Heart Failure
Conditions
Keywords
Machine Learning, Multi-Omics Approaches, Outcomes Research, Patient-Centered Information
Brief summary
The aim of this study is to develop and validate an improved, comprehensive risk assessment algorithm integrating blood RNA-based biomarkers, clinical, and patient-centered data and to assess the incremental predictive value (discrimination and reclassification) compared to a traditional risk model (change in the c-statistics for prediction of the primary endpoint).
Detailed description
The ORACLE study is a multicenter, observational, prospective, cross-sectional and longitudinal study integrated by 3 different cohorts for 1) RNA biomarker discovery (60 nested case-control sample), 2) model derivation (516 nested case-control sample) and 3) external validation (new prospective cohort of 558 consecutive patients recruited in 4 hospital centers) according to a TRIPOD Statement type 3 analysis. In total 1134 consecutive patients with a HF hospitalization or urgent HF visit \< 30 days before inclusion and followed for 6 months will be studied. This study include the discovery of novel RNA-based biomarkers using next-generation sequencing technology to define and validate a new biomarker set and clinical and patient-centered risk determinants definition. A new model will be constructed; training and internal validation in the derivation cohort using machine learning methods, and finally an external validation of the new next generation integrative risk assessment model will be performed.
Interventions
DIAGNOSTIC_TESTComprehensive risk assessment algorithm
Risk assessment algorithm including clinical and patient-centered information and a novel RNA-based biomarkers using next-generation sequencing technology.
Sponsors
Hospital Universitari de Bellvitge
Study design
Observational model
OTHER
Time perspective
CROSS_SECTIONAL
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Age ≥ 18 years old. * Patients with a recent (\<30 days) acute decompensation of HF requiring intravenous diuretic therapy (either hospitalized or in ambulatory care) or intensification of oral diuretics (ambulatory care). * HF diagnosis according to European Society of Cardiology (ESC) criteria. * Written informed consent * Patients receiving oral standard medication for chronic HF.
Exclusion criteria
* Age\<18 years old. * Death before hospital discharge. * The patient is unable or unwilling to give the informed consent to participate. * Unstable patients with signs of fluid overload or low cardiac output at the moment of enrollment.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| To compare the model performance of the comprehensive risk assessment algorithm with a traditional risk model to predict HF related hospitalizations or all-cause death | 6 months | To compare in an external validation cohort of patients a recent acute HF event, the performance (discrimination, additive predictive value, and reclassification ability) of a comprehensive risk assessment algorithm with the performance of a traditional risk model to predict the occurrence of the primary composite clinical end-point of HF related hospitalizations (readmissions) or all-cause death at 180 days after hospital discharge or after an urgent HF visit (acute HF event requiring intravenous administration of diuretics without admission). |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| To compare the model performance of the comprehensive risk assessment algorithm with a traditional risk model to predict the occurrence of HF-related hospitalizations (readmissions) or all-cause death at 30, and 90 days after hospital discharge. | 6 months | The admission will be considered as HF-related if it suits with one of the following conditions: * Cardiovascular cause hospitalization * Complication directly related with prior HF admission * Decompensation of prior disease or comorbidity |
| To compare the model performance of the comprehensive risk assessment algorithm with a traditional risk model to predict the occurrence of HF-related hospitalizations (readmissions) at 30, 90, and 180 days after hospital discharge. | 6 months | — |
| To compare the model performance of the comprehensive risk assessment algorithm with a traditional risk model to predict the occurrence of all-cause death at 30, 90, and 180 days, after hospital discharge. | 6 months | — |
Other
| Measure | Time frame |
|---|---|
| Explore the performance (sensibility and specificity) of the developed model according gender, LVEF (preserved or reduced left ventricle ejection fraction), comorbidity burden (Charlson Comorbidity Index) and frailty. | 6 months |
Countries
Spain
Contacts
Primary ContactJosep Comín Colet, MD, PhD
Backup ContactRaúl Ramos Polo, MD
Outcome results
None listed