Congenital Hyperinsulinism Workup

A number of laboratory studies may be indicated in patients with congenital hyperinsulinism (CHI), also known as persistent hyperinsulinemic hypoglycemia of infancy (PHHI).

Serum glucose, ketone, and insulin levels should be obtained while the patient is hypoglycemic (serum glucose level < 60 mg/dL). The definition of hypoglycemia in neonates is dependent on gestational age, and the threshold may be lower than described here.

The finding of nonketotic hypoglycemia in association with elevated insulin levels (>10 µU/mL) and normal levels of free fatty acid (FFA) supports the diagnosis of hyperinsulinism. The insulin-to-glucose ratio may range from 0.4-2.7 (normal, < 0.3). Sustained glucose use rates in excess of 10 mg/kg/min (evidenced by the need to administer intravenous [IV] glucose at a rate higher than 10 mg/kg/min to maintain normoglycemia) are consistent with exaggerated insulin activity and suggestive of CHI.

Cortisol and growth hormone levels are usually elevated in specimens taken during an episode of hypoglycemia (as an appropriate and normal response to hypoglycemia) and are usually within the reference range during periods of normoglycemia.

Serum metabolic screens, pH, lactate, and ammonia studies may be obtained to exclude other metabolic diseases. The results are expected to be within the reference range in cases of CHI. Urinary ketone, amino acid, and reducing-substance studies may be obtained to exclude other metabolic diseases. The results of these are also expected to be within the reference range in cases of CHI.

Hyperinsulinism with hyperammonemia and elevated levels of FFA suggests a fatty acid oxidation disorder. Hyperinsulinism with hyperammonemia and normal levels of FFA suggest the diagnosis of hyperinsulinism with hyperammonemia, a clinically and genetically distinct variant of CHI. Patients with this disorder usually respond very well to medical therapy alone and are much less likely to require surgical intervention. The hyperammonemia is mild and not symptomatic.

Ultrasonography, computed tomography (CT), and magnetic resonance imaging (MRI) have been used to search for a focal mass in the pancreas; however, in many cases, the lesion is too small to be visible by such techniques. These forms of imaging cannot identify the diffuse form of CHI.

A newer imaging technique has been developed that uses positron emission tomography (PET) in conjunction with coregistered abdominal CT (see the image below) to distinguish between focal and diffuse disease and, in the case of focal disease, localize the lesions.^[7] This technique uses a novel isotope, fluorine-18 L-3,4-dihydroxyphenylalanine (¹⁸ F-L-DOPA), for which neuroendocrine cells have a high affinity.

A study at Children’s Hospital of Philadelphia found that the technique was 96% accurate for diagnosing focal or diffuse disease and 100% accurate in localizing the focal lesion.^[8]

The¹⁸ F-L-DOPA isotope remains investigational and is technically complex to prepare. It is currently available at only a few centers worldwide. However, because of the remarkable results seen in preliminary published trials, physicians treating patients with CHI should strongly consider consultation with an expert at one of these centers.

Catheterization of the portal and pancreatic veins with venous sampling may help distinguish between focal and diffuse CHI. This procedure is well described in the pediatric population.

For venous sampling, a catheter is placed in the pancreatic venous system via a femoral vein or inserted by direct hepatic puncture to enter the portal vein. With the use of fluoroscopic guidance and IV contrast agents, the catheter is advanced into various pancreatic veins, and blood samples are taken to measure glucose, insulin, and C-peptide levels.

If a focal lesion is present, elevated insulin levels are expected in veins draining the area near the lesion, and insulin levels are expected to be within the reference range for CHI in other areas. If a diffuse lesion is present, insulin levels are expected to be high throughout the pancreatic venous bed.

In some cases, the results of this study are difficult to interpret, and correlation of results with pathologic findings remains imperfect. Pathologic examination remains the criterion standard for identification of focal disease. However, pancreatic venous sampling is one of few preoperative techniques available to identify focal lesions in patients in whom findings on conventional imaging are inconclusive.

Pancreatic venous sampling and intraoperative histologic studies should be strongly considered, because the identification of a focal lesion has profound implications for treatment and prognosis.

A test using intra-arterial calcium stimulation has been employed in adults and, to a lesser extent, in children. In this test, a bolus of calcium gluconate is rapidly administered via a catheter in the celiac axis and the splenic, superior mesenteric, and gastroduodenal arteries. Blood samples are obtained through a catheter in the right hepatic vein before injection and at several intervals after injection. These blood samples are then tested for glucose, calcium, and insulin levels.

An excessive insulin response from calcium stimulation in a single artery suggests a focal lesion, and excessive poststimulation insulin secretion associated with all arteries suggests a diffuse form of hyperinsulinism.

Although pancreatic venous sampling has been studied more widely to date, especially in neonates, experience with intra-arterial calcium stimulation in children is increasing. Children’s Hospital of Philadelphia has reported on a large number of cases.

The histology of CHI has been divided into focal and diffuse categories. In the focal form (accounting for one fourth to one half of cases), the focal lesion contains isletlike cell clusters with ductoinsular complexes, hypertrophic cells, and giant nuclei. A well-developed endoplasmic reticulum and prominent Golgi complex are present, suggesting a high level of protein synthetic activity. Immunohistochemical staining shows an increased proportion of insulin-containing cells.

The focal lesion may occur in any part of the pancreas, although the tail and body are the most common locations. The focal lesion is commonly too small to be identified on imaging studies or palpated during surgery. Outside of the area of the focal lesion, the pancreas appears normal. Most patients with the focal form of PHHI have a solitary lesion; however, approximately one fourth of cases are multifocal (ie, contain 2 or more focal lesions).

In the diffuse form of CHI, findings throughout the pancreas are similar to those found within a focal lesion. Again, isletlike cell clusters with ductoinsular complexes, hypertrophic cells, and enlarged, hyperchromatic nuclei are observed; endocrine cells also occur individually (see the images below). The endoplasmic reticulum is well developed, and Golgi complexes are prominent. Results of macroscopic examination are normal.

These histologic findings have also been observed in infants and older children with no known abnormalities of glucose homeostasis.

Some authors suggest that this microscopic appearance may be part of a normal developmental process and that other functional abnormalities may exist in the patient with CHI. Persistent hyperinsulinism, then, may represent a derangement of the developmental process or the extreme end of a spectrum of endocrine cell function. Other authors suggest that these histologic findings may be associated with infants of diabetic mothers or stressed, growth-retarded premature infants.