Asymmetric changes in circulating insulin levels after an increase compared with a reduction in insulin pump basal rate in people with Type 1 diabetes.

AimsTo investigate circulating insulin profiles after a clinically relevant insulin pump basal rate increase vs a reduction, and the associated glucose responses. MethodsA cohort of 12 adults with Type 1 diabetes undertook this two-stage university hospital study using Accu-Chek pumps (Roche Diagnostics, Mannheim, Germany) and insulin aspart. An insulin basal rate change of 0.2 unit/h (increase in first stage, reduction in second stage) was implemented at similar to 09:30 h, after a single overnight basal rate (without bolus insulin), while fasting participants rested. Frequent venous samples for the assessment of plasma free insulin, glucose and cortisol were collected from 60 min before until 300 min after rate change. The primary outcome was time to steady-state insulin. ResultsThe 0.2-unit/h rate change represented a mean sd alteration of 23 6%. After the rate increase, the median (interquartile range) times to 80% and 90% steady-state insulin were 170 (45) min and 197 (87) min, respectively. By contrast, after rate reduction, 80% steady-state insulin was not achieved. After the rate increase, mean se insulin levels increased by 4.3 3.1%, 12.0 2.9% and 25.6 2.6% at 60, 120 and 300 min, respectively (with no significant difference until 180 min). After the rate reduction, insulin decreased by 8.3 +/- 3.0% at 300 min (with no significant difference until 300 min). After rate reduction, glucose levels paradoxically declined by 17.4 +/- 3.7% after 300 min; cortisol levels also fell during observation (P = 0.0003). ConclusionsThe time to circulating insulin change after a 0.2-unit/h basal rate change was substantial, and was greater after a reduction than after an increase. Counter-regulatory hormone circadian variation may affect glycaemia when implementing minor changes at low basal rates. Both direction of basal rate change, and time of day, warrant consideration when anticipating the clinical effects of basal rate changes.