The value of continuous glucose monitoring in qualitative diagnosis of insulinoma

Abstract

Objective To explore the diagnostic value of continuous glucose monitoring(CGM) in insulinoma. Methods Twenty-two insulinoma patients (insulinoma group) , 11 hypoglycemia patients without insulinoma (hypoglycemia group) , and 31 individuals with normal glucose regulation (control group) hospitalized in Chinese PLA General hospital from May 2011 to April 2014 were included. Blood biochemical markers including glucose metabolism markers were tested; CGM data were obtained. The EasyGV Version 9.0 software was used to analyze the CGMS data and to calculate the index of glucose variability. Statistics methods including t test, analysis of variance, and non-parametric test were used. Curve of receiver operating characteristic (ROC) and Youden Index were applied to determine the cut-off value. Results (1)The fasting plasma glucose(FPG) and the lowest blood glucose(LBG) of insulinoma group were significantly lower than those of hypoglycemia group and control group (F=14.72, 9.74, P<0.05) . The fasting insulin (FINS), fasting C-peptide (FCP) and fasting insulin release index(FIRI) of insulinoma group were significantly higher than those of the other two groups (F=10.72, 5.07, 12.8, all P<0.05) ;(2) The continuous overall net glycemic action (CONGA) of insulinoma group was significantly lower than that of control group [3.98 (3.71, 4.63) vs 5.10 (4.78, 5.52) , Z=3.989, all P<0.05], whereas the low blood glucose index (LBGI) and M-value were higher than those of control group [7.38 (4.99, 11.92) vs 1.59 (0.73, 2.70) , 11.99 (8.45, 23.04) vs 1.68 (0.68, 3.27) , Z=2.780-5.163, all P<0.05]. Compared with hypoglycemia group, CONGA of insulinoma group was lower [3.98 (3.71, 4.63) vs 5.03 (4.40, 6.30) , Z=2.595, P<0.05], and LBGI, M-Value were higher[7.38 (4.99, 11.92) vs 1.89 (0.96, 4.02) , 11.99 (8.45, 23.04) vs 2.95 (1.05, 7.60) ; Z= 2.970, 2.526, all P<0.05]; (3)Results from the ROC curve and the Youden Index showed that the cut-off value of LBGI, M-value, CONGA was 4.06, 7.79 or 4.38, respectively. LBGI had the highest diagnostic power of insulinoma with 90.9% sensitivity and 81.8% specificity when using 4.06 as the cut-off point. Conclusions Continuous glucose monitoring data can be used to diagnose insulinoma and fluctuation indicators of blood glucose such as LBGI, M-Value and CONGA might be useful to identify insulinoma. LBGI has the highest diagnostic power of insulinoma. Key words: Hypoglycemia; Insulinoma; Continuous glucose monitoring; Glucose variability