Web site designed and maintained by Chris Cairns  © SICS EBM Group 2004                                  

Up

 

Low dose steroids reduce septic shock mortality.

 

In patients with low / no response to short ACTH test (increment < 250 nmol.l-1).

For every 8 patients given low dose hydrocortisone and fludrocortisone, compared to placebo, 1 less ICU death.     (95% CI 4 to 552)
For every 7 patients given low dose hydrocortisone and fludrocortisone, compared to placebo, 1 less hospital death.   (95% CI 4 to 27)
Level of evidence 1+

 

Citation/s:   Effect of treatment with low doses of hydrocortisone and fludrocortisone on mortality in patients with septic shock. JAMA 2002; 288: 862 - 871.
 

Lead author's name and fax:  Djillali Annane, Service de Reanimation Medicale, Hopital Raymond Poincare, 92380 Garches, France. djillali.annane@rpc.ap-hop-paris.fr

 

Three-part Clinical Question: In patients with septic shock, does the use of low dose steroid compared to standard therapy, affect 28-day mortality?
 

Search Terms: 1. exp sepsis/ or severe sep$.tw or sept$.tw or sepsi$.tw (50301), 2. exp critical care/ or critical ca$.tw or intensive ca$.tw (22553), 3. exp hydrocortisone/ or hydrocort$.tw or exp steroids/ or steroi$.tw (110778), 4. 1 and 2 and 3 (107), 5. therapy filter (652119), 6. 4 and 5 (70)

 

The Study: Double-blinded concealed randomised controlled trial with intention-to-treat.
 

The Study Patients: Carried out in 19 French ICUs, over 4 year period. Age ≥ 18y, and all criteria:1. documented or strong suspicion of infection, 2. temp > 38.3 °C or < 35.6 °C, 3. heart rate > 90.min-1, 4. SBP < 90 mmHg for > 1 h, despite fluid replacement and inotropes, 5. urine < 0.5ml.kg-1 or PaO2/FiO2 < 37.3 kPa, 6. lactate > 2 mmol.l-1, 7. mechanical ventilation.   Exclusions: pregnancy, MI, PE, advanced cancer, AIDS, or contraindication or definite indication for steroids. All enrolled patients had short synacthen test - test results measured at central laboratory – result used to stratify patients – enrolled if peak following ACTH < 250 nmol.l-1.
 

Control group (N = 115; 115 analysed): Identical placebos of hydrocortisione and fludrocortisone. No details on other aspects of ICU care - pragmatic study.
 

Experimental group (N = 114; 114 analysed): Hydrocortisone 50 mg iv 6 hourly, fludrocortisone 50 microg ng (justified by risk of missing primary adrenal insufficiency). No details on other aspects of ICU care - pragmatic study.

 

The Evidence:

Outcome

Time to Outcome

CER

EER

RRR

ARR

NNT

Mortality

28-day

0.635

0.526

17%

0.109

9

95% Confidence Intervals:

-3% to 37%

-0.018 to 0.236

NNT = 4 to INF to NNH = 55

Mortality

ICU

0.704

0.579

18%

0.125

8

95% Confidence Intervals:

0% to 35%

0.002 to 0.248

4 to 552

Mortality

Hospital

0.765

0.61

20%

0.155

6

95% Confidence Intervals:

5% to 36%

0.037 to 0.273

4 to 27

 

Comments:


1. Well-defined group of patients, septic shock (> SCCM definition).  We look after similar patients.
2. Powerful effect, low cost therapy.
3. No evidence of harmful effect of steroids - infection, g-i bleeding or psychiatric disorders.
4. Authors show some slick footwork to display difference in 28-day mortality - limits our certainty of effect.   Event rates closer to 50% increases size of NNT confidence intervals.    But difference in ICU + hospital mortality. 
5. Difference limited to non-responders to short synacthen test (increment of < 250 n.mol.l-1)
6. See - A 3-level prognostic classification in septic shock based on cortisol levels and cortisol response to corticotrophin.  JAMA   2000; 283: 1038 – 1045.

 

Appraised by: Malcolm Daniel, Department of Anaesthesia, Glasgow Royal Infirmary.; Saturday, August 24, 2002.                 Email: md23s@udcf.gla.ac.uk

 

Printer-friendly format