Predicting non-insulin-dependent state in patients with slowly progressive insulin-dependent (type 1) diabetes mellitus or latent autoimmune diabetes in adults. Reply to Sugiyama K and Saisho Y [letter]

概要

When we analyse the clinical data of patients with SPIDDM to elucidate factors influencing ‘progression to an insulin-dependent state,’ we must take into consideration that the initiation of insulin therapy is determined by attending physicians. Indeed, the decision could be affected by clinical data, including low BMI and high levels of HbA1c and GAD antibodies (GADAb), as Sugiyama and Saisho pointed out [1]. In this study, we therefore focused on patients who did not require insulin therapy, rather than those who progressed to an insulin-dependent state [2]. We excluded patients who did not receive insulin therapy but had a mean HbA1c of ≥64 mmol/mol (8.0%) during the last year of the follow-up period because we were unsure whether they were in a non-insulin-dependent state. Among patients who received insulin therapy after the diagnosis of SPIDDM, we excluded those with a mean HbA1c of <64 mmol/mol (8.0%) at the time of insulin initiation from the study because it is possible that the decision regarding the initiation of insulin therapy was influenced by the fact that these patients were positive for GADAb. Patients who were already treated with insulin when they were first found to be positive for GADAb were also excluded because it was not possible to judge the duration from the diagnosis of SPIDDM to the initiation of insulin therapy. We therefore do not believe that there is a selection bias. Rather, we emphasise that the 345 patients were selected for the analyses to avoid the influence of the decision by attending physicians.

We completely agree with the comments by Sugiyama and Saisho that careful assessment of the progression to an insulin-dependent state is important in patients with SPIDDM [1], for which measuring serum C-peptide levels is an option. However, when we analysed the data, we found that serum C-peptide levels were not always measured in patients with SPIDDM. In addition, some were measured in fasting states, whereas others in postprandial states. Analysis of the C-peptide levels was therefore difficult. This is a limitation of our study as mentioned in the discussion [2].