Critical Care Medicine

Rationale. Racial disparities in outcomes among patients with acute respiratory failure are well-described, but the contributions of clinicians to these disparities have not been evaluated. Objectives. Among mechanically ventilated patients, we evaluated racial disparities in severity of illness trajectories and adapted value-added modeling to quantify nurse and physician relationships with these disparities. Methods. In a retrospective cohort of mechanically ventilated patients across five hospitals between 2018 and 2022, we used generalized estimating equations to model the change in Laboratory-based Acute Physiology Score version 2 (LAPS) from the start to end of intensive care unit admission ({Delta}LAPS). Consistent with value-added modeling, we randomly allocated the cohort into development and testing partitions, and fit separate multiple linear regression models of {Delta}LAPS using concurrent nurse and physician assignments (determined at 4-hour intervals), patient race, and clinician-race interaction terms as fixed effects. Clinician-specific and clinician-race interaction coefficients were extracted to determine race-specific value-add for each clinician. We defined the race-contextual value-add difference (RCVAD) as a clinician-level measurement of the difference in that clinician's value-add between Black and White patients in their care; a positive RCVAD indicates a more favorable severity of illness trajectory for Black relative to White patients and vice versa. Measurement and Main Results. Among 6,555 distinct patients, 7,247 clinical encounters, 405 nurses, and 70 physicians, Black patients accounted for 2,926 (40%) encounters. Overall, Black patients had significantly less improvement in {Delta}LAPS than White patients (difference in LAPS decline = 2.26 [0.23, 4.29], p=0.029). In the development partition, median nurse RCVAD was -0.10 (interquartile range [IQR]: -1.17, 1.14) with 191 (47%) nurses having a positive RCVAD; median physician RCVAD was -0.18 (IQR: -1.34, 0.56) with 29 (41%) having a positive RCVAD. Conclusions. Black mechanically ventilated patients experience less improvement in severity of illness during intensive care unit admission than White patients. While the majority of physicians and nurses were associated with disparities-exacerbating illness trajectories, many other clinicians were associated with disparities-mitigating trajectories. Future work to understand practices associated with disparities-exacerbating and disparities-mitigating care profiles could inform interventions to reduce disparities overall.

Background Hypercapnia may indicate a primary ventilatory syndrome, a complication of another illness, or an epiphenomenon of severe disease. The presenting context of hypercapnia is poorly quantified, limiting clinical interpretation and synthesis of epidemiologic studies. Methods We performed a retrospective cross-sectional study of Medical Information Mart for Intensive Care IV (MIMIC-IV) hospital admissions linked to an emergency department (ED) presentation from 2011 through 2019. Admissions were included if the triage chief complaint was not missing and at least one prespecified criterion for hypercapnia was met: an International Classification of Diseases (ICD) code for hypercapnic respiratory failure or obesity hypoventilation syndrome, arterial blood gas (ABG) PCO2 45 mmHg, venous blood gas (VBG) PCO2 50 mmHg, or indeterminate-source blood gas PCO2 50 mmHg. Triage chief-complaint text was classified by natural language processing (NLP) into 17 National Hospital Ambulatory Medical Care Survey reason-for-visit (RFV) categories using a multi-label framework. Primary analyses estimated admission-level RFV category prevalences; secondary analyses compared distributions by overlapping ascertainment indicator, age, and acidemia. Results The total cohort included 11,941 admissions: 1,542 (12.9%) met both blood-gas and ICD-code criteria, 9,958 (83.4%) met blood-gas criteria only, and 441 (3.7%) met ICD-code criteria only. Median age at admission was 68 years (IQR 56-78), and 6,423 admissions (53.8%) were for male patients. Respiratory RFV categories were most prevalent (30.2%), followed by administrative reasons (17.5%), digestive symptoms (14.0%), injuries and adverse effects (14.0%), and nervous-system symptoms (13.8%); categories were not mutually exclusive. Respiratory categories were more common in ICD-positive admissions (50.2%) than in VBG-defined (36.3%) or ABG-defined admissions (27.3%). Injuries and adverse effects were most prevalent among admissions for patients aged 18-39 years (34.4%), whereas respiratory categories increased from 13.7% among admissions for patients aged 18-39 years to 36.5% among admissions for patients aged 80 years. NLP-derived classifications showed mean set-F1 of 0.84 against adjudicated clinician labels in the full annotated benchmark sample. Conclusions Among ED-linked admissions with hypercapnia by diagnosis code, blood gas, or both, respiratory complaints were the most common chief-complaint category but represented fewer than one-third of admissions. Presentation context should be incorporated when defining, comparing, and interpreting hypercapnia cohorts, particularly those ascertained by blood-gas criteria.

Septic shock caused by fungal organisms is characterized by high mortality and diagnostic complexity. We used the Premier Healthcare Database to characterize antifungal use and fungal diagnoses among adults with septic shock requiring vasopressors admitted between October 2022 through July 2024. Among 12.8 million admission at 886 hospitals, 554,948 met septic shock criteria and were included for analysis. A fungal diagnosis was established in 11,405 (2.1%) of encounters; of these, 3,565 (31.3%) received intravenous antifungal therapy within one day of vasopressor initiation. In the overall cohort, antifungal therapy was initiated in 29,824 (5.5%) within one day of vasopressor initiation; of these, 3,656 (12.2%) were ultimately diagnosed with a fungal infection. In the 116 hospitals reporting microbiological data, a subgroup of 489 encounters with septic shock and culture-confirmed candidemia was identified. In this subgroup, intravenous antifungal therapy was initiated in 43.8% within one day, 63.8% within three days, and 78.9% within seven days. These findings highlight a profound decoupling between fungal diagnosis and treatment--few patients receiving antifungals were diagnosed with an infection that would be treated by these agents, while less than half of patients with septic shock and candidemia received timely treatment. Strategies for greater precision in empiric antifungal use in septic shock are needed to improve safety, stewardship, and outcomes.

Background Ventilator-associated pneumonia (VAP) is the most frequent nosocomial infection in critical care, affecting 20-36% of mechanically ventilated patients. Early prediction is hampered by the absence of a reliable, objective diagnostic standard. We developed ADVISE (Automated Dudley Ventilation Infection Series Evaluation), a machine learning model to predict physiological deterioration consistent with developing VAP using routinely collected electronic health record data from a UK NHS intensive care unit. Methods Retrospective observational study of admissions at Russell's Hall Hospital ICU (2008-2026). Following National Data Opt-Out exclusion (158 admissions, 4.2%), 3,566 admissions generated 33,208 candidate 48-hour observation blocks. Six temporal variables - FiO2, ventilator mode, P:F ratio, procalcitonin (PCT), secretion amount, and secretion description - were extracted across the baseline window (hours 1-24). A composite VAP-surrogate outcome required concurrent P:F ratio decline (>=5%) and PCT rise (>=0.5 ng/mL) across the outcome window (hours 25-48). After sequential quality filters, 2,134 blocks (18 positive, 0.84% prevalence) were retained. An XGBoost classifier was trained using nested 5-fold cross-validation with scale_pos_weight=114.0 and ROC-based hyperparameter optimisation on 1,495 training blocks, evaluated on 639 held-out test blocks. Performance was assessed via AUROC, AUPRC, and calibration (Brier score). Bootstrap resampling (1,000 iterations) generated 95% confidence intervals. Results On the held-out test set (n=639, 5 positive outcomes), ADVISE achieved AUROC 0.874 [95% CI: 0.771-0.939] and AUPRC 0.031 [0.008-0.069], representing a 4.0-fold improvement over the no-skill baseline. Nested cross-validation mean AUROC was 0.844 +/- 0.078 (range 0.716-0.915). At the Youden-optimal threshold, sensitivity was 0% with specificity 97.8%, reflecting extreme class imbalance (0.78% test prevalence). A threshold targeting 80% sensitivity achieved sensitivity 80.0% [33.3-100.0%], specificity 87.4% [84.8-89.9%], positive predictive value 4.8% [1.1-9.9%], and negative predictive value 99.8% [99.4-100.0%], detecting 4 of 5 VAP cases with approximately 80 false alarms (12.6% false positive rate). Brier score was 0.0078. Feature importance identified baseline P:F ratio as the dominant predictor (41.3% total gain), followed by ventilator mode (26.1%), secretion amount (13.2%), secretion description (9.1%), procalcitonin (5.9%), and FiO2; (4.5%). Conclusions ADVISE demonstrates that baseline oxygenation trajectory and ventilatory support patterns - derived exclusively from routinely charted ICCA variables - can identify admissions at risk of VAP-related physiological deterioration with meaningful discrimination (AUROC 0.874) despite severe class imbalance. The 80% sensitivity operating point offers a clinically actionable alert rate (12.6% FPR), supporting integration into existing ICU workflows. This proof-of-concept study establishes feasibility; multi-site prospective validation is required before clinical deployment.

Background: The Glasgow Coma Scale (GCS) is a universal neurologic severity score in the intensive care unit and is incorporated into APACHE, SOFA, mortality prediction models, ICU benchmarking, and quality metrics. In mechanically ventilated patients, however, the verbal component cannot be assessed. Common conventions, including assigning a normal total GCS of 15 or excluding patients with missing verbal scores, may misclassify the sickest patients as neurologically normal or remove them from analysis. Objective: To quantify non-assessable verbal GCS examinations after acute brain injury and determine how different handling conventions affect severity scoring and mortality-model performance across two independent critical care databases. Materials and Methods: We conducted a retrospective cohort study of adults with acute brain injury during their first ICU stay in MIMIC-IV, with replication in eICU-CRD. A verbal examination was considered non-assessable when documented as No Response-ETT. We measured the burden and determinants of non-assessability, compared the MIMIC-IV derived GCS convention with a component-aware GCS, and evaluated mortality-model handling strategies. Results: Among 14,230 patients, 45.2% had a non-assessable verbal examination, and 47.5% of ventilated patients had no assessable verbal score in the first 24 hours. Non-assessability was strongly associated with mechanical ventilation and mortality. The MIMIC-IV derived GCS assigned a score of 15 to 42.9% of patients and placed 11.6% in the lowest severity category despite eye and motor findings consistent with GCS [≤]9. Complete-case handling excluded 28.5% of patients, who accounted for 50.2% of deaths. Similar distortions were observed in eICU-CRD/APACHE across 171 hospitals. Discussion: Default-to-normal scoring can make severely ill intubated patients appear neurologically normal, while complete-case analysis removes the highest-risk patients. Conclusion: Non-assessable verbal GCS in mechanically ventilated patients should be explicitly flagged and reported in ICU severity scores, risk-adjusted mortality models, and benchmarking systems.

Background High-risk preoperative triage remains fragmented: existing tools often estimate risk without identifying modifiable mechanisms or linking classification to postoperative monitoring, destination planning, and rescue resources. This protocol describes implementation and evaluation of a Reserve-Stress-Rescue (RSR Framework), pathway that operationalizes perioperative high risk as a mismatch among patient physiologic reserve, procedural stress, and system rescue capacity. Approach RSR is a proposed clinician-facing, modular scoring framework for adults undergoing major surgery, especially patients with frailty, multimorbidity, poor functional capacity, anemia or malnutrition, cardiopulmonary disease, or limited postoperative support. Each domain, Reserve, Stress, and Rescue, is scored from 0 to 4 and recorded as both a three-part profile and a total score from 0 to 12. Scores map to Green, Amber, Red, and Crimson triage bands that trigger escalating actions, including targeted optimization, multidisciplinary review, anesthesia and surgical planning, postoperative destination selection, monitoring intensity, and predefined escalation criteria. Validation Plan The initial phase of this study received an exemption determination from the Yale University Institutional Review Board on June 3, 2026, under IRB Protocol ID 2000042729, with exempt categories 2(ii) and 4(iii), including a waiver of HIPAA authorization for access to and use of protected health information as described in the approved protocol. Evaluation will proceed in stages, assessing feasibility, interrater reliability, completeness, acceptability, discrimination, calibration, and clinical utility. Key outcomes include postoperative complications, unplanned escalation of care, intensive care utilization, failure to rescue, mortality, length of stay, triage burden, low-yield testing cascades, and management-changing pathway activation. Conclusion The RSR pathway reframes high-risk status as a modifiable interaction between vulnerability, operative insult, and rescue capacity rather than a fixed patient label. If feasible and valid, RSR may standardize high-risk identification, align perioperative resources with anticipated physiology, improve communication, and support safer, actionable shared decision-making.

Background: The joint association of the C-reactive protein-triglyceride glucose index-frailty index (CTI-FI) and non-exercise estimated cardiorespiratory fitness (eCRF) with all-cause mortality (ACM) in adults aged [&ge;]45 years with cardiovascular-kidney-metabolic (CKM) syndrome stages 0-3 remains unexplored. Methods: Participants were enrolled from the National Health and Nutrition Examination Survey (NHANES; derivation cohort) and the China Health and Retirement Longitudinal Study (CHARLS; external validation). Covariate selection was performed using LASSO regression. Weighted Cox models were applied across four adjustment models to evaluate the independent associations of CTI-FI and eCRF with ACM. Dose-response patterns were examined with restricted cubic splines (RCS). Subgroup, sex-stratified, and mediation analyses tested robustness and pathways. Results: A total of 6,662 participants from NHANES (median follow-up 10 years; 1,276 ACM, 19.2%) and 3,418 participants from CHARLS (9 years; 391 deaths, 11.4%) were included. Per 1-unit increase in CTI-FI, the risks increased by 44% for ACM (HR 1.44; 95% CI 1.31-1.57) and by 54% for cardiovascular mortality (CVM, HR 1.54; 95% CI 1.33-1.79); per 1-MET increase in eCRF, the risks decreased by 10% (HR 0.90; 95% CI 0.85-0.94) and by 18% (HR 0.82; 95% CI 0.75-0.90), respectively (all P < 0.001). Compared with the low CTI-FI + high eCRF group, the high CTI-FI + low eCRF group was associated with a significantly higher risk of ACM (HR 2.74; 95% CI 2.20-3.40) and CVM (HR 5.04; 95% CI 3.02-8.40). RCS analysis showed a nonlinear CTI-FI-ACM association. The model with CTI-FI and eCRF achieved a C-index of 0.78 for ACM and 0.825 for CVM. CTI-FI and eCRF bidirectionally mediated each other's associations with ACM and CVM. Specifically, eCRF accounted for 16.4%-23.5% of CTI-FI-related mortality risk, whereas CTI-FI accounted for 23.9%-32.2% of eCRF's survival benefit (all P < 0.001). Conclusions: Higher CTI-FI and lower eCRF independently and jointly predict increased mortality, with bidirectional mediation indicating that improving one may partially offset the adverse effect of the other. These findings highlight potential therapeutic targets for early CKM syndrome management.

Background: Manual review of 30-day hospital readmissions can identify actionable quality and safety problems, but it is labor-intensive. We developed and evaluated an agentic AI workflow for evidence-grounded readmission review. Materials and methods: We studied adult patients with unplanned 30-day readmission after discharge from a medicine hospitalist service at a single academic health system. An AI agent using a large language model queried a database containing notes, encounters, procedures, laboratory results, and other clinical data, and completed the same structured readmission-review rubric used by physicians. In the primary comparative evaluation, 20 randomly selected readmissions from 2025 were each reviewed by two physicians and the AI system. Blinded physician evaluators rated review quality. After rubric refinement, the AI workflow was applied to 100 recent readmissions in an exploratory expanded-cohort analysis of recurring improvement opportunities. Results: In the primary comparative evaluation, the AI classified 9/20 readmissions (45%) as preventable, compared with 19/40 physician reviews (47.5%). Blinded overall quality ratings were similar for AI and physician reviews (4.35 vs. 4.20 on a 1-5 scale; mean difference 0.15, 95% CI -0.20 to 0.48; p=0.49), as were factuality/support and usefulness/actionability ratings. No AI hallucinations were identified during factuality review. Agreement on preventability and primary readmission category was low for both AI-human and human-human comparisons. The AI system cost $0.23 per chart; physician reviewers took a median of 15 minutes, corresponding to an estimated $42.43 per chart. In the exploratory expanded-cohort analysis, AI-assisted review identified recurring vulnerabilities in post-discharge follow-up plans, incomplete inpatient workups, medication-safety transitions, and indwelling-device transitions. Conclusions: Agentic AI produced readmission reviews with similar blinded quality ratings to physician reviews in this small single-center primary comparative evaluation and supported identification of recurring quality-improvement themes in the exploratory expanded-cohort analysis. Preventability judgments remained variable among both AI and physicians, underscoring the need for human oversight and prospective evaluation before operational use.

Importance: The transition from medical school to residency forms a national training network, yet its large-scale structure and implications for trainee outcomes remain poorly characterized. Objective: To evaluate the US residency match as a network and assess how institutional position relates to residency placement, educational debt, and specialty choice. Design: Cross-sectional analysis of publicly reported 2025 residency match outcomes. Setting: 107 US MD-granting medical schools and 301 residency institutions with available match data. Participants: 14,616 US MD students matching into residency in 2025 (convenience sample). Exposure: Institutional position within the residency match network, quantified using PageRank network centrality. The relative strength of each school's graduating class was defined as the median centrality of residency destinations across graduates (placement score). Main Outcomes and Measures: Residency placement outcomes, mean medical school debt at graduation, and specialty choice (primary care vs surgical specialties) in relation to institutional position within the residency match network. Network-derived measures were also compared with NIH funding, residency reputation, and student selectivity. Results: Among 14,616 US MD students matched across 107 medical schools and 301 residency institutions (approximately 73.5% of total US MD cohort), network-derived measures of institutional influence closely aligned with benchmarks of institutional standing such as NIH funding, residency reputation, and student selectivity (Spearman's Rho; = 0.72-0.86; all p < .001). Graduate outcomes varied systematically across institutions. Graduates of highly connected medical schools were more likely to match into highly connected residency programs (87.3% for top-quintile vs 41.0% for bottom-quintile schools). Schools with higher placement scores had graduates with lower educational debt, reduced entry into primary care, and increased entry into surgical or competitive specialties. Compared with bottom-decile schools, top-decile schools (stratified by placement score) had 37% lower mean graduate debt, 24% lower primary care entry, and 75% higher surgical specialty entry. Higher educational debt was not associated with entry into higher-compensated specialties. Conclusions and Relevance: The residency match network reflects a hierarchical structure of institutional standing. Graduates of higher- and lower-positioned medical schools experience systematically different residency placement outcomes. These findings provide a population-level, behavior-based perspective on institutional influence and its relationship to training pathways.

Introduction: Generative artificial intelligence (AI) can produce realistic clinical scenarios on demand and deliver immediate, individualized feedback, yet its use to teach ethical reasoning, rather than to address the ethics of AI itself, remains underexplored in interprofessional healthcare education. Aim: This pilot study examined how interprofessional healthcare students perceived an AI-enhanced, case-based platform designed to support ethical decision-making across physical therapy, occupational therapy, speech-language pathology, and audiology. Methods: Students enrolled in an interprofessional education course completed an online module of 20 instructor-vetted, AI-generated ethics cases and an optional post-activity survey of Likert-scale and open-ended items. Quantitative data were analyzed descriptively and qualitative responses were analyzed through content analysis. Results: Ten students responded. Within this small sample, perceptions of platform utility and usability were strongly positive, with all respondents agreeing that immediate feedback and scenario variety supported learning. Perceptions were more divided when the platform was compared directly with traditional classroom learning, and respondents identified pacing and auto-scrolling as usability concerns. Conclusions: These preliminary findings suggest AI-enhanced case-based platforms can engage students and support applied ethics learning but are best positioned to complement rather than replace traditional instruction. Findings are exploratory given the small, demographically limited sample.

Background: In Canada, there is a lack of data relating sociodemographic characteristics to the likelihood of consent and clinical trial participation. Objective: The overall objective of this study is to examine the association of hospital-level sociodemographic variables with a priori informed consent rates for participation in the REVISE trial. Design: This study is a retrospective observational analysis of Canadian sites participating in the international REVISE trial. Methods: Sociodemographic characteristics for 42 hospitals participating in the REVISE trial will be supplemented by national data from the 2021 Canadian Census of Population Profile at the census tract level corresponding to the hospital's location. Hospital level information for Ontario sites will be derived from the Institute for Clinical Evaluate Sciences (ICES) database. Site clustering will be performed using latent class analysis, a flexible clustering technique that identifies meaningful subgroups based on sociodemographic variables purposively selected from data available through the Statistics Canada 2021 census profile, ICES, and hospital-reported data. Clustering analysis will be performed for all Ontario hospitals with available ICES data, followed by a separate analysis for all Canadian REVISE sites using Statistics Canada data. Concordance in the clustering of REVISE sites will be examined by comparing the assignment of hospitals to the latent classes separately identified using ICES and Statistics Canada data. If there is a high degree of agreement between the two datasets, sociodemographic predictors will be analyzed using the clusters identified through ICES for Ontario sites with the concordant classes based on Statistics Canada data for Canadian sites outsite Ontario. If there is disagreement in cluster assignment between the two datasets, separate analyses of sociodemographic factors will be conducted for Ontario sites using ICES data and for all Canadian sites using the 2021 Census Profile. Multivariate linear regression models will be used to analyze the association between hospital-level characteristics and the likelihood of a priori and deferred consent. Results: Results of this study will generate information about the relationship between informed consent to participate in a low-risk critical care clinical trial using different consent models, and socioeconomic patient characteristics at the hospital site level (e.g., educational attainment, knowledge of official languages, citizenship rates, family income, poverty, rurality and immigration patterns). Conclusions: This study will fill an evidence gap by generating information on the relationship between sociodemographic variables and the likelihood of informed consent to participate in a critical care clinical trial in Canada.

Continuous glucose monitoring (CGM) is increasingly used for inpatient glycemic surveillance, but evidence in non-critical care wards remains limited, particularly in real-world public healthcare settings. Intermittent capillary glucose testing may fail to detect transient, nocturnal, or asymptomatic dysglycemia. We sought to evaluate whether CGM improves detection of clinically significant dysglycemia compared with seven-point capillary glucose monitoring in hospitalized patients with type 2 diabetes mellitus or hyperglycemia. This is a prospective, observational, non-randomized, real-world study performed in a tertiary referral center in Mexico. 56 hospitalized patients were included: 28 underwent flash CGM and 28 underwent seven-point capillary glucose monitoring. Patients were followed for up to 6 hospitalization days. The main analytical focus was detection of clinically significant dysglycemia, including hypoglycemia <70 mg/dL, clinically significant hypoglycemia <54 mg/dL, and severe hyperglycemia >250 mg/dL. Secondary outcomes included time in range, mean daily glucose, insulin requirements, infectious complications, length of stay, and mortality. CGM detected more hypoglycemia <70 mg/dL than capillary monitoring (71.4% vs 35.7%, p=0.005), more clinically significant hypoglycemia <54 mg/dL (median 3 [IQR 0-6.5] vs 0, p=0.030), and more severe hyperglycemia >250 mg/dL (median 8.5 [IQR 0.5-17] vs 0 [IQR 0-9.52], p=0.030). Time in range was not significantly different between groups (59.86 +/- 23.46% vs 69.28 +/- 24.99%, p=0.151). After adjustment for age, diabetes duration, and admission hyperglycemia, CGM remained associated with hypoglycemia detection (OR 4.7, 95% CI 1.2-19.0, p=0.027). We concluded that CGM improved detection of clinically significant dysglycemia during up to 6 hospitalization days. Although CGM did not improve time in range or short-term clinical outcomes, it provided superior glycemic surveillance compared with intermittent capillary glucose testing.

Objective: To quantify the burden, structure, and downstream analytic consequences of "Unable to Assess" (UTA) delirium documentation in the intensive care unit (ICU). Design: Retrospective cross-sectional and repeated-measures study. Setting: A single US academic medical center (Medical Information Mart for Intensive Care IV [MIMIC-IV], 2008-2019). Patients: 72,944 adult ICU stays with at least 1 delirium screen. Interventions: None. Measurements and Main Results: Among 610,632 screens, 130,455 (21.4%; 95% CI, 21.0%-21.8%) were recorded as UTA, exceeding the 119,052 (19.5%) scored positive. The UTA fraction rose from 2.0% at a Richmond Agitation-Sedation Scale (RASS) score of 0 to 97.8% at RASS -4; 22.0% of UTA screens occurred in arousable patients, where UTA was associated with mechanical ventilation (odds ratio [OR], 3.43; 95% CI, 3.17-3.71) and non-English primary language (OR, 3.74; 95% CI, 3.43-4.08). Building the delirium label three ways from the same patients shifted prevalence modestly (32.1% to 30.8%) and prediction (area under the curve, 0.737 to 0.719) but most affected the delirium-mortality association: in a baseline-adjusted model the OR was 4.12 (95% CI, 3.88-4.36) under complete-case handling and fell to 2.16 (95% CI, 2.06-2.27) when UTA was recoded as negative. UTA was recoverable from the observed clinical state (area under the curve, 0.95). Conclusions: In this ICU cohort, Unable to Assess was the most common recorded delirium result other than Negative, exceeding positive screens; recoding it as negative roughly halved the apparent delirium-mortality association by relabeling deeply sedated, high-mortality patients. Delirium datasets should preserve and report UTA, whose concentration among arousable non-English-speaking patients is a measurable equity target.

Background: Post-induction hypotension is a predictable haemodynamic hazard in older adults undergoing general anaesthesia. Prevention remains divided among volume optimisation, anaesthetic dose reduction, rescue treatment after hypotension occurs and proactive vasoactive support. Methods: We searched PubMed, Embase, Web of Science, CENTRAL, CNKI, Wanfang and VIP from inception to 30 March 2026. Eligible studies were randomised trials of prophylactic vasoactive drugs given before, during or immediately after induction in older adults. The primary outcome was post-induction hypotension. Secondary outcomes were post-induction mean arterial pressure (MAP), systolic arterial pressure (SBP), heart rate (HR) and reported haemodynamic adverse events. Random-effects network meta-analysis was used, and confidence in network estimates was assessed using CINeMA principles. Results: Thirty-one trials including 2,821 participants were included in the revised network. Compared with placebo/control, all active agents favoured lower post-induction hypotension. The most favourable point estimates were observed for phenylephrine (odds ratio [OR] 0.17, 95% confidence interval [CI] 0.01 to 2.16) and metaraminol (OR 0.19, 95% CI 0.02 to 1.53), although both were imprecise. More precise reductions were observed for methoxamine (OR 0.23, 95% CI 0.13 to 0.43), norepinephrine (OR 0.25, 95% CI 0.13 to 0.47) and ephedrine (OR 0.34, 95% CI 0.19 to 0.63). Phenylephrine ranked highest for MAP support, norepinephrine ranked highest for SBP support, and ephedrine ranked highest for HR preservation. Global inconsistency was detected for SBP but not for hypotension incidence, MAP or HR, supporting cautious profile-based interpretation. Conclusions: Prophylactic vasopressor choice during induction should be guided by haemodynamic phenotype rather than ranking alone. In the revised network, active prophylaxis consistently favoured lower hypotension, but sparse nodes produced uncertainty. Norepinephrine retained a comparatively balanced profile when vasodilatory post-induction hypotension is anticipated, phenylephrine and related alpha-agonists provided stronger pressure support when HR and cardiac-output reserve are preserved, and ephedrine was most relevant when chronotropic support is desired. Keywords: general anaesthesia; induction; hypotension; norepinephrine; phenylephrine; ephedrine; network meta-analysis; older adults.

Objective: This study aims to identify which markers of tissue hypoperfusion - specifically lactate levels, central venous oxygen saturation (ScvO2), and venous arterial carbon dioxide gradient (CO2 gradient) - have the highest sensitivity and specificity in predicting the discharge of postoperative cardiac surgical patients from the ICU within 48 hours. This is an exploratory, hypothesis-generating investigation. Methods: Prospective observational study involving 100 patients in the Surgical ICU at InCor-HCFMUSP undergoing cardiac surgery with cardiopulmonary bypass. Perfusion markers were assessed at ICU admission and 24 hours post-admission. Results: ScvO2 at 24 hours was the only marker significantly associated with ICU discharge (OR=1.096; 95% CI=1.020-1.180; p=0.012). Formal DeLong's test confirmed ScvO2 had significantly superior discriminatory performance compared to lactate (AUC 0.661 vs. 0.428; p=0.004). Lactato and CO2 gap showed no significant associations. Conclusions: In this exploratory cohort, ScvO2 at 24 hours post-admission showed a statistically significant association with early ICU discharge and superior discriminatory performance compared to lactate. These findings are hypothesis-generating and require prospective validation before clinical recommendations can be made.

Background: Longer follow-up periods in clinical trials for S. aureus bacteremia (SAB) may capture unrelated deaths, adding random noise that risks biasing trial results towards the null. Objective: To evaluate the timing and infection-relatedness of deaths within a large SAB clinical trial platform. Design: Blinded duplicate adjudication of trial deaths using a modified 7-point Likert-Scale. A third reviewer settled disagreements. Setting: 37 Canadian hospitals participating in the S. aureus Network Adaptive Platform (SNAP) Trial. Participants: 1515 adult patients recruited to SNAP between February 2022 and May 2026. Measurements: Timing and relatedness of 90-day deaths categorized as at least possibly SAB-related not likely to be SAB-related. Optimal follow-up cut-off was determined using Youden's index and graphically. Results: 247 deaths occurred; 97 (39.3%) were adjudicated as at least possibly SAB-related and 150 (60.7%) as not likely related. For probably/definitely related deaths, interrater agreement was 85.0% (Gwet's AC 0.73, substantial); for at least possibly related, it was 77.3% (Gwet's AC 0.55, moderate). Median survival was significantly shorter for SAB-related deaths (12 vs. 30.5 days; difference: 19 days earlier, 95% CI: 12-26, p<0.0001). Nearly 80% of SAB-related deaths occurred by day 30, whereas 50% of unrelated deaths occurred between days 30 and 90. Youden's index optimized follow-up at 20.5 days. Limitations: Potential for cause of death misclassification and data limited to Canadian sites. Conclusion: Deaths considered attributable to SAB cluster rapidly within the first month, while later deaths are predominantly unrelated. A 30-day all-cause mortality window may be more appropriate than 90 days for primary mortality outcomes in trials evaluating acute SAB therapies with longer follow up reserved for metastatic infection and recurrence.

Background: Computer vision-enabled airway workflows can turn airway video into timestamped model-observation fields, but later blinded review and training studies require source-video linkage, run identifiers, quality-control status, and app/model provenance. Objective: To audit whether a public post-reconciliation 30-run manikin evidence package from a computer vision-enabled video laryngoscopy workflow preserved prespecified, video-linked procedural telemetry in structured JSON, while keeping detection accuracy, report quality, and reviewer agreement outside the current claim. Methods: Thirty manikin runs were captured on a HEALTHIBLE Intubation Simulator using an IntuBlade device connected to an iPhone 15 Pro Max. Six predefined conditions were tested with five runs each in planned round-robin order by a board-certified emergency physician operator. The author-affiliated team analyzed corrected Study Metrics JSON exports, the video manifest, app/model metadata, QC fields, and the frozen package checker after reconciliation against the assigned run guide. Blinded video review, independent analysis, and report-quality adjudication were not performed. Results: After reconciliation, all 30 rows contained parseable Study Metrics JSON, a companion videoFilename, run-named Drive video status, QC pass status, and corrected identifiers matching assigned row labels (30/30 for each completeness field; descriptive exact binomial 95% CI, 88.4% to 100.0%). App/model metadata were complete: appVersion 3.3.0 (75), source revision b94cd63, Navigation model, model version 31, and detection threshold 0.1. The exported JSON target-state flag was true in 25 of 25 target-condition rows (95% CI, 86.3% to 100.0%) and false in 5 of 5 no-target controls (95% CI, 47.8% to 100.0%), with zero glottic-detected frames and zero acceptable-view time in no-target controls. Among target-condition rows, median time to first model-detected glottic target was 2 seconds (IQR 1 to 3), median acceptable-view duration was 2.2 seconds (IQR 1.0 to 3.8), and median glottic visibility was 35.8% (IQR 25.8 to 45.6). Interpretation: The corrected package supports a bounded formative claim: simulated airway video can be represented as specified, video-linked computer-vision procedural telemetry after documented reconciliation. It supports package completeness, traceability, and assigned-condition consistency only; it does not establish native uncorrected export reliability, computer-vision detection accuracy, report quality, reviewer agreement, training effectiveness, autonomous guidance, tube-placement confirmation, clinical efficacy, or patient outcomes.

Background: Predicting 28-day mortality in ICU patients with alcoholic cirrhosis is challenging because clinical deterioration is dynamic and heterogeneous. Methods: Using MIMIC-IV (v3.1), this study included 1,907 patients (training n = 1,334; validation n = 573), engineering 208 temporal and static predictors from 64 base variables and reducing them to 40 through multi-stage selection. Seven classifiers and a weighted gradient-boosting ensemble (XGBoost, CatBoost, LightGBM) were compared with Optuna tuning. Results: The ensemble achieved the highest internal validation AUC (0.9276; 95% CI: 0.9011-0.9507) and lowest Brier score (0.0870), with strong discrimination on eICU-CRD (AUC 0.9347) and related MIMIC-III (AUC 0.9071). Ablation indicated that temporal features, especially deltas, were major contributors ({triangleup}AUC {approx} 0.17 when removed). SHAP highlighted APS III score, anion gap, oxygen saturation (delta), lactate, and INR as leading predictors. Conclusions: The framework supports interpretable, trajectory-informed risk stratification in critically ill cirrhotic patients; prospective validation is needed before clinical use.

Background Multicentre trials in anaesthesia and critical care report low rates of statistically significant differences. This finding may partly reflect conventional sample size methods, which assume a fixed treatment effect. Assurance methods use a design prior to represent uncertainty in the expected treatment effect, which may provide a more realistic way of estimating sample sizes. Methods We calculated power curves across a range of effect sizes, design priors, and sample sizes using frequentist and Bayesian assurance methods and compared the sample sizes required to achieve 80% and 90% power to the conventional method. We standardised the design priors across effect sizes using the coefficient of variation. We derived a theoretical limit for achievable power. We validated a normal approximation to the Bayesian posterior distribution. Results Frequentist and Bayesian assurance methods produced similar power curves across all scenarios. At a coefficient of variation of 0.5 - reflecting realistic prior uncertainty in the expected effect size - both methods required sample sizes that were approximately 1.5 to 3.5 times larger than the conventional method. The theoretical power limit depends only on the coefficient of variation of the design prior and holds true across all effect sizes. The normal approximation to the Bayesian posterior distribution matched the results obtained from Markov chain Monte Carlo sampling. Conclusions Incorporating clinical uncertainty in the expected effect size substantially increases the sample size required to achieve adequate power, which has important implications for the feasibility of randomised trials in anaesthesia and critical care.

Background: Alveolar macrophage (AM) dysfunction contributes to Acute Respiratory Distress Syndrome (ARDS) pathogenesis. We investigated the role of extracellular vesicles (EVs) in mediating this dysfunction. Methods: Pulmonary EVs were isolated from broncho-alveolar lavage and non-directed bronchial lavage samples of ventilated sepsis patients with and without ARDS, and post-operative control patients via ultracentrifugation. AMs were isolated from lung tissue resections of lobectomy patients. AMs were treated with pooled EVs for 24 hours prior to functional, metabolic and autophagy profiling. EV cargo was profiled via small RNA transcriptomics and proteomics. Mechanistic role of EV microRNAs was assessed via mimic / antagomir transfection. Results: Pulmonary EVs from sepsis patients with ARDS impaired AM efferocytosis, and control EVs had no effect. ARDS EV treatment enhanced AM mitochondrial-linked respiration, but not glycolysis. ARDS EV treatment impaired LC3B-II and LAMP1 expression, indicating dysregulated AM autophagy-lysosomal machinery. Proteomics revealed downregulation of innate immune pathways in ARDS EVs. Transcriptomics revealed enrichment of 24 microRNAs in ARDS EVs; miR-652-3p was the most enriched, validated by RT-qPCR. EV miR-652-3p was associated with 90-day mortality (9.20 vs 0.59 RQ, p=0.0295) and inversely correlated with oxygenation (PaO2/FiO2). AM transfection with miR-652-3p mimic induced similar dysregulation of function and autophagy as ARDS EVs. Transfection of ARDS EVs with antagomirs to miR-652-3p prior to AM treatment partially rescued efferocytosis and autophagy. Conclusions: Targeting EV miR-652-3p may restore alveolar macrophage function and reduce excessive inflammation, thus offering a novel therapeutic strategy for patients with ARDS.