Infection, Bacterial, Sepsis, Severe Community-acquired Pneumonia (sCAP)
Conditions
Keywords
Community-acquired Pneumonia, biomarker, clinical data warehouse (CDWH), machine learning, data-driven algorithms, multi-dimensional modelling, sepsis-related mortality, Streptococcus pneumoniae
Brief summary
This multi-center observational case-control study in Intensive Care Unit (ICU) patients is to identify novel biomarkers allowing to recognize severe community acquired pneumonia (sCAP) -associated sepsis at an earlier stage and predict sepsis-related mortality. Patients with sCAP (cases) will be profoundly characterized over time regarding the development of sepsis and compared with control patients. The mechanisms and influencing factors on the clinical course will be explored with most modern -omics technologies allowing a detailed characterisation. These data will be analysed using machine learning algorithms and multi-dimensional mathematical models.
Interventions
compare data patterns by data-driven algorithms including machine learning and multi-dimensional modelling to reliably determine sepsis
compare data patterns by data-driven algorithms including machine learning and multi-dimensional modelling to to predict sepsis-related mortality
Sponsors
Swiss Personalized Health Network (SPHN)
Personalized Health and Related Technologies (PHRT) initiative of ETH Zürich
University Hospital, Basel, Switzerland
Study design
Observational model
CASE_CONTROL
Time perspective
PROSPECTIVE
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Admission to the ICU of one of the participating centers. * Cases: severe community acquired pneumonia with requirement for ICU admission. * Controls: Clinical phenotype of inflammation not due to suspected sepsis In addition, control patients will be patients with fever \>38°C, CRP \>100mg/L, no infection focus expected in ≥ 24h. * All required sample types can most likely be collected within the first 24h visits. * Expected ICU stay of more than 24h.
Exclusion criteria
* Admission to the hospital within the prior 14 days. * Patients with psychosis * Evidence of a hospital acquired pneumonia. * One of the following respiratory conditions: Acute exacerbation of chronic obstructive pulmonary disease (COPD) or bronchiectasis, acute severe asthma, aspiration pneumonia, tuberculosis, clinical suspected viral pneumonia without bacterial infection, cardiogenic pulmonary oedema. * Patients with an acute respiratory distress Syndrome (ARDS). * Patient which can be managed as outpatients and do not require an ICU. * Patient where a transmission to another institution is likely within the next 24h. * Documented rejection of the general consent or participation to research in general. * Patients with a palliative situation and a life expectancy due to other diseases (e.g. progressed cancer) less than 28 days.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Detection of sepsis | within 7 days after study inclusion | Sepsis detection based on new discovered digital biomarkers will be compared to classical sepsis-3 criteria (with an increase of the sequential organ failure assessment (SOFA) score of 2 or larger score points). |
| Sepsis related mortality | within 7 days after study inclusion | Prediction of sepsis related mortality (with \>80% sensitivity and specificity at least 24h prior to event) |
| Time to sepsis detection (minutes after Intensive Care Unit (ICU) admission) | within 7 days after study inclusion | Time to sepsis detection (minutes after ICU admission) based on machine learning |
Countries
Switzerland
Outcome results
None listed