Comparison of adaptive and sliding-scale glycaemic control in critical care and the impact of nutritional inputs

Abstract

Stress-induced hyperglycaemia is
\nprevalent in critical care. Tight glucose control can
\nreduce mortality up to 43%. An adaptive control
\nalgorithm utilising insulin and nutritional feed
\ninputs for targeted glycaemic control in critically ill
\npatients is presented. Validation is performed using
\nretrospective patient data (n=19) in simulated
\nglucoregulatory trials. Conventional sliding-scale
\nand insulin-only methods are compared. Results
\nshow a reduction in mean glucose levels, and
\nvariability. A 312% increase in time spent in the 4-
\n6mmol/L normal glucose band compared to sliding scale
\nand a 240% increase compared to the insulin only
\nprotocol is reported. Results are obtained using
\n60% more insulin and 20% more nutrition across a
\nwide cross-section of ICU patients and patient
\ncondition, indicating that the timing of control input
\nadministration is more crucial than their absolute
\namounts. The results show potential in reducing ICU
\nmortality and the risk of severe complications.