Does X mark the spot? A case series of discrepant preoperative insulinoma localization

Abstract

Summary

We explored the challenges associated with the localization of insulinomas, pancreatic neuroendocrine tumors responsible for hypoglycemia. Insulinomas often present with varied symptoms, leading to potential diagnostic delays. While biochemical evidence confirms the presence of insulinomas, precise preoperative localization remains challenging despite advanced imaging techniques. Our study highlights contradictory findings between intra-arterial calcium stimulation tests and ⁶⁸Ga-NODAGA-Exendin-4-PET/CT in two cases of insulinoma where conventional imaging studies did not reveal the location. We demonstrate that in these cases, surgical exploration resolved the discrepancies and led to a resolution of the disease. Our results also suggest a potential need to revise the threshold for intra-arterial calcium stimulation tests to enhance diagnostic accuracy. This short report emphasizes the complexity of insulinoma localization and the necessity for integrating multiple diagnostic approaches to achieve optimal patient care.

Learning points

• Diagnosis of insulinoma can be delayed and complicated by unspecific symptoms and difficulties in the localization techniques.

• There are different methods for preoperative insulinoma localization which may render conflicting results.

• When conventional imaging studies do not reveal the location and advanced imaging techniques and metabolic tests show discrepant results, surgical exploration may resolve the discrepancies and lead to a resolution of the disease.

• It is necessary to integrate multiple diagnostic approaches to achieve optimal patient care.

Introduction

Insulinomas are pancreatic neuroendocrine tumors leading to hypoglycemia with autonomic and neuroglycopenic symptoms (1, 2) due to uncontrolled insulin secretion. Upon securing biochemical evidence of insulinoma, different methods have been advocated for tumor localization (3). Difficulties may arise from non-aligning localization results. In this case series, we describe challenges in tumor localization with intra-arterial calcium stimulation tests (4, 5, 6, 7) and positron emission tomography–computed tomography (PET/CT) utilizing ⁶⁸Ga-NODAGA-Exendin-4 tracer (8).

Methods

Case series

We present two persons (case 1: male, late 40s, and case 2: female, late 50s) with a history of neuroglycopenic symptoms, syncope, and documented episodes of hypoglycemia. They were admitted for further work up because they met the diagnostic criteria for insulinoma (9, 10, 11).

Case 1 reported recurrent episodes of metamorphopsia and pseudohallucinations of up to 60 min duration. He further described frequent bouts of dizziness, falls, and recurring syncope over the last 9 years. Sugar tablets led to a rapid improvement of these symptoms. Previous neurological, cardiological, and psychiatric examinations were inconclusive. One year before, he was admitted unconscious with a blood glucose level of 16 mg/dL to an emergency department. He promptly recovered after administration of i.v. glucose (12, 13). There were neither pre-existing conditions nor any noteworthy family history.

Case 2 had similar neuroglycopenic symptoms, mainly metamorphopsia and concentration difficulties. She also reported a rapid improvement of symptoms after ingestion of sugar tablets. She had increasing intensity and frequency of symptoms, which led to a referral to our Department of Endocrinology and Diabetology.

Results

Clinical findings

The clinical examination of case 1 (BMI 23.5 kg/m², regular vital signs) and case 2 (BMI 19.9 kg/m², regular vital signs) was unremarkable. No traces of sulfonylurea were detected. Insulin antibody tests ruled out the presence of an insulin autoimmune syndrome.

Standardized fasting test

In case 1, the standardized 72 h fasting test (10) was terminated after 30 min at a blood glucose level of 35 mg/dL. The corresponding serum insulin concentration (18.9 mU/L), C-peptide (3.35 μg/L), and pro-insulin (38.0 pmol/L) were inadequately elevated. In case 2, the patient reached a blood glucose level of 33 mg/dL after 8 h with a corresponding serum insulin concentration of 13 mU/L. Likewise, inadequate elevation of C-peptide (3.30 μg/L) and pro-insulin (48.0 pmol/L) was observed.

Localization

Conventional abdominal ultrasound, endoscopic ultrasound, contrast-enhanced MRI, and ^99mTc-EDDA/HYNIC-TOC (TEKTROTYD) scan, including SPECT/CT, failed to detect an intrapancreatic insulinoma, so that both individuals subsequently underwent intra-arterial calcium stimulation tests. Case 1 featured a significant increase in insulin secretion (up to a five-fold increase over the baseline) primarily in the gastroduodenal and superior mesenteric arteries, which supply the pancreatic head and uncinate process (Fig. 1A and B). Case 2 exhibited a more substantial increase in insulin release (up to a 35-fold increase over the baseline) in the distal segment of the splenic artery (Fig. 1A and C), arguing for hyperresponsive beta cells in the left pancreas.

Figure 1.

Intra-arterial calcium stimulation tests. (A) Anatomic depiction of the arterial vessels supplying the pancreas, with simplified color-coding for the significant vessels tested with intra-arterial calcium stimulation. (B) Serum insulin levels after calcium stimulation in the corresponding vessels for case 1. (C) Serum insulin levels after calcium stimulation for case 2.

Further, thin-slice CT scans revealed no intrapancreatic masses, ⁶⁸Ga-NODAGA-Exendin-4-PET/CT scans were performed subsequently (Fig. 2A and B). These scans highlighted different localizations from the intra-arterial calcium stimulation test results. In case 1, high tracer uptake (SUV_max: 33.6) was noted in the pancreatic tail rather than the pancreatic head and uncinate process, which had been expected from the calcium stimulation test. In case 2, a bifocal tracer uptake was observed both in the pancreatic body (SUV_max: 7.5) and tail (SUV_max: 7.4).

Figure 2.

⁶⁸Ga-NODAGA-Exendin-4-PET/CT. (A) Red arrow indicates the pancreatic tail with high tracer uptake for case 1. (B) Red arrow indicates the bifocal tracer uptake in the pancreas body and pancreatic tail for case 2.

Treatment

To manage recurrent hypoglycemia and reduce the dependence on continuous intravenous glucose, diazoxide was administered to case 1. This led to a decrease in both the frequency and severity of hypoglycemic episodes. However, the patient developed increasing dyspnea due to fluid retention and pulmonary congestion, necessitating the discontinuation of diazoxide. Consequently, the patient was promptly referred for surgical intervention. Given the inconclusive results from localization tests, a laparoscopic exploration was undertaken. This resulted in a left-sided laparoscopic pancreatectomy, in line with the ⁶⁸Ga-NODAGA-Exendin-4-PET/CT findings. Post-surgery, histological analysis confirmed the successful removal of a 1.2 cm, moderately differentiated insulinoma located in the pancreatic tail, with positive insulin immunohistochemical staining.

The interdisciplinary tumor board recommended a resection of the pancreatic tail for case 2 unless otherwise suggested by intraoperative examination. This was based on the excessive stimulation of insulin release combined with one of the ⁶⁸Ga-NODAGA-Exendin-4 uptake areas in this region. A left-sided laparoscopic pancreatectomy was performed, and a complete resection of an insulinoma in the pancreatic tail was also confirmed in this case.

Course and follow-up

Since surgical tumor removal, both patients have not reported any documented hypoglycemia during the 6-month follow-up, indicative of complete remission.

Statement of approval by ethics committee

In this study, we utilized pseudonymized clinical data collected from patients admitted to the Department of Endocrinology and Diabetology at the University Hospital of Düsseldorf with suspected insulinoma. This study is exempt from review board approval. According to the North Rhine-Westphalia Health Data Protection Law § 6 (2), the following applies to data processing for scientific purposes: ‘Without the patient’s consent, the scientific personnel is allowed to use patient data for the purpose of scientific research, to which they already have access in the facilities or public institutions due to their activities according to § 2 (1)’.

Discussion

Although insulinomas are the most common neuroendocrine tumors of the pancreas, they remain a rare entity. Due to their diverse and non-typical symptoms (14, 15), diagnosing insulinomas is still challenging and often delayed.

For effective surgical planning, precise tumor localization following a successful biochemical diagnosis is crucial. This report details extensive localization efforts in two cases, highlighting a common challenge. The localization of insulinomas is challenging because most tumors are tiny, isoattenuating, or located deep in the pancreas. Diffuse beta-cell hyperactivity can also add to the complexity of localizing the tumor. There are also modality-specific technical weaknesses, for example, endoscopic ultrasound blind spots, low somatostatin receptor expression, and variable arterial anatomy.

The sensitivity of endoscopic ultrasound and MRI in these scenarios typically ranges from 54 to 86% (16, 17). The literature suggests that intra-arterial calcium stimulation is more sensitive than abdominal ultrasound, CT, or MRI for preoperative localization of insulinomas (7). This case series emphasizes the relevance of the increase in insulin secretion following calcium stimulation. While a two-fold increase of insulin levels should be indicative of pathologic beta-cell stimulation according to current recommendations (18), even a higher, four-fold increase did not align with the actual insulinoma location in our case 1. In contrast, a 35-fold increase accurately pinpointed the tumor’s location in case 2. More restrictive criteria, such as a four-fold increase with absolute insulin >150 μU/mL, and a dominant-territory logic when multiple arteries are positive, as referenced by previous studies (19, 20) could be considered to increase specificity in future studies. ⁶⁸Ga-NODAGA-Exendin-4-PET/CT was highly sensitive in both cases, yet it indicated bifocal uptake in the body and tail of the pancreas in case 2, contradicting surgical findings, which confirmed a single tumor in the pancreatic tail.

When conventional methods fail to locate an insulinoma, surgical exploration becomes necessary. In this instance, laparoscopic intraoperative localization accurately identified the tumors in both cases. Surgical exploration should therefore be utilized as a diagnostic strategy, especially when preoperative diagnostic tests show discrepancy. Previous studies also underscore the importance of surgical exploration, especially when combined with intraoperative ultrasound (21, 22).

Conclusion

The often varied and nonspecific symptoms associated with insulinoma can result in diagnostic delays. Accurate biochemical diagnosis in specialized endocrinology units is therefore essential. Precise preoperative tumor localization is key to ensuring successful resection with minimal patient risk. Despite advanced diagnostic techniques, pinpointing the tumor’s location can be challenging. In our case series, discrepancies between ⁶⁸Ga-NODAGA-Exendin-4-PET/CT and intra-arterial calcium stimulation test results were effectively resolved through surgical exploration, guiding appropriate surgical intervention. The additional findings from ⁶⁸Ga-NODAGA-Exendin-4-PET/CT allowed for easier localization of the tumor in the operation. In addition, our findings indicate the potential need to revise the threshold to a higher or tiered threshold for the intra-arterial calcium stimulation test to improve diagnostic accuracy.

Declaration of interest

The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the work reported.

Funding

This work did not receive any specific grant from any funding agency in the public, commercial, or not-for-profit sector.

Author contribution statement

SG and OPZ contributed equally, drafted the manuscript, and researched the data. FZ and GA performed and supervised radiological examinations. CK, UH, YM, EM, and FLG performed and supervised nuclear examinations. JP and IE performed and supervised histological analyses. AK and WTK performed and supervised the surgeries. ER, JMD, KBB, JS, and MR performed and supervised the clinical examinations and researched the data. RW is the PI of this study, contributed to the discussion, reviewed, and edited the article.

Guarantor statement

RW is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. All authors gave final approval of this version to be published.