Acute Hypoxemic Respiratory Failure, COVID 19
Conditions
Keywords
Acute Hypoxemic Respiratory Failure, COVID-19, High-Flow Nasal Cannula, Predictive Model, Machine Learning, Metabolomics
Brief summary
The aim of this prospective observational cohort study is to develop a predictive model for high-flow nasal cannula (HFNC) failure in patients with acute hypoxemic respiratory failure (AHRF) caused by SARS-CoV-2. By combining clinical data and metabolomic profiles, the study seeks to answer: \- Can integrating clinical parameters with metabolomic markers enhance the prediction of HFNC failure in AHRF patients? Participants with AHRF due to SARS-CoV-2 receiving HFNC under standard-of-care protocols will be monitored until hospital discharge or death. Blood samples for metabolomic analysis will be collected at inclusion.
Detailed description
This prospective observational cohort study focuses on developing and validating a predictive model for HFNC failure in patients with SARS-CoV-2-related AHRF, combining clinical parameters and metabolomic profiling. Patients received standard-of-care therapy based on hospital protocols, with clinical data and plasma samples collected at inclusion (T0) for laboratory and metabolomic analyses. The study consists of two cohorts: a Training Cohort (March 2020 to April 2021) for model development and a Validation Cohort (May 2021 to October 2021) for model validation and external testing. HFNC outcomes were classified as success or failure, with failure defined as requiring orotracheal intubation (OTI). Advanced machine-learning techniques were employed for feature selection and model optimization, targeting precise integration of clinical and metabolomic data. This model seeks to enhance early identification of HFNC failure risk, enabling timely interventions and personalized management in critically ill AHRF patients.
Interventions
None listed
Sponsors
Fundacio La Marató de TV3
Instituto de Salud Carlos III
European Union
Hospital del Mar Research Institute (IMIM)
Study design
Observational model
COHORT
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Adult patients (≥18 years old) with acute hypoxemic respiratory failure (AHRF) confirmed as secondary to SARS-CoV-2 infection via PCR testing. * Requirement for high-flow nasal cannula (HFNC) initiated within 48 hours of hospital admission.
Exclusion criteria
* Patients under the age of 18 * Patients with active do not resuscitate (DNR) orders at the time of inclusion.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Predictive Model Development | From the initiation of HFNC until the need for intubation, hospital discharge, or death, whichever occurs first (up to 28 days). Completed using the Training Cohort (March 2020 to April 2021). | Area Under the Receiver Operating Characteristic Curve (AUC) of the final predictive model for identifying HFNC failure, measured at the time of study inclusion (T0). |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Identification of Key Metabolites | At study inclusion (T0). | Concentrations of key metabolomic markers (e.g., anthranilic acid, octanoic acid) associated with HFNC failure, measured at T0. |
| Validation of the Predictive Model | From the initiation of HFNC until the need for intubation, hospital discharge, or death, whichever occurs first (up to 28 days). Completed during the Validation Cohort phase (May to October 2021). | Performance of the predictive model in the Validation Cohort. Area Under the Receiver Operating Characteristic Curve (AUC) of the final predictive model for identifying HFNC failure, measured at the time of study inclusion (T0). |
Countries
Spain
Outcome results
None listed