Pediatric Zollinger-Ellison Syndrome Workup

Laboratory studies to confirm the diagnosis of Zollinger-Ellison syndrome (ZES) measurements of the fasting serum and gastrin levels, secretin and calcium stimulation tests, and measurements of the basal acid output.

If the patient has multiple endocrine neoplasia type 1 (MEN-1), other laboratory abnormalities may be suggestive of Zollinger-Ellison syndrome, such as the following:

• High plasma calcium levels
• High parathyroid hormone (PTH) levels
• High prolactin (PL) levels

Conventional radiologic studies (computed tomography [CT] scanning, transabdominal ultrasonography, and magnetic resonance imaging [MRI]) may not be helpful owing to the small size of most gastrinomas (< 2 cm). Nuclear imaging with somatostatin receptor scintigraphy (SRS) (octreotide scanning), however, may be useful in indentifying these tumors.

Operative techniques such as palpation, duodenal transillumination, and intraoperative ultrasonography can be used during laparotomy for the 20% of gastrinomas that somatostatin receptor scintigraphy (SRS) and other imaging studies fail to visualize.

Because patients with MEN-1 and Zollinger-Ellison syndrome are at risk for developing enterochromaffin-like (ECL) cell tumors, they require regular gastroscopy with multiple biopsies of all parts of the stomach and systematic biopsies of all mucosal lesions to detect these carcinoids.

Serum gastrin levels higher than 100 pg/mL are highly suggestive of Zollinger-Ellison syndrome (ZES). If the gastric pH level is less than 2, a gastrin level greater than 1000 pg/mL is diagnostic of this disease.^{[14, 15]}

However, if the patient is not receiving acid-suppressing medication, and the gastric pH levels are higher than 2, Zollinger-Ellison syndrome can be ruled out.

If the gastrin level is in the range of 100-1000 pg/mL, and the pH level is less than 2, a secretin stimulation test must be performed.

Secretin stimulation test

After obtaining blood to measure the basal gastrin level, intravenously (IV) administer 2 IU/kg of secretin. Then, obtain blood at 2.5, 5, 10, 15, and 30 minutes. The increase of the serum gastrin levels greater than 200 pg/mL is diagnostic of Zollinger-Ellison syndrome.

The physiologic mechanism of the secretin test remains unclear; however, it is the most important diagnostic test to exclude other conditions with increased acid secretion, hypergastrinemia, or both.

Clinical conditions in which patients present with hypergastrinemia, such as gastric outlet obstruction, pernicious anemia, renal failure, and achlorhydria due to atrophic gastritis, must be excluded with secretin provocative testing.

Calcium stimulation test

If Zollinger-Ellison syndrome is strongly suspected and the secretin test result is negative, the test can either be repeated or a calcium stimulation test can be performed. Although this test is less commonly used, less sensitive, and has a greater side effect profile with IV calcium infusion, the calcium stimulation test is considered highly specific for gastrinomas.^[7]

Before measurement of the basal acid output (BAO), acid-inhibitory agents must be discontinued, as follows: 2 days for H2-receptor antagonists and 7 days for proton pump inhibitors (PPIs).

In the 24 hours before the test, the patient receives antacids.

A nasogastric tube (NGT) is placed into the antrum, and the stomach is emptied.

Four consecutive samples of gastric fluid are collected; a quadruplicate of each sample is titrated to pH 7 with 0.2 N sodium hydroxide, whereas the basal acid output is determined with a radiometer titrator.

In an unoperated stomach, a basal acid output of more than 15 mEq/h is diagnostic of Zollinger-Ellison syndrome (ZES). If the patient underwent gastric resection for acid reduction, a basal acid output of more than 10 mEq/h is diagnostic for this condition.

Because most gastrinomas are smaller than 2 cm, visualizing them with conventional imaging techniques, such as computed tomography (CT) scanning, transabdominal ultrasonography, and magnetic resonance imaging (MRI), is difficult.

Somatostatin receptor scintigraphy (SRS), or octreotide scanning, can reveal 57-78% of gastrinomas and has a sensitivity of 84-94%.^[16] This technique is currently the single most effective imaging modality for gastrinomas, but it is highly dependent on the tumor size. SRS may not accurately reveal tumor size or location and is best used in conjunction with CT scanning with intravenous (IV) contrast. SRS is also useful, because it allows for whole-body scanning and measurement of whole-body tumor content during a single test. See the following images.

Computed tomography (CT) scanning with intravenous (IV) contrast and magnetic resonance imaging (MRI) are highly specific for gastrinoma, with reports of 83-100% specificity using MRI. However, sensitivities range from 20% to 59%.

Transabdominal ultrasonography has a sensitivity of 0-28%, but this imaging modality may be useful in screening for metastatic disease of the liver, with reported sensitivities of 14-63%. Specificity ranges from 92-100%.

Angiography has been used with limited success because of difficulties in discriminating between the relative vascularity of the gastrinoma lesion and the surrounding tissue. This technique has a sensitivity of 28-68%.

In one study, intra-arterial secretin stimulation with hepatic venous sampling yielded a sensitivity rate of 89%; however, because of its invasive nature, angiography is only used when other imaging techniques are ineffective.

Endoscopy may be used to evaluate recurrent abdominal pain with or without gastrointestinal (GI) bleeding. Regular flexible endoscopy may reveal prominent gastric rugal folds. Ulcers typical of Zollinger-Ellison syndrome (ZES) and histologic findings may help to confirm the diagnosis. Endoscopy is also indicated to stage peptic ulcer disease (PUD). If findings are consistent with Zollinger-Ellison syndrome, immediate treatment with proton pump inhibitors (PPIs) and further diagnostic studies are indicated to localize the tumor.

Endoscopic ultrasonography (EUS) may reveal structures as small as 2 mm, mainly in the pancreas.^[16] Reported sensitivities are 58-100%. One study found a sensitivity of 93% and a specificity of 95% for pancreatic lesions. Specificity is reported to be 84-100%. Unfortunately, EUS is limited in its ability to reveal duodenal gastrinomas and can fail to visualize up to 50% of them. The successful and safe use of EUS in pediatrics, which requires special equipment, has been described in children aged 4-16 years.

Portal venous sampling has been described in adults with Zollinger-Ellison syndrome but not in children, with a positive yield of 46-90%; however, it is associated with a complication rate of 10% and is recommended only as a last resort.

Because patients with multiple endocrine neoplasia type 1 (MEN-1) and Zollinger-Ellison syndrome (ZES) are at risk for developing enterochromaffin-like (ECL) cell tumors, they require regular gastroscopy with multiple biopsies of all parts of the stomach and systematic biopsies of all mucosal lesions to detect these carcinoids.^[17] Both the clinical and laboratory factors and the biopsy results can be used to identify a subgroup of patients that need to be followed more closely, because they have a higher risk for developing carcinoids.

Neuroendocrine tumors (NETs), including gastrinomas, are composed of homogenous sheets of cells with small compact nuclei and prominent nucleoli. The tumors can be glandular or trabecular. Gastrin-producing cells are often well differentiated.

Immunohistochemistry staining may be positive for chromogranin A, neuron-specific enolase, and synaptophysin, as well as for pancreatic peptide, somatostatin, adrenocorticotropic hormone (ACTH), and vasoactive intestinal polypeptide (VIP).

In a series of 57 patients with multiple endocrine neoplasia type 1 (MEN-1) and Zollinger-Ellison syndrome (ZES), no patient had consistently normal gastric biopsy findings, and 47% had diffuse hyperplasia as the most advanced gastric enterochromaffin-like (ECL) cell changes, 25% had linear hyperplasia, 3.5% had micronodular hyperplasia, and 1.8% had dysplasia. In 23% of the patients, (average age at biopsy, 47 y) a carcinoid was found.