A novel, model-based insulin and nutrition delivery controller for glycemic regulation in critically ill patients

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.

Multi-input stochastic prediction of insulin sensitivity for tight glycaemic control using insulin sensitivity and blood glucose data

BACKGROUND

Glycaemic control in the intensive care unit is dependent on effective prediction of patient insulin sensitivity (SI). The stochastic targeted (STAR) controller uses a 2D stochastic model for prediction, with current SI as an input and future SI as an output.

METHODS

This paper develops an extension of the STAR 2D stochastic model into 3D by adding blood glucose (G) as an input. The performance of the 2D and 3D stochastic models is compared over a retrospective cohort of 65,269 data points across 1525 patients.

RESULTS

Under five-fold cross-validation, the 3D model was found to better match the expected potion of data points within, above and below various credible intervals, suggesting it provided a better representation of the underlying probability field. The 3D model was also found to provide an 18.1% narrower 90% credible interval on average, and a narrower 90% credible interval in 96.4% of cases, suggesting it provided more accurate predictions of future SI. Additionally, the 3D stochastic model was found to avoid the undesirable tendency of the 2D model to overestimate SI for patients with high G, and underestimate SI for patients with low G.

CONCLUSIONS

Overall, the 3D stochastic model is shown to provide clear potential benefits over the 2D model for minimal clinical cost or effort, though further exploration into whether these improvements in SI prediction translate into improved clinical outcomes is required.

Hypoglycemia Prevention by Algorithm Design During Intravenous Insulin Infusion

PURPOSE OF REVIEW

This review examines algorithm design features that may reduce risk for hypoglycemia while preserving glycemic control during intravenous insulin infusion. We focus principally upon algorithms in which the assignment of the insulin infusion rate (IR) depends upon maintenance rate of insulin infusion (MR) or a multiplier.

RECENT FINDINGS

Design features that may mitigate risk for hypoglycemia include use of a mid-protocol bolus feature and establishment of a low BG threshold for temporary interruption of infusion. Computer-guided dosing may improve target attainment without exacerbating risk for hypoglycemia. Column assignment (MR) within a tabular user-interpreted algorithm or multiplier may be specified initially according to patient characteristics and medical condition with revision during treatment based on patient response. We hypothesize that a strictly increasing sigmoidal relationship between MR-dependent IR and BG may reduce risk for hypoglycemia, in comparison to a linear relationship between multiplier-dependent IR and BG. Guidelines are needed that curb excessive up-titration of MR and recommend periodic pre-emptive trials of MR reduction. Future research should foster development of recommendations for "protocol maxima" of IR appropriate to patient condition.

An Integrative multi-lineage model of variation in leukopoiesis and acute myelogenous leukemia

Background

Acute myelogenous leukemia (AML) progresses uniquely in each patient. However, patients are typically treated with the same types of chemotherapy, despite biological differences that lead to differential responses to treatment.

Results

Here we present a multi-lineage multi-compartment model of the hematopoietic system that captures patient-to-patient variation in both the concentration and rates of change of hematopoietic cell populations. By constraining the model against clinical hematopoietic cell recovery data derived from patients who have received induction chemotherapy, we identified trends for parameters that must be met by the model; for example, the mitosis rates and the probability of self-renewal of progenitor cells are inversely related. Within the data-consistent models, we found 22,796 parameter sets that meet chemotherapy response criteria. Simulations of these parameter sets display diverse dynamics in the cell populations. To identify large trends in these model outputs, we clustered the simulated cell population dynamics using k-means clustering and identified thirteen ‘representative patient’ dynamics. In each of these patient clusters, we simulated AML and found that clusters with the greatest mitotic capacity experience clinical cancer outcomes more likely to lead to shorter survival times. Conversely, other parameters, including lower death rates or mobilization rates, did not correlate with survival times.

Conclusions

Using the multi-lineage model of hematopoiesis, we have identified several key features that determine leukocyte homeostasis, including self-renewal probabilities and mitosis rates, but not mobilization rates. Other influential parameters that regulate AML model behavior are responses to cytokines/growth factors produced in peripheral blood that target the probability of self-renewal of neutrophil progenitors. Finally, our model predicts that the mitosis rate of cancer is the most predictive parameter for survival time, followed closely by parameters that affect the self-renewal of cancer stem cells; most current therapies target mitosis rate, but based on our results, we propose that additional therapeutic targeting of self-renewal of cancer stem cells will lead to even higher survival rates.

Electronic supplementary material

The online version of this article (doi:10.1186/s12918-017-0469-2) contains supplementary material, which is available to authorized users.

Continuous versus bolus tube feeds: Does the modality affect glycemic variability, tube feeding volume, caloric intake, or insulin utilization?

Introduction:

Enteral nutrition (EN) is very important to optimizing outcomes in critical illness. Debate exists regarding the best strategy for enteral tube feeding (TF), with concerns that bolus TF (BTF) may increase glycemic variability (GV) but result in fewer nutritional interruptions than continuous TF (CTF). This study examines if there is a difference in GV, insulin usage, TF volume, and caloric delivery among intensive care patients receiving BTF versus CTF. We hypothesize that there are no significant differences between CTF and BTF when comparing the above parameters.

Materials and Methods:

Prospective, randomized pilot study of critically ill adult patients undergoing percutaneous endoscopic gastrostomy (PEG) placement for EN was performed between March 1, 2012 and May 15, 2014. Patients were randomized to BTF or CTF. Glucose values, insulin use, TF volume, and calories administered were recorded. Data were organized into 12-h epochs for statistical analyses and GV determination. In addition, time to ≥80% nutritional delivery goal, demographics, Acute Physiology and Chronic Health Evaluation II scores, and TF interruptions were examined. When performing BTF versus CTF assessments, continuous parameters were compared using Mann–Whitney U-test or repeated measures t-test, as appropriate. Categorical data were analyzed using Fisher's exact test.

Results:

No significant demographic or physiologic differences between the CTF (n = 24) and BTF (n = 26) groups were seen. The immediate post-PEG 12-h epoch showed significantly lower GV and median TF volume for patients in the CTF group. All subsequent epochs (up to 18 days post-PEG) showed no differences in GV, insulin use, TF volume, or caloric intake. Insulin use for both groups increased when comparing the first 24 h post-PEG values to measurements from day 8. There were no differences in TF interruptions, time to ≥80% nutritional delivery goal, or hypoglycemic episodes.

Conclusions:

This study demonstrated no clinically relevant differences in GV, insulin use, TF volume or caloric intake between BTF and CTF groups. Despite some shortcomings, our data suggest that providers should not feel limited to BTF or CTF because of concerns for GV, time to goal nutrition, insulin use, or caloric intake, and should consider other factors such as resource utilization, ease of administration, and/or institutional/patient characteristics.

Blood glucose measurement in the intensive care unit: what is the best method?

Abnormal glucose measurements are common among intensive care unit (ICU) patients for numerous reasons and hypoglycemia is especially dangerous because these patients are often sedated and unable to relate the associated symptoms. Additionally, wide swings in blood glucose have been closely tied to increased mortality. Therefore, accurate and timely glucose measurement in this population is critical. Clinicians have several choices available to assess blood glucose values in the ICU, including central laboratory devices, blood gas analyzers, and point-of-care meters. In this review, the method of glucose measurement will be reviewed for each device, and the important characteristics, including accuracy, cost, speed of result, and sample volume, will be reviewed, specifically as these are used in the ICU environment. Following evaluation of the individual measurement devices and after considering the many features of each, recommendations are made for optimal ICU glucose determination.

Pilot study of the SPRINT glycemic control protocol in a Hungarian medical intensive care unit

INTRODUCTION

Stress-induced hyperglycemia increases morbidity and mortality. Tight control can reduce mortality but has proven difficult to achieve. The SPRINT (Specialized Relative Insulin and Nutrition Tables) protocol is the only protocol that reduced both mortality and hypoglycemia by modulating both insulin and nutrition, but it has not been tested in independent hospitals.

METHODS

SPRINT was used for 12 adult intensive care unit patients (949 h) at Kálmán Pándy Hospital (Gyula, Hungary) as a clinical practice assessment. Insulin recommendations (0-6 U/h) were administered via constant infusion rather than bolus delivery. Nutrition was administered per local standard protocol, weaning parenteral to enteral nutrition, but was modulated per SPRINT recommendations. Measurement was every 1 to 2 h, per protocol. Glycemic performance is assessed by percentage of blood glucose (BG) measurements in glycemic bands for the cohort and per patient. Safety from hypoglycemia is assessed by numbers of patients with BG < 2.2 (severe) and %BG < 3.0 and < 4.0 mmol/liter (moderate and light). Clinical effort is assessed by measurements per day. Results are median (interquartile range).

RESULTS

There were 742 measurements over 1088 h of control (16.4 measurements/day), which is similar to clinical SPRINT results (16.2/day). Per-patient hours of control were 65 (50-95) h. Initial per-patient BG was 10.5 (7.9-11.2) mmol/liter. All patients (100%) reached 6.1 mmol/liter. Cohort BG was 6.3 (5.5-7.5) mmol/liter, with 42.2%, 65.1% and 77.6% of BG in the 4.0-6.1, 4.0-7.0, and 4.0-8.0 mmol/liter bands. Per-patient, median percentage time in these bands was 40.2 (26.7-51.5)%, 62.5 (46.0-75.7)%, and 74.7 (61.6.8-87.8)%, respectively. No patients had BG < 2.2 mmol/liter, and the %BG < 4.0 mmol/liter was 1.9%. These results were achieved using 3.0 (3.0-5.0) U/h of insulin with 7.4 (4.4-10.2) g/h of dextrose administration (all sources) for the cohort. Per-patient median insulin administration was 3.0 (3.0-3.0) U/h and 7.1 (3.4-9.6) g/h dextrose. Higher carbohydrate nutrition formulas than were used in SPRINT are offset by slightly higher insulin administration in this study.

CONCLUSIONS

The glycemic performance shows that using the SPRINT protocol to guide insulin infusions and nutrition administration provided very good glycemic control in initial pilot testing, with no severe hypoglycemia. The overall design of the protocol was able to be generalized with good compliance and outcomes across geographically distinct clinical units, patients, and clinical practice.

Variability of insulin sensitivity during the first 4 days of critical illness: implications for tight glycemic control

Background

Effective tight glycemic control (TGC) can improve outcomes in critical care patients, but it is difficult to achieve consistently. Insulin sensitivity defines the metabolic balance between insulin concentration and insulin-mediated glucose disposal. Hence, variability of insulin sensitivity can cause variable glycemia. This study quantifies and compares the daily evolution of insulin sensitivity level and variability for critical care patients receiving TGC.

Methods

This is a retrospective analysis of data from the SPRINT TGC study involving patients admitted to a mixed medical-surgical ICU between August 2005 and May 2007. Only patients who commenced TGC within 12 hours of ICU admission and spent at least 24 hours on the SPRINT protocol were included (N = 164). Model-based insulin sensitivity (SI) was identified each hour. Absolute level and hour-to-hour percent changes in SI were assessed on cohort and per-patient bases. Levels and variability of SI were compared over time on 24-hour and 6-hour timescales for the first 4 days of ICU stay.

Results

Cohort and per-patient median SI levels increased by 34% and 33% (p < 0.001) between days 1 and 2 of ICU stay. Concomitantly, cohort and per-patient SI variability decreased by 32% and 36% (p < 0.001). For 72% of the cohort, median SI on day 2 was higher than on day 1. The day 1–2 results are the only clear, statistically significant trends across both analyses. Analysis of the first 24 hours using 6-hour blocks of SI data showed that most of the improvement in insulin sensitivity level and variability seen between days 1 and 2 occurred during the first 12–18 hours of day 1.

Conclusions

Critically ill patients have significantly lower and more variable insulin sensitivity on day 1 than later in their ICU stay and particularly during the first 12 hours. This rapid improvement is likely due to the decline of counter-regulatory hormones as the acute phase of critical illness progresses. Clinically, these results suggest that while using TGC protocols with patients during their first few days of ICU stay, extra care should be afforded. Increased measurement frequency, higher target glycemic bands, conservative insulin dosing, and modulation of carbohydrate nutrition should be considered to minimize safely the outcome glycemic variability and reduce the risk of hypoglycemia.

Pilot proof of concept clinical trials of Stochastic Targeted (STAR) glycemic control

Introduction

Tight glycemic control (TGC) has shown benefits but has been difficult to achieve consistently. STAR (Stochastic TARgeted) is a flexible, model-based TGC approach directly accounting for intra- and inter- patient variability with a stochastically derived maximum 5% risk of blood glucose (BG) < 4.0 mmol/L. This research assesses the safety, efficacy, and clinical burden of a STAR TGC controller modulating both insulin and nutrition inputs in pilot trials.

Methods

Seven patients covering 660 hours. Insulin and nutrition interventions are given 1-3 hourly as chosen by the nurse to allow them to manage workload. Interventions are calculated by using clinically validated computer models of human metabolism and its variability in critical illness to maximize the overlap of the model-predicted (5-95^th percentile) range of BG outcomes with the 4.0-6.5 mmol/L band while ensuring a maximum 5% risk of BG < 4.0 mmol/L. Carbohydrate intake (all sources) was selected to maximize intake up to 100% of SCCM/ACCP goal (25 kg/kcal/h). Maximum insulin doses and dose changes were limited for safety. Measurements were made with glucometers. Results are compared to those for the SPRINT study, which reduced mortality 25-40% for length of stay ≥3 days. Written informed consent was obtained for all patients, and approval was granted by the NZ Upper South A Regional Ethics Committee.

Results

A total of 402 measurements were taken over 660 hours (~14/day), because nurses showed a preference for 2-hourly measurements. Median [interquartile range, (IQR)] cohort BG was 5.9 mmol/L [5.2-6.8]. Overall, 63.2%, 75.9%, and 89.8% of measurements were in the 4.0-6.5, 4.0-7.0, and 4.0-8.0 mmol/L bands. There were no hypoglycemic events (BG < 2.2 mmol/L), and the minimum BG was 3.5 mmol/L with 4.5% < 4.4 mmol/L. Per patient, the median [IQR] hours of TGC was 92 h [29-113] using 53 [19-62] measurements (median, ~13/day). Median [IQR] results: BG, 5.9 mmol/L [5.8-6.3]; carbohydrate nutrition, 6.8 g/h [5.5-8.7] (~70% goal feed median); insulin, 2.5 U/h [0.1-5.1]. All patients achieved BG < 6.1 mmol/L. These results match or exceed SPRINT and clinical workload is reduced more than 20%.

Conclusions

STAR TGC modulating insulin and nutrition inputs provided very tight control with minimal variability by managing intra- and inter- patient variability. Performance and safety exceed that of SPRINT, which reduced mortality and cost in the Christchurch ICU. The use of glucometers did not appear to impact the quality of TGC. Finally, clinical workload was self-managed and reduced 20% compared with SPRINT.

Physiological modeling, tight glycemic control, and the ICU clinician: what are models and how can they affect practice?

Critically ill patients are highly variable in their response to care and treatment. This variability and the search for improved outcomes have led to a significant increase in the use of protocolized care to reduce variability in care. However, protocolized care does not address the variability of outcome due to inter- and intra-patient variability, both in physiological state, and the response to disease and treatment. This lack of patient-specificity defines the opportunity for patient-specific approaches to diagnosis, care, and patient management, which are complementary to, and fit within, protocolized approaches.Computational models of human physiology offer the potential, with clinical data, to create patient-specific models that capture a patient's physiological status. Such models can provide new insights into patient condition by turning a series of sometimes confusing clinical data into a clear physiological picture. More directly, they can track patient-specific conditions and thus provide new means of diagnosis and opportunities for optimising therapy.This article presents the concept of model-based therapeutics, the use of computational models in clinical medicine and critical care in specific, as well as its potential clinical advantages, in a format designed for the clinical perspective. The review is presented in terms of a series of questions and answers. These aspects directly address questions concerning what makes a model, how it is made patient-specific, what it can be used for, its limitations and, importantly, what constitutes sufficient validation.To provide a concrete foundation, the concepts are presented broadly, but the details are given in terms of a specific case example. Specifically, tight glycemic control (TGC) is an area where inter- and intra-patient variability can dominate the quality of care control and care received from any given protocol. The overall review clearly shows the concept and significant clinical potential of using computational models in critical care medicine.

Tight glycemic control in critical care--the leading role of insulin sensitivity and patient variability: a review and model-based analysis

Tight glycemic control (TGC) has emerged as a major research focus in critical care due to its potential to simultaneously reduce both mortality and costs. However, repeating initial successful TGC trials that reduced mortality and other outcomes has proven difficult with more failures than successes. Hence, there has been growing debate over the necessity of TGC, its goals, the risk of severe hypoglycemia, and target cohorts. This paper provides a review of TGC via new analyses of data from several clinical trials, including SPRINT, Glucontrol and a recent NICU study. It thus provides both a review of the problem and major background factors driving it, as well as a novel model-based analysis designed to examine these dynamics from a new perspective. Using these clinical results and analysis, the goal is to develop new insights that shed greater light on the leading factors that make TGC difficult and inconsistent, as well as the requirements they thus impose on the design and implementation of TGC protocols. A model-based analysis of insulin sensitivity using data from three different critical care units, comprising over 75,000h of clinical data, is used to analyse variability in metabolic dynamics using a clinically validated model-based insulin sensitivity metric (S(I)). Variation in S(I) provides a new interpretation and explanation for the variable results seen (across cohorts and studies) in applying TGC. In particular, significant intra- and inter-patient variability in insulin resistance (1/S(I)) is seen be a major confounder that makes TGC difficult over diverse cohorts, yielding variable results over many published studies and protocols. Further factors that exacerbate this variability in glycemic outcome are found to include measurement frequency and whether a protocol is blind to carbohydrate administration.

Induced L₂-norm minimization of glucose-insulin system for Type I diabetic patients

Using induced L₂-norm minimization, a robust controller was developed for insulin delivery in Type I diabetic patients. The high-complexity nonlinear diabetic patient Sorensen-model was considered and Linear Parameter Varying methodology was used to develop open-loop model and robust H(∞) controller. Considering the normoglycaemic set point (81.1 mg/dL), a polytopic set was created over the physiologic boundaries of the glucose-insulin interaction of the Sorensen-model. In this way, Linear Parameter Varying model formalism was defined. The robust control was developed considering input and output multiplicative uncertainties with two additional uncertainties from those used in the literature: sensor noise and worst-case design for meal disturbance (60 g carbohydrate). Simulation scenario on large meal absorption illustrates the applicability of the robust LPV control technique, while patient variability is tested with real data taken from the SPRINT clinical protocol on ICU patients.

A physiological Intensive Control Insulin-Nutrition-Glucose (ICING) model validated in critically ill patients

Intensive insulin therapy (IIT) and tight glycaemic control (TGC), particularly in intensive care unit (ICU), are the subjects of increasing and controversial debate in recent years. Model-based TGC has shown potential in delivering safe and tight glycaemic management, all the while limiting hypoglycaemia. A comprehensive, more physiologically relevant Intensive Control Insulin-Nutrition-Glucose (ICING) model is presented and validated using data from critically ill patients. Two existing glucose-insulin models are reviewed and formed the basis for the ICING model. Model limitations are discussed with respect to relevant physiology, pharmacodynamics and TGC practicality. Model identifiability issues are carefully considered for clinical settings. This article also contains significant reference to relevant physiology and clinical literature, as well as some references to the modeling efforts in this field. Identification of critical constant population parameters was performed in two stages, thus addressing model identifiability issues. Model predictive performance is the primary factor for optimizing population parameter values. The use of population values are necessary due to the limited clinical data available at the bedside in the clinical control scenario. Insulin sensitivity, S(I), the only dynamic, time-varying parameter, is identified hourly for each individual. All population parameters are justified physiologically and with respect to values reported in the clinical literature. A parameter sensitivity study confirms the validity of limiting time-varying parameters to S(I) only, as well as the choices for the population parameters. The ICING model achieves median fitting error of <1% over data from 173 patients (N=42,941 h in total) who received insulin while in the ICU and stayed for ≥ 72 h. Most importantly, the median per-patient 1-h ahead prediction error is a very low 2.80% [IQR 1.18, 6.41%]. It is significant that the 75th percentile prediction error is within the lower bound of typical glucometer measurement errors of 7-12%. These results confirm that the ICING model is suitable for developing model-based insulin therapies, and capable of delivering real-time model-based TGC with a very tight prediction error range. Finally, the detailed examination and discussion of issues surrounding model-based TGC and existing glucose-insulin models render this article a mini-review of the state of model-based TGC in critical care.

Computer-assisted glucose regulation during rapid step-wise increases of parenteral nutrition in critically ill patients: a proof of concept study

BACKGROUND

Early delivery of calories is important in critically ill patients, and the administration of parenteral nutrition (PN) is sometimes required to achieve this goal. However, PN can induce acute hyperglycemia, which is associated with adverse outcome. We hypothesized that initiation of PN using a rapid "step-up" approach, coupled with a computerized insulin-dosing protocol, would result in a desirable caloric intake within 24 hours without causing hyperglycemia.

METHODS

In our surgical intensive care unit (ICU), glucose is regulated by a nurse-centered computerized glucose regulation program. When adequate enteral feeding was not possible, PN was initiated according to a simple step-up rule at an infusion rate of 10 mL/h (approximately 10 kcal/h) and subsequently increased by steps of 10 mL/h every 4 hours, provided glucose was <10 mmol/L, until the target caloric intake (1 kcal/kg/h) was reached. All glucose levels and insulin doses were collected during the step-up period and for 24 hours after achieving target feeding.

RESULTS

In all 23 consecutive patients requiring PN, mean intake was 1 kcal/kg/h within 24 hours. Of the 280 glucose samples during the 48-hour study period, mean ± standard deviation glucose level was 7.4 ± 1.4 mmol/L. Only 4.5% of glucose measurements during the step-up period were transiently ≥10 mmol/L. After initiating PN, the insulin requirement rose from 1.1 ± 1.5 units/h to 2.9 ± 2.5 units/h (P < .001).

CONCLUSIONS

This proof of concept study shows that rapid initiation of PN using a step-up approach coupled with computerized glucose control resulted in adequate caloric intake within 24 hours while maintaining adequate glycemic control.

Automated near-continuous glucose monitoring measured in plasma using mid-infrared spectroscopy

OBJECTIVE

There are increasing calls for a precise, automated system to enable tight glycemic control and to avoid hypoglycemia in an intensive care unit setting. OptiScan Biomedical has developed a glucose monitor based on mid-infrared spectroscopy that withdraws blood samples (120 µl) and measures plasma glucose. The goal of this study was to validate the performance of the OptiScan Model 5000 over a wide range of glycemic levels in patients.

RESEARCH DESIGN AND METHODS

Sixty people with type 1 (n = 18) or type 2 (n = 42) diabetes who were otherwise healthy were connected to OptiScanners. Their blood glucose concentrations were kept in a euglycemic, hypoglycemic (<75 mg/dl), and hyperglycemic (>180 mg/dl) range by intravenous administrations of insulin and glucose. OptiScanner venous blood samples were automatically withdrawn every 15 minutes. Reference measurements were done using the YSI 2300 glucose analyzer.

RESULTS

The aggregate data points (1155 paired readings) were within International Organization for Standardization standards, with 98.6% of the glucose values within ±20% above 75 mg/dl and ±15 mg/dl below this value. A Clarke error grid analysis showed a total of 1139 points (98.6%) in zone A. Points outside of A exceeded the A zone boundary by an average of 4.3%. The r(2) was 0.99. The total coefficient for variance was 6.4%.

CONCLUSIONS

These results show that the OptiScanner is highly accurate in healthy patients with diabetes across a wide range of glucose values. Mid-infrared spectroscopy may become the method of choice for highly accurate, high frequency, automated glucose measurements and may thus enable better glycemic control in critically ill patients.

Validation of a model-based virtual trials method for tight glycemic control in intensive care

BACKGROUND

In-silico virtual patients and trials offer significant advantages in cost, time and safety for designing effective tight glycemic control (TGC) protocols. However, no such method has fully validated the independence of virtual patients (or resulting clinical trial predictions) from the data used to create them. This study uses matched cohorts from a TGC clinical trial to validate virtual patients and in-silico virtual trial models and methods.

METHODS

Data from a 211 patient subset of the Glucontrol trial in Liege, Belgium. Glucontrol-A (N = 142) targeted 4.4-6.1 mmol/L and Glucontrol-B (N = 69) targeted 7.8-10.0 mmol/L. Cohorts were matched by APACHE II score, initial BG, age, weight, BMI and sex (p > 0.25). Virtual patients are created by fitting a clinically validated model to clinical data, yielding time varying insulin sensitivity profiles (SI(t)) that drives in-silico patients.Model fit and intra-patient (forward) prediction errors are used to validate individual in-silico virtual patients. Self-validation (tests A protocol on Group-A virtual patients; and B protocol on B virtual patients) and cross-validation (tests A protocol on Group-B virtual patients; and B protocol on A virtual patients) are used in comparison to clinical data to assess ability to predict clinical trial results.

RESULTS

Model fit errors were small (<0.25%) for all patients, indicating model fitness. Median forward prediction errors were: 4.3, 2.8 and 3.5% for Group-A, Group-B and Overall (A+B), indicating individual virtual patients were accurate representations of real patients. SI and its variability were similar between cohorts indicating they were metabolically similar.Self and cross validation results were within 1-10% of the clinical data for both Group-A and Group-B. Self-validation indicated clinically insignificant errors due to model and/or clinical compliance. Cross-validation clearly showed that virtual patients enabled by identified patient-specific SI(t) profiles can accurately predict the performance of independent and different TGC protocols.

CONCLUSIONS

This study fully validates these virtual patients and in silico virtual trial methods, and clearly shows they can accurately simulate, in advance, the clinical results of a TGC protocol, enabling rapid in silico protocol design and optimization. These outcomes provide the first rigorous validation of a virtual in-silico patient and virtual trials methodology.

International recommendations for glucose control in adult non diabetic critically ill patients

INTRODUCTION

The purpose of this research is to provide recommendations for the management of glycemic control in critically ill patients.

METHODS

Twenty-one experts issued recommendations related to one of the five pre-defined categories (glucose target, hypoglycemia, carbohydrate intake, monitoring of glycemia, algorithms and protocols), that were scored on a scale to obtain a strong or weak agreement. The GRADE (Grade of Recommendation, Assessment, Development and Evaluation) system was used, with a strong recommendation indicating a clear advantage for an intervention and a weak recommendation indicating that the balance between desirable and undesirable effects of an intervention is not clearly defined.

RESULTS

A glucose target of less than 10 mmol/L is strongly suggested, using intravenous insulin following a standard protocol, when spontaneous food intake is not possible. Definition of the severe hypoglycemia threshold of 2.2 mmol/L is recommended, regardless of the clinical signs. A general, unique amount of glucose (enteral/parenteral) to administer for any patient cannot be suggested. Glucose measurements should be performed on arterial rather than venous or capillary samples, using central lab or blood gas analysers rather than point-of-care glucose readers.

CONCLUSIONS

Thirty recommendations were obtained with a strong (21) and a weak (9) agreement. Among them, only 15 were graded with a high level of quality of evidence, underlying the necessity to continue clinical studies in order to improve the risk-to-benefit ratio of glucose control.

Organ failure and tight glycemic control in the SPRINT study

Introduction

Intensive care unit mortality is strongly associated with organ failure rate and severity. The sequential organ failure assessment (SOFA) score is used to evaluate the impact of a successful tight glycemic control (TGC) intervention (SPRINT) on organ failure, morbidity, and thus mortality.

Methods

A retrospective analysis of 371 patients (3,356 days) on SPRINT (August 2005 - April 2007) and 413 retrospective patients (3,211 days) from two years prior, matched by Acute Physiology and Chronic Health Evaluation (APACHE) III. SOFA is calculated daily for each patient. The effect of the SPRINT TGC intervention is assessed by comparing the percentage of patients with SOFA ≤5 each day and its trends over time and cohort/group. Organ-failure free days (all SOFA components ≤2) and number of organ failures (SOFA components >2) are also compared. Cumulative time in 4.0 to 7.0 mmol/L band (cTIB) was evaluated daily to link tightness and consistency of TGC (cTIB ≥0.5) to SOFA ≤5 using conditional and joint probabilities.

Results

Admission and maximum SOFA scores were similar (P = 0.20; P = 0.76), with similar time to maximum (median: one day; IQR: [1,3] days; P = 0.99). Median length of stay was similar (4.1 days SPRINT and 3.8 days Pre-SPRINT; P = 0.94). The percentage of patients with SOFA ≤5 is different over the first 14 days (P = 0.016), rising to approximately 75% for Pre-SPRINT and approximately 85% for SPRINT, with clear separation after two days. Organ-failure-free days were different (SPRINT = 41.6%; Pre-SPRINT = 36.5%; P < 0.0001) as were the percent of total possible organ failures (SPRINT = 16.0%; Pre-SPRINT = 19.0%; P < 0.0001). By Day 3 over 90% of SPRINT patients had cTIB ≥0.5 (37% Pre-SPRINT) reaching 100% by Day 7 (50% Pre-SPRINT). Conditional and joint probabilities indicate tighter, more consistent TGC under SPRINT (cTIB ≥0.5) increased the likelihood SOFA ≤5.

Conclusions

SPRINT TGC resolved organ failure faster, and for more patients, from similar admission and maximum SOFA scores, than conventional control. These reductions mirror the reduced mortality with SPRINT. The cTIB ≥0.5 metric provides a first benchmark linking TGC quality to organ failure. These results support other physiological and clinical results indicating the role tight, consistent TGC can play in reducing organ failure, morbidity and mortality, and should be validated on data from randomised trials.

A fast generalizable solution method for glucose control algorithms

In critical care tight control of blood glucose levels has been shown to lead to better clinical outcomes. The need to develop new protocols for tight glucose control, as well as the opportunity to optimize a variety of other drug therapies, has led to resurgence in model-based medical decision support in this area. One still valid hindrance to developing new model-based protocols using so-called virtual patients, retrospective clinical data, and Monte Carlo methods is the large amount of computational time and resources needed. This paper develops fast analytical-based methods for insulin-glucose system model that are generalizable to other similar systems. Exploiting the structure and partial solutions in a subset of the model is the key in finding accurate fast solutions to the full model. This approach successfully reduced computing time by factors of 5600-144000 depending on the numerical error management method, for large (50-164 patients) virtual trials and Monte Carlo analysis. It thus allows new model-based or model-derived protocols to be rapidly developed via extensive simulation. The new method is rigorously compared to existing standard numerical solutions and is found to be highly accurate to within 0.2%.

What makes tight glycemic control tight? The impact of variability and nutrition in two clinical studies

INTRODUCTION

Tight glycemic control (TGC) remains controversial while successful, consistent, and effective protocols remain elusive. This research analyzes data from two TGC trials for root causes of the differences achieved in control and thus potentially in glycemic and other outcomes. The goal is to uncover aspects of successful TGC and delineate the impact of differences in cohorts.

METHODS

A retrospective analysis was conducted using records from a 211-patient subset of the GluControl trial taken in Liege, Belgium, and 393 patients from Specialized Relative Insulin Nutrition Titration (SPRINT) in New Zealand. Specialized Relative Insulin Nutrition Titration targeted 4.0-6.0 mmol/liter, similar to the GluControl A (N = 142) target of 4.4-6.1 mmol/liter. The GluControl B (N = 69) target was 7.8-10.0 mmol/liter. Cohorts were matched by Acute Physiology and Chronic Health Evaluation II score and percentage males (p > .35); however, the GluControl cohort was slightly older (p = .011). Overall cohort and per-patient comparisons (median, interquartile range) are shown for (a) glycemic levels achieved, (b) nutrition from carbohydrate (all sources), and (c) insulin dosing for this analysis. Intra- and interpatient variability were examined using clinically validated model-based insulin sensitivity metric and its hour-to-hour variation.

RESULTS

Cohort blood glucose were as follows: SPRINT, 5.7 (5.0-6.6) mmol/liter; GluControl A, 6.3 (5.3-7.6) mmol/liter; and GluControl B, 8.2 (6.9-9.4) mmol/liter. Insulin dosing was 3.0 (1.0-3.0), 1.5 (0.5-3), and 0.7 (0.0-1.7) U/h, respectively. Nutrition from carbohydrate (all sources) was 435.5 (259.2-539.1), 311.0 (0.0-933.1), and 622.1 (103.7-1036.8) kcal/day, respectively. Median per-patient results for blood glucose were 5.8 (5.3-6.4), 6.4 (5.9-6.9), and 8.3 (7.6-8.8) mmol/liter. Insulin doses were 3.0 (2.0-3.0), 1.5 (0.8-2.0), and 0.5 (0.0-1.0) U/h. Carbohydrate administration was 383.6 (207.4-497.7), 103.7 (0.0-829.4), and 207.4 (0.0-725.8) kcal/day. Overall, SPRINT gave approximately 2x more insulin with a 3-4x narrower, but generally non-zero, range of nutritional input to achieve equally TGC with less hypoglycemia. Specialized Relative Insulin Nutrition Titration had much less hypoglycemia (<2.2 mmol/liter), with 2% of patients, compared to GluControl A (7.7%) and GluControl B (2.9%), indicating much lower variability, with similar results for glucose levels <3.0 mmol/liter. Specialized Relative Insulin Nutrition Titration also had less hyperglycemia (>8.0 mmol/liter) than groups A and B. GluControl patients (A+B) had a approximately 2x wider range of insulin sensitivity than SPRINT. Hour-to-hour variation was similar. Hence GluControl had greater interpatient variability but similar intrapatient variability.

CONCLUSION

Protocols that dose insulin blind to carbohydrate administration can suffer greater outcome glycemic variability, even if average cohort glycemic targets are met. While the cohorts varied significantly in model-assessed insulin resistance, their variability was similar. Such significant intra- and interpatient variability is a further significant cause and marker of glycemic variability in TGC. The results strongly recommended that TGC protocols be explicitly designed to account for significant intra- and interpatient variability in insulin resistance, as well as specifying or having knowledge of carbohydrate administration to minimize variability in glycemic outcomes across diverse cohorts and/or centers.

Hypoglycemia detection in critical care using continuous glucose monitors: an in silico proof of concept analysis

BACKGROUND

Tight glycemic control (TGC) in critical care has shown distinct benefits but has also been proven difficult to obtain. The risk of severe hypoglycemia (<40 mg/dl) has been increased significantly in several, but not all, studies, raising significant concerns for safety. Continuous glucose monitors (CGMs) offer frequent measurement and thus the possibility of using them for early detection alarms to prevent hypoglycemia.

METHODS

This study used retrospective clinical data from the Specialized Relative Insulin Nutrition Titration TGC study covering seven patients who experienced severe hypoglycemic events. Clinically validated metabolic system models were used to recreate a continuous blood glucose profile. In silico analysis was enabled by using a conservative single Gaussian noise model based on reported CGM clinical data from a critical care study [mean absolute percent error (MAPE) 17.4%]. A novel median filter was implemented and further smoothed with a least mean squares-fitted polynomial to reduce sensor noise. Two alarm approaches were compared. An integral-based method is presented that examined the area between a preset threshold and filtered simulated CGM data. An alarm was raised when this value became too low. A simple glycemic threshold method was also used for comparison. To account for random noise skewing the results, each patient record was Monte Carlo simulated 100 times with a different random noise profile for a total of 700 runs. Different alarm thresholds were analyzed parametrically. Results are reported in terms of detection time before the clinically measured event and any false alarms. These retrospective clinical data were used with approval from the New Zealand South Island Regional Ethics Committee.

RESULTS

The median filter reduced MAPE from 17.4% [standard deviation (SD) 13%] to 9.3% (SD 7%) over the cohort. For the integral-based alarm, median per-patient detection times ranged, t, from -35 minutes (before event) to -170 minutes, with zero to two false alarms per patient over the cohort and different alarm parameters. For a simple glycemic threshold alarm (three consecutive values below threshold), median per-patient alarm times were -10 to -75 minutes and false alarms were zero to seven; however, in one case, five of seven subjects never alarmed at all, despite the hypoglycemic event.

CONCLUSIONS

A retrospective study used clinical hypoglycemic events from a TGC study to develop and analyze an integral-based hypoglycemia alarm for use in critical care TGC studies. The integral-based approach was accurate, provided significant lead time before a hypoglycemic event, alarmed at higher glycemic levels, was robust to sensor noise, and had minimal false alarms. The approach is readily generalizable to similar scenarios, and results would justify a pilot clinical trial to verify this study.

Towards closed-loop glycaemic control

Blood glucose control performed by intensive care unit (ICU) nurses is becoming standard practice for critically ill patients. New algorithms, ranging from basic protocols to elementary computerized protocols to advanced computerized protocols, have been presented during the last years aiming to reduce the workload of the medical team. This paper gives an overview of the different types of algorithms and their features. Performance comparisons between different algorithms are avoided as blood glucose sampling frequencies and protocol durations were not similar among different studies and even within studies. Particularly advanced computerized protocols can potentially be introduced as fully-automated blood glucose algorithms when accurate and reliable near-continuous glucose sensor devices are available. Furthermore, it is surprising to consider in some of the described protocols that the original blood glucose target ranges (80-110 mg/dl) were increased (due to fear of hypoglycaemia) and/or that glycaemia levels were determined in capillary blood samples.

Safety constraints in an artificial pancreatic beta cell: an implementation of model predictive control with insulin on board

BACKGROUND

Type 1 diabetes mellitus (T1DM) is characterized by the destruction of pancreatic beta cells, resulting in the inability to produce sufficient insulin to maintain normoglycemia. As a result, people with T1DM depend on exogenous insulin that is given either by multiple daily injections or by an insulin pump to control their blood glucose. A challenging task is to design the next step in T1DM therapy: a fully automated insulin delivery system consisting of an artificial pancreatic beta cell that shall provide both safe and effective therapy. The core of such a system is a control algorithm that calculates the insulin dose based on automated glucose measurements.

METHODS

A model predictive control (MPC) algorithm was designed to control glycemia by controlling exogenous insulin delivery. The MPC algorithm contained a dynamic safety constraint, insulin on board (IOB), which incorporated the clinical values of correction factor and insulin-to-carbohydrate ratio along with estimated insulin action decay curves as part of the optimal control solution.

RESULTS

The results emphasized the ability of the IOB constraint to significantly improve the glucose/insulin control trajectories in the presence of aggressive control actions. The simulation results indicated that 50% of the simulations conducted without the IOB constraint resulted in hypoglycemic events, compared to 10% of the simulations that included the IOB constraint.

CONCLUSIONS

Achieving both efficacy and safety in an artificial pancreatic beta cell calls for an IOB safety constraint that is able to override aggressive control moves (large insulin doses), thereby minimizing the risk of hypoglycemia.

Glucose control in sepsis

Hyperglycemia is common during the course of critical illness and is associated with adverse clinical outcomes. Randomized controlled trials and large observational trials of insulin therapy titrated to achieve glucose values approximating the normal range (80 to 110 mg/dL) demonstrate improved morbidity and mortality in heterogeneous populations and have led to recommendations for improved glucose control. Patients who have septic shock, however, appear to be at higher risk for hypoglycemia, and a recent randomized trial focusing exclusively on patients who had severe sepsis did not show benefit. The recent Surviving Sepsis consensus statement recommends insulin therapy using validated protocols to lower glucose (less than 150 mg/dL) pending the results of adequately powered trials to determine if normalization (less than 110 mg/dL) of glucose is needed to optimize outcomes in patients who have severe sepsis.

A systematic review on quality indicators for tight glycaemic control in critically ill patients: need for an unambiguous indicator reference subset

Introduction

The objectives of this study were to systematically identify and summarize quality indicators of tight glycaemic control in critically ill patients, and to inspect the applicability of their definitions.

Methods

We searched in MEDLINE^® for all studies evaluating a tight glycaemic control protocol and/or quality of glucose control that reported original data from a clinical trial or observational study on critically ill adult patients.

Results

Forty-nine studies met the inclusion criteria; 30 different indicators were extracted and categorized into four nonorthogonal categories: blood glucose zones (for example, 'hypoglycaemia'); blood glucose levels (for example, 'mean blood glucose level'); time intervals (for example, 'time to occurrence of an event'); and protocol characteristics (for example, 'blood glucose sampling frequency'). Hypoglycaemia-related indicators were used in 43 out of 49 studies, acting as a proxy for safety, but they employed many different definitions. Blood glucose level summaries were used in 41 out of 49 studies, reported as means and/or medians during the study period or at a certain time point (for example, the morning blood glucose level or blood glucose level upon starting insulin therapy). Time spent in the predefined blood glucose level range, time needed to reach the defined blood glucose level target, hyperglycaemia-related indicators and protocol-related indicators were other frequently used indicators. Most indicators differ in their definitions even when they are meant to measure the same underlying concept. More importantly, many definitions are not precise, prohibiting their applicability and hence the reproducibility and comparability of research results.

Conclusions

An unambiguous indicator reference subset is necessary. The result of this systematic review can be used as a starting point from which to develop a standard list of well defined indicators that are associated with clinical outcomes or that concur with clinicians' subjective views on the quality of the regulatory process.

A simulation model of glucose regulation in the critically ill

Focused research is underway to improve the delivery of tight glycaemic control at the intensive care unit. A major component is the development of safe, efficacious and effective insulin titration algorithms, which are normally evaluated in time-consuming resource-demanding clinical studies. Simulation studies with virtual critically ill patients can substantially accelerate the development process. For this purpose, we created a model of glucoregulation in the critically ill. The model includes five submodels: a submodel of endogenous insulin secretion, a submodel of insulin kinetics, a submodel of enteral glucose absorption, a submodel of insulin action and a submodel of glucose kinetics. Model parameters are estimated utilizing prior knowledge and data collected routinely at the intensive care unit to represent the high intersubject and temporal variation in insulin needs in the critically ill. Bayesian estimation combined with the regularization method is used to estimate (i) time-invariant model parameters and (ii) a time-varying parameter, the basal insulin concentration, which represents the temporal variation in insulin sensitivity. We propose a validation process to validate virtual patients developed for the purpose of testing glucose controllers. The parameter estimation and the validation are exemplified using data collected in six critically ill patients treated at a medical intensive care unit. In conclusion, a novel glucoregulatory model has been developed to create a virtual population of critically ill facilitating in silico testing of glucose controllers at the intensive care unit.

A subcutaneous insulin pharmacokinetic model for computer simulation in a diabetes decision support role: model structure and parameter identification

OBJECTIVE

The goal of this study was to develop a unified physiological subcutaneous (SC) insulin absorption model for computer simulation in a clinical diabetes decision support role. The model must model the plasma insulin appearance of a wide range of current insulins, especially monomer insulin and insulin glargine, utilizing common chemical states and transport rates, where appropriate.

METHODS

A compartmental model was developed with 13 patient-specific model parameters covering six diverse insulin types [rapid-acting, regular, neutral protamine Hagedorn (NPH), lente, ultralente, and glargine insulin]. Model parameters were identified using 37 sets of mean plasma insulin time-course data from an extensive literature review via nonlinear optimization methods.

RESULTS

All fitted parameters have a coefficient of variation <100% (median 51.3%, 95th percentile 3.6-60.6%) and can be considered a posteriori identifiable.

CONCLUSION

A model is presented to describe SC injected insulin appearance in plasma in a diabetes decision support role. Clinically current insulin types (monomeric insulin, regular insulin, NPH, insulin, and glargine) and older insulin types (lente and ultralente) are included in a unified framework that accounts for nonlinear concentration and dose dependency. Future work requires clinical validation using published pharmacokinetic studies.

A benchmark data set for model-based glycemic control in critical care

BACKGROUND

Hyperglycemia is prevalent in critical care. That tight control saves lives is becoming more clear, but the "how" and "for whom" in repeating the initial results remain elusive. Model-based methods can provide tight, patient-specific control, as well as providing significant insight into the etiology and evolution of this condition. However, it is still often difficult to compare results due to lack of a common benchmark. This article puts forward a benchmark data set for critical care glycemic control in a medical intensive care unit (ICU). Based on clinical patient data from SPecialized Relative Insulin and Nutrition Tables (SPRINT) studies, it provides a benchmark for comparing and analyzing performance in model-based glycemic control.

METHODS

Data from 20 of the first 150 postpilot patients treated under SPRINT are presented. All patients had longer than a 5-day length of stay (LoS) in the Christchurch ICU. The benchmark data set matches overall patient data and glycemic control results for the entire cohort and this particular LoS >5-day group. The mortality outcome (n =3, 15%) also matches SPRINT results for this patient group.

RESULTS

Data cover 20 patients and 6372 total patient hours with an average of 339.4 hours per patient. It includes insulin and nutrition inputs along with 4182 blood glucose measurements at an average of 224.3 measurements per patient, averaging a measurement approximately every 1.5 hours (16 per day). Data are available via download in a Microsoft Excel format. A series of cumulative distribution functions and tables are used to summarize data in this article.

CONCLUSION

Model-based methods can provide tighter, more adaptable "one method fits all" solutions using methods that enable patient-specific modeling and control. A benchmark data set will enable easier model and protocol development for groups lacking clinical data, as well as providing a benchmark to compare results of different protocols on a single (virtual) cohort based on real clinical data.

Tight glucose control: should we move from intensive insulin therapy alone to modulation of insulin and nutritional inputs?

The report by Chase and coworkers in the previous issue of Critical Care describes the implementation into clinical practice of the Specialized Relative Insulin Nutrition Table (SPRINT) for tight glycaemic control in critically ill patients. SPRINT is a simple, wheel-based system that modulates both insulin rate and nutritional inputs. It achieved a better glycaemic control in a severely ill critical cohort than their previous method for glycaemic control in a matched historical cohort. Reductions in mortality were also observed.

The impact of parameter identification methods on drug therapy control in an intensive care unit

This paper investigates the impact of fast parameter identification methods, which do not require any forward simulations, on model-based glucose control, using retrospective data in the Christchurch Hospital Intensive Care Unit. The integral-based identification method has been previously clinically validated and extensively applied in a number of biomedical applications; and is a crucial element in the presented model-based therapeutics approach. Common non-linear regression and gradient descent approaches are too computationally intense and not suitable for the glucose control applications presented. The main focus in this paper is on better characterizing and understanding the importance of the integral in the formulation and the effect it has on model-based drug therapy control. As a comparison, a potentially more natural derivative formulation which has the same computation speed advantages is investigated, and is shown to go unstable with respect to modelling error which is always present clinically. The integral method remains robust.

Development of a clinical type 1 diabetes metabolic system model and in silico simulation tool

OBJECTIVES

The goal of this study was to develop a system model of type 1 diabetes for the purpose of in silico simulation for the prediction of long-term glycemic control outcomes.

METHODS

The system model was created and identified on a physiological cohort of virtual type 1 diabetes patients (n = 40). Integral-based identification was used to develop (n = 40) insulin sensitivity profiles.

RESULTS

The n = 40 insulin sensitivity profiles provide a driving input for virtual patient trials using the models developed. The identified models have a median (90% range) absolute percentage error of 1.33% (0.08-7.20%). The median (90% range) absolute error was 0.12 mmol/liter (0.01-0.56 mmol/liter). The model and integral-based identification of SI captured all patient dynamics with low error, which would lead to more physiological behavior simulation.

CONCLUSIONS

A simulation tool incorporating n = 40 virtual patient data sets to predict long-term glycemic control outcomes from clinical interventions was developed based on a physiological type 1 diabetes metabolic system model. The overall goal is to utilize this model and insulin sensitivity profiles to develop and optimize self-monitoring blood glucose and multiple daily injection therapy.

Model-based insulin sensitivity as a sepsis diagnostic in critical care

BACKGROUND

Timely diagnosis and treatment of sepsis in critical care require significant clinical effort, experience, and resources. Insulin sensitivity is known to decrease with worsening condition and could thus be used to aid diagnosis. Some glycemic control protocols are able to identify insulin sensitivity in real time.

METHODS

Receiver operating characteristic curves and cutoff insulin sensitivity values for diagnosing sepsis were calculated for model-based insulin sensitivity (S(I)) and a simpler metric (SS(I)) that was estimated from glycemic control data of 30 patients with sepsis and can be calculated in real time without use of a computer. Results were compared to the insulin sensitivity profiles of a general intensive care unit population of 113 patients without sepsis and 30 patients with sepsis, comprising a total of 26,453 patient hours. Patients with sepsis were identified as having sepsis based on a sepsis score (ss) of 3 or higher (ss = 0 - 4 for increasing severity). Patients with type I or type II diabetes were excluded. Ethics approval for this study was granted by the South Island Regional Ethics Committee.

RESULTS

Receiver operating characteristic cutoff values of S(I) = 8 x 10-5 liter mU(-1) min(-1) and SS(I) = 2.8 x 10-4 liter mU(-1) min(-1) were determined for ss > or = 3. The model-based S(I) fell below this value in 15% of all patient hours. The S(I) test had a negative predictive value of 99.8%. The test sensitivity was 78% and specificity was 82%. However, the positive predictor value was 2.8%. Slightly lower sensitivity (68.8%) and specificity (81.7%), but equally good negative prediction (99.7%), were obtained for the estimated SS(I).

CONCLUSIONS

Insulin sensitivity provides a negative predictive diagnostic for sepsis. High insulin sensitivity rules out sepsis for the majority of patient hours and may be determined noninvasively in real time from glycemic control protocol data. Low insulin sensitivity is not an effective diagnostic, as it can equally mark the presence of sepsis or other conditions.

Impact of human factors on clinical protocol performance: a proposed assessment framework and case examples

BACKGROUND

Hyperglycemia is prevalent in critical care and tight control can save lives. Current ad-hoc clinical protocols require significant clinical effort and can often produce highly variable results. Thus, tight control remains elusive as there is not enough understanding of the relationship between control performance and protocol design, particularly with regard to how a given protocol is implemented.

METHODS

This article examines the role of human factors and how individuals relate to technological protocols in clinical settings. The study consists of an overall brief review that is used to create a first graphical representation of the impact of human factors in clinical medical protocol implementations. This initial framework is examined in the context of two similar, but different, case studies-the specialized relative insulin and nutrition tables glycemic control protocol and the TREAT system for antibiotic selection.

RESULTS

A graphical framework relating the human factors impact on medical protocol implementation is created. This framework describes the primary impacts on performance as resulting from clinical burden and protocol transparency. Their primary effect is on compliance with the protocol, which directly affects performance and outcome, particularly in long-term studies versus short pilot studies.

SUMMARY

Compliance is a key element in obtaining the best clinical outcome that a given protocol can provide. The issues that most affect compliance are quite often unrelated to the patient or treatment, but are a function of the protocol design and its ability to integrate (by its design) into a given clinical setting. A framework for examining these issues in design and in post-hoc assessment is therefore proposed and examined in two brief case studies.

Multicenter validation of a computer-based clinical decision support tool for glucose control in adult and pediatric intensive care units

INTRODUCTION

Hyperglycemia during critical illness is common, and intravenous insulin therapy (IIT) to normalize blood glucose improves outcomes in selected populations. Methods differ widely in complexity, insulin dosing approaches, efficacy, and rates of hypoglycemia. We developed a simple bedside-computerized decision support protocol (eProtocol-insulin) that yields promising results in the development center. We examined the effectiveness and safety of this tool in six adult and five pediatric intensive care units (ICUs) in other centers.

METHODS

We required attending physicians of eligible patients to independently intend to use intravenous insulin to normalize blood glucose. We used eProtocol-insulin for glucose control for a duration determined by the clinical caregivers. Adults had an anticipated length of stay of 3 or more days. In pediatric ICUs, we also required support or intended support with mechanical ventilation for greater than 24 hours or with a vasoactive infusion. We recorded all instances in which eProtocol-insulin instructions were not accepted and all blood glucose values. An independent data safety and monitoring board monitored study results and subject safety. Bedside nurses were selected randomly to complete a paper survey describing their perceptions of quality of care and workload related to eProtocol-insulin use.

RESULTS

Clinicians accepted 93% of eProtocol-insulin instructions (11,773/12,645) in 100 adult and 48 pediatric subjects. Forty-eight percent of glucose values were in the target range. Both of these results met a priori-defined efficacy thresholds. Only 0.18% of glucose values were < or =40 mg/dl. This is lower than values reported in prior IIT studies. Although nurses reported eProtocol-insulin required as much work as managing a mechanical ventilator, most nurses felt eProtocol-insulin had a low impact on their ability to complete non-IIT nursing activities.

CONCLUSIONS

A multicenter validation demonstrated that eProtocol-insulin is a valid, exportable tool that can assist clinicians in achieving control of glucose in critically ill adults and children.

Implementation and evaluation of the SPRINT protocol for tight glycaemic control in critically ill patients: a clinical practice change

Introduction

Stress-induced hyperglycaemia is prevalent in critical care. Control of blood glucose levels to within a 4.4 to 6.1 mmol/L range or below 7.75 mmol/L can reduce mortality and improve clinical outcomes. The Specialised Relative Insulin Nutrition Tables (SPRINT) protocol is a simple wheel-based system that modulates insulin and nutritional inputs for tight glycaemic control.

Methods

SPRINT was implemented as a clinical practice change in a general intensive care unit (ICU). The objective of this study was to measure the effect of the SPRINT protocol on glycaemic control and mortality compared with previous ICU control methods. Glycaemic control and mortality outcomes for 371 SPRINT patients with a median Acute Physiology And Chronic Health Evaluation (APACHE) II score of 18 (interquartile range [IQR] 15 to 24) are compared with a 413-patient retrospective cohort with a median APACHE II score of 18 (IQR 15 to 23).

Results

Overall, 53.9% of all measurements were in the 4.4 to 6.1 mmol/L band. Blood glucose concentrations were found to be log-normal and thus log-normal statistics are used throughout to describe the data. The average log-normal glycaemia was 6.0 mmol/L (standard deviation 1.5 mmol/L). Only 9.0% of all measurements were below 4.4 mmol/L, with 3.8% below 4 mmol/L and 0.1% of measurements below 2.2 mmol/L. On SPRINT, 80% more measurements were in the 4.4 to 6.1 mmol/L band and standard deviation of blood glucose was 38% lower compared with the retrospective control. The range and peak of blood glucose were not correlated with mortality for SPRINT patients (P >0.30). For ICU length of stay (LoS) of greater than or equal to 3 days, hospital mortality was reduced from 34.1% to 25.4% (-26%) (P = 0.05). For ICU LoS of greater than or equal to 4 days, hospital mortality was reduced from 34.3% to 23.5% (-32%) (P = 0.02). For ICU LoS of greater than or equal to 5 days, hospital mortality was reduced from 31.9% to 20.6% (-35%) (P = 0.02). ICU mortality was also reduced but the P value was less than 0.13 for ICU LoS of greater than or equal to 4 and 5 days.

Conclusion

SPRINT achieved a high level of glycaemic control on a severely ill critical cohort population. Reductions in mortality were observed compared with a retrospective hyperglycaemic cohort. Range and peak blood glucose metrics were no longer correlated with mortality outcome under SPRINT.

Inpatient insulin therapy

PURPOSE OF REVIEW

In a 2001 report from a surgical intensive care unit in Leuven, Belgium, intravenous insulin infusion targeting blood glucose 80-110 mg/dl reduced patient mortality and morbidities. Subsequent research has failed to define glycemic targets necessary or sufficient for attainment of desired health outcomes in other inpatient settings, but a large body of evidence suggests hospital outcomes are related to hyperglycemia.

RECENT FINDINGS

Recent literature describes observational evidence for hypoglycemia as an independent predictor of mortality in a general medical intensive care unit; superiority of performance of computerized intravenous insulin algorithms in comparison to earlier manual algorithms; acceptability of early transition to scheduled basal prandial correction subcutaneous insulin analog therapy for maintenance of glycemic targets after induction of euglycemia by intravenous insulin infusion, among cardiothoracic surgery patients; inferiority of sliding scale insulin compared to basal prandial correction therapy; and feasibility of diabetes patient self-management in the hospital setting.

SUMMARY

With development of improved insulin administration strategies problems of hypoglycemia and variability of glycemic control are reduced. Investigators and care providers need to achieve glycemic targets to optimize patient outcomes.

Clinical validation of a model-based glycaemic control design approach and comparison to other clinical protocols

Hyperglycaemia is prevalent in critical care and tight control can reduce mortality from 9-43% depending on the level of control and the cohort. This research presents a table-based method that varies both insulin dose and nutritional input to achieve tight control. The system mimics a previously validated model-based system, but can be used for long term, large patient number clinical evaluation. This paper evaluates this method in simulation using retrospective data and then compares clinical measurements over 15,000 patient hours to validate the models and development approach. This validation thus also validates the in silico comparison to the landmark clinical tight glycaemic control protocols. Overall, an average clinical glucose level is 5.9 +/- 1.0 mmol/L, matching simulation, however the overall clinical glucose distribution is slightly tighter than that obtained in simulation, indicating that the retrospective virtual trial design approach is slightly conservative. Finally, the model based approach is shown to have tighter control than existing, more ad-hoc clinical approaches based on the simulation results that qualitatively match reported clinical results, but also show significant variation around the average levels obtained in both the hypo-and hyperglycaemic ranges.

Monte Carlo analysis of a new model-based method for insulin sensitivity testing

Insulin resistance (IR), or low insulin sensitivity, is a major risk factor in the pathogenesis of type 2 diabetes and cardiovascular disease. A simple, high resolution assessment of IR would enable earlier diagnosis and more accurate monitoring of intervention effects. Current assessments are either too intensive for clinical settings (Euglycaemic Clamp, IVGTT) or have too low resolution (HOMA, fasting glucose/insulin). Based on high correlation of a model-based measure of insulin sensitivity and the clamp, a novel, clinically useful test protocol is designed with: physiological dosing, short duration (<1 h), simple protocol, low cost and high repeatability. Accuracy and repeatability are assessed with Monte Carlo analysis on a virtual clamp cohort (N=146). Insulin sensitivity as measured by this test has a coefficient of variation (CV) of CV(SI)=4.5% (90% CI: 3.8-5.7%), slightly higher than clamp ISI (CV(ISI)=3.3% (90% CI: 3.0-4.0%)) and significantly lower than HOMA (CV(HOMA)=10.0% (90% CI: 9.1-10.8%)). Correlation to glucose and unit normalised ISI is r=0.98 (90% CI: 0.97-0.98). The proposed protocol is simple, cost effective, repeatable and highly correlated to the gold-standard clamp.

Stochastic modelling of insulin sensitivity and adaptive glycemic control for critical care

Targeted, tight model-based glycemic control in critical care patients that can reduce mortality 18-45% is enabled by prediction of insulin sensitivity, S(I). However, this parameter can vary significantly over a given hour in the critically ill as their condition evolves. A stochastic model of S(I) variability is constructed using data from 165 critical care patients. Given S(I) for an hour, the stochastic model returns the probability density function of S(I) for the next hour. Consequently, the glycemic distribution following a known intervention can be derived, enabling pre-determined likelihoods of the result and more accurate control. Cross validation of the S(I) variability model shows that 86.6% of the blood glucose measurements are within the 0.90 probability interval, and 54.0% are within the interquartile interval. "Virtual Patients" with S(I) behaving to the overall S(I) variability model achieved similar predictive performance in simulated trials (86.8% and 45.7%). Finally, adaptive control method incorporating S(I) variability is shown to produce improved glycemic control in simulated trials compared to current clinical results. The validated stochastic model and methods provide a platform for developing advanced glycemic control methods addressing critical care variability.

Nurse-led implementation of an insulin-infusion protocol in a general intensive care unit: improved glycaemic control with increased costs and risk of hypoglycaemia signals need for algorithm revision

Background

Strict glycaemic control (SGC) has become a contentious issue in modern intensive care. Physicians and nurses are concerned about the increased workload due to SGC as well as causing harm through hypoglycaemia. The objective of our study was to evaluate our existing degree of glycaemic control, and to implement SGC safely in our ICU through a nurse-led implementation of an algorithm for intensive insulin-therapy.

Methods

The study took place in the adult general intensive care unit (11 beds) of a 44-bed department of intensive care at a tertiary care university hospital. All patients admitted during the 32 months of the study were enrolled. We retrospectively analysed all arterial blood glucose (BG) results from samples that were obtained over a period of 20 months prior to the implementation of SGC. We then introduced an algorithm for intensive insulin therapy; aiming for arterial blood-glucose at 4.4 – 6.1 mmol/L. Doctors and nurses were trained in the principles and potential benefits and risks of SGC. Consecutive statistical analyses of blood samples over a period of 12 months were used to assess performance, provide feedback and uncover incidences of hypoglycaemia.

Results

Median BG level was 6.6 mmol/L (interquartile range 5.6 to 7.7 mmol/L) during the period prior to implementation of SGC (494 patients), and fell to 5.9 (IQR 5.1 to 7.0) mmol/L following introduction of the new algorithm (448 patients). The percentage of BG samples > 8 mmol/L was reduced from 19.2 % to 13.1 %. Before implementation of SGC, 33 % of samples were between 4.4 to 6.1 mmol/L and 12 patients (2.4 %) had one or more episodes of severe hypoglycaemia (< 2.2 mmol/L). Following implementation of SGC, 45.8 % of samples were between 4.4 to 6.1 mmol/L and 40 patients (8.9 %) had one or more episodes of severe hypoglycaemia. Of theses, ten patients died while still hospitalised (all causes).

Conclusion

The retrospective part of the study indicated ample room for improvement. Through the implementation of SGC the fraction of samples within the new target range increased from 33% to 45.8%. There was also a significant increase in severe hypoglycaemic episodes. There continues to be potential for improved glycaemic control within our ICU. This might be achieved through an improved algorithm and continued efforts to increase nurses' confidence and skills in achieving SGC.

Analysis of algorithms for intensive care unit blood glucose control

Intensive care unit (ICU) blood glucose control algorithms were reviewed and analyzed in the context of linear systems theory and classical feedback control algorithms. Closed-loop performance was illustrated by applying the algorithms in simulation studies using an in silico model of an ICU patient. Steady-state and dynamic input-output analysis was used to provide insight about controller design and potential closed-loop performance. The proportional-integral-derivative, columnar insulin dosing (CID, Glucommander-like), and glucose regulation for intensive care patients (GRIP) algorithms were shown to have similar features and performance. The CID strategy is a time-varying proportional-only controller (no integral action), whereas the GRIP algorithm is a nonlinear controller with integral action. A minor modification to the GRIP algorithm was suggested to improve the closed-loop performance. Recommendations were made to guide control theorists on important ICU control topics worthy of further study.