GLUCOSAFE 2 - A New Tool for Nutritional Management and Insulin-therapy in the Intensive Care Unit (ICU)

HEIDEGGER CP213 enrolled

Overview

The survival and the outcomes of critically ill patients are strongly influenced by insulin-therapy and nutritional support. The GLUCOSAFE 2 pilot study, aims to test the performance and the security of the new GLUCOSAFE 2 software, developed by the model-based medical decision support of Aalborg University (Denmark) and adapted to the clinical needs in the intensive care unit (ICU) of the Geneva University Hospital (HUG). This new device is based on a mathematical model of the glucose-insulin metabolism and attempts to give advices for better glycaemia control and nutritional therapy. The GLUCOSAFE 2 study hypothesizes that the use of the Glucosafe 2 software will allow better glycaemia ("Time-in-target") control and better achievement of nutritional energy and protein targets in comparison to the local protocols.

Study Type

INTERVENTIONAL

Allocation

RANDOMIZED

Purpose

TREATMENT

Masking

NONE

Enrollment

213

Conditions

Critical Illness Energy Supply; Deficiency, Severe Protein Deficiency

Interventions

GLUCOSAFE 2DEVICE

Use of GLUCOSAFE 2 software for nutrition management and insulin-therapy

Local protocol control group with routine careDEVICE

Use of the local protocols (electronic or paper version) for nutrition management and blood glucose control.

Historical control group with routine careDEVICE

Use of the local protocols (electronic or paper version) for nutrition management and blood glucose control.

Eligibility

Sex: ALLMin age: 18 Years

Medical Language ↔ Plain English

Inclusion Criteria: All patients ≥ 18 years admitted to the ICU with * An expected length of stay ≥ 72h * At least 1 blood glucose (BG) measurement ≥10 mmol/l or 2 BG measurement ≥ 8.5 mmol/l * Informed Consent signed by the subject/ legal representative, except for patients in the historical control group Exclusion Criteria: * Lack of legal consent or consent withdrawn, except for patients in the historical control group * Pregnant or breast feeding * Diabetic ketoacidosis or hyperosmolar state * Oral feeding * Fulminant hepatic failure * Medically contraindicated to receive rapidly acting insulin by intravenous (iv) infusion or iv injection

Locations (2)

Hôpitaux Universitaire de Genève

Geneva, Canton of Geneva, Switzerland

Service of Intensive Care, Geneva University Hospital,

Geneva, Canton of Geneva, Switzerland

Outcomes

Primary Outcomes

Time-in-target (range: 5.0 - 8.5 mmol/l)

Time spent in the glycaemia range of 5.0 - 8.5 mmol/l per day, per patient and in the cohort

Time frame: During ICU stay, up to 15 days post-randomization.

Secondary Outcomes

Overall number of mild (<3.3 mmol/l) and severe (<2.2 mmol/l) hypoglycemia events

Overall number of mild (\<3.3 mmol/l) and severe (\<2.2 mmol/l) hypoglycemia events (per patient and in the cohort)

Time frame: During ICU stay, up to 15 days post-randomization.

Overall percentage of mild (<3.3 mmol/l) and severe (<2.2 mmol/l) hypoglycemia events

Overall percentage of mild (\<3.3 mmol/l) and severe (\<2.2 mmol/l) hypoglycemia events (per patient and in the cohort)

Time frame: During ICU stay, up to 15 days post-randomization.

Number of mild (<3.3 mmol/l) and severe (<2.2 mmol/l) hypoglycemia events due to non-compliance

Overall number of mild (\<3.3 mmol/l) and severe (\<2.2 mmol/l) hypoglycemia events due to non-compliance (only in intervention arm)

Time frame: During ICU stay, up to 15 days post-randomization.

Percentage of mild (<3.3 mmol/l) and severe (<2.2 mmol/l) hypoglycemia events due to non-compliance

Percentage of mild (\<3.3 mmol/l) and severe (\<2.2 mmol/l) hypoglycemia events due to non-compliance (only in intervention arm)

Time frame: During ICU stay, up to 15 days post-randomization.

Time to normalize blood glucose (5.0-8.5 mmol/l)

Three values \< 8.5 mmol/l as indicator for normalization

Time frame: During ICU stay, up to 15 days post-randomization.

Percentage of time in the ICU (per patient and in the cohort) with hyperglycaemia (BG > 8.5 mmol/l)

Percentage of time with BG \> 8.5 mmol/l before and after normalization per patient and in the cohort.

Time frame: During ICU stay, up to 15 days post-randomization.

Percentage of time in the ICU (per patient and in the cohort) with hyperglycaemia (BG > 8.5 mmol/l) due to non compliance

Percentage of time with BG \> 8.5 mmol/l before and after normalization per patient and in the cohort due to non compliance (only in the intervention arm).

Time frame: During ICU stay, up to 15 days post-randomization.

Number of hyperglycemic episodes after normalization (> 8.5 mmol/l)

Three values \< 8.5 mmol/h as indicator for normalization

Time frame: From normalization time during ICU stay, up to 15 days post-randomization.

Number of hyperglycemic episodes after normalization (> 8.5 mmol/l) due to non compliance

Three values \< 8.5 mmol/h as indicator for normalization

Time frame: From normalization time during ICU stay, up to 15 days post-randomization.

Number of episodes (per patient and in the cohort) where a BG measurements is not followed up within 30 minutes by a request for Glucosafe 2 advice

Time frame: During ICU stay, up to 15 days post-randomization.

Number of episodes (per patient and in the cohort) where pumps were not set within 30 min according to Glucosafe2 advice accepted by the nurse.

Number of episodes (per patient and in the cohort) when pumps were not set within 30 min according to Glucosafe2 advice accepted by the nurse.

Time frame: During ICU stay, up to 15 days post-randomization.

Number of advices given by GS2 which were accepted, accepted with modification, or rejected.

Time frame: During ICU stay, up to 15 days post-randomization.

Frequency of daily and cumulated BG measurements per patient and in the cohort

Frequency of BG measurements (per patient and in the cohort)

Time frame: During ICU stay, up to 15 days post-randomization.

Frequency of daily and cumulated adjustments of insulin and nutrition pump settings (per patient and in the cohort)

Time frame: During ICU stay, up to 15 days post-randomization.

Glycaemic variability

Mean and standard deviation (SD) of blood glucose measurements. Daily maximum blood glucose difference.

Time frame: During ICU stay, up to 15 days post-randomization.

Protein goal achievements (80-100% of accumulated target) with a target of 1.3 g/kg of body weight per day.

Percentage of proteins received per day and at the end of the ICU stay with a target of 1.3 g/kg of body weight per day.

Time frame: During ICU stay, up to 15 days post-randomization.

Caloric goal achievements (80-100% of the accumulated target) by indirect calorimetry (IC) or predictive formula if IC not feasible.

Percentage of nutritional and non-nutritional calories received per day and at the end of the ICU stay with a target defined by IC or predictive formula if IC not feasible.

Time frame: During ICU stay, up to 15 days post-randomization.

Energy debt: difference between the defined energy target (80-100%, defined by IC or predictive formula) and the energy received (nutritional and non-nutritional)

Per day and at the end of the ICU stay.

Time frame: During ICU stay, up to 15 days post-randomization.

Protein debt: difference between the defined protein target (1.3 g/kg of body weight/day) and the proteins received

Per day and at the end of the ICU stay.

Time frame: During ICU stay, up to 15 days post-randomization.

Prediction of BG

Prediction error as a function of time elapsed since last BG measurement, per patient, per cohort.measurement (per patient and per cohort).

Time frame: During ICU stay, up to 15 days post-randomization.