Search

Summary: Background: While many patients seem to recover from SARS-CoV-2 infections, many patients report experiencing SARS-CoV-2 symptoms for weeks or months after their acute COVID-19 ends, even developing new symptoms weeks after infection. These long-term effects are called post-acute sequelae of SARS-CoV-2 (PASC) or, more commonly, Long COVID. The overall prevalence of Long COVID is currently unknown, and tools are needed to help identify patients at risk for developing long COVID. Methods: A working group of the Rapid Acceleration of Diagnostics-radical (RADx-rad) program, comprised of individuals from various NIH institutes and centers, in collaboration with REsearching COVID to Enhance Recovery (RECOVER) developed and organized the Long COVID Computational Challenge (L3C), a community challenge aimed at incentivizing the broader scientific community to develop interpretable and accurate methods for identifying patients at risk of developing Long COVID. From August 2022 to December 2022, participants developed Long COVID risk prediction algorithms using the National COVID Cohort Collaborative (N3C) data enclave, a harmonized data repository from over 75 healthcare institutions from across the United States (U.S.). Findings: Over the course of the challenge, 74 teams designed and built 35 Long COVID prediction models using the N3C data enclave. The top 10 teams all scored above a 0.80 Area Under the Receiver Operator Curve (AUROC) with the highest scoring model achieving a mean AUROC of 0.895. Included in the top submission was a visualization dashboard that built timelines for each patient, updating the risk of a patient developing Long COVID in response to clinical events. Interpretation: As a result of L3C, federal reviewers identified multiple machine learning models that can be used to identify patients at risk for developing Long COVID. Many of the teams used approaches in their submissions which can be applied to future clinical prediction questions. Funding: Research reported in this RADx® Rad publication was supported by the National Institutes of Health. Timothy Bergquist, Johanna Loomba, and Emily Pfaff were supported by Axle Subcontract: NCATS-STSS-P00438.

BackgroundPostacute sequelae of COVID-19 (PASC), also known as long COVID, is a broad grouping of a range of long-term symptoms following acute COVID-19. These symptoms can occur across a range of biological systems, leading to challenges in determining risk factors for PASC and the causal etiology of this disorder. An understanding of characteristics that are predictive of future PASC is valuable, as this can inform the identification of high-risk individuals and future preventative efforts. However, current knowledge regarding PASC risk factors is limited. ObjectiveUsing a sample of 55,257 patients (at a ratio of 1 patient with PASC to 4 matched controls) from the National COVID Cohort Collaborative, as part of the National Institutes of Health Long COVID Computational Challenge, we sought to predict individual risk of PASC diagnosis from a curated set of clinically informed covariates. The National COVID Cohort Collaborative includes electronic health records for more than 22 million patients from 84 sites across the United States. MethodsWe predicted individual PASC status, given covariate information, using Super Learner (an ensemble machine learning algorithm also known as stacking) to learn the optimal combination of gradient boosting and random forest algorithms to maximize the area under the receiver operator curve. We evaluated variable importance (Shapley values) based on 3 levels: individual features, temporal windows, and clinical domains. We externally validated these findings using a holdout set of randomly selected study sites. ResultsWe were able to predict individual PASC diagnoses accurately (area under the curve 0.874). The individual features of the length of observation period, number of health care interactions during acute COVID-19, and viral lower respiratory infection were the most predictive of subsequent PASC diagnosis. Temporally, we found that baseline characteristics were the most predictive of future PASC diagnosis, compared with characteristics immediately before, during, or after acute COVID-19. We found that the clinical domains of health care use, demographics or anthropometry, and respiratory factors were the most predictive of PASC diagnosis. ConclusionsThe methods outlined here provide an open-source, applied example of using Super Learner to predict PASC status using electronic health record data, which can be replicated across a variety of settings. Across individual predictors and clinical domains, we consistently found that factors related to health care use were the strongest predictors of PASC diagnosis. This indicates that any observational studies using PASC diagnosis as a primary outcome must rigorously account for heterogeneous health care use. Our temporal findings support the hypothesis that clinicians may be able to accurately assess the risk of PASC in patients before acute COVID-19 diagnosis, which could improve early interventions and preventive care. Our findings also highlight the importance of respiratory characteristics in PASC risk assessment. International Registered Report Identifier (IRRID)RR2-10.1101/2023.07.27.23293272

OBJECTIVES:. Elevated intracranial pressure (ICP) is a potentially devastating complication of neurologic injury. Developing an ICP prediction algorithm to help the clinician adjust treatments and potentially prevent elevated ICP episodes. DESIGN:. Retrospective study. SETTING:. Three hundred thirty-five ICUs at 208 hospitals in the United States. SUBJECTS:. Adults patients from the electronic ICU (eICU) Collaborative Research Database was used to train an ensemble machine learning model to predict the ICP 30 minutes in the future. Predictive performance was evaluated using a left-out test dataset and externally evaluated on the Medical Information Mart for Intensive Care-III (MIMIC-III) Matched Waveform Database. INTERVENTIONS:. None. MEASUREMENTS AND MAIN RESULTS:. Predictors included age, assigned sex, laboratories, medications and infusions, input/output, Glasgow Coma Scale (GCS) components, and time-series vitals (heart rate, ICP, mean arterial pressure, respiratory rate, and temperature). Each patient ICU stay was divided into successive 95-minute timeblocks. For each timeblock, the model was trained on nontime-varying covariates as well as on 12 observations of time-varying covariates at 5-minute intervals and asked to predict the 5-minute median ICP 30 minutes after the last observed ICP value. Data from 931 patients with ICP monitoring in the eICU dataset were extracted (46,207 timeblocks). The root mean squared error was 4.51 mm Hg in the eICU test set and 3.56 mm Hg in the MIMIC-III dataset. The most important variables driving ICP prediction were previous ICP history, patients’ temperature, weight, serum creatinine, age, GCS, and hemodynamic parameters. CONCLUSIONS:. IntraCranial pressure prediction AlgoRithm using machinE learning, an ensemble machine learning model, trained to predict the ICP of a patient 30 minutes in the future based on baseline characteristics and vitals data from the past hour showed promising predictive performance including in an external validation dataset.

OBJECTIVES:. To develop proof-of-concept algorithms using alternative approaches to capture provider sentiment in ICU notes. DESIGN:. Retrospective observational cohort study. SETTING:. The Multiparameter Intelligent Monitoring of Intensive Care III (MIMIC-III) and the University of California, San Francisco (UCSF) deidentified notes databases. PATIENTS:. Adult (≥18 yr old) patients admitted to the ICU. MEASUREMENTS AND MAIN RESULTS:. We developed two sentiment models: 1) a keywords-based approach using a consensus-based clinical sentiment lexicon comprised of 72 positive and 103 negative phrases, including negations and 2) a Decoding-enhanced Bidirectional Encoder Representations from Transformers with disentangled attention-v3-based deep learning model (keywords-independent) trained on clinical sentiment labels. We applied the models to 198,944 notes across 52,997 ICU admissions in the MIMIC-III database. Analyses were replicated on an external sample of patients admitted to a UCSF ICU from 2018 to 2019. We also labeled sentiment in 1,493 note fragments and compared the predictive accuracy of our tools to three popular sentiment classifiers. Clinical sentiment terms were found in 99% of patient visits across 88% of notes. Our two sentiment tools were substantially more predictive (Spearman correlations of 0.62–0.84, p values < 0.00001) of labeled sentiment compared with general language algorithms (0.28–0.46). CONCLUSION:. Our exploratory healthcare-specific sentiment models can more accurately detect positivity and negativity in clinical notes compared with general sentiment tools not designed for clinical usage.

Introduction There is little consensus and high heterogeneity on the optimal set of relevant clinical outcomes in research studies regarding extubation in neurocritical care patients with brain injury undergoing mechanical ventilation. The aims of this study are to: (1) develop a core outcome set (COS) and (2) reach consensus on a hierarchical composite endpoint for such studies.Methods and analysis The study will include a broadly representative, international panel of stakeholders with research and clinical expertise in this field and will involve four stages: (1) a scoping review to generate an initial list of outcomes represented in the literature, (2) an investigator meeting to review the outcomes for inclusion in the Delphi surveys, (3) four rounds of online Delphi consensus-building surveys and (4) online consensus meetings to finalise the COS and hierarchical composite endpoint.Ethics and dissemination This study received ethical approval from the French Society of Anesthesia and Critical Care Medicine Institutional Review Board (SFAR CERAR-IRB 00010254‐2023-029). The study results will be disseminated through communication to stakeholders, publication in a peer-reviewed journal, and presentations at conferences.Trial registration number This study is registered with the Core Outcome Measures in Effectiveness Trials (COMET) Initiative.

Abstract Machine learning (ML) and artificial intelligence (AI) algorithms have the potential to derive insights from clinical data and improve patient outcomes. However, these highly complex systems are sensitive to changes in the environment and liable to performance decay. Even after their successful integration into clinical practice, ML/AI algorithms should be continuously monitored and updated to ensure their long-term safety and effectiveness. To bring AI into maturity in clinical care, we advocate for the creation of hospital units responsible for quality assurance and improvement of these algorithms, which we refer to as “AI-QI” units. We discuss how tools that have long been used in hospital quality assurance and quality improvement can be adapted to monitor static ML algorithms. On the other hand, procedures for continual model updating are still nascent. We highlight key considerations when choosing between existing methods and opportunities for methodological innovation.

Objective To investigate risk factors and subphenotypes associated with long term symptoms and outcomes after hospital admission for covid-19.Design Prospective, multicentre observational study.Setting 93 hospitals in France.Participants Data from 2187 adults admitted to hospital with covid-19 in France between 1 February 2020 and 30 June 2021.Main outcome measures Primary endpoint was the total number of persistent symptoms at six months after hospital admission that were not present before admission. Outcomes examined at six months were persistent symptoms, Hospital Anxiety and Depression Scale, six minute walk test distances, 36-Item Short Form Health Survey scores, and ability to resume previous professional activities and self-care. Secondary endpoints included vital status at six months, and results of standardised quality-of-life scores. Additionally, an unsupervised consensus clustering algorithm was used to identify subphenotypes based on the severity of hospital course received by patients.Results 1109 (50.7%) of 2187 participants had at least one persistent symptom. Factors associated with an increased number of persistent symptoms were in-hospital supplemental oxygen (odds ratio 1.12, 95% confidence interval 1 to 1.24), no intensive care unit admission (1.15, 1.01 to 1.32), female sex (1.33, 1.22 to 1.45), gastrointestinal haemorrhage (1.51, 1.02 to 2.23), a thromboembolic event (1.66, 1.17 to 2.34), and congestive heart failure (1.76, 1.27 to 2.43). Three subphenotypes were identified: including patients with the least severe hospital course (based on ventilatory support requirements). Although Hospital Anxiety and Depression Scale scores were within normal values for all groups, patients of intermediate severity and more comorbidities had a higher median Hospital Anxiety and Depression Scale score than did the other subphenotypes. Patients in the subphenotype with most severe hospital course had worse short form-36 scores and were less able to resume their professional activity or care for themselves as before compared with other subphenotypes.Conclusions Persistent symptoms after hospital admission were frequent, regardless of acute covid-19 severity. However, patients in more severe subphenotypes had a significantly worse functional status and were less likely to resume their professional activity or able to take care of themselves as before.Trial registration NCT04262921.

Acute kidney injury (AKI) is a major postoperative complication that lacks established intraoperative predictors. Our objective was to develop a prediction model using preoperative and high-frequency intraoperative data for postoperative AKI. In this retrospective cohort study, we evaluated 77,428 operative cases at a single academic center between 2016 and 2022. A total of 11,212 cases with serum creatinine (sCr) data were included in the analysis. Then, 8519 cases were randomly assigned to the training set and the remainder to the validation set. Fourteen preoperative and twenty intraoperative variables were evaluated using elastic net followed by hierarchical group least absolute shrinkage and selection operator (LASSO) regression. The training set was 56% male and had a median [IQR] age of 62 (51–72) and a 6% AKI rate. Retained model variables were preoperative sCr values, the number of minutes meeting cutoffs for urine output, heart rate, perfusion index intraoperatively, and the total estimated blood loss. The area under the receiver operator characteristic curve was 0.81 (95% CI, 0.77–0.85). At a score threshold of 0.767, specificity was 77% and sensitivity was 74%. A web application that calculates the model score is available online. Our findings demonstrate the utility of intraoperative time series data for prediction problems, including a new potential use of the perfusion index. Further research is needed to evaluate the model in clinical settings.

Background: Several continuous monitoring solutions, including wireless wearable sensors, are available or being developed to improve patient surveillance on surgical wards. We designed a survey to understand the current perception and expectations of anaesthesiologists who, as perioperative physicians, are increasingly involved in postoperative care. Methods: The survey was shared in 40 university hospitals from Western Europe and the USA. Results: From 5744 anaesthesiologists who received the survey link, there were 1158 valid questionnaires available for analysis. Current postoperative surveillance was mainly based on intermittent spot-checks of vital signs every 4–6 h in the USA (72%) and every 8–12 h in Europe (53%). A majority of respondents (91%) considered that continuous monitoring of vital signs should be available on surgical wards and that wireless sensors are preferable to tethered systems (86%). Most respondents indicated that oxygen saturation (93%), heart rate (80%), and blood pressure (71%) should be continuously monitored with wrist devices (71%) or skin adhesive patches (54%). They believed it may help detect clinical deterioration earlier (90%), decrease rescue interventions (59%), and decrease hospital mortality (54%). Opinions diverged regarding the impact on nurse workload (increase 46%, decrease 39%), and most respondents considered that the biggest implementation challenges are economic (79%) and connectivity issues (64%). Conclusion: Continuous monitoring of vital signs with wireless sensors is wanted by most anaesthesiologists from university hospitals in Western Europe and in the USA. They believe it may improve patient safety and outcome, but may also be challenging to implement because of cost and connectivity issues.

Abstract Background The use of hydroxocobalamin has long been advocated for treating suspected cyanide poisoning after smoke inhalation. Intravenous hydroxocobalamin has however been shown to cause oxalate nephropathy in a single-center study. The impact of hydroxocobalamin on the risk of acute kidney injury (AKI) and survival after smoke inhalation in a multicenter setting remains unexplored. Methods We conducted a multicenter retrospective study in 21 intensive care units (ICUs) in France. We included patients admitted to an ICU for smoke inhalation between January 2011 and December 2017. We excluded patients discharged at home alive within 24 h of admission. We assessed the risk of AKI (primary endpoint), severe AKI, major adverse kidney (MAKE) events, and survival (secondary endpoints) after administration of hydroxocobalamin using logistic regression models. Results Among 854 patients screened, 739 patients were included. Three hundred six and 386 (55.2%) patients received hydroxocobalamin. Mortality in ICU was 32.9% (n = 243). Two hundred eighty-eight (39%) patients developed AKI, including 186 (25.2%) who developed severe AKI during the first week. Patients who received hydroxocobalamin were more severe and had higher mortality (38.1% vs 27.2%, p = 0.0022). The adjusted odds ratio (95% confidence interval) of AKI after intravenous hydroxocobalamin was 1.597 (1.055, 2.419) and 1.772 (1.137, 2.762) for severe AKI; intravenous hydroxocobalamin was not associated with survival or MAKE with an adjusted odds ratio (95% confidence interval) of 1.114 (0.691, 1.797) and 0.784 (0.456, 1.349) respectively. Conclusion Hydroxocobalamin was associated with an increased risk of AKI and severe AKI but was not associated with survival after smoke inhalation. Trial registration ClinicalTrials.gov, NCT03558646

Abstract Background This study aimed at describing usual anesthetic practices for brain-dead donors (BDD) during an organ procurement (OP) procedure and to assess the knowledge and self-confidence of French anesthesiologists with this practice. Methods An electronic and anonymous survey with closed-questions about anesthetic management of BDD was distributed to French anesthesiologists via the mailing list of the French Society of Anesthesiology and Intensive Care Medicine. Results Four hundred fifty-eight responses were analyzed. Respondents were mainly attending physicians with more than 10 years of clinical experience. 78% of them declared being cognizant of guidelines regarding management of BDD. Advanced hemodynamic monitoring and endocrine substitution were rarely considered by respondents (31 and 35% of respondents, respectively). 98% of the respondents used crystalloids for fluid resuscitation. During the procedure, use of neuromuscular blockers, opioids and sedative agents were considered by respectively 84, 61 and 27% of the respondents. A very high level of agreement (10 [8–10], on a ten-points Likert-style scale) was reported concerning the expected impact of intraoperative anesthetic management on the primary function of grafts. Conclusions Declared anesthetic practice appeared in accordance with guidelines concerning organ donor management in the ICU. Further studies are needed to evaluate the specific impact of intraoperative management during this procedure and thus the need for specific anesthetic guidelines.

Introduction The benefits and risks of low-dose hydrocortisone in patients with septic shock have been investigated in numerous randomised controlled trials and trial-level meta-analyses. Yet, the routine use of this treatment remains controversial. To overcome the limitations of previous meta-analyses inherent to the use of aggregate data, we will perform an individual patient data meta-analysis (IPDMA) on the effect of hydrocortisone with or without fludrocortisone compared with placebo or usual care on 90-day mortality and other outcomes in patients with septic shock.Methods and analysis To assess the benefits and risks of hydrocortisone, with or without fludrocortisone for adults with septic shock, we will search major electronic databases from inception to September 2020 (Cochrane Central Register of Controlled Trials, MEDLINE, EMBASE and Latin American Caribbean Health Sciences Literature), complimented by a search for unpublished trials. The primary analysis will compare hydrocortisone with or without fludrocortisone to placebo or no treatment in adult patients with septic shock. Secondary analyses will compare hydrocortisone to placebo (or usual care), hydrocortisone plus fludrocortisone to placebo (or usual care), and hydrocortisone versus hydrocortisone plus fludrocortisone. The primary outcome will be all cause mortality at 90 days. We will conduct both one-stage IPDMA using mixed-effect models and machine learning with targeted maximum likelihood analyses. We will assess the risk of bias related to unshared data and related to the quality of individual trial.Ethics and dissemination This IPDMA will use existing data from completed randomised clinical trials and will comply with the ethical and regulatory requirements regarding data sharing for each of the component trials. The findings of this study will be submitted for publication in a peer-review journal with straightforward policy for open access.PROSPERO registration number CRD42017062198.

Abstract Background Haemorrhagic shock is the leading cause of early preventable death in severe trauma. Delayed treatment is a recognized prognostic factor that can be prevented by efficient organization of care. This study aimed to develop and validate Red Flag, a binary alert identifying blunt trauma patients with high risk of severe haemorrhage (SH), to be used by the pre-hospital trauma team in order to trigger an adequate intra-hospital standardized haemorrhage control response: massive transfusion protocol and/or immediate haemostatic procedures. Methods A multicentre retrospective study of prospectively collected data from a trauma registry (Traumabase®) was performed. SH was defined as: packed red blood cell (RBC) transfusion in the trauma room, or transfusion ≥ 4 RBC in the first 6 h, or lactate ≥ 5 mmol/L, or immediate haemostatic surgery, or interventional radiology and/or death of haemorrhagic shock. Pre-hospital characteristics were selected using a multiple logistic regression model in a derivation cohort to develop a Red Flag binary alert whose performances were confirmed in a validation cohort. Results Among the 3675 patients of the derivation cohort, 672 (18%) had SH. The final prediction model included five pre-hospital variables: Shock Index ≥ 1, mean arterial blood pressure ≤ 70 mmHg, point of care haemoglobin ≤ 13 g/dl, unstable pelvis and pre-hospital intubation. The Red Flag alert was triggered by the presence of any combination of at least two criteria. Its predictive performances were sensitivity 75% (72–79%), specificity 79% (77–80%) and area under the receiver operating characteristic curve 0.83 (0.81–0.84) in the derivation cohort, and were not significantly different in the independent validation cohort of 2999 patients. Conclusion The Red Flag alert developed and validated in this study has high performance to accurately predict or exclude SH.

Abstract Background Intracranial hypertension (ICH) is a major cause of death after traumatic brain injury (TBI). Continuous hyperosmolar therapy (CHT) has been proposed for the treatment of ICH, but its effectiveness is controversial. We compared the mortality and outcomes in patients with TBI with ICH treated or not with CHT. Methods We included patients with TBI (Glasgow Coma Scale ≤ 12 and trauma-associated lesion on brain computed tomography (CT) scan) from the databases of the prospective multicentre trials Corti-TC, BI-VILI and ATLANREA. CHT consisted of an intravenous infusion of NaCl 20% for 24 hours or more. The primary outcome was the risk of survival at day 90, adjusted for predefined covariates and baseline differences, allowing us to reduce the bias resulting from confounding factors in observational studies. A systematic review was conducted including studies published from 1966 to December 2016. Results Among the 1086 included patients, 545 (51.7%) developed ICH (143 treated and 402 not treated with CHT). In patients with ICH, the relative risk of survival at day 90 with CHT was 1.43 (95% CI, 0.99–2.06, p = 0.05). The adjusted hazard ratio for survival was 1.74 (95% CI, 1.36–2.23, p

Abstract Background Intravascular haemolysis has been associated with acute kidney injury (AKI) in different clinical settings (cardiac surgery, sickle cell disease). Haemolysis occurs frequently in critically ill burn patients. The aim of this study was to assess the predictive value of haptoglobin at admission to predict major adverse kidney events (MAKE) and AKI in critically ill burn patients. Methods We conducted a retrospective, single-centre cohort study in a burn critical care unit in a tertiary centre, including all consecutive severely burned patients (total burned body surface > 20% and/or shock and/or mechanical ventilation at admission) from January 2012 to April 2017 with a plasmatic haptoglobin dosage at admission. Results A total of 130 patients were included in the analysis. Their mean age was 49 (34–62) years, their median total body surface area burned was 29% (15–51%) and the intensive care unit (ICU) mortality was 25%. Early haemolysis was defined as an undetectable plasmatic haptoglobin at admission. We used logistic regression to identify MAKE and AKI risk factors. In multivariate analysis, undetectable haptoglobin was associated with MAKE and AKI (respectively, OR 6.33, 95% CI 2.34–16.45, p

Introduction: The Glasgow Coma Scale (GCS) score is the primary method of assessing consciousness after traumatic brain injury (TBI), and the clinical standard for classifying TBI severity. There is scant literature discerning the influence of circadian rhythms or emergency department (ED) arrival hour on this important clinical tool. Methods: Retrospective cohort analysis of adult patients suffering blunt TBI using the National Sample Program of the National Trauma Data Bank, years 2003–2006. ED arrival GCS score was characterized by midday (10 a.m.–4 p.m.) and midnight (12 a.m.–6 a.m.) cohorts (N=24548). Proportions and standard errors are reported for descriptive data. Multivariable regressions using odds ratios (OR), mean differences (B), and their associated 95% confidence intervals [CI] were performed to assess associations between ED arrival hour and GCS score. Statistical significance was assessed at p