Septic shock continues to be a significant cause of morbidity and mortality despite the use of broad-spectrum antibiotics, modern intensive care unit (ICU) management, and treatment based on specific guidelines (1-5). Current guidelines stress the importance of early recognition of sepsis, prompt institution of effective antibiotics, and aggressive source control when indicated (6, 7). In addition, studies have emphasized the importance of early goal-directed resuscitation and the use of specific bundles of care designed to improve compliance and early administration of effective therapeutic support strategies (8, 9). After initial fluid resuscitation, the use of vasopressor support to achieve and maintain adequate blood pressure and organ perfusion is the cornerstone of the hemodynamic support strategy (3). Optimum hemodynamic support would be expected to maintain organ perfusion and function, therefore reducing the development of one of the most common causes of death in the ICU, multiple organ dysfunction syndrome (10, 11).
Currently, the optimum vasopressor agent for patients with septic shock remains an area of controversy (7, 12-14). Although almost all agree that vasopressor therapy should not be started until there has been adequate volume resuscitation, currently defined by a central venous pressure (CVP) of 8- to 12-mmHg measurement, there is no current consensus as to the "best vasopressor" drug (3, 7). In 2004, the Society of Critical Care Medicine published consensus guidelines for hemodynamic support in septic shock (3). The guideline recommended initiation of either dopamine (DA) or norepinephrine (NE). If further hemodynamic support is necessary, the recommendation is to add NE (if not initially started), fixed, low-dose vasopressin (0.01 - 0.04 U min−1), phenylephrine, or epinephrine. In 2008, the Surviving Sepsis Campaign pronounced that DA and NE were equally effective first-line agents to use for vasopressor support in septic shock (7).
The dominance of DA began to change as observational trials demonstrated improved outcome with NE use, and at the same time, there were reports of increased rhythm disturbances and lack of a renal protection benefit associated with DA use (13, 15, 16). Two of the studies that suggested a potential survival benefit associated with NE treatment were conducted by Martin et al. (17, 18). One study randomized 32 septic shock patients to receive either DA or NE (17). If the hemodynamic goals were not achieved within 6 h, another agent was added. In the DA group, NE was added. The survival rate was 59% in the NE group compared with 17% with DA treatment; however, the small number of patients limited the strength of this observation (17). In a subsequent observational study of 97 septic shock patients treated with high-dose DA, the patients either continued on high-dose DA with epinephrine or were changed to NE (18). The patients who received NE had a significantly lower mortality rate (62% vs. 82%; P < 0.001) compared with high-dose DA treatment (18). Unfortunately, the treatment choice was not randomized or controlled. The European Sepsis Occurrence in Acutely Ill Patients study of 1,058 patients with shock, of which only 462 had septic shock, reported an increase in mortality associated with DA treatment (15).
Levy et al. (19) reported on a two-phase, prospective, observational study of 110 patients with septic shock. Phase 1 of the study assigned all patients to DA (5-20mcg kg−1 min−1) to achieve and maintain a MAP greater than 70 mmHg within a 30-min window. Phase 2 evaluated those patients who did not achieve the target MAP and were classified as DA-resistant shock. These patients required the addition of either NE or epinephrine to achieve the MAP goal. The DA resistance patients had a significantly higher mortality rate of 78% (52/66) versus 16% (7/44; P = 0.0006) (19).
Dopamine and NE have been the most commonly used initial vasopressors agents in the United States. The objective of this trial was to determine if there was an efficacy or safety benefit to the use of DA versus NE as the initial vasopressor in patients with septic shock. The primary outcome variable was 28-day all-cause mortality between the two treatment strategies.
MATERIALS AND METHODS
This trial was a single-center, prospective, randomized, protocol-controlled, open-label comparison of DA versus NE as the initial vasopressor for adult patients presenting with septic shock. This study took place from September 1, 2003 to June 30, 2008 at Rush University Medical Center, a 750-bed, tertiary-care, academic institution located in Chicago, Ill. The study was approved by the institutional review board for human experimentation. All of the patients in this study were managed in the 14-bed medical ICU (MICU) of Rush University Medical Center. The enrollment criteria are displayed in Table 1. Patients were eligible for inclusion in the study if they were 18 years or older, presented with a diagnosis of systemic inflammatory response syndrome (SIRS) plus a suspected or documented source of infection, and were admitted to the medical ICU (10). Eligible patients had to be in shock (MAP <60 mmHg and/or systolic blood pressure <90 mmHg) after adequate fluid resuscitation, as determined by a CVP greater than 8 mmHg for nonmechanically ventilated patients (12-15 mmHg for patients requiring mechanical ventilation) and require the administration of a vasopressor for management. Patients were excluded from the study if they were found to have alternative causes of their shock (i.e., hypovolemic, hemorrhagic, cardiogenic, anaphylactic, and/or obstructive shock) or another cause of their SIRS. Patients who were allergic to DA or NE were excluded, as were patients who were on vasopressors for more than 6 h before enrollment (see Fig. 1 for study flow). This study was approved by our Institutional Human Use Committee and informed consent was obtained.
Adult patients with septic shock were randomized to receive either DA (5-20 mcg kg−1 min−1) or NE (5-20 mcg min−1) as the first-line vasopressor therapy for septic shock. Randomization was based on whether the patient presented on an odd or even calendar day of the month. For example, if the patient had presented on the third day of the month, then they would be randomized to DA, and if they had presented on the fourth day of the month, then they would be randomized to the NE treatment arm. This randomization scheme was chosen to facilitate rapid institution of assigned vasopressor therapy and to avoid either delay in administration of vasopressor or the need to switch vasopressor therapy associated with use of a more time-consuming randomization system.
All patients were treated according to our sepsis management algorithm that incorporated early goal-directed therapy, including fluid resuscitation, early and appropriate antimicrobial therapy (with a goal of <1 h for administration), source control as indicated, glycemic control (blood glucose concentration <150 mg dL−1), and consideration of steroid replacement for patients with "relative" adrenal insufficiency. Suspected or confirmed septic shock patients were initially resuscitated with either crystalloid or colloid infusions to a CVP greater than or equal to 8 mmHg. If they continued to have a MAP less than 60 mmHg or a systolic blood pressure less than 90 mmHg after adequate fluid resuscitation, they were considered candidates for randomized to receive vasopressor therapy and enrolled in our study. A vasopressor administration protocol (Fig. 2) guided the administration and dosing titration of vasopressor agents to achieve a MAP greater than or equal to 60 mmHg or a systolic pressure greater than or equal to 90 mmHg. If the predetermined maximum dose was reached for the initial vasopressor (DA, 20 mcg kg−1 min−1 or NE, 20 mcg min−1), then the addition of vasopressin at a continuous infusion dose (0.04 U min−1) was initiated. Patients who required additional hemodynamic support to meet the goals were then started on an infusion of phenylephrine (25-200 mcg min−1), which was titrated to reach the goal hemodynamic parameters (Fig. 2). Those patients who were found to have a low central venous oxygen saturation (ScvO2 <70%) were given dobutamine. Activated protein C was used when indicated based on Acute Physiologic and Chronic Health Evaluation (APACHE) II score and patient characteristics on day of entry into the clinical trial. To ensure similarity in ventilator management strategies, patients in the study were managed with our sedation and analgesia, acute lung injury/adult respiratory distress syndrome, and weaning protocols, as indicated.
The primary end point was all-cause 28-day mortality. Secondary end points included length of stay in the ICU, organ dysfunction/failure (as measured by Sequential Organ Failure Assessment [SOFA] score), length of stay in the hospital, and the occurrence of cardiac arrhythmias. Data collected included baseline characteristics (sex, age, underlying illness, site of infection, organism, history of previous cardiac disease or cardiac arrhythmia), laboratory parameters and APACHE II score (20), occurrence of arrhythmias, and survival.
Arrhythmias were defined as sinus tachycardia, atrial fibrillation and flutter, ventricular fibrillation and flutter, and premature ventricular contractions (PVCs). Sinus tachycardia was defined as greater than 20% increase in heart rate from baseline after the start of vasopressor for a period of greater than 8 to 12 h. Atrial arrhythmia was defined as the presence of an abnormal atrial rhythm such as atrial fibrillation, atrial flutter, or paroxysmal supraventricular tachycardia. Ventricular arrhythmia was defined as ventricular tachycardia, ventricular fibrillation, or torsade de pointes. Ventricular tachycardia was recorded if greater than six consecutive beats with a coinciding heart rate greater than 100. Premature ventricular contractions (defined as more than two consecutive PVCs during the course of a nursing shift of 8- to 12-h period) based on electrocardiograph record. All patients were continuously monitored via electrocardiogram leads and the Solar 7000/8000 system, which alerts the nurse via an alarm mechanism for abnormal cardiac rhythms. Patients enrolled in this trial were followed for 28 days, until discharge from the hospital, or death, whichever came first.
The primary outcome variable was 28-day survival. The exposure variable was whether the patient received DA or NE for hemodynamic support in the setting of septic shock. All statistical evaluations were performed using intention-to-treat analysis. The sample size for this study was based on a power analysis that assumed the expected mortality of septic shock was approximately 40% to 60%. We were looking for 20% reduction in mortality rate that would require n = 240 for our sample size to achieve a power of 80% (21). The primary outcome of survival was compared via a chi-square test. Comparison of baseline categorical and continuous data was completed using a chi-square and Student t test, respectively. The continuous secondary outcome parameters such as length of ICU and hospital stay and SOFA score (as a measure of organ dysfunction/failure) were analyzed using a t test. The development of arrhythmias was compared in both groups using a chi-square test. Kaplan-Meier survival curves were constructed and compared via log-rank analysis for the NE and DA treatment groups. Multivariate analysis was performed by constructing a model consisting of variables commonly known to affect mortality and variables significant at the P < 0.2 in univariate analysis with a plausible relationship with morbidity and mortality. All statistical analyses were performed using SPSS software (v15; Chicago, Ill), and statistical significance was judged by a P ≤ 0.05.
A total of 252 consecutive septic shock patients were enrolled in this trial during a 5-year period (Fig. 1). Baseline characteristics at the time of enrollment were balanced in terms of patient demographics, severity of illness (APACHE II scores), multiorgan dysfunction/failure (SOFA scores), immunosuppressive disease states (i.e., cancer), and type of bacterial infection (gram-positive and gram-negative; see Table 2). In addition, the use of corticosteroids and recombinant activated protein C was similar in both groups. The overall mortality from the septic shock was 47% (118/252). The mortality rate in the patients who received DA as the initial vasopressor was 50% (67/134) as compared with 43% (51/118) for NE treatment (P = 0.282). Kaplan-Meier survival curves for the 28-day study period are displayed in Figure 3. The secondary end points of multiorgan dysfunction (as assessed by SOFA scores) and length of stay (ICU and hospital) were not significantly different between the DA and NE treatment arms (Table 3).
A similar number of patients in both groups required the addition of vasopressin and phenylephrine to the initial vasopressor strategy. Importantly, there was a significant difference in the occurrence of arrhythmias between the two vasopressor treatment arms. This difference is reflected in the increased need to change vasopressors in the DA-treated arm of the study (Table 4). The incidence of arrhythmias in the DA-treated group was 38% (51/134) versus 11.8% (14/118) in the NE-treated patients (P < 0.0001). Table 5depicts the arrhythmias that were noted in the two study populations. The presence of a previous cardiac history and/or arrhythmia was not significantly different between the DA and NE treatment arms (P = 0.975 and P = 0.842, respectively; Table 5).
In an attempt to determine the presence of additional predictors of mortality, a multiple logistic regression analysis was performed. The logistic regression analysis demonstrated that the APACHE II score (P < 0.0001) and arrhythmias (P < 0.015) were significant predictors of mortality.
This study was designed to evaluate both the efficacy and safety of NE and DA in the treatment of adults with fluid-resuscitated septic shock. There was no difference observed in the primary end point, 28-day mortality rate. However, the use of DA was associated with significantly increased rate of arrhythmias. The management of septic shock entails many different components that focus on appropriate and timely administration of antimicrobial agents, fluid resuscitation, source control, hemodynamic and respiratory support as indicated, and prevention of common complications of critical illness (3-5, 7). The controversy concerning superiority of one vasopressor strategy over others continues. This trial demonstrated that both DA and NE as initial vasopressors in a vasopressor treatment algorithm lead to similar 28-day survival rates. Despite the similarity in overall survival, NE seemed to be the safer agent as evidenced by a reduced number of cardiac arrhythmias.
The significant difference in the occurrence of arrhythmias between the DA and NE treatment arms does raise some important safety considerations. The increased occurrence of excessive sinus tachycardia, atrial fibrillation, and premature ventricular contractions may potentially pose additional risks for critically ill patients with septic shock requiring vasopressor agents for hemodynamic support. In a study of 1,341 critically ill patients, Annane et al. (22) found sustained arrhythmias occurred in 12% of the population. Supraventricular arrhythmias occurred in 8%, ventricular arrhythmias occurred in 2%, and conduction abnormalities occurred in 2% of the patients. The mortality rate in patients without arrhythmias was 17% but increased to 73% in those patients with ventricular arrhythmias and 60% in those with conduction abnormalities. Both of these arrhythmias significantly increased the risk of death. Neurologic complications were also more likely to occur in those patients who manifested ventricular arrhythmias (22).
The occurrence of increased arrhythmias associated with the use of DA is not a new finding. The Australia-New Zealand Critical Care Trials group found that the use of "renal dose" DA was associated with increased arrhythmias (16). Similarly, Argalious et al. (23) noted an odds ratio of 1.74 for the development of atrial fibrillation when "renal dose" DA was given to patients after coronary artery bypass surgery. However, it should be noted that Levy et al. (19) did not demonstrate an increased risk for cardiac arrhythmias with the use of DA in the trial. The use of NE and epinephrine in the French vasopressor study did not pose an increase risk for the development of cardiac arrhythmias or adverse neurologic or ischemic events (24). The vasopressin study of Russell et al. (25) also reported a low incidence of cardiac arrhythmias. Our finding of arrhythmias in 38% of the DA group in contrast to approximately 12% of the NE group we think is noteworthy and should prompt a change in vasopressor selection. The 12% incidence of arrhythmias in the NE group was similar to those results reported by Annane et al. (22) in their study of sustained arrhythmias in critically ill patients.
Although the recognition of increased cardiac rhythm disturbances associated with DA use did not result in an increase in mortality, this observation raises concern over the continued use of this agent when there are alternative vasopressors that do not seem to pose the same risk. Current Society of Critical Care Medicine guidelines do not recommend one catecholamine over another, but the consensus is that there are physiologic advantages of both medications in the setting of septic shock (7, 26, 27) The mechanism by which DA is thought to precipitate more arrhythmias has been linked to its properties and action on the myocardium via the β1 receptor activity. In addition to increased heart rate, there is also concern for the effects on the hormones controlled by the anterior pituitary and adverse effects from low-dose DA (26, 27).
There are several strengths of this study that we would like to emphasize. First, there have been few prospective, randomized, protocol-controlled trials of NE versus DA in patients with septic shock. Previous trials included small numbers of patients and/or were observational studies that limit their clinical utility (17, 18). Our study prospectively evaluated DA versus NE in 252 patients with fluid-resuscitated septic shock. The study was performed in a single MICU that allowed for a more uniform approach to patient management and one that used a specific vasopressor dosing and titration protocol to minimize variation of management. In addition to the protocol to guide vasopressor management, this study incorporated protocols and "bundles of care" to guide sepsis management, ventilator management, sedation, and glucose control. Specific components of our sepsis protocol included goal-directed fluid administration, early administration of effective antibiotic treatment, assessment of adequacy of source control, adrenal assessment and replacement therapy when indicated, and assessment for use of activated protein C therapy, in addition to the goal-directed vasopressor administration protocol. We also included a safety analysis along with the efficacy evaluation. Because vasopressors may contribute to the development of arrhythmias and other untoward events, we evaluated the patients in this trial for the occurrence of rhythm changes or other cardiac events that might lead to a change in therapy or an adverse outcome.
One of the limitations of this study was the use of a randomization scheme based on the date. Although this form of randomization has the potential to introduce bias, it was thought to be necessary for the successful conduct of this study and to facilitate the early administration of the necessary vasopressor support. A more complex randomization procedure may have produced a delay in the start of necessary vasopressor support and would have made it difficult to begin correct vasopressor support in the emergency room or hospital ward before the patient arrived in the MICU. Despite our use of the even/odd enrollment procedure, it does seem that the randomization was effective, as demonstrated by Table 2.
Another potential limitation of our trial was the use of a single center. However, we would suggest that the external validity of our study was not compromised by this approach because our study population is representative of the typical university or tertiary-care MICU. Our patient population was critically ill with an APACHE II score in the 27 to 28 range. This high level of acuity is higher than typical multicenter trials investigating innovative therapeutic strategies in patients with severe sepsis and septic shock because we only enrolled patients with septic shock and did not exclude patients who had preexisting malignancy and/or immunocompromised state, or chronic organ dysfunction (28, 29). Therefore, it may be argued that our population is more reflective of a true MICU septic shock population as opposed to the patients enrolled in most multicenter pharmaceutical trials.
In addition, the use of a MAP of greater than 60 mmHg or a systolic blood pressure greater than 90 mmHg both represent accepted guideline goals for resuscitation. The use of higher MAP (e.g., >70 mmHg) or SBP (e.g., >100 mmHg), although possibly associated with improved perfusion, has not necessarily been associated with improved outcome (30, 31).
Finally, it may be argued that a systematic search for arrhythmias was not conducted on a daily basis during the infusion of vasopressors. All of the patients enrolled in the trial underwent continuous electrocardiogram and hemodynamic monitoring. Any arrhythmia would have been recognized and printed by the arrhythmia monitoring system, and adverse hemodynamic outcomes were called to the attention of the nursing and medical staff by the alarm system. In addition, it is known that there can be some degree of sinus tachycardia that can result from NE infusions; however, there was a threshold set by the investigators of a 20% increase for a period of 8 to 12 h to stay consistent with the nursing monitoring and working shifts.
In clinical practice, there may be advantages to choosing one vasopressor over another. The results of this study of MICU patients with septic shock demonstrated no significant 28-day survival benefit for NE compared with DA as the initial vasopressor after adequate volume resuscitation. However, the finding of significantly more cardiac arrhythmias associated with DA administration is concerning and should prompt further study and caution when DA is used in patients with a history of cardiac disease or arrhythmia.
The authors thank the nurses and physicians (including the internal medicine interns and residents) who worked in our MICU and their adherence to the authors' vasopressor protocol.
1. Angus DC, Linde-Zwirble WT, Lidicker J, Clermont G, Carcillo J, Pinsky MR: Epidemiology of severe sepsis in the United States: analysis of incidence, outcome, and associated costs of care. Crit Care Med
2. Balk RA: Optimum treatment of severe sepsis and septic shock: evidence in support of the recommendations. Dis Mon
3. Dellinger RP, Carlet JM, Masur H, Gerlach H, Calandra T, Cohen J, Gea-Banacloche J, Keh D, Marshall JC, Parker MM, et al: Surviving Sepsis Campaign Management Guidelines Committee: Surviving Sepsis Campaign guidelines for management of severe sepsis and septic shock. Crit Care Med
4. Wheeler AP: Recent developments in the diagnosis and management of severe sepsis. Chest
5. Russell JA: Management of sepsis. N Engl J Med
6. Kumar A, Roberts D, Wood KE, Light B, Parrillo JE, Sharma S, Suppes R, Feinstein D, Zanotti S, Taiberg L, et al: Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med
7. Dellinger RP, Levy MM, Carlet JM, Bion J, Parker MM, Jaeschke R, Reinhart K, Angus DC, Brun-Buisson C, Beale R, et al: International Surviving Sepsis Campaign Guidelines Committee; American Association of Critical-Care Nurses; American College of Chest Physicians; American College of Emergency Physicians; Canadian Critical Care Society; European Society of Clinical Microbiology and Infectious Diseases; European Society of Intensive Care Medicine; European Respiratory Society; International Sepsis Forum; Japanese Association for Acute Medicine; Japanese Society of Intensive Care Medicine; Society of Critical Care Medicine; Society of Hospital Medicine; Surgical Infection Society; World Federation of Societies of Intensive and Critical Care Medicine: Surviving Sepsis Campaign: International guidelines for management of severe sepsis and septic shock. Crit Care Med
8. Rivers M, Ngyuen B, Havstad S, Ressler J, Muzzin A, Knoblich B, Peterson E, Tomlanovich M: Early Goal-Directed Therapy Collaborative Group: Early goal directed therapy for the treatment of severe sepsis and septic shock. N Engl J Med
9. Nguyen HB, Corbett SW, Steele R, Banta J, Clark RT, Hayes SR, Edwards J, Cho TW, Wittlake WA: Implementation of a bundle of quality indicators for the early management of sepsis and septic shock is associated with decreased mortality. Crit Care Med
10. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA, Schein RM, Sibbald WJ: Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. Chest
11. Balk R: Severe sepsis and septic shock. Crit Care Clin
12. Hollenberg SM: Vasopressor support in septic shock. Chest
13. Mullner M, Vrbank B, Havel C, Losert H, Waechter F, Gamper G: Vasopressors for shock. Cochrane Database Syst Rev
14. Beale RJ, Hollenberg SM, Vincent JL, Parrillo JE: Vasopressor and inotropic support in septic shock: an evidence-based review. Crit Care Med
15. Sakr Y, Reinhart K, Vincent JL, Sprung CL, Moreno R, Ranieri VM, De Backer D, Payen D: Does dopamine administration in shock influence outcome? Results of the Sepsis Occurrence in Acutely Ill Patients (SOAP) Study. Crit Care Med
16. Australian and New Zealand Intensive Care Society (ANZICS) Clinical Trials Group: Low dose dopamine in patients with early renal dysfunction: a placebo controlled trial. Lancet
17. Martin C, Papazian L, Perrin G, Saux P, Gouin F: Norepinephrine or dopamine for the treatment of hyperdynamic septic shock. Chest
18. Martin C, Viviand X, Leone M, Thirion X: Effect of norepinephrine on outcome of septic shock. Crit Care Med
19. Levy B, Dusang B, Annane D, Gibot S, Bollaert PE: College Interregional des Réanimateurs du Nord-Est: Cardiovascular response to dopamine and early prediction of outcome in septic shock: a prospective multi-center study. Crit Care Med
20. Knaus WA, Draper EA, Wagner DP, Zimmerman JE: APACHE II: a severity of disease classification system. Crit Care Med
21. Dupont WD, Plummer WD: Power and sample size calculations: a review and computer program. Controlled Clin Trials
22. Annane D, Sebille V, Duboc D, Le Heuzey JY, Sadoul N, Bouvier E, Bellissant E: Incidence and prognosis of sustained arrhythmias in critically ill patients. Am J Respir Crit Care Med
23. Argalious M, Motta P, Khandwala F, Samuel S, Koch CG, Gillinov AM, Yared JP, Starr NJ, Bashour CA: "Renal dose" dopamine is associated with the risk of new-onset atrial fibrillation after cardiac surgery. Crit Care Med
24. Annane D, Vignon P, Renault A, Bollaert PE, Charpentier C, Martin C, Troché G, Ricard JD, Nitenberg G, Papazian L, et al: CATS Study Group: Norepinephrine plus dobutamine versus epinephrine alone for the management of septic shock: A randomized trial. Lancet
25. Russell JA, Walley KR, Singer J, Gordon AC, Hébert PC, Cooper DJ, Holmes CL, Mehta S, Granton JT, Storms MM, et al: VASST Investigators: Vasopressin versus norepinephrine infusion in patients with septic shock. N Engl J Med
26. Hollenberg SM, Aherns TS, Annane D, Astiz ME, Chalfin DB, Dasta JF, Heard SO, Martin C, Napolitano LM, Susla GM, et al: Practice parameters for hemodynamic support of sepsis in adult patients: 2004 update. Crit Care Med
27. Overgaard CB, Dzavik V: Inotropes and vasopressors: review of physiology and clinical use in cardiovascular disease. Circulation
28. Bernard GR, Vincent JL, Laterre PF, LaRosa SP, Dhainaut JF, Lopez-Rodriguez A, Steingrub JS, Garber GE, Helterbrand JD, Ely EW, et al: Recombinant Human Protein C Worldwide Evaluation in Severe Sepsis (PROWESS) Study Group: Efficacy and safety
of recombinant human activated protein C for severe sepsis. N Engl J Med
29. Warren BL, Eid A, Singer P, Pillay SS, Carl P, Novak I, Chalupa P, Atherstone A, Pénzes I, Kübler A, et al: KyberSept Trial Study Group: High-dose antithrombin III in severe sepsis. JAMA
30. Bourgoin A, Leone M, Delmas A, Garnier F, Albanèse J, Martin C: Increasing mean arterial pressure in patients with septic shock: effects on oxygen variables and renal function. Crit Care Med
31. LeDoux D, Astiz ME, Carpati CM, Rackow EC: Effects of perfusion pressure on tissue perfusion in septic shock. Crit Care Med