Glycaemia and cardiac rhythm disturbances in survivors of critical illness with pre-existing diabetes

Thirty patients with diabetes diagnosed previously or on this admission (HbA1c > 6.5%) who were admitted to ICU with at least 24 h of organ failure and require either insulin or sulponylureas following discharge from ICU and who are anticipated to remain an inpatient for at least 48 h after ICU discharge will be eligible. All patients enrolled into the study must be able to provide informed consent. All patients will have simultaneous 12-lead Holter and continuous interstitial glucose (IG) monitoring (CGM) from enrolment until discharge or for a maximum of 5 days. There will be no change in treatment instituted, e.g. patients will carry on with their usual activities and diabetes treatments. Twelve-lead ambulatory ECGs (Lifecard 12; Spacelabs Healthcare, Hertford, U.K.) will be recorded at a sampling rate of 128 Hz with electrodes in a Mason-Likar configuration. Patients will also have a time-synchronized CGM attached (FreeStyle Navigator Continuous Glucose Monitoring System; Abbott Diabetes Care, Maidenhead, U.K.). Calibrations will ideally be performed daily (but no less frequent than 48 hourly) according to the manufacturer’s instructions. Studies will occur in ‘blinded mode’ which means that alarms do not sound and patients and health care practitioners cannot view CGM glucose values. Patients/health care practitioners will be also asked to keep a record of any symptomatic hypoglycaemia. An episode of IG < 3.5 mmol/L on CGM without simultaneous self-report of symptoms is regarded as asymptomatic.