J Crit Care
· 2026 Aug · PMID 41880712
·
Publisher ↗
This letter comments on the study by Turan et al., which evaluates the efficacy of large language models (LLMs) in predicting ICU admission needs. While commending the rigorous benchmark established for off-the-shelf LLM...This letter comments on the study by Turan et al., which evaluates the efficacy of large language models (LLMs) in predicting ICU admission needs. While commending the rigorous benchmark established for off-the-shelf LLMs, we propose methodological directions to shift the evaluation paradigm from validating performance against local decisions toward demonstrating generalizable clinical value. Key suggestions include: aligning validation with patient-centered outcomes and dynamic risk prediction; enhancing transparency and reproducibility for evaluating LLMs on unstructured data; exploring ensemble and privacy-preserving multi-center frameworks to improve robustness and generalizability; and prioritizing research on human-AI collaborative decision-making. These reflections aim to guide the development of reliable, interpretable, and clinically integrated intelligent aids.
Kanji S, Murthy S, Zytaruk N
… +7 more, Heels-Ansdell D, Deane A, Mehta S, Adhikari NKJ, Poole A, Cook D, REVISE Investigators and the Canadian Critical Care Trials Group
J Crit Care
· 2026 Aug · PMID 41880711
·
Publisher ↗
PURPOSE: More environmentally sustainable clinical trials are needed. This case study simulation, based on a secondary analysis of the Re-Evaluating the Inhibition of Stress Erosions (REVISE) trial, was designed to estim...PURPOSE: More environmentally sustainable clinical trials are needed. This case study simulation, based on a secondary analysis of the Re-Evaluating the Inhibition of Stress Erosions (REVISE) trial, was designed to estimate the carbon emission and monetary impact if the proton pump inhibitor had been switched from intravenous to enteral administration as soon as possible. METHODS: In REVISE, only intravenous study drug was used, to avoid additional pharmacy costs associated with preparing study drug and placebo for enteral administration, to streamline trial management, and minimize transition-associated administration errors. The median duration of intravenous pantoprazole (or placebo) administration was 5 days; we assumed patients would transition to enteral lansoprazole on day 2 for the remaining intervention period. Greenhouse gas emissions from the intervention and waste generated were estimated using cost-based emissions factors for drugs and weight-based emission factors for waste. Monetary costs were derived from average acquisition costs from REVISE participating centers. RESULTS: Allowing step-down to enteral lansoprazole on day 2 of an average 5-day course would have reduced greenhouse gas emissions by 28% compared with a 5-day intravenous pantoprazole regimen (9.84 vs. 13.6 kgCO₂e). Considering the entire trial, a complex intravenous/enteral intervention would have saved approximately 9000 kgCO2e and halved drug-related costs including aquisition, preparation and administration supplies. CONCLUSIONS: This simulation demonstrates that enabling transition from intravenous pantoprazole to enteral lansoprazole may have yielded substantial reductions in both greenhouse gas emissions and costs. Considering environmental stewardship represents an important step towards more sustainable clinical trials.
O'Brien SJ, Egan R, Martin-Loeches I
… +1 more, Keane S
J Crit Care
· 2026 Aug · PMID 41875519
·
Publisher ↗
BACKGROUND: Penicillin allergy (PenA) labels are common and often do not represent true immune-mediated allergy, yet are not routinely assessed or tested in the ICU. With greater demand for effective antimicrobial stewar...BACKGROUND: Penicillin allergy (PenA) labels are common and often do not represent true immune-mediated allergy, yet are not routinely assessed or tested in the ICU. With greater demand for effective antimicrobial stewardship, and growing support for non-allergist performed PenA assessment, enthusiasm for routine ICU-led PenA evaluation is increasing. However, we do not presently understand key factors needed to plan and deliver such a change. METHODS: We conducted a convergent mixed-methods study combining an electronic survey of ICU healthcare staff with follow-up interviews. Participants included those working in ICUs with and without PenA assessment programmes. Survey data were analysed descriptively. Interview and free-text data underwent thematic analysis guided by the Theoretical Domains Framework (TDF). Quantitative and qualitative findings were integrated at the interpretation stage. RESULTS: Overall, 113 healthcare staff completed the survey and 13 participated in interviews. Barriers to implementation included limited training, poor guideline awareness, low confidence in risk stratification and drug challenge procedures, perceived medico-legal risk, unclear ownership, and workload pressures. Enablers included clear governance structures, robust protocols, decision support tools, access to specialist advice, local champions, and audit and feedback mechanisms. Participants working in ICUs with established programmes described greater confidence and integration of PenA assessment into routine practice. CONCLUSIONS: Variation in ICU-led PenA assessment appears to reflect organisational and behavioural influences rather than intrinsic clinical barriers. Differences between settings were associated with governance structures, leadership, education, and social reinforcement. Addressing these determinants may support more consistent integration of PenA assessment into routine ICU practice.
J Crit Care
· 2026 Aug · PMID 41875518
·
Publisher ↗
We have commented on the recent study by Li et al., demonstrate a significant association between ketamine administration and reduced 7-day ICU mortality. Although this study is commendable for its large sample size, we...We have commented on the recent study by Li et al., demonstrate a significant association between ketamine administration and reduced 7-day ICU mortality. Although this study is commendable for its large sample size, we would like to offer some constructive comments that could further strengthen the clinical implications of this work. Stratified analysis of subgroup population by race and SOFA or APACHEII. These key points emphasize the need to improve multi angle analysis by incorporating the results of stratified analysis.
Brender TD, Stumpf N, Vossler K
… +12 more, Kim S, Mills H, Lee A, Yang HY, Heintz T, Espejo E, Boscardin J, Pantilat S, Chapman AC, Widera E, Smith AK, Cobert J
J Crit Care
· 2026 Aug · PMID 41875516
·
Publisher ↗
BACKGROUND: ICU patients frequently receive treatments clinicians perceive are futile which can cause conflict between clinicians, patients and families. Medical futility lacks a consensus definition, yet this ambiguous...BACKGROUND: ICU patients frequently receive treatments clinicians perceive are futile which can cause conflict between clinicians, patients and families. Medical futility lacks a consensus definition, yet this ambiguous and contentious term is used in medical notes. RESEARCH QUESTION: What themes are associated with futility mentions in ICU notes? How have themes' frequencies changed over time? STUDY DESIGN/METHODS: Mixed methods study of ICU notes (e.g., H&P, progress notes) written by clinicians (e.g., physicians, nurses) for adult patients at a large hospital system from 2010 to 2020. Neural network models identified terms most associated with "futile" or "futility." Distributional semantic analysis grouped terms into themes. Regression modeling explored longitudinal changes in themes' frequencies. RESULTS: Across 2,460,169 notes for 9912 patients, the annual average count of unique notes with futility mentions was 137 per 100,000 and unchanged from 2010 to 2020. 8 themes were identified among terms most associated with the words "futile" or "futility." The most represented themes were Decision Making (annual average 18% [95% CI: 16%-19%]), Assessing, Prognosticating, and End-of-Life Outcomes (15%, 13%-16%), and Identifying Sentiments (13%, 10%-15%). Recording Code Status was the least represented theme in 2010 (4%) and increased over time (9% in 2020, P = 0.001). INTERPRETATION: Use of futility was rare and stable across a decade of ICU notes. Semantic analysis indicates clinicians use futility in heterogeneous contexts. Changes in themes' frequencies may reflect clinicians' evolving conceptions of medical futility. These findings could guide development of EHR-based interventions to address perceived futile treatments which contribute to clinicians' moral distress.
Fennessy A, Slattery L, Shelley O
… +2 more, Reyes LF, Martin-Loeches I
J Crit Care
· 2026 Aug · PMID 41861747
·
Publisher ↗
BACKGROUND: Severe burn injury is associated with profound physiological derangement and remains a major cause of infection-related morbidity and mortality worldwide. Disruption of the skin barrier sustained immune dysre...BACKGROUND: Severe burn injury is associated with profound physiological derangement and remains a major cause of infection-related morbidity and mortality worldwide. Disruption of the skin barrier sustained immune dysregulation, prolonged intensive care unit (ICU) exposure, and extensive use of invasive devices create a uniquely infection-prone host environment. Infectious complications, particularly those caused by multidrug-resistant organisms (MDROs), continue to account for a substantial proportion of deaths in critically ill burn patients despite advances in surgical and critical care management. OBJECTIVES: This narrative review aims to provide a comprehensive, clinically focused overview of infectious complications in critically ill burn patients, integrating current evidence on epidemiology, pathophysiology, microbial dynamics, diagnostic strategies, and contemporary management approaches relevant to daily ICU practice. SOURCES OF EVIDENCE: A narrative synthesis of the published literature was performed, including international guidelines, observational studies, randomised trials, systematic reviews, and translational research focusing on burn-related infections, antimicrobial resistance, diagnostics, and emerging therapies. CONTENT: The review examines the multifactorial pathophysiology underlying infection susceptibility following major burns, including loss of the cutaneous barrier, hyperinflammatory responses followed by immune paralysis, and burn-induced hypermetabolism. Dynamic patterns of microbial colonisation, biofilm formation, microbiome disruption, and the global rise of MDROs are explored. Diagnostic challenges in distinguishing colonisation from invasive infection are discussed, alongside traditional and advanced diagnostic modalities such as quantitative tissue cultures, biomarkers, multiplex molecular assays, and next-generation sequencing. Contemporary management strategies are reviewed, emphasising early surgical source control, pharmacokinetically optimised antimicrobial therapy, antimicrobial stewardship, and rigorous infection prevention and control practices. Emerging adjunctive therapies, including bacteriophage therapy, nanotechnology-based antimicrobials, microbiome-directed interventions, and immunomodulatory approaches, are also highlighted. IMPLICATIONS: Effective infection management in burn patients requires an integrated, multidisciplinary approach that combines rapid diagnosis, early surgical intervention, tailored antimicrobial therapy, and robust infection prevention strategies. Advances in molecular diagnostics, precision medicine, and microbiome science hold promise for improving outcomes and mitigating the growing burden of antimicrobial resistance in burn ICUs. CONCLUSIONS: Infectious complications remain a leading determinant of outcome following severe burn injury. Optimising infection care through early recognition, precise diagnostics, coordinated surgical and antimicrobial strategies, and emerging precision-based interventions is essential to reduce infection-related morbidity and mortality in this vulnerable patient population.
Matin N, Beqiri E, Wahlster S
… +2 more, Newcombe V, Robba C
J Crit Care
· 2026 Aug · PMID 41855823
·
Publisher ↗
Despite extensive research, traumatic brain injury (TBI) imposes significant mortality and morbidity burden globally. This narrative review provides a comprehensive overview of how TBI classification, surgical and critic...Despite extensive research, traumatic brain injury (TBI) imposes significant mortality and morbidity burden globally. This narrative review provides a comprehensive overview of how TBI classification, surgical and critical care management, and physiological treatment goals have evolved based on recent evidence, shaping future directions for TBI care. We highlight the conceptual shift in TBI classification from the Glasgow Coma Scale to a new framework that integrates clinical and imaging features, biomarkers, and modifiers to improve disease characterization and describe disease trajectories. We review landmark trials that refine the role and timing of surgical interventions in TBI, and emphasize the shift from fixed intracranial pressure goals to physiology-driven approaches with dynamic individualized autoregulation metrics and cerebral oxygen monitoring. We also highlight recent changes in critical care management, including the shift from restrictive transfusion strategies to more liberal transfusion goals, and deliberate monitoring and adjustment of ventilatory settings, hemodynamic goals, and volume status. An enhanced understanding of the prolonged recovery journey and recognition of cognitive motor dissociation may further inform our neuroprognostication and cautions against early withdrawal of life sustaining treatment. Artificial intelligence offers the integration of multimodal data from large databases and enables dynamic patient-specific management. Innovative trial design, incorporating multifaceted, individualized strategies targeting multiple pathophysiological mechanisms, may better account for the complexity and heterogeneity of TBI.
Zerem Y, Finkelstein A, Weiniger CF
… +3 more, Schvartz R, Sela HY, Einav S
J Crit Care
· 2026 Aug · PMID 41855822
·
Publisher ↗
INTRODUCTION: The rate of intensive care unit (ICU) admission of obstetric patients in high-income countries is approximately 0.4%. ICU admission can negatively affect quality of life (QoL). AIM: To assess survival statu...INTRODUCTION: The rate of intensive care unit (ICU) admission of obstetric patients in high-income countries is approximately 0.4%. ICU admission can negatively affect quality of life (QoL). AIM: To assess survival status of obstetric patients admitted to ICU. Secondary outcomes included domiciliary status and QoL versus matched controls without ICU admission. METHODS: Bi-center, cross-sectional questionnaire study of patients admitted to an ICU and controls of non-ICU patients, matched for date of admission, demographics, obstetric history and neonatal outcome. Survival was determined via the Central National Population Registry. QoL and postpartum depression (PPD) were assessed at hospital discharge and at time of interview. Post-Traumatic Stress Disorder (PTSD) was assessed up to the time of interview. RESULTS: Among 216 ICU admission, seven died (3.2%). Overall, 128 patients were interviewed (64 ICU patients vs. 64 non-ICU patients). All participants were living at home. ICU survivors had lower QoL at the time of hospital discharge (42.6 ± 22.4 vs. 60.2 ± 18.9, p < 0.001) compared to non-ICU controls, but by the time of interview QoL was similar between cohorts. ICU survivors also had higher rates of PPD at hospital discharge (51.6% vs. 17.2%, p < 0.001) and at interview (12.5% vs. 0%, p = 0.003) and higher rates of PTSD versus controls (10.9% vs. 1.6%, p = 0.030). CONCLUSION: In this study, one in thirty ICU obstetric patients died. Critical illness with ICU admission carried short- and long-term implications in terms of QoL and psychological morbidity. Most patients reported improvement over time, but this process spanned months to years.
Nakano H, Horie R, Gondo T
… +4 more, Inoue Y, Inokuchi R, Doi K, CHANGE Study Group
J Crit Care
· 2026 Aug · PMID 41844479
·
Publisher ↗
PURPOSE: Investigate the association between the phenotypic classification of laboratory data and mortality in adult patients with acute kidney injury (AKI) undergoing continuous renal replacement therapy (CRRT). METHODS...PURPOSE: Investigate the association between the phenotypic classification of laboratory data and mortality in adult patients with acute kidney injury (AKI) undergoing continuous renal replacement therapy (CRRT). METHODS: This post-hoc analysis of a retrospective multicenter cohort study examined adult AKI patients treated with CRRT at 18 university hospitals in Japan. The primary outcome was mortality within 7 days of CRRT termination. The phenotypes were classified by analyzing the laboratory data until day 3 following CRRT initiation using a group-based multitrajectory (GBMT) analysis. External validation was performed using the Medical Information Mart for Intensive Care (MIMIC)-IV database. RESULTS: Altogether, 550 patients were categorized into the following three groups based on their lactate (Lac) levels and serum urea nitrogen-to-creatinine ratio (UCR): group 1, high Lac levels; group 2, high UCR; and group 3, normal UCR and Lac levels. Groups 1 and 2 exhibited higher mortality rates than group 3 (33.3% vs. 29.3% vs. 13.5%, p < 0.01), although group 3 demonstrated the highest serum creatinine level (2.3 vs. 2.1 vs. 3.0 mg/dL, p < 0.001). External validation using the MIMIC-IV database showed similar results. CONCLUSIONS: UCR and Lac were distinctive factors in the GBMT analysis. The groups with high UCR or high Lac levels had significantly higher mortality rates than the groups with normal UCR and Lac levels. UCR may reflect increased protein metabolism and be a useful risk factor for mortality even in patients with severe AKI requiring CRRT.
Stegink D, Roos-Blom MJ, Dongelmans DA
… +14 more, Klopotowska JE, Sneekes P, van den Berg R, Kusadasi N, van Essen EHR, Arbous MS, den Uil CA, Festen-Spanjer B, Dormans T, van den Berg LAE, van Tellingen M, Deetman NPE, Schouten JA, de Keizer NF
J Crit Care
· 2026 Aug · PMID 41833176
·
Publisher ↗
PURPOSE: Antimicrobial resistance is a major concern in intensive care units (ICUs), where infections and antibiotic use are common. Antimicrobial stewardship programs (ASPs) promote optimal infectious disease management...PURPOSE: Antimicrobial resistance is a major concern in intensive care units (ICUs), where infections and antibiotic use are common. Antimicrobial stewardship programs (ASPs) promote optimal infectious disease management through interventions such as Therapeutic Drug Monitoring (TDM) and selective digestive/oropharyngeal decontamination (SDD/SOD), but data on ICU-specific quality indicators (QIs) and between-centre variation are limited. This study aimed to evaluate infectious disease management in Dutch ICUs by assessing adherence to, and variation in, TDM and SDD/SOD practices. METHODS: In this retrospective multi-centre study, infectious disease management was evaluated across eleven ICUs using routine data from 2019 to 2023. One process QI for TDM and two for SDD/SOD administration and surveillance were assessed. QI performance was calculated per ICU, and adjusted variation was quantified using the median odds ratio (MOR) from multilevel logistic regression models. RESULTS: Among 64,858 ICU admissions, 78.8% received antibiotics and 35.7% received SDD/SOD. Median performance for timely TDM was 73.6%, with 75.6% for vancomycin, 36.8% for voriconazole, and 23.5% for aminoglycosides. SDD/SOD was administered in 73.9% of admissions, and surveillance cultures were collected in 92.4% of cases when SDD/SOD was administered. Substantial variation was observed between ICUs (MOR = 1.73, 2.42, and 1.23 for TDM, SDD/SOD administration, and SDD/SOD surveillance), which persisted after adjustment for patient characteristics. CONCLUSIONS: Substantial variation was observed between ICUs, indicating room for quality improvement. This could not be explained by case-mix, which suggests that organizational factors are possible drivers of this variation.
Fresilli S, Labanca R, Pontillo D
… +8 more, Belletti A, Mariotti C, Kargbo I, Rusconi F, Bruni A, Zangrillo A, Landoni G, 4S Group Collaborators
J Crit Care
· 2026 Aug · PMID 41831334
·
Publisher ↗
PURPOSE: This study aimed to evaluate sex-specific differences in 28-day mortality among young adults (18-55 years) with septic shock using a large propensity score-matched cohort. MATERIALS AND METHODS: We performed a r...PURPOSE: This study aimed to evaluate sex-specific differences in 28-day mortality among young adults (18-55 years) with septic shock using a large propensity score-matched cohort. MATERIALS AND METHODS: We performed a retrospective propensity score-matched analysis using the TriNetX network. Adults 18-55 years old developing septic shock between January 2011 and January 2020 were identified. Women were matched 1:1 to men. The primary outcome was 28-day all-cause mortality; 90- and 180-day mortality were prespecified secondary outcomes. RESULTS: A total of 100,894 young adults (≥18 and ≤ 55 years) with septic shock were analyzed. After 1:1 propensity score matching, 86,748 patients (43,374 males and 43,374 females) were included in the analyses. The 28-day mortality rate was lower in young women compared to men (6630/43,374 [15.3%] vs. 7192/43,374 [16.6%]; RR = 0.92, 95% CI 0.89-0.95, p < 0.001). Similar results were observed for 90-day and 180-day mortality. CONCLUSIONS: Mortality in young women with septic shock was lower than in young men suggesting sex-related differences and warranting further studies to explore therapeutic implications.