There is no role for steroids in sepsis in the absence of shock, nor is there a role for high-dose steroids in sepsis.
by Nikhil R. Gandhi, MD, FACP, Deepak G. Asudani, MD, FHM
An 81-year-old woman with diabetes mellitus presents with a three-day history of fever, chills, left-side flank pain, and dysuria. Her blood pressure upon presentation is 75/45 mm/Hg, her heart rate is 120 beats per minute, and she has a temperature of 103.1°`F and a respiratory rate of 22 breaths/minute. On physical examination, she is an ill-appearing elderly woman, with dry oral mucosa and left costo-vertebral angle tenderness. Lab work shows leukocytosis of 18,000 mg/dL with 88% polymorphonuclear leukocyte (PMN), the urine analysis is consistent with a urinary tract infection, and a chemistry panel reveals elevated BUN and creatinine levels of 52 mg/dL and 2.4 mg/dL, respectively. In the emergency department, she is given a bolus of 2 liters normal saline, but her blood pressure remains 78/49 mm/Hg. She is then started on broad-spectrum antibiotics and a norepinephrine drip, and is admitted to the ICU.
What role would steroids add to her management?
Sepsis is the clinical syndrome defined by the presence of systemic inflammatory response syndrome (SIRS) in the setting of an infection. SIRS is defined by the presence of at least two of the following: fever or hypothermia; leukocytosis, leukopenia, or bandemia; heart rate >90 bpm; or tachypnea or hypocapnia.
When acute organ dysfunction, such as acute renal failure, altered mental status, or acute lung injury (hypoxemia), is present, sepsis is classified as severe.
Septic shock is a state of sepsis associated with acute circulatory collapse characterized by persistent arterial hypotension (defined as a systolic blood pressure <90 mmHg, a mean arterial pressure <60 mmHg, or a reduction in systolic blood pressure of >40 mmHg from baseline) despite fluid resuscitation attempts.1
The incidence and mortality due to sepsis and septic shock is directly related to the age of the patient, many of whom require ICU hospitalization.2 Clinically, this portends a great challenge, as the incidence of sepsis is likely to increase as the U.S. population ages.
Initial management of a patient with sepsis/septic shock is goal-directed therapy, which consists of early administration of broad-spectrum antibiotics, crystalloid or colloid fluid resuscitation, and use of vasopressor support to improve hemodynamics and maintain a mean arterial
pressure ≥65 mmHg. Patients with acute lung injury may also require prompt ventilator support.
The role of steroids in sepsis is controversial.
Steroids have long been known for their anti-inflammatory properties. From the 1950s to the 1980s, high-dose steroids (methylprednisolone 30mg/kg and dexamethasone 3 mg/kg to 6 mg/kg in divided doses) were used in the management of sepsis. This was based on a study by Schumer that showed steroids reduced mortality to 10% from 38%.3
Later, Sprung and colleagues demonstrated reversal of shock and improved short-term survival with high-dose steroids in patients with sepsis, but subsequent prospective randomized trials did not support this beneficial effect of high-dose steroids.4-6 In fact, two meta-analyses in 1995 concluded that high-dose steroids are ineffective and potentially harmful, and associated with higher mortality, secondary infections, and renal and hepatic dysfunction.7,8 Thereafter, the use of high-dose steroids fell into disfavor.
In the early 2000s, there was an emergence in the use of low-dose steroids in patients with sepsis. This was based on various trials showing the benefit of the use of low-dose steroids in the reversal of septic shock without significant side effects, discussed further below.
Steroids improve hemodynamic parameters. In an animal model, Hinshaw and colleagues induced septic shock in adrenalectomized dogs by infusing lethal doses of E. coli. The untreated dogs died within hours, whereas the dogs treated with antibiotics and steroids had a complete recovery from shock, and survived more than 100 hours.9
During sepsis, endotoxins induce nitric oxide synthase, which produces relaxation of vascular smooth muscle tone, with resultant hypotension and reduced contractility response to norepinephrine.10 Corticosteroids prevent induction of nitric oxide synthase and enhance the vaso-active response to catecholamines through the glucocorticoid receptors. In vascular endothelial cells, glucocorticoids also inhibit serum phospholipase A2, reducing the production of vasodilators, such as prostacyclin and prostaglandin E1.11,12
Steroids reduce inflammation. Sepsis is driven by a systemic inflammatory response, in which components of the outer-cell membrane of both gram-positive and gram-negative bacteria and endotoxins induce the production of inflammatory cytokines, such as tumor necrosis factor alpha (TNF-alpha) and interleukin-1 (IL-1).13 These cytokines have a direct toxic effect on various tissues. In addition, inflammatory cytokines suppress adrenal response to adrenocorticotropic hormone (ACTH), which results in decreased endogenous cortisol production, and compete with glucocorticoids for their receptors, inducing resistance to the action of steroids at the tissue level.14
In healthy volunteers challenged with bacterial endotoxins, low-dose steroids (~10 mg of prednisolone) suppress the release of proinflammatory cytokines and prevent the activation of endothelial cells and neutrophils.15 Steroids also inhibit the release of toxic enzymes, such as lysozymes and superoxides from neutrophils.16,17
Mortality data. Two major studies evaluated the effect of low-dose steroids in patients with septic shock. Annane and colleagues conducted a placebo-controlled, blinded trial and divided the study population into “responders” and “nonresponders” based on their response to ACTH stimulation test. Within the “nonresponder” group, steroids reduced the risk of mortality by 16% (63% mortality in the placebo group and 53% mortality in the corticosteroid group, P=0.02).16 Steroids also significantly reduced ICU mortality (70% versus 58%), hospital mortality (72% versus 61%) and one-year mortality (77% versus 68%) compared with placebo. No statistically significant difference in mortality between steroids and placebo was seen in the “responder” group.16
The CORTICUS trial, a multicenter, randomized, double-blind, placebo-controlled trial, showed no significant difference in 28-day mortality between those treated with corticosteroids (39.2%) and those receiving a placebo (36%, P=0.069).17 There was also no significant difference in either hospital or ICU mortality in this study.
A recent meta-analysis demonstrated no significant effect of corticosteroid treatment on 28-day mortality, ICU mortality, or hospital mortality in septic shock. However, subanalysis of trials using a prolonged course (>5 days) of low-dose steroids (300 mg of hydrocortisone or equivalent) showed a significant reduction in 28-day all-cause mortality (P=0.02) and hospital mortality (P =0.05), and a decrease in ICU length of stay.18
Reversal of shock. Various studies have shown a decrease in the time necessary to reverse septic shock with the use of low-dose steroids. Annane and colleagues showed the median time to vasopressor therapy withdrawal was seven days in the group treated with steroids versus nine days in the placebo group (P=0.01).16 The CORTICUS study demonstrated significantly shorter times to reversal of shock in the group treated with hydrocortisone compared with the placebo group—3.3 days versus 5.8 days (P<0.001).17 In a smaller study, 68% of hydrocortisone-treated patients achieved seven-day shock reversal compared with 21% in the placebo group, a difference of 47% in the rate of reversal of shock.18
In which sepsis patients should I use steroids? The large clinical trials that found a benefit to low-dose steroids included patients with a systolic blood pressure <90 mm/Hg for more than one hour, despite aggressive fluid and vasopressor therapy. Based on these and other smaller trials, the Surviving Sepsis Campaign recommends the addition of IV steroids to those patients with septic shock who don’t respond to adequate fluid and vasopressor resuscitation.19
Should I obtain ACTH stimulation test in these patients? Although the Annane study showed that “nonresponders”—those who did not achieve ≥9 mcg/dL increase in cortisol level after 30 to 60 minutes of ACTH administration—were more likely to benefit from steroids, the overall trial population appeared to benefit regardless of the ACTH response.16
Furthermore, the CORTICUS study showed no difference between the corticotropin responder and nonresponder group. Also, most cortisol immunoassays measure total cortisol (free and protein-bound), whereas the free cortisol level is the more clinically relevant measurement. Hence, current guidelines from the American College of Critical Care Medicine and Surviving Sepsis Campaign do not recommend performing an ACTH stimulation test prior to administration of steroids.20
What type of steroids should I use, and at what dose? Current guidelines recommend IV hydrocortisone in a dose of 200 mg/day to 300 mg/day given as 50 mg every six hours or 100 mg every eight hours for at least seven days before tapering. Alternatively, IV hydrocortisone can be given as a bolus of 100 mg followed by a continuous infusion at 10 mg/hr (240 mg/day). Hydrocortisone at this dose has intrinsic mineralocorticoid activity, obviating the need for adding fludrocortisone. Fludrocortisone may otherwise be added as 50 mcg daily, if using a corticosteroid without significant mineralocorticoid activity. Patients with septic shock should not be treated with dexamethasone, which causes immediate and prolonged suppression of the hypothalamic-pituitary-adrenal axis.21
Do I need to taper off the steroids? It is recommended to wean the steroids after seven or more days of use, when vasopressors are no longer required. Keh and colleagues noted a 30% recurrence of shock in patients when the steroids were not tapered.22 There was also evidence of immunologic rebound after abrupt cessation of steroids, with an increase in inflammatory markers.23 The taper should decrease the dose every two or three days in small steps.
What potential side effects should I be concerned about? Overall, the use of higher dose corticosteroids is associated with significant potential side effects, including a worsening of the underlying infection, new infection, hyperglycemia, hypernatremia, and gastrointestinal bleeding. In a meta-analysis of nine clinical trials with high-dose corticosteroids (a starting dose of ~30mg/kg/day of methylprednisolone), Cronin and colleagues found a trend toward increased mortality due to secondary infections (relative risk 1.70; 95% confidence interval, 0.70 to 4.12).24 A recent meta-analysis of 15 trials found low-dose corticosteroids reduced ICU mortality and increased the proportion of shock reversal by Day 7 and by Day 28 without increasing the rate of gastroduodenal bleeding, super-infection, or hyperglycemia.25
Our patient was admitted to the medical ICU. After obtaining urine and blood cultures, she was started on IV levofloxacin. She remained hypotensive despite IV fluids and IV norepinephrine. She was started on IV hydrocortisone 50 mg every six hours. Over the next 48 hours, her hemodynamic parameters improved. Urine and blood cultures came back positive for E. coli. Her BUN and creatinine decreased to 24 mg/dL and 1.4 mg/dL, respectively.
Later, she was weaned off norepinephrine and transferred out of the ICU. On hospital Day 7, a slow taper of her hydrocortisone initiated, and antibiotics were switched to oral levofloxacin. She was later discharged home in stable condition.
In patients with septic shock that is unresponsive to IV fluid resuscitation and vasopressors, the addition of low-dose corticosteroids is relatively safe and can improve rate of reversal of shock, reduce time to reversal of shock, decrease ICU length of stay, and potentially lower mortality.
Drs. Gandhi and Asudani are health science assistant professors of medicine in the Division of Hospital Medicine at the University of California at San Diego.
The Hospitalist newsmagazine reports on issues and trends in hospital medicine. The Hospitalist reaches more than 25,000 hospitalists, physician assistants, nurse practitioners, residents, and medical administrators interested in the practice and business of hospital medicine.