Decision support for optimized blood glucose control and nutrition in a neurotrauma intensive care unit: preliminary results of clinical advice and prediction accuracy of the Glucosafe system. J Clin Monit Comput. 2012;26:319-328. [PMID: 22581038 DOI: 10.1007/s10877-012-9364-y]

Society of Critical Care Medicine Guidelines on Glycemic Control for Critically Ill Children and Adults 2024

RATIONALE

Maintaining glycemic control of critically ill patients may impact outcomes such as survival, infection, and neuromuscular recovery, but there is equipoise on the target blood levels, monitoring frequency, and methods.

OBJECTIVES

The purpose was to update the 2012 Society of Critical Care Medicine and American College of Critical Care Medicine (ACCM) guidelines with a new systematic review of the literature and provide actionable guidance for clinicians.

PANEL DESIGN

The total multiprofessional task force of 22, consisting of clinicians and patient/family advocates, and a methodologist applied the processes described in the ACCM guidelines standard operating procedure manual to develop evidence-based recommendations in alignment with the Grading of Recommendations Assessment, Development, and Evaluation Approach (GRADE) methodology. Conflict of interest policies were strictly followed in all phases of the guidelines, including panel selection and voting.

METHODS

We conducted a systematic review for each Population, Intervention, Comparator, and Outcomes question related to glycemic management in critically ill children (≥ 42 wk old adjusted gestational age to 18 yr old) and adults, including triggers for initiation of insulin therapy, route of administration, monitoring frequency, role of an explicit decision support tool for protocol maintenance, and methodology for glucose testing. We identified the best available evidence, statistically summarized the evidence, and then assessed the quality of evidence using the GRADE approach. We used the evidence-to-decision framework to formulate recommendations as strong or weak or as a good practice statement. In addition, "In our practice" statements were included when the available evidence was insufficient to support a recommendation, but the panel felt that describing their practice patterns may be appropriate. Additional topics were identified for future research.

RESULTS

This guideline is an update of the guidelines for the use of an insulin infusion for the management of hyperglycemia in critically ill patients. It is intended for adult and pediatric practitioners to reassess current practices and direct research into areas with inadequate literature. The panel issued seven statements related to glycemic control in unselected adults (two good practice statements, four conditional recommendations, one research statement) and seven statements for pediatric patients (two good practice statements, one strong recommendation, one conditional recommendation, two "In our practice" statements, and one research statement), with additional detail on specific subset populations where available.

CONCLUSIONS

The guidelines panel achieved consensus for adults and children regarding a preference for an insulin infusion for the acute management of hyperglycemia with titration guided by an explicit clinical decision support tool and frequent (≤ 1 hr) monitoring intervals during glycemic instability to minimize hypoglycemia and against targeting intensive glucose levels. These recommendations are intended for consideration within the framework of the patient's existing clinical status. Further research is required to evaluate the role of individualized glycemic targets, continuous glucose monitoring systems, explicit decision support tools, and standardized glycemic control metrics.

Development, Design and Utilization of a CDSS for Refeeding Syndrome in Real Life Inpatient Care-A Feasibility Study

BACKGROUND

The refeeding syndrome (RFS) is an oftentimes-unrecognized complication of reintroducing nutrition in malnourished patients that can lead to fatal cardiovascular failure. We hypothesized that a clinical decision support system (CDSS) can improve RFS recognition and management.

METHODS

We developed an algorithm from current diagnostic criteria for RFS detection, tested the algorithm on a retrospective dataset and combined the final algorithm with therapy and referral recommendations in a knowledge-based CDSS. The CDSS integration into clinical practice was prospectively investigated for six months.

RESULTS

The utilization of the RFS-CDSS lead to RFS diagnosis in 13 out of 21 detected cases (62%). It improved patient-related care and documentation, e.g., RFS-specific coding (E87.7), increased from once coded in 30 month in the retrospective cohort to four times in six months in the prospective cohort and doubled the rate of nutrition referrals in true positive patients (retrospective referrals in true positive patients 33% vs. prospective referrals in true positive patients 71%).

CONCLUSION

CDSS-facilitated RFS diagnosis is possible and improves RFS recognition. This effect and its impact on patient-related outcomes needs to be further investigated in a large randomized-controlled trial.

Weighted entropy deep features on hybrid RNN with LSTM for glucose level and diabetes prediction

Glucose level regulation with essential advice regarding diabetes must be provided to the patients to maintain their diet for diabetes treatment. Therefore, the academic community has focused on implementing novel glucose prediction techniques for decision support systems. Recent computational techniques for diagnosing diabetes have certain limitations, and also they are not evaluated under various datasets obtained from the different people of various countries. This generates inefficiency in the prediction systems to apply it in real-time applications. This paper plans to suggest a hybrid deep learning model for diabetes prediction and glucose level classification. Two benchmark datasets are used in the data collection process for experimenting. Initially, the deep selected features were extracted by the Convolutional Neural Network (CNN). Further, weighted entropy deep features are extracted, where the tuning of weight is taken place by the Modified Escaping Energy-based Harris Hawks Optimization. These features are processed in the glucose level classification using the modified Fuzzy classifier for classifying the high-level and low-level glucose. Further, glucose prediction is done by the Hybrid Recurrent Neural Network (RNN), and Long Short Term Memory (LSTM) termed R-LSTM with parameter optimization. From the experimental result, In the dataset 2 analyses on SMAPE, the MEE-HHO-R-LSTM is 12.5%, 87.5%, 50%, 12.5%, and 2.5% better than SVM, LSTM, DNN, RNN, and RNN-LSTM, at the learning percentage of 75%. The analytical results enforce that the suggested methods attain enhanced prediction performance concerning the evaluation metrics compared to conventional prediction models.

Enabling a learning healthcare system with automated computer protocols that produce replicable and personalized clinician actions

Clinical decision-making is based on knowledge, expertise, and authority, with clinicians approving almost every intervention-the starting point for delivery of "All the right care, but only the right care," an unachieved healthcare quality improvement goal. Unaided clinicians suffer from human cognitive limitations and biases when decisions are based only on their training, expertise, and experience. Electronic health records (EHRs) could improve healthcare with robust decision-support tools that reduce unwarranted variation of clinician decisions and actions. Current EHRs, focused on results review, documentation, and accounting, are awkward, time-consuming, and contribute to clinician stress and burnout. Decision-support tools could reduce clinician burden and enable replicable clinician decisions and actions that personalize patient care. Most current clinical decision-support tools or aids lack detail and neither reduce burden nor enable replicable actions. Clinicians must provide subjective interpretation and missing logic, thus introducing personal biases and mindless, unwarranted, variation from evidence-based practice. Replicability occurs when different clinicians, with the same patient information and context, come to the same decision and action. We propose a feasible subset of therapeutic decision-support tools based on credible clinical outcome evidence: computer protocols leading to replicable clinician actions (eActions). eActions enable different clinicians to make consistent decisions and actions when faced with the same patient input data. eActions embrace good everyday decision-making informed by evidence, experience, EHR data, and individual patient status. eActions can reduce unwarranted variation, increase quality of clinical care and research, reduce EHR noise, and could enable a learning healthcare system.

Usability study of a new tool for nutritional and glycemic management in adult intensive care: Glucosafe 2

The new decision support tool Glucosafe 2 (GS2) is based on a mathematical model of glucose and insulin dynamics, designed to assist caregivers in blood glucose control and nutrition. This study aims to assess end-user acceptance and usability of this bedside decision support tool in an adult intensive care setting. Caregivers were first trained and then invited to trial GS2 prototype on bedside computers. Data for qualitative analysis were collected through semi-structured interviews from twenty users after minimum three trial days. Most caregivers (70%) rated GS2 as convenient and believed it would help improving adherence to current guidelines (85%). Moreover, most nurses (80%) believed that GS2 would be timesaving. Nurses' risk perceptions and manual data entry emerged as central barriers to use GS2 in routine practice. Issues emerged from the caregivers were compiled into a list of 12 modifications of the GS2 prototype to increase end-user acceptance and usability. This usability study showed that GS2 was considered by ICU caregivers as helpful in daily clinical practice, allowing time-saving and better standardization of ICU patient's care. Important issues were raised by the users with implications for the development and deployment of GS2. Integrating the technology into existing IT infrastructure may facilitate caregivers' acceptance. Further clinical studies of the performance and potential health outcomes are warranted.

Human Cognitive Limitations. Broad, Consistent, Clinical Application of Physiological Principles Will Require Decision Support

Our education system seems to fail to enable clinicians to broadly understand core physiological principles. The emphasis on reductionist science, including "omics" branches of research, has likely contributed to this decrease in understanding. Consequently, clinicians cannot be expected to consistently make clinical decisions linked to best physiological evidence. This is a large-scale problem with multiple determinants, within an even larger clinical decision problem: the failure of clinicians to consistently link their decisions to best evidence. Clinicians, like all human decision-makers, suffer from significant cognitive limitations. Detailed context-sensitive computer protocols can generate personalized medicine instructions that are well matched to individual patient needs over time and can partially resolve this problem.

Next-generation, personalised, model-based critical care medicine: a state-of-the art review of in silico virtual patient models, methods, and cohorts, and how to validation them

Critical care, like many healthcare areas, is under a dual assault from significantly increasing demographic and economic pressures. Intensive care unit (ICU) patients are highly variable in response to treatment, and increasingly aging populations mean ICUs are under increasing demand and their cohorts are increasingly ill. Equally, patient expectations are growing, while the economic ability to deliver care to all is declining. Better, more productive care is thus the big challenge. One means to that end is personalised care designed to manage the significant inter- and intra-patient variability that makes the ICU patient difficult. Thus, moving from current "one size fits all" protocolised care to adaptive, model-based "one method fits all" personalised care could deliver the required step change in the quality, and simultaneously the productivity and cost, of care. Computer models of human physiology are a unique tool to personalise care, as they can couple clinical data with mathematical methods to create subject-specific models and virtual patients to design new, personalised and more optimal protocols, as well as to guide care in real-time. They rely on identifying time varying patient-specific parameters in the model that capture inter- and intra-patient variability, the difference between patients and the evolution of patient condition. Properly validated, virtual patients represent the real patients, and can be used in silico to test different protocols or interventions, or in real-time to guide care. Hence, the underlying models and methods create the foundation for next generation care, as well as a tool for safely and rapidly developing personalised treatment protocols over large virtual cohorts using virtual trials. This review examines the models and methods used to create virtual patients. Specifically, it presents the models types and structures used and the data required. It then covers how to validate the resulting virtual patients and trials, and how these virtual trials can help design and optimise clinical trial. Links between these models and higher order, more complex physiome models are also discussed. In each section, it explores the progress reported up to date, especially on core ICU therapies in glycemic, circulatory and mechanical ventilation management, where high cost and frequency of occurrence provide a significant opportunity for model-based methods to have measurable clinical and economic impact. The outcomes are readily generalised to other areas of medical care.

Untangling glycaemia and mortality in critical care

Background

Hyperglycaemia is associated with adverse outcomes in the intensive care unit, and initial studies suggested outcome benefits of glycaemic control (GC). However, subsequent studies often failed to replicate these results, and they were often unable to achieve consistent, safe control, raising questions about the benefit or harm of GC as well as the nature of the association of glycaemia with mortality and clinical outcomes. In this study, we evaluated if non-survivors are harder to control than survivors and determined if glycaemic outcome is a function of patient condition and eventual outcome or of the glycaemic control provided.

Methods

Clinically validated, model-based, hour-to-hour insulin sensitivity (SI) and its hour-to-hour variability (%ΔSI) were identified over the first 72 h of therapy in 145 patients (119 survivors, 26 non-survivors). In hypothesis testing, we compared distributions of SI and %ΔSI in 6-hourly blocks for survivors and non-survivors. In equivalence testing, we assessed if differences in these distributions, based on blood glucose measurement error, were clinically significant.

Results

SI level was never equivalent between survivors and non-survivors (95% CI of percentage difference in medians outside ±12%). Non-survivors had higher SI, ranging from 9% to 47% higher overall in 6-h blocks, and this difference became statistically significant as glycaemic control progressed. %ΔSI was equivalent between survivors and non-survivors for all 6-hourly blocks (95% CI of difference in medians within ±12%) and decreased in general over time as glycaemic control progressed.

Conclusions

Whereas non-survivors had higher SI levels, variability was equivalent to that of survivors over the first 72 h. These results indicate survivors and non-survivors are equally controllable, given an effective glycaemic control protocol, suggesting that glycaemia level and variability, and thus the association between glycaemia and outcome, are essentially determined by the control provided rather than by underlying patient or metabolic condition.

Electronic supplementary material

The online version of this article (doi:10.1186/s13054-017-1725-y) contains supplementary material, which is available to authorized users.

Nonlinear model predictive glycemic control of critically ill patients using online identification of insulin sensitivity

In critically ill patients suffering from hyperglycemia, it has been recently shown that mortality and morbidity can be reduced by keeping blood glucose within the range of 80-110 mg/dL. However, maintaining glycemia within such range is difficult due to the time variability in insulin sensitivity in critically ill patients. In this paper, we propose a novel glycometabolism model of critically ill patients with an insulin sensitivity parameter and develop a nonlinear model predictive glycemic control system with online identification of insulin sensitivity at one-hour intervals. Simulation results show that our system keeps 70% of BG measurements within the range of 80-110 mg/dL without any severe hypoglycemic incidents, which indicates the effectiveness and safety of our system.

An in silico method to identify computer-based protocols worthy of clinical study: An insulin infusion protocol use case

OBJECTIVE

Develop an efficient non-clinical method for identifying promising computer-based protocols for clinical study. An in silico comparison can provide information that informs the decision to proceed to a clinical trial. The authors compared two existing computer-based insulin infusion protocols: eProtocol-insulin from Utah, USA, and Glucosafe from Denmark.

MATERIALS AND METHODS

The authors used eProtocol-insulin to manage intensive care unit (ICU) hyperglycemia with intravenous (IV) insulin from 2004 to 2010. Recommendations accepted by the bedside clinicians directly link the subsequent blood glucose values to eProtocol-insulin recommendations and provide a unique clinical database. The authors retrospectively compared in silico 18,984 eProtocol-insulin continuous IV insulin infusion rate recommendations from 408 ICU patients with those of Glucosafe, the candidate computer-based protocol. The subsequent blood glucose measurement value (low, on target, high) was used to identify if the insulin recommendation was too high, on target, or too low.

RESULTS

Glucosafe consistently provided more favorable continuous IV insulin infusion rate recommendations than eProtocol-insulin for on target (64% of comparisons), low (80% of comparisons), or high (70% of comparisons) blood glucose. Aggregated eProtocol-insulin and Glucosafe continuous IV insulin infusion rates were clinically similar though statistically significantly different (Wilcoxon signed rank test P = .01). In contrast, when stratified by low, on target, or high subsequent blood glucose measurement, insulin infusion rates from eProtocol-insulin and Glucosafe were statistically significantly different (Wilcoxon signed rank test, P < .001), and clinically different.

DISCUSSION

This in silico comparison appears to be an efficient nonclinical method for identifying promising computer-based protocols.

CONCLUSION

Preclinical in silico comparison analytical framework allows rapid and inexpensive identification of computer-based protocol care strategies that justify expensive and burdensome clinical trials.

Computerized advice on drug dosage to improve prescribing practice

BACKGROUND

Maintaining therapeutic concentrations of drugs with a narrow therapeutic window is a complex task. Several computer systems have been designed to help doctors determine optimum drug dosage. Significant improvements in health care could be achieved if computer advice improved health outcomes and could be implemented in routine practice in a cost-effective fashion. This is an updated version of an earlier Cochrane systematic review, first published in 2001 and updated in 2008.

OBJECTIVES

To assess whether computerized advice on drug dosage has beneficial effects on patient outcomes compared with routine care (empiric dosing without computer assistance).

SEARCH METHODS

The following databases were searched from 1996 to January 2012: EPOC Group Specialized Register, Reference Manager; Cochrane Central Register of Controlled Trials (CENTRAL), MEDLINE, Ovid; EMBASE, Ovid; and CINAHL, EbscoHost. A "top up" search was conducted for the period January 2012 to January 2013; these results were screened by the authors and potentially relevant studies are listed in Studies Awaiting Classification. The review authors also searched reference lists of relevant studies and related reviews.

SELECTION CRITERIA

We included randomized controlled trials, non-randomized controlled trials, controlled before-and-after studies and interrupted time series analyses of computerized advice on drug dosage. The participants were healthcare professionals responsible for patient care. The outcomes were any objectively measured change in the health of patients resulting from computerized advice (such as therapeutic drug control, clinical improvement, adverse reactions).

DATA COLLECTION AND ANALYSIS

Two review authors independently extracted data and assessed study quality. We grouped the results from the included studies by drug used and the effect aimed at for aminoglycoside antibiotics, amitriptyline, anaesthetics, insulin, anticoagulants, ovarian stimulation, anti-rejection drugs and theophylline. We combined the effect sizes to give an overall effect for each subgroup of studies, using a random-effects model. We further grouped studies by type of outcome when appropriate (i.e. no evidence of heterogeneity).

MAIN RESULTS

Forty-six comparisons (from 42 trials) were included (as compared with 26 comparisons in the last update) including a wide range of drugs in inpatient and outpatient settings. All were randomized controlled trials except two studies. Interventions usually targeted doctors, although some studies attempted to influence prescriptions by pharmacists and nurses. Drugs evaluated were anticoagulants, insulin, aminoglycoside antibiotics, theophylline, anti-rejection drugs, anaesthetic agents, antidepressants and gonadotropins. Although all studies used reliable outcome measures, their quality was generally low.This update found similar results to the previous update and managed to identify specific therapeutic areas where the computerized advice on drug dosage was beneficial compared with routine care:1. it increased target peak serum concentrations (standardized mean difference (SMD) 0.79, 95% CI 0.46 to 1.13) and the proportion of people with plasma drug concentrations within the therapeutic range after two days (pooled risk ratio (RR) 4.44, 95% CI 1.94 to 10.13) for aminoglycoside antibiotics;2. it led to a physiological parameter more often within the desired range for oral anticoagulants (SMD for percentage of time spent in target international normalized ratio +0.19, 95% CI 0.06 to 0.33) and insulin (SMD for percentage of time in target glucose range: +1.27, 95% CI 0.56 to 1.98);3. it decreased the time to achieve stabilization for oral anticoagulants (SMD -0.56, 95% CI -1.07 to -0.04);4. it decreased the thromboembolism events (rate ratio 0.68, 95% CI 0.49 to 0.94) and tended to decrease bleeding events for anticoagulants although the difference was not significant (rate ratio 0.81, 95% CI 0.60 to 1.08). It tended to decrease unwanted effects for aminoglycoside antibiotics (nephrotoxicity: RR 0.67, 95% CI 0.42 to 1.06) and anti-rejection drugs (cytomegalovirus infections: RR 0.90, 95% CI 0.58 to 1.40);5. it tended to reduce the length of time spent in the hospital although the difference was not significant (SMD -0.15, 95% CI -0.33 to 0.02) and to achieve comparable or better cost-effectiveness ratios than usual care;6. there was no evidence of differences in mortality or other clinical adverse events for insulin (hypoglycaemia), anaesthetic agents, anti-rejection drugs and antidepressants.For all outcomes, statistical heterogeneity quantified by I(2) statistics was moderate to high.

AUTHORS' CONCLUSIONS

This review update suggests that computerized advice for drug dosage has some benefits: it increases the serum concentrations for aminoglycoside antibiotics and improves the proportion of people for which the plasma drug is within the therapeutic range for aminoglycoside antibiotics.It leads to a physiological parameter more often within the desired range for oral anticoagulants and insulin. It decreases the time to achieve stabilization for oral anticoagulants. It tends to decrease unwanted effects for aminoglycoside antibiotics and anti-rejection drugs, and it significantly decreases thromboembolism events for anticoagulants. It tends to reduce the length of hospital stay compared with routine care while comparable or better cost-effectiveness ratios were achieved.However, there was no evidence that decision support had an effect on mortality or other clinical adverse events for insulin (hypoglycaemia), anaesthetic agents, anti-rejection drugs and antidepressants. In addition, there was no evidence to suggest that some decision support technical features (such as its integration into a computer physician order entry system) or aspects of organization of care (such as the setting) could optimize the effect of computerized advice.Taking into account the high risk of bias of, and high heterogeneity between, studies, these results must be interpreted with caution.

Computerized decision support in adult and pediatric critical care

Computerized decision support (CDS) is the most advanced form of clinical decision support available and has evolved with innovative technologies to provide meaningful assistance to medical professionals. Critical care clinicians are in unique environments where vast amounts of data are collected on individual patients, and where expedient and accurate decisions are paramount to the delivery of quality healthcare. Many CDS tools are in use today among adult and pediatric intensive care units as diagnostic aides, safety alerts, computerized protocols, and automated recommendations for management. Some CDS use have significantly decreased adverse events and improved costs when carefully implemented and properly operated. CDS tools integrated into electronic health records are also valuable to researchers providing rapid identification of eligible patients, streamlining data-gathering and analysis, and providing cohorts for study of rare and chronic diseases through data-warehousing. Although the need for human judgment in the daily care of critically ill patients has limited the study and realization of meaningful improvements in overall patient outcomes, CDS tools continue to evolve and integrate into the daily workflow of clinicians, and will likely provide advancements over time. Through novel technologies, CDS tools have vast potential for progression and will significantly impact the field of critical care and clinical research in the future.

The effects of on-screen, point of care computer reminders on processes and outcomes of care

BACKGROUND

The opportunity to improve care by delivering decision support to clinicians at the point of care represents one of the main incentives for implementing sophisticated clinical information systems. Previous reviews of computer reminder and decision support systems have reported mixed effects, possibly because they did not distinguish point of care computer reminders from e-mail alerts, computer-generated paper reminders, and other modes of delivering 'computer reminders'.

OBJECTIVES

To evaluate the effects on processes and outcomes of care attributable to on-screen computer reminders delivered to clinicians at the point of care.

SEARCH STRATEGY

We searched the Cochrane EPOC Group Trials register, MEDLINE, EMBASE and CINAHL and CENTRAL to July 2008, and scanned bibliographies from key articles.

SELECTION CRITERIA

Studies of a reminder delivered via a computer system routinely used by clinicians, with a randomised or quasi-randomised design and reporting at least one outcome involving a clinical endpoint or adherence to a recommended process of care.

DATA COLLECTION AND ANALYSIS

Two authors independently screened studies for eligibility and abstracted data. For each study, we calculated the median improvement in adherence to target processes of care and also identified the outcome with the largest such improvement. We then calculated the median absolute improvement in process adherence across all studies using both the median outcome from each study and the best outcome.

MAIN RESULTS

Twenty-eight studies (reporting a total of thirty-two comparisons) were included. Computer reminders achieved a median improvement in process adherence of 4.2% (interquartile range (IQR): 0.8% to 18.8%) across all reported process outcomes, 3.3% (IQR: 0.5% to 10.6%) for medication ordering, 3.8% (IQR: 0.5% to 6.6%) for vaccinations, and 3.8% (IQR: 0.4% to 16.3%) for test ordering. In a sensitivity analysis using the best outcome from each study, the median improvement was 5.6% (IQR: 2.0% to 19.2%) across all process measures and 6.2% (IQR: 3.0% to 28.0%) across measures of medication ordering. In the eight comparisons that reported dichotomous clinical endpoints, intervention patients experienced a median absolute improvement of 2.5% (IQR: 1.3% to 4.2%). Blood pressure was the most commonly reported clinical endpoint, with intervention patients experiencing a median reduction in their systolic blood pressure of 1.0 mmHg (IQR: 2.3 mmHg reduction to 2.0 mmHg increase).

AUTHORS' CONCLUSIONS

Point of care computer reminders generally achieve small to modest improvements in provider behaviour. A minority of interventions showed larger effects, but no specific reminder or contextual features were significantly associated with effect magnitude. Further research must identify design features and contextual factors consistently associated with larger improvements in provider behaviour if computer reminders are to succeed on more than a trial and error basis.