THERAPY

Criteria abstracted from The Users' Guides to the Medical Literature series in JAMA

Am J Respir Crit Care Med. 2006 Feb 1;173(3):276-80..[abstract]

Review posted August 23, 2007

1. What is being studied?:
   
   
   1. The study objective:
      
      To determine if stress dose corticosteroid supplementation during periods of adrenal insufficiency, would improve ventilator weaning and clinical outcome in ICU patients with respiratory failure as the primary etiology. 
      
      The rationale behind conducting this trial was that adrenal insufficiency (AI) is common in intensive care units (ICUs) and corticosteroid supplementation may be beneficial during periods of physiologic stress as reported in trials by Annane (1).  The authors argue that the relationship between the success of ventilator weaning and AI is unknown.   The authors further hypothesize that patient morbidity and mortality can be positively influenced in the setting of corticosteroid supplementation where adrenal reserve has been found insufficient. 
      
      
   2. The study design:
      
      The study was designed as a prospective, randomized, double blinded single 3,000 bed tertiary teaching hospital in Taiwan with a 26 bed ICU.  Data collection began May 1, 2003 and ended December 31, 2003.  The hospital ethics committee approved the study and informed consent was obtained from the patient or close friend.
      
      
   3. The patients included:
      
      Patients included in the study were those deemed to be in the ventilator weaning phase:
      
      
      • those under mechanical ventilation for more than 72 hours
      • primary etiology of respiratory failure resolving
      • vasopressors and sedatives discontinued for at least 24 hours
      • stable hemodynamics
      • neurologically intact (Glasgow coma score >11)
      • n= 93
        
        
   4. The patients excluded:
      
      Patients were excluded for the study secondary to:
      
      
      • unstable condition (FI02 > 50%, vasopressors requirement to maintain blood pressure, body temp > 39C or < 36C, heart rate > 150 beats/min or < 40 beats/min, or respiratory rate > 35 breaths/min or < 8 breaths/min)
      • receiving corticosteroids
      • poor neurological condition
      • no cough reflex
      • n = 86
        
        
   5. The interventions compared:
      
      The study randomized patients with adrenal insufficiency—defined as AM cortisol less than 25 mcg/dl AND failure to have at least a 9 mcg/dl increase with cosyntropin stimulation test (250 mcg IM cosyntropin injection with blood samples taken 60 min after injection)  to receive either 50 mg hydrocortisone every 6 hours during ventilator weaning  or a normal saline placebo.
      Comparisons were made with adrenally intact patients (n=23) who where not randomized.
      
      
   6. The outcomes evaluated:
      
      Primary outcome evaluated:
      

   • success of ventilator wean to extubation  (defined as: not requiring reintubation or additional respiratory support within 48 hours of extubation.  In addition, if weaning was unsuccessful and total ventilator time was more than 14 days- this was defined as weaning failure.)

   Secondary outcomes evaluated:

   • duration of ventilator wean
   • ICU length of stay
   • hospital length of stay
   • hospital mortality

2. Are the results of the study valid? 
   
   
   1. Was the assignment of patients to treatments randomized?
      
      Yes adrenal insufficient patients were randomized via computer generated random number table.   35 patients received the treatment of 50mg cortisol every 6 hours and 35 patients received the placebo.
      
      
   2. Were all patients who entered the trial properly accounted for and attributed at its conclusion?
      
      
      1. Was follow-up complete?
         
         Yes, all patients were accounted for and follow- up was complete. 
         
         
      2. Were patients analyzed in the groups to which they were randomized?
         
         Yes patients were analyzed in the groups to which they were randomized.
      
      
   3. Were patients, health workers, and study personnel "blind" to treatment?
      
      Yes, the study was double-blinded.
      
      
   4. Were the groups similar at the start of the trial?
      
      There were no statistically significant differences of the three study groups at the start of the study- included in this analysis were age, sex, APACHE III, rapid shallow breathing index (RSI- defined as a respiratory rate greater than 35 breaths/min for more than 5 minutes or less than 8 breaths/min), ventilator days prior to randomization, or PaO2/FIO2.  The means were compared among the three groups using one-way analysis of variance testing.  As would be expected from the study design, morning cortisol levels were found to be higher in the adequate adrenal reserve group (38.3± 20.8) than with the treatment (13.7 ± 6.0) and placebo (17.0 ± 5.0) groups with a statistical significance of <0.001.  This difference of cortisol levels was not found when comparing the treatment and placebo groups.
      
      
   5. Aside from the experimental intervention, were the groups treated equally?
      
      Yes the ventilator weaning protocol for the ICU was followed by the respiratory therapists; all sedatives and vasopressor agents commenced 24 hours from the study.
      
      
3. What were the results?
   
   
   1. How large was the treatment effect?

   
   The authors defined successful weaning as not requiring reintubation or additional respiratory support within 48 hours of extubation. In addition, if weaning was unsuccessful and total ventilator time was more than 14 days- this defined a weaning failure. Of the normal adrenal function group (n= 23), 20 were successfully weaned from the ventilator with 3 failures post extubation.  The corticosteroid treatment group had 32 successfully ventilator weans, 1 failed T-piece trial, and 2 failed post extubation.  The placebo group had 24 successful ventilator weans, 2 failed T-piece trials, and 9 failures post extubation.
   
   91% of  the patients in the corticosteroid treatment group were successfully weaned from the ventilator (95% CI = 82 to 97%) ; compared to a 69% success rate (95% CI = 63 to 78%) in the placebo group (p=0.035).   The adequate adrenal reserve group had a 87% success rate in ventilator weaning  (95% CI = 79 to 95%). 
   
   The steroid treatment reduced the relative risk of non successful weaning (treatment vs placebo groups)  (RR = 0.27; 95% CI = 0.09 to 0.81).  There was a 23% absolute risk reduction (ARR = 23%; 95% CI = 4 to 38%) and a 73% relative risk reduction (95% CI = 19 to 92%) when patients were treated for AI. 
   
   Duration of weaning for the treatment group had a mean of 3.4 days (95% CI = 1.3 to 5.7 days) vs 6.5 days (95% CI 1.8  to  11.2  days) among the placebo group.  The adequate adrenal reserve group had a duration of weaning of 2.9 days (95% CI =   1.2  to 4.6 days).  The patients with adequate adrenal reserve group and those with AI in the treatment group had a statistically significant shorter duration of weaning than those with AI treated with placebo (p = 0.044).
   
   The differences among the secondary outcome measures ICU length of stay, days of hospital stay, and hospital mortality were not significant among the three groups.

2. How precise was the estimate of the treatment effect?
   
   The relative risk of unsuccessful weaning was 0.27 in favor of treating AI patients with corticosteroid replacement, and the 95% confidence intervals, 0.09 to 0.81, do not cross 1, further suggesting that steroid treatment of AI will result in more successful weaning. 
   
   The effect of successfully weaning from the ventilator with corticosteroid treatment in those patients with AI is also evident in the confidence interval calculated for the NNT (3 to 24).
   
   
4. Will the results help me in caring for my patients?
   
   
   1. Can the results be applied to my patient care?
      
      Maybe. Adrenal insufficiency has been observed in pediatric patients in several studies.  In 2005 Pizarro et al. (2) showed that absolute and relative adrenal insufficiency is common in children with catecholamine-resistant shock; similar results were found in Sarthi et al. (3) study in 2007.   Suominen et al. (4) published study outcomes in 2005 a positive response to hydrocortisone supplementation in those pediatric patients in a low cardiac output state post cardiothoracic surgery who were not responding to increasing inotropic support or fluid administration.  Ando et al. (5) looked at a similar population of pediatric patients in 2005 and also found positive results when stress dose hydrocortisone was supplemented in adrenal insufficient patients, they also found as part of a secondary outcome mechanical ventilation time shortened from 138.2 ± 89.7 to 83.5 ± 42.1 hours in the hydrocortisone group.  In summary, adrenal insufficiency is an observed phenomenon in pediatrics during ICU admissions though perhaps not in “stable” patients as described in this paper of adults.  The incidence and significance of adrenally insufficient children with primary resolving respiratory failure is unknown.  There are times of physiological stress when hydrocortisone supplementation has shown to improve outcomes in pediatrics; larger studies need to be conducted in pediatrics looking specifically at hydrocortisone treatment effects including outcomes related to ventilator weaning.
      
      
   2. Were all clinically important outcomes considered?
      
      Yes.  Primary as well as secondary outcomes were addressed and analyzed.  
      
      
   3. Are the likely treatment benefits worth the potential harms and costs?
      
      I would obtain an AM cortisol with ACTH stimulation testing on the pediatric patients who are in the weaning stages of the ventilator. Although long term effects of hydrocortisone supplementation have not been studied, the doses given are physiologic (although there is ongoing debate on appropriate dosing); therefore, there is not expected harm from this therapy.  Patients have an increased risk of morbidity and mortality with a prolonged stay in the ICU related to a variety of causes, including nosocomial infections.  That being said I feel this would decrease the risk to benefit ratio. There is also considerable cost associated with prolonged ventilation and prolonged length of stay in the ICU.  This study did report complications, such as new onset hyperglycemia, nosocomial infections, and gastrointestinal bleeding, which were not found to be significantly increased.

             References

1. Annane D. Corticosteroids for septic shock.  Critical Care Medicine. 29(7 Suppl):S117-20, 2001 Jul.
   
   
2. Pizarro C. F., Troster E. J., Damiani D., Carcillo J. A.  Absolute and relative adrenal insufficiency in children with septic shock.  Critical Care Medicine. 33(4):855-9, 2005 Apr.
   
   
3. Sarthi M., Lodha R,. Vivekanandhan S., Arora NK. Adrenal status in children with septic shock using low-dose stimulation test.  Pediatric Critical Care Medicine. 8(1):23-8, 2007 Jan.
   
   
4. Suominen P. K., Dickerson H. A., Moffett B. S., Ranta S. O., Mott A. R., Price J. F., Heinle J. S., McKenzie E. D., Fraser C. D., & Chang A. C.   Hemodynamic effects of rescue protocol hydrocortisone in neonates with low cardiac output syndrome after cardiac surgery.  Pediatric Critical Care Medicine. 6(6): 655-659, 2005.
   
   
5. Ando M., Park I., Wada N., & Takahashi Y.  Steroid supplementation: A legitimate pharmacotherapy after neonatal open heart surgery.  The Annals of Thoracic Surgery. 80 (5): 1672-1678, 2005.

Last Updated: August 23, 2007

Fill out our Feedback Form | Editorial Notes | About Us