Dataset Information

Predicting critical illness on initial diagnosis of COVID-19 based on easily obtained clinical variables: development and validation of the PRIORITY model.

ABSTRACT:

Objectives

We aimed to develop and validate a prediction model, based on clinical history and examination findings on initial diagnosis of coronavirus disease 2019 (COVID-19), to identify patients at risk of critical outcomes.

Methods

We used data from the SEMI-COVID-19 Registry, a cohort of consecutive patients hospitalized for COVID-19 from 132 centres in Spain (23rd March to 21st May 2020). For the development cohort, tertiary referral hospitals were selected, while the validation cohort included smaller hospitals. The primary outcome was a composite of in-hospital death, mechanical ventilation, or admission to intensive care unit. Clinical signs and symptoms, demographics, and medical history ascertained at presentation were screened using least absolute shrinkage and selection operator, and logistic regression was used to construct the predictive model.

Results

There were 10 433 patients, 7850 in the development cohort (primary outcome 25.1%, 1967/7850) and 2583 in the validation cohort (outcome 27.0%, 698/2583). The PRIORITY model included: age, dependency, cardiovascular disease, chronic kidney disease, dyspnoea, tachypnoea, confusion, systolic blood pressure, and SpO₂ ≤93% or oxygen requirement. The model showed high discrimination for critical illness in both the development (C-statistic 0.823; 95% confidence interval (CI) 0.813, 0.834) and validation (C-statistic 0.794; 95%CI 0.775, 0.813) cohorts. A freely available web-based calculator was developed based on this model (https://www.evidencio.com/models/show/2344).

Conclusions

The PRIORITY model, based on easily obtained clinical information, had good discrimination and generalizability for identifying COVID-19 patients at risk of critical outcomes.

SUBMITTER: Martinez-Lacalzada M

PROVIDER: S-EPMC8280376 | biostudies-literature | 2021 Dec

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Predicting critical illness on initial diagnosis of COVID-19 based on easily obtained clinical variables: development and validation of the PRIORITY model.

Martínez-Lacalzada Miguel M Viteri-Noël Adrián A Manzano Luis L Fabregate Martin M Rubio-Rivas Manuel M Luis García Sara S Arnalich-Fernández Francisco F Beato-Pérez José Luis JL Vargas-Núñez Juan Antonio JA Calvo-Manuel Elpidio E Espiño-Álvarez Alexia Constanza AC Freire-Castro Santiago J SJ Loureiro-Amigo Jose J Pesqueira Fontan Paula Maria PM Pina Adela A Álvarez Suárez Ana María AM Silva-Asiain Andrea A García-López Beatriz B Luque Del Pino Jairo J Sanz-Cánovas Jaime J Chazarra-Pérez Paloma P García-García Gema María GM Núñez-Cortés Jesús Millán JM Casas-Rojo José Manuel JM Gómez-Huelgas Ricardo R

Clinical microbiology and infection : the official publication of the European Society of Clinical Microbiology and Infectious Diseases 20210715 12

<h4>Objectives</h4>We aimed to develop and validate a prediction model, based on clinical history and examination findings on initial diagnosis of coronavirus disease 2019 (COVID-19), to identify patients at risk of critical outcomes.<h4>Methods</h4>We used data from the SEMI-COVID-19 Registry, a cohort of consecutive patients hospitalized for COVID-19 from 132 centres in Spain (23rd March to 21st May 2020). For the development cohort, tertiary referral hospitals were selected, while the validat ...[more]

PMID: 34274525

Similar Datasets

Project description:Annual seasonal influenza epidemics of variable severity result in significant morbidity and mortality in the United States (U.S.) and worldwide. In temperate climate countries, including the U.S., influenza activity peaks during the winter months. Annual influenza vaccination is recommended for all persons in the U.S. aged 6 months and older, and among those at increased risk for influenza-related complications in other parts of the world (e.g. young children, elderly). Observational studies have reported effectiveness of influenza vaccination to reduce the risks of severe disease requiring hospitalization, intensive care unit admission, and death. A diagnosis of influenza should be considered in critically ill patients admitted with complications such as exacerbation of underlying chronic comorbidities, community-acquired pneumonia, and respiratory failure during influenza season. Molecular tests are recommended for influenza testing of respiratory specimens in hospitalized patients. Antigen detection assays are not recommended in critically ill patients because of lower sensitivity; negative results of these tests should not be used to make clinical decisions, and respiratory specimens should be tested for influenza by molecular assays. Because critically ill patients with lower respiratory tract disease may have cleared influenza virus in the upper respiratory tract, but have prolonged influenza viral replication in the lower respiratory tract, an endotracheal aspirate (preferentially) or bronchoalveolar lavage fluid specimen (if collected for other diagnostic purposes) should be tested by molecular assay for detection of influenza viruses.Observational studies have reported that antiviral treatment of critically ill adult influenza patients with a neuraminidase inhibitor is associated with survival benefit. Since earlier initiation of antiviral treatment is associated with the greatest clinical benefit, standard-dose oseltamivir (75 mg twice daily in adults) for enteric administration is recommended as soon as possible as it is well absorbed in critically ill patients. Based upon observational data that suggest harms, adjunctive corticosteroid treatment is currently not recommended for children or adults hospitalized with influenza, including critically ill patients, unless clinically indicated for another reason, such as treatment of asthma or COPD exacerbation, or septic shock. A number of pharmaceutical agents are in development for treatment of severe influenza.

Project description:ObjectivesProlonged critical illness after congenital heart surgery disproportionately harms patients and the healthcare system, yet much remains unknown. We aimed to define prolonged critical illness, delineate between nonmodifiable and potentially preventable predictors of prolonged critical illness and prolonged critical illness mortality, and understand the interhospital variation in prolonged critical illness.DesignObservational analysis.SettingPediatric Cardiac Critical Care Consortium clinical registry.PatientsAll patients, stratified into neonates (≤28 d) and nonneonates (29 d to 18 yr), admitted to the pediatric cardiac ICU after congenital heart surgery at Pediatric Cardiac Critical Care Consortium hospitals.InterventionsNone.Measurements and main resultsThere were 2,419 neonates and 10,687 nonneonates from 22 hospitals. The prolonged critical illness cutoff (90th percentile length of stay) was greater than or equal to 35 and greater than or equal to 10 days for neonates and nonneonates, respectively. Cardiac ICU prolonged critical illness mortality was 24% in neonates and 8% in nonneonates (vs 5% and 0.4%, respectively, in nonprolonged critical illness patients). Multivariable logistic regression identified 10 neonatal and 19 nonneonatal prolonged critical illness predictors within strata and eight predictors of mortality. Only mechanical ventilation days and acute renal failure requiring renal replacement therapy predicted prolonged critical illness and prolonged critical illness mortality in both strata. Approximately 40% of the prolonged critical illness predictors were nonmodifiable (preoperative/patient and operative factors), whereas only one of eight prolonged critical illness mortality predictors was nonmodifiable. The remainders were potentially preventable (postoperative critical care delivery variables and complications). Case-mix-adjusted prolonged critical illness rates were compared across hospitals; six hospitals each had lower- and higher-than-expected prolonged critical illness frequency.ConclusionsAlthough many prolonged critical illness predictors are nonmodifiable, we identified several predictors to target for improvement. Furthermore, we observed that complications and prolonged critical care therapy drive prolonged critical illness mortality. Wide variation of prolonged critical illness frequency suggests that identifying practices at hospitals with lower-than-expected prolonged critical illness could lead to broader quality improvement initiatives.

Project description:IntroductionGuidelines recommend follow-up for patients after an intensive care unit (ICU) stay. Methods for identifying patients with psychological problems after intensive care would be of value, to optimize treatment and to improve adequate resource allocation in ICU follow-up of ICU survivors. The aim of the study was to develop a predictive screening instrument, for use at ICU discharge, to identify patients at risk for post-traumatic stress, anxiety or depression.MethodsTwenty-one potential risk factors for psychological problems - patient characteristics and ICU-related variables - were prospectively collected at ICU discharge. Two months after ICU discharge 252 ICU survivors received the questionnaires Post-Traumatic Stress Symptom scale -10 (PTSS-10) and Hospital Anxiety and Depression Scale (HADS) to estimate the degree of post-traumatic stress, anxiety and depression.ResultsOf the 150 responders, 46 patients (31%) had adverse psychological outcome, defined as PTSS-10 >35 and/or HADS subscales ?8. After analysis, six predictors were included in the screening instrument: major pre-existing disease, being a parent to children younger than 18 years of age, previous psychological problems, in-ICU agitation, being unemployed or on sick-leave at ICU admission and appearing depressed in the ICU. The total risk score was related to the probability for adverse psychological outcome in the individual patient. The predictive accuracy of the screening instrument, as assessed with area under the receiver operating characteristic curve, was 0.77. When categorizing patients in three risk probability groups - low (0 to 29%), moderate (30 to 59%) high risk (60 to 100%), the actual prevalence of adverse psychological outcome in respective groups was 12%, 50% and 63%.ConclusionThe screening instrument developed in this study may aid ICU clinicians in identifying patients at risk for adverse psychological outcome two months after critical illness. Prior to wider clinical use, external validation is needed.

Project description:BackgroundCirculating levels of soluble urokinase plasminogen activation receptor (suPAR) have been proposed as a prognostic biomarker in patients with critical illness and sepsis. However, the origin of suPAR in sepsis has remained obscure. We investigated the potential cellular sources of suPAR by analyzing membrane-bound urokinase plasminogen activator receptor (uPAR, CD87) and evaluated its clinical relevance in critically ill patients.MethodsWe studied 87 critically ill patients (44 with sepsis, 43 without sepsis) from the medical intensive care unit (ICU) in comparison to 48 standard care patients with infections and 27 healthy controls in a prospective single-center non-interventional cohort study. Cellular uPAR expression of different immune cell subsets (by flow cytometry from peripheral blood) and corresponding serum suPAR concentrations were determined upon ICU admission and at day 3. Furthermore, we analyzed the effects of serum from sepsis patients on the activation and uPAR cleavage of primary human neutrophils and macrophages in vitro.ResultsIn healthy controls, uPAR (CD87) expression was detected on nearly all blood neutrophils and monocytes, but only scarcely on lymphocytes. While uPAR expression on monocytes was maintained in ICU patients, only 58% of neutrophils from critically ill patients expressed uPAR, which was significantly lower than in healthy controls or standard care patients. Concomitantly, serum suPAR levels were significantly increased in ICU patients. We noted a clear inverse correlation between low neutrophilic uPAR and high serum suPAR in standard care and ICU patients, indicating that shedding of uPAR from activated neutrophils represents a main source of suPAR in systemic inflammation. Both low uPAR and high suPAR were closely associated with mortality in critically ill patients. Furthermore, serum from sepsis patients induced uPAR protein expression and subsequent receptor shedding on isolated primary neutrophils, but not on macrophages, in vitro.ConclusionsThe inverse correlation between low uPAR surface expression on neutrophils and high serum suPAR in critically ill patients supports that neutrophils are a main source of shed suPAR proteins in systemic inflammation. Furthermore, high suPAR levels and low neutrophilic uPAR expression predict mortality in ICU patients.

Project description:BackgroundTo develop machine learning classifiers at admission for predicting which patients with coronavirus disease 2019 (COVID-19) who will progress to critical illness.MethodsA total of 158 patients with laboratory-confirmed COVID-19 admitted to three designated hospitals between December 31, 2019 and March 31, 2020 were retrospectively collected. 27 clinical and laboratory variables of COVID-19 patients were collected from the medical records. A total of 201 quantitative CT features of COVID-19 pneumonia were extracted by using an artificial intelligence software. The critically ill cases were defined according to the COVID-19 guidelines. The least absolute shrinkage and selection operator (LASSO) logistic regression was used to select the predictors of critical illness from clinical and radiological features, respectively. Accordingly, we developed clinical and radiological models using the following machine learning classifiers, including naive bayes (NB), linear regression (LR), random forest (RF), extreme gradient boosting (XGBoost), adaptive boosting (AdaBoost), K-nearest neighbor (KNN), kernel support vector machine (k-SVM), and back propagation neural networks (BPNN). The combined model incorporating the selected clinical and radiological factors was also developed using the eight above-mentioned classifiers. The predictive efficiency of the models is validated using a 5-fold cross-validation method. The performance of the models was compared by the area under the receiver operating characteristic curve (AUC).ResultsThe mean age of all patients was 58.9±13.9 years and 89 (56.3%) were males. 35 (22.2%) patients deteriorated to critical illness. After LASSO analysis, four clinical features including lymphocyte percentage, lactic dehydrogenase, neutrophil count, and D-dimer and four quantitative CT features were selected. The XGBoost-based clinical model yielded the highest AUC of 0.960 [95% confidence interval (CI): 0.913-1.000)]. The XGBoost-based radiological model achieved an AUC of 0.890 (95% CI: 0.757-1.000). However, the predictive efficacy of XGBoost-based combined model was very close to that of the XGBoost-based clinical model, with an AUC of 0.955 (95% CI: 0.906-1.000).ConclusionsA XGBoost-based based clinical model on admission might be used as an effective tool to identify patients at high risk of critical illness.

Dataset Information

Predicting critical illness on initial diagnosis of COVID-19 based on easily obtained clinical variables: development and validation of the PRIORITY model.

Objectives

Methods

Results

Conclusions

Publications

Predicting critical illness on initial diagnosis of COVID-19 based on easily obtained clinical variables: development and validation of the PRIORITY model.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets