A 56-year-old man with brittle type 1 diabetes and unaware of the effects of hypoglycemia was started on a continuous subcutaneous insulin infusion (CSII) in April 2000. After 12 months, he achieved excellent glycemic control, and his HbA1c values averaged 6.5%. During this time, however, the patient required admission to the hospital on four separate occasions for diabetic ketoacidosis despite frequent self-monitored blood glucose (SMBG) (six to eight times per day) and frequent catheter insertion site changes. The patient insisted that he administered subcutaneous injections, as directed, when there was any question of pump dysfunction. Medical teams noted that the patient developed diabetic ketoacidosis very rapidly on several occasions. During one admission, he reported an SMBG value of 99 mg/dl at 10:00 a.m. Within 95 min, the patient was brought to the emergency room with a glucose level of 510 mg/dl and an anion gap of 35 mmol/l. Because of the frequent episodes of diabetic ketoacidosis, the patient’s insulin therapy was switched from CSII to multiple daily insulin injections. However, the patient preferred CSII therapy for the quality-of-life benefits provided by the insulin pump, particularly the greater flexibility in meal planning, fewer subcutaneous injections, and less frequent hypoglycemic episodes. To accommodate the patient’s wishes and prevent diabetic ketoacidosis, we devised the following treatment strategy. Sixty percent of basal insulin was provided by a daily injection of glargine insulin, and his bolus requirements were provided by the insulin pump. The basal rate of the pump was programmed for 0.2 units/h to prevent the insulin from crystallizing within the catheter. After 18 months, the patient has experienced no further episodes of diabetic ketoacidosis and has maintained acceptable glycemic control with HbA1c values averaging 7.1%.
With CSII treatment, our patient had frequent occurrences of diabetic ketoacidosis, which is a morbid and potentially lethal consequence of the failure to deliver adequate amounts of insulin. When basal insulin infusion rates are interrupted in patients treated with CSII, the subcutaneous reserves of short-acting insulin are insufficient to prevent the metabolic processes that lead to hyperglycemia and ketogenesis (1). Glargine insulin is an alternative to CSII therapy for mimicking physiological basal insulin secretion. Glargine insulin kinetics demonstrate relatively consistent insulin levels for ≥24 h after a single subcutaneous injection (2,3). In our patient, glargine insulin limited the ketosis and the associated complications that occurred with temporary infusion interruptions with the CSII. By combining daily glargine insulin injections with short-acting insulin boluses from an insulin pump, our patient had no episodes of diabetic ketoacidosis and maintained the lifestyle benefits provided by the insulin pump.
We must note that ketoacidosis rates have diminished in patients treated with CSII. Currently, the rates of diabetic ketoacidosis are similar in patients treated with CSII or multiple daily injections (4). However, our strategy may benefit some patients who have recurrent diabetic ketoacidosis on insulin pump therapy.
