Efficacy and safety of glucose control with Space GlucoseControl in the medical intensive care unit--an open clinical investigation

What, why and how to monitor blood glucose in critically ill patients

Critically ill patients are prone to high glycemic variations irrespective of their diabetes status. This mandates frequent blood glucose (BG) monitoring and regulation of insulin therapy. Even though the most commonly employed capillary BG monitoring is convenient and rapid, it is inaccurate and prone to high bias, overestimating BG levels in critically ill patients. The targets for BG levels have also varied in the past few years ranging from tight glucose control to a more liberal approach. Each of these has its own fallacies, while tight control increases risk of hypoglycemia, liberal BG targets make the patients prone to hyperglycemia. Moreover, the recent evidence suggests that BG indices, such as glycemic variability and time in target range, may also affect patient outcomes. In this review, we highlight the nuances associated with BG monitoring, including the various indices required to be monitored, BG targets and recent advances in BG monitoring in critically ill patients.

Artificial intelligence in critically ill diabetic patients: current status and future prospects

Recent years have witnessed increasing numbers of artificial intelligence (AI) based applications and devices being tested and approved for medical care. Diabetes is arguably the most common chronic disorder worldwide and AI is now being used for making an early diagnosis, to predict and diagnose early complications, increase adherence to therapy, and even motivate patients to manage diabetes and maintain glycemic control. However, these AI applications have largely been tested in non-critically ill patients and aid in managing chronic problems. Intensive care units (ICUs) have a dynamic environment generating huge data, which AI can extract and organize simultaneously, thus analysing many variables for diagnostic and/or therapeutic purposes in order to predict outcomes of interest. Even non-diabetic ICU patients are at risk of developing hypo or hyperglycemia, complicating their ICU course and affecting outcomes. In addition, to maintain glycemic control frequent blood sampling and insulin dose adjustments are required, increasing nursing workload and chances of error. AI has the potential to improve glycemic control while reducing the nursing workload and errors. Continuous glucose monitoring (CGM) devices, which are Food and Drug Administration (FDA) approved for use in non-critically ill patients, are now being recommended for use in specific ICU populations with increased accuracy. AI based devices including artificial pancreas and CGM regulated insulin infusion system have shown promise as comprehensive glycemic control solutions in critically ill patients. Even though many of these AI applications have shown potential, these devices need to be tested in larger number of ICU patients, have wider availability, show favorable cost-benefit ratio and be amenable for easy integration into the existing healthcare systems, before they become acceptable to ICU physicians for routine use.

The development of a glucose prediction model in critically ill patients

PURPOSE

The aim of the current study is to develop a prediction model for glucose levels applicable for all patients admitted to the ICU with an expected ICU stay of at least 24 h. This model will be incorporated in a closed-loop glucose system to continuously and automatically control glucose values.

METHODS

Data from a previous single-center randomized controlled study was used. All patients received a FreeStyle Navigator II subcutaneous CGM system from Abbott during their ICU stay. The total dataset was randomly divided into a training set and a validation set. A glucose prediction model was developed based on historical glucose data. Accuracy of the prediction model was determined using the Mean Squared Difference (MSD), the Mean Absolute Difference (MAD) and a Clarke Error Grid (CEG).

RESULTS

The dataset included 94 ICU patients with a total of 134,673 glucose measurements points that were used for modelling. MSD was 0.410 ± 0.495 for the model, the MAD was 5.19 ± 2.63 and in the CEG 99.8% of the data points were in the clinically acceptable regions.

CONCLUSION

In this study a glucose prediction model for ICU patients is developed. This study shows that it is possible to accurately predict a patient's glucose 30 min ahead based on historical glucose data. This is the first step in the development of a closed-loop glucose system.

Preventing undesirable behavior of intelligent machines

Intelligent machines using machine learning algorithms are ubiquitous, ranging from simple data analysis and pattern recognition tools to complex systems that achieve superhuman performance on various tasks. Ensuring that they do not exhibit undesirable behavior-that they do not, for example, cause harm to humans-is therefore a pressing problem. We propose a general and flexible framework for designing machine learning algorithms. This framework simplifies the problem of specifying and regulating undesirable behavior. To show the viability of this framework, we used it to create machine learning algorithms that precluded the dangerous behavior caused by standard machine learning algorithms in our experiments. Our framework for designing machine learning algorithms simplifies the safe and responsible application of machine learning.

Feasibility of Blood Glucose Management Using Intra-Arterial Glucose Monitoring in Combination with an Automated Insulin Titration Algorithm in Critically Ill Patients

Background: This two-center pilot study combined for the first time an intra-arterial glucose sensor with a decision support system for insulin dosing (SGCplus system) in critically ill patients with hyperglycemia. Methods: Twenty-two patients who were equipped with an arterial line and required iv insulin therapy were managed by the SGCplus system during their medical treatment at the intensive care unit. Results: Time to target was 111 ± 195 min (80-150 mg/dL) and 135 ± 267 min (100-160 mg/dL) in the lower and higher glucose target group. Mean blood glucose (BG) was 142 ± 32 mg/dL with seven BG values <70 mg/dL. Mean daily insulin dose was 62 ± 38 U and mean daily carbohydrate intake 148 ± 50 g/day (enteral nutrition) and 102 ± 58 g/day (parenteral nutrition). Acceptance of SGCplus suggestions was high (93%). Conclusions: The SGCplus system can be safely applied in critically ill patients with hyperglycemia and enables good glycemic control.

Pre-operative optimisation of the surgical patient with diagnosed and undiagnosed diabetes: a practical review

Peri-operative hyperglycaemia, whether the cause is known diabetes, undiagnosed diabetes or stress hyperglycaemia, is a risk factor for harm, increased length of stay and death. There is increasing evidence that peri-operative hyperglycaemia is a modifiable risk factor, and many of the interventions required to improve the outcome of surgery must be instituted before the actual surgical admission. These interventions depend on communication and collaboration within the multidisciplinary team along each stage of the patient journey to ensure that integration of care occurs across the whole of the patient-centred care pathway.

Nutrition delivery, workload and performance in a model-based ICU glycaemic control system

BACKGROUND AND OBJECTIVE

Hyperglycaemia is commonplace in the adult intensive care unit (ICU), and has been associated with increased morbidity and mortality. Effective glycaemic control (GC) can reduce morbidity and mortality, but has proven difficult. STAR is a model-based GC protocol that uniquely maintains normoglycaemia by changing both insulin and nutrition interventions, and has been proven effective in controlling blood glucose (BG) in the ICU. However, most ICU GC protocols only change insulin interventions, making the variable nutrition aspect of STAR less clinically desirable. This paper compares the performance of STAR modulating only insulin, with three simpler alternative nutrition protocols in clinically evaluated virtual trials.

METHODS

Alternative nutrition protocols are fixed nutrition rate (100% caloric goal), CB (Cahill et al. best) stepped nutrition rate (60%, 80% and 100% caloric goal for the first 3 days of GC, and 100% thereafter) and SLQ (STAR lower quartile) stepped nutrition rate (65%, 75% and 85% caloric goal for the first 3 days of GC, and 85% thereafter). Each nutrition protocol is simulated with the STAR insulin protocol on a 221 patient virtual cohort, and GC performance, safety and total intervention workload are assessed.

RESULTS

All alternative nutrition protocols considerably reduced total intervention workload (14.6-19.8%) due to reduced numbers of nutrition changes. However, only the stepped nutrition protocols achieved similar GC performance to the current variable nutrition protocol. Of the two stepped nutrition protocols, the SLQ nutrition protocol also improved GC safety, almost halving the number of severe hypoglycaemic cases (5 vs. 9, P = 0.42).

CONCLUSIONS

Overall, the SLQ nutrition protocol was the best alternative to the current variable nutrition protocol, but either stepped nutrition protocol could be adapted by STAR to reduce workload and make it more clinically acceptable, while maintaining its proven performance and safety.

Interpretation of Retrospective BG Measurements

BACKGROUND

This study investigates blood glucose (BG) measurement interpolation techniques to represent intermediate BG dynamics, and the effect resampling of retrospective BG data has on key glycemic control (GC) performance results. GC protocols in the ICU have varying BG measurement intervals ranging from 0.5 to 4 hours. Sparse data pose problems, particularly in comparing GC performance or model fitting, and thus interpolation is required.

METHODS

Retrospective data from SPRINT in Christchurch Hospital Intensive Care Unit (ICU) (2005-2007) were used to analyze several interpolation techniques. Piecewise linear, spline, and cubic interpolation functions, which force interpolation through measured data, as well as 1st and 2nd Order B-spline basis functions, are used to identify the interpolated trace. Dense data were thinned to increase sparsity and obtain measurements (Hidden Measurements) for comparison after interpolation. Performance is assessed based on error in capturing hidden measurements. Finally, the effect of minutely versus hourly sampling of the interpolated trace on key GC performance statistics was investigated using retrospective data received from STAR GC in Christchurch Hospital ICU, New Zealand (2011-2015).

RESULTS

All of the piecewise functions performed considerably better than the fitted interpolation functions. Linear piecewise interpolation performed the best having a mean RMSE 0.39 mmol/L, within 2 standard deviations of the BG sensor error. Minutely sampled BG resulted in significantly different key GC performance values when compared to raw sparse BG measurements.

CONCLUSION

Linear piecewise interpolation provides the best estimate of intermediate BG dynamics and all analyses comparing GC protocol performance should use minutely linearly interpolated BG data.

Comparison of Space Glucose Control and Routine Glucose Management Protocol for Glycemic Control in Critically Ill Patients: A Prospective, Randomized Clinical Study

Background:

The Space Glucose Control (SGC) system is a computer-assisted device combining infusion pumps with the enhanced Model Predictive Control algorithm to achieve the target blood glucose (BG) level safely. The objective of this study was to evaluate the efficacy and safety of glycemic control by SGC with customized BG target range of 5.8–8.9 mmol/L in the critically ill patients.

Methods:

It is a randomized controlled trial of seventy critically ill patients with mechanical ventilation and hyperglycemia (BG ≥ 9.0 mmol/L). Thirty-six patients in the SGC group and 34 in the routine glucose management group were observed for three consecutive days. Target BG for both groups was 5.8–8.9 mmol/L. The primary outcome was the percentage time in the target range.

Results:

The percentage time within BG target range in the SGC group (69 ± 15%) was significantly higher than in the routine management group (52 ± 24%; P < 0.01). No measurement was ≤2.2 mmol/L, and there was only one episode of hypoglycemia (2.3–3.3 mmol/L) in each group. The average BG was significantly lower in the SGC group (7.8 ± 0.7 mmol/L) than in the routine management group (9.1 ± 1.6 mmol/L, P < 0.001). Target BG level was reached earlier in the SGC group than routine management group (2.5 ± 2.9 vs. 12.1 ± 15.3 h, P = 0.001). However, the SGC group performed worse for daily insulin requirement (59.8 ± 39.3 vs. 28.4 ± 36.7 U, P = 0.001) and sampling interval (2.0 ± 0.5 vs. 3.7 ± 0.5 h, P < 0.001) than the routine management group did. Multiple linear regression showed that the intervention group remained a significant individual predictor (P < 0.001) of the percentage time in target range.

Conclusions:

The SGC system, with a BG target of 5.8–8.9 mmol/L, resulted in effective and reliable glycemic control with few hypoglycemic episodes in critically ill patients with mechanical ventilation and hyperglycemia. However, the workload was increased.

Trial Registration:

http://www.clinicaltrials.gov, NCT 02491346; https://www.clinicaltrials.gov/ct2/show/NCT02491346?term=NCT02491346&cond=Hyperglycemia&cntry1=ES%3ACN&rank=1.

Indicators for best practices in glycemic control in the intensive care unit

Abstract Objective: To understand the perception of the nursing team' regarding the management of the intensive glycemic control protocol in order to obtain better practices in glycemic control in the Intensive Care Unit of a University Hospital. Method: A qualitative, convergent care study carried out in an Intensive Care Unit of a university hospital. The data were collected through interviews and workshops and analyzed through thematic analysis. Thirty nursing professionals participated in the study. Results: The importance of the glycemic control protocol which standardizes and guides care was reported by the participants, however they indicated that the used protocol is confusing, difficult to understand and does not include some important guidelines. Restructuring was recommended by adding aspects such as: the desired glycemic value, spaces between glycaemia recording, diet and others; as well as training for its application. Conclusion: The participants recognized the weaknesses of the protocol, and reaffirmed the potentialities of this instrument and defended permanent education as a fundamental factor for the best practices in intensive care.

Nutrition delivery of a model-based ICU glycaemic control system

BACKGROUND

Hyperglycaemia is commonplace in the adult intensive care unit (ICU), associated with increased morbidity and mortality. Effective glycaemic control (GC) can reduce morbidity and mortality, but has proven difficult. STAR is a proven, effective model-based ICU GC protocol that uniquely maintains normo-glycaemia by changing both insulin and nutrition interventions to maximise nutrition in the context of GC in the 4.4-8.0 mmol/L range. Hence, the level of nutrition it provides is a time-varying estimate of the patient-specific ability to take up glucose.

METHODS

First, the clinical provision of nutrition by STAR in Christchurch Hospital, New Zealand (N = 221 Patients) is evaluated versus other ICUs, based on the Cahill et al. survey of 158 ICUs. Second, the inter- and intra- patient variation of nutrition delivery with STAR is analysed. Nutrition rates are in terms of percentage of caloric goal achieved.

RESULTS

Mean nutrition rates clinically achieved by STAR were significantly higher than the mean and best ICU surveyed, for the first 3 days of ICU stay. There was large inter-patient variation in nutrition rates achieved per day, which reduced overtime as patient-specific metabolic state stabilised. Median intra-patient variation was 12.9%; however, the interquartile range of the mean per-patient nutrition rates achieved was 74.3-98.2%, suggesting patients do not deviate much from their mean patient-specific nutrition rate. Thus, the ability to tolerate glucose intake varies significantly between, rather than within, patients.

CONCLUSIONS

Overall, STAR's protocol-driven changes in nutrition rate provide higher nutrition rates to hyperglycaemic patients than those of 158 ICUs from 20 countries. There is significant inter-patient variability between patients to tolerate and uptake glucose, where intra-patient variability over stay is much lower. Thus, a best nutrition rate is likely patient specific for patients requiring GC. More importantly, these overall outcomes show high nutrition delivery and safe, effective GC are not exclusive and that restricting nutrition for GC does not limit overall nutritional intake compared to other ICUs.

Safety, efficacy and clinical generalization of the STAR protocol: a retrospective analysis

Background

The changes in metabolic pathways and metabolites due to critical illness result in a highly complex and dynamic metabolic state, making safe, effective management of hyperglycemia and hypoglycemia difficult. In addition, clinical practices can vary significantly, thus making GC protocols difficult to generalize across units.The aim of this study was to provide a retrospective analysis of the safety, performance and workload of the stochastic targeted (STAR) glycemic control (GC) protocol to demonstrate that patient-specific, safe, effective GC is possible with the STAR protocol and that it is also generalizable across/over different units and clinical practices.

Methods

Retrospective analysis of STAR GC in the Christchurch Hospital Intensive Care Unit (ICU), New Zealand (267 patients), and the Gyula Hospital, Hungary (47 patients), is analyzed (2011–2015). STAR Christchurch (BG target 4.4–8.0 mmol/L) is also compared to the Specialized Relative Insulin and Nutrition Tables (SPRINT) protocol (BG target 4.4–6.1 mmol/L) implemented in the Christchurch Hospital ICU, New Zealand (292 patients, 2005–2007). Cohort mortality, effectiveness and safety of glycemic control and nutrition delivered are compared using nonparametric statistics.

Results

Both STAR implementations and SPRINT resulted in over 86 % of time per episode in the blood glucose (BG) band of 4.4–8.0 mmol/L. Patients treated using STAR in Christchurch ICU spent 36.7 % less time on protocol and were fed significantly more than those treated with SPRINT (73 vs. 86 % of caloric target). The results from STAR in both Christchurch and Gyula were very similar, with the BG distributions being almost identical. STAR provided safe GC with very few patients experiencing severe hypoglycemia (BG < 2.2 mmol/L, <5 patients, 1.5 %).

Conclusions

STAR outperformed its predecessor, SPRINT, by providing higher nutrition and equally safe, effective control for all the days of patient stay, while lowering the number of measurements and interventions required. The STAR protocol has the ability to deliver high performance and high safety across patient types, time, clinical practice culture (Christchurch and Gyula) and clinical resources.

Continuous Glucose Monitoring in Patients Undergoing Extracorporeal Ventricular Assist Therapy

Background

Dysregulations of blood glucose (BG) are associated with adverse outcome in critical illness; controlling BG to target appears to improve outcome. Since BG-control is challenging in daily intensive care practice BG-control remains poor especially in patients with rapidly fluctuating BG. To improve BG-control and to avoid deleterious hypoglycemia, automated online-measurement tools are advocated. We thus evaluated the point-accuracy of the subcutaneous Sentrino^® Continuous Glucose Monitoring System (CGM, Medtronic Diabetes, Northridge, California) in patients undergoing extracorporeal cardiac life support (ECLS) for cardiogenic shock.

Methods

Management of BG was performed according to institute’s standard aiming at BG-levels between 100–145 mg/dl. CGM-values were recorded without taking measures into therapeutic account. Point-accuracy in comparison to intermittent BG-measurement by the ABL-blood-gas analyzer was determined.

Results

CGM (n = 25 patients) correlated significantly with ABL-values (r = 0.733, p<0.001). Mean error from standard was 15.0 mg/dl (11.9%). 44.2% of the readings were outside a 15% range around ABL-values. In one of 635 paired data-points, ABL revealed hypoglycemia (BG 32 mg/dl) whereas CGM did not show hypoglycemic values (132mg/dl).

Conclusions

CGM reveals minimally invasive BG-values in critically ill adults with dynamically impaired tissue perfusion. Because of potential deviations from standard, CGM-readings must be interpreted with caution in specific ICU-populations.

Space GlucoseControl system for blood glucose control in intensive care patients--a European multicentre observational study

BACKGROUND

Glycaemia control (GC) remains an important therapeutic goal in critically ill patients. The enhanced Model Predictive Control (eMPC) algorithm, which models the behaviour of blood glucose (BG) and insulin sensitivity in individual ICU patients with variable blood samples, is an effective, clinically proven computer based protocol successfully tested at multiple institutions on medical and surgical patients with different nutritional protocols. eMPC has been integrated into the B.Braun Space GlucoseControl system (SGC), which allows direct data communication between pumps and microprocessor. The present study was undertaken to assess the clinical performance and safety of the SGC for glycaemia control in critically ill patients under routine conditions in different ICU settings and with various nutritional protocols.

METHODS

The study endpoints were the percentage of time the BG was within the target range 4.4 - 8.3 mmol.l(-1), the frequency of hypoglycaemic episodes, adherence to the advice of the SGC and BG measurement intervals. BG was monitored, and insulin was given as a continuous infusion according to the advice of the SGC. Nutritional management (enteral, parenteral or both) was carried out at the discretion of each centre.

RESULTS

17 centres from 9 European countries included a total of 508 patients, the median study time was 2.9 (1.9-6.1) days. The median (IQR) time-in-target was 83.0 (68.7-93.1) % of time with the mean proposed measurement interval 2.0 ± 0.5 hours. 99.6% of the SGC advices on insulin infusion rate were accepted by the user. Only 4 episodes (0.01% of all BG measurements) of severe hypoglycaemia <2.2 mmol.l(-1) in 4 patients occurred (0.8%; 95% CI 0.02-1.6%).

CONCLUSION

Under routine conditions and under different nutritional protocols the Space GlucoseControl system with integrated eMPC algorithm has exhibited its suitability for glycaemia control in critically ill patients.

TRIAL REGISTRATION

ClinicalTrials.gov NCT01523665.

Perioperative glycemic control with a computerized algorithm versus conventional glycemic control in cardiac surgical patients undergoing cardiopulmonary bypass with blood cardioplegia

OBJECTIVE

In critical illness, hypoglycemia and hyperglycemia seem to influence outcome. While hypoglycemia can lead to organ dysfunction, hyperglycemia can lead to surgical site infections (SSI). In cardiac surgery, the use of blood cardioplegia is associated with high blood glucose levels. A computer-based algorithm (CBA) for guiding insulin towards normoglycemia might be beneficial. The authors' primary study end-point was the duration in a predefined blood glucose target range of 80 mg/dL to 150 mg/dL. Patients with conventional therapy served as controls.

DESIGN

Prospective, randomized trial.

SETTING

University hospital.

PARTICIPANTS

Seventy-five patients.

INTERVENTIONS

The start of therapy was the beginning of cardiopulmonary bypass. Group A: Therapy with CBA and measurement of blood glucose every 30 minutes. Group B: Measurement of blood glucose every 15 minutes using the identical CBA. Group C: Conventional therapy using a fixed insulin dosing scheme. End of therapy was defined as discharge from ICU.

MEASUREMENT AND MAIN RESULTS

Glucose administration during cardioplegia did not differ between groups (A: 33 ± 12 g; B: 32 ± 12 g; C: 38 ± 20 g). Glucose levels in groups A and B stayed significantly longer in the target interval compared with group C (A: 75 ± 20%; B: 72 ± 19%; C: 50 ± 34%, p < 0.01 n = 25, respectively). There were no significant differences regarding ICU stay and SSI rates.

CONCLUSIONS

Early computer-based insulin therapy allows practitioners to better achieve normoglycemia in patients undergoing major cardiac surgery with the use of blood cardioplegia.

Glucose control in intensive care: usability, efficacy and safety of Space GlucoseControl in two medical European intensive care units

BACKGROUND

The Space GlucoseControl system (SGC) is a nurse-driven, computer-assisted device for glycemic control combining infusion pumps with the enhanced Model Predictive Control algorithm (B. Braun, Melsungen, Germany). We aimed to investigate the performance of the SGC in medical critically ill patients.

METHODS

Two open clinical investigations in tertiary centers in Graz, Austria and Zurich, Switzerland were performed. Efficacy was assessed by percentage of time within the target range (4.4-8.3 mmol/L; primary end point), mean blood glucose, and sampling interval. Safety was assessed by the number of hypoglycemic episodes (≤2.2 mmol/L) and the percentage of time spent below this cutoff level. Usability was analyzed with a standardized questionnaire given to involved nursing staff after the trial.

RESULTS

Forty medical critically ill patients (age, 62 ± 15 years; body mass index, 30.0 ± 8.9 kg/m2; APACHE II score, 24.8 ± 5.4; 27 males; 8 with diabetes) were included for a period of 6.5 ± 3.7 days (n = 20 in each center). The primary endpoint (time in target range 4.4 to 8.3 mmol/l) was reached in 88.3% ± 9.3 of the time and mean arterial blood glucose was 6.7 ± 0.4 mmol/l. The sampling interval was 2.2 ± 0.4 hours. The mean daily insulin dose was 87.2 ± 64.6 IU. The adherence to the given insulin dose advice was high (98.2%). While the percentage of time spent in a moderately hypoglycemic range (2.2 to 3.3 mmol/L) was low (0.07 ± 0.26% of the time), one severe hypoglycemic episode (<2.2 mmol/L) occurred (2.5% of patients or 0.03% of glucose readings).

CONCLUSIONS

SGC is a safe and efficient method to control blood glucose in critically ill patients as assessed in two European medical intensive care units.

The future is now: software-guided intensive insulin therapy in the critically ill

Since the development of intensive insulin therapy for the critically ill adult, tight glycemic control (TGC) has become increasingly complicated to apply and achieve. Software-guided (SG) algorithms for insulin dosing represent a new method to achieve euglycemia in critical illness. We provide an overview of the state of SG TGC with an eye to the future. The current milieu is disorganized, with little research that incorporates newer variables of dysglycemia, such as glycemic variability. To develop and implement better algorithms, scientists, programmers, and clinicians need to standardize measurements and variables.