Early Intravenous Hydrocortisone in Sepsis

ENROLLMENT AND RANDOMIZATION Once patients meet the eligibility criteria, the treating physicians will notify one of the study investigators to invite the patients or the legal representatives to participate in the study. Written informed consent will be obtained before randomization. Each patient will be randomly assigned in a 1:1 ratio by the sequential enrollment numbers to receive early intravenous low-dose hydrocortisone (the EH group) or standard care. Randomization is performed using a computer-generated sequence in the permuted block of 4 by the investigator who is not involving patient enrollment and assessment. The other investigators, the patients or their representatives, and the treating physicians are all blinded to the group assignment. Patients may withdraw from the study for any reason at any time. The investigators or the treating physician may also withdraw the patients from the study for safety reasons at any time.

STUDY PROTOCOL AND INTERVENTIONS Blood will be drawn from the patients in both groups at the time of randomization for baseline interleukin-6, interleukin-10, tumor necrosis factor, procalcitonin, C-reactive protein, and cortisol levels. All patients will receive standard treatment and investigation for septic shock including standard laboratory analysis, antibiotics, infection source control, intravenous crystalloids, vasopressors, and organ support as directed by the treating physicians according to the Surviving Sepsis Campaign: International Guidelines for Management of Sepsis and Septic Shock 2021.

The study drug (low-dose hydrocortisone versus placebo) will be prepared by a pharmacist who has no role in the study, according to the sequential enrollment numbers. The study drugs will be packaged in identical containers labeled with sequential enrollment numbers. After randomization, the study drugs will be administered as soon as possible by the nurses who have no role in the study. For the early hydrocortisone (EH) group, 50 mg of hydrocortisone in 10 ml of normal saline will be given as an intravenous bolus, then 200 mg of hydrocortisone in 100 ml of normal saline will be given as continuous intravenous infusion in 24 hours for 2 consecutive days (total 250 mg in day 1 and 200 mg in day 2). For the standard care group, a bolus of 10 ml of normal saline will be given, then 100 ml of normal saline will be given as continuous intravenous infusion in 24 hours for 2 consecutive days as a sham control. After completion of the study drugs for 2 days, the study drug will be discontinued without tapering.

An open-label 50 mg of hydrocortisone given as an intravenous bolus followed by intravenous hydrocortisone 200 mg/day given as continuous infusion or divided bolus administration is suggested to be commenced in both study arms if the hemodynamic goal of the patient is not reached despite the dose of norepinephrine or epinephrine ≥ 0.25 mcg/kg/min at least 4 hours after the initiation of the vasopressors as recommended by the guideline. The study drug in each arm will be discontinued once an open-label hydrocortisone is initiated. Capillary or venous blood glucose will be tested at least every 6 hours for 2 days then at least once daily or more as appropriate for the first 7 days as a part of the study protocol.

The group allocation will be unblinded per request of the treating physicians or the principal investigators if the patient develops one of the following conditions: hyperglycemic emergencies, hypoglycemia, or active gastrointestinal hemorrhage within 7 days after randomization, undifferentiated hypotension or suspected adrenal insufficiency within 72 hours after the cessation of the study drug, and conditions related to hydrocortisone that the treating physicians concern to be harmful or potentially harmful to the patients.

The study protocol will be terminated if hyperglycemic emergencies, or active gastrointestinal hemorrhage develop during receiving the study drug. The investigators or the treating physician may also stop the study drug for safety reasons at any time.

STATISTICAL ANALYSIS Data will be analyzed according to an intention-to-treat principle for the primary analysis. The EH group will be compared against the usual care group for the main analysis of the primary outcomes and secondary outcomes. Continuous outcomes will be compared using an independent t-test or the Mann-Whitney U test as appropriate. The result will be reported as an absolute difference with a corresponding 95% confidence interval (CI). Chi-square or Fisher exact test will be employed as appropriate for comparing binary outcomes. Multivariate regression analyses will also be performed for the primary and secondary outcomes adjusted for potential confounders. Potential factors will be tested and selected as confounders for the model when the p-value is less than 0.2 in the univariate analysis. The result will be reported as adjusted OR (aOR) with 95% CI. For all statistical analyses, a p-value of less than 0.05 is considered statistically significant unless defined otherwise.

Prespecified subgroup analyses will be conducted based on the hypothesis that patients with higher inflammatory biomarkers would benefit more from early hydrocortisone. The median value of each inflammatory biomarker and cortisol level will be used as cut-off values for the exposures (highly-elevated versus low inflammatory biomarker groups). Subgroup analyses will be performed using multivariable logistic regression models adjusted for confounders with interaction terms. The result will be reported as adjusted OR (aOR) with 95% CI for each inflammatory biomarker.

A single interim analysis for risk monitoring and safety will be performed after a total recruitment of 115 patients (50% of the total sample size). A significance level of 0.001 is selected as a stopping rule for harm, based on Haybittle-Peto method. The trial will be stopped early if there is a significantly higher rate of 28-day mortality at the significance level of ≤ 0.001.