There are no clear criteria for the diagnosis of RYGB-related hyperinsulinemic hypoglycemia as the recognition of this syndrome has been evolving over the past 10 years. Here, we detail a reasonable approach to identify these patients and a diagnostic algorithm to clarify the diagnosis and rule out other causes of hypoglycemia.

The first step is recognition of the symptoms in a patient who has had RYGB. Typically, the episodes of hypoglycemia are a late complication, occurring 1–4 years after the initial surgery. As discussed later in this chapter, the symptoms can overlap with those of “dumping syndrome.” A careful history should be performed, including what symptoms are occurring, whether they coincide with a low blood sugar, what time of day they occur, whether they occur when fasting or postprandial, what foods trigger symptoms, and whether symptoms resolve with food intake. Typically, patients with RYGB-related hypoglycemia will describe symptoms beginning 1–2 h after a meal with minimal fasting symptoms. High carbohydrate intake is often a trigger for symptoms. Symptoms should improve within 15 min of food intake, but then may recur again an hour later. After many recurrent episodes of hypoglycemia, the symptoms may lessen as the patient develops hypoglycemia unawareness.

In order to diagnose hypoglycemia of any cause, it is important to identify Whipple’s triad. Whipple’s triad includes (1) the presence of classic hypoglycemia symptoms, (2) a low plasma glucose (not capillary glucometer reading) at the time of symptoms, and (3) resolution of symptoms with food intake. Therefore, laboratory testing must begin with documentation of low plasma glucose during a symptomatic episode, typically less than 55 mg/dL. Although patients may be provided with a glucometer to test capillary glucose during symptomatic episodes, these readings should not be considered diagnostic. In the case of RYGB-related hypoglycemia, symptoms are typically postprandial, so it makes sense to provoke symptoms in order to document hypoglycemia in a controlled setting. Although mixed meal tests are sometimes used, we typically perform an oral glucose tolerance test (OGTT) to assist in making the diagnosis (Fig. 21.1). The OGTT provides a higher carbohydrate load and is more likely to induce hypoglycemia and related symptoms. This provocative test can be performed on an outpatient basis and consists of administration of 50–100 g of oral glucose after a 12 h fast. The serum glucose, insulin, and C-peptide are measured at the start of the test and every 30 min thereafter for 2–3 h after administration. A typical pattern in RYGB-related hypoglycemia is a rapid rise in glucose, insulin, and C-peptide in the first 30 min, followed by a rapid decline. Although the insulin and C-peptide levels often remain elevated or inappropriately normal at the time of the hypoglycemia, they should demonstrate a rapid decrease. Persistent hyperinsulinism is not consistent with reactive hypoglycemia. Fasting hyperinsulinemia with fasting hypoglycemia would also not be expected, and would prompt consideration of alternate causes of hypoglycemia, such as insulinoma. Importantly, fasting hyperinsulinemia with normal or elevated fasting glucose is indicative of insulin resistance or diabetes, not insulinoma. A failure to see a coincident elevation in C-peptide levels would raise concern that the hypoglycemia was induced by exogenously administered insulin. Consultation with an endocrinologist is recommended if an alternate cause is considered [42]. The OGTT test should be considered confirmatory for RYGB-related hyperinsulinemic hypoglycemia if glucose levels are less than 55 mg/dL and the patient developed symptoms during testing consistent with their described ambulatory symptoms. Importantly, it should be noted that the OGTT will induce hypoglycemia in up to 12.5 % of control patients and up to 72 % of RYGB patients [6, 43]. Therefore, the lab results should be interpreted with consideration of the clinical presentation of the individual patient.

Several algorithms for the diagnosis of hypoglycemia involve additional testing to rule out other causes, including insulinoma. It is reasonable to exclude alternate causes of hypoglycemia in RYGB patients when the testing is relatively straightforward and noninvasive. A TSH will screen for hyperthyroidism and an early morning cortisol will screen for adrenal insufficiency. Liver and kidney function should be examined to rule out contributions from severe renal insufficiency or liver disease. However, exclusion of insulinoma is a much more involved process. Insulinoma should be considered if fasting hypoglycemia and hyperinsulinemia are present or if symptoms are not clearly postprandial in nature. However, careful consideration of the expense and the invasiveness of a complete insulinoma rule out should take place if the patient has classic RYGB-related postprandial hyperinsulinemic hypoglycemia symptoms and testing.

Ruling out an insulinoma involves performing a 72-h diagnostic fast. In hyperinsulinemic hypoglycemia, this test should be negative as the hypoglycemia in these patients follows the consumption of food. However, if hypoglycemia and symptoms are not induced with OGTT or a shorter 12 h fast, then a prolonged fast would be indicated. This testing requires inpatient admission and careful coordination to manage the needs for frequent blood draws. The serum glucose, insulin, C-peptide, and proinsulin are measured every 6 h or every 2 h if the glucose level drops below 60 mg/dL. The test is stopped if the patient’s glucose levels drop below 45 mg/dL with the development of neuroglycopenic symptoms. The patient is then tested for serum insulin, C-peptide, proinsulin, beta-hydroxy-butyrate, and sulfonylurea levels. One mg of glucagon is administered and the serum glucose levels are checked at 10, 20 and 30 min [44].

A computed tomography scan of the abdomen and pelvis should be obtained to evaluate for masses in the pancreas if fasting or persistent hyperinsulinemia with hypoglycemia is seen. Some surgeons will also evaluate the pancreas with endoscopic ultrasound. If there is high suspicion, calcium-stimulated arterial angiography can be used to identify more precisely the location of an insulin-secreting pancreatic lesion.