Objective: To observe β cell function in newly diagnosed diabetic patients with ketosis before and in remission period and evaluate its classification and predictive value.
Methods: A total of 206 patients newly diagnosed as diabetic ketosis who had been treated with intensive insulin therapy in our hospital and entered in the "honeymoon" after the withdraw of insulin therapy were followed for 36 months from onset of diabetes. They were divided into two groups of type 1 and type 2 diabetes (group A and B), according to the dependence or independence on insulin treatment. The β cell function of the two groups before and in remission period was compared by oral glucose tolerance test (OGTT). β cell function was measured with the AUC of insulin and C-peptide and homeostatic model assessment β-cell function (HOMA-β), while homeostatic model assessment insulin resistant (HOMA-IR) for insulin resistant. The duration of the "honeymoon" and the change of insulin and C-peptide curve before and in "honeymoon" were also observed.
Results: The AUC of insulin and C-peptide, the HOMA-β and the HOMA-IR before and after the intensive insulin treatment were lower in group A than that in group B [before the insulin treatment: (10.18 ± 2.36) mIU×h×L(-1) vs (20.28 ± 6.89) mIU×h×L(-1), (1.56 ± 0.53) µg×h×L(-1) vs (3.75 ± 0.67) µg×h×L(-1), 3.68 ± 1.08 vs 18.20 ± 6.59, 1.22 ± 0.49 vs 3.06 ± 1.54, respectively; after the insulin treatment: (29.86 ± 8.65) mIU×h×L(-1) vs (93.35 ± 19.42) mIU×h×L(-1), (3.99 ± 0.79) µg×h×L(-1) vs (12.54 ± 3.83) µg×h×L(-1), 8.50 ± 2.46 vs 56.17 ± 19.42, 0.63 ± 0.56 vs 1.42 ± 0.78, respectively]. The duration of the "honeymoon" in group A was significantly shorter than in group B [(7.9 ± 5.2) months vs (20.9 ± 9.9) months]. In oral glucose insulin and C-peptide release test, the peak of insulin and C-peptide releasing curve in group A was brought forward by a half to 1 hour after intensive treatment while delayed in group B by 1 or 2 hours. The releasing peak of insulin and C-peptide in group A was less than two folds of the basic value, while four to ten fold of the basic value in group B. The positive ratio of glutamic acid decarboxylase antibody, insulin autoantibody and insular cellular antibody in group A and group B were 21.2% vs 4.8%, 18.1% vs 3.3%, 9.2% vs 10.6%, respectively.
Conclusions: Of all the patients newly diagnosed as diabetes ketosis who had entered into the honeymoon after intensive insulin therapy, 91% were type 2 diabetes. Inferior β cell function before insulin therapy, weaker remission after insulin therapy and shorter duration of remission period suggest the classification of type 1 diabetes.