E-Note for Adult
Stat | Lytes | Drugs | ID | Heart | Lungs | Kidneys | GI | Rheum | Heme-Onc | Endo | Neuro | Derm | Misc. | Resources | Home
The New England Journal of Medicine -- August 26, 1999 -- Vol. 341, No. 9 (Editorial)
Surgery for the Zollinger-Ellison Syndrome
In 1955, Zollinger and Ellison described the clinical syndrome of virulent peptic ulcer disease associated with marked gastric hyperacidity and pancreatic islet-cell tumors. (1) A year earlier, Wermer had reported a familial disease characterized by hyperparathyroidism, pituitary tumors, and pancreatic islet-cell tumors: the disease is now known as multiple endocrine neoplasia type 1 (MEN-1). (2) Since these two landmark discoveries in endocrine oncology, we have learned a great deal about the pathophysiology, clinical presentation, diagnosis, and treatment of the Zollinger-Ellison syndrome.
The most important advance was the discovery that gastrin, a polypeptide hormone and potent acid secretagogue, is the cause of the severe peptic ulcer disease in patients with the Zollinger-Ellison syndrome. (3) Subsequently, a series of remarkable drugs that inhibited gastric acid secretion were developed, including histamine H2-receptor antagonists and, more recently, H+/K+-ATPase inhibitors; these drugs proved highly effective in controlling the ulcerative diathesis. As a result, surgery for peptic ulcers, the complications of which cause death in approximately half the patients with the Zollinger-Ellison syndrome, became unnecessary, and attention was focused on the primary treatment of the gastrin-secreting islet-cell tumors, or gastrinomas. It then became known that gastrinomas originate not only in the pancreas but also in the duodenal submucosa, that they rarely originate in lymph nodes or other anatomical sites, and that most are malignant. (4)
What are the important considerations in the surgical treatment of patients with the Zollinger-Ellison syndrome? Do they apply equally to patients who have sporadic gastrinomas and patients who have gastrinomas and associated MEN-1? In this issue of the Journal, Norton et al. (5) report that the prognosis for patients with sporadic gastrinomas is similar to that for patients with gastrinomas associated with MEN-1.
In terms of death due to manifestations of the Zollinger-Ellison syndrome, the 10-year survival rates among patients with sporadic gastrinomas and patients with gastrinomas associated with MEN-1 were 95 percent and 86 percent, respectively. (5) In another study from the same clinic, in which 212 consecutive patients with the Zollinger-Ellison syndrome were followed for a mean of 13.8 years, 31 percent died, half from causes related to the Zollinger-Ellison syndrome or malignant gastrinomas. (6) Important prognostic factors in patients who died of the Zollinger-Ellison syndrome, as compared with those who died from causes unrelated to the syndrome, were a primary tumor in the pancreas; a tumor 3 cm or more in diameter; metastases to lymph nodes, liver, or bone; and the presence of the ectopic corticotropin syndrome. (6) Accurate indexes of the growth of malignant gastrinomas would be useful in planning treatment, but there is no reliable method for obtaining useful prognostic information when the Zollinger-Ellison syndrome is first diagnosed.
The treatment of patients with the Zollinger-Ellison syndrome who have sporadic gastrinomas is relatively clear. Virtually all are symptomatic at the time of presentation, and the biochemical diagnosis is straightforward. The initial step is to control the acid hypersecretion by the administration of appropriate drugs and then to localize the tumor. In the present study, somatostatin-receptor scintigraphy was the most useful imaging procedure; it detected gastrinoma tissue in 80 percent of the patients with sporadic gastrinomas in whom it was performed. This procedure has high sensitivity, and it detects primary or metastatic gastrinoma tissue at anatomical sites other than the pancreas or duodenum. At surgical exploration, care must be taken to resect the primary tumor, to perform a duodenotomy and resect identified tumor foci, to resect all enlarged lymph nodes and normal-appearing lymph nodes in the peripancreatic region, and to explore the liver and remainder of the abdomen for evidence of metastatic disease. In the current study, 51 percent of the patients with sporadic gastrinomas were free of disease immediately after surgery, and the majority of these patients remained so 10 years later. In selected patients who have persistent or recurrent gastrinomas postoperatively, there is a proven benefit in resecting localized hepatic metastases.
Patients with MEN-1 would seem to have a therapeutic advantage, because screening for a gastrinoma or other pancreatic tumor can be undertaken when they are found to have primary hyperparathyroidism or a pituitary adenoma or when they undergo direct DNA testing as members of a kindred with MEN-1. (7) However, it does not necessarily follow that screening (for example, by periodic measurements of serum gastrin) is useful, unless one can offer early therapy that is beneficial to the patient. A prophylactic total or subtotal pancreatectomy in patients with no clinical or biochemical evidence of pancreatic endocrine disease is unwarranted. Aggressive pancreatic resection in asymptomatic patients who have high serum concentrations of gastrin or other islet-cell hormones has been advocated, (8) but the value of this approach is unknown.
Because disease in the pancreas can be multicentric and because of the high frequency of regional lymph-node metastases in patients with duodenal gastrinomas, the likelihood that patients with MEN-1 and the Zollinger-Ellison syndrome will be biochemically free of disease after surgery is small. However, survival may be prolonged if the primary gastrinoma is resected before it spreads to the liver. Furthermore, after surgery, patients could be prospectively evaluated for new primary tumors or metastatic disease and could undergo one or more additional operations, if indicated.
In the present study, the authors followed a policy of operating on patients with MEN-1 and the Zollinger-Ellison syndrome when the tumor was 3 cm or more in diameter. One would expect, from their data and those of others, that the size of the tumor is directly related to the stage of disease and that, as with other solid malignant tumors, the rate of cure is higher if the tumor is resected when small. Given the current low rates of morbidity and mortality associated with pancreatoduodenectomy and lesser pancreatic resections, it would appear that early intervention is warranted in patients with MEN-1 and the Zollinger-Ellison syndrome. But only a clinical trial can answer the question of whether the cure rate will be higher with this approach than with the approach, advocated by Norton and colleagues, of not operating until the tumor is larger.
Gastrinomas are uncommon but fascinating tumors that can cause devastating peptic ulcer disease. They can be small, located in many different places in the upper abdomen, and hard to find. The extensive experience of Norton et al. has made it clear that despite the small size of the tumors, they can often be localized preoperatively, allowing effective surgical treatment.
Samuel A. Wells, Jr., M.D. American College of Surgeons - Chicago, IL 60611