Comparison of the post-meal glucose response to different insulin bolus waveforms in insulin pump- and pre-meal pramlintide-treated type 1 diabetes patients.

BACKGROUND: Both pramlintide and insulin pump waveforms separately provide improved post-meal glucose control. However, when used together there may be a mismatch in actions leading to hypoglycemia. We studied the three currently available waveforms and a "modified combination wave" (MC) in pramlintide-treated patients. The MC was a "square" (SQ) wave combined with a "standard" (ST) bolus that was delayed 1 h into the mealtime.

METHODS: Using the CGMS Gold (Medtronics, Northridge, CA) we measured the glucose response 0-4 h after the beginning of a meal and 15 min after the initiation of the insulin bolus wave and pramlintide (60 microg), bolus. Pump-treated type 1 diabetes subjects were randomized to one bolus waveform for one full day of three meals and then crossed over to the other bolus waveforms. In the first study nine subjects were randomized to the ST, SQ, or "combination" (C) wave consisting of a ST wave at the beginning of a SQ wave. In the second study nine subjects were randomized to SQ or MC.

RESULTS: With ST and C waves glucose fell approximately 40 mg/dL 0-2 h post-meal and then returned to baseline by the fourth hour. Initially isoglycemic, the SQ wave increased approximately 20 mg/dL in the late meal period. The MC demonstrated minimal (approximately 10 mg/dL) change during the entire post-meal period.

CONCLUSIONS: For currently available insulin pump bolus waves, SQ and MC may be least likely to cause hypoglycemia in pramlintide-treated patients.