patient’s prior positive findings, a repeat gastroscopy was performed. Findings included chronic atrophic gastritis with erosions, duodenitis, and histologically confirmed chronic active gastritis with metaplasia. Notably, the gastric antrum (Figure 3D) and the duodenal bulb (Figure 3E) biopsies revealed significant eosinophilic infiltration (gastric antrum: 25 eosinophils/HPF, duodenum bulb: 73 eosinophils/HPF), further supporting a diagnosis of EGE (Figure 3). EUS showed mild dilatation of the pancreatic duct at the neck, with a maximum cross-sectional diameter of approximately 3.3 mm. The pancreatic duct at the head appeared tortuous and dilated, measuring 4.1 mm in maximum cross-sectional diameter, with no abnormalities detected in the visualized common bile duct. The visualized pancreatic body and tail were unremarkable (Figure 4). Endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) was performed using a 22-gauge EchoTip Ultra endoscopic ultrasound needle (Model ECHO-3-22; Cook Medical, Bloomington, IN, USA) to obtain tissue samples from the pancreatic head. Pancreatic head biopsy (via gastric/duodenal approach) revealed fibroinflammatory tissue with scant acinar epithelium and eosinophils (15/HPF) (Figure 5). Importantly, the specimen lacked the characteristic histological features of autoimmune pancreatitis, such as dense lymphoplasmacytic infiltration, storiform fibrosis, or obliterative phlebitis. No worsening of postoperative symptoms, such as abdominal pain and distension, was observed after the EUS-FNA procedure.

Based on the patient’s clinical course and investigation results, the final diagnosis was synthesized from a structured diagnostic framework consisting of 5 key elements: (1) the clinical presentation of recurrent acute pancreatitis; (2) persistent peripheral eosinophilia coinciding with symptom onset; (3) the systematic exclusion of common etiologies including biliary, alcoholic, and autoimmune causes; (4) evidence of concomitant eosinophilic gastroenteritis indicated by infiltration in gastric and duodenal biopsies; and (5) definitive histological confirmation of eosinophilic infiltration within the pancreatic parenchyma obtained via EUS-FNA.

Based on these findings, the patient was ultimately diagnosed with EP and EGE. He was treated with intravenous methylprednisolone (40 mg/day for 2 days), followed by intramuscular diphenhydramine (20 mg/day) as adjunctive symptomatic therapy, and achieved sustained remission. Abdominal pain and inflammatory markers significantly decreased. A notable correlation between peripheral eosinophil counts and disease activity was observed throughout the clinical course. During the initial admission 1 year prior, the eosinophil count was 1.23×10⁹/L (11%), which decreased to 0.29×10⁹/L (3.9%) after treatment. Upon the current recurrence, the count was similarly elevated at 0.82×10⁹/L (12.2%). After initiation of corticosteroid therapy, a rapid normalization was observed, with the count dropping to 0.26×10⁹/L (5.1%) within 1 week, paralleling the complete resolution of abdominal symptoms. The patient was discharged on day 8.

One week later, he was evaluated in the hematology outpatient clinic. The hematologist confirmed our preliminary diagnosis and prescribed oral prednisone acetate 30 mg once daily with a tapering regimen. A repeat MRCP performed after 2 weeks showed a reduction in the pancreatic duct diameter to 3.5 mm. Given the patient’s age and the recurrence of pancreatitis requiring systemic steroids twice within 1 year, the potential future use of steroid-sparing agents (eg, azathioprine) was considered as a long-term management strategy to prevent further relapses. At this stage, our clinical strategy was to first confirm a consistent and effective response to corticosteroids, which is the cornerstone of treatment. Given the patient’s advanced age, we opted for a cautious, stepwise approach to avoid the potential adverse effects associated with immunosuppressive steroid-sparing agents. Therefore, an immediate initiation was deferred in favor of observing the response to the current tapering steroid regimen.

Discussion

This case illustrates the critical importance of utilizing a comprehensive diagnostic framework for eosinophilic pancreatitis (EP). The definitive diagnosis in our patient was anchored on the convergence of a recurrent clinical course, documented peripheral eosinophilia, rigorous exclusion of alternative etiologies, and decisive histological confirmation.

Eosinophilic infiltration in pancreatic tissue is a rare pathological finding, with few documented cases in the literature [2]. Among the potential causes of pancreatic eosinophilia, EP is exceptionally uncommon. The incidence of an increased number of eosinophils is reported to be less than 1% in all pancreatic specimens [3,4]. This case of recurrent EP with concurrent, subclinical EGE contributes to the literature by highlighting the systemic nature of some eosinophilic disorders and underscoring the diagnostic utility of proactive endoscopy in pancreatitis of unknown etiology.

A key step in diagnosing EP is distinguishing it from other systemic eosinophilic syndromes where the pancreas may be just one of several affected organs. For a complete differential diagnosis, it is critical to evaluate for involvement of other common targets of eosinophilic inflammation, such as the cardiovascular and respiratory systems. In our patient, a thorough evaluation revealed no signs of cardiorespiratory involvement; he was asymptomatic with no cough, dyspnea, or wheezing, and a chest CT was unremarkable, showing no evidence of pulmonary infiltrates. Furthermore, a screening echocardiogram was normal, showing no signs of myocardial infiltration or dysfunction. The absence of these findings was pivotal in excluding a broader systemic vasculitis or hypereosinophilic syndrome, thereby strengthening the diagnosis of a process localized to the gastro-pancreatic system.

The clinical presentation of EP is highly variable, often mimicking more common conditions and posing a significant diagnostic challenge. In a systematic review, EP typically presents with symptoms such as jaundice and abdominal pain, but can also have atypical manifestations like nausea, fatigue, fever, vomiting, diarrhea, anorexia, and weight loss [1,5,6]. When analyzed by the affected organs, intrapancreatic usually manifests as acute pancreatitis, pancreatic enlargement, and resulting obstructive symptoms, while extrapancreatic is usually characterized by peripheral eosinophilia and eosinophilic infiltration into other organs [2]. The present case was characterized by recurrent pancreatitis as the predominant manifestation, with only mild peripheral eosinophilia. Thus, the diagnosis was more consistent with intrapancreatic.

The imaging findings of eosinophilic pancreatitis (EP) are often non-specific, potentially mimicking autoimmune pancreatitis, chronic pancreatitis, or even a pancreatic mass concerning for malignancy [7–9]. In our patient, while pancreatic duct dilation initially suggested chronic pancreatitis, a diagnosis of recurrent acute pancreatitis was ultimately favored. This conclusion was supported by a triad of evidence: the absence of specific imaging hallmarks of chronic disease (eg, calcifications), the lack of clinical pancreatic insufficiency, and radiological improvement following corticosteroid treatment [10]. The latter confirmed that the dilation was a reversible inflammatory consequence of eosinophilic infiltration rather than irreversible fibrosis. Nevertheless, it is crucial to recognize that recurrent acute pancreatitis from any cause, including EP, carries a risk of progression to chronic pancreatitis, which involves irreversible glandular damage. This long-term risk underscores the importance of a timely and accurate diagnosis, not only to rule out malignancy, but also to initiate appropriate anti-inflammatory therapy aimed at preventing permanent pancreatic injury. However, in some EP patients presenting with a pancreatic mass and complications, even pancreatic biopsy may not reliably distinguish the condition from pancreatic cancer. Consequently, this profound diagnostic uncertainty, driven by the imperative to exclude malignancy, has led to unnecessary surgical interventions, including radical pancreaticoduodenectomy, for a condition ultimately confirmed to be benign [8,9]. A key turning point in our case was the endoscopic discovery of EGE during the initial admission for pancreatitis. This finding in an otherwise asymptomatic patient was pivotal, shifting the differential diagnosis toward a systemic eosinophilic disorder and providing the justification for a targeted EUS-FNA during the subsequent admission. This approach secured a definitive histological diagnosis, thereby averting a more invasive and unnecessary surgical procedure.

Eosinophilic pancreatitis (EP) and autoimmune pancreatitis (AIP) share striking clinical similarities, including pancreatic duct dilation, pancreatic enlargement, and a dramatic therapeutic response to corticosteroids. However, relying solely on these shared features carries a significant risk of misdiagnosis. Corticosteroids act as non-specific anti-inflammatory agents equally effective in treating AIP, and imaging studies are often inadequate to distinguish EP from AIP, occult pancreatic malignancies, or lymphoma. Consequently, EP remains fundamentally a diagnosis of exclusion that requires histological confirmation.

While type 1 AIP is characterized by elevated serum IgG4 levels and lymphoplasmacytic infiltration with storiform fibrosis [11], EP is distinguished by peripheral eosinophilia and tissue infiltration predominantly by eosinophils, lacking the fibrosis typical of AIP. In our case, the normal serum IgG4 levels and the absence of AIP-specific histological features served as the pivotal basis for exclusion. This underscores that clinical symptoms and steroid responsiveness alone are insufficient for diagnosis; demonstrating eosinophilic infiltration while ruling out other etiologies is essential, highlighting the indispensable role of endoscopic ultrasound-guided fine-needle aspiration (EUS-FNA) in establishing a definitive diagnosis.

This case critically adds to our understanding of the relationship between pancreatic and gastrointestinal eosinophilic inflammation. While some reports describe a cause-and-effect mechanism, in which EGE-induced duodenal edema leads to pancreatic duct obstruction and subsequent pancreatitis [6], our patient lacked any obstructive features or significant gastrointestinal symptoms. This presentation suggests that the EP and EGE were parallel manifestations of a single systemic inflammatory process, rather than one causing the other. These findings align with the hypothesis that EP and EGE can exist within a broader spectrum, which we propose tentatively terming ‘eosinophilic gastro-pancreatic disease’ (EGPD), rather than strictly as isolated clinical entities.

The mechanism by which eosinophils are engaged in this type of pancreatitis remains unclear. Although EGPD is not yet an established disease entity, the underlying pathophysiology of this hypothesized spectrum is postulated to be driven primarily by a type I hypersensitivity reaction [12,13]. Our patient’s history of a seafood allergy and elevated serum IgE levels supports a systemic Th2-mediated immune response as the central mechanism [14,15]. In this pathway, activated Th2 cells secrete cytokines such as IL-4, IL-5, and IL-13. IL-4 and IL-13 are potent inducers of eotaxin-3 expression, which has been demonstrated in human pancreatic myofibroblasts [3,16,17]. Eotaxin-3 is a powerful chemokine, recruiting eosinophils to the tissue via the CCR3 receptor. The migration and adhesion of these eosinophils are further facilitated by the integrin β7/MAdCAM-1 pathway [16,18,19]. Once localized in the pancreas and gastrointestinal tract, activated eosinophils release cytotoxic granules (eg, major basic protein) and pro-inflammatory cytokines, causing direct tissue damage and perpetuating the inflammatory cycle [12,13].

Given the confirmed eosinophilic inflammatory nature of the disease, immunomodulation with corticosteroids was the logical therapeutic approach. Corticosteroids are potent inhibitors of eosinophil production, recruitment, and activation. Our patient’s rapid and sustained clinical and radiological improvement after methylprednisolone therapy is consistent with numerous reports [8,9].

The recurrence of pancreatitis in our patient raises the critical question of whether an earlier definitive diagnosis and a more prolonged steroid course could have prevented the relapse. In our case, the initial short course provided symptomatic relief but likely failed to eradicate the underlying eosinophilic burden. Had EP been confirmed via EUS-FNA initially, a structured tapering regimen would have been justified, which might have averted the second episode. However, it is important to acknowledge the lack of established guidelines for EP management regarding treatment duration. Our case suggests that in patients with EGE and concurrent pancreatitis, clinicians might consider a lower threshold for diagnostic biopsy and subsequent prolonged steroid therapy to mitigate the risk of recurrence.

The primary clinical lesson from this case is that clinicians should maintain a high index of suspicion for systemic eosinophilic disease in patients with recurrent pancreatitis of unknown etiology, especially when accompanied by peripheral eosinophilia. We propose that upper endoscopy with biopsies should be considered as part of the diagnostic workup in such patients, irrespective of the presence of gastrointestinal symptoms. This proactive approach can lead to an earlier diagnosis of EP and/or EGE, guiding appropriate therapy and preventing misdiagnosis or unnecessary surgery.

Our case highlights the utility of peripheral eosinophilia as a surrogate marker for monitoring disease activity. The temporal association was distinct: eosinophil counts surged during both acute exacerbations and were correlated with the recurrence of symptoms, while their normalization paralleled the clinical response to corticosteroids. However, it is crucial to recognize that while peripheral eosinophilia is supportive, it is not definitive. It can be absent in localized pancreatic disease or present in other systemic conditions. Thus, while valuable for tracking the clinical course, it does not replace histological confirmation.

Future research should focus on identifying non-invasive biomarkers (eg, serum eotaxin-3 levels) to aid in the diagnosis and monitoring of EP. Furthermore, prospective studies are needed to establish optimal long-term management strategies, including the efficacy of steroid-sparing agents and biologic therapies targeting eosinophilic pathways, to minimize the cumulative toxicity of corticosteroids.

Conclusions

EP is a rare yet clinically important subtype of pancreatitis that responds remarkably well to immunomodulatory therapy. This case supports the hypothesis that EP and EGE can be organ-specific manifestations of a shared eosinophilic inflammatory process, rather than independent occurrences. Consequently, identifying this link underscores the value of proactive endoscopy in unexplained pancreatitis. An accurate histological diagnosis is key to initiating effective corticosteroid therapy, preventing disease progression and averting unnecessary surgery. Future research should aim to develop non-invasive diagnostic markers and establish evidence-based guidelines for long-term management to improve outcomes for patients with this rare disease.