Can Insulin Glucose Infusions be safely managed in General Wards?  — ASN Events
  1. Schedule
  2. ADS Clinical Poster Viewing: Diabetes in Hospital
  3. Can Insulin Glucose Infusions be safely managed in General Wards? 

Can Insulin Glucose Infusions be safely managed in General Wards?  (#346)

Wenlin Cecilia Chi 1 , David Chipps 1 2 , N Wah Cheung 1 2 , Sylvia Lim-Tio 1 2
  1. Department of Diabetes and Endocrinology, Westmead Hospital, Westmead, NSW, Australia
  2. University of Sydney, Sydney, NSW, Australia

Background: Inpatients with diabetes have increased mortality. Both hypoglycaemia, and hyperglycaemia, are independent risk factors for mortality. Insulin glucose infusions offer the most flexible means of optimising glycaemic control, but require frequent monitoring and staff training; hence is restricted to intensive care settings in many hospitals.

Objectives:

(i)                Review use and safety of insulin infusions in surgical inpatients with diabetes

(ii)               Identify common problems and their impact on glycaemic outcomes.

Methods: Our institutional guidelines recommend perioperative insulin infusions to optimise glycaemic control in patients with insulin-requiring diabetes. We reviewed 53 consecutive surgical inpatients with diabetes who met the criteria for requiring insulin infusion. Data was collected for protocol deviations and glycaemic outcomes.

Results: 90 insulin infusions were, or should have been, started in 53 patients. 46 patients had type 2 diabetes; 7 had type 1 diabetes. Mean age was 64±11.9 years. Mean HbA1c was 8.7±1.9%. Protocol deviations occurred with 54 infusions (40%), most commonly failure to overlap insulin infusion with the 1st dose of subcutaneous insulin, followed by inadequate BSL monitoring (10%). Errors in insulin dose titration following our algorithm were rare (2%). Most protocol deviations (44%) occurred in intensive care/high dependency units, rather than general wards. Of these, 25% occurred during ward transfer.

There were 20 adverse glycaemic outcomes: 1 case of diabetic ketoacidosis, 13 cases of moderate hyperglycaemia (BSL >15.0mmol/l), 3 cases of minor hypoglycaemia and 3 cases of major hypoglycaemia. Patients with type 1 diabetes had a higher risk of having an adverse outcome as a result of protocol deviation (p=0.017).

Conclusions: Protocol deviations frequently occurred during insulin infusions. Contrary to common expectations, protocol deviations occurred more commonly in critical care units than in general wards; and more commonly at stepdown. This audit provides reassurance that insulin infusions can be equally safely used in general wards as well as critical care units.