ABSTRACT
Primary hepatic neuroendocrine tumors (PHNETs) are rare, often presenting with abdominal pain, carcinoid syndrome, or incidental imaging findings. PHNET-associated portal hypertension is uncommon. A 45-year-old man with known PHNET presented with massive upper gastrointestinal bleeding from a duodenal ectopic varix and hemorrhagic shock. Emergent esophagogastroduodenoscopy achieved hemostasis with hemoclips. He subsequently developed high-serum-ascites albumin gradient (1.4 g/dL); imaging showed a 23-cm hepatic mass, portal vein dilatation, and splenomegaly. Liver biopsy confirmed a well-differentiated neuroendocrine neoplasm without cirrhosis, supporting tumor-related, noncirrhotic portal hypertension likely from mechanical vascular compression. This case highlights the rare cause of variceal bleeding due to noncirrhotic portal hypertension in patients with PHNET.
KEYWORDS: primary hepatic neuroendocrine tumor, noncirrhotic portal hypertension, gastrointestinal bleeding, ectopic duodenal varices
INTRODUCTION
Primary hepatic neuroendocrine tumors (PHNETs) are rare neoplasms arising from neuroendocrine cells within the liver. Unlike metastatic neuroendocrine tumors, which commonly involve the liver, PHNETs pose diagnostic challenges owing to their rarity and nonspecific presentation, which includes abdominal pain, carcinoid syndrome, and incidental imaging findings.1
Portal hypertension is most often associated with cirrhosis; however, noncirrhotic causes such as hepatic tumors, vascular disorders, and infiltrative diseases should also be considered. Tumor-related portal hypertension is typically attributed to mechanical compression or distortion of the portal venous system.
We report a case of PHNET presenting with massive upper gastrointestinal bleeding from an ectopic duodenal varix secondary to tumor-related, noncirrhotic portal hypertension.
CASE REPORT
A 45-year-old man with a known PHNET presented to the emergency department with hematemesis and melena complicated by hemodynamic instability. The PHNET had been diagnosed in India in 2016 as a large, well-differentiated Grade 1 tumor (Ki-67 2%). He received sequential systemic therapy: octreotide (2016–2017), peptide receptor radionuclide therapy (4 cycles, 2017–2018), capecitabine/temozolomide for documented progression (December 2022–March 2023), bevacizumab with octreotide (March–June 2023; 4 cycles), and cisplatin from July 2023, complicated by neutropenia. He subsequently relocated to Qatar; this was his first presentation, with no history of gastrointestinal bleeding.
On arrival, he was critically ill, with tachycardia (130 beats per minute), hypotension (mean arterial pressure ∼50 mm Hg), cold clammy extremities, and hepatosplenomegaly. Investigations showed severe anemia (hemoglobin 1.6 g/dL), acute kidney injury (creatinine 145 µmol/L), and mildly elevated alanine aminotransferase (71 U/L). Coagulation studies showed international normalized ratio 2.5, elevated D-dimer, and reduced fibrinogen, attributed to impaired hepatic synthetic function and consumptive coagulopathy in the setting of acute hemorrhagic shock. He was admitted to the intensive care unit and managed with a massive transfusion protocol, intravenous fluids, and proton-pump inhibitor therapy.
Emergent esophagogastroduodenoscopy identified a bleeding vessel in the second part of the duodenum, consistent with an ectopic varix. Hemostasis was achieved with 2 endoscopic hemoclips, deployed because of active bleeding and hemodynamic instability (Figure 1). Small esophageal varices without active bleeding or red wale signs were also identified (Figure 2).
Figure 1.

Esophagogastroduodenoscopy demonstrating an ectopic duodenal varix (enclosed area) following successful hemostasis with hemoclip placement.
Figure 2.

Esophagogastroduodenoscopy demonstrating small esophageal varices without red wale sign.
He subsequently developed new-onset ascites. Paracentesis showed an ascitic albumin of 0.39 g/dL and total protein of 1.14 g/dL; with a serum albumin of 1.8 g/dL, the serum-ascites albumin gradient was 1.4 g/dL (≥1.1 g/dL), consistent with portal hypertension-related ascites, while the low ascitic total protein further supported a transudative etiology. Ascitic white blood cell count was 112 cells/µL with 15% neutrophils; cultures and cytology were negative, excluding malignant ascites.
Abdominal ultrasound demonstrated the previously known PHNET, now measuring approximately 23 cm; a dilated portal vein (17 mm); and splenomegaly (20 cm), with coarse hepatic echotexture but no radiological evidence of cirrhosis. Magnetic resonance imaging revealed a large, complex, heterogeneous hepatic mass with solid, cystic, and hemorrhagic components (Figure 3); portal vein thrombus was absent. Gallium-68 DOTA-(Tyr3)-octreotate positron emission tomography ruled out extrahepatic metastasis.
Figure 3.

Magnetic resonance imaging of the liver demonstrating a large heterogeneous neuroendocrine tumor (arrow) with solid, cystic, and hemorrhagic components.
Endoscopic ultrasound, performed after stabilization, confirmed ectopic varices in the second portion of the duodenum (Figure 4). Histopathology of the liver mass and surrounding parenchyma confirmed a well-differentiated, low-grade neuroendocrine neoplasm without features of cirrhosis. He was started on a nonselective beta-blocker and diuretics, and the oncology multidisciplinary team resumed octreotide; interval DOTATATE PET showed stable disease, with planned surveillance using chromogranin A, DOTATATE PET, and alpha-fetoprotein.
Figure 4.

Endoscopic ultrasound demonstrating Doppler flow within ectopic duodenal varices, confirming their vascular nature.
DISCUSSION
PHNETs are an exceptionally rare subset of neuroendocrine neoplasms, in contrast to the far more common hepatic metastasis from extrahepatic primaries; only a few hundred cases have been reported.1–3 The typical age at diagnosis is 45–50 years, consistent with our patient.1
Radiological findings are nonspecific: PHNETs may appear as heterogeneous masses with solid, cystic, or necrotic components, making differentiation from hepatocellular carcinoma or other hepatic malignancies challenging. In our patient, MRI showed a large heterogeneous lesion with solid, cystic, and hemorrhagic components, and liver biopsy remained essential for definitive diagnosis, demonstrating a well-differentiated NET (Ki-67 2%).4 Clinically, PHNETs may present with vague abdominal pain, incidental imaging findings, carcinoid syndrome, or ascites; our patient presented with upper gastrointestinal bleeding from a duodenal ectopic varix, a complication of the disease.4
Surgical resection is the mainstay of treatment and may offer long-term survival or potential cure in localized disease.4,5 In unresectable or advanced disease, options include transarterial chemoembolization, systemic chemotherapy, and other antitumor therapies depending on tumor grade and disease extent.3 Our patient had received an extensive sequential regimen with documented progression; the resulting increase in tumor size likely produced significant mass effect on adjacent vascular structures.
Portal hypertension secondary to hepatic malignancy is uncommon and has been described more often in the context of metastatic disease than in primary hepatic tumors.6,7 Pathophysiologically, it differs from cirrhosis: Tumor-related distortion of hepatic vascular architecture increases intrahepatic resistance, a mechanism termed “pseudocirrhosis” because of its radiological and clinical resemblance to cirrhosis in the absence of classical cirrhotic pathology.8,9
Several features support a tumor-compression mechanism of noncirrhotic portal hypertension in our patient: the absence of cirrhosis on biopsy and imaging; a 23-cm tumor with adjacent portal vein dilatation and splenomegaly consistent with mechanical compression; exclusion of portal vein thrombus on cross-sectional imaging; and high-serum-ascites albumin gradient with negative cytology confirming portal hypertension-related rather than malignant ascites.
In summary, this case highlights a rare and previously unreported association between PHNET and clinically significant portal hypertension with variceal hemorrhage. Clinicians should consider noncirrhotic causes of portal hypertension in patients with large hepatic tumors presenting with ascites or gastrointestinal bleeding, in whom early recognition and combined endoscopic and pharmacologic management are essential to reduce morbidity and prevent recurrent bleeding. Further studies are needed to characterize the vascular consequences of PHNETs.
DISCLOSURES
Author contributions: MU Saddique: Contributed to manuscript writing and data collection; AAA Ali: Contributed to manuscript writing. SM Ghazouani: Served as faculty supervisor, performed the endoscopy, and revised the manuscript. All authors have reviewed and approved the final version of the manuscript. MU Saddique is the article guarantor.
Acknowledgment: The authors acknowledge the patient and their family for their kind permission to publish this case report.
Financial disclosure: None to report.
Ethics approval and consent to participate: This study was reviewed and approved by the Medical Research Council of Hamad Medical Corporation, Doha, Qatar, under the condition that no personal identifying information would be included in the publication. Approval number: MRC-04-24-797.
Data availability: All data generated or analyzed during this study are included in this published article. Additional details are available from the corresponding author upon reasonable request, in accordance with institutional policies and patient confidentiality requirements.
Informed consent was obtained for this case report.
ABBREVIATIONS:
- AFP
alpha-fetoprotein
- HCC
hepatocellular carcinoma
- PHNET
primary hepatic neuroendocrine tumor
- SAAG
serum–ascites albumin gradient
Contributor Information
Ahmed Abdulhafedh Ahmed Ali, Email: Aali106@hamad.qa.
Souraia Mezhoud Ghazouani, Email: SGhazouani@hamad.qa.
REFERENCES
- 1.Elayan A, Batah H, Badawi M, Saadeh A, Abdel Hafez S. Primary hepatic neuroendocrine tumor: A case report and literature review. Cureus. 2022;14(2):e22370. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Shah D, Mandot A, Cerejo C, Amarapurkar D, Pal A. The outcome of primary hepatic neuroendocrine tumors: A single-center experience. J Clin Exp Hepatol. 2019;9(6):710–5. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3.Nikfarjam M, Muralidharan V, Christophi C. Primary hepatic carcinoid tumours. HPB (Oxford). 2004;6(1):13–7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Wang LX, Liu K, Lin GW, Jiang T. Primary hepatic neuroendocrine tumors: Comparing CT and MRI features with pathology. Cancer Imaging. 2015;15(1):13. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Sugawara T, Shindoh J, Hoshi D, Hashimoto M. Intrahepatic cholangiocarcinoma and portal hypertension developing in a patient with multicystic biliary microhamartomas. Malays J Pathol. 2018;40(3):331–5. [PubMed] [Google Scholar]
- 6.Oliver TI, Sharma B, John S. Portal hypertension. In: StatPearls. StatPearls Publishing, 2026. Accessed April 15, 2026. http://www.ncbi.nlm.nih.gov/books/NBK507718/ [PubMed] [Google Scholar]
- 7.Theophilidou E, Waraich N, Raza T, Agarwal PK. Liver metastases, a rare cause of portal hypertension and stoma bleeding: Brief review of literature. Int J Surg Case Rep. 2012;3(5):173–6. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Kang SP, Taddei T, McLennan B, Lacy J. Pseudocirrhosis in a pancreatic cancer patient with liver metastases: A case report of complete resolution of pseudocirrhosis with an early recognition and management. World J Gastroenterol. 2008;14(10):1622–4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Villani R, Di Cosimo F, Sangineto M, Romano AD, Serviddio G. Pseudocirrhosis and portal hypertension in patients with metastatic cancers: A systematic review and meta-analysis. Sci Rep. 2022;12(1):19865. [DOI] [PMC free article] [PubMed] [Google Scholar]
