Natural course of remission in IDDM during 1st yr after diagnosis

OBJECTIVE - To describe the natural course of clinical remission in insulin-dependent diabetes mellitus (IDDM) when insulin dose is minimized without loss of target glycemia and to identify factors that predict clinical remission. RESEARCH DESIGN AND METHODS - Ninety-five patients, who were placebo-treated control subjects in the Canadian-European multicenter randomized trial of cyclosporin A in recent-onset IDDM, were studied. RESULTS - The mean insulin dose decreased during the first months after diagnosis, with a nadir at 3 mo, when 27% of the patients did not require insulin to maintain target glycemia. At 1 yr, 10% of patients still did not need insulin. Patients not receiving insulin who had glycosylated hemoglobin within the normal range were called remitters. Mean basal and glucagon-stimulated C-peptide values were significantly (P < 0.025) higher in remitters than nonremitters at the start of the study. Therefore, all patients were divided into those with values above the mean stimulated C-peptide (0.4 nM) and those with values below the mean at entry. The probability of entering a remission with a stimulated C-peptide >0.4 nM was 10 times as high (P < 0.05) as for those with a stimulated C-peptide below this level. Surprisingly, the beginning and end of the remission were associated with neither major changes in C-peptide levels nor islet cell antibody and insulin-antibody titer. A more rapid loss of stimulated C-peptide occurred in patients who lacked HLA-DR3 and -DR4 (P < 0.05 at mo 9). CONCLUSIONS - This study shows a higher spontaneous clinical remission rate than expected during the 1st yr after diagnosis. Preserved β-cell function at entry predicts a greater chance of entering a remission, and a more rapid loss of β-cell function was seen in patients without HLA-DR3 and -DR4.