With combination of intraoperative ultrasound and palpation, more than 90-95% of all insulinomas will be found during exploration. But even in experienced hands some are not detected. When a familial multiple endocrine neoplasia type 1 (MEN-1) has been excluded, the question arises intraoperatively, whether we just failed to find an insulinoma or whether another illness is causing the disease. The latter may be a noninsulinoma pancreatic hypoglycemia caused by focal hyperplastic of islets or by neoplastic isleta with or without microadenomas (synonymous: “adult nesidioblastosis” or “insulinomatosis” by pathologists and “NIPHS” by clinicians).

Our own experience with 20 out of 125 patients with pancreatic hyperinsulinism is demonstrated, where we did not find an insulinoma intraoperatively. In some of them operative flaws led to this result and in the majority preoperative diagnosis of NIPHS was confirmed. Thus NIPHS accounts for 16% of all forms of sporadic pancreatic hyperinsulinism and has to be integrated into our daily preoperative work-up and intraoperative management.

The charts of 125 adult patients with documented endogenous hyperinsulinemic hypoglycemia were extracted, operated on between 1986 and 2008. All patients with benign or malignant solid insulinoma and all patients with familial MEN-1 were excluded, leaving 20 patients (xx%) with sporadic disease, in whom no insulinoma was detected during exploration. These were 4 men and 16 women with a mean age of 45.4 ± 14.8 years (range 18 to 76 years).

Eleven patients underwent operation for a presumed insulinoma, of which the final diagnoses were insulinoma in 4 (three at the head and one at the tail of pancreas) and NIPHS in 7, respectively. The other 9 patients underwent operation for preoperatively diagnosed NIPHS and were all proven to have NIPHS, postoperatively. Patients with insulinoma were all cured by removal of their tumors. The 11 patients with NIPHS were treated by a partial or subtotal pancreatectomy and none had reported further episodes of neuroglycopenia after 77 months of follow-up. Three of them developed postoperative insulin-dependent diabetes mellitus.

When a circumscribed insulinoma cannot be detected by preoperative localization studies in patients with proven endogenous hyperinsulinimic hypoglycemia the biochemical results of oral glucose tolerance test (OGTT) and 72 hours fast should be reassessed, carefully. When the data clearly point to an insulinoma, a thorough surgical examination undertaken, with special focus of attention on the head and uncinate process of the pancreas. If still no tumor is found, the operation should be terminated. When results of OGTT and 72 hours fast assume NIPHS a selective arterial calcium stimulation test is indicated, since the necessary 70-80% pancreatectomy can be guided by results of the stimulated insulin gradient. Following these principles all 20 patients were cured in a mean follow-up of 7.5 years. In 2 patients (10%), however, extensive partial pancreatectomy resulted in a mild insulin dependent diabetes mellitus.

Altogether sixteen out of 20 patients (%) with sporadic endogenous pancreatic hyperinsulinemia and failed demonstration of a single tumor proved to have NIPHS. Even when an insulinoma is suspected after biochemical analysis and preoperative localization study as well as intraoperative findings failed to show a tumor (n = 11) 7 patients (%) demonstrated to have NIPHS.