diagnosis of insulinoma and tests

Diagnosis

1. Insulinoma is the most frequent hormone secreting tumor of the pancreas featuring an overproduction of insulin and / or proinsulin.

The mostly benign tumor is found twice as often in female patients as compared to male patients. The median of age at diagnosis is 50 years. Principally, the tumor may be found at any age of life.

2. The most important chemical finding are repetitively low blood glucose levels around or below 40 mg/dl (hypoglycemia), predominantly during fasting or physical activites.

3. Serum insulin concentrations are inadaequately high in relation to blood glucose levels due to absent or incomplete suppression of insulin secretion

the basic tool in endocrinological diagnostics of hypoglycemia is the

standardized supervised fasting test

For decades this test is established as the gold standard in the endocrinological diagnostic work-up of hypoglycemia.

Usually the test has to be performed on an in-patient basis requiring a specialized and trained hospital setting.

By means of the test adaequate physiological suppression of the endogenous insulin concentration into the range of basal concentrations of 3 - 5 µU/ml ( 18-30 pmol/l ) during fasting conditions is examined. This occurs regularly in every individual between daily meals, but predominantly during night rest.

OGTT and Fasting Test (diagnosis of insulinoma) :

Many endocrinologists omit the oral glucose load before a fasting test due to a misconception when both tests are being viewed at independently.

An oral glucose loading test performed before or after a fasting test on a different day is superfluos and not indicated. We discourage fasting tests being started sometimes after omission of food and intended "to wait for hypoglycemia".

The fasting test reflects a classical endocrinological suppression test in the situation of potential hyperinsulinemia.

Suppressibility of the ambient insulin concentration to normal levels ( adaequate insulin secretion in the normal situation or inadaequate secretion in insulinoma ) right after a glucose-induced stimulation with oral 75-100 g of glucose is being tested.

Thus, a standardized condition may be created allowing correct interpretation of the test already early into the test phase by the endocrinologist.

Fasting-Test / Sample-Diagram:

We start every fasting test in the morning after overnight fast by means of oral administration of 100 g of glucose = last defined meal at a known time point.

Onset of blood sampling as late as after a 12h-fasting period overnight with eventual blood glucose levels in the range of 40-60 mg/dl does not seem to be adaequate in the patient with an insulinoma about to be proven.

oral glucose load (100 g) immediately before fasting test (full size 46 Kb)

fasting test in insulinoma and exclusion of insulinoma (full size ca 50 Kb)

In patients with insulinoma (n = 120) a fasting test usually requires 15 hours (median) mean: 17 ± 10 h ; range 3 - 66 h

Other tests, e.g. proposed hyperinsulinemic euglycemic clamps - eventually in the presence of a hypoglycemic condition - are laborious and do not play a significant role in the diagnostics of insulinoma clinically.

Principally these are modified controlled C-peptide suppression tests.

The C-peptide suppression test (Saddig et al. JOP. J Pancreas 2002) has apredominant role in the diagnostic exclusion of insulinoma, when the disease seems to be highly unlikely.

Stimulation tests, such as the formerly described i.v.-tolbutamide test are obsolete and have been abandoned.
No advantage has been shown with the intraarterial calcium infusion (stimulation) test (SIPS) which is propagated mainly in the U.S. during invasive arterial angiography only to roughly regionalize an eventual tumor).
see also Insulinoma localisation

Interpretation of fasting test results:

Besides logistics the correct interpretation of fasting test results is of fundamental importance for a correct diagnosis (insulin - secretion) - misinterpretations often are the cause of a doubtful or even wrong diagnosis.

The reproducibility of typical insulin and blood glucose levels, especially before discontinuation or at the end of the fasting test should be of prime concern. This applies similarly to reliable estimations of absolute insulin concentrations.

! Exact mathematical values such as the insulin / glucose-ratio (I / G-ratio), are neither reliable nor too helpful in inspection of the data.

A limit of 0.25 ( µU/ml / mg/dl ) for the insulin / glucose ratio - as indicated in many textbooks and review articles - should not be regarded as a safe discrimination between healthy and insulinoma-bearing patients.

Most patients with an insulinoma present ratio values clearly > 2.00 during fasting.

A ratio of 0.25 in the presence of blood glucose levels of 40 mg/dl would consider insulin concentrations of 10 µU/ml as being mathematically normal.

At blood glucose levels of 30 mg/dl insulin concentrations of 7.5 µU/ml would equally be "normal". Both constellations, however, have to classified as being pathological.

Calculation of the I / G-ratio when the blood glucose level is within the normal range of > 50 mg/dl is redundant and without any diagnostic value.

We have seen many patients bearing an insulinoma in whom the correct diagnosis had been ruled out on the basis of a "normal" insulin / glucose ratio.

Patients with insulinoma may present an insulin/glucose ratio clearly below 0.2 during fasting.

Patients without Insulinoma may present blood glucose levels < 50 mg/dl during fasting, not so rarely even < 40 mg/dl, if fasted under supervision for more than 48 hours.

Source: data from > 150 fasting tests.

So far no data are available as to differentiate insulinoma from rare islet cell hyperplasia in adults (beta cell hyperplasia, focal hyperplasia, "nesidioblastosis") by means of a fasting test (so called "noninsulinoma pancreatogenous hyperinsulinemic hypoglycemia").

Sensitivity and specificity of the fasting test

Adaequately performed the standarized 72h-fasting test shows a diagnostic accuracy of close to 100 % according to sensitivity and specificity

End points:
Correct positive endocrinological diagnosis and successful surgical removal of an insulinoma

Sensitivity : Percentage of patients with an insulinoma and a positive (pathological) result druing the fasting test

Specificity: Percentage of healthy patients (without insulinoma) with a negative (normal) result in the fasting test

Sensitivity ( #1 + #2 ) and specificity ( #3 + #4) cannot be calculated by means of strict mathematics used in evidence-based medicine, due to

1. there is no clear definition of exact mathematical ranges for a pathological fasting test (e.g. insulin/glucose-ratio)

2. surgical detection depends upon surgical experience

3. none of the imaging techniques allows definitive exclusion of an insulinoma

4. clearly healthy patients according to the test results cannot be surgically explorated on ethical grounds.

Valid studies addressing follow-up of patients with an earlier negative result in the fasting test so far have not been done.

C - Peptide Suppressiontest