Aims: Half of the patients with newly diagnosed type 2 diabetes who received short-term continuous subcutaneous insulin infusion (CSII) were able to achieve a long-term drug-free glycemic remission. It is meaningful to develop an estimation formula to predict the glycemic remission.
Methods: We collected data from three prospective trials conducted from 2002 to 2015. A 2∼3 weeks of CSII was provided for the patients with newly diagnosed diabetes to achieve and maintain normal glucose. The one-year drug-free remission rate was observed. Differences between patients with remission and those without remission were evaluated. Estimation formula was developed by using the logistic regression models. The receiver operating characteristic (ROC) curve was used to determine the sensitivity and specificity of the estimation formula at different score cutoffs.
Results: Data of 244 patients were analyzed. After comparisons, age, gender, body mass index (BMI), high-density lipoprotein cholesterol (HDL-C), red cell distribution width (RDW), ratio of fasting plasma glucose to HbA1c (FPG/HbA1c) at baseline, FPG, postprandial plasma glucose (PPG) and the homeostasis model assessment of insulin resistance (HOMA-IR) after short-term CSII treatment were analyzed. We developed a formula as follows: R=11.936-1.626*HDL-1.56*FPG/HbA1c-0.14*RDW-0.496*FPG’ (FPG’ means FPG after short-term CSII). When R is greater than 0.5, it indicates a possible glycemic remission. The sensitivity, specificity, accuracy, positive predictive value and negative predictive value were 82.8%, 62.2%, 73.7%, 73.3% and 74.2%, respectively.
Conclusions: We provide a prediction formula for long-term glycemic remission. For those who is predicted glycemic non-remission, sequential treatment after CSII secession may be beneficial for the long-term glucose control.
L. Xu: None. L. Liu: None. X. Wan: None. Z. Huang: None. Y. Li: None.