An observational study in Sepsis and Critical Illness, sponsored by Shanghai Zhongshan Hospital. Completed at 1 site in China. Per ClinicalTrials.gov, last updated 2026-11-03.
Sponsored by Shanghai Zhongshan Hospital · Observational
This retrospective multicohort observational study evaluates whether the PaO2/FiO2 ratio (P/F ratio) organizes heterogeneity in the association between early systemic corticosteroid exposure and mortality among critically ill patients with sepsis.
The study uses one internal critical-care cohort (ICCA) and two external critical-care databases (MIMIC and eICU). Analyses include internal discovery, broad external testing, reconstruction of external cohorts toward vasopressor-anchored shock-decision contexts, realized-exposure observational analyses, and strategy-level analyses using a simplified clone-censor-weighting framework (Lite-CCW).
The primary objective is to characterize whether corticosteroid-associated outcome patterns differ across P/F strata and whether similar directional patterns are observed across independent ICU datasets after decision-context reconstruction. The study is hypothesis-generating and does not assign corticosteroid treatment or define a clinical treatment-selection rule.
2,416 studies on the registry are indexed under Sepsis; 541 are open to participants now.
This study's enrollment of 2,693 is above the median of 150 across 1,162 observational studies indexed under Sepsis.
Browse Sepsis studies →Shanghai Zhongshan Hospital is the lead sponsor of 652 studies on the registry; 291 are open to participants now.
Counted across the registry records on this site, refreshed daily.
The study population consists of critically ill patients with sepsis identified from three ICU databases: the internal ICCA cohort and the external MIMIC and eICU cohorts. The ICCA cohort included ICU stays meeting the finalized sepsis phenotype based on organ dysfunction and infection-related evidence. MIMIC and eICU cohorts were used for broad external testing and subsequently reconstructed toward vasopressor-anchored shock-decision contexts. All data were obtained retrospectively from routinely collected clinical records or existing critical-care databases, and treatment was not assigned by the investigators.
Go through the criteria one by one — yes, no or not sure — and take the list to your doctor or the study team. Kept in this browser only.
Exclusion Criteria:
Analysis-specific exclusions:
Underlined terms in the criteria open these too. Explained by clinicaltrials.gg, not part of the registry record. The study team decides who can take part.
Anti-inflammatory medicines such as prednisone, prednisolone and dexamethasone — not the anabolic steroids used for muscle.
Studies differ in what they allow: check which steroid medicines, routes (tablets, injections, creams, inhalers, eye drops) and doses this study accepts.
Corticosteroids (steroids)
Anti-inflammatory medicines such as prednisone, prednisolone and dexamethasone — not the anabolic steroids used for muscle.
Studies differ in what they allow: check which steroid medicines, routes (tablets, injections, creams, inhalers, eye drops) and doses this study accepts.
Explained by clinicaltrials.gg, not part of the registry record. The study team decides who can take part.
Critically ill patients meeting the finalized ICCA sepsis phenotype. Realized systemic corticosteroid exposure was assessed within 0-72 hours after cohort time zero. The finalized manuscript-facing ICCA cohort included 2,693 ICU stays.
Drug: Systemic Corticosteroids
Patients represented in the broad MIMIC ICU cohort, with reconstructed main and shock-aligned subsets used for external decision-context analyses. Realized corticosteroid exposure was assessed within 0-24 hours after time zero.
Drug: Systemic Corticosteroids
Patients represented in the broad eICU critical-care cohort, with reconstructed main and shock-aligned subsets used for external decision-context analyses. Realized corticosteroid exposure was assessed within 0-24 hours after time zero.
Drug: Systemic Corticosteroids
Systemic corticosteroid exposure occurring as part of routine clinical care. Treatment was not assigned by the investigators. For realized-exposure observational analyses, corticosteroid exposure was defined within 0-72 hours after time zero in the ICCA cohort and within 0-24 hours in the MIMIC and eICU cohorts. Strategy-level analyses evaluated corticosteroid initiation within 24 hours versus no initiation during the corresponding 24-hour strategy window.
ICU Mortality
Death during the index ICU stay, based on the ICU mortality variable recorded in each database.
Time frame: From ICU admission (cohort entry) until ICU discharge or death, whichever occurred first, during the index ICU stay. The maximum observed follow-up duration was 90 days.
ICU Mortality by PaO2/FiO2 (P/F) Stratum
ICU mortality evaluated separately in patients with baseline PaO₂/FiO₂ ≤200 mmHg and \>200 mmHg.
Time frame: From ICU admission (cohort entry) until ICU discharge or death, whichever occurred first, during the index ICU stay. The maximum observed follow-up duration was 90 days.
Strategy-Level Association Between Early Corticosteroid Initiation and ICU Mortality
Comparison of ICU mortality between an early corticosteroid initiation strategy (within 24 hours after cohort entry) and a no-early-initiation strategy during the same 24-hour window, using the limited clone-censor-weighting framework.
Time frame: From cohort entry (ICU admission) until ICU discharge or death, whichever occurred first, during the index ICU stay. The treatment-strategy assignment window was the first 24 hours after cohort entry. The maximum observed follow-up duration was 90 days.
P/F-Stratified Associations Between Corticosteroid Exposure and ICU Mortality Across Cohorts
Adjusted odds ratios (ORs) with 95% confidence intervals (CIs) for the association between systemic corticosteroid exposure and ICU mortality were compared across ICCA, MIMIC, and eICU cohorts within P/F ≤200 mmHg and P/F \>200 mmHg strata. Exposure windows followed source-specific definitions (0-72 hours in ICCA and within 24 hours in MIMIC and eICU). Broad and reconstructed external cohorts were evaluated separately. These exploratory comparisons did not constitute formal tests of cross-cohort replication, transportability, or causal effects.
Time frame: From ICU admission or cohort entry (time zero) until ICU discharge or death, whichever occurred first, during the index ICU stay.
Ran from Aug 2026 to primary completion in Sep 2026, 1 month.
How these figures are worked out, and which studies are left out of them.
From the registry record's own update history. This site started tracking changes on Sep 25, 2026; for anything earlier, see the record history on ClinicalTrials.gov ↗
This study is completed, as verified in Nov 2026. You cannot join it, but the record below documents what was studied.
Get an email when the registry record changes — status, dates, results — or when someone posts here.
Sign in to followQuestions and observations about this study, from anyone following it. Not medical advice, and not a channel to the study team — their contact details are on the registry record.
Sign in to join the discussion. Reading takes no account; posting does. You choose a display name, and a pseudonym is the default.
Nothing here yet. If you are running this trial, taking part in it, or weighing whether to, this is the place to say so.