Clinical outcomes of medical therapy, EUS-guided ethanol ablation, and surgical resection in insulinoma: a clinical epidemiology study.
Journal
Frontiers in Endocrinology
Journal Volume
17
Start Page
Article number 1848169
ISSN
1664-2392
Date Issued
2026
Author(s)
Lu, Jin-Ying
Chou, Fan
Wen, Fang-Yu
Chiu, Wei-Yi
Kuo, Ting-Chun
Wu, Wan-Chen
Tien, Yu-Wen
Hu, Fu-Chang
Chen, Pei-Lung
Shih, Shyang-Rong
Wang, Chih-Yuan
Wang, Hsiu-Po
Abstract
Introduction – Insulinoma is a rare pancreatic neuroendocrine tumor causing hypoglycemia. Surgery is the standard treatment, while endoscopic ultrasound (EUS)-guided ethanol ablation and medical therapy are alternatives in some patients. Methods – In this retrospective study, we recruited 19 patients with insulinoma treated between 2011 and 2025 at National Taiwan University Hospital. The final treatment strategies were surgery (n = 13), EUS-guided ethanol ablation (n = 3), and medical therapy (n = 3). The study outcomes included post-procedure stays (PPS), treatment-related complications, and follow-up results. New-onset diabetes and complete remission rates for insulinoma were examined and compared across treatment modalities. A Firth’s bias-reduced logistic regression analysis was performed to identify predictors of complete remission in 19 patients with insulinoma. Results – Patients could receive more than one treatment. Among the three final treatment modalities, surgical resection achieved the highest complete remission rate (13/13; 100%) but was associated with a longer PPS (mean ± standard deviation [SD] = 13.46 ± 7.59 days) and a higher incidence of new-onset diabetes (2/13, 15.38%). EUS-guided ethanol ablation demonstrated moderate efficacy (2/3; 66.67%) with a comparable PPS (18.33 ± 27.43 days). Lastly, medical therapy relieved symptoms but yielded no complete remissions in insulinoma (0/3; 0%). Paradoxical hypoglycemia occurred in 2 of 8 patients (25.0%) during octreotide testing or after direct octreotide treatment. One patient with insulin autoimmune syndrome improved after drug withdrawal and multimodal therapy. Logistic regression analysis identified a positive linear effect of baseline glucose levels (mg/dL) on insulinoma complete remission (estimated adjusted odds ratio [aOR] = 1.2140, p = 0.0461) and ever surgery (yes vs. no) as the most effective treatment for complete remission in insulinoma (aOR = 296.0707, p = 0.0015). Conclusion – Although surgery was the most effective treatment, it was associated with longer recovery times and potential metabolic risks. EUS-guided ethanol ablation provided a less invasive alternative with a moderate success rate, whereas medical therapy primarily offered symptomatic control rather than a definitive complete remission. Notably, lower baseline glucose levels were associated with treatment resistance, suggesting a potential role in guiding individualized therapeutic strategies.
Subjects
Firth’s bias-reduced logistic regression model
ethanol ablation
insulin autoimmune syndrome
insulinoma
somatostatin receptor ligands
surgery
Publisher
Frontiers Media SA
Type
journal article