Sign In
Individual Sign In
Create an Account
Institutional Sign In
OpenAthens Shibboleth
Views 11,694
Citations 0
October 03, 2016

Evaluating Glucocorticoids for SepsisTime to Change Course

Author Affiliations
  • 1Clinical Research, Investigation, and Systems Modeling of Acute Illness (CRISMA) Center, Department of Critical Care Medicine, University of Pittsburgh, Pittsburgh, Pennsylvania
  • 2VA Pittsburgh Healthcare System, Pittsburgh, Pennsylvania
  • 3Division of Pulmonary and Critical Care Medicine, Massachusetts General Hospital, Boston
  • 4Harvard Medical School, Boston, Massachusetts

Copyright 2016 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.

JAMA. Published online October 3, 2016. doi:10.1001/jama.2016.13904

For decades, researchers have tested glucocorticoids such as hydrocortisone, methylprednisolone, prednisolone, and dexamethasone, in varying doses, for the related conditions of pneumonia, sepsis, septic shock, and acute respiratory distress syndrome.14 The overarching rationale has been that glucocorticoids could mitigate exaggerated and potentially deleterious aspects of the host’s inflammatory response. Corticosteroids have typically been prescribed to target the acute phase of these conditions, although they have also been used later to improve resolution of inflammation and organ dysfunction in instances such as nonresolving acute respiratory distress syndrome.

However, the results of these studies have been mixed. For example, a meta-analysis of more than 30 trials in septic shock suggested that glucocorticoids reduce the duration of septic shock and may reduce mortality, particularly in patients who receive prolonged therapy.4 Other studies in infected patients who may not have been in septic shock are contradictory. For example, a meta-analysis of trials in pneumonia,2 the most common cause of sepsis, reported significant heterogeneity in terms of both patient characteristics and type of glucocorticoids used and found an uncertain effect on mortality but consistent reduction in duration of mechanical ventilation, length of hospitalization, and time to clinical stability. Consequently, the efficacy of glucocorticoids remains unclear, treatment guidelines can be inconsistent, and clinicians continue to struggle with decisions regarding optimal approaches.

A key problem is that corticosteroids induce a vast array of biological effects, and the relative balance of these effects can be difficult to predict. For example, corticosteroids, in addition to broad anti-inflammatory properties, modulate blood pressure through multiple mechanisms. Therefore, the potential benefits of corticosteroids for a condition such as septic shock could be through mitigation of unwanted inflammation or resolution of hypotension. The problem is compounded by recent appreciation that both inflammation and immunosuppression occur early in the course of sepsis, suggesting that a blanket approach to suppress inflammation is inadequately tailored for restoration of immune system balance.5

Against this backdrop, in this issue of JAMA, Keh et al6 present findings of the Hydrocortisone for Prevention of Septic Shock (HYPRESS) randomized clinical trial. In contrast to prior trials that focused on community-acquired infections, approximately half of the patients in the HYPRESS trial had hospital-acquired sepsis, a group that may be more immunosuppressed compared with those with community-acquired sepsis, and all 380 patients had at least 1 organ dysfunction. Some had transient hypotension prior to enrollment, but none met criteria for septic shock, defined as hypotension or need for vasopressors for 4 hours or longer. Patients were randomly assigned either to receive a continuous infusion of 200 mg of hydrocortisone for 5 days followed by dose tapering until day 11 (n = 190) or to receive placebo (n = 190). The primary outcome was development of septic shock within 14 days.

Patients randomized to the hydrocortisone group did not have a lower risk of progression to septic shock within 14 days (36 of 170 patients [21.2%] in the hydrocortisone group and 39 of 170 patients [22.9%] in the placebo group developed septic shock; difference, −1.8%; 95% CI, −10.7% to 7.2%; P = .70). In addition, there were no significant differences between the hydrocortisone and placebo groups for the use of mechanical ventilation (53.2% vs 59.9%), mortality at 28 days (8.8% vs 8.2%) or up to 180 days (26.8% vs 22.2%), intensive care unit length of stay (median [interquartile range], 8 [5-15] vs 9 [6-17] days), or hospital length of stay (median [interquartile range], 26 [16-46] vs 25 [16-40] days). Outcomes were similar in various subgroup analyses, including those examining patients with critical illness–related corticosteroid insufficiency (CIRCI). Hyperglycemia was the most common adverse event, and glucocorticoids did not increase the risk of secondary infections.

There are some important limitations of this trial. First, a large number of patients (61%) with sepsis who either presented with septic shock or developed septic shock prior to study entry were excluded, thus limiting the generalizability of the findings to those with delayed septic shock development. Second, CIRCI status was determined only at selected sites and could be determined only among patients who remained free of septic shock long enough for the postcorticotropin cortisol value to be obtained and prior to receiving the randomized treatment, further limiting generalizability. However, baseline characteristics were similar for participants with and without CIRCI. A major strength of this well-conducted trial was that a comprehensive and relevant set of outcome and safety measures were assessed, thus strengthening the conclusion of no effect, with the exception of hyperglycemia.

Concrete recommendations regarding the use of corticosteroids in sepsis and related conditions remain difficult to make. Current evidence and guidelines7 support limiting the use of intravenous hydrocortisone for patients with septic shock to those instances when fluid resuscitation and vasopressor therapy are inadequate to restore hemodynamic stability. For patients who are not in septic shock, the HYPRESS trial suggests no role for glucocorticoids for prevention of shock, at least among patients similar to those in this trial (ie, with evidence of a systemic response to infection and organ dysfunction present for not more than 48 hours), and also supports the current recommendation not to perform a corticotropin stimulation test to determine CIRCI. The efficacy of glucocorticoids on mortality in patients not in septic shock remains uncertain, and only modest efficacy was observed on other outcomes, such as duration of mechanical ventilation and length of stay.2,4 Thus, unresolved questions remain regarding which patients without septic shock are likely to benefit or be harmed from glucocorticoids and which dose and duration of treatment, if any, should be used.

Several additional trials are under way, and their results should be available in the next few years ( identifier NCT01283009 for severe community-acquired pneumonia, n = 1450; identifier NCT01284452 for sepsis and acute respiratory distress syndrome, n = 197; and identifier NCT01448109 for septic shock, n = 3800). Similar to the heterogeneity observed in prior trials,24 these trials are enrolling patients with varying illness severity and use different types of glucocorticoids and durations of therapy. If these trials run to completion, they will more than double the aggregate size of every corticosteroid study for these indications conducted to date, which may help answer some long-standing questions about use of corticosteroids in critical illness. However, simply increasing the number of patients studied without changing the approach may not be enough to change the outcome of patients with septic shock.

Recently, a precision medicine approach showed enrichment for a glucocorticoid response among patients with community-acquired pneumonia and an exaggerated immune response as evidenced by high circulating C-reactive protein levels.8 Are there other biomarkers that can be used to identify patients who are likely to benefit from glucocorticoids in patients with an infection? The HYPRESS study and a prior trial9 suggest that CIRCI, as evidenced by low circulating cortisol levels after a corticotropin stimulation test, may not be the best enrichment approach. In another study of pediatric sepsis, repression of innate immune and corticosteroid signaling pathways was associated with a 4-fold increase in the odds of death if corticosteroids were used.10 Additional precision medicine approaches may involve evaluations of tissue glucocorticoid resistance. Glucocorticoid resistance may occur owing to alterations in glucocorticoid cellular signaling caused by differences in glucocorticoid receptor α expression or genetic variants within candidate genes that may play a role in glucocorticoid signaling.11

What is the way forward if candidate biomarkers for glucocorticoid response endotypes are identified? Findings from secondary analyses of existing randomized clinical trials have to be validated, and uncertainties regarding dosing and the right clinical subtype are unlikely to be completely resolved by secondary analyses. The next step could be to consider an adaptive trial design used in cancer.12 Different doses tested in different subtypes identified by clinical features and biomarkers, derived from a better understanding of the host response to infection and glucocorticoids, would be tested in parallel. Response-adaptive randomization and enrichment during the trial could help identify promising and potentially harmful dose and subtype combinations early, thus increasing trial efficiency and exposing fewer patients to inferior treatments.

Glucocorticoids are clearly immunomodulatory during sepsis, and they are inexpensive. Research has failed to demonstrate unequivocal efficacy for patients with sepsis, and larger trials should help address outstanding questions. However, if past is prologue, it may be time to change direction and consider alternative trial designs to identify glucocorticoid-responsive subtypes and enable greater precision in current intensive care unit practice.

Back to top
Article Information

Corresponding Author: Sachin Yende, MD, MS, VA Pittsburgh Healthcare System, University Drive C, Room 2A124, Pittsburgh, PA 15240 (

Published Online: October 3, 2016. doi:10.1001/jama.2016.13904

Conflict of Interest Disclosures: Both authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr Yende reported receiving grants from the National Institute of General Medical Sciences, National Institutes of Health and from Bristol-Myers Squibb. Dr Thompson reported receiving grants from the National Heart, Lung, and Blood Institute and receiving personal fees for consulting from Alexion, Asahi Kasei, Boehringer Ingelheim, GlaxoSmithKline, and Regeneron.

Disclaimer: The views expressed in this article are those of the authors and do not necessarily reflect the views or policies of the Department of Veterans Affairs or the US government.

Correction: This article was corrected online October 17, 2016, to fix an error in the text and an error in the reference list.

Hahn  EO, Houser  HB, Rammelkamp  CH  Jr, Denny  FW, Wannamaker  LW.  Effect of cortisone on acute streptococcal infections and poststreptococcal complications. J Clin Invest. 1951;30(3):274-281.
Chen  Y, Li  K, Pu  H, Wu  T.  Corticosteroids for pneumonia. Cochrane Database Syst Rev. 2011;3(3):CD007720.
Meduri  GU, Bridges  L, Shih  MC, Marik  PE, Siemieniuk  RA, Kocak  M.  Prolonged glucocorticoid treatment is associated with improved ARDS outcomes: analysis of individual patients’ data from four randomized trials and trial-level meta-analysis of the updated literature. Intensive Care Med. 2016;42(5):829-840.
Annane  D, Bellissant  E, Bollaert  PE, Briegel  J, Keh  D, Kupfer  Y.  Corticosteroids for treating sepsis. Cochrane Database Syst Rev. 2015;12(12):CD002243.
Angus  DC, van der Poll  T.  Severe sepsis and septic shock. N Engl J Med. 2013;369(9):840-851.
Keh  D, Trips  E, Marx  G,  et al; SepNet–Critical Care Trials Group.  Effect of hydrocortisone on development of shock among patients with severe sepsis: the HYPRESS randomized clinical trial. JAMA. 2016. doi:10.1001/jama.2016.14799
Dellinger  RP, Levy  MM, Rhodes  A,  et al; Surviving Sepsis Campaign Guidelines Committee Including the Pediatric Subgroup.  Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock, 2012. Intensive Care Med. 2013;39(2):165-228.
Torres  A, Sibila  O, Ferrer  M,  et al.  Effect of corticosteroids on treatment failure among hospitalized patients with severe community-acquired pneumonia and high inflammatory response: a randomized clinical trial. JAMA. 2015;313(7):677-686.
Sprung  CL, Annane  D, Keh  D,  et al; CORTICUS Study Group.  Hydrocortisone therapy for patients with septic shock. N Engl J Med. 2008;358(2):111-124.
Wong  HR, Cvijanovich  NZ, Anas  N,  et al.  Developing a clinically feasible personalized medicine approach to pediatric septic shock. Am J Respir Crit Care Med. 2015;191(3):309-315.
De Iudicibus  S, Franca  R, Martelossi  S, Ventura  A, Decorti  G.  Molecular mechanism of glucocorticoid resistance in inflammatory bowel disease. World J Gastroenterol. 2011;17(9):1095-1108.
Park  JW, Liu  MC, Yee  D,  et al; I-SPY 2 Investigators.  Adaptive randomization of neratinib in early breast cancer. N Engl J Med. 2016;375(1):11-22.