Intensive Care Unit, Aberdeen Royal Infirmary, Aberdeen AB25 2ZN, UK
* Corresponding author. E-mail: david.noble{at}arh.grampian.scot.nhs.uk
Septic shock has a crude mortality rate of 45% and claims the lives of 90 000 people each year in the USA alone.1 An epidemiological study from France of over 100 000 intensive care unit (ICU) admissions indicates the incidence of septic shock before or following admission to ICU is rising and now affects almost 10% of this patient population.2 Given the scale and associated costs of this problem,3 4 it is not surprising that developing solutions has been a focus of researchers, clinicians, and the pharmaceutical industry. Despite many past disappointments particularly with antagonists of endogenous mediators,5 some recent approaches have shown promise in prevention or treatment of sepsis and septic shock. They include tight glycaemic control, early haemodynamic goal-directed therapy, infusion of activated protein C, and use of corticosteroids.6
Corticosteroids have a long history of use in intensive care for septic shock and were extensively used in high dose for a short duration until the mid-1980s when large multicentre trials showed they were of no benefit.7 8 This finding was confirmed in subsequent meta-analyses.9 10 Their use for septic shock was largely abandoned.
However, since then, small explanatory trials of physiological stress doses of corticosteroids have demonstrated a reduced need for vasopressor drugs to maintain cardiovascular homeostasis, a possible surrogate for improved clinical outcomes.1113 The benefits of steroids may result from suppression of over-exuberant and dysregulated immune responses, suppression of inflammatory responses through a variety of mechanisms, and up-regulation of adrenoreceptor function.1417 However, adverse sequelae are well described and effects of corticosteroids on development of nosocomial infection, reactivation of latent infection, hyperglycaemia, bone metabolism, and psychosis as well as intensive care associated paresis must also be considered.18 19 Thus, surrogate measures describing benefit may be misleading and adequately designed and powered clinical outcome studies are essential to assure that potential benefits outweigh known adverse effects.20
One moderately large outcome study has recently been reported.21 This prospective randomized, controlled multicentre trial conducted in France reported the effects of hydrocortisone and fludrocortisone for patients with septic shock requiring vasopressors and mechanical ventilation as part of their treatment. After data modelling, this study of 299 evaluable patients showed statistically significant improved survival in the large subset of 229 patients who did not adequately respond to a short corticotropin test. The absolute reduction in mortality was 11% and the adjusted number needed to treat value of these non-responders was impressive at only seven patients to save one life at 28 days. The survival advantages of corticosteroids for septic shock in this subgroup appear to exceed those of activated protein C (drotrecogin alfa) for severe sepsis.22 Notably, there was no benefit of corticosteroids in the complementary sub-group of 70 patients who responded to corticotropin. They actually suffered an increase in mortality of 8%, which was statistically insignificant in this relatively small patient subset.
Given the apparently large survival benefit, as well as the ease of administration and low costs of this treatment, which contrast with some other evidence-based approaches, these findings could assume great importance for the 77% of patients with insufficient adrenal reserve as defined by their response to corticotropin. Several recent reviews and commentaries have since highlighted these beneficial results.6 23 24 Indeed, the trial findings were well-known to the intensive care community in advance of publication through presentations at international meetings,24 25 and re-adoption of steroids for septic shock has been swift. Even before publication, 60% of UK intensive care units responding to a survey indicated their use of corticosteroids for treatment of septic shock.26
However, there are problems with this influential study, highlighted in subsequent correspondence, which suggest that caution may be prudent. First the data are not so robust that these findings remain significant with a simpler statistical approach looking at crude outcome data27 and reliance on sub-group analysis invariably raises concerns.28 Secondly, there is no obvious pharmacological rationale for the use of oral fludrocortisone given the intrinsic mineralocorticoid activity of hydrocortisone at the dosage used.29 Thirdly, the increased mortality in the small sub-group of corticotropin responsive patients, although statistically insignificant, is of concern or even worrisome to some observers.6 30 Of fundamental importance however, is the study's failure to exclude patients who had been given the anaesthetic agent etomidate before the corticotropin test. This drug is a potent inhibitor of adrenal steroidogenesis primarily through its inhibition of the enzyme 11ß-hydroxylase. It is more potent than metyrapone,31 and even single doses may affect adrenal function for more than 24 h.32 Sixty-eight out of 72 (94%) etomidate-treated patients were non-responders to the corticotropin test.33 The problem of etomidate was tacitly acknowledged by the investigators as they changed the exclusion criteria in the second half of the study to preclude its use within 6 h of enrolment. This was arguably too little too late to safeguard the integrity of what could have been a landmark trial. Use of etomidate might account for 30% or more of the patients in their non-responder group.
Notably, etomidate by infusion had been associated previously with a 44% increased mortality in intensive care patients.34 Importantly, only months earlier other investigators from the same ICU, which used etomidate infusions, had reported a 7-fold better survival rate in a variety of patients with inadequate adrenal reserve when treated with corticosteroids.35 They were unaware, at the time, of etomidate's effect on adrenal function. Thus, an alternative explanation of the French study's apparently favourable results might reasonably be that appropriate treatment of iatrogenic etomidate-induced adrenal suppression improves outcome in patients with septic shock. Extensive use of etomidate will also have artificially inflated the proportion of non-responders to corticotropin. This further reduces the applicability of the results for individuals, institutions, and countries where etomidate is used infrequently or not at all.36 At best, the published results of this multicentre trial of corticosteroid supplementation are compromised by inclusion of patients given a known and potent adrenal suppressant.
Fortunately, there is a pan-European clinical trial, CORTICUS, of corticosteroids in septic shock, which is currently recruiting patients.37 This initiative deserves full support as results are urgently needed to inform whether the considerable promise of stress doses of corticosteroids, an inexpensive and widely available therapy, will ultimately be fulfilled in terms of meaningful outcomes for patients suffering from septic shock. In the absence of new robust and uncontaminated data, we remain reliant on surrogate measures from very small studies to guide practice. In terms of clinically relevant and important patient-orientated outcomes, key workers agree that equipoise remains.37 Despite renewed enthusiasm, presently it must be concluded there is insufficient scientifically secure evidence to support corticosteroids as a standard of care for patients with septic shock.
References
1 Wenzel RP. Treating sepsis. N Engl J Med 2002; 347: 9667
2 Annane D, Aegerter P, Jans-Guincestre MC, Guidet B. Current epidemiology of septic shock. The CUB-Réa Network. Am J Respir Crit Care Med 2003; 168: 16572
3 Edbrooke DL, Hibbert CL, Kingsley JM, Smith S, Bright NM, Quinn M. The patient-related costs of care for sepsis patients in a United Kingdom adult general intensive care unit. Crit Care Med 1999; 27: 17607[ISI][Medline]
4 Letarte J, Longo CJ, Pelletier J, Nabonne B, Fisher HN. Patient characteristics and costs of severe sepsis and septic shock in Quebec. J Crit Care 2002; 17: 3949[CrossRef][ISI][Medline]
5 Dinarello CA, Abraham E. Does blocking cytokines in sepsis work? Am J Respir Crit Care Med 2002; 166: 11567
6 Hotchkiss RS, Karl IE. The pathophysiology and treatment of sepsis. N Engl J Med 2003; 348: 13850
7 The Veterans Administration Systemic Sepsis Cooperative Study Group. Effect of high-dose glucocorticosteroid therapy on mortality in patients with clinical signs of sepsis. N Engl J Med 1987; 317: 65965[Abstract]
8 Bone RC, Fisher CJ, Clemmer TP, et al. A controlled clinical trial of high-dose methylprednisolone in the treatment of severe sepsis and septic shock. N Engl J Med 1987; 317: 6538[Abstract]
9 Lefering R, Neugebauer EAM. Steroid controversy in sepsis and septic shock: a meta-analysis. Crit Care Med 1995; 23: 1294303[ISI][Medline]
10 Cronin L, Cook DJ, Carlet J, et al. Corticosteroids treatment for sepsis: a critical appraisal and meta-analysis of the literature. Crit Care Med 1995; 23: 14309[ISI][Medline]
11 Annane D, Bellissant E, Sebille V, et al. Impaired pressor sensitivity to norepinephrine in septic shock patients with and without impaired adrenal reserve. Br J Clin Pharmacol 1998; 46: 58997[CrossRef][ISI][Medline]
12 Bollaert PE, Charpentier C, Levy B, Debouverie M, Audibert G, Larcan A. Reversal of late septic shock with supraphysiologic doses of hydrocortisone. Crit Care Med 1998; 26: 64550[ISI][Medline]
13 Briegel J Forst H, Haller M, et al. Stress doses of hydrocortisone reverse hyperdynamic septic shock: a prospective randomized, double blind, single-center study. Crit Care Med 1999; 27: 72332[ISI][Medline]
14 Keh D, Boehnke T, Weber-Cartens S, et al. Immunologic and hemodynamic effects of low-dose hydrocortisone in septic shock. A double-blind, randomized placebo-controlled crossover study. Am J Respir Crit Care Med 2003; 167: 51220
15 Baue AE. The horror autotoxicus and multiple-organ failure. Arch Surg 1992; 127: 145162[Abstract]
16 Barnes PJ, Adcock IM. How do corticosteroids work in asthma? Ann Intern Med 2003; 139: 35970
17 Annane D. Corticosteroids for septic shock. Crit Care Med 2001; 29: s11720[CrossRef][ISI][Medline]
18 McCombs CC, deShazo RD. Immune system disorders. In: Speight TM, Holford NHG, eds. Drug Treatment, 4th Edn. Chester: Adis International Ltd, 1997; 1304
19 De Jonghe B. Sharsar T, Lefaucheur JP, et al. Paresis acquired in the intensive care unit. JAMA 2002; 288: 285967
20 Fleming TR, DeMets DL. Surrogate end points in clinical trials: are we being misled? Ann Intern Med 1996; 125: 60513
21 Annane D, Sebille V, Charpentier C, et al. Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA 2002; 288: 86271
22 Bernard G, Vincent JL, Laterre PF, et al. Efficacy and safety of recombinant human activated protein c for severe sepsis. N Engl J Med 2001; 344: 699709
23 Cooper MS, Stewart PM. Corticosteroid insufficiency in acutely ill patients. N Engl J Med 2003; 348: 72734
24 Nasraway SA. The problems and challenges of immunotherapy in sepsis. Chest 2003; 123: 4519s[CrossRef]
25 Webster NR. Evidence based practice in intensive carelight on the horizon? Br J Anaesth 2001; 87: 3779
26 Bourne RS, Webber SJ, Hutchison SP. Adrenal axis testing and corticosteroid replacement therapy in septic shock patientslocal and national perspectives. Anaesthesia 2003; 58: 5916[CrossRef][ISI][Medline]
27 Millo J. Corticosteroids for patients with septic shock. JAMA 2003; 289: 41
28 Oxman AD, Guyatt GH. A consumer's guide to subgroup analysis. Ann Intern Med 1992; 116: 7884[ISI][Medline]
29 Bloomfield R, MacMillan M, Noble DW. Corticosteroid insufficiency in acutely ill patients. N Engl J Med 2003; 348: 2158
30 Beigel J, Eichacker P. Hydrocortisone and fludrocortisone improved 28-day survival in septic shock and adrenal insufficiency. ACP J Club 2003; 138: 44[Medline]
31 Owen H, Spence AA. Etomidate. Br J Aanaesth 1984; 56: 5557
32 Absalom A, Pledger D, Kong A. Adrenocortical function in critically ill patients 24 h after a single dose of etomidate. Anaesthesia 1999; 54: 8617[CrossRef][ISI][Medline]
33 Annane D, Sebille V, Bellisant E. Corticosteroids for patients with septic shock. JAMA 2003; 289: 434
34 Ledingham I, Watt I. Influence of sedation on mortality in critically ill multiple trauma patients. Lancet 1983; i: 1270
35 McKee JL, Finlay WEL. Cortisol replacement in severely stressed patients. Lancet 1983; i: 484
36 Ball C, Westhorpe R. Intravenous induction agents: etomidate. Anaesth Intens Care 2002; 30: 405[ISI][Medline]
37 Annane D, Briegel J, Sprung CL. Corticosteroid insufficiency in acutely ill patients. N Engl J Med 2003; 348: 21578