Abstract Background Portable chest radiographs (CXRs) obtained at presentation of acute respiratory failure (ARF) are interpreted qualitatively to assess airspace disease (ASD) and identify features consistent with acute respiratory distress syndrome (ARDS). The Radiographic Assessment of Lung Edema (RALE) score offers a semiquantitative measure of radiographic ASD, and has demonstrated prognostic value in ARDS, but its ability to quantify radiographic ASD burden across the full spectrum of ARF presentations remains uncertain. Methods We analyzed 4,259 portable CXRs from 814 critically ill adults with expert‑adjudicated ARF subtypes, including ARDS, at‑risk for ARDS, cardiogenic pulmonary edema, acute exacerbation of interstitial lung disease (AE‑ILD), acute‑on‑chronic hypercapnic respiratory failure, and airway‑protection intubations. Trained clinicians, blinded to clinical data, assigned RALE scores. Analyses addressed (1) whether RALE distinguishes CXRs without ASD (airway‑protection intubations and acute-on-chronic hypercapnic respiratory failure) from those with any ASD, and (2) among CXRs with ASD, whether RALE quantifies diffuse bilateral opacification, distinguishing extensive from limited parenchymal involvement. Discrimination was evaluated using receiver operating characteristic curves, with sensitivity analyses assessing the influence of image‑quality features. Results RALE scores increased stepwise from ASD absent to ASD limited to ASD diffuse, discriminated CXRs with any ASD from those without (area under the curve [AUC] 0.81, 95% confidence interval [CI] 0.77–0.85; rule‑out threshold RALE ≥ 7 yielded 96% sensitivity) and quantified diffuse bilateral opacification among ASD-positive CXRs with moderate accuracy (AUC 0.79, 95% CI 0.75–0.82). RALE distributions overlapped substantially across ARDS, cardiogenic pulmonary edema, and AE‑ILD, reflecting shared radiographic appearances rather than etiologic specificity. Suboptimal penetration inflated RALE scores and markedly reduced discrimination (AUC 0.59 vs 0.83 with adequate penetration). Conclusions RALE quantifies radiographic ASD burden across the full spectrum of ARF at intubation — discriminating ASD-absent from ASD-present CXRs and stratifying the extent of parenchymal involvement among those with ASD. However, it did not distinguish between clinical syndromes that share similar diffuse radiographic presentations, highlighting the need to interpret RALE within the broader clinical context and ensure adequate image quality for accurate scoring. These findings reflect RALE’s nature as a radiographic burden metric rather than a diagnostic classifier.

BACKGROUND Acute respiratory failure (ARF) requiring invasive mechanical ventilation affects approximately 1.5 million adult Americans annually. While acute respiratory distress syndrome (ARDS) is well-characterized, non-ARDS ARF—comprising the majority of mechanically ventilated patients—lacks systematic characterization of clinical features, biological profiles, and long-term outcomes. RESEARCH QUESTION How do ARF subtypes differ in clinical, physiologic, and biological characteristics, and long-term survival? STUDY DESIGN AND METHODS Prospective cohort study of 826 mechanically ventilated adults enrolled at a single academic center from 2011-2024. Patients were systematically classified into seven predefined ARF subtypes using consensus methodology. We compared physiology, biological subphenotypes, and 3-year survival. RESULTS Patients were classified as ARDS (36.1%), at-risk for ARDS (ARFA, 34.1%), airway controls (10.7%), other/multifactorial (7.4%), congestive heart failure (CHF, 6.2%), acute-on-chronic hypercapnic respiratory failure (AoCHRF, 3.3%), and acute exacerbation of interstitial lung disease (AEILD, 2.3%). Subtypes differed significantly in gas exchange (hypoxemia and ventilatory ratio), respiratory mechanics (normalized elastance), and radiographic severity (all p<0.01). Plasma inflammatory biomarkers were highest in ARDS and ARFA patients, with hyperinflammatory subphenotype prevalence ranging from 0% (AoCHRF) to 31% (ARFA). Three-year survival ranged from 10.5% (AEILD) to 67.0% (airway controls): 37.7% (other/multifactorial), 48.1% (AoCHRF), 56.9% (CHF), 53.2% (ARFA), 51.3% (ARDS). Hyperinflammatory subphenotype classification predicted lower survival (42.9% vs. 54.9%, p<0.01), predominantly in ARDS and ARFA, while other subtypes showed minimal subphenotype-associated survival differences. INTERPRETATION Mechanically ventilated patients with ARF demonstrate marked heterogeneity in physiology, biology, and long-term outcomes. Overall, fewer than half of patients survived to three years, but survival varied markedly across clinical subtypes beyond traditional ARDS categorization. Clinical subtype classification identifies distinct populations with divergent long-term prognoses, providing information complementary to biological subphenotyping. These findings provide quantitative outcome data to support prognostic counseling and establish a reproducible framework for clinical research in ARF from heterogeneous etiologies.

Background: Chronic critical illness (CCI) is marked by persistent respiratory failure, with patients commonly receiving tracheostomies for ongoing ventilation across ICU and post-ICU settings. Well-documented communication and support breakdowns across CCI worsen family distress and may perpetuate care misaligned with patient values and goals. Yet, effective interventions supporting surrogate decision-makers throughout CCI are lacking. Research Question: What are the communication and support processes that families view as essential to enabling them as active surrogate decision-makers across the patient’s trajectory to better inform future intervention development? Study Design and Methods: We conducted semistructured interviews with surrogate decision-makers of patients who received a tracheostomy for persistent respiratory failure after an acute illness at 2 time points (first, 2 weeks to 6 months after tracheostomy; second, weeks to months later). We used deductive and inductive coding, followed by thematic analysis of interview transcripts. Results: We interviewed 23 surrogate decision-makers of 19 medical and surgical patients with moderate-high illness severity (mean Acute Physiology and Chronic Health Evaluation II score, 21.5 ± 5.3). Five themes were generated from family perspectives, representing the essential communication and support processes to enable families as active decision-makers across time and settings in CCI. They included being (1) engaged proactively during transitions; (2) connected to tailored practical resources that enable their presence, engagement, and space for self-care across the CCI trajectory; (3) actively engaged in ongoing and iterative anticipatory guidance and reassessment; (4) recognized as the expert and advocate for the patient; and (5) treated with compassion and respect. Interpretation: Our findings suggest that future interventions should deliver decisional support in CCI through longitudinal, relationship-centered strategies integrating iterative care planning, compassionate and inclusive family engagement, and tailored practical resources as core intervention components.

Biological heterogeneity in host inflammatory responses to severe pneumonia predicts clinical outcomes and may influence the effectiveness of immunomodulatory therapy. The upstream drivers of this heterogeneity remain poorly defined. We hypothesized that microbial translocation from the lungs to the bloodstream, detectable via multi-compartment metagenomic analysis, contributes to divergent host responses in pneumonia. In this nested case–control study of mechanically ventilated patients with severe pneumonia, we collected paired plasma and endotracheal aspirate samples at baseline. Plasma samples underwent microbial cell-free DNA (mcfDNA) sequencing, and endotracheal aspirates were analyzed by Nanopore metagenomic sequencing. Host-response biomarkers were measured in both plasma and endotracheal aspirate samples. Microbial translocation of pulmonary origin was defined by the genus-level concordance of detectable taxa between matched endotracheal aspirate and plasma samples. Among 98 patients (76 pneumonia, 22 controls), plasma mcfDNA was markedly higher in microbiologically confirmed pneumonia compared with culture-negative pneumonia (median 4015 vs. 210 molecules/μL, p = 0.0006). Pulmonary microbial translocation was identified in 31 (41

BACKGROUND:Reproducible quantification of pulmonary edema on chest radiographs is challenging for bedside clinicians. The Radiographic Assessment of Lung Edema (RALE) score offers a structured, semi-quantitative approach with prognostic implications in acute respiratory failure (ARF), yet its clinical use is limited by the absence of standardized training. OBJECTIVE:To determine whether a standardized RALE training program, combined with expert feedback, improves inter-rater reliability across clinician experience levels. METHODS:We analyzed 4487 chest radiographs from 864 critically ill patients with ARF enrolled in a prospective registry. Sixteen physicians (1 expert and 15 reviewers-including 5 interns, 4 residents, 3 fellows, and 3 attendings) completed a 3-phase RALE training program: (i) independent review of instructional materials, (ii) a live virtual session led by the expert, and (iii) iterative practice with expert feedback on discrepant scores. All scoring was performed using the Pulmo-Annotator platform. We assessed inter-rater reliability using intraclass correlation coefficients (ICC), Bland-Altman analysis, and Deming regression, while stratifying performance by experience level and self-reported confidence. RESULTS:Inter-rater reliability improved following the training program, with ICC increasing from 0.89 (95% CI: 0.85-0.92) to 0.93 (95% CI: 0.90-0.95) after expert feedback. Fellows demonstrated the largest gain of ICC from 0.89 to 0.97, while interns and attendings maintained high reliability. Post-feedback analyses indicated reduced systematic bias. Residents exhibited the greatest variability and revised fewer scores post-feedback, despite equivalent exposure to feedback. Score distributions also varied by self-reported confidence, with greater variability among less experienced reviewers. CONCLUSION:Following a structured training program, radiographic edema scoring using the RALE framework demonstrated excellent inter-rater reliability, which further improved with targeted expert feedback. This structured training program, which incorporated standardized instructional materials and iterative feedback, demonstrated that RALE scoring is teachable across different levels of clinical experience and can be integrated into educational curricula to enhance consistency in chest radiograph interpretation. These findings provide a foundation for future work exploring the role of RALE training in broader research and clinical applications.

Binary inflammatory subphenotype classification (hyperinflammatory vs. hypoinflammatory) may guide trial enrollment in acute hypoxemic respiratory failure (AHRF), but assumes within-category homogeneity. We determined whether continuous probabilities reveal clinically meaningful heterogeneity. We analyzed 575 critically ill adults with AHRF (Pittsburgh Acute Lung Injury Registry) and validated findings in 1134 patients from the EDEN trial, the COVID-19 cohorts, and the RoCI registry. Continuous subphenotype probabilities were calculated using a parsimonious biomarker model (IL-6, sTNFR-1, bicarbonate; probability threshold 0.5). The primary outcome was 90-day mortality. Among 575 patients, 77 patients (13

Objective: To determine whether ivermectin improves outcomes for critically and noncritically ill hospitalized patients with COVID-19.Design: An ongoing international, multifactorial, adaptive platform, randomized, controlled trial.Setting: Hospitals in Pakistan, India, and Ireland between June 11, 2021, and September 9, 2022.Patients: Critically and noncritically ill patients.Interventions: Randomized to ivermectin or no ivermectin (control).Measurements and Main Results: The primary outcome was respiratory and cardiovascular organ support-free days, assessed on an ordinal scale combining in-hospital death (assigned a value of -1) and days free of organ support through day 21 in survivors. Analyses used a Bayesian cumulative logistic model. Enrollment was closed for operational futility, following external evidence suggesting no benefit with ivermectin in nonhospitalized patients with COVID-19. Among 61 critically ill patients, the median number of organ support-free days was -1, indicating death was the most common vital outcome (interquartile range [IQR], -1 to 17), for the ivermectin group and -1 (IQR, -1 to 17.25) for the control group (adjusted proportional odds ratio [OR], 0.94; 95% credible interval [CrI], 0.40-2.07) and the posterior probability of superiority to control was 44.2%. Among 89 noncritically ill patients, the median number of organ support-free days was 22 (IQR, 18.5-22) for ivermectin and 22 (IQR, 16-22) for control (adjusted proportional OR, 1.04; 95% CrI, 0.48-2.34) and the posterior probability of superiority was 53.7%. Among critically ill patients, hospital survival was 35.1% (13/37) for ivermectin and 37.5% (9/24) for control (adjusted OR, 1.00; 95% CrI, 0.39-2.32), posterior probability of superiority was 50.0%. Among noncritically ill patients, hospital survival was 84.1% (37/44) for ivermectin and 77.8% (35/45) for control (adjusted OR, 1.16; 95% CrI, 0.5-3.07), posterior probability of superiority was 63.3%.Conclusions: For critically and noncritically ill hospitalized patients with COVID-19, ivermectin was unlikely to improve the primary composite outcome of organ support-free days and hospital survival.

BACKGROUND:Liberation from invasive mechanical ventilation (IMV) is a central therapeutic objective in acute respiratory failure (ARF). While lung-protective ventilation strategies and diaphragm function have received considerable attention as determinants of liberation success, the prognostic role of extrapulmonary thoracic tissue composition remains poorly defined. We examined whether chest computed tomography (CT)-derived thoracic skeletal muscle mass (TMM), subcutaneous fat mass (SFM), and their ratio (muscle-fat ratio: MFR) are associated with IMV outcomes in ARF. METHODS:This secondary analysis of a prospective observational ARF cohort includes patients with a chest CT obtained within 7 days before or 3 days after intubation. TMM, SFM and MFR were derived using validated three-dimensional convolutional neural network-based volumetric segmentation spanning the lung apices to bases. Co-primary outcomes were time to successful ventilator liberation within 60 days and 60-day all-cause mortality. Multivariable Cox and logistic regression models with restricted cubic splines assessed linear and non-linear associations, adjusted for age, sex, and key comorbidities. We performed prespecified BMI-stratified and ARDS subgroup analyses. RESULTS:Among 826 patients with ARF requiring IMV, 238 patients with CT imaging were included for analysis (median age 59.5 years, median BMI 28.7 kg/m²). A higher MFR was independently associated with faster ventilator liberation (hazard ratio 1.16, 95% confidence interval [CI] 1.02-1.32, p=0.023). TMM demonstrated a significant J-shaped non-linear relationship with time to liberation (p for non-linearity=0.019), with delayed liberation at both low and high extremes of TMM. Higher SFM was independently associated with increased 60-day mortality (odds ratio 1.09, 95% CI 1.00-1.18, p=0.049). In 81 patients with ARDS, TMM also demonstrated a significant J-shaped non-linear relationship with 60-day mortality (p for non-linearity=0.037). BMI was not significantly associated with either outcome. CONCLUSIONS:Volumetric CT-derived thoracic body composition metrics, particularly MFR and TMM, are independently associated with ventilator liberation and 60-day mortality in ARF, capturing prognostic information not reflected by BMI alone. These findings support the incorporation of automated thoracic body composition analysis into early risk stratification frameworks for mechanically ventilated patients and highlight extrapulmonary tissue composition as an underexplored determinant of IMV outcomes. Key Points:Question: In adults with acute respiratory failure on invasive mechanical ventilation, are CT-derived thoracic skeletal muscle mass (TMM), subcutaneous fat mass (SFM), and muscle-fat ratio (MFR) associated with ventilator liberation and 60-day mortality?Findings: In this analysis of a prospective cohort of 238 patients, higher MFR was associated with faster liberation and TMM showed a J-shaped association with liberation. Higher SFM predicted greater 60-day mortality, while BMI category predicted neither.Meaning: Automated CT-derived thoracic body composition adds prognostic information beyond BMI and may aid early risk stratification in mechanically ventilated ARF patients.

Background COVID‐19 and other respiratory viral infections can cause cardiovascular complications. SARS‐CoV‐2 virions are found in the blood, and circulating viral RNA levels are associated with death. We hypothesized that viremia can induce thrombotic endotheliopathy that contributes to death and studied this relationship in patients hospitalized for COVID‐19 and enrolled in the ACTIV‐4a (Accelerating COVID‐19 Therapeutic Interventions and Vaccines) randomized trial of antithrombotic therapy. Methods We quantified SARS‐CoV‐2 nucleocapsid RNA and protein in plasma and measured their associations with clinical outcomes and biomarkers of thromboinflammation and endotheliopathy. We used Cox regression and Fine–Gray competing risk models to analyze survival and thrombosis. We conducted causal mediation analysis to explore whether thrombotic endotheliopathy mediates the relationship between viral RNA and death. Results In 93 patients, SARS‐CoV‐2 RNA and N‐antigen were higher in nonsurvivors. Baseline viral RNA levels were associated with increased 90‐day death (hazard ratio [HR], 1.31 [95% CI, 1.17–1.46]) and thrombosis (HR, 1.35 [95% CI, 1.24–1.47]). Viral RNA more effectively predicted survivorship than N‐antigen levels. Soluble thrombomodulin, a biomarker of thrombotic endotheliopathy, positively correlated with viral RNA levels. Mediation analysis revealed that soluble thrombomodulin accounts for 12.2% (95% CI, 0.1%–32.3%; P=0.048) after adjustment for age and sex of the relationship between viral RNA and the 90‐day mortality rate. Conclusions Elevated plasma SARS‐CoV‐2 RNA levels are associated with death in ACTIV‐4a, which is causally mediated in part by soluble thrombomodulin. We propose that lung–blood viral dissemination is a potential mechanism for cardiovascular complications of respiratory viruses. Registration URL: https://www.clinicaltrials.gov; Unique Identifier: NCT04505774.

RATIONALE:The Early versus Delayed Enteral Nutrition (EDEN) trial found no significant difference in mortality between trophic and full enteral nutrition strategies in acute respiratory distress syndrome (ARDS) patients. Heterogeneity of treatment effect (HTE) has been identified in prior ARDS trials. We previously identified intestine-derived incretin hormones (glucose-dependent insulinotropic peptide [GIP] and glucagon-like peptide [GLP]-1 as potential predictive biomarkers in the response to nutrition. OBJECTIVES:To investigate incretins as biomarkers predictive of HTE in EDEN. METHODS:GIP, GLP-1, and host immune response biomarkers were measured from pre-intervention EDEN plasma samples. We investigated HTE with 60-day mortality as a primary outcome by testing interaction of treatment with circulating incretin levels in analyses adjusted for demographics, severity of illness, diabetes mellitus, and circulating interleukin-6, and assessed mortality by treatment arm across incretin tertiles. We additionally tested for HTE by de novo ARDS subphenotypes and by risk of mortality. MEASUREMENTS & MAIN RESULTS:889 participants were included (452 randomized to trophic and 437 to full enteral nutrition). GIP predicted HTE to enteral nutrition strategies (adjusted interaction p-value 0.01) with lower mortality from trophic feeds (14.1% vs 27.2% in full) in patients in the highest GIP tertile but similar mortality in other tertiles. GLP-1, ARDS subphenotypes, and baseline risk of mortality did not predict HTE. CONCLUSIONS:GIP was unique among incretins in predicting HTE to enteral nutrition strategies in EDEN. Further studies are needed to validate our findings as GIP might serve as a biomarker to guide level of enteral nutrition for ARDS patients.

BACKGROUND:The optimal thromboprophylaxis among critically ill adults with COVID-19 is uncertain. OBJECTIVES:To determine the effectiveness and safety of intermediate-dose heparin compared with standard low-dose thromboprophylaxis. METHODS:In an ongoing adaptive platform trial (randomized embedded multifactorial adaptive platform for community-acquired pneumonia), critically ill patients with COVID-19 were randomized to intermediate-dose heparin or standard low-dose thromboprophylaxis. Interventions were continued in hospital for up to 14 days. The primary endpoint was organ support-free days (OSFDs), an ordinal outcome combining in-hospital survival and the number of days free of intensive care unit-based respiratory or cardiovascular organ support through 21 days. The primary analysis was an adjusted Bayesian hierarchical cumulative logistic model. An odds ratio (OR) > 1.0 represents an improved outcome with intermediate-dose heparin. RESULTS:Between April 27, 2021 and November 25, 2023, 1255 critically ill adults with COVID-19 were enrolled from 78 sites in 15 countries, of whom 1254 completed follow-up (n = 572 intermediate-dose, n = 682 low-dose). Enrollment was terminated prior to reaching a prespecified statistical trigger due to declining case numbers and slow recruitment. Median age was 59 years, and 36.7% were female (n = 461/1255). The probability that intermediate-dose heparin improved OSFDs was 73.5% (OR, 1.06; 95% credible interval, 0.87, 1.30), which did not meet the prespecified superiority threshold of 99%. Hospital survival was 77.1% (441/572) and 76.7% (523/682) in the intermediate- and low-dose heparin groups, respectively (median adjusted OR, 1.14; 95% credible interval, 0.86, 1.52). Major bleeding occurred in 10 of 572 (1.7%) and 14 of 682 (2.1%) patients receiving intermediate and standard low doses, respectively. CONCLUSION:Intermediate-dose heparin did not improve OSFDs or survival compared with standard thromboprophylaxis in critically ill patients with COVID-19. (ClinicalTrials.gov number: CT02735707).

PURPOSE:Subphenotype classifiers for acute respiratory distress syndrome (ARDS) dichotomise patients into hyperinflammatory versus hypoinflammatory subgroups. These models demonstrated prognostic and predictive values but were developed primarily in Caucasian populations. Generalisability of these models in Asian patients, who experience worse clinical outcomes, has not been established. We aimed to profile host responses in Asian patients with ARDS and evaluate the generalisability of established classifiers in this understudied population compared with a Caucasian cohort. METHODS:We prospectively enrolled patients with ARDS from medical intensive care units in Beijing, China, and Pittsburgh, Pennsylvania, USA. In the Beijing cohort, 37 protein biomarkers were measured, with 10 overlapping biomarkers measured in the Pittsburgh cohort. Six established subphenotype models were assessed for generalisability and intermodel agreement. Sensitivity analyses, including latent class analysis, were conducted to explore biological heterogeneity within Asians. RESULTS:Between 2011 and 2020, a total of 356 patients with ARDS (83% meeting the Berlin Definition; the rest on high-flow nasal cannula (HFNC) meeting the New Global Definition) were enrolled across Beijing (97% Han Asian) and Pittsburgh (90% Caucasian) sites, with comparable baseline hypoxaemia severity but disparate outcome. While the proportion of hyperinflammatory versus hypoinflammatory subphenotypes was predicted to be overall similar across different cohorts per each model, we observed poor intermodel agreement. We observed heightened inflammation in Berlin patients with ARDS compared with HFNC-ARDS within our Asian cohort. CONCLUSION:Established subphenotype classifiers demonstrated similar distribution of subphenotypes in Asian patients with ARDS. However, poor intermodel agreement highlights the need for further investigation into model variability with models coming closer to bedside implementation. TRIAL REGISTRATION NUMBER:NCT02975908.

BACKGROUND: Acute respiratory failure is a common and deadly diagnosis among patients receiving intensive care. Physicians have a variety of oxygenation support devices available to support patients, ranging from low-flow oxygen to high-flow nasal cannula, noninvasive ventilation, and invasive mechanical ventilation. Clinical equipoise exists regarding selection of a specific device, spurred by varying physician practice patterns and preferences. RESEARCH QUESTIONS: Do physician preferences for oxygenation support vary within a single institution? STUDY DESIGN AND METHODS: A survey distributed to 100 critical care physicians across 5 academic and community hospitals within an integrated health care system in Western Pennsylvania. The survey contained 5 real-world clinical scenarios across 5 domains: (1) a patient experiencing an acute COPD exacerbation with refractory hypercapnia, (2) a patient with worsening hypoxemia secondary to streptococcal pneumonia, (3) an immunosuppressed host with severe COVID-19 pneumonia, (4) an older patient with multiple cardiopulmonary comorbidities demonstrating hypoxemia and septic shock, and (5) a patient with flash pulmonary edema in the setting of multiple cardiac comorbidities. Physicians were asked to provide an oxygenation support recommendation from the following: (1) low-flow or facemask oxygen, (2) heated high-flow nasal cannula, (3) bilevel pressure ventilation, or (4) endotracheal intubation to initiate mechanical ventilation. We assessed interrater agreement within each scenario by percent agreement and performed a sensitivity analysis dichotomizing clinician recommendation around endotracheal intubation. RESULTS: A total of 44 (44%) physicians completed the survey instrument. Initiating mechanical ventilation was chosen variably by physicians ranging from 5% (n = 2, scenario 2) to 61% (n = 27, scenario 3). Interrater agreement was poor across scenarios 1, 3, 4, and 5 (range, 0.32-0.56) and moderate only in scenario 2 (0.83). Dichotomizing the clinical decision to pursue endotracheal intubation or not improved agreement slightly in scenario 2 (0.92) and scenario 5 (0.73). INTERPRETATION: Our results show that critical care physicians are unlikely to agree on oxygenation support device for patients receiving intensive care, confirming clinical equipoise. CHEST Critical Care 2026; 4(1):100224

Patients receiving prolonged mechanical ventilation are often discharged to long-term acute care hospitals (LTACHs) with hopes of recovery and ultimately return to the community. Among those who survive and undergo tracheostomy, little is known about their quality of life and social outcomes after LTACH discharge. Measure health related quality of life in a cohort of critical illness survivors who underwent tracheostomy and an LTACH stay Single center, prospective observational cohort study conducted at a long-term acute care hospital between 2022-2024. Adults with prolonged mechanical ventilation requiring tracheostomy were eligible. Survivors or surrogates completed a telephone survey 3-6 months after LTACH admission. Surveys included the Katz Index of Independence in Activities of Daily Living (ADL) and Patient-Reported Outcome Measurement Information System (PROMIS) measures of physical, mental, and social health. Descriptive statistics summarized scores; exploratory analyses examined associations between persistent tracheostomy and residence. Seventy participants were enrolled (median age 64 years; 58.6% male). Median ICU length of stay was 40 days prior to LTACH transfer. By LTACH discharge, 39 people (56%) had been decannulated. At a median follow-up of 5.5 months, 50 participants (71%) were alive, 39 completed the survey. Most respondents reported impairments in feeding, dressing, and bathing. PROMIS t-scores demonstrated severe impairments in physical function (median 26.7), and ability to participate in social roles (37.0), with high symptom burden of depression (61.9) and anxiety (58.3). Greater impairment was observed among participants with ongoing tracheostomy or were not residing at home. Home residence did not differ significantly by decannulation status (61.8% decannulated vs. 46.7% not decannulated; OR 1.85 (95% CI 0.54, 6.30)). Survivors of PMV and LTACH admission experience marked functional impairments, restricted participation in social roles, and prominent symptoms of depression and anxiety—particularly among patients not living at home or with persistent tracheostomy. These data may help clinicians prepare patients and families for life after tracheostomy and guide tailored support addressing multifaceted needs. National Institutes of Health

OBJECTIVES:Delirium is common and harmful in the ICU. The Intensive Care Delirium Screening Checklist (ICDSC) and Confusion Assessment Method for the ICU (CAM-ICU) are validated tools recommended for delirium identification. However, the accuracy of bedside nurse-documented delirium assessments in the ICU is inconsistent, limiting utility in clinical research. We sought to evaluate and optimize agreement between bedside nurse-documented and trained researcher delirium assessments. DESIGN, SETTING, AND PATIENTS:Critically ill adults with acute respiratory failure or sepsis in ICUs in large academic hospitals in a southwestern Pennsylvania health system were assessed daily for delirium by bedside nurses (using the ICDSC) and trained researchers (using the CAM-ICU). Using matched nurse-to-researcher delirium assessments, we categorized delirium status using validated cutoffs and evaluated agreement using Cohen's kappa. We derived and compared logistic regression models that used ICDSC documentation, mechanical ventilation status, and admission Sequential Organ Failure Assessment to predict delirium in noncomatose patients, using researcher CAM-ICU assessments as the reference standard. We internally validated models using ten-fold cross-validation. INTERVENTIONS:None. MEASUREMENTS AND MAIN RESULTS:From a sample of 1535 matched assessments of 279 patients, there was moderate agreement between bedside nurse assessments using the established ICDSC delirium/normal cutoff (ICDSC ≥ 4) and trained researcher assessments using the CAM-ICU (Cohen's kappa = 0.42). A logistic regression model informed by individual ICDSC components and clinical data predicted a positive research CAM-ICU with good discrimination (area under the curve = 0.87) and performed well in cross-validation (F1 score = 0.72). In sensitivity analyses, models with more limited ICDSC information demonstrated fair to good discriminatory ability (F1 = 0.60-0.70), with the validated cutoff model having the lowest performance. CONCLUSIONS:A delirium model informed by bedside nurse ICDSC findings and clinical variables improves accuracy of delirium detected in the ICU and can be used in future pragmatic research that leverages large clinical datasets to advance understanding of delirium mechanisms, trajectories, and outcomes.

Rationale: The Radiographic Assessment of Lung Edema (RALE) score is an established tool for quantifying lung edema severity in patients with Acute Respiratory Distress Syndrome (ARDS). However, the RALE score alone may not fully capture the complexity of radiographic findings associated with prognosis in critically ill patients with other forms of acute respiratory failure (ARF). Additional chest X-ray (CXR) features, such as the presence of atelectasis, along with physician gestalt—expert clinical impressions formed from viewing CXRs—may provide valuable prognostic insights that complement RALE scoring and reveal subtle cues predictive of outcomes. We aimed to evaluate the prognostic value of specific CXR features beyond RALE scores in ARF. Methods: We conducted a prospective, multisite observational study involving 827 critically ill patients with ARF admitted to ICUs. Baseline CXRs upon ICU admission and follow-up CXRs (days 5-7) were independently evaluated by at least two trained raters who assessed RALE scores, presence of atelectasis, artifacts, and other radiographic features. Raters provided survival predictions using an ordinal scale (1 = imminent death, 2 = 30-day mortality, 3 = 90-day mortality, 4 = 1-year mortality, 5 = long-term survival). We examined the relationship between CXR features, RALE scores, physician predictions, and 90-day mortality using Kaplan-Meier curves and Cox regression models. Results: We analyzed the first available (baseline) CXR upon ICU admission in 827 ARF patients (34% ARDS; median[IQR] RALE: 22.6[15.8]) and the first available CXR obtained on days 5-7 post-baseline (n = 442; 44% ARDS; median[IQR] RALE: 23.6[13.3]). Both baseline and follow-up RALE scores were prognostic of 90-day survival (hazard ratio [95% CI] = 1.02[1.01-1.03] and 1.04[1.02-1.06], p<0.0001, respectively). Bilateral or left-sided atelectasis at baseline was significantly associated with worse survival, whereas right-sided atelectasis had similar survival to patients without atelectasis. At follow-up, atelectasis was no longer predictive of survival. The presence of artifacts related to critical care (such as defibrillator pads or external tubing) did not correlate with survival outcomes. At both baseline and follow-up, physician predictions of survival were significantly associated with observed survival. However, physician predictions were significantly influenced by RALE scores; when adjusted for these scores in Cox models, their predictive significance diminished. Conclusions: The RALE score is a valuable tool for assessing lung edema in critically ill patients with ARF, yet may not fully capture complex prognostic factors. Integrating specific CXR features and expert assessments with traditional scoring systems could enhance prognostic accuracy and improve patient management in ARF populations.

Introduction: Acute respiratory failure is a common diagnosis in the acutely ill, with an inpatient mortality rate of 12%. Multiple options are available to support patients with acute respiratory failure, including high flow nasal cannula, non-invasive ventilation, and invasive ventilation. At the moment of patient decompensation, however, there may be variability in the timing and use of organ support, and little known about factors that dictate clinicians’ decision to perform endotracheal intubation. Methods: We surveyed critical care clinicians in five academic and community hospitals within an integrated healthcare system in Western Pennsylvania in September 2024. Participants each read five clinical vignettes describing the laboratory parameters, vital signs, and radiographic findings of hypothetical patients experiencing decompensated acute respiratory failure who were amenable to life sustaining treatment. The vignettes ranged from COPD exacerbation with refractory, symptomatic hypercapnia to an immunosuppressed with severe COVID-19 pneumonia on heated high flow nasal cannula. Participants were asked to indicate their preferred next step in terms of respiratory support: low-flow oxygen, heated high flow oxygen, non-invasive ventilation or perform endotracheal intubation and initiate invasive ventilation. Data were analyzed for intragroup interrater agreement using percent agreement where > 0.8 was very good agreement, 0.7-0.8 was good agreement, 0.6-0.7 was moderate agreement and < 0.6 was poor agreement. Results: We received 37 complete responses out of 100 surveyed individuals (response rate of 37%), with 11 (30%) being clinical trainees and 12 (32%) with more than five years of experience. The pooled percent agreement across all cases was 0.24 (SD, 0.21-0.27). Pair-wise comparison of percent agreement between cases was universally poor (< 0.5) except between cases one and five (0.62, SD, 0.46-0.79) given good agreement on non-invasive initiation (Table 1). In the vignette concerning an acutely hypoxemic patient with underlying idiopathic pulmonary fibrosis, clinicians recommended endotracheal intubation in 22%, high flow oxygen in 49%, and NIV in 8%. Alternatively, in the vignette concerning a young patient with streptococcal pneumonia, there was near universal to initiate high flow-nasal cannula (89%) (Table 1). Conclusions: In common clinical vignettes of decompensating patients with respiratory failure, clinicians’ decision about the amount of respiratory support and endotracheal intubation was variable with poor interrater agreement. These results suggest clinical equipoise in current management of acute respiratory failure. Identification of salient respiratory subtypes or leveraging patient data beyond typical physiological parameters may prove useful in personalizing respiratory support.