After providing written informed consent, the first 20 Subjects meeting Inclusion/Exclusion Criteria will be consecutively enrolled in the Standard of Care cohort. The moderate treatment cohort will then be consecutively enrolled, followed by the tight glycemic control cohort.

After sensor insertion, baseline evaluations including APACHE II, SOFA, and laboratory evaluations will be determined.

Subjects enrolled in the standard of care cohort will be treated according to the institution's protocol for measuring glucose and managing insulin. These subjects will be monitored on a GlucoClear System but they will not be managed based on the values or trends of the GlucoClear system.

Subjects enrolled in the treatment cohorts will be monitored and managed with a special version of the GlucoClear continuous monitoring system. This system contains the GlucoClear Insulin Dosing Algorithm providing insulin dosing recommendations to enable the clinician to manage patient glucose within pre-specified target levels. These recommendations are presented on screen for a clinical professional to approve or override.

Subjects in the moderate treatment cohort will have their glucose managed in the range of 120 - 180 mg/dl. Subjects in the tight glycemic control treatment cohort will be managed in the range of 80 - 120 mg/dl.

After discharge from the ICU, subjects will followed for adverse events and mortality at 30 days, either by telephone contact or office visit.

Hyperglycemia, hypoglycemia and glycemic variability, the three components of dysglycemia are independently associated with morbidity and mortality of critically ill patients. Blood glucose (BG) control with insulin has the potential to decrease morbidity and mortality of intensive care unit (ICU) patients. Blood glucose control with insulin, however, is associated with an increased risk of hypoglycemia and its effect on glycemic variability is uncertain. BG control with insulin utilizing manual systems for glucose measurement is blood-consuming and time-consuming, since frequent blood draws for glucose measurements are necessary in order to achieve blood glucose control.

Severe hypoglycemia (blood glucose level < 40 mg/dl) is a feared complication of blood glucose control with insulin. Undoubtedly, with implementation of blood glucose with insulin the incidence of hypoglycemia increases. Reported incidences of severe hypoglycemia rise by 5 to 10-fold as compared to conventional glucose control strategies in randomized controlled trials. Several reports showed a significant association between hypoglycemia and patient outcomes. Recently, an association has even been suggested between moderate and mild hypoglycemia (blood glucose level between 40-69 mg/dL) and patient outcomes. Incidences of moderate hypoglycemia are more prevalent than severe hypoglycemia. The risk of developing (severe or moderate) hypoglycemia hampers, at least in part, broad implementation of blood glucose control with insulin, in particular when aiming at normal blood glucose levels [17,18].

Recent studies showed also significant associations between glycemic variability and patient outcomes. BG control algorithms, if properly applied, could decrease glycemic variability. Experimentally, rodent experiment showed that brain damage was not associated with the duration of severe hypoglycemia, but instead with its correction (mainly overcorrection) with intravenous dextrose causing formation of radicals. Hence, a close glucose monitoring to prevent overcorrection is mandatory.

These different arguments strongly support the need for reliable and accurate CGM. The GlucoClearTM Continuous Glucose Monitoring system from Edwards Lifesciences measures blood sugar by Glucose Oxidase Sensing Technology through in-blood measurement. Blood is automatically drawn and analyzed every 5 minutes from a peripheral venous catheter, with real time graphical display. Blood is then returned to the patient and the system automatically self-calibrates. The GlucoClear CGM is designed to be highly accurate. In a recent study performed in critically ill patients, it was shown to comply with the required standards of quality.

The main research question addressed by the present study is: "To which extent a CGM-guided strategy improve the quality of BG control performance"?

Inclusion Criteria:

• Have an anticipated ICU stay of at least 48 hours;
• Participate or have authorized representative participate in the Informed Consent process and sign/date the approved informed consent form.
• Have an expected ICU stay of ≥ 24 hours up to 72 hours
• Have an APACHE II score of ≥ 10, within the first 24 hours of ICU admission.
• Not be participating in any other investigational interventional study.
• Have hyperglycemia (BG > 150 mg/dl) up to 6 hours after admission and / or ongoing insulin therapy.
• Access to a large peripheral vein

Exclusion Criteria:

• End stage pre-existing medical conditions that would preclude the subject from being able to complete up to 72 hours of glucose monitoring and follow up (14 ± 3 days after Sensor removal); Restricted peripheral venous access, inadequate access for reference sampling or any contraindication to the placement of peripheral IV catheters.
• Skin conditions or existing (or planned) medical instrumentation and/or dressings that preclude placing the peripheral IV catheters or dressings (e.g., extensive psoriasis, recent burns or severe sunburn, extensive eczema, extensive scarring, extensive tattoos, dermatitis herpetiformis, or surgical dressings).
• A known history of heparin-induced thrombocytopenia or any contraindication for anti-coagulation therapy.
• Intolerance to Lactated Ringer's solution ((mEq/liter): Sodium 130; Potassium 4; Calcium 3; Chloride 110; Lactate 28).
• A positive pregnancy test 72 hours prior to Sensor insertion (for subjects of child bearing potential).
• Any condition that, in the opinion of the Investigator, would interfere with their participation in the Study.
• Participated in any other drug, device, or biologic Study within the last 30 days (prior to Sensor insertion) which may clinically interfere with this Study or have not completed the required Protocol follow-up period.