Tight glycaemic control (TGC) in critical care has shown distinct benefits, but has also proven difficult to obtain. The risk of severe hypoglycaemia (< 2.2mmol/L) raises significant concerns for safety. Continuous Glucose Monitors (CGMs) offer frequent, though potentially noisy, automated measurement and thus the possibility of using them for early detection and intervention of hypoglycaemic events. This in-silico study investigates the potential of CGM devices to maintain control, prevent hypoglycaemia and reduce clinical effort. Retrospective clinical data from the SPRINT TGC study covering 26 patients was used with clinically validated metabolic system models and 3 different stochastic noise models (two Gaussian and one first-order autoregressive.) The noisy, virtual CGM blood glucose (BG) values were filtered and used to drive the SPRINT TGC protocol. A simple threshold alarm was used to trigger glucose interventions to avert potential hypoglycaemia. Monte Carlo analysis was used to get robust results from the stochastic noise models. Using SPRINT with simulated CGM noise, the BG time in the 4.4-6.1mmol/L band was reduced no more than 3% from 45.2% obtained with glucometer sensors. The number of patients experiencing severe hypoglycaemia was reduced by 0-30%. Duration of hypoglycaemic events was reduced by 19-65%. Finally, nurse workload was reduced by approximately 20 minutes per patient, per day. The results of this proof of concept study justify a pilot clinical study for verification in a clinical setting.