1046-P: Day-to-Day Variability of Insulin Requirements in Inpatients on General Wards

Aim: To characterize variability of exogenous insulin requirements during fully closed-loop insulin delivery in hospitalized patients with hyperglycaemia requiring subcutaneous insulin.

Methods: We retrospectively analyzed data from randomized trials investigating fully closed-loop insulin delivery for up to 15 days in inpatients with hyperglycaemia on general wards (type 1 diabetes excluded). We calculated insulin amounts delivered by closed-loop overnight (2300-0449), during breakfast (0500-1059), lunch (1100-1659), and evening meal (1700-2249). The coefficient of variation (CV) quantified between day and between night variability of insulin requirements in inpatients using closed-loop for at least 5 days.

Results: Data from 487 days in 55 participants were analyzed [age 69(8)years, weight 100(32)kg, baseline HbA1c 8.0(2.1)%, duration of diabetes 18(13)years, duration of insulin 9(10)years, mean(SD)]; 14.5% of participants received dialysis, 29.1% enteral/parenteral nutrition and 21.8% steroids. Closed-loop directed insulin infusion rate was 1.6(0.8), 2.9(1.2), 3.1(1.4), 2.1(1.0)units/kg/h during overnight, breakfast, lunch, and evening meal periods, respectively. The CV of insulin delivery was higher between nights than between any of the daytime periods [overnight 54(21)% compared with 42(21)% at breakfast, 41(16)% lunchtime and 48(17)% dinner; p=0.001].

Conclusions: Overnight exogenous insulin requirements are more variable than daytime requirements. Particular attention needs to be paid to glucose management overnight to prevent dysglycaemia in inpatients receiving insulin therapy on general wards.