From Wiki Journal Club
Jump to: navigation, search
Keh D, et al. "Effect of hydrocortisone on development of shock among patients with severe sepsis". Journal of the American Medical Association. 2016. 316(17):1775-1785.
PubMedFull textPDF

Clinical Question

In patients with severe sepsis, does early hydrocortisone therapy reduce progression to septic shock compared to placebo?

Bottom Line

In patients with severe sepsis, hydrocortisone in the first 48 hours did not reduce progression to septic shock at 14 days compared to placebo. Hydrocortisone therapy also failed to result in improvement in time to septic shock or short- or intermediate-term mortality. Hydrocortisone was associated with a 10% absolute increase in hyperglycemia and an absolute 13% reduction in delirium.

Major Points

Multiple RCTs have investigated the potential role for steroid therapy in patients with septic shock. The Annane Trial in 2002 demonstrated a short-term mortality benefit with IV hydrocortisone and fludrocortisone among patients with evidence of adrenal insufficiency on ACTH stimulation testing. The more recent CORTICUS study investigated hydrocortisone in patients with and without adrenal insufficiency and found no benefit in either subgroup with suggestion of increased infection rates in patients receiving hydrocortisone. Since the patient population in the Annane Trial were more critically ill, a benefit to steroid therapy in sicker patients could not be excluded. In addition, in both trials shock was more rapidly reversed with the use of corticosteroids. As a result of these conflicting data with possible benefit, the 2012 Surviving Sepsis Campaign Guidelines provide a weak Grade 2C recommendation for the use of IV corticosteroids in septic shock refractory to IV fluids and pressor therapy.

More recently, in a smaller RCT in patients with community-acquired pneumonia, hydrocortisone significantly improved survival and prevented progression to septic shock.[1] Further data investigating the effect of corticosteroids on patients with severe infection with regard to delaying or preventing progression to septic shock were needed.

The 2016 Hydrocortisone for Prevention of Septic Shock (HYPRESS) trial randomized 380 patients with severe sepsis without shock to IV hydrocortisone or placebo and assessed short-term prevalence of septic shock as the primary outcome. At 14 days, hydrocortisone had no beneficial effect on the development of septic shock versus placebo. Furthermore, there was no difference with regard to 28- or 90-day mortality or time to development of shock. Interestingly, there was a 13% absolute reduction in delirium in the corticosteroid group. Generally, hydrocortisone was well-tolerated with no evidence of excess infections with hydrocortisone therapy although there was a 10% absolute increase in hyperglycemia in the corticosteroid group. Ultimately, HYPRESS provides further evidence that corticosteroid therapy should be reserved for patients with refractory septic shock and patients with severe sepsis without shock do not appear to benefit from corticosteroid therapy. These findings are somewhat in conflict with emerging data suggesting benefit of corticosteroid therapy in patients with severe pneumonia, but this discrepancy can likely be explained by inclusion of a broad range of infections in the HYPRESS trial (only 45% of patients had PNA).


As of November 2016, no guidelines have been published that reflect the results of this trial.


  • Multicenter, placebo-controlled, double-blind RCT
  • N=380
    • IV hydrocortisone (n=190)
    • Placebo (n=190)
  • Setting: 34 sites in Germany
  • Enrollment: January 13, 2009 to August 27, 2013
  • Duration of follow-up: 14 days (primary outcome)
  • Analysis: Intention-to-treat
  • Primary outcome: Prevalence of septic shock at 14 days


Inclusion Criteria

  • Evidence of infection
  • Evidence of systemic response to infection, defined as two or more of
    • WBC > 12 or < 4 or > 13% bands
    • Temp ≥ 38 or ≤ 36
    • HR ≥ 90
    • Tachypnea, hypocapnia, or mechanical ventilation
  • Evidence of organ dysfunction for not longer than 48 hours
  • Provided informed consent

Exclusion Criteria

  • Septic shock
  • Age <18
  • Known hypersensitivity to hydrocortisone or mannitol (placebo)
  • History of glucocorticoid medication with indication for continuing therapy or other indications for treatment with glucocorticoids

Baseline Characteristics

From the placebo arm

  • Demographics: Age 65, male 63%
  • Type of admission: Elective surgery 24%, emergency surgery 18%, elective nonsurgery 2%, emergency nonsurgery 56%
  • Sepsis severity: SOFA 2.4, APACHE II 6.0, SAPS II 9.9, SAPS 3 11.0
  • SIRS criteria: Fever 80%, tachycardia 93%, tachypnea 89%, leukocytosis/leukopenia 75.0%
  • Shock parameters: Refractory hypotension 54%, hyperlactemia 65%, mean lactate 7 mmol/L, supplemental O2 74%, ICU need 67%, APACHE II score 18.7, MEDS score 8.0, SOFA score 4.2
  • Organ dysfunction: CNS 27%, coagulation 15%, pulmonary 68%, renal 42%, microcirculatory 30%
  • Source of infection: Community 47%, nosocomial ICU 30%, nosocomial ward 23%
  • Focus of infection: Pneumonia 53%, other respiratory 7%, thoracic 1%, GI 5%, intra-abdominal 16%, primary bacteremia 1%, bones/soft tissue 10%, surgical 2%, urogenital 14%, catheter 2%


  • 1:1 randomization to IV hydrocortisone versus placebo
    • Randomization stratified by participating center and sex
  • Septic shock defined as sepsis-induced hypotension despite adequate volume status for longer than 4 hours
    • Hypotension defined as mean arterial pressure < 65mmHg, systolic arterial pressure < 90mmHg, or the use of vasopressors to keep mean arterial pressure ≥ 90mmHg
    • Adequate volume status defined as CVP of 8mmHg or greater (≥ 12mmHg in ventilated patients) and a ScvO2 > 70%
  • Hydrocortisone was administered as an IV bolus of 50mg followed by a 24-hour continuous infusion of 200mg for days 1-5, 100mg for days 6-7, 50mg for days 8-9, and 25mg for days 10-11


Comparisons are hydrocortisone versus placebo.

Primary Outcomes

Septic shock at 14 days
21.2% [95% CI 15.6-28.1] vs. 22.9% [95% CI 17.2-30.0] (P=0.70)

Secondary Outcomes

Mortality at 28 days
8.8% [95% CI 5.4-14.0] vs. 8.2% [95% CI 5.0-13.4] (P=0.86)
Mortality at 180 days
26.8% [95% CI 20.7-34.0] vs. 22.2% [95% CI 16.5-29.0] (P=0.32)
SOFA score at day 14
4.7 [95% CI 3.5-6.5] vs. 5.0 [95% CI 3.5-6.8] (P=0.69)
11.2% [95% CI 6.4-19.0] vs. 24.5% [95% CI 17.2-33.7] (P=0.01)

Subgroup Analysis

  • There was no significant difference regarding the primary and secondary end points between those with and without relative adrenal insufficiency who received hydrocortisone versus placebo

Adverse Events

Secondary infection
40 (21.5%) vs. 32 (16.9%) (P=0.26)
169 (90.9%) vs. 154 (81.5%) (P=0.009)


  • The trial was powered to detect an absolute difference of 15% in the development of septic shock at 14 days. As a result, a smaller benefit with hydrocortisone cannot be ruled out.
  • The observed rate of septic shock in the placebo group (23%) was lower than originally presumed for sample size calculation (40%), again reducing sensitivity for this outcome.
  • Patients were enrolled with a variety of infections. As a result, a benefit of corticosteroid therapy in reducing septic shock in specific populations (e.g., patients with pneumonia), cannot be ruled out.
  • Secondary analyses not corrected for multiple comparisons, making certain findings (e.g., benefit of steroids on delirium) possibly due to chance.


  • Study supported by the German Federal Ministry of Education and Research

Further Reading

  1. Confalonieri M et al. Hydrocortisone infusion for severe community-acquired pneumonia: a preliminary randomized study. Am. J. Respir. Crit. Care Med. 2005. 171:242-8.