Early Prediction for Sepsis Induced Acute Lung Injury
Conditions
Keywords
radiomics, sepsis, acute respiratory distress syndrome, deep learning, prediction model
Brief summary
Introduction: Sepsis-induced acute respiratory distress syndrome(SI-ARDS) is a common complication of severe sepsis and is an independent contributor to poor prognosis of patients. It remains a clinical challenge to identify the SI-ARDS early and accurately, which could optimize the treatment strategy and reduce the mortality risk. Radiomics high-dimensional features extracted from CT images offer an insight into microvascular damage of SI-ARDS that are imperceptible to human eyes and aspects of intra-alveolar heterogeneity with potential prognostic relevance. Methods: Study design Investigators screened all patients with sepsis and septic shock who are treated in Sun Yat-sen Memorial Hospital, Sun Yat-sen University during the period from 1 May 2015 and 30 May 2022. Patients were recruited retrospectively from May 2015 to April 2021 as discovering group, and prospectively during the period from May 2021 to May 2022 as validation group. Follow-up will conducted until April 2023. Cohort descriptions and definitions Investigators plan to recruit 160 patients in discovering group, 40 patients in internal validation group, and 100 patients in external validation group. Patients between 18 and 80 years of age with sepsis and septic shock will be screened for eligibility. SI-ARDS is defined by sequential occurrence of the sepsis-3 consensus criteria for sepsis and the Berlin Definition for ARDS. The exclusion criteria are： 1. admission stay \<24hours, 2. the presence of end-stage lung disease or long-term oxygen therapy, 3. critically ill patients who have started mechanical ventilation caused by SI-ARDS before admission, 4. a history of lung transplantation and chronic obstructive pulmonary disease, 5. cancer patients not/have received chemotherapy. Outcome measures In this study, the primary outcome measure was the occurrence rates of acute respiratory distress syndrome(ARDS). It refers to the occurrence of sepsis patients progressed into ARDS. Secondary outcome measures were as follows: 1.28-day mortality 2.ventilator-free days 3.respiratory failure-free days Data collection All clinical data were collected by investigators and trained personnel. Each participant's data will be filled in electronic case report forms (CRF) and store online using REDCap (Research Electronic Data Capture). Discussion: SI-ARDS is one common severe complication with critically ill sepsis patients, which causes high mortality and poor prognosis. Early ARDS patient(arterial oxygen tension/inspired oxygen fraction \[PaO2/FIO2\] ≤ 300 mmHg but \> 200 mmHg) may not require invasive mechanical ventilation, and is more readily reversible than acute respiratory distress syndrome(ARDS). In this ambispecive cohort study, investigators developed and validated novel nomograms incorporating the radiomics signature and clinical signature to provide an easy-to-use and individualized prediction of SI-ARDS occurrence and severe degree in patients with early stage.
Interventions
DIAGNOSTIC_TESTradiomic of chest CT
radiomics as a quantitative imaging method for early detection, risk assessment, and treatment decisions for early predicting sepsis induced acute respiratory distress syndrome.
Sponsors
Sun Yat-Sen Memorial Hospital of Sun Yat-Sen University
Study design
Observational model
COHORT
Time perspective
OTHER
Eligibility
Sex/Gender
ALL
Age
18 Years to 90 Years
Healthy volunteers
No
Inclusion criteria
* Clinical diagnosis of sepsis * Clinical diagnosis of ARDS * Written informed consent in external validation cohort
Exclusion criteria
* admission stay \<24hours, * the presence of end-stage lung disease or long-term oxygen therapy, * critically ill patients who have started mechanical ventilation caused by SA-ARDS before admission, * a history of lung transplantation and chronic obstructive pulmonary disease, * cancer patients not/have received chemotherapy
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| occurrence rates of acute respiratory distress syndrome(ARDS) | from date of admission until the date of disease progressing to ARDS, assessed up to 28days | It refers to the occurrence of sepsis patients progressed into ARDS in all patients enrolled into the study |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| 28-day mortality | from date of admission until the date of death from any cause, assessed up to 28days | a cox proportion hazards regression model was used to analysis the survival outcome of patients with sepsis within 28 days according to radiomics characteristics |
| ventilator-free days | from date of admission until the date of ventilation or discharge from hospital, assessed up to 28days | defined as the number of calendar days after initiating ventilator unassisted breathing to day 28 after admission |
| respiratory failure-free days | from date of admission until the initiating date of respiratory failure, assessed up to 28days | defined as the day without evidence of non-respiratory organ failure |
Countries
China
Contacts
Primary ContactLI LI
Outcome results
None listed