Slow response to loss of glycemic control in type 2 diabetes mellitus

Background: To achieve glycemic control in type 2 diabetes mellitus, the American Diabetes Association (ADA) recommends intensification of glucose-lowering therapy when the glycosylated hemoglobin (HbA(1c)) level exceeds 8.0%. Objective: To, study glycemic control before and after initiation of secondary antihyperglycemic therapy to better understand the pace and patterns of therapeutic failure and clinical responses to failure. Study Design: A retrospective, population-based observational study. Patients and Methods: From a 12-year-old diabetes registry of members of Kaiser Permanente Northwest, a large group-model HMO, we tracked the glycemic control histories of all 570 registrants who, in 1998, added metformin therapy to sulphonylurea monotherapy. Results: The last HbA(1c) level before metformin use averaged 9.4%. Metabolic decompensation accelerated over time: Patients typically spent numerous months at and had several measurements of HbA(1c) >8.0% before a final glycemic spike to >9.0%. Persons experiencing more gradual failure accumulated greater glycemic burdens before changing therapy. Conclusions: The level of HbA(1c) that seemed to trigger glucose-lowering action was 9.0% or higher, not 8.0% as recommended by the ADA. A substantial hyperglycemic peak preceded change in therapy even in this relatively tightly controlled population with type 2 diabetes mellitus. Earlier therapeutic changes, but not more frequent testing, would prevent the glycemic excursions we observed. Low mean HbA(1c) levels in populations do not necessarily indicate that loss of glycemic control is being rapidly addressed for most patients. More research is needed to estimate the impact of these peaks on current well-being and future complications.