Successful Localization by Hyperselective Arterial Calcium Injection Test for Surgical Resection of Insulinoma: A Case Report
2024 Volume 46 Issue 3 Pages 263-269
Details
2024 Volume 46 Issue 3 Pages 263-269
Surgery is the main treatment for insulinoma, and precise preoperative localization is important to determine the extent of resection and to rule out multiple lesions. The selective arterial calcium injection (SACI) test is instrumental in the localization of insulinoma. Here we report a patient in whom the exact location of pancreatic insulinoma could not be determined by the conventional SACI test, and thus surgery was replaced with oral diazoxide. The hyperselective SACI test subsequently localized the lesion accurately, allowing surgical resection of the pancreatic body and tail while preserving the pancreatic head. We recommend the use of the hyperselective SACI test when the conventional SACI test fails to accurately determine the location of insulinoma lesions within the pancreas.
Insulinoma is an insulin-secreting pancreatic tumor derived from the beta cells of pancreatic Langerhans islets. The clinical features of insulinoma include those related to the effects of excessive insulin secretion on the autonomic and central nervous systems. The diagnosis of insulinoma may be delayed in patients with long history of recurrent hypoglycemia, due to the presence of nonspecific symptoms [1]. Insulinoma is often solitary and develops in the pancreas in 99% of cases. The tumor size is often as small as 2 cm or less [2]. Approximately 90% of insulinomas are benign and can often be radically cured by surgery [3]. Following the diagnosis of insulinoma, precise preoperative localization is important to determine the extent of surgical resection and to rule out the presence of multiple lesions. With regard to the detection of insulinoma, the selective arterial calcium injection (SACI) test is indispensable for the localization of insulinoma due to its high sensitivity (≥94%), which surpasses that of computed tomography (CT) and magnetic resonance imaging (MRI) [4, 5]. Here we report the failure of the conventional SACI test to localize insulinoma in a female patient, whereas the hyperselective SACI test determined the precise location of the lesion and subsequent surgery allowed the removal of the lesion and the preservation of the pancreatic head.
A 76-year-old woman was taken to a local hospital by ambulance for presyncope in year X-3. Despite a blood glucose level of 46 mg/dl, a blood insulin level was 9.0 μU/ml. Insulinoma was suspected, based on the lack of insulin secretion suppression. MRI showed a mass lesion measuring 11 mm in the pancreatic head and body. The patient was transferred to our hospital for detailed examination of hypoglycemia, including insulinoma. Her past medical history included acute appendicitis at the age of 38 years, together with total hysterectomy and adnexectomy for endometrial cancer at the age of 53 years. Her family history was negative for diabetes mellitus, endocrinopathies, and malignancies.
The patient’s physical findings included height of 161 cm, weight of 60.4 kg, body mass index (BMI) of 23.3 kg/m2, no thyromegaly, surgical scar in the mid-abdomen, and no tremors. Blood counts and biochemical test results were unremarkable (Table 1). An endocrinological survey showed low levels of glycated hemoglobin (HbA1c) at 4.8% and fasting plasma glucose at 53 mg/dl, but high levels of immunoreactive insulin (IRI) at 6.8 μU/ml and C peptide immunoreactivity (CPR) at 3.29 ng/ml. The anti-insulin antibody titer was within the reference range, and the baseline levels of the other endocrine-related hormones were normal. In the fasting test, hypoglycemia of 44 mg/dl was observed after 6 hours of fasting, and the blood glucose level increased by ≥25 mg/dl after intravenous injection of glucagon.
| Complete blood count | Biochemistry | Diabetes-related | |||
|---|---|---|---|---|---|
| Leukocytes (/mm3) | 6300 | TP (g/dl) | 6.4 | FPG (mg/dl) | 53 |
| Neutrophils (%) | 69.4 | Alb (g/dl) | 3.3 | PPG (mg/dl) | 97 |
| Eosinophils (%) | 2.2 | AST (U/l) | 18 | HbA1c (%) | 4.8 |
| Basophils (%) | 0.8 | ALT (U/l) | 16 | IRI (µU/ml) | 6.8 |
| Lymphocytes (%) | 23.2 | γ-GTP (U/l) | 14 | CPR (ng/ml) | 3.29 |
| Monocytes (%) | 4.4 | LDH (U/l) | 172 | IAA (U/ml) | <0.4 |
| Erythrocytes (×104/mm3) | 400 | ALP (U/l) | 16 | Endocrine-related | |
| Hemoglobin (g/dl) | 11.5 | CK (U/l) | 76 | ACTH (pg/ml) | 24.7 |
| Hematocrit (%) | 35.7 | LDL-C (mg/dl) | 91 | Cortisol (µg) | 8.3 |
| Platelets (×104/mm3) | 26.4 | TG (mg/dl) | 76 | TSH (µU/ml) | 3.12 |
| Urine | HDL-C (mg/dl) | 49 | FT4 (ng/ml) | 1.36 | |
| pH | 7.0 | BUN (mg/dl) | 13 | GH (ng/dl) | 0.32 |
| Glucose | (–) | Cr (mg/dl) | 0.79 | IGF-1 (ng/dl) | 141 |
| Protein | (–) | eGFR (ml/min) | 59.35 | ||
| Ketone | (–) | Na (mEq/ml) | 143 | ||
| Occult blood | (–) | K (mEq/ml) | 4.2 | ||
| Cl (mEq/ml) | 106 | ||||
| UA (mg/dl) | 3.0 | ||||
| Ca (mg/dl) | 9.9 | ||||
| iP (mg/dl) | 4.5 | ||||
*normal range
pH: hydrogen potential, TP: total protein, Alb: albumin, AST: aspartate aminotransferase, ALT: alanine transaminase, γ-GTP: γ-glutamyl transpeptidase, LDH: lactate dehydrogenase, ALP: alkaline phosphatase, CK: creatine kinase, LDL-C: low density lipoprotein cholesterol, TG: triglyceride, HDL-C: high density lipoprotein cholesterol, BUN: blood urea nitrogen, Cr: creatinine, eGFR: estimated glomerular filtration rate, Na: sodium, K: potassium, Cl: chloride, UA: Uric acid, Ca: calcium, iP: inorganic phosphorus, FPG: fasting plasma glucose, PPG: postprandial plasma glucose, HbA1c: glycated hemoglobin, IRI: immunoreactive insulin, CPR: C peptide immunoreactivity, IAA: anti-insulin antibody, ACTH: adrenocorticotropic hormone, TSH: thyroid-stimulating hormone, FT4: free thyroxine, GH: growth hormone, IGF-1: insulin-like growth factor I
These results indicated autonomous insulin secretion, consistent with features of insulinoma. The next step was the localization of the insulinoma. Contrast-enhanced CT identified a mass shadow in the pancreatic head and body (Figure 1). MRI showed a mass lesion at the site identified on the CT, with low signal intensity on T1-weighted images and high signal intensity on T2-weighted images (Figure 2). Contrast-enhanced CT and MRI showed a mass shadow in the pancreatic head and body. Endoscopic ultrasonography (EUS) also showed a small mass shadow in the pancreatic tail (Figure 3).
Based on the failure of the imaging studies to precisely localize the mass, we next used the SACI test. In this test, calcium gluconate hydrate (0.025 mEq/kg) was injected as a stimulant from the arteries supplying each region of the pancreas (gastroduodenal artery, proper hepatic artery, splenic artery, and superior mesenteric artery). Several blood samples were collected from a catheter placed in the hepatic vein (at 0, 30, 60, 90, and 120 seconds) to measure insulin (IRI) levels. Insulin secretion was clearly enhanced in the splenic artery, suggesting the presence of insulinoma in the pancreatic body-tail (Figure 4), but this site was inconsistent with the site of the mass shadow identified on the CT and MRI. While the test added confirmation to the presence of the insulinoma, it failed to precisely localize the mass, which is an important aspect of pre-surgical assessment.
Based on the above limitations, we decided to apply drug therapy instead of surgical therapy, with oral administration of diazoxide, with the aim of preventing insulin release from the pancreas and helping to return the blood glucose to normal levels. The dose of diazoxide was gradually increased up to 200 mg, resulting in a decrease in the frequency and level of hypoglycemia. However, pedal edema was noted to worsen gradually, with an increase in body weight by about 10 kg within approximately 1 year, as well as a gradual increase in the frequency of hypoglycemic episodes. Thus we decided to reconsider surgical treatment.
In order to get a more precise localization of the lesion preoperatively, we performed the hyperselective SACI test for the splenic artery region since the previous test showed increased insulin secretion in the splenic artery. Blood samples were collected from the dorsal pancreatic artery, which is located proximal to the splenic artery, and calcium was injected for stimulation at the following five sites: the peripheral splenic artery, dorsal pancreatic artery, superior mesenteric artery, proper hepatic artery, and gastroduodenal artery (Figure 5). Insulin secretion increased markedly in these arteries after stimulation with calcium injection. In addition, angiography of the dorsal pancreatic artery showed a densely stained area consistent with the mass being present in the pancreatic body (Figure 6). The lesion in the pancreatic body appeared to be the culprit lesion, and the insulinoma was considered to be located in the pancreatic body. Laparoscopic spleen-preserving resection of the pancreatic body and tail was performed at the Department of Surgery of our hospital (neuroendocrine tumor, G2 such as insulinoma, T1N0M0, Stage I). Hypoglycemia was resolved after the surgery and was not observed even after discontinuation of diazoxide.
We experienced a case of insulinoma that could be localized precisely in the pancreatic body by the hyperselective SACI test, but not by the conventional SACI test. Although many reports have described the usefulness of the SACI test for the localization of insulinoma [4, 5], there are no reports on the usefulness of the hyperselective SACI test.
Insulinoma is a relatively rare condition with an incidence rate of new cases of 10,000 to 40,000 persons/year [6], but it is the most common pancreatic neuroendocrine tumor (pNET), accounting for 20.9% [7]. Its symptoms vary widely, from central nervous system symptoms caused by hypoglycemia induced by excessive insulin secretion (e.g., headache, dizziness, impaired consciousness, convulsion, and coma) to autonomic clinical features of hunger, sweating, tremor, and malaise [8]. The threshold of blood glucose for developing hypoglycemic symptoms varies greatly among individual patients, and symptoms are also often nonspecific; for these reasons, delayed diagnosis is not uncommon. Persistence of hypoglycemic symptoms over a long period of time often leads to weight gain due the suppressive effect of eating on clinical symptoms.
In our patient, even after the diagnosis of insulinoma, we could not identify the exact location of the mass within the pancreas by the conventional SACI test, resulting in switching the clinical management to diazoxide. Although this reduced the frequency of hypoglycemia, that effect was only temporary. Our patient gained weight (approximately 10 kg in 1 year). Although hypoglycemic symptoms can be controlled by oral administration of diazoxide in approximately 50–60% of patients with insulinoma [9], it is necessary to reexamine whether surgery is indicated in patients like ours who still develop hypoglycemic symptoms after initiating diazoxide therapy.
Surgical resection is the treatment of choice for insulinoma, and surgical removal of the tumor resolves hypoglycemic symptoms in most patients [10]. In our case, the hypoglycemic symptoms were completely resolved after surgery. Localization of the insulinoma is very important in preoperative evaluation. The sensitivity of CT and MRI for localization of insulinoma is only 80% [11] and 70% [12], respectively. Conversely, the sensitivity of EUS and the SACI test for localizing insulinoma is higher, at 80–95% [13] and 96% [14], respectively. The latter procedures are considered very useful for localization. In our patient, although CT and MRI identified the lesion in the pancreatic head and body, the results of EUS indicated it was in the pancreatic tail. The conventional SACI test showed enhanced insulin secretion in the splenic artery, suggesting the presence of insulinoma in the pancreatic body-tail. Precise localization could not be achieved because the site suggested by this test was inconsistent with those by CT and MRI.
The hyperselective SACI test is comparable to the SACI test in technical difficulty, with no increase in risk. In our patient, the surgical resection option was abandoned following the failure of the SACI test to precisely localize the lesion. We believe that when the SACI test results on tumor location are inconclusive, as in our case, the hyperselective SACI test should be used for detailed evaluation of the lesion.
We reported a patient with insulinoma in whom the hyperselective SACI test helped in precisely localizing the lesion. Such localization played a part in clinical decision making, allowing surgical removal of the body and tail of the pancreas while preserving the pancreatic head. Although the SACI test is instrumental in localizing insulinoma, a more selective SACI test is useful for determining the extent of surgical resection for insulinoma that is difficult to localize by imaging studies.
We express our sincere appreciation to Dr. Satoshi Igawa (Department of Radiology, University of Occupational and Environmental Health, Japan) for the inspiring guidance and encouragement in the preparation of this case report.
The authors declare no conflict of interest.
The data that support the findings of this study are available on request from the corresponding author. The data are not publicly available due to their containing information that could compromise the privacy of research participants.
