SUN-125 Role of Various Insulin Assays in the Diagnosis of Factitious Hypoglycemia

Abstract

Introduction Factitious hypoglycemia, caused by the surreptitious use of insulin, is an uncommon cause of hypoglycemia, but is similar in presentation to insulinoma and is often a major diagnostic challenge to the physicians. We present the case of a 48-year-old woman who presented with recurrent episodes of hypoglycemia, associated with an elevated plasma insulin level on one assay, but low on another. Factitious illness was denied by the patient until it was definitively proven by mass spectroscopy that the insulin circulating at the time of hypoglycemia was a synthetic insulin analog. Case Report A 48-year-old woman with no significant past medical history presented with recurrent episodes of light headedness, shakiness and sweating of 2 weeks duration that resolved with eating. Her blood pressure was 106/70, pulse 60/min, rest of the physical examination was unremarkable. Her fasting blood glucose was 40 mg/dL with insulin of 81.4 µIU/mL and C peptide 0.71 ng/mL, yet she denied exposure to insulin. Shortly after initiating a 72-hour fast she developed symptomatic hypoglycemia to 45 mg/dL. After 1 mg glucagon, glucose rose to 96 and 83 mg/dL at 10 and 20 minutes, respectively. Her insulin level by chemiluminescent assay was 5 uU/mL (<15 uU/mL fasting, Yale Labs), but 40.2 uU/mL by immunoassay (Quest). C peptide and proinsulin were <0.1 ng/mL, and < 5 pM, respectively. Because of discordant results we further measured free insulin and the beta chain of insulin which were 43 uU/mL (1.5-14.9 uU/mL) by radioimmunoassay, and 53.8 uU/mL (<13.7 uU/mL) by mass spectroscopy. An oral glucose tolerance test was performed when she presented with fasting hypoglycemia with measurement of C peptide, insulin and proinsulin along with mass spectroscopy to identify if an analog insulin was present vs. another substance accounting for discordant results between the two insulin assays. Mass spectroscopy identified the substance as insulin lispro which declined during the glucose tolerance test, while endogenous insulin, C peptide and proinsulin increased, confirming the diagnosis of surreptitious insulin use. Conclusion A high index of clinical suspicion is key to the diagnosis of factitious hypoglycemia as assays specific for human insulin may report low insulin levels when an analog is used. We were able to make a diagnosis of factitious hypoglycemia by measuring insulin using two assays, one specific for human insulin, and the other nonspecific, resulting in discordant findings suspicious for surreptitious use of insulin, which we then confirmed by mass spectroscopy to be insulin lispro. Therefore, in evaluating hypoglycemia, it is important to be aware of variability in the detection of insulin analogues by different assays and the opportunity for definitive identification by mass spectroscopy.