GARCHES, FRANCE—The use of corticosteroids in patients with septic shock has long been a point of contention. A recent randomized controlled trial was conducted by a French team who theorized that a short course of corticosteroids in physiologic doses would improve 28-day survival in patients with septic shock and “relative” adrenal insufficiency.[1]

“This is an important study that attempts to address the controversy concerning steroid replacement therapy in severe sepsis and septic shock,” said Robert A. Balk, MD, Director of Pulmonary and Critical Care Medicine at Rush-Presbyterian-St. Luke’s Medical Center and Cook County Hospital in Chicago. He added, though, that the results do not completely answer the question the study was designed to address; in fact, they raise “a new controversy concerning the definition of ‘relative adrenal insufficiency.’ ”

THE STUDY

Nineteen intensive care units (ICUs) enrolled a total of 299 patients between October 9, 1995, and February 23, 1999. Patients were randomized to the treatment or placebo group within eight hours of septic shock onset. A short corticotropin test was performed on all patients, and blood samples were taken before and 30 and 60 minutes after administration of 250 µg of adrenocorticotropic hormone. Patients with a response of 9 µg/dL or less were classified as nonresponders (that is, they were defined as having “relative” adrenal insufficiency).

The treatment group was given 50 mg of intravenous hydrocortisone every six hours and 50 µg of fludrocortisone once daily via a nasogastric tube. Treatment was continued for seven days. The primary end point of the study was 28-day survival in the nonresponder group; the secondary end point was 28-day survival in the responders.

BENEFIT FOR NONRESPONDERS

In the study, there were 229 nonresponders (115 in the placebo group and 114 in the treatment group) and 70 responders (34 in the placebo group and 36 in the treatment group).

Among nonresponders, 63% of patients in the placebo group had died at 28 days compared with 53% in the treatment group. The median time to death in the placebo and treatment groups was 12 and 24 days, respectively.

Fifty-three percent and 61% of responders had died by day 28 in the placebo and treatment groups, respectively. The median time to death was 14 days in the placebo group and 16.5 days in the treatment group.

By the end of the ICU stay, mortality among nonresponders had increased to 70% in the placebo group and 58% in the treatment group. Among responders, mortality in the respective groups was 59% and 67%.

The differences in mortality among the nonresponders were statistically significant, although the confidence intervals neared 1. Neither of the mortality differences among the responders even approached significance.

A LOT OF UNKNOWNS

The authors recommended that all patients with catecholamine-dependent septic shock be given hydrocortisone/fludrocortisone therapy immediately after a short corticotropin test. When the test results become available, treatment can be withdrawn in responders, whereas nonresponders should continue on corticosteroids for seven days.

Dr. Balk observed that “this study targeted a very select group of patients with vasodepressor-dependent septic shock. All of the patients enrolled in the study were intubated, ventilated, and resuscitated according to a defined protocol and guided by the use of a pulmonary artery catheter. In most centers, cortisol levels are not available on a ‘stat’ basis.”

“The drawback to starting everyone on [cortico]steroids and stopping if they are found to be adrenally sufficient is that there are no data … to tell us that it is truly safe,” said Dr. Balk. However, he pointed out, there is a wealth of data showing that higher corticosteroid doses lack benefit and may increase toxicity compared with a placebo.

“I think we need a greater understanding of the physiology of the normal adrenal response in the setting of septic shock and critical illness before we can even speculate whether one steroid replacement strategy is better than another,” concluded Dr. Balk.

—Gale Jurasek

Reference
1. Annane D, Sébille 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.

Return to table of contents