Published in the May 2006 issue of Today’s Hospitalist
In an effort to reduce the mortality and morbidity of septic shock, a condition that kills nearly half of its victims, intensivists are rediscovering an inexpensive therapy that was the gold standard only 20 years ago.
Until the late 1980s, physicians regularly gave high, anti-inflammatory doses of steroids to patients in septic shock. But when research suggested the therapy wasn’t helping patients and could be harming them, the therapy fell out of favor.
Now, however, new studies re-evaluating the relationship between steroids and sepsis and inflammation have shown that the therapy works in treating septic shock when the patient is also suffering from relative adrenal insufficiency.
While steroids are back in favor with physicians who spend most of their time in the ICU, experts say that many nonintensivists who care for patients in septic shock “a group that includes hospitalists “aren’t taking full advantage of the therapy for treating septic shock.
And while intensivists want more of their colleagues outside of critical care to give steroids to septic shock patients, there are some significant caveats in using the therapy. Corticosteroids do not work for everybody, for example, and they may even harm some sepsis patients.
The key to appropriately treating septic shock with steroids, experts say, is discerning how to identify patients that need the therapy “and when steroids should be stopped after an initial trial.
Going beyond early recognition
Intensivists agree that one of the keys to treating sepsis is to detect the condition early. Early recognition is the best way to keep the condition from progressing and damaging precious tissue.
(For more on early recognition strategies, see “Time is tissue: Why emerging evidence on sepsis urges physicians to watch the clock” in the August 2005 issue of Today’s Hospitalist.)
While prevention through early recognition is the rallying cry of sepsis-treatment initiatives like the Surviving Sepsis Campaign, experts say that physicians need to do more when it comes to treatment. In particular, a growing chorus of experts is calling on physicians to give steroids to patients who require vasopressors to maintain adequate blood pressure after they’ve received aggressive IV-fluid resuscitation. The target is the patient who has not responded to lots of fluids and thus requires vasopressors to treat the shock.
“For the first few hours that you have a patient with sepsis, you need to concentrate on fluid resuscitation and vasopressors,” explains Margaret M. Parker, MD, professor of pediatrics, medicine and anesthesia at the State University of New York at Stony Brook. “After two or three hours of fluid resuscitation, and after you have had an hour or more of vasopressors, you should start steroids.”
Evidence backing up that strategy comes from an article in the Aug. 21, 2002, Journal of the American Medical Association, which was led by Djillali Annane, MD. That study supported starting 200 mg of daily hydrocortisone immediately (administered intravenously as 50 mg every six hours) and continuing the therapy for five to seven days. Some experts call for tapering off the drugs over a few days “either once septic shock has resolved or at the end of seven days “just as you would do with other steroid regimens.
A growing body of research has similarly found a benefit when steroids were used to treat septic shock. A number of smaller studies, as well as a meta-analysis that looked at many of the studies published since 1987, found that low-dose steroid regimens have consistently reported a beneficial effect for steroids.
William Janssen, MD, a critical care specialist and instructor of medicine at the University of Colorado Health Sciences Center, says that steroid therapy for septic shock not only reduces mortality, but does so fairly efficiently. “The number needed to treat to save one life is seven,” he says.
Intensivists say that there is plenty of anecdotal evidence about the drugs’ effectiveness. “I have seen plenty of people who were on vasopressors,” says Dr. Parker, who is a member of the Surviving Sepsis Campaign steering group, which has come out in favor of steroid therapy. “We couldn’t get them off, and when we gave them steroids, they came off within a relatively small number of hours. I was not doing this [five years ago]. Now I do this.”
While the evidence on steroids and sepsis is convincing enough that organizations around the world have revised their clinical practice guidelines to recommend low-dose steroids for the treatment of septic shock, the data leave some experts hungry for more information about the exact role that steroids play in septic shock.
Dr. Parker, for example, says that despite her anecdotal experience with steroids, she is still not sure exactly how effective the drugs are at combating sepsis. That’s why she, along with many others in the critical care community, are looking forward to seeing the results of the CORTICUS trial, a large, multicenter study that is underway in Europe. Results are expected later this year.
The role of testing
One key issue the CORTICUS trial may address is the role of testing, says Curtis Sessler, MD, the Orhan Muren professor of medicine and medical director of critical care at Virginia Commonwealth University in Richmond.
While Dr. Annane’s study found a clear benefit to treating septic shock with steroids, it also raised some important questions. While the study found an overall mortality benefit from giving low doses of steroids to patients in septic shock, the benefit was limited to a subgroup of patients whose adrenal glands either did not naturally produce (or produced very little) extra cortisol when stimulated by a large dose of adrenocorticotropic hormone, commonly known as ACTH or corticotropin.
Dr. Annane referred to study subjects whose cortisol levels did not increase by more than nine points as “nonresponders,” meaning they have relative adrenal insufficiency. The adrenal glands of these patients have exhausted their ability to produce additional steroids despite the body’s demand for more in the face of septic shock.
Dr. Janssen from the University of Colorado explains that Dr. Annane’s research found that “patients who exhibited an appropriate response to ACTH by increasing their cortisol levels didn’t really benefit from steroids.” In nonresponders, on the other hand, Dr. Janssen explains, “the adrenal glands aren’t producing the hormone, and we need to give the body a boost by giving mineralocorticoids.”
Steven M. Hollenberg, MD, director of the coronary care unit at Cooper Health System in Camden, N.J., says that septic patients may be producing a relatively “normal” level of corticosteroids, but the question is whether they should be producing significantly more steroids when they’re sick.
“That’s the concept of relative adrenal insufficiency,” explains Dr. Hollenberg, who is author of a Society of Critical Care Medicine practice guideline on hemodynamic support of adult patients with sepsis. “You should be making more than you are.”
Absolute cortisol levels
While that line of thinking explains why Dr. Annane and others examine patients’ cortisol levels after an ACTH stimulation test, other experts prefer to focus on patients’ absolute cortisol level, not changes produced by an ACTH stimulation test.
Dr. Sessler, for example, says that simply taking a random cortisol level for every patient in septic shock would probably be just as useful. A random cortisol level would let physicians see how much cortisol patients are able to make for themselves when their body is in a stressed state of shock. He notes that giving an extra dose of ACTH may not provide significantly more information.
“Critics argue that the supra-physiologic dose of ACTH is far more than what the body would actually generate,” Dr. Sessler says. “Therefore it is not a good test of what their adrenal function should be in a stressed state like septic shock.” If a random cortisol level in a critically ill patient shows a level of less than 25 (or, some argue, less than 15), he explains, the patient could be suffering from relative adrenal insufficiency and simply need hydrocortisone and not ACTH testing.
There are other problems with giving every septic shock patient an ACTH stimulation test. A very real concern is that it is simply not feasible in the real world.
The test requires timed blood draws (30 and 60 minutes after ACTH is administered), something that is not always possible in every hospital. In addition, some labs do not have the ability to turn the result around quick enough to be useful when time is of the essence.
Skewed test results
There is another problem with ACTH testing: If patients are already taking steroids, the test results may be inaccurate. As a result, experts say physicians who are considering performing an ACTH stimulation test should avoid giving hydrocortisone. DexaÂ¬methasone is less likely to interfere with the test and can be given instead. Once the final cortisol level has been drawn, hydrocortiÂ¬sone can be started.
If physicians feel strongly about performing an ACTH stimulaÂ¬tion after starting a patient on steroids, says Philip Dellinger, MD, director of the critical care section at Cooper University Hospital in Camden, N.J., the best bet is to initiate dexamethasone, but physiÂ¬cians need to realize that it will still affect the adrenal axis.
“My experience is that some of my attendings do the test, and some don’t,” says Dr. Dellinger, who is also a member of the executive committee of the Surviving Sepsis Campaign. “I think most people feel that the overwhelming majority of septic shock patients that we see are not at risk [of suffering serious side effects] from seven days of low-dose steroids, as long as you are also giving them antibiotics to cover the pathogen that is causing the process.”
Anticipating side effects
And while common steroid side effects, including secondary infections, gastrointestinal bleeding and increased blood glucose, are a concern, most experts say they can be managed. In an encouraging meta-analysis of clinical trials examining steroid use in sepsis patients, Dr. Sessler explains, the Cochrane Review group found gastrointestinal bleeding, secondary infections and troublesome hyperglycemia to be no more common with steroids than placebo.
“To be fair, it is very difficult to figure out a steroid side effect in a critically ill, septic patient,” says Dr. Hollenberg from Cooper Health. “Lots of things go wrong. These patients are receiving eight drugs and they have six problems. It’s hard to point to one drug and say that’s the problem. That’s why we need a large trial.”
In some patients, however, side effects do merit more serious conÂ¬sideration. Dr. Dellinger says that there are patients who face a greater risk of complications of steroid therapy in which the side effects of a steroid regimen would make him think twice about using the drugs. These include patients with major burns (at risk of secondary infections) or patients on neuromuscular blockers (at risk of prolonged neuromuscular blockade following discontinuation of steroids).
“These are two circumstances in which I would still give the steroids,” he says, “but I would do it under cover of an ACTH stimulation test, and I would stop the steroids if they are a responder.”
Knowing when to stop therapy
What’s the bottom line when it comes to treating septic shock with steroids? Because Dr. Annane’s work showed that patients in septic shock are more likely to be ACTH nonresponders than responders, most experts say they give all patients in sepÂ¬tic shock low doses of corticosteroids as they wait for results of the ACTH stimulation test. If the test results indicate that the patient is a responder, they stop the steroids before the seven-day course is finished.
For many intensivists, knowing when to stop the therapy is more important than deciding when to start it. “I would support starting steroids in all patients who have pressor-dependent septic shock,” Dr. Sessler explains. “The question is when do you discontinue steroids.”
“I would argue that patients who have an impressive hemodynamic response” “they need reduced doses of vasopressors after starting the steroids “”should have the steroids continued until recovery regardless of their adrenal axis testing,” he says. “If we stop steroids in those patients, their blood pressure status will deteriorate in many cases.”
For now, however, there is little in the way of definite guidance, even from organizations like the Surviving Sepsis Campaign.
“It’s hard to say what the recipe should be, because the studÂ¬ies have used slightly different doses or slightly different definiÂ¬tions or slightly different lengths of time,” Dr. Parker says. The campaign’s guidelines follow Dr. Annane’s protocol, she says, but experts say the pending results of the CORTICUS trial could change some details.
Dr. Sessler agrees: “It’s an important development for good patient outcomes, but we need more research to guide some of the details.”
Deborah Gesensway is a freelance writer who reports on U.S. health care from Toronto, Canada.
The statistics on sepsis: a look at incidence, outcomes and cost
With mortality rates for various forms of sepsis on the rise, the statistics on this deadly condition are not encouraging.
A number of recent studies that have quantified the incidence, outcomes and cost of severe sepsis (systemic inflammatory response syndrome with infection and organ dysfunction) in the United States have found the condition accounts for 2.26 cases of every 100 hospital discharges annually. The condition kills 20 percent to 30 percent of its victims. Once severe sepsis progresses into septic shock, it becomes even more deadly; about half of the patients with septic shock die.
Adding to the bad news, the incidence of severe sepsis and septic shock is projected to increase by as much as 1.5 percent each year. That trend is being fueled in part by the aging of the populaÂ¬tion, antibiotic resistance, immunosuppression and more aggressive use of invasive procedures.
For more information
For guidance on how to manage patients with sepsis, physicians can turn to the Surviving Sepsis Campaign’s Web site. In addition to clinical guidelines, the Web site contains an easy link to the Institute for HealthÂ¬care Improvement’s “Severe Sepsis Bundles,” which includes advice on how to administer low-dose steroids for septic shock.