Reinforcement learning for intensive care medicine: actionable clinical insights from novel approaches to reward shaping and off-policy model evaluation.

BACKGROUND: Reinforcement learning (RL) holds great promise for intensive care medicine given the abundant availability of data and frequent sequential decision-making. But despite the emergence of promising algorithms, RL driven bedside clinical decision support is still far from reality. Major challenges include trust and safety. To help address these issues, we introduce cross off-policy evaluation and policy restriction and show how detailed policy analysis may increase clinical interpretability. As an example, we apply these in the setting of RL to optimise ventilator settings in intubated covid-19 patients. METHODS: With data from the Dutch ICU Data Warehouse and using an exhaustive hyperparameter grid search, we identified an optimal set of Dueling Double-Deep Q Network RL models. The state space comprised ventilator, medication, and clinical data. The action space focused on positive end-expiratory pressure (peep) and fraction of inspired oxygen (FiO2) concentration. We used gas exchange indices as interim rewards, and mortality and state duration as final rewards. We designed a novel evaluation method called cross off-policy evaluation (OPE) to assess the efficacy of models under varying weightings between the interim and terminal reward components. In addition, we implemented policy restriction to prevent potentially hazardous model actions. We introduce delta-Q to compare physician versus policy action quality and in-depth policy inspection using visualisations. RESULTS: We created trajectories for 1118 intensive care unit (ICU) admissions and trained 69,120 models using 8 model architectures with 128 hyperparameter combinations. For each model, policy restrictions were applied. In the first evaluation step, 17,182/138,240 policies had good performance, but cross-OPE revealed suboptimal performance for 44% of those by varying the reward function used for evaluation. Clinical policy inspection facilitated assessment of action decisions for individual patients, including identification of action space regions that may benefit most from optimisation. CONCLUSION: Cross-OPE can serve as a robust evaluation framework for safe RL model implementation by identifying policies with good generalisability. Policy restriction helps prevent potentially unsafe model recommendations. Finally, the novel delta-Q metric can be used to operationalise RL models in clinical practice. Our findings offer a promising pathway towards application of RL in intensive care medicine and beyond.

Artificial intelligence to advance acute and intensive care medicine.

PURPOSE OF REVIEW: This review explores recent key advancements in artificial intelligence for acute and intensive care medicine. As artificial intelligence rapidly evolves, this review aims to elucidate its current applications, future possibilities, and the vital challenges that are associated with its integration into emergency medical dispatch, triage, medical consultation and ICUs. RECENT FINDINGS: The integration of artificial intelligence in emergency medical dispatch (EMD) facilitates swift and accurate assessment. In the emergency department (ED), artificial intelligence driven triage models leverage diverse patient data for improved outcome predictions, surpassing human performance in retrospective studies. Artificial intelligence can streamline medical documentation in the ED and enhances medical imaging interpretation. The introduction of large multimodal generative models showcases the future potential to process varied biomedical data for comprehensive decision support. In the ICU, artificial intelligence applications range from early warning systems to treatment suggestions. SUMMARY: Despite promising academic strides, widespread artificial intelligence adoption in acute and critical care is hindered by ethical, legal, technical, organizational, and validation challenges. Despite these obstacles, artificial intelligence's potential to streamline clinical workflows is evident. When these barriers are overcome, future advancements in artificial intelligence have the potential to transform the landscape of patient care for acute and intensive care medicine.

Does Reinforcement Learning Improve Outcomes for Critically Ill Patients? A Systematic Review and Level-of-Readiness Assessment.

OBJECTIVE: Reinforcement learning (RL) is a machine learning technique uniquely effective at sequential decision-making, which makes it potentially relevant to ICU treatment challenges. We set out to systematically review, assess level-of-readiness and meta-analyze the effect of RL on outcomes for critically ill patients. DATA SOURCES: A systematic search was performed in PubMed, Embase.com, Clarivate Analytics/Web of Science Core Collection, Elsevier/SCOPUS and the Institute of Electrical and Electronics Engineers Xplore Digital Library from inception to March 25, 2022, with subsequent citation tracking. DATA EXTRACTION: Journal articles that used an RL technique in an ICU population and reported on patient health-related outcomes were included for full analysis. Conference papers were included for level-of-readiness assessment only. Descriptive statistics, characteristics of the models, outcome compared with clinician's policy and level-of-readiness were collected. RL-health risk of bias and applicability assessment was performed. DATA SYNTHESIS: A total of 1,033 articles were screened, of which 18 journal articles and 18 conference papers, were included. Thirty of those were prototyping or modeling articles and six were validation articles. All articles reported RL algorithms to outperform clinical decision-making by ICU professionals, but only in retrospective data. The modeling techniques for the state-space, action-space, reward function, RL model training, and evaluation varied widely. The risk of bias was high in all articles, mainly due to the evaluation procedure. CONCLUSION: In this first systematic review on the application of RL in intensive care medicine we found no studies that demonstrated improved patient outcomes from RL-based technologies. All studies reported that RL-agent policies outperformed clinician policies, but such assessments were all based on retrospective off-policy evaluation.

Augmented intelligence facilitates concept mapping across different electronic health records.

INTRODUCTION: With the advent of artificial intelligence, the secondary use of routinely collected medical data from electronic healthcare records (EHR) has become increasingly popular. However, different EHR systems typically use different names for the same medical concepts. This obviously hampers scalable model development and subsequent clinical implementation for decision support. Therefore, converting original parameter names to a so-called ontology, a standardized set of predefined concepts, is necessary but time-consuming and labor-intensive. We therefore propose an augmented intelligence approach to facilitate ontology alignment by predicting correct concepts based on parameter names from raw electronic health record data exports. METHODS: We used the manually mapped parameter names from the multicenter "Dutch ICU data warehouse against COVID-19" sourced from three types of EHR systems to train machine learning models for concept mapping. Data from 29 intensive care units on 38,824 parameters mapped to 1,679 relevant and unique concepts and 38,069 parameters labeled as irrelevant were used for model development and validation. We used the Natural Language Toolkit (NLTK) to preprocess the parameter names based on WordNet cognitive synonyms transformed by term-frequency inverse document frequency (TF-IDF), yielding numeric features. We then trained linear classifiers using stochastic gradient descent for multi-class prediction. Finally, we fine-tuned these predictions using information on distributions of the data associated with each parameter name through similarity score and skewness comparisons. RESULTS: The initial model, trained using data from one hospital organization for each of three EHR systems, scored an overall top 1 precision of 0.744, recall of 0.771, and F1-score of 0.737 on a total of 58,804 parameters. Leave-one-hospital-out analysis returned an average top 1 recall of 0.680 for relevant parameters, which increased to 0.905 for the top 5 predictions. When reducing the training dataset to only include relevant parameters, top 1 recall was 0.811 and top 5 recall was 0.914 for relevant parameters. Performance improvement based on similarity score or skewness comparisons affected at most 5.23% of numeric parameters. CONCLUSION: Augmented intelligence is a promising method to improve concept mapping of parameter names from raw electronic health record data exports. We propose a robust method for mapping data across various domains, facilitating the integration of diverse data sources. However, recall is not perfect, and therefore manual validation of mapping remains essential.

Determining and assessing characteristics of data element names impacting the performance of annotation using Usagi.

INTRODUCTION: Hospitals generate large amounts of data and this data is generally modeled and labeled in a proprietary way, hampering its exchange and integration. Manually annotating data element names to internationally standardized data element identifiers is a time-consuming effort. Tools can support performing this task automatically. This study aimed to determine what factors influence the quality of automatic annotations. METHODS: Data element names were used from the Dutch COVID-19 ICU Data Warehouse containing data on intensive care patients with COVID-19 from 25 hospitals in the Netherlands. In this data warehouse, the data had been merged using a proprietary terminology system while also storing the original hospital labels (synonymous names). Usagi, an OHDSI annotation tool, was used to perform the annotation for the data. A gold standard was used to determine if Usagi made correct annotations. Logistic regression was used to determine if the number of characters, number of words, match score (Usagi's certainty) and hospital label origin influenced Usagi's performance to annotate correctly. RESULTS: Usagi automatically annotated 30.5% of the data element names correctly and 5.5% of the synonymous names. The match score is the best predictor for Usagi finding the correct annotation. It was determined that the AUC of data element names was 0.651 and 0.752 for the synonymous names respectively. The AUC for the individual hospital label origins varied between 0.460 to 0.905. DISCUSSION: The results show that Usagi performed better to annotate the data element names than the synonymous names. The hospital origin in the synonymous names dataset was associated with the amount of correctly annotated concepts. Hospitals that performed better had shorter synonymous names and fewer words. Using shorter data element names or synonymous names should be considered to optimize the automatic annotating process. Overall, the performance of Usagi is too poor to completely rely on for automatic annotation.

Assessing the FAIRness of databases on the EHDEN portal: A case study on two Dutch ICU databases.

OBJECTIVE: To address the growing need for effective data reuse in health research, healthcare institutions need to make their data Findable, Accessible, Interoperable, and Reusable (FAIR). A prevailing method to model databases for interoperability is the Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM), developed by the Observational Health Data Sciences and Informatics (OHDSI) initiative. A European repository for OMOP CDM-converted databases called the "European Health Data & Evidence Network (EHDEN) portal" was developed, aiming to make these databases Findable and Accessible. This paper aims to assess the FAIRness of databases on the EHDEN portal. MATERIALS AND METHODS: Two researchers involved in the OMOP CDM conversion of separate Dutch Intensive Care Unit (ICU) research databases each manually assessed their own database using seventeen metrics. These were defined by the FAIRsFAIR project as a list of minimum requirements for a database to be FAIR. Each metric is given a score from zero to four based on how well the database adheres to the metric. The maximum score for each metric varies from one to four based on the importance of the metric. RESULTS: Fourteen out of the seventeen metrics were unanimously rated: seven were rated the highest score, one was rated half of the highest score, and five were rated the lowest score. The remaining three metrics were assessed differently for the two use cases. The total scores achieved were 15.5 and 12 out of a maximum of 25. CONCLUSION: The main omissions in supporting FAIRness were the lack of globally unique identifiers such as Uniform Resource Identifiers (URIs) in the OMOP CDM and the lack of metadata standardization and linkage in the EHDEN portal. By implementing these in future updates, the EHDEN portal can be more FAIR.

A guide to sharing open healthcare data under the General Data Protection Regulation.

Sharing healthcare data is increasingly essential for developing data-driven improvements in patient care at the Intensive Care Unit (ICU). However, it is also very challenging under the strict privacy legislation of the European Union (EU). Therefore, we explored four successful open ICU healthcare databases to determine how open healthcare data can be shared appropriately in the EU. A questionnaire was constructed based on the Delphi method. Then, follow-up questions were discussed with experts from the four databases. These experts encountered similar challenges and regarded ethical and legal aspects to be the most challenging. Based on the approaches of the databases, expert opinion, and literature research, we outline four distinct approaches to openly sharing healthcare data, each with varying implications regarding data security, ease of use, sustainability, and implementability. Ultimately, we formulate seven recommendations for sharing open healthcare data to guide future initiatives in sharing open healthcare data to improve patient care and advance healthcare.

Evolution of Clinical Phenotypes of COVID-19 Patients During Intensive Care Treatment: An Unsupervised Machine Learning Analysis.

BACKGROUND: Identification of clinical phenotypes in critically ill COVID-19 patients could improve understanding of the disease heterogeneity and enable prognostic and predictive enrichment. However, previous attempts did not take into account temporal dynamics with high granularity. By including the dimension of time, we aim to gain further insights into the heterogeneity of COVID-19. METHODS: We used granular data from 3202 adult COVID patients in the Dutch Data Warehouse that were admitted to one of 25 Dutch ICUs between February 2020 and March 2021. Parameters including demographics, clinical observations, medications, laboratory values, vital signs, and data from life support devices were selected. Twenty-one datasets were created that each covered 24 h of ICU data for each day of ICU treatment. Clinical phenotypes in each dataset were identified by performing cluster analyses. Both evolution of the clinical phenotypes over time and patient allocation to these clusters over time were tracked. RESULTS: The final patient cohort consisted of 2438 COVID-19 patients with a ICU mortality outcome. Forty-one parameters were chosen for cluster analysis. On admission, both a mild and a severe clinical phenotype were found. After day 4, the severe phenotype split into an intermediate and a severe phenotype for 11 consecutive days. Heterogeneity between phenotypes appears to be driven by inflammation and dead space ventilation. During the 21-day period, only 8.2% and 4.6% of patients in the initial mild and severe clusters remained assigned to the same phenotype respectively. The clinical phenotype half-life was between 5 and 6 days for the mild and severe phenotypes, and about 3 days for the medium severe phenotype. CONCLUSIONS: Patients typically do not remain in the same cluster throughout intensive care treatment. This may have important implications for prognostic or predictive enrichment. Prominent dissimilarities between clinical phenotypes are predominantly driven by inflammation and dead space ventilation.

Predicting Readmission or Death After Discharge From the ICU: External Validation and Retraining of a Machine Learning Model.

OBJECTIVES: Many machine learning (ML) models have been developed for application in the ICU, but few models have been subjected to external validation. The performance of these models in new settings therefore remains unknown. The objective of this study was to assess the performance of an existing decision support tool based on a ML model predicting readmission or death within 7 days after ICU discharge before, during, and after retraining and recalibration. DESIGN: A gradient boosted ML model was developed and validated on electronic health record data from 2004 to 2021. We performed an independent validation of this model on electronic health record data from 2011 to 2019 from a different tertiary care center. SETTING: Two ICUs in tertiary care centers in The Netherlands. PATIENTS: Adult patients who were admitted to the ICU and stayed for longer than 12 hours. INTERVENTIONS: None. MEASUREMENTS AND MAIN RESULTS: We assessed discrimination by area under the receiver operating characteristic curve (AUC) and calibration (slope and intercept). We retrained and recalibrated the original model and assessed performance via a temporal validation design. The final retrained model was cross-validated on all data from the new site. Readmission or death within 7 days after ICU discharge occurred in 577 of 10,052 ICU admissions (5.7%) at the new site. External validation revealed moderate discrimination with an AUC of 0.72 (95% CI 0.67-0.76). Retrained models showed improved discrimination with AUC 0.79 (95% CI 0.75-0.82) for the final validation model. Calibration was poor initially and good after recalibration via isotonic regression. CONCLUSIONS: In this era of expanding availability of ML models, external validation and retraining are key steps to consider before applying ML models to new settings. Clinicians and decision-makers should take this into account when considering applying new ML models to their local settings.

Right dose, right now: bedside, real-time, data-driven, and personalised antibiotic dosing in critically ill patients with sepsis or septic shock-a two-centre randomised clinical trial.

BACKGROUND: Adequate antibiotic dosing may improve outcomes in critically ill patients but is challenging due to altered and variable pharmacokinetics. To address this challenge, AutoKinetics was developed, a decision support system for bedside, real-time, data-driven and personalised antibiotic dosing. This study evaluates the feasibility, safety and efficacy of its clinical implementation. METHODS: In this two-centre randomised clinical trial, critically ill patients with sepsis or septic shock were randomised to AutoKinetics dosing or standard dosing for four antibiotics: vancomycin, ciprofloxacin, meropenem, and ceftriaxone. Adult patients with a confirmed or suspected infection and either lactate > 2 mmol/L or vasopressor requirement were eligible for inclusion. The primary outcome was pharmacokinetic target attainment in the first 24 h after randomisation. Clinical endpoints included mortality, ICU length of stay and incidence of acute kidney injury. RESULTS: After inclusion of 252 patients, the study was stopped early due to the COVID-19 pandemic. In the ciprofloxacin intervention group, the primary outcome was obtained in 69% compared to 3% in the control group (OR 62.5, CI 11.4-1173.78, p < 0.001). Furthermore, target attainment was faster (26 h, CI 18-42 h, p < 0.001) and better (65% increase, CI 49-84%, p < 0.001). For the other antibiotics, AutoKinetics dosing did not improve target attainment. Clinical endpoints were not significantly different. Importantly, higher dosing did not lead to increased mortality or renal failure. CONCLUSIONS: In critically ill patients, personalised dosing was feasible, safe and significantly improved target attainment for ciprofloxacin. TRIAL REGISTRATION: The trial was prospectively registered at Netherlands Trial Register (NTR), NL6501/NTR6689 on 25 August 2017 and at the European Clinical Trials Database (EudraCT), 2017-002478-37 on 6 November 2017.

Fluid Management, Intra-Abdominal Hypertension and the Abdominal Compartment Syndrome: A Narrative Review.

BACKGROUND: General pathophysiological mechanisms regarding associations between fluid administration and intra-abdominal hypertension (IAH) are evident, but specific effects of type, amount, and timing of fluids are less clear. OBJECTIVES: This review aims to summarize current knowledge on associations between fluid administration and intra-abdominal pressure (IAP) and fluid management in patients at risk of intra-abdominal hypertension and abdominal compartment syndrome (ACS). METHODS: We performed a structured literature search from 1950 until May 2021 to identify evidence of associations between fluid management and intra-abdominal pressure not limited to any specific study or patient population. Findings were summarized based on the following information: general concepts of fluid management, physiology of fluid movement in patients with intra-abdominal hypertension, and data on associations between fluid administration and IAH. RESULTS: We identified three randomized controlled trials (RCTs), 38 prospective observational studies, 29 retrospective studies, 18 case reports in adults, two observational studies and 10 case reports in children, and three animal studies that addressed associations between fluid administration and IAH. Associations between fluid resuscitation and IAH were confirmed in most studies. Fluid resuscitation contributes to the development of IAH. However, patients with IAH receive more fluids to manage the effect of IAH on other organ systems, thereby causing a vicious cycle. Timing and approach to de-resuscitation are of utmost importance, but clear indicators to guide this decision-making process are lacking. In selected cases, only surgical decompression of the abdomen can stop deterioration and prevent further morbidity and mortality. CONCLUSIONS: Current evidence confirms an association between fluid resuscitation and secondary IAH, but optimal fluid management strategies for patients with IAH remain controversial.

Assess and validate predictive performance of models for in-hospital mortality in COVID-19 patients: A retrospective cohort study in the Netherlands comparing the value of registry data with high-granular electronic health records.

PURPOSE: To assess, validate and compare the predictive performance of models for in-hospital mortality of COVID-19 patients admitted to the intensive care unit (ICU) over two different waves of infections. Our models were built with high-granular Electronic Health Records (EHR) data versus less-granular registry data. METHODS: Observational study of all COVID-19 patients admitted to 19 Dutch ICUs participating in both the national quality registry National Intensive Care Evaluation (NICE) and the EHR-based Dutch Data Warehouse (hereafter EHR). Multiple models were developed on data from the first 24 h of ICU admissions from February to June 2020 (first COVID-19 wave) and validated on prospective patients admitted to the same ICUs between July and December 2020 (second COVID-19 wave). We assessed model discrimination, calibration, and the degree of relatedness between development and validation population. Coefficients were used to identify relevant risk factors. RESULTS: A total of 1533 patients from the EHR and 1563 from the registry were included. With high granular EHR data, the average AUROC was 0.69 (standard deviation of 0.05) for the internal validation, and the AUROC was 0.75 for the temporal validation. The registry model achieved an average AUROC of 0.76 (standard deviation of 0.05) in the internal validation and 0.77 in the temporal validation. In the EHR data, age, and respiratory-system related variables were the most important risk factors identified. In the NICE registry data, age and chronic respiratory insufficiency were the most important risk factors. CONCLUSION: In our study, prognostic models built on less-granular but readily-available registry data had similar performance to models built on high-granular EHR data and showed similar transportability to a prospective COVID-19 population. Future research is needed to verify whether this finding can be confirmed for upcoming waves.

Co-infection and ICU-acquired infection in COIVD-19 ICU patients: a secondary analysis of the UNITE-COVID data set.

BACKGROUND: The COVID-19 pandemic presented major challenges for critical care facilities worldwide. Infections which develop alongside or subsequent to viral pneumonitis are a challenge under sporadic and pandemic conditions; however, data have suggested that patterns of these differ between COVID-19 and other viral pneumonitides. This secondary analysis aimed to explore patterns of co-infection and intensive care unit-acquired infections (ICU-AI) and the relationship to use of corticosteroids in a large, international cohort of critically ill COVID-19 patients. METHODS: This is a multicenter, international, observational study, including adult patients with PCR-confirmed COVID-19 diagnosis admitted to ICUs at the peak of wave one of COVID-19 (February 15th to May 15th, 2020). Data collected included investigator-assessed co-infection at ICU admission, infection acquired in ICU, infection with multi-drug resistant organisms (MDRO) and antibiotic use. Frequencies were compared by Pearson's Chi-squared and continuous variables by Mann-Whitney U test. Propensity score matching for variables associated with ICU-acquired infection was undertaken using R library MatchIT using the "full" matching method. RESULTS: Data were available from 4994 patients. Bacterial co-infection at admission was detected in 716 patients (14%), whilst 85% of patients received antibiotics at that stage. ICU-AI developed in 2715 (54%). The most common ICU-AI was bacterial pneumonia (44% of infections), whilst 9% of patients developed fungal pneumonia; 25% of infections involved MDRO. Patients developing infections in ICU had greater antimicrobial exposure than those without such infections. Incident density (ICU-AI per 1000 ICU days) was in considerable excess of reports from pre-pandemic surveillance. Corticosteroid use was heterogenous between ICUs. In univariate analysis, 58% of patients receiving corticosteroids and 43% of those not receiving steroids developed ICU-AI. Adjusting for potential confounders in the propensity-matched cohort, 71% of patients receiving corticosteroids developed ICU-AI vs 52% of those not receiving corticosteroids. Duration of corticosteroid therapy was also associated with development of ICU-AI and infection with an MDRO. CONCLUSIONS: In patients with severe COVID-19 in the first wave, co-infection at admission to ICU was relatively rare but antibiotic use was in substantial excess to that indication. ICU-AI were common and were significantly associated with use of corticosteroids. Trial registration ClinicalTrials.gov: NCT04836065 (retrospectively registered April 8th 2021).

Intravenous fluid therapy in perioperative and critical care setting-Knowledge test and practice: An international cross-sectional survey.

PURPOSE: In the absence of recent international recommendations supported by scientific societies like Anesthesiology or Intensive Care Medicine, healthcare professionals (HCP) knowledge on IV fluid is expected to vary. We undertook a cross-sectional survey, aiming to assess prescription patterns and test the knowledge of HCP for IV fluid use in the operating room (OR) and intensive care unit (ICU). METHODS: An online international cross-sectional survey was conducted between October 20, 2019, and November 27, 2021. The survey included multiple-choice questions on demographics, practice patterns and knowledge of IV fluids, and a hemodynamically unstable patient assessment. RESULTS: 1045 HCP, from 97 countries responded to the survey. Nearly three-quarters reported the non-existence of internal hospital or ICU-based guidelines on IV fluids. The respondents' mean score on the knowledge assessment questions was 46.4 ± 14.4. The cumulative mean scores were significantly higher among those supervising trainees (p = 0.02), specialists (p < 0.001) and those working in high-income (p < 0.001) countries. Overall performance of respondents on the knowledge testing for IV fluid was unsatisfactory with only 6.5% respondents performed above average. CONCLUSION: There is a wide difference in the knowledge and prescription of IV fluids among the HCP surveyed. These findings reflect the urgent need for education on IV fluids.

Ischaemic electrocardiogram patterns and its association with survival in out-of-hospital cardiac arrest patients without ST-segment elevation myocardial infarction: a COACT trials' post-hoc subgroup analysis.

AIMS: ST-depression and T-wave inversion are frequently present on the post-resuscitation electrocardiogram (ECG). However, the prognostic value of ischaemic ECG patterns is unknown. METHODS AND RESULTS: In this post-hoc subgroup analysis of the Coronary Angiography after Cardiac arrest (COACT) trial, the first in-hospital post-resuscitation ECG in out-of-hospital cardiac arrest patients with a shockable rhythm was analysed for ischaemic ECG patterns. Ischaemia was defined as ST-depression of ≥0.1 mV, T-wave inversion in ≥2 contiguous leads, or both. The primary endpoint was 90-day survival. Secondary endpoints were rate of acute unstable lesions, levels of serum troponin-T, and left ventricular function. Of the 510 out-of-hospital cardiac arrest patients, 340 (66.7%) patients had ischaemic ECG patterns. Patients with ischaemic ECG patterns had a worse 90-day survival compared with those without [hazard ratio 1.51; 95% confidence interval (CI) 1.08-2.12; P = 0.02]. A higher sum of ST-depression was associated with lower survival (log-rank = 0.01). The rate of acute unstable lesions (14.5 vs. 15.8%; odds ratio 0.90; 95% CI 0.51-1.59) did not differ between the groups. In patients with ischaemic ECG patterns, maximum levels of serum troponin-T (µg/L) were higher [0.595 (interquartile range 0.243-1.430) vs. 0.359 (0.159-0.845); ratio of geometric means 1.58; 1.13-2.20] and left ventricular function (%) was worse (44.7 ± 12.5 vs. 49.9 ± 13.3; mean difference -5.13; 95% CI -8.84 to -1.42). Adjusted for age and time to return of spontaneous circulation, ischaemic ECG patterns were no longer associated with survival. CONCLUSION: Post-arrest ischaemic ECG patterns were associated with worse 90-day survival. A higher sum of ST-depression was associated with lower survival. Adjusted for age and time to return of spontaneous circulation, ischaemic ECG patterns were no longer associated with survival.

Pooled Population Pharmacokinetic Analysis for Exploring Ciprofloxacin Pharmacokinetic Variability in Intensive Care Patients.

BACKGROUND AND OBJECTIVE: Previous pharmacokinetic (PK) studies of ciprofloxacin in intensive care (ICU) patients have shown large differences in estimated PK parameters, suggesting that further investigation is needed for this population. Hence, we performed a pooled population PK analysis of ciprofloxacin after intravenous administration using individual patient data from three studies. Additionally, we studied the PK differences between these studies through a post-hoc analysis. METHODS: Individual patient data from three studies (study 1, 2, and 3) were pooled. The pooled data set consisted of 1094 ciprofloxacin concentration-time data points from 140 ICU patients. Nonlinear mixed-effects modeling was used to develop a population PK model. Covariates were selected following a stepwise covariate modeling procedure. To analyze PK differences between the three original studies, random samples were drawn from the posterior distribution of individual PK parameters. These samples were used for a simulation study comparing PK exposure and the percentage of target attainment between patients of these studies. RESULTS: A two-compartment model with first-order elimination best described the data. Inter-individual variability was added to the clearance, central volume, and peripheral volume. Inter-occasion variability was added to clearance only. Body weight was added to all parameters allometrically. Estimated glomerular filtration rate on ciprofloxacin clearance was identified as the only covariate relationship resulting in a drop in inter-individual variability of clearance from 58.7 to 47.2%. In the post-hoc analysis, clearance showed the highest deviation between the three studies with a coefficient of variation of 14.3% for posterior mean and 24.1% for posterior inter-individual variability. The simulation study showed that following the same dose regimen of 400 mg three times daily, the area under the concentration-time curve of study 3 was the highest with a mean area under the concentration-time curve at 24 h of 58 mg·h/L compared with that of 47.7 mg·h/L for study 1 and 47.6 mg·h/L for study 2. Similar differences were also observed in the percentage of target attainment, defined as the ratio of area under the concentration-time curve at 24 h and the minimum inhibitory concentration. At the epidemiological cut-off minimum inhibitory concentration of Pseudomonas aeruginosa of 0.5 mg/L, percentage of target attainment was only 21%, 18%, and 38% for study 1, 2, and 3, respectively. CONCLUSIONS: We developed a population PK model of ciprofloxacin in ICU patients using pooled data of individual patients from three studies. A simple ciprofloxacin dose recommendation for the entire ICU population remains challenging owing to the PK differences within ICU patients, hence dose individualization may be needed for the optimization of ciprofloxacin treatment.

The Potential Cost-Effectiveness of a Machine Learning Tool That Can Prevent Untimely Intensive Care Unit Discharge.

OBJECTIVES: The machine learning prediction model Pacmed Critical (PC), currently under development, may guide intensivists in their decision-making process on the most appropriate time to discharge a patient from the intensive care unit (ICU). Given the financial pressure on healthcare budgets, this study assessed whether PC has the potential to be cost-effective compared with standard care, without the use of PC, for Dutch patients in the ICU from a societal perspective. METHODS: A 1-year, 7-state Markov model reflecting the ICU care pathway and incorporating the PC decision tool was developed. A hypothetical cohort of 1000 adult Dutch patients admitted in the ICU was entered in the model. We used the literature, expert opinion, and data from Amsterdam University Medical Center for model parameters. The uncertainty surrounding the incremental cost-effectiveness ratio was assessed using deterministic and probabilistic sensitivity analyses and scenario analyses. RESULTS: PC was a cost-effective strategy with an incremental cost-effectiveness ratio of €18 507 per quality-adjusted life-year. PC remained cost-effective over standard care in multiple scenarios and sensitivity analyses. The likelihood that PC will be cost-effective was 71% at a willingness-to-pay threshold of €30 000 per quality-adjusted life-year. The key driver of the results was the parameter "reduction in ICU length of stay." CONCLUSIONS: We showed that PC has the potential to be cost-effective for Dutch ICUs in a time horizon of 1 year. This study is one of the first cost-effectiveness analyses of a machine learning device. Further research is needed to validate the effectiveness of PC, thereby focusing on the key parameter "reduction in ICU length of stay" and potential spill-over effects.