Background: A C-peptide testing carried out several years after a clinical diagnosis of T1DM could uncover a misdiagnosis in some cases. T1DM in Asian population are extremely low in prevalence and display some unique features. This study aims to determine the prevalence and characteristics of Thai adult patients with clinically misdiagnosis with T1DM.

Materials and Methods: A cross-sectional study of Thai patients diagnosed with T1DM registered at the Theptarin Hospital, Thailand was performed. Random plasma C-peptide were measured in all cases and mixed meal tolerance test (MMTT) were performed if random plasma C-peptide was in between 0.1-0.5 ng/mL. Clinically misdiagnosis with T1DM defined by preserved endogenous insulin secretion (random C-peptide ≥ 0.6 ng/mL or stimulated C-peptide after MMTT ≥ 0.6 ng/mL) after at least 3 years after onset of DM.

Results: A total of 73 patients (females 52.1%, mean age 42.2±12.5 years, BMI 23.2±3.3 kg/m2, A1C 7.8±1.3%) with duration of diabetes 20.3±11.3 years were studied. Only 42.5% of patients had measurement of pancreatic auto-antibodies. The prevalence of anti-GAD and anti-IA2 were 53.3% and 20.8%, respectively. Preserved endogenous insulin secretion (random C-peptide ≥ 0.6 ng/mL) were found in 8 patients. The overall prevalence of clinically misdiagnosis with T1DM was 11.0%. When compared with truly T1DM, misdiagnosed patients had higher prevalence of hypertension and diabetic complications. Five were reclassified as having T2DM and 3 patients were suspected to have monogenic diabetes.

Conclusions: Our results were consistent with previous studies from Caucasian patients that about 10-15% of clinically diagnosed T1DM had been wrongly diagnosed. In clinical practice, positivity for pancreatic auto-antibodies was less helpful than a simple random C-peptide measurement. C-peptide testing a few years after a diagnosis of T1DM should be a routine procedure to identify some patients who might have been misdiagnosed.

Disclosure

Y. Thewjitcharoen: None. S. Krittiyawong: None.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.