Study protocol for ADAPT-TDM: A beta-lactam antibiotic Dose AdaPtation feasibility randomised controlled Trial using Therapeutic Drug Monitoring.

INTRODUCTION: Critically ill patients are at risk of suboptimal beta-lactam antibiotic (beta-lactam) exposure due to the impact of altered physiology on pharmacokinetics. Suboptimal concentrations can lead to treatment failure or toxicity. Therapeutic drug monitoring (TDM) involves adjusting doses based on measured plasma concentrations and individualising dosing to improve the likelihood of improving exposure. Despite its potential benefits, its adoption has been slow, and data on implementation, dose adaptation and safety are sparse. The aim of this trial is to assess the feasibility and fidelity of implementing beta-lactam TDM-guided dosing in the intensive care unit setting. METHODS AND ANALYSIS: A beta-lactam antibiotic Dose AdaPtation feasibility randomised controlled Trial using Therapeutic Drug Monitoring (ADAPT-TDM) is a single-centre, unblinded, feasibility randomised controlled trial aiming to enroll up to 60 critically ill adult participants (≥18 years). TDM and dose adjustment will be performed daily in the intervention group; the standard of care group will undergo plasma sampling, but no dose adjustment. The main outcomes include: (1) feasibility of recruitment, defined as the number of participants who are recruited from a pool of eligible participants, and (2) fidelity of TDM, defined as the degree to which TDM as a test is delivered as intended, from accurate sample collection, sample processing to result availability. Secondary outcomes include target attainment, uptake of TDM-guided dosing and incidence of neurotoxicity, hepatotoxicity and nephrotoxicity. ETHICS AND DISSEMINATION: This study has been approved by the Alfred Hospital human research ethics committee, Office of Ethics and Research Governance (reference: Project No. 565/22; date of approval: 22/11/2022). Prospective consent will be obtained and the study will be conducted in accordance with the Declaration of Helsinki. The finalised manuscript, including aggregate data, will be submitted for publication in a peer reviewed journal. ADAPT-TDM will determine whether beta-lactam TDM-guided dose adaptation is reproducible and feasible and provide important information required to implement this intervention in a phase III trial. TRIAL REGISTRATION NUMBER: Australian New Zealand Clinical Trials Registry, ACTRN12623000032651.

Heparin is an effective treatment for preventing liver failure after hepatectomy.

BACKGROUND: Posthepatectomy liver failure (PHLF) is one of the most important causes of death following liver resection. Heparin, an established anticoagulant, can protect liver function through a number of mechanisms, and thus, prevent liver failure. AIM: To look at the safety and efficacy of heparin in preventing hepatic dysfunction after hepatectomy. METHODS: The data was extracted from Multiparameter Intelligent Monitoring in Intensive Care III (MIMIC-III) v1. 4 pinpointed patients who had undergone hepatectomy for liver cancer, subdividing them into two cohorts: Those who were injected with heparin and those who were not. The statistical evaluations used were unpaired t-tests, Mann-Whitney U tests, chi-square tests, and Fisher's exact tests to assess the effect of heparin administration on PHLF, duration of intensive care unit (ICU) stay, need for mechanical ventilation, use of continuous renal replacement therapy (CRRT), incidence of hypoxemia, development of acute kidney injury, and ICU mortality. Logistic regression was utilized to analyze the factors related to PHLF, with propensity score matching (PSM) aiming to balance the preoperative disparities between the two groups. RESULTS: In this study, 1388 patients who underwent liver cancer hepatectomy were analyzed. PSM yielded 213 matched pairs from the heparin-treated and control groups. Initial univariate analyses indicated that heparin potentially reduces the risk of PHLF in both matched and unmatched samples. Further analysis in the matched cohorts confirmed a significant association, with heparin reducing the risk of PHLF (odds ratio: 0.518; 95% confidence interval: 0.295-0.910; P = 0.022). Additionally, heparin treatment correlated with improved short-term postoperative outcomes such as reduced ICU stay durations, diminished requirements for respiratory support and CRRT, and lower incidences of hypoxemia and ICU mortality. CONCLUSION: Liver failure is an important hazard following hepatic surgery. During ICU care heparin administration has been proved to decrease the occurrence of hepatectomy induced liver failure. This indicates that heparin may provide a hopeful option for controlling PHLF.

Factors Associated With Quality-of-Dying-and-Death Classes Among Critically Ill Patients.

Importance: Improving end-of-life care in the intensive care unit (ICU) is a priority, but clinically modifiable factors of quality of dying and death (QODD) are seldom identified. Objectives: To comprehensively identify factors associated with QODD classes of dying ICU patients, emphasizing clinically modifiable factors based on the integrative framework of factors associated with for bereavement outcomes. Design, Setting, and Participants: This observational cohort study was conducted at medical ICUs of 2 Taiwanese medical centers from January 2018 to March 2020 with follow-up through December 2022. Eligible participants included primary family surrogates responsible for decision making for critically ill ICU patients at high risk of death (Acute Physiology and Chronic Health Evaluation II score >20) but who survived more than 3 days after ICU admission. Data analysis was conducted from July to September 2023. Main Outcomes and Measures: QODD was measured by the 23-item ICU-QODD questionnaire. Factors associated with patient membership in 4 previously determined QODD classes (high, moderate, poor to uncertain, and worst) were examined using a 3-step approach for latent class modeling with the high QODD class as the reference category. Results: A total of 309 family surrogates (mean [SD] age, 49.83 [12.55] years; 184 women [59.5%] and 125 men [40.5%]) were included in the study. Of all surrogates, 91 (29.4%) were the patients' spouse and 66 (53.7%) were the patients' adult child. Patient demographics were not associated with QODD class. Two family demographics (age and gender), relationship with the patient (spousal or adult-child), and length of ICU stay were associated with QODD classes. Patients of surrogates perceiving greater social support were less likely to be in the poor to uncertain (adjusted odds ratio [aOR], 0.89; 95% CI, 0.83-0.94) and worst (aOR, 0.92; 95% CI, 0.87-0.96) QODD classes. Family meetings were associated with the poor to uncertain QODD class (aOR, 8.61; 95% CI, 2.49-29.74) and worst QODD class (aOR, 7.28; 95% CI, 1.37-38.71). Death with cardiopulmonary resuscitation was associated with the worst QODD class (aOR, 7.51; 95% CI, 1.12-50.25). Family presence at patient death was uniformly negatively associated with the moderate QODD class (aOR, 0.16; 95% CI, 0.05-0.54), poor to uncertain QODD class (aOR, 0.21; 95% CI, 0.05-0.82), and worst QODD class (aOR, 0.08; 95% CI, 0.02-0.38). Higher family satisfaction with ICU care was negatively associated with the poor to uncertain QODD class (aOR, 0.93; 95% CI, 0.87-0.98) and worst QODD class (aOR, 0.86; 95% CI, 0.81-0.92). Conclusions and Relevance: In this cohort study of critically ill patients and their family surrogates, modifiable end-of-life ICU-care characteristics played a more significant role in associations with patient QODD class than did immutable family demographics, preexisting family health conditions, patient demographics, and patient clinical characteristics, thereby illuminating actionable opportunities to improve end-of-life ICU care.

My Magical Day in the Trauma ICU.

Healing is a difficult concept to describe, quantify, or replicate. It is a complex mixture of personal contributions from the professional providing care, including competence, compassion, and empathy, that conjoins with the needs, sensitivities, and receptivity of the one who is receiving the care. Although it may be difficult to predict all the elements that come together to initiate sustained healing, as well as the long-term impact, it is important to observe the moments that make a difference. For those who study the nature of healing, a patient's reflections can surface the kinds of elements that are present when healing is sustained. [J Contin Educ Nurs. 2024;55(7):326-327.].

The effect of web-based training given to nurses on the prevention of pressure injury in patients hospitalized in the anesthesiology and reanimation intensive care unit: quasi-experimental pre-test post-test research.

BACKGROUND: The knowledge, attitudes, and behaviors of intensive care nurses concerning the prevention of pressure injury (PI) may be positively affected by education. PURPOSE: To evaluate the effect of web-based training given to nurses on their knowledge of, attitudes about, and behaviors in the prevention of PI. METHODS: This study was conducted between May 2019 and December 2019 with a pre-test and post-test design. The study sample consisted of 22 nurses and 80 patients. A link to the educational video prepared for the prevention of PIs was sent to the nurses' mobile phones. RESULTS: The training had a significant positive effect on nurses' level of knowledge of and attitudes toward PI prevention (P < .001 and P = .042, respectively). In group 1, comprising 40 patients who received treatment before nurses' training, 2.5% of patients had stage 1 PI on day 1 and 7.5% had stage 1 PI on day 7, and 2.5% had stage 4 PI on day 7. In group 2, comprising 40 patients who received treatment after nurses' training, 2.5% of patients had stage 1 PI on day 1 and 2.5% had stage 1 PI on day 7. CONCLUSION: Nurses' knowledge of and attitudes and behaviors toward PI prevention were improved following the web-based training, and the stage and rate of PI were lower in patients who received care after nurses received the training.

MORTALITY RISK ASSESSMENT AMONG PATIENTS, HOSPITALIZED FOR COVID-19.

In Georgia, the number of confirmed cases of Coronavirus were 1,85,5289. Among them, 17 132 patients died. Information on risk factors for mortality is insufficient. The purpose of our research is to evaluate clinical features of heavy patients with severe COVID and determine prognostic factors of outcome. Factors associated with critical COVID-19 included older age and certain chronic medical conditions. The clinical material of 250 chronically ill COVID-19 patients admitted to the intensive care unit was retrospectively studied. We divided the patients into two groups. The dead and the survivors. Demographic data, comorbidities, chronic diseases, results of ultrasound, cardiography, computed tomography and laboratory characteristics were studied. In patients with chronic diseases, in the intensive care unit during COVID-19, the relative chance of survival decreases: CRP3 - OR=0.98(95% CI:0.97-0.99Hydrothorax- OR=0.24(95% CI:0.06-0.95); Sepsis/Septic shock - OR=0.07(95% CI:0.01-0.39); WBC - OR=0.86(95% CI:0.74-0.99); Mechanical lung ventilation - OR=0.01(95% CI:0.00-0.05)); increase survival relative chance- pO2 - OR=1.03(95% CI:1.0-1.06). Predictors of mortality in patients with chronic diseases: coagulation characteristics, inflammatory markers, sepsis, and artificial lung ventilation. Risk factors for covid-19 mortality need to be studied to increase pandemic preparedness.

Association between admission pan-immune-inflammation value and short-term mortality in septic patients: a retrospective cohort study.

Pan-Immune-Inflammation Value (PIV) has recently received more attention as a novel indicator of inflammation. We aimed to evaluate the association between PIV and prognosis in septic patients. Data were extracted from the Medical Information Mart for Intensive Care IV database. The primary and secondary outcomes were 28-day and 90-day mortality. The association between PIV and outcomes was assessed by Kaplan-Meier curves, Cox regression analysis, restricted cubic spline curves and subgroup analysis. A total of 11,331 septic patients were included. Kaplan-Meier curves showed that septic patients with higher PIV had lower 28-day survival rate. In multivariable Cox regression analysis, log2-PIV was positively associated with the risk of 28-day mortality [HR (95% CI) 1.06 (1.03, 1.09), P < 0.001]. The relationship between log2-PIV and 28-day mortality was non-linear with a predicted inflection point at 8. To the right of the inflection point, high log2-PIV was associated with an increased 28-day mortality risk [HR (95% CI) 1.13 (1.09, 1.18), P < 0.001]. However, to the left of this point, this association was non-significant [HR (95% CI) 1.01 (0.94, 1.08), P = 0.791]. Similar results were found for 90-day mortality. Our study showed a non-linear relationship between PIV and 28-day and 90-day mortality risk in septic patients.

Performance of the MRI lesion pattern score in predicting neurological outcome after out of hospital cardiac arrest: a retrospective cohort analysis.

BACKGROUND: Despite advances in resuscitation practice, patient survival following cardiac arrest remains poor. The utilization of MRI in neurological outcome prognostication post-cardiac arrest is growing and various classifications has been proposed; however a consensus has yet to be established. MRI, though valuable, is resource-intensive, time-consuming, costly, and not universally available. This study aims to validate a MRI lesion pattern score in a cohort of out of hospital cardiac arrest patients at a tertiary referral hospital in Switzerland. METHODS: This cohort study spanned twelve months from February 2021 to January 2022, encompassing all unconscious patients aged ≥ 18 years who experienced out-of-hospital cardiac arrest of any cause and were admitted to the intensive care unit (ICU) at Inselspital, University Hospital Bern, Switzerland. We included patients who underwent the neuroprognostication process, assessing the performance and validation of a MRI scoring system. RESULTS: Over the twelve-month period, 137 patients were admitted to the ICU, with 52 entering the neuroprognostication process and 47 undergoing MRI analysis. Among the 35 MRIs indicating severe hypoxic brain injury, 33 patients (94%) experienced an unfavourable outcome (UO), while ten (83%) of the twelve patients with no or minimal MRI lesions had a favourable outcome. This yielded a sensitivity of 0.94 and specificity of 0.83 for predicting UO with the proposed MRI scoring system. The positive and negative likelihood ratios were 5.53 and 0.07, respectively, resulting in an accuracy of 91.49%. CONCLUSION: We demonstrated the effectiveness of the MLP scoring scheme in predicting neurological outcome in patients following cardiac arrest. However, to ensure a comprehensive neuroprognostication, MRI results need to be combined with other assessments. While neuroimaging is a promising objective tool for neuroprognostication, given the absence of sedation-related confounders-compared to electroencephalogram (EEG) and clinical examination-the current lack of a validated scoring system necessitates further studies. Incorporating standardized MRI techniques and grading systems is crucial for advancing the reliability of neuroimaging for neuroprognostication. TRIAL REGISTRATION: Registry of all Projects in Switzerland (RAPS) 2020-01761.

Vitamin K: a potential missing link in critical illness-a scoping review.

BACKGROUND: Vitamin K is essential for numerous physiological processes, including coagulation, bone metabolism, tissue calcification, and antioxidant activity. Deficiency, prevalent in critically ill ICU patients, impacts coagulation and increases the risk of bleeding and other complications. This review aims to elucidate the metabolism of vitamin K in the context of critical illness and identify a potential therapeutic approach. METHODS: In December 2023, a scoping review was conducted using the PRISMA Extension for Scoping Reviews. Literature was searched in PubMed, Embase, and Cochrane databases without restrictions. Inclusion criteria were studies on adult ICU patients discussing vitamin K deficiency and/or supplementation. RESULTS: A total of 1712 articles were screened, and 13 met the inclusion criteria. Vitamin K deficiency in ICU patients is linked to malnutrition, impaired absorption, antibiotic use, increased turnover, and genetic factors. Observational studies show higher PIVKA-II levels in ICU patients, indicating reduced vitamin K status. Risk factors include inadequate intake, disrupted absorption, and increased physiological demands. Supplementation studies suggest vitamin K can improve status but not normalize it completely. Vitamin K deficiency may correlate with prolonged ICU stays, mechanical ventilation, and increased mortality. Factors such as genetic polymorphisms and disrupted microbiomes also contribute to deficiency, underscoring the need for individualized nutritional strategies and further research on optimal supplementation dosages and administration routes. CONCLUSIONS: Addressing vitamin K deficiency in ICU patients is crucial for mitigating risks associated with critical illness, yet optimal management strategies require further investigation. IMPACT RESEARCH: To the best of our knowledge, this review is the first to address the prevalence and progression of vitamin K deficiency in critically ill patients. It guides clinicians in diagnosing and managing vitamin K deficiency in intensive care and suggests practical strategies for supplementing vitamin K in critically ill patients. This review provides a comprehensive overview of the existing literature, and serves as a valuable resource for clinicians, researchers, and policymakers in critical care medicine.

POStoperative INTELLiVENT-adaptive support VEntilation in cardiac surgery patients (POSITiVE) II-study protocol of a randomized clinical trial.

BACKGROUND: One single-center randomized clinical trial showed that INTELLiVENT-adaptive support ventilation (ASV) is superior to conventional ventilation with respect to the quality of ventilation in post-cardiac surgery patients. Other studies showed that this automated ventilation mode reduces the number of manual interventions at the ventilator in various types of critically ill patients. In this multicenter study in patients post-cardiac surgery, we test the hypothesis that INTELLiVENT-ASV is superior to conventional ventilation with respect to the quality of ventilation. METHODS: "POStoperative INTELLiVENT-adaptive support VEntilation in cardiac surgery patients II (POSITiVE II)" is an international, multicenter, two-group randomized clinical superiority trial. In total, 328 cardiac surgery patients will be randomized. Investigators screen patients aged > 18 years of age, scheduled for elective cardiac surgery, and expected to receive postoperative ventilation in the ICU for longer than 2 h. Patients either receive automated ventilation by means of INTELLiVENT-ASV or ventilation that is not automated by means of a conventional ventilation mode. The primary endpoint is quality of ventilation, defined as the proportion of postoperative ventilation time characterized by exposure to predefined optimal, acceptable, and critical (injurious) ventilatory parameters in the first two postoperative hours. One major secondary endpoint is ICU team staff workload, captured by the ventilator software collecting manual settings on alarms. Patient-centered endpoints include duration of postoperative ventilation and length of stay in ICU. DISCUSSION: POSITiVE II is the first international, multicenter, randomized clinical trial designed to confirm that POStoperative INTELLiVENT-ASV is superior to non-automated conventional ventilation and secondary to determine if this closed-loop ventilation mode reduces ICU team staff workload. The results of POSITiVE II will support intensive care teams in their choices regarding the use of automated ventilation in postoperative care of uncomplicated cardiac surgery patients. TRIAL REGISTRATION: Clinicaltrials.gov NCT06178510 . Registered on December 4, 2023.

Construction and validation of a risk prediction model for postoperative ICU admission in patients with colorectal cancer: clinical prediction model study.

BACKGROUND: Transfer to the ICU is common following non-cardiac surgeries, including radical colorectal cancer (CRC) resection. Understanding the judicious utilization of costly ICU medical resources and supportive postoperative care is crucial. This study aimed to construct and validate a nomogram for predicting the need for mandatory ICU admission immediately following radical CRC resection. METHODS: Retrospective analysis was conducted on data from 1003 patients who underwent radical or palliative surgery for CRC at Ningxia Medical University General Hospital from August 2020 to April 2022. Patients were randomly assigned to training and validation cohorts in a 7:3 ratio. Independent predictors were identified using the least absolute shrinkage and selection operator (LASSO) and multivariate logistic regression in the training cohort to construct the nomogram. An online prediction tool was developed for clinical use. The nomogram's calibration and discriminative performance were assessed in both cohorts, and its clinical utility was evaluated through decision curve analysis (DCA). RESULTS: The final predictive model comprised age (P = 0.003, odds ratio [OR] 3.623, 95% confidence interval [CI] 1.535-8.551); nutritional risk screening 2002 (NRS2002) (P = 0.000, OR 6.129, 95% CI 2.920-12.863); serum albumin (ALB) (P = 0.013, OR 0.921, 95% CI 0.863-0.982); atrial fibrillation (P = 0.000, OR 20.017, 95% CI 4.191-95.609); chronic obstructive pulmonary disease (COPD) (P = 0.009, OR 8.151, 95% CI 1.674-39.676); forced expiratory volume in 1 s / Forced vital capacity (FEV1/FVC) (P = 0.040, OR 0.966, 95% CI 0.935-0.998); and surgical method (P = 0.024, OR 0.425, 95% CI 0.202-0.891). The area under the curve was 0.865, and the consistency index was 0.367. The Hosmer-Lemeshow test indicated excellent model fit (P = 0.367). The calibration curve closely approximated the ideal diagonal line. DCA showed a significant net benefit of the predictive model for postoperative ICU admission. CONCLUSION: Predictors of ICU admission following radical CRC resection include age, preoperative serum albumin level, nutritional risk screening, atrial fibrillation, COPD, FEV1/FVC, and surgical route. The predictive nomogram and online tool support clinical decision-making for postoperative ICU admission in patients undergoing radical CRC surgery. TRIAL REGISTRATION: Despite the retrospective nature of this study, we have proactively registered it with the Chinese Clinical Trial Registry. The registration number is ChiCTR2200062210, and the date of registration is 29/07/2022.

Development of best evidence-based practice protocols for central venous catheter placement and maintenance to reduce CLABSI.

Although evidence-based interventions can reduce the incidence of central line-associated bloodstream infection (CLABSI), there is a large gap between evidence-based interventions and the actual practice of central venous catheter (CVC) care. Evidence-based interventions are needed to reduce the incidence of CLABSI in intensive care units (ICU) in China. Professional association, guidelines, and database websites were searched for data relevant to CLABSI in the adult ICUs from inception to February 2020. Checklists were developed for both CVC placement and maintenance. Based on the Integrated Promoting Action on Research Implementation in Health Services framework, a questionnaire collected the cognition and practice of ICU nursing and medical staff on the CLABSI evidence-based prevention guidelines. From January 2018 to December 2021, ICU CLABSI rates were collected monthly. Ten clinical guidelines were included after the screening and evaluation process and used to develop the best evidence-based protocols for CVC placement and maintenance. The CLABSI rates in 2018, 2019, and 2020 were 2.98‰ (9/3021), 1.83‰ (6/3276), and 1.69‰ (4/2364), respectively. Notably, the CLABSI rate in 2021 was 0.38‰ (1/2607). In other words, the ICU CLABSI rate decreased from 1.69‰ to 0.38‰ after implementation of the new protocols. Additionally, our data suggested that the use of ultrasound-guidance for catheter insertion, chlorhexidine body wash, and the use of a checklist for CVC placement and maintenance were important measures for reducing the CLABSI rate. The evidence-based processes developed for CVC placement and maintenance were effective at reducing the CLABSI rate in the ICU.

Palliative Care for SARS-CoV-2 Patients in the Intensive Care Unit: A Comprehensive Study.

OBJECTIVE: To comprehend the multiprofessional actions regarding palliative care for patients in the Intensive Care Unit affected by SARS-CoV-2. METHODS: A comprehensive qualitative study conducted with 31 professionals from the Intensive Care Units of a university hospital, based on the Theory of Peaceful End of Life. RESULTS: The analysis of the discourse led to the identification of two categories: "Multidisciplinary actions to promote comfort at the end of life" and "Palliative care during the pandemic period and the promotion of emotional and spiritual comfort." FINAL CONSIDERATIONS: It became evident that local administration needs to invest in measures that reduce barriers to the implementation of palliative care during times of crisis. Understanding the discourse highlighted that non-specialized professionals can provide basic palliative care appropriately, without diminishing the importance and necessity of the presence of palliative care specialists in various hospital areas.

The respiratory oxygenation index for identifying the risk of orotracheal intubation in COVID-19 patients receiving high-flow nasal cannula oxygen.

OBJECTIVE: To assess whether the respiratory oxygenation index (ROX index) measured after the start of high-flow nasal cannula oxygen therapy can help identify the need for intubation in patients with acute respiratory failure due to coronavirus disease 2019. METHODS: This retrospective, observational, multicenter study was conducted at the intensive care units of six Brazilian hospitals from March to December 2020. The primary outcome was the need for intubation up to 7 days after starting the high-flow nasal cannula. RESULTS: A total of 444 patients were included in the study, and 261 (58.7%) were subjected to intubation. An analysis of the area under the receiver operating characteristic curve (AUROC) showed that the ability to discriminate between successful and failed high-flow nasal cannula oxygen therapy within 7 days was greater for the ROX index measured at 24 hours (AUROC 0.80; 95%CI 0.76 - 0.84). The median interval between high-flow nasal cannula initiation and intubation was 24 hours (24 - 72), and the most accurate predictor of intubation obtained before 24 hours was the ROX index measured at 12 hours (AUROC 0.75; 95%CI 0.70 - 0.79). Kaplan-Meier curves revealed a greater probability of intubation within 7 days in patients with a ROX index ≤ 5.54 at 12 hours (hazard ratio 3.07; 95%CI 2.24 - 4.20) and ≤ 5.96 at 24 hours (hazard ratio 5.15; 95%CI 3.65 - 7.27). CONCLUSION: The ROX index can aid in the early identification of patients with acute respiratory failure due to COVID-19 who will progress to the failure of high-flow nasal cannula supportive therapy and the need for intubation.

Extracorporeal Carbon Dioxide Removal With the Hemolung in Patients With Acute-on-Chronic Respiratory Failure: A Multicenter Retrospective Cohort Study.

Extracorporeal carbon dioxide removal (ECCO2R) devices are increasingly used in treating acute-on-chronic respiratory failure caused by chronic lung diseases. There are no large studies that investigated safety, efficacy, and the independent association of prognostic variables to survival that could define the role of ECCO2R devices in such patients. This multicenter, multinational, retrospective study investigated the efficacy, safety of a single ECCO2R device (Hemolung) in patients with acute on chronic respiratory failure and identified variables independently associated with intensive care unit (ICU) survival. The primary outcome was improvement in blood gasses with the use of Hemolung. Secondary outcomes included reduction in tidal volume, respiratory rate, minute ventilation, survival to ICU discharge, and complication profile. Multivariable regression analysis was used to identify variables that are independently associated with ICU survival. A total of 62 patients were included. There was a significant improvement in pH and partial pressure of carbon dioxide in arterial blood (PaCO2) along with a reduction in respiratory rate, tidal volume, and minute ventilation with Hemolung therapy. The complication profile did not differ between survivors and nonsurvivors. Multivariable analysis identified the duration of Hemolung therapy to be independently associated with survival to ICU discharge (adjusted odds ratio = 1.21; 95% confidence interval [CI] = 1.040-1.518; p = 0.01).

A privacy-preserving platform oriented medical healthcare and its application in identifying patients with candidemia.

Federated learning (FL) has emerged as a significant method for developing machine learning models across multiple devices without centralized data collection. Candidemia, a critical but rare disease in ICUs, poses challenges in early detection and treatment. The goal of this study is to develop a privacy-preserving federated learning framework for predicting candidemia in ICU patients. This approach aims to enhance the accuracy of antifungal drug prescriptions and patient outcomes. This study involved the creation of four predictive FL models for candidemia using data from ICU patients across three hospitals in China. The models were designed to prioritize patient privacy while aggregating learnings across different sites. A unique ensemble feature selection strategy was implemented, combining the strengths of XGBoost's feature importance and statistical test p values. This strategy aimed to optimize the selection of relevant features for accurate predictions. The federated learning models demonstrated significant improvements over locally trained models, with a 9% increase in the area under the curve (AUC) and a 24% rise in true positive ratio (TPR). Notably, the FL models excelled in the combined TPR + TNR metric, which is critical for feature selection in candidemia prediction. The ensemble feature selection method proved more efficient than previous approaches, achieving comparable performance. The study successfully developed a set of federated learning models that significantly enhance the prediction of candidemia in ICU patients. By leveraging a novel feature selection method and maintaining patient privacy, the models provide a robust framework for improved clinical decision-making in the treatment of candidemia.

Machine learning-based prediction of clinical outcomes after traumatic brain injury: Hidden information of early physiological time series.

AIMS: To assess the predictive value of early-stage physiological time-series (PTS) data and non-interrogative electronic health record (EHR) signals, collected within 24 h of ICU admission, for traumatic brain injury (TBI) patient outcomes. METHODS: Using data from TBI patients in the multi-center eICU database, we focused on in-hospital mortality, neurological status based on the Glasgow Coma Score (mGCS) motor subscore at discharge, and prolonged ICU stay (PLOS). Three machine learning (ML) models were developed, utilizing EHR features, PTS signals collected 24 h after ICU admission, and their combination. External validation was performed using the MIMIC III dataset, and interpretability was enhanced using the Shapley Additive Explanations (SHAP) algorithm. RESULTS: The analysis included 1085 TBI patients. Compared to individual models and existing scoring systems, the combination of EHR and PTS features demonstrated comparable or even superior performance in predicting in-hospital mortality (AUROC = 0.878), neurological outcomes (AUROC = 0.877), and PLOS (AUROC = 0.835). The model's performance was validated in the MIMIC III dataset, and SHAP algorithms identified six key intervention points for EHR features related to prognostic outcomes. Moreover, the EHR results (All AUROC >0.8) were translated into online tools for clinical use. CONCLUSION: Our study highlights the importance of early-stage PTS signals in predicting TBI patient outcomes. The integration of interpretable algorithms and simplified prediction tools can support treatment decision-making, contributing to the development of accurate prediction models and timely clinical intervention.

Development and validation of a prediction model for mortality in critically ill COVID-19 patients.

Background: Early prediction of prognosis may help early treatment measures to reduce mortality in critically ill coronavirus disease (COVID-19) patients. The study aimed to develop a mortality prediction model for critically ill COVID-19 patients. Methods: This retrospective study analyzed the clinical data of critically ill COVID-19 patients in an intensive care unit between April and June 2022. Propensity matching scores were used to reduce the effect of confounding factors. A predictive model was built using logistic regression analysis and visualized using a nomogram. Calibration and receiver operating characteristic (ROC) curves were used to estimate the accuracy and predictive value of the model. Decision curve analysis (DCA) was used to examine the value of the model for clinical interventions. Results: In total, 137 critically ill COVID-19 patients were enrolled; 84 survived, and 53 died. Univariate and multivariate logistic regression analyses revealed that aspartate aminotransferase (AST), creatinine, and myoglobin levels were independent prognostic factors. We constructed logistic regression prediction models using the seven least absolute shrinkage and selection operator regression-selected variables (hematocrit, red blood cell distribution width-standard deviation, procalcitonin, AST, creatinine, potassium, and myoglobin; Model 1) and three independent factor variables (Model 2). The calibration curves suggested that the actual predictions of the two models were similar to the ideal predictions. The ROC curve indicated that both models had good predictive power, and Model 1 had better predictive power than Model 2. The DCA results suggested that the model intervention was beneficial to patients and patients benefited more from Model 1 than from Model 2. Conclusion: The predictive model constructed using characteristic variables screened using LASSO regression can accurately predict the prognosis of critically ill COVID-19 patients. This model can assist clinicians in implementing early interventions. External validation by prospective large-sample studies is required.

Onset of Prolonged High-Potency Benzodiazepine Use Among ICU Survivors: A Nationwide Cohort Study.

OBJECTIVES: Exposure to critical illness and intensive care may lead to long-term psychologic and physical impairments. To what extent ICU survivors become prolonged users of benzodiazepines after exposure to critical care is not fully explored. This study aimed to describe the extent of onset of prolonged high-potency benzodiazepine use among ICU survivors not using these drugs before admission, identify factors associated with this use, and analyze whether such usage is associated with increased mortality. DESIGN: Retrospective cohort study. SETTING: Sweden, including all registered ICU admissions between 2010 and 2017. PATIENTS: ICU patients surviving for at least 3 months, not using high-potency benzodiazepine before admission, were eligible for inclusion. INTERVENTIONS: Admission to intensive care. MEASUREMENTS AND MAIN RESULTS: A total of 237,904 patients were screened and 137,647 were included. Of these 5338 (3.9%) became prolonged users of high-potency benzodiazepines after ICU discharge. A peak in high-potency benzodiazepine prescriptions was observed during the first 3 months, followed by sustained usage throughout the follow-up period of 18 months. Prolonged usage was associated with older age, female sex, and a history of both somatic and psychiatric comorbidities, including substance abuse. Additionally, a longer ICU stay, a high estimated mortality rate, and prior consumption of low-potency benzodiazepines were associated with prolonged use. The risk of death between 6 and 18 months post-ICU admission was significantly higher among high-potency benzodiazepine users, with an adjusted hazard ratio of 1.8 (95% CI, 1.7-2.0; p < 0.001). No differences were noted in causes of death between users and nonusers. CONCLUSIONS: Despite the lack of evidence supporting long-term treatment, prolonged usage of high-potency benzodiazepines 18 months following ICU care was notable and associated with an increased risk of death. Considering the substantial number of ICU admissions, prevention of benzodiazepine misuse may improve long-term outcomes following critical care.