Professional Continuous Glucose Monitoring for the Identification of Type 1 Diabetes Mellitus Among Subjects With Insulin Therapy

The identification of type 1 diabetes in diabetic subjects receiving insulin therapy is sometimes difficult. The purpose of this study is to evaluate whether results of professional continuous glucose monitoring can improve the identification of type 1 diabetes. From 2007 to 2012, 119 adults receiving at least twice-daily insulin therapy and professional continuous glucose monitoring were recruited. Type 1 diabetes was diagnosed by endocrinologists according to American Diabetes Association standards, including a very low C-peptide level (<0.35 pg/mL) or the presence of diabetic ketoacidosis. Continuous glucose monitoring was applied for 3 days. Among 119 subjects, 86 were diagnosed with type 1 diabetes. Subjects with type 1 diabetes were younger (33.8 vs 52.3 years old, P < 0.001), had lower body mass index (BMI, 21.95 vs 24.42, P = 0.003), lower serum creatinine (61.77 vs 84.65 μmol/L, P = 0.001), and higher estimated glomerular filtration rate (108.71 vs 76.48 mg/mL/min/1.73m2, P < 0.001) than subjects with type 2 diabetes. Predictive scores for identification of type 1 diabetes were constructed, including age, BMI, average mean amplitude of glucose excursion in days 2 and 3, and the area under the curve of nocturnal hyperglycemic and hypoglycemic states. The area under the receiver operating characteristic curve was 0.90. With the cutoff of 0.58, the sensitivity was 86.7% and the specificity was 80.8%. The good performance was validated by the leave-one-out method (sensitivity 83.3%, specificity 73.1%). Professional continuous glucose monitoring is a useful tool that improves identification of type 1 diabetes among diabetic patients receiving insulin therapy.