CORTICUS was the long-awaited trial addressing the use of corticosteroids in sepsis that was published in the NEJM this past January. Months prior to the leading auther Charles Sprung publishing it, the Tisch and Bellevue intensive care units halted corticotropin stimulation testing. Corticosteroids have warranted much publicity since CORTICUS came out—and rightly so as practice across the country has changed because of it. The Survinig Sepsis Campaign has now downgraded the recommendation on the use of corticosteroids in sepsis from C down to 2C based on this trial, which did not show a mortality benefit in those with septic shock. Should practice change as a result?
Don’t get me wrong, I was thrilled as an intern when I realized that I didn’t have to jump from bed 1 to bed 6 to bed 13 administering corticotropin and timing the cortisol draws at time 0, 30, and 60. But again, should practice change universally based on this? The beauty about treating the critically ill is that their physiology is so incredibly complex and altered that assessing and treating them well is a daunting task; and every small intervention could reduce time off the ventilator, pressors or at best decrease overall mortality. What makes these patients so fascinating to treat is the very reason studying these patients is tremendously difficult as well. And unfortunately, the trials looking at the use of steroids in sepsis are all quite different. And let’s face it, trying to figure out if steroids are effective in septic patients is nothing new. The trials really started back in the 1960’s when the Cooperative Study Group demonstrated that steroids were not beneficial with patients with “severe infection”. It was not until the 1980’s that we had large RCTs addressing this issue—both of which were negative for the use of steroids in sepsis. But it was the Annane studies in 2000 and 2002 that determined first the prognostic value in corticotropin stimulation as well as the fact that treatment of those relatively adrenally insufficient led to decreased time off vasopressors, reduced 28-day and 1-year mortality rates.
So why should we think twice about using corticosteroids in septic patients since CORTICUS? Corticosteroids showed no mortality benefit and the corticotropin test was of no value in predicting shock reversal. In fact, those receiving corticosteroids had higher rates of superinfection, new episodes of sepsis or septic shock, and higher rates of hyperglycemia as well. Simon Finfer’s editorial in the same issue of NEJM elaborates further on these findings. as well as the comparison between the CORTICUS findings vs. Annane’s. There are a couple of key differences that bear mentioning. For one, patients enrolled in the Annane study were certainly sicker, as they were hypotensive refractory to both fluids and vasopressors after 8 hours whereas Sprung’s patients were enrolled after remaining hypotensive on vasopressors 72 hours after adequate fluid resuscitation. Also, fludrocortisone (for additional mineralocorticoid activity) was used in addition to hydrocortisone in the Annane study—the effects of which are not completely clear. Finfer justifies CORTICUS since Annane’s study may have been flawed with the use of etomidate in these patients (for intubation purposes), which selectively inhibits adrenal corticosteroid synthesis, which possibly led to increased non-responders to corticotropin stimulation in the trial. It was only after statistical adjustment that Annane’s results were statistically significant that made it a positive study. So the CORTICUS trial was certainly warranted; however, critics of CORTICUS are swift to point out that the study was significantly underpowered—recruiting only 500 patients when they needed 800 patients to achieve 80% power. The study was stopped because of failure of enrollment and lack of finances to further the study.
So where does that leave us? The answer, inevitably, is that we’re not certain. In fact, the Campaign for Surviving Sepsis was just released in January 2008 and there appears to have been significant conflict within the committee regarding what recommendations to give. Some wanted to give two separate recommendations based on two different findings from the Annane and Sprung trial and others wanted one clear standing on the issue. The final outcome: 31-to-19 in favor of one recommendation—a 2C recommendation “that intravenous hydrocortisone be given only to adult septic shock patients with blood pressure poorly responsive to fluid resuscitation and vasopressor therapy.” The recommendation seems clear-cut, but the road leading to it has been conflicting, uncertain, and questionable. But this is the beauty of studying the critically ill. Every moment and hour that passes could sway a patient’s clinical status in either direction depending on the intervention administered or withheld. Endless questions remain regarding the timing of therapy, proper dosages, etc. Without question, the topic of sepsis is sexy, its management challenging, and its future to be determined. Unfortunately, the true role of corticosteroids in the septic patient is still a small part of this puzzle that is yet to be definitively resolved and will require ongoing investigation until it is pieced together correctly.
Sprung CL et al. Hydrocortisone therapy for patients with septic shock. N Engl J Med 2008 Jan 10; 358:111
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; 288:862–871
Dellinger RP et al. Surviving Sepsis Campaign: international guidelines for management of severe sepsis and septic shock: 2008. Critical Care Medicine 2008;36(1):296-327
Image courtesy of Ehrman Medical Library, Bellevue Hospital Men’s Ward, time period is circa 1870-1880.