Protamine-containing insulin but not analog insulin and duration of insulin use are risk factors for the production of insulin autoantibodies in insulin-treated patients with diabetes mellitus.

Insulin autoantibodies can be produced by insulin injections but rarely cause severe side effects such as glucose instability and insulin allergy. We study the characteristics of insulin autoantibody-positive diabetic patients with a medical history of insulin therapy using single and multiple (adjusted for age, sex, type of diabetes) logistic regression analyses. Associations between insulin autoantibodies and age, sex, type of diabetes, HbA1c, and serum creatinine were not significant, but the association between insulin autoantibodies and duration of insulin use was significant. Unadjusted and adjusted odds ratios were 1.08 (1.02-1.14) and 1.07 (1.01-1.14), respectively. Unadjusted and adjusted odds ratios for protamine-containing insulin were 3.08 (1.49-6.34) and 4.27 (1.90-9.58), respectively. The adjusted odds ratios for premixed biphasic insulin and intermediate-acting insulin were 2.21 (1.03-4.73) and 2.35 (1.01-5.49), respectively. Associations between insulin autoantibodies and any insulin analog were not significant. These results suggest that protamine-containing insulin and duration of insulin use are risk factors for the production of insulin autoantibodies. If patients with poorly controlled diabetes have a history of protamine-containing insulin therapy over a long time, the appearance of insulin autoantibodies should be monitored.