Response to "Preserved" glucagon secretion in fulminant type 1 diabetes.

We compared the glucagon responses to arginine stimulation in five patients with fulminant type 1 diabetes and five age- and diabetes duration-matched acute-onset type 1 diabetes, and the curves of the glucagon responses were similar. In our study, we measured the glucagon levels by not only radioimmunoassay, but also novel enzyme-linked immunosorbent assay.

We thank Murase‐Mishiba et al. for their comments1 on our case report2. The Authors reported that glucagon responses to mixed meal test were not impaired in patients with fulminant type 1 diabetes whose diabetes duration was >4 years. We evaluated glucagon responses to arginine stimulation in five patients with fulminant type 1 diabetes, and those in five age‐ and diabetes duration‐matched patients with acute‐onset type 1 diabetes who were enrolled in our study3 (their age and diabetes duration were 54.8 ± 20.9 vs 55.6 ± 18.1 years and 3.0 ± 6.3 vs 3.2 ± 6.9 years, respectively; data were presented as mean ± standard deviation). As shown in Figure 1, the curves of glucagon levels were similar between the two groups. Another 44‐year‐old woman underwent an arginine stimulation test at the onset of fulminant type 1 diabetes and 11 months later, and her glucagon response did not decrease (her glucagon levels at pre‐loading, at peak and the area under the curve of the glucagon were; 139 vs 204 pg/mL, 406 vs 425 pg/mL and 3.7 × 104 vs 4.6 × 104 pg/mL/min at the onset of diabetes and 11 months later, respectively). On the contrary, another study reported that immunohistochemically stained glucagon‐positive cell areas in the pancreas of patients with fulminant type 1 diabetes were significantly lower than those with autoimmune type 1 diabetes for short duration of 2.8 ± 1.9 months after the onset of diabetes4. The change in pancreatic α‐cell function might be heterogenous among patients with fulminant type 1 diabetes.

Figure 1

Glucagon response to arginine stimulation in five patients with (a) fulminant type 1 diabetes, and (b) five age‐ and diabetes duration‐matched patients with acute‐onset type 1 diabetes. Solid lines with circles show glucagon levels measured by radioimmunoassay (RIA; Sceti Medical Labo, Tokyo, Japan; the intra‐ and interassay coefficients of variation were <20 and <15%, respectively). Dashed lines with squares show glucagon levels measured by enzyme‐linked immunosorbent assay (ELISA; Mercodia AB, Sweden; the intra‐ and interassay coefficients of variation were 7.3–9.4% and 7.5–8.5%, respectively). Patients were enrolled in our study3. Variables are presented as mean ± standard deviation.

Regarding the methods to measure the glucagon level, Murase‐Mishiba et al. used conventional radioimmunoassay (RIA). In our case report, we measured the glucagon levels by not only RIA, but also with quantitative sandwich enzyme‐linked immunosorbent assays (ELISA), and the curves of the glucagon levels were similar. We also measured the glucagon levels of patients with fulminant and acute‐onset type 1 diabetes by ELISA, as shown in Figure 1. Their glucagon levels measured by RIA and ELISA were similar. However, a recent study showed that the trend of glucagon response to a mixed meal test measured by RIA was different from that measured by ELISA and novel liquid chromatography‐high resolution mass spectroscopy, which can measure the glucagon levels more specifically5. Further prospective, long‐term longitudinal studies with large numbers of patients are warranted to gain a complete understanding of the pancreatic α‐cell function in fulminant type 1 diabetes.

Disclosure

The authors declare no conflict of interest.