Severe Recurrent Gastrointestinal Bleeding and Diagnostic Dilemmas in Zollinger-Ellison Syndrome: A Case Report

Abstract

Zollinger-Ellison Syndrome (ZES) is a rare condition caused by gastrin-secreting neuroendocrine tumours (gastrinoma) that induce excessive gastric acid production, leading to severe peptic ulcer disease (PUD) and gastrointestinal (GI) complications. Its nonspecific presentation often delays diagnosis, increasing the risk of life-threatening outcomes. We report the case of a 53-year-old man with chronic gastroesophageal reflux disease and long-standing unexplored diarrhoea who presented with worsening abdominal pain and severe diarrhoea. Laboratory tests showed acute kidney injury, and CT revealed gastric wall thickening and a hyperdense nodule at the gastroduodenal junction, raising suspicion for ZES. Despite proton pump inhibitor (PPI) therapy, he developed recurrent upper GI bleeding requiring intensive care unit admission. Hormonal and imaging studies confirmed ZES, and he underwent antrectomy with lymphadenectomy. Postoperatively, he experienced further GI bleeding managed with over-the-scope clip (OTSC®) haemostasis and gastroduodenal artery ligation. Pathology confirmed a well-differentiated neuroendocrine tumour (pT2N1, G1). One year later, he remained asymptomatic with no evidence of recurrence. This case illustrates the diagnostic and therapeutic challenges of ZES, underscoring the risks associated with PPI withdrawal during evaluation and the need for multidisciplinary management. Recurrent GI bleeding in ZES requires a coordinated approach integrating medical, endoscopic, and surgical therapies. Early recognition is essential to prevent severe complications, and a high index of suspicion is warranted in patients with refractory PUD, chronic diarrhoea, or unexplained GI bleeding.

Introduction

Zollinger-Ellison Syndrome (ZES) is characterized by hypergastrinemia leading to excessive gastric acid secretion and the development of classic manifestations, including atypical peptic ulcer disease (PUD), diarrhoea, abdominal pain, and gastroesophageal reflux disease (GERD). The condition results from a gastrinoma – most often duodenal – located within the gastrinoma triangle. With an estimated annual incidence of 1–1.5 cases per million, ZES is predominantly sporadic (80%), while 20–30% of cases occur in the context of multiple endocrine neoplasia type 1 (MEN-1). Because symptoms are nonspecific, diagnosis is frequently delayed for more than 5 years, making clinical suspicion crucial. The classic biochemical diagnosis requires fasting serum gastrin (FSG) >1,000 pg/mL with gastric pH ≤2 [1]. To improve accuracy, proton pump inhibitors (PPIs) should ideally be discontinued for at least 7 days before testing [2–4].

However, given the risks of PPI withdrawal, alternative diagnostic criteria have been proposed to avoid stopping therapy, as PPIs are essential for symptom control and for preventing ulcer-related complications – the main cause of short-term mortality. These criteria allow diagnosis without FSG or gastric pH assessment. Imaging is fundamental for tumour localization and staging, with modalities such as endoscopic ultrasonography and 68 Ga-DOTATOC PET offering high sensitivity [1].

Management depends on the underlying aetiology (sporadic vs. MEN-1) and disease stage, requiring multidisciplinary discussion. Treatment may involve surgery or medical therapy [5].

In patients without liver metastases, 15-year survival exceeds 80%. Nonetheless, long-term follow-up is essential due to the risk of nodal or distant metastases and to monitor potential adverse effects associated with chronic PPI therapy [2].

Case Report

We present the case of a 53-year-old male with a history of GERD complicated by peptic oesophagitis, previously treated with esomeprazole 20 mg twice daily but discontinued 1 week prior to admission. He also reported a 3-year history of chronic diarrhoea that had not been previously investigated.

He presented to the emergency department with abdominal pain and worsening non-bloody chronic diarrhoea over the preceding 5 days (>10 stools per day), associated with significant unintentional weight loss (6 kg over 3 months). On examination, GCS 15, visibly dehydrated, hypotensive (80/50 mm Hg) and tachycardic (HR 110 bpm). The abdomen was soft, non-distended and depressible, with tenderness in the epigastric region but no rebound tenderness.

Initial laboratory evaluation revealed haemoglobin of 16 g/dL (MCV 86 fL, haematocrit 50%), no leucocytosis, and a C-reactive protein level of 1 mg/dL. Acute kidney injury was present (creatinine 5.35 mg/dL, urea 90 mg/dL – AKIN III), without hyperkalaemia or metabolic acidosis. Amylase was elevated (180 U/L). Stool cultures were negative. Abdominopelvic CT demonstrated diffuse gastric wall thickening and a hyperdense 20-mm nodule at the gastroduodenal junction (shown in Fig. 1).

Fig. 1.

Axial abdominal CT images. a Diffuse gastric wall thickening (arrow). b Nodule at the gastroduodenal junction (circle).

A diagnosis of ZES was considered, and treatment with pantoprazole 80 mg twice daily and intravenous fluid resuscitation was initiated. This resulted in improvement of abdominal pain, reduction of diarrhoea (to 2–3 stools per day), and normalization of renal function (creatinine 0.73 mg/dL, urea 54 mg/dL). After 1 week of PPI therapy, pantoprazole was replaced with a histamine-H₂-receptor antagonist (H₂RA; famotidine 20 mg twice daily) to enable hormonal testing. However, 72 h after this switch, the patient developed melena with haemodynamic instability. Laboratory results showed haemoglobin of 8 g/dL, creatinine 1.12 mg/dL and urea 140 mg/dL.

Upper GI endoscopy (UGIE) revealed severe oesophagitis with necrotic mucosal changes, as well as extensive and diffuse duodenal ulceration, including Forrest IIc and III ulcers (shown in Fig. 2). High-dose pantoprazole infusion (240 mg/day) was restarted. Subsequent gastric pH and fasting serum gastrin (FSG) levels were 2.9 and 4,390 ng/L, respectively. Chromogranin A was elevated at 880.5 nmol/L, and urinary 5-HIAA was 12 mg/24 h (creatinine 0.87 mg/dL).

Fig. 2.

Endoscopic findings. a Severe oesophagitis with necrotic mucosal alterations. b Extensive duodenal PUD Forrest IIc and III ulcers.

Anti-parietal cell and anti-intrinsic factor antibodies were negative. Screening for MEN-1-associated neoplasms was unremarkable. Given the strong clinical suspicion of ZES, a 68 Ga-DOTATOC PET scan was requested to confirm tumour localization and assess staging.

On the 5-day of hospitalization, the patient experienced recurrent upper GI bleeding (UGIB), haematochezia and haemodynamic instability, requiring admission to the intensive care unit (ICU). UGIE showed a Forrest IIa duodenal ulcer at the bulb-D2 junction, which was successfully treated with combined haemostatic therapy (adrenaline injection and hemoclip placement) (shown in Fig. 3). Due to the recurrent UGIB, octreotide (100 µg subcutaneously every 8 h) was added to the PPI regimen. The 68 Ga-DOTATOC PET confirmed a hypermetabolic nodule at the gastroduodenal junction with two locoregional lymph nodes. After multidisciplinary discussion, a diagnosis of sporadic ZES was established, and the patient underwent antrectomy with lymphadenectomy (Roux-en-Y reconstruction), without perioperative complications. Notably, the duodenal ulcer was not included in the resected specimen, as its excision would have required a Whipple procedure.

Fig. 3.

Endoscopic findings. a Forrest IIa duodenal ulcer. b After combined haemostatic endoscopic therapy.

Two days after surgery, the patient developed a second UGIB with haematochezia and haemodynamic instability. Rebleeding from the bulb-D2 ulcer – previously treated but not resected – was suspected. Endoscopic access to D2 was initially impossible due to large clots. A laparotomy with gastroduodenal artery ligation and intraoperative retrograde enteroscopy via enterostomy is a specialized surgical technique. Endoscopically, a Forrest Ib ulcer, 1 cm from the duodenal stump, was identified and successfully treated with OTSC placement (OTSC^®, Ovesco Endoscopy AG, Tübingen, Germany) (shown in Fig. 4). Immediate high postoperative gastrin levels, failure of previous endoscopic therapy and exclusion of the ulcer from the resected specimen likely contributed to this recurrence.

Fig. 4.

Endoscopic findings. a Forrest Ib duodenal ulcer. b After haemostatic endoscopic therapy with over-the-scope clip.

The patient recovered well, remained hemodynamically stable without further UGIB, and showed a normalization of postoperative FSG level. Histopathology confirmed a well-differentiated neuroendocrine tumour (NET), pT2N1 (G1), with a low mitotic rate (<1/2 mm²), Ki-67 <3%, and 1/15 metastatic lymph nodes (R2 resection) (Fig. 5). After multidisciplinary discussion, given the R2 status, a watchful-waiting approach with continued PPI therapy was adopted. At 1-year follow-up, the patient remained asymptomatic, with improved quality of life, maintained on pantoprazole 80 mg twice daily, without recurrence.

Fig. 5.

Histopathological study. a HE. ×10. Neoplastic cells . b Chromogranin positive (×10). c Synaptophysin positive (×10). d Ki-67 inferior to 3% (×10).

Discussion

We present a rare and successfully managed case of ZES. This case underscores the diagnostic and staging challenges of gastrinoma and highlights the need for a coordinated multidisciplinary approach to address one of its most severe complications.

In this case, the combination of GERD and chronic diarrhoea raised clinical suspicion 0f ZES. GERD is a common ZES manifestation and often presents with typical symptoms; however, patients have a higher risk of oesophageal complications, including oesophagitis and peptic strictures [2, 6]. NETs are a rare cause of chronic diarrhoea but should be considered when common aetiologies are excluded. Among duodenal-pancreatic NET syndromes, chronic diarrhoea occurs in 100% of vipomas, 65% of gastrinomas, and 15% of glucagonomas [7, 8].

ZES-related PUD shows characteristic, though not fully specific, endoscopic features, including multiple ulcers, involvement beyond the first or second duodenal portion, and complications such as bleeding, penetration/perforation, or strictures. Prominent gastric folds on UGIE are a sensitive finding, reported in up to 94% of patients, and are generally absent in other causes of hypergastrinemia [6, 9].

Because ZES symptoms arise from acid hypersecretion, PPIs effectively relieve symptoms and reduce PUD-related complications. However, PPI therapy complicates ZES diagnosis, creating a key dilemma: should PPIs be stopped before measuring FSG and gastric pH? Off antisecretory therapy, FSG has high sensitivity (98–100%) for ZES, allowing exclusion when normal. However, elevated FSG lacks specificity to confirm the diagnosis [1, 10].

FSG elevation can be classified as: (1) appropriate – a physiological response to hypo- or achlorhydria; or (2) inappropriate – hypergastrinemia in the presence of hyperchlorhydria, as in ZES. When FSG is elevated, chronic atrophic gastritis, Helicobacter pylori-associated pangastritis, and antisecretory therapy must be excluded. Renal dysfunction, although less common, is another relevant cause. In this case, renal function had normalized by the time of measurement [11–13].

The classic diagnostic criteria for ZES include FSG >10× the upper limit (>1,000 pg/mL) and gastric pH ≤2 in the absence of antisecretory therapy. When FSG is elevated but <1,000 pg/mL, or when measured on-PPI therapy, alternative diagnostic strategies are required [1, 11, 14, 15]. When these criteria were first established, to minimize false positives, PPIs were traditionally stopped for at least 1 week and replaced with an H₂RA, which was then discontinued 12 h before sampling [1, 4, 13].

However, due to the risks of PPI withdrawal, alternative diagnostic criteria have been proposed. Metz et al. [1] suggest that ZES is likely in two scenarios: (1) a patient without MEN-1, with active or recent PUD, whose diarrhoea improves on-PPI therapy and with a positive 68 Ga-DOTATOC PET; or (2) a patient with or without MEN-1, with active or recent PUD, whose diarrhoea improves on-PPI therapy and with histological or cytological confirmation of a NET.

With advances in endoscopic and imaging techniques – UGIE with EUS, CT/MRI, and somatostatin receptor imaging – current guidelines have largely replaced the secretin stimulation test and measurement of gastric basal acid output [15]. A proposed diagnostic algorithm is shown in Figure 6.

Fig. 6.

Proposed diagnostic algorithm.

In our case, a classic diagnostic approach was used, and a temporal association between PPI withdrawal and bleeding suggested a possible causal relation. After the first on-PPI UGIB, FSG was >1,000 pg/mL and gastric pH was 2.9. Because EUS was not feasible due to active PUD, a PET scan was requested.

The authors believe that PPI discontinuation in this case triggered PUD-related UGIB, with two recurrences requiring ICU admission. This raised a second dilemma: what is the optimal management strategy for such complications?

The first recurrence was controlled with standard endoscopic haemostasis (through-the-scope hemoclip). The second, complicated by postoperative anatomy, required a combined surgical-endoscopic approach to minimize further recurrence and morbidity. Because the ulcer was at the bulb-D2 transition – a high-risk site with prior severe bleeding – OTSC therapy (OTSC^®, Ovesco Endoscopy AG, Tübingen, Germany) was selected. The OTSC system has transformed GI bleeding management and is an effective rescue therapy, lowering rebleeding rates and the need for urgent surgery. It may be used as first-line therapy for high-risk gastroduodenal ulcers or second-line after failure, with low failure rates of 6.2% (95% CI: 3.1–9.2%) and 16.9% (95% CI: 9.3–24.5%) [16–18]. OTSC deployment via enterostomy was technically demanding due to limited luminal distension but was both technically and clinically successful. The patient remained on optimized PPI therapy, and octreotide was continued until surgery. This strategy achieved haemostasis after the second recurrence.

Gastrinomas are associated with MEN-1 in 20–30% of cases, although not in this patient. MEN-1-related ZES typically presents with multiple primary tumours in different organs, earlier onset (<40 years), and a characteristic inheritance pattern [15]. All ZES patients should be screened with serum parathyroid hormone, calcium, and prolactin. Gastrinomas in the MEN-1 setting show improved overall and progression-free survival [2, 19, 20].

However, it remains uncertain whether duodenotomy with lymphadenectomy or pancreaticoduodenectomy offers superior long-term cure. Retrospective studies often favour duodenotomy with lymphadenectomy, as survival outcomes do not differ significantly from pancreaticoduodenectomy, while complication rates are lower [15, 21].

In this case, a noncurative resection was performed, leaving one locoregional lymph node in situ in an asymptomatic patient with indolent disease and favourable histopathology. This raised a third dilemma: should adjuvant therapy be considered, or is surveillance appropriate?

For residual disease in G1 duodenal NETs, both ENETS and ESMO guidelines report insufficient evidence to support or oppose adjuvant therapy; therefore, it is not recommended [15, 21, 22]. In nonsurgical candidates, medical therapy aims to control both hormone-related symptoms and tumour growth. In gastrinoma, PPIs are first-line for symptom control, with somatostatin analogues recommended as second-line when symptoms persist despite PPI therapy [15]. In our case, given the R2 resection, PPI therapy was maintained.

Duodenal gastrinomas show locoregional lymph-node positivity in 50–67% of surgical specimens, although its prognostic significance remains uncertain and depends on additional tumour characteristics [23, 24]. Poor prognostic factors include rapid progression, pancreatic location, tumour size ≥20 mm, liver metastases at diagnosis, and a high Ki-67 index [12, 14, 24, 25].

In this patient, surveillance was chosen given the indolent course, absence of symptoms, normalized gastrin levels, and favourable histopathology. No standardized surveillance protocol exists, but clinical evaluation, FSG measurement, and imaging every 3–9 months are suggested for G1-G2 NETs. If disease remains stable, intervals may be extended, with no proven benefit to follow-up beyond 10 years. For patients with a positive 68 Ga-DOTATOC PET, repeat imaging every 2 years – or earlier if recurrence is suspected – is recommended [14, 22].

Long-term PPI therapy remains essential after surgical treatment, at least for 3–6 months following curative resection, as some patients may continue to exhibit gastric acid hypersecretion [14, 15]. PPIs are preferred over H₂RAs due to their superior potency, longer duration of action, and simple twice-daily regimen [4]. In the long term, it is important to consider dose adjustment based on efficacy and adverse effects, including vitamin B12 deficiency and hypomagnesemia. This has recently prompted evaluation of routine gastric acid testing to individualize PPI therapy, further supporting the benefit of managing these patients in specialized centres [2].

This case underscores the diagnostic and therapeutic complexity of ZES, illustrating the hazards of PPI withdrawal, the challenges of managing severe PUD-related bleeding, and the nuances of treating duodenal NETs. It reinforces the need for early recognition, individualized acid-suppression strategies, and careful integration of endoscopic, surgical, and imaging-based approaches. Optimal management – particularly following noncurative resection of indolent NETs – remains uncertain, highlighting the importance of multidisciplinary decision-making and long-term, risk-adapted surveillance. Continued research is essential to refine treatment algorithms and improve outcomes in this rare but potentially life-threatening entity.

Statement of Ethics

The presentation/publication of this case report was approved by the Health Ethics Committee of the Local Health Unit of Amadora/Sintra, with registration ID number 035/2024. The CARE Checklist has been completed by the authors for this case report and is attached as online supplementary material (for all online suppl. material, see https://doi.org/10.1159/000550071). Verbal and written informed consent was obtained from the patient for presentation/publication of this case report and any accompanying images.

Conflict of Interest Statement

The authors have no conflicts of interest to declare.

Funding Sources

This study was not supported by any sponsor or funder.

Author Contributions

Sofia Bragança: writing – original draft. Ana Maria Oliveira: supervision and writing – review and editing. Luísa Martins Figueiredo, Carolina Padrão, Mariana Nuno Costa, and David Horta: writing – review and editing.

Funding Statement

This study was not supported by any sponsor or funder.

Data Availability Statement

Not applicable.