Introduction: Sepsis is a life-threatening syndrome that remains a major cause of morbidity and mortality in the intensive care unit (ICU). The 2016 Surviving Sepsis Campaign guidelines recommend tapering corticosteroids once vasopressors are no longer needed, but evidence on proper daily dosing and duration is limited. This study evaluates corticosteroid tapering strategies in septic shock patients. Methods: A retrospective analysis, approved by the UNC Human Research Review Committee, was conducted on critically ill patients with septic shock who received vasopressors and hydrocortisone. Patients were divided into two groups based on the duration of their steroid taper: rapid taper ( 180 mg/dL), hypernatremia (sodium > 145 mEq/L), and upper gastrointestinal (UGI) bleed during admission. Adults over 18 who received vasopressors and hydrocortisone for septic shock were included. Exclusions were pregnancy, prior steroid use, history of adrenal insufficiency, use of corticosteroids other than hydrocortisone, non-standard hydrocortisone regimens, and inability to wean steroids during intensive care unit admission. Results: A total of 312 patients were analyzed: 222 received a rapid steroid taper (< 72 hours) and 90 received a prolonged taper (≥72 hours). Vasopressor reinitiation within 72 hours occurred in 48% of the rapid taper group and 66% of the prolonged group (p=0.006). Within 24 hours, reinitiation was 42% in the rapid group versus 56% in the prolonged group (p=0.039). Mortality trended lower in the rapid group (32% vs. 41%, p=0.185). Mean ICU LOS was 215 hours for the rapid group and 418 for the prolonged group (p< 0.001). Mean hospital LOS was 447 hours for the rapid group and 734 for the prolonged group (p< 0.001). Hypernatremia was less frequent in the rapid group (29% vs. 47%, p=0.004). Hyperglycemia was similar between groups (24% vs. 21%, p=0.327). Conclusions: Rapid corticosteroid tapering in septic shock patients reduces vasopressor reinitiation, shortens ICU and hospital stays, and decreases hypernatremia.
Denninger et al. (Sun,) studied this question.