No full-text available
To read the full-text of this research,
you can request a copy directly from the authors.
Unsupervised machine learning analysis to identify patterns of ICU medication use for fluid overload prediction
Abstract
Background
Fluid overload (FO) in the intensive care unit (ICU) is common, serious, and may be preventable. Intravenous medications (including administered volume) are a primary cause for FO but are challenging to evaluate as a FO predictor given the high frequency and time‐dependency of their use and other factors affecting FO. We sought to employ unsupervised machine learning methods to uncover medication administration patterns correlating with FO.
Methods
This retrospective cohort study included 927 adults admitted to an ICU for ≥72 h. FO was defined as a positive fluid balance ≥7% of admission body weight. After reviewing medication administration record data in 3‐h periods, medication exposure was categorized into clusters using principal component analysis (PCA) and Restricted Boltzmann Machine (RBM). Medication regimens of patients with and without FO were compared within clusters to assess their temporal association with FO.
Results
FO occurred in 127 (13.7%) of 927 included patients. Patients received a median (interquartile range) of 31(13–65) discrete intravenous medication administrations over the 72‐h period. Across all 47,803 intravenous medication administrations, 10 unique medication clusters, containing 121 to 130 medications per cluster, were identified. The mean number of Cluster 7 medications administered was significantly greater in the FO cohort compared with patients without FO (25.6 vs.10.9, p < 0.0001). A total of 51 (40.2%) of 127 unique Cluster 7 medications were administered in more than five different 3‐h periods during the 72‐h study window. The most common Cluster 7 medications included continuous infusions, antibiotics, and sedatives/analgesics. Addition of Cluster 7 medications to an FO prediction model including the Acute Physiologic and Chronic Health Evaluation (APACHE) II score and receipt of diuretics improved model predictiveness from an Area Under the Receiver Operation Characteristic (AUROC) curve of 0.719 to 0.741 ( p = 0.027).
Conclusions
Using machine learning approaches, a unique medication cluster was strongly associated with FO. Incorporation of this cluster improved the ability to predict FO compared to traditional prediction models. Integration of this approach into real‐time clinical applications may improve early detection of FO to facilitate timely intervention.
ResearchGate has not been able to resolve any citations for this publication.
The TRIPOD (Transparent Reporting of a multivariable prediction model for Individual Prognosis Or Diagnosis) statement was published in 2015 to provide the minimum reporting recommendations for studies developing or evaluating the performance of a prediction model. Methodological advances in the field of prediction have since included the widespread use of artificial intelligence (AI) powered by machine learning methods to develop prediction models. An update to the TRIPOD statement is thus needed. TRIPOD+AI provides harmonised guidance for reporting prediction model studies, irrespective of whether regression modelling or machine learning methods have been used. The new checklist supersedes the TRIPOD 2015 checklist, which should no longer be used. This article describes the development of TRIPOD+AI and presents the expanded 27 item checklist with more detailed explanation of each reporting recommendation, and the TRIPOD+AI for Abstracts checklist. TRIPOD+AI aims to promote the complete, accurate, and transparent reporting of studies that develop a prediction model or evaluate its performance. Complete reporting will facilitate study appraisal, model evaluation, and model implementation.
Fluid overload, while common in the ICU and associated with serious sequelae, is hard to predict and may be influenced by ICU medication use. Machine learning (ML) approaches may offer advantages over traditional regression techniques to predict it. We compared the ability of traditional regression techniques and different ML-based modeling approaches to identify clinically meaningful fluid overload predictors. This was a retrospective, observational cohort study of adult patients admitted to an ICU ≥ 72 h between 10/1/2015 and 10/31/2020 with available fluid balance data. Models to predict fluid overload (a positive fluid balance ≥ 10% of the admission body weight) in the 48–72 h after ICU admission were created. Potential patient and medication fluid overload predictor variables (n = 28) were collected at either baseline or 24 h after ICU admission. The optimal traditional logistic regression model was created using backward selection. Supervised, classification-based ML models were trained and optimized, including a meta-modeling approach. Area under the receiver operating characteristic (AUROC), positive predictive value (PPV), and negative predictive value (NPV) were compared between the traditional and ML fluid prediction models. A total of 49 of the 391 (12.5%) patients developed fluid overload. Among the ML models, the XGBoost model had the highest performance (AUROC 0.78, PPV 0.27, NPV 0.94) for fluid overload prediction. The XGBoost model performed similarly to the final traditional logistic regression model (AUROC 0.70; PPV 0.20, NPV 0.94). Feature importance analysis revealed severity of illness scores and medication-related data were the most important predictors of fluid overload. In the context of our study, ML and traditional models appear to perform similarly to predict fluid overload in the ICU. Baseline severity of illness and ICU medication regimen complexity are important predictors of fluid overload.
Unsupervised clustering of intensive care unit (ICU) medications may identify unique medication clusters (i.e., pharmacophenotypes) in critically ill adults. We performed an unsupervised analysis with Restricted Boltzmann Machine of 991 medications profiles of patients managed in the ICU to explore pharmacophenotypes that correlated with ICU complications (e.g., mechanical ventilation) and patient-centered outcomes (e.g., length of stay, mortality). Six unique pharmacophenotypes were observed, with unique medication profiles and clinically relevant differences in ICU complications and patient-centered outcomes. While pharmacophenotypes 2 and 4 had no statistically significant difference in ICU length of stay, duration of mechanical ventilation, or duration of vasopressor use, their mortality differed significantly (9.0% vs. 21.9%, p < 0.0001). Pharmacophenotype 4 had a mortality rate of 21.9%, compared with the rest of the pharmacophenotypes ranging from 2.5 to 9%. Phenotyping approaches have shown promise in classifying the heterogenous syndromes of critical illness to predict treatment response and guide clinical decision support systems but have never included comprehensive medication information. This first-ever machine learning approach revealed differences among empirically-derived subgroups of ICU patients that are not typically revealed by traditional classifiers. Identification of pharmacophenotypes may enable enhanced decision making to optimize treatment decisions.
While medication regimen complexity, as measured by a novel medication regimen complexity-intensive care unit (MRC-ICU) score, correlates with baseline severity of illness and mortality, whether the MRC-ICU improves hospital mortality prediction is not known. After characterizing the association between MRC-ICU, severity of illness and hospital mortality we sought to evaluate the incremental benefit of adding MRC-ICU to illness severity-based hospital mortality prediction models. This was a single-center, observational cohort study of adult intensive care units (ICUs). A random sample of 991 adults admitted ≥ 24 h to the ICU from 10/2015 to 10/2020 were included. The logistic regression models for the primary outcome of mortality were assessed via area under the receiver operating characteristic (AUROC). Medication regimen complexity was evaluated daily using the MRC-ICU. This previously validated index is a weighted summation of medications prescribed in the first 24 h of ICU stay [e.g., a patient prescribed insulin (1 point) and vancomycin (3 points) has a MRC-ICU = 4 points]. Baseline demographic features (e.g., age, sex, ICU type) were collected and severity of illness (based on worst values within the first 24 h of ICU admission) was characterized using both the Acute Physiology and Chronic Health Evaluation (APACHE II) and the Sequential Organ Failure Assessment (SOFA) score. Univariate analysis of 991 patients revealed every one-point increase in the average 24-h MRC-ICU score was associated with a 5% increase in hospital mortality [Odds Ratio (OR) 1.05, 95% confidence interval 1.02–1.08, p = 0.002]. The model including MRC-ICU, APACHE II and SOFA had a AUROC for mortality of 0.81 whereas the model including only APACHE-II and SOFA had a AUROC for mortality of 0.76. Medication regimen complexity is associated with increased hospital mortality. A prediction model including medication regimen complexity only modestly improves hospital mortality prediction.
Background
Identifying patterns within ICU medication regimens may help artificial intelligence algorithms to better predict patient outcomes; however, machine learning methods incorporating medications require further development, including standardized terminology. The Common Data Model for Intensive Care Unit (ICU) Medications (CDM-ICURx) may provide important infrastructure to clinicians and researchers to support artificial intelligence analysis of medication-related outcomes and healthcare costs. Using an unsupervised cluster analysis approach in combination with this common data model, the objective of this evaluation was to identify novel patterns of medication clusters (termed ‘pharmacophenotypes’) correlated with ICU adverse events (e.g., fluid overload) and patient-centered outcomes (e.g., mortality).
Methods
This was a retrospective, observational cohort study of 991 critically ill adults. To identify pharmacophenotypes, unsupervised machine learning analysis with automated feature learning using restricted Boltzmann machine and hierarchical clustering was performed on the medication administration records of each patient during the first 24 h of their ICU stay. Hierarchical agglomerative clustering was applied to identify unique patient clusters. Distributions of medications across pharmacophenotypes were described, and differences among patient clusters were compared using signed rank tests and Fisher's exact tests, as appropriate.
Results
A total of 30,550 medication orders for the 991 patients were analyzed; five unique patient clusters and six unique pharmacophenotypes were identified. For patient outcomes, compared to patients in Clusters 1 and 3, patients in Cluster 5 had a significantly shorter duration of mechanical ventilation and ICU length of stay ( p < 0.05); for medications, Cluster 5 had a higher distribution of Pharmacophenotype 1 and a smaller distribution of Pharmacophenotype 2, compared to Clusters 1 and 3. For outcomes, patients in Cluster 2, despite having the highest severity of illness and greatest medication regimen complexity, had the lowest overall mortality; for medications, Cluster 2 also had a comparably higher distribution of Pharmacophenotype 6.
Conclusion
The results of this evaluation suggest that patterns among patient clusters and medication regimens may be observed using empiric methods of unsupervised machine learning in combination with a common data model. These results have potential because while phenotyping approaches have been used to classify heterogenous syndromes in critical illness to better define treatment response, the entire medication administration record has not been incorporated in those analyses. Applying knowledge of these patterns at the bedside requires further algorithm development and clinical application but may have the future potential to be leveraged in guiding medication-related decision making to improve treatment outcomes.
Machine learning can predict outcomes and determine variables contributing to precise prediction, and can thus classify patients with different risk factors of outcomes. This study aimed to investigate the predictive accuracy for mortality and length of stay in intensive care unit (ICU) patients using machine learning, and to identify the variables contributing to the precise prediction or classification of patients. Patients (n = 12,747) admitted to the ICU at Chiba University Hospital were randomly assigned to the training and test cohorts. After learning using the variables on admission in the training cohort, the area under the curve (AUC) was analyzed in the test cohort to evaluate the predictive accuracy of the supervised machine learning classifiers, including random forest (RF) for outcomes (primary outcome, mortality; secondary outcome, length of ICU stay). The rank of the variables that contributed to the machine learning prediction was confirmed, and cluster analysis of the patients with risk factors of mortality was performed to identify the important variables associated with patient outcomes. Machine learning using RF revealed a high predictive value for mortality, with an AUC of 0.945 (95% confidence interval [CI] 0.922–0.977). In addition, RF showed high predictive value for short and long ICU stays, with AUCs of 0.881 (95% CI 0.876–0.908) and 0.889 (95% CI 0.849–0.936), respectively. Lactate dehydrogenase (LDH) was identified as a variable contributing to the precise prediction in machine learning for both mortality and length of ICU stay. LDH was also identified as a contributing variable to classify patients into sub-populations based on different risk factors of mortality. The machine learning algorithm could predict mortality and length of stay in ICU patients with high accuracy. LDH was identified as a contributing variable in mortality and length of ICU stay prediction and could be used to classify patients based on mortality risk.
Introduction
De-resuscitation practices in septic patients with heart failure (HF) are not well characterized. This study aimed to determine if diuretic initiation within 48 hours of intensive care unit (ICU) admission was associated with a positive fluid balance and patient outcomes.
Methods
This single-center, retrospective cohort study included adult patients with an established diagnosis of HF admitted to the ICU with sepsis or septic shock. The primary outcome was the incidence of positive fluid balance in patients receiving early (<48 hours) versus late (>48 hours) initiation of diuresis. Secondary outcomes included hospital mortality, ventilator-free days, and hospital and ICU length of stay. Continuous variables were assessed using independent t-test or Mann-Whitney U, while categorical variables were evaluated using the Pearson Chi-squared test.
Results
A total of 101 patients were included. Positive fluid balance was significantly reduced at 72 hours (−139 mL vs 4370 mL, P < .001). The duration of mechanical ventilation (4 vs 5 days, P = .129), ventilator-free days (22 vs 18.5 days, P = .129), and in-hospital mortality (28 (38%) vs 12 (43%), P = .821) were similar between groups. In a subgroup analysis excluding patients not receiving renal replacement therap (RRT) (n = 76), early diuretics was associated with lower incidence of mechanical ventilation (41 [73.2%] vs 20 (100%), P = .01) and reduced duration of mechanical ventilation (4 vs 8 days, P = .018).
Conclusions
Diuretic use within 48 hours of ICU admission in septic patients with HF resulted in less incidence of positive fluid balance. Early diuresis in this unique patient population warrants further investigation.
Background:
The detrimental impact of fluid overload (FO) on intensive care unit (ICU) morbidity and mortality is well known. However, research to identify subgroups of patients particularly prone to fluid overload is scarce. The aim of this cohort study was to derive "FO phenotypes" in the critically ill by using machine learning techniques.
Methods:
Retrospective single center study including adult intensive care patients with a length of stay of ≥3 days and sufficient data to compute FO. Data was analyzed by multivariable logistic regression, fast and frugal trees (FFT), classification decision trees (DT), and a random forest (RF) model.
Results:
Out of 1772 included patients, 387 (21.8%) met the FO definition. The random forest model had the highest area under the curve (AUC) (0.84, 95% CI 0.79-0.86), followed by multivariable logistic regression (0.81, 95% CI 0.77-0.86), FFT (0.75, 95% CI 0.69-0.79) and DT (0.73, 95% CI 0.68-0.78) to predict FO. The most important predictors identified in all models were lactate and bicarbonate at admission and postsurgical ICU admission. Sepsis/septic shock was identified as a risk factor in the MV and RF analysis.
Conclusion:
The FO phenotypes consist of patients admitted after surgery or with sepsis/septic shock with high lactate and low bicarbonate.
Sepsis is a leading cause of death in hospitals. Early prediction and diagnosis of sepsis, which is critical in reducing mortality, is challenging as many of its signs and symptoms are similar to other less critical conditions. We develop an artificial intelligence algorithm, SERA algorithm, which uses both structured data and unstructured clinical notes to predict and diagnose sepsis. We test this algorithm with independent, clinical notes and achieve high predictive accuracy 12 hours before the onset of sepsis (AUC 0.94, sensitivity 0.87 and specificity 0.87). We compare the SERA algorithm against physician predictions and show the algorithm’s potential to increase the early detection of sepsis by up to 32% and reduce false positives by up to 17%. Mining unstructured clinical notes is shown to improve the algorithm’s accuracy compared to using only clinical measures for early warning 12 to 48 hours before the onset of sepsis.
This article is one of ten reviews selected from the Annual Update in Intensive Care and Emergency Medicine 2020. Other selected articles can be found online at https://www.biomedcentral.com/collections/annualupdate2020. Further information about the Annual Update in Intensive Care and Emergency Medicine is available from http://www.springer.com/series/8901.
Intensive-care clinicians are presented with large quantities of measurements from multiple monitoring systems. The limited ability of humans to process complex information hinders early recognition of patient deterioration, and high numbers of monitoring alarms lead to alarm fatigue. We used machine learning to develop an early-warning system that integrates measurements from multiple organ systems using a high-resolution database with 240 patient-years of data. It predicts 90% of circulatory-failure events in the test set, with 82% identified more than 2 h in advance, resulting in an area under the receiver operating characteristic curve of 0.94 and an area under the precision-recall curve of 0.63. On average, the system raises 0.05 alarms per patient and hour. The model was externally validated in an independent patient cohort. Our model provides early identification of patients at risk for circulatory failure with a much lower false-alarm rate than conventional threshold-based systems. A machine-learning algorithm based on an array of demographic, physiological and clinical information is able to predict, hours in advance, circulatory failure of patients in the intensive-care unit.
Objective:
Administration of diuretics has been shown to assist fluid management and improve clinical outcomes in the critically ill post-shock resolution. Current guidelines have not yet included standardization or guidance for diuretic-based de-resuscitation in critically ill patients. This study aimed to evaluate the impact of a multi-disciplinary protocol for diuresis-guided de-resuscitation in the critically ill.
Methods:
This was a pre-post single-center pilot study within the medical intensive care unit (ICU) of a large academic medical center. Adult patients admitted to the Medical ICU receiving mechanical ventilation with either (1) clinical signs of volume overload via chest radiography or physical exam or (2) any cumulative fluid balance ≥ 0 mL since hospital admission were eligible for inclusion. Patients received diuresis per clinician discretion for a 2-year period (historical control) followed by a diuresis protocol for 1 year (intervention). Patients within the intervention group were matched in a 1:3 ratio with those from the historical cohort who met the study inclusion and exclusion criteria.
Results:
A total of 364 patients were included, 91 in the protocol group and 273 receiving standard care. Protocolized diuresis was associated with a significant decrease in 72-h post-shock cumulative fluid balance [median, IQR - 2257 (- 5676-920) mL vs 265 (- 2283-3025) mL; p < 0.0001]. In-hospital mortality in the intervention group was lower compared to the historical group (5.5% vs 16.1%; p = 0.008) and higher ICU-free days (p = 0.03). However, no statistically significant difference was found in ventilator-free days, and increased rates of hypernatremia and hypokalemia were demonstrated.
Conclusions:
This study showed that a protocol for diuresis for de-resuscitation can significantly improve 72-h post-shock fluid balance with potential benefit on clinical outcomes.
Intravenous fluids (IVFs) are the most common drugs administered in the intensive care unit. Despite the ubiquitous use, IVFs are not benign and carry significant risks associated with under- or overadministration. Hypovolemia is associated with decreased organ perfusion, ischemia, and multi-organ failure. Hypervolemia and volume overload are associated with organ dysfunction, delayed liberation from mechanical ventilation, and increased mortality. Despite appropriate provision of IVF, adverse drug effects such as electrolyte abnormalities and acid–base disturbances may occur. The management of volume status in critically ill patients is both dynamic and tenuous, a process that requires frequent monitoring and high clinical acumen. Because patient-specific considerations for fluid therapy evolve across the continuum of critical illness, a standard approach to the assessment of fluid needs and prescription of IVF therapy is necessary. We propose the principle of “fluid stewardship,” guided by 4 rights of medication safety: right patient, right drug, right route, and right dose. The successful implementation of fluid stewardship will aid pharmacists in making decisions regarding IVF therapy to optimize hemodynamic management and improve patient outcomes. Additionally, we highlight several areas of focus for future research, guided by the 4 rights construct of fluid stewardship.
Background and objectives
Excess fluid balance in acute kidney injury (AKI) may be harmful, and conversely, some patients may respond to fluid challenges. This study aimed to develop a prediction model that can be used to differentiate between volume-responsive (VR) and volume-unresponsive (VU) AKI.
Methods
AKI patients with urine output < 0.5 ml/kg/h for the first 6 h after ICU admission and fluid intake > 5 l in the following 6 h in the US-based critical care database (Medical Information Mart for Intensive Care (MIMIC-III)) were considered. Patients who received diuretics and renal replacement on day 1 were excluded. Two predictive models, using either machine learning extreme gradient boosting (XGBoost) or logistic regression, were developed to predict urine output > 0.65 ml/kg/h during 18 h succeeding the initial 6 h for assessing oliguria. Established models were assessed by using out-of-sample validation. The whole sample was split into training and testing samples by the ratio of 3:1.
Main results
Of the 6682 patients included in the analysis, 2456 (36.8%) patients were volume responsive with an increase in urine output after receiving > 5 l fluid. Urinary creatinine, blood urea nitrogen (BUN), age, and albumin were the important predictors of VR. The machine learning XGBoost model outperformed the traditional logistic regression model in differentiating between the VR and VU groups (AU-ROC, 0.860; 95% CI, 0.842 to 0.878 vs. 0.728; 95% CI 0.703 to 0.753, respectively).
Conclusions
The XGBoost model was able to differentiate between patients who would and would not respond to fluid intake in urine output better than a traditional logistic regression model. This result suggests that machine learning techniques have the potential to improve the development and validation of predictive modeling in critical care research.
Principal component analysis is one of the most important and powerful methods in chemometrics as well as in a wealth of other areas. This paper provides a description of how to understand, use, and interpret principal component analysis. The paper focuses on the use of principal component analysis in typical chemometric areas but the results are generally applicable.
Principal component analysis of a data matrix extracts the dominant patterns in the matrix in terms of a complementary set of score and loading plots. It is the responsibility of the data analyst to formulate the scientific issue at hand in terms of PC projections, PLS regressions, etc. Ask yourself, or the investigator, why the data matrix was collected, and for what purpose the experiments and measurements were made. Specify before the analysis what kinds of patterns you would expect and what you would find exciting.The results of the analysis depend on the scaling of the matrix, which therefore must be specified. Variance scaling, where each variable is scaled to unit variance, can be recommended for general use, provided that almost constant variables are left unscaled. Combining different types of variables warrants blockscaling.In the initial analysis, look for outliers and strong groupings in the plots, indicating that the data matrix perhaps should be “polished” or whether disjoint modeling is the proper course.For plotting purposes, two or three principal components are usually sufficient, but for modeling purposes the number of significant components should be properly determined, e.g. by cross-validation.Use the resulting principal components to guide your continued investigation or chemical experimentation, not as an end in itself.
Much biomedical research is observational. The reporting of such research is often inadequate, which hampers the assessment of its strengths and weaknesses and of a study's generalisability. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) Initiative developed recommendations on what should be included in an accurate and complete report of an observational study. We defined the scope of the recommendations to cover three main study designs: cohort, case–control and cross-sectional studies. We convened a 2-day workshop in September 2004, with methodologists, researchers, and journal editors to draft a checklist of items. This list was subsequently revised during several meetings of the coordinating group and in e-mail discussions with the larger group of STROBE contributors, taking into account empirical evidence and methodological considerations. The workshop and the subsequent iterative process of consultation and revision resulted in a checklist of 22 items (the STROBE Statement) that relate to the title, abstract, introduction, methods, results, and discussion sections of articles. 18 items are common to all three study designs and four are specific for cohort, case–control, or cross-sectional studies. A detailed Explanation and Elaboration document is published separately and is freely available on the websites of PLoS Medicine, Annals of Internal Medicine and Epidemiology. We hope that the STROBE Statement will contribute to improving the quality of reporting of observational studies.
Objective
Common data models provide a standard means of describing data for artificial intelligence (AI) applications, but this process has never been undertaken for medications used in the intensive care unit (ICU). We sought to develop a common data model (CDM) for ICU medications to standardize the medication features needed to support future ICU AI efforts.
Materials and Methods
A 9-member, multi-professional team of ICU clinicians and AI experts conducted a 5-round modified Delphi process employing conference calls, web-based communication, and electronic surveys to define the most important medication features for AI efforts. Candidate ICU medication features were generated through group discussion and then independently scored by each team member based on relevance to ICU clinical decision-making and feasibility for collection and coding. A key consideration was to ensure the final ontology both distinguished unique medications and met Findable, Accessible, Interoperable, and Reusable (FAIR) guiding principles.
Results
Using a list of 889 ICU medications, the team initially generated 106 different medication features, and 71 were ranked as being core features for the CDM. Through this process, 106 medication features were assigned to 2 key feature domains: drug product-related (n = 43) and clinical practice-related (n = 63). Each feature included a standardized definition and suggested response values housed in the electronic data library. This CDM for ICU medications is available online.
Conclusion
The CDM for ICU medications represents an important first step for the research community focused on exploring how AI can improve patient outcomes and will require ongoing engagement and refinement.
Purpose
Intravenous fluids are the most commonly prescribed medication in the intensive care unit (ICU) and can have a negative impact on patient outcomes if not utilized properly. Fluid stewardship aims to heighten awareness and improve practice in fluid therapy. This report describes a practical construct for implementation of fluid stewardship services and characterizes the pharmacist’s role in fluid stewardship practice.
Summary
Fluid stewardship services were integrated into an adult medical ICU at a large community hospital. Data characterizing these services over a 2-year span are reported and categorized based on the 4 rights (right patient, right drug, right route, right dose) and the ROSE (rescue, optimization, stabilization, evacuation) model of fluid administration. The review encompassed 305 patients totaling 905 patient days for whom 2,597 pharmacist recommendations were made, 19% of which were related to fluid stewardship. This corresponded to an average of 1.52 fluid stewardship recommendations per patient. Within the construct of the 4 rights, 39% of recommendations were related to the right patient, 33% were related to the right route, 17% were related to the right drug, and 11% were related to the right dose. By the ROSE model, 1% of recommendations were related to the rescue phase, 3% were related to optimization, 79% were related to stabilization, and 17% were related to evacuation.
Conclusion
Implementation of fluid stewardship pharmacy services in a community hospital medical ICU is feasible. Integration of this practice contributed to 19% of pharmacy recommendations. The most common recommendations involved evaluation of the patient for the appropriateness of fluid therapy during the stabilization phase. The impact of fluid stewardship on patient outcomes needs to be explored.
Despite the frequent use of maintenance intravenous fluids (mIVF) in critically ill patients, limited guidance is available. Notably, fluid overload secondary to mIVF mismanagement is associated with significant adverse patient outcomes. The Four Rights (right drug, right dose, right duration, right patient) construct of fluid stewardship has been proposed for the safe evaluation and use of fluids. The purpose of this evidence-based review is to offer practical insights for the clinician regarding mIVF selection, dosing, and duration in line with the Four Rights of Fluid Stewardship.
Background
Critically ill patients are at increased risk for fluid overload, but objective prediction tools to guide clinical decision-making are lacking. The MRC-ICU scoring tool is an objective tool for measuring medication regimen complexity.
Objective
To evaluate the relationship between MRC-ICU score and fluid overload in critically ill patients.
Methods
In this multi-center, retrospective, observational study, the relationship between MRC-ICU and the risk of fluid overload was examined. Patient demographics, fluid balance at day 3 of ICU admission, MRC-ICU score at 24 hours, and clinical outcomes were collected from the medical record. The primary outcome was relationship between MRC-ICU and fluid overload. To analyze this, MRC-ICU scores were divided into tertiles (low, moderate, high), and binary logistic regression was performed. Linear regression was performed to determine variables associated with positive fluid balance.
Results
A total of 125 patients were included. The median MRC-ICU score at 24 hours of ICU admission for low, moderate, and high tertiles were 9, 15, and 21, respectively. For each point increase in MRC-ICU, a 13% increase in the likelihood of fluid overload was observed (OR 1.128, 95% CI 1.028-1.238, p = 0.011). The MRC-ICU score was positively associated with fluid balance at day 3 (β-coefficient 218.455, 95% CI 94.693-342.217, p = 0.001) when controlling for age, gender, and SOFA score.
Conclusions
Medication regimen complexity demonstrated a weakly positive correlation with fluid overload in critically ill patients. Future studies are necessary to establish the MRC-ICU as a predictor to identify patients at risk of fluid overload.
Predicting the future course of critical conditions involves personal experience, heuristics and statistical models. Although these methods may perform well for some cases and population averages, they suffer from substantial shortcomings when applied to individual patients. The reasons include methodological problems of statistical modeling as well as limitations of cross-sectional data sampling. Accurate predictions for individual patients become crucial when they have to guide irreversible decision-making. This notably applies to triage situations in response to a lack of healthcare resources. We will discuss these issues and argue that analysing longitudinal data obtained from time-limited trials in intensive care can provide a more robust approach to individual prognostication.
Despite evidence highlighting harms of fluid overload, minimal guidance exists on counteraction via utilization of diuretics in the de-resuscitation phase. While diuretics have been shown to decrease net volume and improve clinical outcomes in the critically ill, a lack of standardization surrounding selection of diuretic regimen or monitoring of de-resuscitation exists. Current monitoring parameters of de-resuscitation often rely on clinical signs of fluid overload, end organ recovery and other biochemical surrogate markers which are often deemed unreliable. The majority of evidence suggests that achieving a net-negative fluid balance within 72 h after shock resolution may be of benefit; however, approaches to such goal are uncertain. Loop diuretics are a widely available type of diuretic for removal of volume in patients with sufficient kidney function, with the potential for adjunct diuretics in special circumstances. At present, administration of diuretics within the broad critically ill population fails to find uniformity and often efficacy. Given the lack of randomized controlled trials in this susceptible population, we aim to provide a thorough therapeutic understanding of diuretic pharmacotherapy which is necessary in order to achieve desired goal of fluid balance and improve overall outcomes.
The Restricted Boltzmann Machine (RBM) has aroused wide interest in machine learning fields during the past decade. This review aims to report the recent developments in theoretical research and applications of the RBM. We first give an overview of the general RBM from the theoretical perspective, including stochastic approximation methods, stochastic gradient methods, and preventing overfitting methods. And then this review focuses on the RBM variants which further improve the learning ability of the RBM under general or specific applications. The RBM has recently been extended for representational learning, document modeling, multi-label learning, weakly supervised learning and many other tasks. The RBM and RBM variants provide powerful tools for representing dependency in the data, and they can be used as the basic building blocks to create deep networks. Apart from the Deep Belief Network (DBN) and the Deep Boltzmann Machine (DBM), the RBM can also be combined with the Convolutional Neural Network (CNN) to create deep networks. This review provides a comprehensive view of these advances in the RBM together with its future perspectives.
Rationale:
Survivors of septic shock have impaired functional status. Volume overload is associated with poor outcomes in patients with septic shock, but the impact of volume overload on functional outcome and discharge destination of survivors is unknown.
Objectives:
This study describes patterns of fluid management both during and after septic shock, and examines factors associated with volume overload on intensive care unit (ICU) discharge. We then examine associations between volume overload on ICU discharge and mobility limitation and discharge to a healthcare facility in septic shock survivors, with the hypothesis that volume overload is associated with increased odds of these outcomes.
Methods:
We retrospectively reviewed the medical records of 247 patients admitted with septic shock to an academic county hospital from June 2009 to April 2012, who survived to ICU discharge. We defined volume overload as a fluid balance expected to increase the subject's admission weight by 10%. Statistical methods included unadjusted analyses and multivariable logistic regression.
Measurements and main results:
Eighty-six percent of patients had a positive fluid balance and 35% had volume overload on ICU discharge. Factors associated with volume overload in unadjusted analyses included more severe illness, cirrhosis, blood transfusion during shock, and higher volumes of fluid administration both during and after shock. Blood transfusion during shock was independently associated with increased odds of volume overload (odds ratio [OR] 2.65, 95% confidence interval [CI] 1.33-5.27, p=0.01) after adjusting for pre-existing conditions and severity of illness. Only 42% of patients received at least one dose of diuretic during their hospitalization. Volume overload on ICU discharge was independently associated with inability to ambulate on hospital discharge (OR 2.29, 95% CI 1.24-4.25, p=0.01) and, in patients admitted from home, discharge to a healthcare facility (OR 2.34, 95% CI 1.1-4.98, p=0.03).
Conclusions:
Volume overload is independently associated with impaired mobility and discharge to a healthcare facility in survivors of septic shock. Prevention and treatment of volume overload in patients with septic shock warrants further investigation.
Restricted Boltzmann machines (RBMs) have been used as generative models of many different types of data. RBMs are usually trained using the contrastive divergence learning procedure. This requires a certain amount of practical experience to decide how to set the values of numerical meta-parameters. Over the last few years, the machine learning group at the University of Toronto has acquired considerable expertise at training RBMs and this guide is an attempt to share this expertise with other machine learning researchers.
Fluid accumulation is associated with adverse outcomes in critically ill patients. Here, we sought to determine if fluid accumulation is associated with mortality and non-recovery of kidney function in critically ill adults with acute kidney injury. Fluid overload was defined as more than a 10% increase in body weight relative to baseline, measured in 618 patients enrolled in a prospective multicenter observational study. Patients with fluid overload experienced significantly higher mortality within 60 days of enrollment. Among dialyzed patients, survivors had significantly lower fluid accumulation when dialysis was initiated compared to non-survivors after adjustments for dialysis modality and severity score. The adjusted odds ratio for death associated with fluid overload at dialysis initiation was 2.07. In non-dialyzed patients, survivors had significantly less fluid accumulation at the peak of their serum creatinine. Fluid overload at the time of diagnosis of acute kidney injury was not associated with recovery of kidney function. However, patients with fluid overload when their serum creatinine reached its peak were significantly less likely to recover kidney function. Our study shows that in patients with acute kidney injury, fluid overload was independently associated with mortality. Whether the fluid overload was the result of a more severe renal failure or it contributed to its cause will require clinical trials in which the role of fluid administration to such patients is directly tested.
Improving irregular temporal modeling by integrating synthetic data to the electronic medical record using conditional GANs: a case study of fluid overload prediction in the intensive care unit
- Rafiei A
Syuand/unsupervisedfoprediction-: A time-dependent unsupervised machine learning (ML) analysis was conducted to determine patterns of IV medication use associated with fo
- Dengs