The importance of glucose control in the intensive care unit (ICU) setting was first demonstrated by Furnary 1 and later confirmed by Van den Berghe 2 in a large, prospective, randomized control study. However, almost all of the clinical studies that have been published since then have struggled to achieve superior glucose control while avoiding the glucose metrics that actually increase mortality rates-hypoglycemia, hyperglycemia, variability, and low time in range.3-6 A Japanese study that utilized a closed loop glucose control system that minimized hypoglycemia while at the same time achieving a high percentage of time in range showed the true promise of effective glucose control.7 To achieve tight glucose control while at the same time minimizing glucose metrics that increase mortality rates, a closed loop system will need to be created. This system should mimic the workings of the native system which have been previously reviewed. 8 To date, most attempts at creating a closed loop glucose control system for use in the ICU setting have utilized the standard engineering control techniques of proportional integral derivative (PID) or model predictive control (MPC). [9][10][11][12] PID controllers typically control to a set point and are challenged by the fact they rely solely on insulin for control, thus they do not take advantage of the counter-regulatory effects of intravenous (IV) glucose. MPC controllers rely on a complex mathematical model of the glucose-insulin system. MPC model parameters are refitted every 1 to 4 hours, in an iterative fashion, based on the difference between the predicted and measured glucose level. Control recommendations are made from the resulting best fit model. MPC controllers can be designed to control to a set point or to a desired range and Background: Effective glucose control in the intensive care unit (ICU) setting has the potential to decrease morbidity and mortality rates which should in turn lead to decreased health care expenditures. Current ICU-based glucose controllers are mathematically derived, and tend to be based on proportional integral derivative (PID) or model predictive control (MPC). Artificial intelligence (AI)-based closed loop glucose controllers may have the ability to achieve control that improves on the results achieved by either PID or MPC controllers. Method:We conducted an in silico analysis of an AI-based glucose controller designed for use in the ICU setting. This controller was tested using a mathematical model of the ICU patient's glucose-insulin system. A total of 126 000 unique 5-day simulations were carried out, resulting in 107 million glucose values for analysis. Results:For the 7 control ranges tested, with a sensor error of ±10%, the following average results were achieved: (1) time in control range, 94.2%, (2) time in range 70-140 mg/dl, 97.8%, (3) time in hyperglycemic range (>140 mg/dl), 2.1%, and (4) time in hypoglycemic range (<70 mg/dl), 0.09%. In addition, the average coefficient of variation (CV) was 11.1%. Conclusions:This in sili...

The novel scoring metric employed in this study shows promise as a means for evaluating new and existing ICU-based glucose controllers, and it could be used in the future to compare results of glucose control studies in critical care. The IMT AI-based glucose controller demonstrated the most consistent performance results based on this new metric.

Intensive care unit (ICU) patients develop stress induced insulin resistance causing hyperglycemia, large glucose variability and hypoglycemia. These glucose metrics have all been associated with increased rates of morbidity and mortality. The only way to achieve safe glucose control at a lower glucose range (e.g., 4.4-6.6 mmol/L) will be through use of an autonomous closed loop glucose control system (artificial pancreas). Our goal with the present study was to assess the safety and performance of an artificial pancreas system, composed of the EIRUS (Maquet Critical Care AB) continuous glucose monitor (CGM) and novel artificial intelligence-based glucose control software, in a swine model using unannounced hypo-and hyperglycemia challenges. Fourteen piglets (6 control, 8 treated) underwent sequential unannounced hypoglycemic and hyperglycemic challenges with 3 IU of NovoRapid and a glucose infusion at 17 mg/kg/min over the course of 5 h. In the Control animals an experienced ICU physician used every 30-min blood glucose values to maintain control to a range of 4.4-9 mmol/L. In the Treated group the artificial pancreas system attempted to maintain blood glucose control to a range of 4.4-6.6 mmol/L. Five of six Control animals and none of eight Treated animals experienced severe hypoglycemia (< 2.22 mmol/L). The area under the curve 3.5 mmol/L was 28.9 (21.1-54.2) for Control and 4.8 (3.1-5.2) for the Treated animals. The total percent time within tight glucose control range, 4.4-6.6 mmol/L, was 32.8% (32.4-47.1) for Controls and 55.4% (52.9-59.4) for Treated (p < 0.034). Data are median and quartiles. The artificial pancreas system abolished severe hypoglycemia and outperformed the experienced ICU physician in avoiding clinically significant hypoglycemic excursions.

Effective glucose control in the Intensive Care Unit (ICU) setting has the potential to lower mortality rates [1], shorten length of stay [2] and decrease overall cost of care [3]. Yet the goal of achieving this control remains elusive due to the limitations of our current open loop methods [4] that still require manual testing of glucose values, entry of the measured value into local or web based glucose control software, and manual adjustment of the intravenous pumps infusing insulin into the ICU patient. In order to improve overall glucose control, ICU care givers will need to be empowered with a closed loop glucose control system.The three main components of a closed loop glucose control system are a glucose sensor(s), control algorithm, and intravenous pump(s). Current intravenous pumps are accurate and reliable enough for a closed loop system, so the two components preventing completion of the system are the glucose sensor(s) and controller. An accurate and reliable glucose sensor array is a must, as trying to control a system without real time knowledge of mission critical sensor data will invariably lead to unacceptable outcomes, as has been seen in the aerospace industry [5]. In fact, the aerospace industry provides an excellent example of how to properly engineer a safe and effective control system, as they routinely build in redundancy of mission critical system components to improve reliability and will also use different methods of measurement to improve overall accuracy.Studies on type I diabetics have shown that a multi-sensor array improves overall accuracy of the system [6]. The following equation can be used as a guide to determine the number of sensors N, needed given a known sensor failure rate p and a desired uptime rate q, where p, q ∈ [0,1]:As can be seen from Table 1 if the closed loop system requirement is to have a complete sensor failure rate limited to less than 10 min over a typical 96 h ICU length of stay, which is a greater than 99.8% uptime rate, and the known sensor failure rate is 1%, then the system requirement is to have two independent glucose sensors in place.Current CE marked blood based glucose sensors designed for use in the ICU setting have already been shown to be both highly accurate and reliable, with an uptime rate that exceeds 99% [7]. In addition, the next generation interstitial continuous glucose monitoring (CGM) system that Dexcom is developing with assistance from Google will not be affected by acetaminophen, making it a potential excellent candidate as the second sensor in a glucose array [8]. The ability to avoid interference from acetaminophen is important as this medication is known to affect the accuracy of current interstitial CGM systems, yet is ubiquitously used in the hospital setting. As sensor reliability is far more important than accuracy when it comes to closed loop glucose control [9], the improved uptime rate of a two sensor array will more than offset the decreased accuracy that will occur by averaging simultaneous blood and interstitial glucos...

Introduction: Intensive care unit (ICU) patients develop stress induced insulin resistance causing hyperglycemia, large glucose variability and hypoglycemia, all of which increase mortality rates. Tight glucose control (TGC) in the ICU is challenging, and usually significantly increases the risk for treatment induced hypoglycemia. The only way to achieve safe TGC will be through use of an artificial pancreas (AP) system. Our goal with the present study was to assess the safety and performance of an AP system, composed of the EIRUS (Getinge AB) continuous glucose monitor (CGM) and novel artificial intelligence-based (AI-based) glucose control software, in a swine model using unannounced hypo- and hyperglycemia challenges. Methods: Fourteen piglets (6 control, 8 treated) underwent sequential unannounced (to the AP system) hypoglycemic and hyperglycemic challenges with 3 U of insulin aspart and a large glucose infusion over the course of 5 hours. In the control animals an ICU physician used every 30-minute arterial blood glucose values to maintain control to a range of 4.4 - 9 mmol/L through use of 20% dextrose for hypoglycemia and insulin aspart for hyperglycemia. In the treated group the AP system attempted to maintain blood glucose control to a range of 4.4 - 6.6 mmol/L in a fully autonomous mode. Results: Five of six control animals and none of eight treated animals experienced severe hypoglycemia (&lt; 2.22 mmol/L). The area under the curve 3.5 mmol/L was 28.9 (21.1 - 54.2) for control and 4.8 (3.1 - 5.2) for the treated animals. The percent time in range 4.4 - 6.6 mmol/L was 32.8 (32.4 - 47.1) for the control and 55.4 (52.9 - 59.4) for the treated animals. After the start of the experiment the control animals underwent 4 (3-5) human interventions to maintain control whereas the treated animals underwent none. Data are median (25-75). Conclusions The AP system abolished severe hypoglycemia and outperformed the ICU physician both in avoiding hypoglycemic excursions and maximizing time in range. Disclosure J. DeJournett: Employee; Self; Ideal Medical Technologies. L. DeJournett: Stock/Shareholder; Self; Ideal Medical Technologies.