Evaluation of Three Cytomegalovirus IgG Lateral Flow Assays for Rapid Determination of CMV Serostatus.

Background: Cytomegalovirus (CMV) serostatus is a major determinant of CMV infection, disease risk, and transplant outcomes. Current clinical serology assays are limited by relatively slow turnaround time, design for batched testing, need for trained personnel, and/or specialized equipment. Rapid diagnostic assays in development have a role in emerging settings, such as critically ill patients, but have not been systematically evaluated. Methods: We assessed the performance of 3 rapid lateral flow assays (LFAs) for the detection of CMV immunoglobulin (Ig)G antibodies compared with a reference commercially available CMV IgG enzyme-linked immunosorbent assay in residual serum samples from 200 consecutive adults who underwent clinical CMV serology testing. Samples with discrepant results between the LFA and reference assay were tested by a second reference assay. A subset of serum samples was assessed for interoperator variability. Operating characteristics of the QooLabs LFA were separately assessed in plasma samples. Results: The sensitivity and specificity of the individual LFA assays using serum varied significantly: 86%/83%, 99/93%, and 57/97%, for Healgen, QNow automated reader, and nanoComposix, respectively, compared with the reference assay. Results for the QNow assay were comparable between automated and manual reads. Among a subset of 10 serum samples assessed by 5 individual operators, 44 of 50 (88%) results were concordant. Among 50 plasma samples assessed by the QooLabs LFA, the sensitivity and specificity were 72% and 96%. Conclusions: The ease of performance, rapid turnaround time, and good operating characteristics provide the rationale for further evaluation of the Qoolabs QNow LFA in specialized settings where rapid assessment of CMV serostatus would be advantageous.

Machine-learning assisted swallowing assessment: a deep learning-based quality improvement tool to screen for post-stroke dysphagia.

Introduction: Post-stroke dysphagia is common and associated with significant morbidity and mortality, rendering bedside screening of significant clinical importance. Using voice as a biomarker coupled with deep learning has the potential to improve patient access to screening and mitigate the subjectivity associated with detecting voice change, a component of several validated screening protocols. Methods: In this single-center study, we developed a proof-of-concept model for automated dysphagia screening and evaluated the performance of this model on training and testing cohorts. Patients were admitted to a comprehensive stroke center, where primary English speakers could follow commands without significant aphasia and participated on a rolling basis. The primary outcome was classification either as a pass or fail equivalent using a dysphagia screening test as a label. Voice data was recorded from patients who spoke a standardized set of vowels, words, and sentences from the National Institute of Health Stroke Scale. Seventy patients were recruited and 68 were included in the analysis, with 40 in training and 28 in testing cohorts, respectively. Speech from patients was segmented into 1,579 audio clips, from which 6,655 Mel-spectrogram images were computed and used as inputs for deep-learning models (DenseNet and ConvNext, separately and together). Clip-level and participant-level swallowing status predictions were obtained through a voting method. Results: The models demonstrated clip-level dysphagia screening sensitivity of 71% and specificity of 77% (F1 = 0.73, AUC = 0.80 [95% CI: 0.78-0.82]). At the participant level, the sensitivity and specificity were 89 and 79%, respectively (F1 = 0.81, AUC = 0.91 [95% CI: 0.77-1.05]). Discussion: This study is the first to demonstrate the feasibility of applying deep learning to classify vocalizations to detect post-stroke dysphagia. Our findings suggest potential for enhancing dysphagia screening in clinical settings. https://github.com/UofTNeurology/masa-open-source.

Early and late withdrawal of life-sustaining treatment after out-of-hospital cardiac arrest in the United Kingdom: Institutional variation and association with hospital mortality.

AIM: Frequency and timing of Withdrawal of Life-Sustaining Treatment (WLST) after Out-of-Hospital Cardiac Arrest (OHCA) vary across Intensive Care Units (ICUs) in the United Kingdom (UK) and may be a marker of lower healthcare quality if instituted too frequently or too early. We aimed to describe WLST practice, quantify its variability across UK ICUs, and assess the effect of institutional deviation from average practice on patients' risk-adjusted hospital mortality. METHODS: We conducted a retrospective multi-centre cohort study including all adult patients admitted after OHCA to UK ICUs between 2010 and 2017. We identified patient and ICU characteristics associated with early (within 72 h) and late (>72 h) WLST and quantified the between-ICU variation. We used the ICU-level observed-to-expected (O/E) ratios of early and late-WLST frequency as separate metrics of institutional deviation from average practice and calculated their association with patients' hospital mortality. RESULTS: We included 28,438 patients across 204 ICUs. 10,775 (37.9%) had WLST and 6397 (59.4%) of them had early-WLST. Both WLST types were strongly associated with patient-level demographics and pre-existing conditions but weakly with ICU-level characteristics. After adjustment, we found unexplained between-ICU variation for both early-WLST (Median Odds Ratio 1.59, 95%CrI 1.49-1.71) and late-WLST (MOR 1.39, 95%CrI 1.31-1.50). Importantly, patients' hospital mortality was higher in ICUs with higher O/E ratio of early-WLST (OR 1.29, 95%CI 1.21-1.38, p < 0.001) or late-WLST (OR 1.39, 95%CI 1.31-1.48, p < 0.001). CONCLUSIONS: Significant variability exists between UK ICUs in WLST frequency and timing. This matters because unexplained higher-than-expected WLST frequency is associated with higher hospital mortality independently of timing, potentially signalling prognostic pessimism and lower healthcare quality.

Estimating the attributable fraction of mortality from acute respiratory distress syndrome to inform enrichment in future randomised clinical trials.

BACKGROUND: Efficiency of randomised clinical trials of acute respiratory distress syndrome (ARDS) depends on the fraction of deaths attributable to ARDS (AFARDS) to which interventions are targeted. Estimates of AFARDS in subpopulations of ARDS could improve design of ARDS trials. METHODS: We performed a matched case-control study using the Large observational study to UNderstand the Global impact of Severe Acute respiratory FailurE cohort. Primary outcome was intensive care unit mortality. We used nearest neighbour propensity score matching without replacement to match ARDS to non-ARDS populations. We derived two separate AFARDS estimates by matching patients with ARDS to patients with non-acute hypoxaemic respiratory failure (non-AHRF) and to patients with AHRF with unilateral infiltrates only (AHRF-UL). We also estimated AFARDS in subgroups based on severity of hypoxaemia, number of lung quadrants involved and hyperinflammatory versus hypoinflammatory phenotypes. Additionally, we derived AFAHRF estimates by matching patients with AHRF to non-AHRF controls, and AFAHRF-UL estimates by matching patients with AHRF-UL to non-AHRF controls. RESULTS: Estimated AFARDS was 20.9% (95% CI 10.5% to 31.4%) when compared with AHRF-UL controls and 38.0% (95% CI 34.4% to 41.6%) compared with non-AHRF controls. Within subgroups, estimates for AFARDS compared with AHRF-UL controls were highest in patients with severe hypoxaemia (41.1% (95% CI 25.2% to 57.1%)), in those with four quadrant involvement on chest radiography (28.9% (95% CI 13.4% to 44.3%)) and in the hyperinflammatory subphenotype (26.8% (95% CI 6.9% to 46.7%)). Estimated AFAHRF was 33.8% (95% CI 30.5% to 37.1%) compared with non-AHRF controls. Estimated AFAHRF-UL was 21.3% (95% CI 312.8% to 29.7%) compared with non-AHRF controls. CONCLUSIONS: Overall AFARDS mean values were between 20.9% and 38.0%, with higher AFARDS seen with severe hypoxaemia, four quadrant involvement on chest radiography and hyperinflammatory ARDS.

Linking Acute Physiology to Outcomes in the ICU: Challenges and Solutions for Research.

ICU clinicians rely on bedside physiological measurements to inform many routine clinical decisions. Because deranged physiology is usually associated with poor clinical outcomes, it is tempting to hypothesize that manipulating and intervening on physiological parameters might improve outcomes for patients. However, testing these hypotheses through mathematical models of the relationship between physiology and outcomes presents a number of important methodological challenges. These models reflect the theories of the researcher and can therefore be heavily influenced by one's assumptions and background beliefs. Model building must therefore be approached with great care and forethought, because failure to consider relevant sources of measurement error, confounding, coupling, and time dependency or failure to assess the direction of causality for associations of interest before modeling may give rise to spurious results. This paper outlines the main challenges in analyzing and interpreting these models and offers potential solutions to address these challenges.

Mortality As a Measure of Treatment Effect in Clinical Trials Recruiting Critically Ill Patients.

OBJECTIVES: All-cause mortality is a common measure of treatment effect in ICU-based randomized clinical trials (RCTs). We sought to understand the performance characteristics of a mortality endpoint by evaluating its temporal course, responsiveness to differential treatment effects, and impact when used as an outcome measure in trials of acute illness. DATA SOURCES: We searched OVID Medline for RCTs published from 1990 to 2018. STUDY SELECTION: We reviewed RCTs that had randomized greater than or equal to 100 patients, were published in one of five high-impact general medical or eight critical care journals, and reported mortality at two or more distinct time points. We excluded trials recruiting pediatric or neonatal patients and cluster RCTs. DATA EXTRACTION: Mortality by randomization group was recorded from the article or estimated from survival curves. Trial impact was assessed by inclusion of results in clinical practice guidelines. DATA SYNTHESIS: From 2,592 potentially eligible trials, we included 343 RCTs (228,784 adult patients). While one third of all deaths by 180 days had occurred by day 7, the risk difference between study arms continued to increase until day 60 (p = 0.01) and possibly day 90 (p = 0.07) and remained stable thereafter. The number of deaths at ICU discharge approximated those at 28-30 days (95% [interquartile range [IQR], 86-106%]), and deaths at hospital discharge approximated those at 60 days (99% [IQR, 94-104%]). Only 13 of 43 interventions (30.2%) showing a mortality benefit have been adopted into widespread clinical practice. CONCLUSIONS: Our findings provide a conceptual framework for choosing a time horizon and interpreting mortality outcome in trials of acute illness. Differential mortality effects persist for 60 to 90 days following recruitment. Location-based measures approximate time-based measures for trials conducted outside the United States. The documentation of a mortality reduction has had a modest impact on practice.

Rethinking Acute Respiratory Distress Syndrome after COVID-19: If a "Better" Definition Is the Answer, What Is the Question?

The definition of acute respiratory distress syndrome (ARDS) has a somewhat controversial history, with some even questioning the need for the term "ARDS." This controversy has been amplified by the coronavirus disease (COVID-19) pandemic given the marked increase in the incidence of ARDS, the relatively new treatment modalities that do not fit neatly with the Berlin definition, and the difficulty of making the diagnosis in resource-limited settings. We propose that attempts to revise the definition of ARDS should apply the framework originally developed by psychologists and social scientists and used by other medical disciplines to generate and assess definitions of clinical syndromes that do not have gold standards. This framework is structured around measures of reliability, feasibility, and validity. Future revisions of the definition of ARDS should contain the purpose, the methodology, and the framework for empirically testing any proposed definition. Attempts to revise critical illness syndromes' definitions usually hope to make them "better"; our recommendation is that future attempts use the same criteria used by other fields in defining what "better" means.

Protocol and statistical analysis plan for the identification and treatment of hypoxemic respiratory failure and acute respiratory distress syndrome with protection, paralysis, and proning: A type-1 hybrid stepped-wedge cluster randomised effectiveness-implementation study.

Objective: To describe a study protocol and statistical analysis plan (SAP) for the identification and treatment of hypoxemic respiratory failure (HRF) and acute respiratory distress syndrome (ARDS) with protection, paralysis, and proning (TheraPPP) study prior to completion of recruitment, electronic data retrieval, and analysis of any data. Design: TheraPPP is a stepped-wedge cluster randomised study evaluating a care pathway for HRF and ARDS patients. This is a type-1 hybrid effectiveness-implementation study design evaluating both intervention effectiveness and implementation; however primarily powered for the effectiveness outcome. Setting: Seventeen adult intensive care units (ICUs) across Alberta, Canada. Participants: We estimate a sample size of 18816 mechanically ventilated patients, with 11424 patients preimplementation and 7392 patients postimplementation. We estimate 2688 sustained ARDS patients within our study cohort. Intervention: An evidence-based, stakeholder-informed, multidisciplinary care pathway called Venting Wisely that standardises diagnosis and treatment of HRF and ARDS patients. Main outcome measures: The primary outcome is 28-day ventilator-free days (VFDs). The primary analysis will compare the mean 28-day VFDs preimplementation and postimplementation using a mixed-effects linear regression model. Prespecified subgroups include sex, age, HRF, ARDS, COVID-19, cardiac surgery, body mass index, height, illness acuity, and ICU volume. Results: This protocol and SAP are reported using the Standard Protocol Items: Recommendations for Interventional Trials guidance and the Guidelines for the Content of Statistical Analysis Plans in Clinical Trials. The study received ethics approval and was registered (ClinicalTrials.gov-NCT04744298) prior to patient enrolment. Conclusions: TheraPPP will evaluate the effectiveness and implementation of an HRF and ARDS care pathway.

Prevalence of Clostridioides difficile Infection in Critically Ill Patients with Extreme Leukocytosis and Diarrhea.

While early empiric antibiotic therapy is beneficial for patients presenting with sepsis, the presentation of sepsis from Clostridioides difficile (formerly Clostridium difficile) infection (CDI) has not been well studied in large cohorts. We sought to determine whether the combination of extreme leukocytosis and diarrhea was strongly predictive of CDI in a cohort of 8659 patients admitted to the intensive care unit. We found that CDI was present in 15.0% (95% CI, 12.1-18.3%) of patients with extreme leukocytosis and diarrhea and that mortality for those with CDI, diarrhea, and extreme leukocytosis was 33.8% (95% CI, 23.2-44.3%). These data support consideration of empiric treatment for CDI in unstable critically ill patients with extreme leukocytosis and diarrhea, along with treatment of other possible sources of sepsis as appropriate. Empiric treatment for CDI can usually be discontinued promptly, along with narrowing of other broad-spectrum antimicrobial coverage, if a sensitive C. difficile test is negative.

Outcome of acute hypoxaemic respiratory failure: insights from the LUNG SAFE Study.

BACKGROUND: Current incidence and outcome of patients with acute hypoxaemic respiratory failure requiring mechanical ventilation in the intensive care unit (ICU) are unknown, especially for patients not meeting criteria for acute respiratory distress syndrome (ARDS). METHODS: An international, multicentre, prospective cohort study of patients presenting with hypoxaemia early in the course of mechanical ventilation, conducted during four consecutive weeks in the winter of 2014 in 459 ICUs from 50 countries (LUNG SAFE). Patients were enrolled with arterial oxygen tension/inspiratory oxygen fraction ratio ≤300 mmHg, new pulmonary infiltrates and need for mechanical ventilation with a positive end-expiratory pressure of ≥5 cmH2O. ICU prevalence, causes of hypoxaemia, hospital survival and factors associated with hospital mortality were measured. Patients with unilateral versus bilateral opacities were compared. FINDINGS: 12 906 critically ill patients received mechanical ventilation and 34.9% with hypoxaemia and new infiltrates were enrolled, separated into ARDS (69.0%), unilateral infiltrate (22.7%) and congestive heart failure (CHF; 8.2%). The global hospital mortality was 38.6%. CHF patients had a mortality comparable to ARDS (44.1% versus 40.4%). Patients with unilateral-infiltrate had lower unadjusted mortality, but similar adjusted mortality compared to those with ARDS. The number of quadrants on chest imaging was associated with an increased risk of death. There was no difference in mortality comparing patients with unilateral-infiltrate and ARDS with only two quadrants involved. INTERPRETATION: More than one-third of patients receiving mechanical ventilation have hypoxaemia and new infiltrates with a hospital mortality of 38.6%. Survival is dependent on the degree of pulmonary involvement whether or not ARDS criteria are reached.

Clinical trials in critical care: can a Bayesian approach enhance clinical and scientific decision making?

Recent Bayesian reanalyses of prominent trials in critical illness have generated controversy by contradicting the initial conclusions based on conventional frequentist analyses. Many clinicians might be sceptical that Bayesian analysis, a philosophical and statistical approach that combines prior beliefs with data to generate probabilities, provides more useful information about clinical trials than the frequentist approach. In this Personal View, we introduce clinicians to the rationale, process, and interpretation of Bayesian analysis through a systematic review and reanalysis of interventional trials in critical illness. In the majority of cases, Bayesian and frequentist analyses agreed. In the remainder, Bayesian analysis identified interventions where benefit was probable despite the absence of statistical significance, where interpretation depended substantially on choice of prior distribution, and where benefit was improbable despite statistical significance. Bayesian analysis in critical care medicine can help to distinguish harm from uncertainty and establish the probability of clinically important benefit for clinicians, policy makers, and patients.