Reducing Mortality in Severe Sepsis and Septic Shock

Andrew T. Levinson, M.D., M.P.H.; Brian P. Casserly, M.D.; Mitchell M. Levy, M.D.


Semin Respir Crit Care Med. 2011;32(2):195-205. 

In This Article


Vasopressors are used in patients with septic shock if fluid resuscitation is not successful in restoring arterial pressure and tissue perfusion. The goals of vasopressor therapy are to improve tissue perfusion and cellular derangements caused by septic shock. Different vasopressor have different effects on different adrenergic receptors, resulting in a heterogeneity of physiological effects.[67] Although there has been a long-standing debate,[68] no single pressor has been definitively shown to have a mortality benefit over another in patients with septic shock. To help better answer the question of whether there is a mortality benefit from the initial vasopressor used, the recently published Sepsis Occurrence in Acutely Ill Patients II (SOAP II) study randomized 1679 patients with shock to norepinephrine or dopamine as the initial pressor. The study found no difference between the two groups in 28-day mortality, although there were significantly more cardiac arrhythmias in the dopamine group.[69] The SSC recommends maintaining a mean arterial pressure (MAP) of ≥65 and using either norepinephrine or dopamine as the initial vasopressor in patients with septic shock.[14]

There is also no clear evidence that there is a survival benefit with the administration of low-dose vasopressin. Vasopressin is a peptide hormone released by the pituitary in response to decreased intravascular volume, and decreased levels have been noted in patients with septic shock.[67,70] An RCT with 788 patients in septic shock receiving norepinephrine and low-dose vasopressin (0.01 to 0.03 Units/minute) or norepinephrine alone found no difference in mortality between the two groups.[71] A subsequent retrospective post hoc analysis found that in the subset of patients that received vasopressin and corticosteroids there was an associated reduction in mortality, suggesting the need for a study specifically designed to address the possible synergistic mortality benefit of vasopressin and corticosteroids.[72]


Recombinant Human Activated Protein C (RHAPC)

The administration of recombinant human activated protein C (rhAPC) to patients with a high likelihood of death has been shown to result in a significant reduction in mortality.[73] Protein C, one of our natural anticoagulants, is produced by the liver and activated in the circulation, where it acts by cleavage and inhibition of factors Va and VIIIA.[74] Activated protein C (APC) also appears to exert an indirect antiinflammatory action by inhibition of thrombin generation, and it can also modulate inflammation through direct effects.[75] Conversely, levels of endogenous APC are reduced in sepsis and associated with poor outcome.[36,76,77] In the Recombinant Human Activated Protein C Worldwide Evaluation in Severe Sepsis (PROWESS) study, 1690 patients with severe sepsis were randomized to rhAPC infusion for 96 hours or placebo. The study was stopped early for efficacy after demonstrating an absolute risk reduction of death of 6.1% with the administration of rhAPC.[78] Subsequent subgroup analyses of the PROWESS data found that patients with multisystem organ failure or very high risk of death, as determined by Acute Physiology and Chronic Health Evaluation (APACHE) II Score, had the highest reduction in mortality with the administration of rhAPC.[79] Although the utility of this subgroup analysis remains controversial, the findings led to the current recommendation of rhAPC for patients with sepsis with a high risk of death, as defined by an APACHE II score of 25 or greater or multiorgan failure.[14] An open-label safety study (ENHANCE), Extended Evaluation of Recombinant Human Activated Protein C, found a similar mortality rate and that the early administration of rhAPC within the first 24 hours was associated with greater reductions in mortality.[80] A subsequent study, Administration of Drotrecogin Alfa (Activated) in Early Stage Severe Sepsis (ADDRESS), of 2640 patients with sepsis and a low risk of death randomized to rhAPC or placebo found no reduction in mortality in patients with severe sepsis with a low likelihood of death.[81] More recent evidence for a survival benefit with the administration of rhAPC comes from a nonrandomized, propensity-matched analysis of 33,749 patients with severe sepsis, which found a significant reduction in hospital mortality with the administration of rhAPC.[82] The current SSC guidelines recommend the use of rhAPC for septic patients at high risk of death if there is no contraindication.[14]

Steroids A common feature of sepsis is adrenal insufficiency.[22] About 30% of all critically ill patients show adrenal dysfunction, and this percentage rises to 50 to 60% in septic shock.[83] The presence of adrenal insufficiency in septic shock patients has been associated with worse outcomes, including higher mortality rates and prolonged requirements for vasopressors.[83] However, the benefits of administering exogenous steroids in septic shock remain an enduring controversy. There is good evidence that the administration of high-dose corticosteroids (> 300 mg hydrocortisone per day) has no mortality benefit or may even lead to increased mortality.[84,85,86] The utility of low-dose corticosteroids in reducing mortality is less clear cut (< 300 mg hydrocortisone per day). Annane and colleagues randomized 300 septic patients in 19 French ICUs unresponsive to vasopressor therapy with a positive cosytropin stimulation test to low-dose hydrocortisone and fludrocortisone or placebo. They found that those patients initially unresponsive to vasopressors with a positive cosytropin stimulation test randomized to low-dose hydrocortisone and fludrocortisone had a significantly lower risk of death.[87] In the Corticosteroid Therapy of Septic Shock (CORTICUS) study, a subsequent multicountry trial of 499 patients randomized to low-dose hydrocortisone alone or placebo, there was no reduction in mortality in the group that received hydrocortisone. It is difficult to compare these two trials because the CORTICUS study included patients with septic shock, regardless of whether their blood pressure responded initially to vasopressors.[23] A recent RCT found no improvement in in-hospital mortality by adding fludrocortisone when hydrocortisone is administered to patients with severe sepsis.[83] In addition, two recent systematic reviews found no overall survival benefit by administering low-dose steroids.[11,88] The current SSC guidelines recommend using low-dose hydrocortisone only when there is a poor initial response to vasopressors.[14]

Glucose Control At the beginning of the decade there was significant interest in very tight glucose control in the hopes of reducing mortality of critically ill patients. Most of this interest was generated by the results of a large, single-center study by van der Berghe and colleagues of 1548 cardiac surgery ICU patients randomized either to very tight glucose control (80 to 110 mg/dL) via intensive insulin therapy (IIT) or to conventional treatment (280 to 200 mg/dL).[89] The study found a very significant 3.4% absolute mortality reduction among all patients and a 9.6% absolute risk reduction in mortality for patients who remained in the ICU for greater than 5 days. When the same IIT protocols by the same authors were applied to nonsurgical patients in a medical ICU there was no reduction in mortality.[90] In recent years there has been an accumulation of evidence from RCTs that tight glucose control in nonsurgical, critically ill patients does not have a significant mortality benefit and may in fact be quite harmful. A systematic review of 29 RCTs involving 8432 patients found there was no reduction in mortality with IIT compared with standard treatment.[91] In one study of 532 patients with severe sepsis randomized to IIT or conventional insulin therapy (180 to 200 mg/dL) and various combinations of crystalloids or colloids, there was no difference in mortality with IIT or conventional therapy but a significant increase in severe hypoglycemia in the IIT group. The study was stopped early because of increased significant hypoglycemic events in the IIT group.[92] Another, even larger, study, the Normoglycemia in Intensive Care Evaluation-Survival Using Glucose Algorithm Regulation (NICE-SUGAR) study, randomized 6104 medical and surgical patients within 24 hours of ICU admission to IIT or conventional therapy (180 mg/dL or less). The study found a significantly increased risk of death (OR 1.14) in the group of patients who received IIT. The increased mortality with IIT was found in both surgical and medical patients.[93] The current SSC guidelines recommend a glucose target of <180 mg/dL for patients with severe sepsis and septic shock.

Factors that Affect Outcome

There have now been several studies which have shown a survival benefit of utilizing both the resuscitation and the management bundles of the SSC for patients with severe sepsis and septic shock.[39,94,95] Castellanos-Ortega and colleagues analyzed the outcomes of 384 patients with severe sepsis admitted to a 30-bed ICU at an academic medical center in Spain who received care according to the surviving sepsis bundles. Their outcomes were compared with historical controls. The authors found a significant decrease in mortality with the implementation of the SSC (57 vs 38% mortality). They also found that compliance with six or more components of the resuscitation bundle was an independent predictor of survival.[95] A meta-analysis of studies that implemented bundled care for sepsis also found a significant improvement in survival if bundled care for resuscitation was received.[28] Two large RCTs, the Protocolized Care for Early Septic Shock (ProCESS) trial in the United States and the Australasian Resuscitation In Sepsis Evaluation (ARISE) in Australia, are currently being conducted and may help to better elucidate which components of resuscitation are most important.

Which components of resuscitation and management of patients with severe sepsis and septic shock in the intensive care unit are most important remains controversial. It is unclear whether success in reducing mortality is dependent on individual bundled components or whether implantation of the entire protocol is necessary. Castellanos-Ortega et al, in their study, found that, although improvement in survival was dependent on the number of bundle elements completed, the only single bundle component that separately resulted in reduction in mortality was ScvO2.[95] In contrast, in one meta-analysis that analyzed primarily studies that focused on EGDT and the resuscitation bundles, the timely administration of antibiotics was the only intervention that improved outcome.[28]

In the largest evaluation of the SSC to date, however, components of both the resuscitation bundle and the management bundle were associated with a reduction in mortality. Data from 15,022 subjects at 165 sites participating in the SSC campaign were analyzed. After adjustment for baseline characteristics, administration of broad-spectrum antibiotics (OR 0.86), obtaining blood cultures before their initiation (OR 0.76), and maintaining blood glucose control (OR 0.67) were all significantly associated with lower hospital mortality. Measuring lactate was not associated with improved outcome. The administration of drotrecogin alfa in the first 24 hours was associated with improved survival in those with shock (OR 0.81). For those who required mechanical ventilation, achieving plateau pressure control was associated with improved outcome (OR 0.70). Unadjusted hospital mortality decreased from 37.0% in the first quarter in the Campaign to 30.8% by 2 years (p = 0.001). The longer the involvement of a site in the SSC the greater reduction there was in mortality (Fig. 2).[96] Taken together, there is an increasing body of published, peer-reviewed studies that suggests standardizing management of sepsis through the use of performance metrics, checklists, or protocols can improve survival for these critically ill patients with severe sepsis and septic shock.

Figure 2.

Surviving Sepsis Campaign compliance and mortality change over time. (A) Change in the percentage of patients compliant with all elements of the resuscitation bundle (dotted line) and the management bundle (solid line) over 2 years of data collection (*p < 0.01 compared with first quarter). Note that both y axes are truncated at 40% to emphasize relative change over time as opposed to absolute change. (B) Change in hospital mortality over time (*p < 0.01 compared with first quarter). (From Levy MM, Dellinger RP, Townsend SR, et al. The Surviving Sepsis Campaign: results of an international guideline-based performance improvement program targeting severe sepsis. Critical Care Medicine 2010;38(2):371. Reproduced with permission. With kind permission of Wolters Kluwer Health.)


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.