Secondary Logo

Journal Logo

News: Hydrocortisone Fails to Prevent Septic Shock

Sorelle, Ruth MPH

doi: 10.1097/01.EEM.0000512765.77635.c8

Hydrocortisone failed to reduce the risk of septic shock in severe sepsis patients in recent research, though its authors held out hope that a larger study could show a better result. Current guidelines recommend hydrocortisone for treating sepsis, but whether it prevents progression to septic shock has been controversial. The HYPRESS study, which its authors conceded was underpowered, found no benefit.

The Hydrocortisone for Prevention of Septic Shock (HYPRESS) study of 380 patients with severe sepsis treated in 34 intermediate and intensive care units in Germany showed that adjunctive hydrocortisone did not prevent progression to septic shock. (JAMA 2016;316[17]:1775.) The double-blind, randomized, controlled trial assigned one group of patients with severe sepsis to receive a continuous infusion of hydrocortisone for five days followed by a tapering dose until day 11. The second group received placebo during the same period. Neither the physicians nor the patients knew who received which treatment.

“It is simply not useful to treat them with hydrocortisone,” said Didier Keh, MD, an author of the report and a member of the faculty in anesthesiology and intensive care medicine at Charité-Universitätsmedizin in Berlin. “It might be useful for subgroups (of severe sepsis patients), perhaps with community-acquired pneumonia, but for general hospital patients, we would say no.”

Septic shock appeared in 36 of the 170 hydrocortisone patients (an intent-to-treat group) and in 39 of 170 patients in the placebo group. No significant difference in mortality was seen between the groups. Ninety-day mortality was essentially the same.

Dr. Keh said the difference between the percentage of placebo patients expected to develop septic shock (40%) and those who actually did (23%) occurred when researchers attempted to predict the percentage from earlier studies, he said. The study may have been underpowered, and Dr. Keh said it was possible that more patients could shift the answer in another direction.

He said new definitions of sepsis and septic shock that were promulgated at the consensus conference called Sepsis-3 were not part of this study, which was completed before the new definitions were published. “We had to use the old definition of sepsis for that reason,” he said.

Back to Top | Article Outline

No Reduction

Salim Rezaie, MD, assessing the study in the blog R.E.B.E.L. EM, noted that the study was unexpectedly underpowered, which made its clinical significance cloudy. ( “The use of hydrocortisone in adults with severe sepsis not in septic shock did not reduce the risk of septic shock within 14 days, but the results of this study should be interpreted with caution as this study was underpowered to detect this outcome,” wrote Dr. Rezaie, an emergency physician with the Greater San Antonio Emergency Physician Group and the creator and founder of R.E.B.E.L. EM.

Answering the question asked in the study is important, he said. Current treatment guidelines recommend use of hydrocortisone in patients with vasopressor-dependent refractory sepsis, but the use of hydrocortisone in patients with severe sepsis without shock remains controversial, Dr. Rezaie said.

“Anything we can do to stop progression on the spectrum from severe sepsis to septic shock means reducing the risk of mortality,” he said. The farther along the spectrum a patient travels, the more likely he is to die.

“What we have not known is that if we give steroids earlier, will it prevent people from going into septic shock,” Dr. Rezaie said. Studying ever-decreasing numbers of patients in niche groups may make such questions unanswerable, however.

“I don't think we got a good answer to the question” in this study, he said, adding that as things stand, he recommends that physicians stick to the basics of sepsis treatment: fluid resuscitation, appropriate empiric antibiotics, and identifying patients as early as possible.

“If you have done all these things or if you are getting to the patient late, then you can think about using steroids,” said Dr. Rezaie. “If you are starting to think about vasopressors, giving steroids won't hurt.” He said researchers can later think about doing a study similar to HYPRESS with larger numbers of patients.

This is a matter of how to treat the tree in a forest of sepsis, or how to treat a single patient at a particular stage of disease. “The individual trees are not as big as the entire forest,” he said.

Changing the definitions within the sepsis spectrum is less important to physicians and patients, he said. “Do the early stuff first — the ABCs,” he said. Dr. Rezaie said he sees no indication that an early dose of steroids is harmful, particularly if the physician is considering vasopressors. Survival of sepsis has improved over the past 15 years, he said. “It means we have done something right.”

Share this article on Twitter and Facebook.

Access the links in EMN by reading this on our website or in our free iPad app, both available at

Comments? Write to us at

Copyright © 2017 Wolters Kluwer Health, Inc. All rights reserved.