Prescribing patterns of hydrocortisone in septic shock: a single-center experience of how surviving sepsis guidelines are interpreted and translated into bedside practice.

Objectives: The Surviving Sepsis Campaign suggests giving hydrocortisone to septic patients only if their "blood pressure is poorly responsive to fluid resuscitation and vasopressor therapy." Because the definition of "poorly responsive" is not provided, the purpose of this study was to identify prescribing triggers for hydrocortisone in septic shock.

Design: Retrospective chart review of patients with septic shock over 17 months, who received hydrocortisone, followed by a survey of all intensivists who attended in the study ICUs to determine whether provider attitudes matched clinical practice.

Setting: Eight ICUs in an academic hospital and a hybrid academic/community hospital.

Patients: A total of 155 patients with septic shock in whom vasopressors were initiated and hydrocortisone was prescribed.

Measurements And Main Results: Ninety-nine patients (64%) were already receiving two vasopressors before hydrocortisone was prescribed. An additional 22 patients were on a single high-dose vasopressor prior to corticosteroid initiation. Of patients who survived to have their hydrocortisone dose changed, 57% had their corticosteroids tapered, whereas 43% were abruptly discontinued. Seventy-six percent of patients were no longer on vasopressors when the first dosing change was made. Twenty-seven out of 36 intensivists (75%) completed the survey. The majority (72%) defined "poorly responsive to vasopressors" as the presence of two vasopressors, and 70% stated that they required patients to be off vasopressors prior to altering the corticosteroid dose.

Conclusions: Significant variability exists when corticosteroids are prescribed for septic shock, with the most common interpretation in our institution of "poorly responsive to fluid resuscitation and vasopressor therapy" being the presence of two vasopressors. The method and timing of corticosteroid discontinuation also differed among providers. Self-described prescribing patterns from intensivists closely matched their actual behavior, suggesting variability is due to differing interpretations of the guidelines themselves, rather than a deficit in knowledge translation.