Searchable abstracts of presentations at key conferences in endocrinology
Endocrine Abstracts (2021) 73 OC9.5 | DOI: 10.1530/endoabs.73.OC9.5


1Cochin Hospital, Hormonology, France; 2Cochin Hospital, Endocrinology, France; 3Cochin Hospital, Anatomopathology, France


Introduction

Fasting trial remains the gold standard to confirm the diagnosis of insulinoma, based on low blood glucose level concomitant with inadequate high insulin level. However, diagnostic criteria are not consensual. Glycemia and insulin thresholds differ between the different consensus statements : endogenous hyperinsulism diagnosis relies on a glycemia < 3 mmol/l associated with an insulin level > 3 mUI/l in Endocrine Society guidelines (2009) whereas in NANETS 2010 and ENETS 2012 consensus guidelines, diagnosis is based on a glycemia < 2, 2 mmol/l concomitant with an insulin level > 6 mUI/l or > 3 mUI/l, depending on the assays. It has been admitted that the previously used diagnostic insulin level cut-off is challenged by the more specific and sensitive currently routinely used assays.

Methods

We retrospectively analyzed fasting trials results from the 124 patients explored in the Endocrinology department of Cochin Hospital from February 2012 to July 2020 to reevaluate insulinoma diagnosis criteria in terms of glycemia, insulin and C-peptide levels. Insulin and C-peptide levels were determined by automated chemiluminescent sandwich immunoassays.

Results

On the 124 patients included, 51 presented an hypoglycemia < 2.5 mmol/l during fasting trial and 19 were diagnosed with insulinoma, histologically proven. In all insulinoma cases, glycemia nadir was < 2.5 mmol/l (min =1.3 mmol/l; max =2.3 mmol/l; median = 1.7 mmol/l) concomitant with an insulin level ranging from 2.6 to 51.7 mUI/l (median = 10.3 mUI/l) and a C-peptide level ranging from 0.22 to 1.58 nmol/l (median = 0.6 nmol/l). The only insulinoma case presenting an insulinemia < 3 mUI/l at the time of hypoglycemia, harbored a concomitant C-peptide level at 1.21 nmol/l. Such a discordance between insulin and C-peptide levels could only be explained by an hemolyzed blood sample, insulin being very sensible to hemolysis. Insulinoma diagnosis criteria based on a glycemia nadir < 2, 5 mmol/l concomitant with an insulin level > 3 mUI/l and a C-peptide level > 0.2 nmol/l, applied to our retrospective cohort, had a sensibility of 95%, a specificity of 97%, a positive predictive value of 86% and a negative predictive value of 99%.

Conclusion

Using current insulin immunoassays, it seems reasonable to make the diagnosis of insulinoma on the occurrence of a glycemia < 2, 5 mmol/l concomitant with a C-peptide level > 0, 2 nmol/l associated with an insulin level > 3 mUI/l on non-hemolyzed blood samples.

Volume 73

European Congress of Endocrinology 2021

Online
22 May 2021 - 26 May 2021

European Society of Endocrinology 

Browse other volumes

Article tools

My recent searches

No recent searches.