http://jic.sagepub.com/Journal of Intensive Care Medicine

http://jic.sagepub.com/content/early/2012/08/17/0885066612453025The online version of this article can be found at:

 DOI: 10.1177/0885066612453025

published online 17 August 2012J Intensive Care MedYang and Emanuel P. Rivers

Davis, Bonnie Turman, Jordan Weingarten, Truman J. Milling, Jr, Nathan Lidsky, Victor Coba, Arturo Suarez, James J. Chad M. Cannon, Christopher V. Holthaus, Marc T. Zubrow, Pat Posa, Satheesh Gunaga, Vipul Kella, Ron Elkin, Scott

Improvement CollaborativeThe GENESIS Project (GENeralized Early Sepsis Intervention Strategies) : A Multicenter Quality

  

Published by:

http://www.sagepublications.com

can be found at:Journal of Intensive Care MedicineAdditional services and information for    

  http://jic.sagepub.com/cgi/alertsEmail Alerts:

 

http://jic.sagepub.com/subscriptionsSubscriptions:  

http://www.sagepub.com/journalsReprints.navReprints:  

http://www.sagepub.com/journalsPermissions.navPermissions:  

What is This? 

- Aug 17, 2012OnlineFirst Version of Record >>

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

The GENESIS Project (GENeralizedEarly Sepsis Intervention Strategies):A Multicenter Quality ImprovementCollaborative

Chad M. Cannon, MD1, Christopher V. Holthaus, MD2, Marc T. Zubrow, MD3,Pat Posa, RN, BSN, MSA4, Satheesh Gunaga, DO5, Vipul Kella, MD5,Ron Elkin, MD6, Scott Davis, MD7, Bonnie Turman, BSN, RN8,Jordan Weingarten, MD9, Truman J. Milling, Jr, MD9, Nathan Lidsky, MD10,Victor Coba, MD11, Arturo Suarez, MD11, James J. Yang, PhD11, andEmanuel P. Rivers, MD, MPH11

AbstractBackground: Improved outcomes for severe sepsis and septic shock have been consistently observed with implementation ofearly best practice intervention strategies or the 6-hour resuscitation bundle (RB) in single-center studies. This multicenter studyexamines the in-hospital mortality effect of GENeralized Early Sepsis Intervention Strategies (GENESIS) when utilized in commu-nity and tertiary care settings. Methods: This study was comprised of 2 strategies to assess treatment. The first was a prospectivebefore-and-after observational comparison of historical controls to patients receiving the RB after implementation of GENESIS in4 community and 4 tertiary hospitals. The second was a concurrent examination comparing patients not achieving all componentsof the RB to those achieving all components of the RB in 1 community and 2 tertiary care hospitals after implementation of GEN-ESIS. These 4 subgroups merged to comprise a control (historical controls treated before GENESIS and RB not achieved afterGENESIS) group and treatment (patients treated after GENESIS and RB achieved after GENESIS) group for comparison. Results:The control group comprised 1554 patients not receiving the RB (952 before GENESIS and 602 RB not achieved after GENESIS).The treatment group comprised 4801 patients receiving the RB (4109 after GENESIS and 692 RB achieved after GENESIS).Patients receiving the RB (treatment group) experienced an in-hospital mortality reduction of 14% (42.8%-28.8%, P < .001) anda 5.1 day decrease in hospital length of stay (20.7 vs 15.6, P < .001) compared to those not receiving the RB (control group). Similarmortality reductions were seen in the before-and-after (43% vs 29%, P < .001) or concurrent RB not achieved versus achieved(42.5% vs 27.2%, P < .001) subgroup comparisons. Conclusions: Patients with severe sepsis and septic shock receiving the RB incommunity and tertiary hospitals experience similar and significant reductions in mortality and hospital length of stay. These find-ings remained consistent when examined in both before-and-after and concurrent analyses. Early sepsis intervention strategiesare associated with 1 life being saved for every 7 treated.

1 University of Kansas Hospital, Kansas City, KS, USA2 Washington University, Barnes-Jewish Hospital, St Louis, MO, USA3 Christiana Care Health System, Newark, DE, USA4 St. Joseph Mercy Hospital, Ann Arbor, MI, USA5 Henry Ford Hospital–Wyandotte, Wyandotte, MI, USA6 California Pacific Medical Center, San Francisco, CA, USA7 St Cloud Hospital, St. Cloud, MN, USA8 Porter Hospital, Valparaiso, IN, USA9 University Medical Center at Brackenridge, Austin, TX, USA10 Northwest Community Hospital, Arlington Heights, IL, USA11 Henry Ford Hospital, Main Campus, Detroit, MI, USA

Corresponding Author:

Christopher V. Holthaus, Division of Emergency Medicine, Washington University School of Medicine, Campus Box 8072, 660 S Euclid Ave, St. Louis, MO 63110, USA

Email: holthausc@wusm.wustl.edu

Journal of Intensive Care Medicine00(0) 1-14ª The Author(s) 2012Reprints and permission:sagepub.com/journalsPermissions.navDOI: 10.1177/0885066612453025http://jicm.sagepub.com

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

Keywordsearly goal-directed therapy, resuscitation bundle, sepsis, severe sepsis, septic shock, quality improvement, emergency medicine,critical care, shock

Introduction

There are more than 500,000 cases of severe sepsis and septic

shock annually in the United States, with a mortality of 20%and 45%, respectively.1 In the last decade, hospital costs have

increased 183%, faster than any cause for hospitalization with

an annual cost of over $54 billion.2 The majority of patients

with sepsis are identified in the emergency department (ED)

with the remainder coming from other inpatient settings.3

Improved survival rates with diseases of similar volume

such as stroke, acute myocardial infarction, and trauma have

been achieved through early intervention strategies comprising

early detection, risk stratification, and rapid therapeutic inter-

ventions. In 2001, a similar strategy for severe sepsis and shock

termed early goal-directed therapy (EGDT) resulted in an in-

hospital absolute mortality reduction of 16% when compared

to controls receiving standard care.4 Combining EGDT with

other early evidence-based strategies has resulted in the creation

of the sepsis resuscitation bundle ([RB] Figure 1).5 Over the last

decade, numerous studies using a before-and-after implementa-

tion of the RB have shown similar reductions in mortality, organ

failure, and health care resource consumption.6

The purpose of this multicenter collaborative was to exam-

ine the impact of real-time quality initiatives in the form of

GENeralized Early Sepsis Intervention Strategies (GENESIS)

on in-hospital mortality, morbidity, and health care resource

consumption in community and tertiary care hospitals in the

United States.

Methods

Institutional Review Board Approval

The GENESIS was approved by the institutional review board

at each participating center (Appendix A).

Study Setting

The GENESIS was a continuous quality improvement (CQI)

initiative implemented at 5 community and 6 tertiary US

hospitals. The GENESIS is comprised of a comprehensive

strategy which includes (1) an institutional assessment of the

sepsis prevalence and mortality, (2) identification of high-

risk patients or a sepsis alert,7 (3) mobilization of resources,

(4) timely intervention of the 6-hour sepsis bundle via a sepsis

team or sepsis order sets, (5) quality indicators to assess com-

pliance, (6) quantification of health care resource consumption,

(7) assessment of outcomes, and (8) a CQI program which

includes feedback and continuing education.8

Study Design

Evaluation of GENESIS consisted of 2 assessment strategies.

The first was a before-and-after RB implementation strategy

conducted in 8 hospitals, analyzing outcomes between histori-

cal controls and patients after RB implementation. The second

was to assess the impact of complete versus incomplete bundle

compliance through examining a concurrent RB implementa-

tion strategy in 3 hospitals (Figure 2). These 4 groups gave rise

to the control and treatment groups for comparison. The control

group comprised patients not receiving the RB before imple-

mentation or group I (historical controls-IA and concurrent

RB not achieved-IB). The treatment group comprised

patients receiving the RB or group II (after implementation-

IIA and concurrent RB achieved-IIB; Figure 2).

Participants

Patients were routinely screened in the ED, general floor

(general practice unit [GPU]), operating room, and intensive

Sepsis Resuscitation Bundle (6 - hour bundle)

1. Suspected infection:Confirmed or suspected source with� 2 systemic inflammatory response syndrome criteria with persistent hypotension (systolic bloodpressure [SBP] < 90 mm Hg or mean arterial pressure [MAP] < 65 mm Hg), lactate � 4 mmol/L, vasopressor use, or organ dysfunction.

2. Serum lactate measured3. Blood cultures obtained before antibiotic administration4. Broad-spectrum antibiotics, from the time of presentation, administered within 3 hours for ED admissions and 1 hour for non-ED ICU

admissions5. In the event of hypotension and/or lactate � 4 mmol/L (36 mg/dL)� Deliver an initial minimum of 20 mL/kg of crystalloid (or colloid equivalent)� Apply vasopressors for hypotension not responding to initial fluid resuscitation to maintain mean arterial pressure > 65 mm Hg

6 In the event of persistent hypotension despite fluid resuscitation (septic shock) and/or lactate � 4 mmol/L (36 mg/dL)� Achieve central venous pressure of � 8 mm Hg� Achieve central venous oxygen saturation (ScvO2) of � 70%

Figure 1. Sepsis resuscitation bundle elements.

2 Journal of Intensive Care Medicine 00(0)

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

care units (ICUs) based on regional practices. Historical

controls were found via chart reviews using International

Classification of Diseases, Ninth Revision codes for severe

sepsis (995.92) or septic shock (785.52). Inclusion criteria were

a sepsis diagnosis with a lactate �4 mmol/L, vasopressor

use, or organ dysfunction (Figure 1). Exclusion criteria were

age <18 years, shock suspected secondary to cause/causes other

than sepsis, or advanced directives which compromised the

intended care.

Statistical Analysis

Data are presented as the mean + standard deviation.

Continuous and categorical variables were analyzed with a 2-

tailed Student t test, Mann-Whitney U test, Kruskal-Wallis,

or Chi-square test as appropriate. Multivariate logistic regres-

sion with risk-adjusted odds ratios (ORs) and 95% confidence

intervals was conducted to eliminate confounding between

variables and multiple test performance. A P value <.05 was

statistically significant. Statistical analysis was performed by

biostatisticians from Washington University, Henry Ford

Hospital, and the University of Kansas Medical Center.

Results

The size of the community hospital beds ranged from 301 to

431 beds, with 33,000 to 95,000 annual ED visits. Tertiary

hospital size ranged from 440 to 1200 beds with 46,000 to

160,000 annual ED visits.

Patient Groups

There were 6,624 patients assessed for eligibility. Due to missing

mortality data, 59 (3.7%) and 210 (4.3%) patients were excluded

from group I and group II, respectively. Group I, the control group

(n ¼ 1,554) consisted of patients before RB implementation or

group IA (n ¼ 952) and concurrent RB not achieved or group

IB (n¼ 602). Group II, the treatment group (n¼ 4,801), consisted

of patients after RB implementation or group IIA (n¼ 4,109) and

concurrent RB achieved or group IIB (n ¼ 692; Figure 2).

Patients assessed for eligibilityN=6624

Excluded from analysis(mortality missing)

N=269 (4.1%)

Group-IBRB not achieved

N=602Mortality 42.5%

Group-IIBRB achieved

N=692Mortality 27.2%

GENESISImplemented in 11 Hospitals

Group-IIAAfter implementation

RB receivedN=4109

Mortality 29%

Group-IABefore implementation

RB not receivedN=952

Mortality 43%

Concurrent Implementation3 Hospitals

N=1294

Before and After Implementation8 Hospitals

N=5061 ControlGroup-I A&B

N=1554Mortality 42.8%

TreatmentGroup-II A&B

N=4801Mortality 28.8%

Figure 2. The GENeralized Early Sepsis Intervention Strategies (GENESIS) flow diagram.Abbreviations: RB, Resuscitation Bundle

Cannon et al 3

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

Presentation Characteristics

Group II patients were significantly older (2.8 years) and had

more females (3%) and less Caucasians (5.2%) when compared

to group I (all P < .03). Group II had a significantly higher

percentage of patients originating from the ED (15.8%) and a

lower percentage from the GPU (8.3%) and ICUs (7.5%; all

P < .006). Group II also had a significantly higher temperature

(0.15�C, P < .001). In both groups, the lung was the most

common source of infection followed by urinary, gastrointest-

inal, blood, catheter, and other sources (Table 1).

Baseline Resuscitation Parameters and Organ Function

In group II patients, the baseline serum bicarbonate was

0.5 mEq/L higher (P ¼ .047), central venous oxygen saturation

[ScvO2] was 2.6% higher (P ¼ .01), and lactate was

0.42 mmol/L lower (P ¼ .006). There were no significant dif-

ferences in the percentage of patients presenting with baseline

hypotension (systolic blood pressure [SBP] <90 mm Hg or

mean arterial pressure (MAP) <65 mm Hg) or baseline lactate

�4 mmol/L between groups (Table 1). Group II baseline organ

dysfunction or Acute Physiology and Chronic Health Evalua-

tion (APACHE)-II scores were 11.2% higher (P < .001) and

baseline Sequential Organ Failure Assessment (SOFA) scores

were 12.9% higher (P ¼ .002) than group I (Table 2).

Resuscitation Parameters at 6 and 24 Hours

At 6 hours, group II had a significantly higher SBP (2.7 mm

Hg, P ¼ .047), ScvO2 (4.4%, P < .001), lower lactate

(0.67 mmol/L, P ¼ .002), greater urine output (10.6 mL/h,

P ¼ .023), and greater lactate clearance (19%, P ¼ .044;

Table 1). No significant differences existed for MAP or central

venous pressure (CVP) between groups (Table 1). At 24 hours,

group II had a significantly lower heart rate (11.7 beats/min,

P < .001), 8 mm Hg higher MAP (P < .001), 0.10 units lower

shock index (P ¼ .002), 0.7 mEq/L higher bicarbonate level

(P ¼ .044), 0.75 mmol/L lower lactate (P < .001), and 26%greater lactate clearance (P ¼ .003; Table 1).

Organ Function Over the First 24 Hours

Group II APACHE-II scores significantly decreased by 15.7%compared to an 11.7% increase in group I over 24 hours

(P < .001). Similarly, the SOFA scores significantly decreased

by 15.4% in group II compared to an 18.6% increase in group I

over 24 hours (P < .001; Table 2). The absolute improvement in

organ dysfunction at 24 hours in group II compared to group I

was 27.4% (APACHE-II) and 34% (SOFA).

Mortality Overall

There was a significant 14.0% absolute and a 32.7% relative

risk reduction for in-hospital mortality (42.8% vs 28.8%,

P < .001) between groups I and II. Similar and significant in-

hospital mortality reductions were seen between groups IA and

IIA (14%) and between groups IB and IIB (15.3%), both

P < .001 (Figure 2 and Table 2). A cumulative examination

of mortality showed a consistent decrease in mortality from

2003 to 2009 (Figure 3). Multivariate regression was used to

control for baseline differences in age, sex, race, origin at pre-

sentation, temperature, bicarbonate, lactate—0 hour, ScvO2—0

hour, APACHE-II score—0 hour, and SOFA score—0 hour. Of

these variables, the SOFA score—0 hour remained signifi-

cantly associated with in-hospital mortality with an adjusted

OR of 1.14 (1.03-1.27, P ¼ .01). Overall, the treatment group

(group II) had a significantly greater likelihood of death at

baseline compared to those not receiving the RB (group I).

Mortality by Hemodynamic Subgroups

Group II consistently showed statistically significant lower mor-

talities across all hemodynamic subgroup strata of lactate levels

and/or hypotension (SBP <90 mm Hg or MAP <65). Group II

absolute in-hospital mortality reductions for the following sub-

group strata are as follows (all P < .02): lactate �4 mmol/L,

18.4%; hypotensive, 17.7%; lactate �4 mmol/L or hypotensive,

15.7%; lactate �4 mmol/L and hypotensive, 22.2%; lactate

�4 mmol/L and no hypotension, 17.9%; and lactate �2 mmol/

L and hypotension, 5.7% (Table 2).

Mortality by Hospital Location and Type

The absolute in-hospital mortality reduction in group II versus

group I for patients diagnosed in the following locations was

12.1% for the ED (P < .001), 15.4% for the GPU (P ¼ .006),

and 13.9% for the ICU (P < .001; Table 2). Participating cen-

ters experienced an overall absolute in-hospital mortality

reduction ranging from 8.1% to 25.7% and a relative risk

reduction from 21.3% to 48.9% (Table 3).

Duration of Mechanical Ventilation, Length of HospitalStay, and Hospital Charges

Group II had significantly less mechanical ventilation (5.0%)

from 0 to 6 hours (P ¼ .02) and 6.2% less from 6 to 72 hours

(P ¼ .02). There was a 1.3-day trend toward a shorter duration

of mechanical ventilation in group II (P ¼ .06). There was no

significant difference in mean ED length of stay. Mean hospital

length of stay was significantly shorter (5.1 days) and hospital

charges were $47,923 less in group II (both P < .001; Table 2).

Time to Completion of Therapeutic Interventions

Time to completion of interventions and reaching hemody-

namic targets in group II versus group I comparisons were sig-

nificantly reduced (hours) for a fluid challenge (1.18),

obtaining a lactate (3.07), antibiotic administration (1.21),

reaching the CVP target (2.98), and reaching the ScvO2 target

(1.95; all P � .002). There was a trend in decreasing the time

(0.77 hours) to achieving the MAP target (P ¼ .05; Table 1).

4 Journal of Intensive Care Medicine 00(0)

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

Table 1. Patient Demographic and Resuscitation Characteristics Group I and Group II

Characteristic

Group I Group II

P ValueN Value (+SD) N Value (+SD)

DemographicsTotal number of patients 1554 4801Age, y 1524 64 (17) 4776 62 (17) <.001Sex, % female 746 48 2164 45 <.001Body weight, kg 838 82 (28) 2061 81 (27) NSRace, % White/Other 1554 75.1 4801 80.3 .003

Origin at presentation, %Emergency department 558 50.1 1987 65.9 <.001General practice unit 139 15.4 168 7.1 .006Intensive care unit 331 34.5 697 27.0 <.001

SIRS characteristicsTemperature, �C 1070 37.1 (1.6) 2207 37.3 (1.6) .008Heart rate, beats/min 1034 107.5 (24.7) 1916 109.1 (26.1) NSRespiratory rate, breaths/min 1096 24.1 (8.4) 2290 23.6 (8.2) NSWhite blood cell count 1081 16.5 (14.1) 2228 15.9 (12.6) NS

Source of infection and culture results, %Lung 208 42.5 492 38.7 NSUrinary 111 22.7 281 22.1 NSGastrointestinal 67 13.7 215 16.9 NSBlood 43 8.8 98 7.7 NSCatheter 16 3.3 33 2.6 NSOther sourcesa 44 9.0 153 12.0 NS

Positive cultures that were sent, % 336 50.4 656 53.2 NSPositive blood cultures that were sent, % 224 37.7 430 38.4 NSLaboratories—baseline, 0 hour

Hemoglobin, g/dL 235 10.9 (2.5) 830 11.3 (2.7) NSPlatelet count, �1000 per cubic mL 1042 230.6 (138.7) 1784 230.5 (137.0) NSBUN/creatinine 806 19.6 (11.6) 1846 19.1 (12.2) NSTotal bilirubin, mg/dL 748 2.1 (3.5) 1281 1.8 (3.1) NSAlbumin, g/dL 629 2.50 (0.77) 1289 2.51 (0.76) NSArterial pH 814 7.32 (0.15) 1125 7.31 (0.15) NSBicarbonate, mEq/L 690 20.5 (6.5) 1291 21.0 (6.0) .047

Baseline hemodynamic variablesLactate, mmol/L 907 4.8 (4.1) 2608 4.4 (3.6) .006Systolic BP, mm Hg 896 97.8 (27.71) 2287 99.6 (29.0) NSMAP, mm Hg 1055 68.7 (19.9) 2054 69.7 (20.6) NSCVP, mm Hg 355 10.5 (6.9) 1096 10.8 (6.6) NSScvO2, % 318 66.9 (17.0) 985 69.5 (13.8) .02Heart rate, beats/min 1034 107.5 (24.7) 1916 109.1 (26.1) NSShock index (heart rate/systolic BP) 813 1.17 (0.41) 1902 1.16 (0.43) NS% with systolic BP <90 or MAP <65 (%) 351 56.16 775 51.91 NS% with lactate �4-0 hour (%) 421 46.42 1122 43.02 NS

6 hours hemodynamic variablesLactate, mmol/L 414 3.84 (3.58) 893 3.17 (3.28) .002Systolic BP, mm Hg 446 109.1 (23.2) 1014 111.7 (23.7) .05MAP, mm Hg 771 72.5 (15.3) 1975 73.4 (16.4) NSCVP, mm Hg 311 12.2 (5.8) 850 11.7 (5.1) NSScvO2, % 242 69.3 (12.5) 740 73.8 (10.4) <.001Urine output, ml/hour (0-6 hours) 401 65.73 (78.0) 989 76.3 (80.6) .03Lactate clearance (0-6 hours) 404 0.02 (1.83) 857 0.21 (0.87) .04

24 hour hemodynamic variablesLactate, mmol/L 257 3.16 (3.01) 414 2.41 (2.55) <.001Systolic BP, mm Hg 130 113.2 (21.3) 343 117.9 (21.0) .03MAP, mm Hg 409 64.8 (19.5) 419 72.8 (17.1) <.001Heart rate, beats/min 413 113.4 (26.9) 462 101.7 (23.9) <.001Shock index (heart rate/systolic BP) 130 0.99 (0.31) 182 0.89 (0.28) .001pH 286 7.31 (0.15) 612 7.32 (0.13) NS

(continued)

Cannon et al 5

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

Antibiotic Administration and Culture Rates

Group II had 5.2% less antibiotic appropriateness (93.2% vs

88.1%, P ¼ .004) with a 2.13-day longer duration of antibio-

tics, P < .001. There were no differences in the overall culture

and primary blood culture positivity rates between groups

(Table 1).

Fluid therapy, Vasopressors, Inotropes, and Red BloodCell Transfusions

Group II received 0.2, 0.8, and 1.1 L more fluid during the 0 to

6, 7 to 72, and 0 to 72 time periods, respectively (all P < .001).

There was no significant difference in vasopressor use during

the 0- to 6-hour time period; however, there was a 15.7%absolute reduction in vasopressor use in group II over the 0-

to 72-hour time period (P < .001). There was no significant

difference in the use of inotropes or red blood cell transfu-

sions between groups (Table 1).

Analysis of Interventions on Mortality

The overall individual RB bundle element compliance was

67.5% in the concurrent group (groups IB and IIB). Obtaining

a lactate (OR ¼ 2.18, P < .001), achieving a MAP �65 mm Hg

(OR ¼ 0.59, P < .001), and achieving a ScvO2 �70%(OR ¼ 0.75, P ¼ .048) were significantly associated with in-

hospital mortality (Table 4).

Discussion

The GENESIS is associated with significantly decreased in-

hospital mortality. The impact on in-hospital mortality was

equally observed in a before-and-after and concurrent obser-

vational analysis based on completing the RB. These results

are similar and consistent with the previous studies.9-14

These ‘‘real-life’’ salutary effects are seen even after inclu-

sion of comorbidities (ie, cancer, end-stage renal, and liver

disease) which could potentially diminish the treatment

effect.

Table 1. (continued)

Characteristic

Group I Group II

P ValueN Value (+SD) N Value (+SD)

Bicarbonate, mEq/L 328 20.5 (5.7) 660 21.1 (5.6) .04BUN/creatinine 548 20.1 (12.5) 1101 19.0 (11.5) .08Lactate clearance (0-24 hours) 216 0.11 (1.12) 234 0.39 (0.50) <.001

Time from diagnosis to intervention, hLactate obtained 494 7.09 (11.38) 1148 4.02 (8.37) <.001Blood cultures 632 7.7 (12.1) 1706 6.1 (10.4) .002Antibiotics 436 2.87 (5.11) 863 1.66 (3.28) <.001Fluid challenge, 20-40 mL/kg 373 3.51 (5.48) 1141 2.33 (4.35) <.001MAP >65 mm Hg 564 6.04 (7.9) 1182 5.27 (6.80) .05CVP >8 mm Hg 302 8.27 (10.96) 912 5.29 (7.27) <.001ScvO2 >70% 219 8.61 (8.52) 808 6.66 (7.22) <.002

Antibiotic and culturesCorrectness of antibiotics, % 386 93.2 799 88.1 .004Duration of antibiotics, days 499 9.73 (10.37) 793 11.86 (11.04) .001

Fluid administration, L0-6 hours 854 2.5 (1.8) 2029 2.7 (1.7) <.0017-72 hours 667 7.6 (5.1) 1415 8.4 (5.6) <.0010-72 hour total fluid volume 642 10.1 (5.6) 1393 11.2 (6.2) <.001

Vasopressor useAdministered 0-6 hours, % patients 410 47.5 825 44.2 NSAdministered 0-72 hours, % patients 663 88.1 1178 72.4 <.001

Inotropic therapy, % of patientsAdministered 0-6 hours 64 9.9 192 11.6 NSAdministered 6-72 hours 85 13.5 179 12.1 NS

PRBC transfusionTotal used 0-72 hours, units 254 1.3 (2.1) 524 1.3 (1.9) NS

Glucose levels, mg/dLBaseline 908 158.1 (116.9) 1806 167.6 (132.7) NS24 hour 582 133.3 (65.7) 1388 130.3 (60.8) NS

Cortisol level, mg/dL 413 37.3 (37.1) 640 39.6 (34.9) NS

Abbreviations: BP, blood pressure; BUN, blood urea nitrogen; CVP, central venous pressure; MAP, mean arterial pressure; NS, not significant; ScvO2, centralvenous oxygen saturation; SD, standard deviation.a Musculoskeletal, central nervous system, endocarditis, ear nose and throat, obstetrics and gynecology, and wound sources of infection.

6 Journal of Intensive Care Medicine 00(0)

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

As with comparable diseases such as acute myocardial

infarction, stroke, and trauma, a standard operating procedure

that includes early detection, risk stratification, and early inter-

vention decreases morbidity and in-hospital mortality. The

15.8% increase (from 50.1% to 65.9%) in the patients identified

in the ED resulted in a significant decrease in patients diagnosed

in the GPU and ICU. These patients (ICU and GPU) usually face

a 3 times higher mortality risk if they develop septic shock.15

This earlier identification was accompanied by a significant

GPU absolute in-hospital mortality reduction (48.9%-33.5%)

followed by the ICU (45.2%-31.3%) and the ED (35.9%-

23.8%, all P � .006; Table 2).

The GENESIS resulted in a significant decrease in time-

to-fluid challenge, lactate measurement, antibiotic therapy,

and hemodynamic target attainment. Although the total

Table 2. Mortality, Organ Dysfunction and Health Care Resource Consumption

Group I Group IIRelative Risk

(95% CI) P ValueNa Value (%) Na Value (%)

In-hospital mortalityGroup I vs group II 665 42.8 1383 28.8 0.67 (0.63-0.72) <.001Group IA vs IIA (before vs after implementation of RB) 409 43.0 1192 29.0 0.68 (0.62-0.74) <.001Group IB vs IIB (concurrent RB not achieved vs achieved) 256 42.5 188 27.2 0.64 (0.55-0.74) <.001

Mortality by sepsis inclusion category (at baseline)Lactate �4 mmol/L 229 54.4 404 36.0 0.66 (0.59-0.74) <.001SBP <90 mm Hg or MAP <65 243 47.3 357 29.6 0.63 (0.55-0.71) <.001SBP <90 or MAP <65 mm Hg or lactate �4 mmol/L 362 47.2 596 31.5 0.67 (0.60-0.74) <.001SBP <90 or MAP <65 mm Hg and lactate �4 mmol/L 222 48.0 343 30.4 0.63 (0.56-0.72) <.001SBP >90 or MAP >65 mm Hg and lactate �4 mmol/L 203 42.8 312 24.9 0.58 (0.50-0.67) <.001SBP <90 or MAP <65 mm Hg and lactate <2 mmol/L 286 52.2 662 46.5 0.89 (0.81-0.99) .02

Mortality by hospital locationED 208 35.9 486 23.8 0.66 (0.58-0.75) <.001GPU 69 48.9 55 33.5 0.66 (0.50-0.87) .006ICU 150 45.2 218 31.3 0.69 (0.59-0.81) <.001

Mortality by centers, %Tertiary care 522 43.2 1006 29.0 0.66 (0.61-0.72) <.001Community 143 41.2 375 29.3 0.71 (0.61-0.83) <.001

Organ dysfunction (SD)Baseline APACHE score 483 20.6 (7.6) 806 22.9 (8.0) – <.00124-Hour APACHE score 482 23.0 (8.5) 409 19.3 (7.3) – <.001Baseline SOFA score 406 7.0 (3.8) 299 7.9 (4.0) – .00224-Hour SOFA score 407 8.3 (4.2) 296 6.5 (3.3) – <.001Baseline PaO2/FiO2 577 264.6 (158.6) 820 262.9 (158.9) – NS24-Hour PaO2/FiO2 348 240.7 (135.5) 630 298.9 (158.2) – <.001

Duration of mechanical ventilation –0-6 hours, % of patients 281 38.5 549 33.5 – .026-72 hours, % of patients 326 50.9 433 44.7 – .02Duration, days (SD) 391 9.7 (12.5) 933 8.4 (9.8) – .06

Length of stay (SD) –ED, hours 414 5.7 (4.5) 1584 5.4 (4.3) – NSHospital length of stay, days 1554 20.7 (30.7) 4809 15.6 (20.0) – <.001

Financial costs (SD)Hospital charges, $ 723 143,949 (188 295) 1182 96,026 (141 139) – <.001

Abbreviations: APACHE, Acute Physiology and Chronic Health Evaluation study; CI, confidence interval; ED, emergency department; FiO2, fraction of inspiredoxygen; GPU, general practice unit; ICU, intensive care unit; MAP, mean arterial pressure; PaO2, partial pressure of arterial oxygen; RB, resuscitation bundle; SBP,systolic blood pressure; SD, standard deviation; SOFA, Sequential Organ Failure Assessment.a For mortality N equals the number of deaths.

Figure 3. Cumulative mortality over the study duration.

Cannon et al 7

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

amount of fluid differed by only 1.1 L over 72 hours between

groups I and II, the time to receiving a fluid challenge and meeting

CVP goals was significantly reduced.16 Early and aggressive fluid

administration modulates early inflammation and reduces vaso-

pressor support (associated with decreased mortality) and qualifi-

cation for corticosteroids.17-21,22 Consistent with previous

studies, we found a significant association between ScvO2, MAP,

lactate measured, and mortality reduction.20,23,24

Group I (control group) in-hospital mortality, in all hemody-

namic subgroups, exceeded 42% (Table 2). When compared to

group II (treatment group), there was a significant in-hospital

mortality reduction after receiving the RB. The in-hospital mor-

tality for patients in this study versus the original EGDT study of

the same hemodynamic subgroup (lactate >4 mmol/L or

SBP <90 mm Hg) was 47.2% versus 46.5% in the control and

31.5% versus 30.5% in the treatment groups, respectively.4 In

an examination of 15,022 patients over a similar time frame,

Levy et al showed that a SBP <90 mm Hg or

MAP <65 mm Hg and lactate �4 mmol/L was associated with

a 46.1% hospital mortality which corresponds to a 48.0%observed mortality in the same subgroup in this study.3 Of par-

ticular interest are normotensive patients with an elevated lactate

who are described as ‘‘cryptic shock’’ and who experienced an

in-hospital mortality reduction from 42.8% to 24.9%

(P < .001).25 These patients who appear stable by traditional

vital signs have an underappreciated illness severity and often

later suffer cardiopulmonary collapse or multiorgan failure

due to persistently untreated hypoperfusion.26,27

Despite significantly higher baseline organ dysfunction

scores (APACHE-II and SOFA) in group II, there was still a

significantly greater improvement over the first 24 hours indi-

cating improved organ function compared to group I. For

example, while baseline pulmonary function via partial pres-

sure of arterial oxygen/fraction of inspired oxygen (PaO2/FiO2)

ratios were equal, group I significantly worsened by 9.0%while group II improved by 13.7% at 24 hours (both

P < .001). Paralleling these findings were a lower use of

mechanical ventilation in group II. Early detection and treat-

ment of shock on hospital admission are associated with

decreased need for mechanical ventilation.27

Lactate clearance is significantly associated with the

modulation of inflammatory biomarkers, organ failure, and

outcome.28 In this study, the lactate clearance in group I versus

group II significantly improved over the first 6 hours (2% vs

21%, P ¼ .044) and over 24 hours (11% vs 39%, P < .001).

However, the alactemic patients (lactate <2 mmol/L with

hypotension) had a baseline in-hospital mortality of 52.2% and

were less responsive to the in-hospital mortality reduction of

Table 3. Individual Study Center Settings and Mortalitiesa

Center Hospital LocationsGroup IPeriod

Group IIPeriod

Group IMortality

Group IIMortality

Absolute MortalityReduction

Relative MortalityReduction

A ED/GMF/OR/ICU 2003-2005 2006-2007 43.0 33.6 9.5 22.0B ED/ICU 2006 2007-2009 36.7 28.6 8.1 22.1C ED/ICU 2004 2005-2007 35.3 27.8 7.5 21.3D ED/ICU 2005 2006-2008 41.3 25.3 16.0 38.7E ICU 2004-2005 2006-2007 41.6 31.8 9.8 23.6F ED/GMF/OR/ICU 2003-2005 2006-2009 38.8 30.7 8.1 20.8G ED/ICU 2004-2004 2005-2008 36.8 18.8 18.0 48.9H ED/GMF/ICU 2003-2004 2005-2008 55.9 30.2 25.7 46.0Ib ED/GMF/ICU 2005-2008 56.3 40.0 16.3 28.9Jb ED/GMF/OR/ICU 2005-2008 36.5 21.4 15.1 41.5Kb ED/GMF/OR/ICU 2006-2009 43.0 23.2 19.8 46.0

Abbreviations: ED, emergency department; GMF, general medical floor; OR, operating room; ICU, intensive care unit.a The identity of the hospitals are blinded.b Concurrent centers.

Table 4. Adjusted Predictors of Mortality–Early Therapeutic Interventions

Therapeutic Interventions Variable in Database Odds Ratio 95% Confidence Interval P Value

6-Hour resuscitation bundleSerum lactate measured Lactate—0 hours 2.18 1.89-2.52 <.001Blood cultures before antibiotics Blood culture in <3 hours 1.01 0.84-1.21 .92Early treatment with antibiotics Antibiotics in <3 hours 1.00 0.85-1.18 .96Intravenous fluids delivered Fluid challenge in <3 hours 1.26 0.96-1.66 .09Mean arterial pressure �65 mm Hg achieved MAP �65 mm Hg—6 hours 0.59 0.50-0.70 <.001Central venous pressure �8 mm Hg achieved CVP � 8 mm Hg—6 hours 1.17 0.88-1.57 .28Central venous oxygen saturation �70% achieved ScvO2 �70%—6 hours 0.75 0.56-0.99 .047

Abbreviations: CVP, central venous pressure; MAP, mean arterial pressure; ScvO2, central venous oxygen saturation.

8 Journal of Intensive Care Medicine 00(0)

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

GENESIS than any other hemodynamic subgroup. This alacte-

mic patient population raises caution with the use of lactate

clearance as an isolated diagnostic and therapeutic end point.29

There was a 24.6% reduction in duration of hospital stay and

a 33.3% reduction in total hospital charges per admission

following GENESIS implementation which are similar to the

previous reports.30-32 Ross et al found that the association

between hospital volume and 30-day mortality is maximized

when the volume threshold for hospitalization exceeds 210

patients per year for pneumonia in particular.33 Applying this

economic model to our study means that 1071 hospital days

could potentially be saved resulting in $10,063,830 in-hospital

charges saved per year.

Limitations

This study was not a prospective randomized trial, thus, it has

the capacity to reveal associations but not causal relationships.

Additional unmeasured procedural changes at institutions, such

as implementation of other quality initiatives, may have led to

unaccounted differences between groups over time. These

include other components of sepsis management in the

subsequent 24 hours such as corticosteroids, protective lung

strategies, glucose control, recombinant activated protein C,

gastrointestinal ulcer, and deep venous prophylaxis.34 These

interventions (some debatable) and the diagnostic coding for

sepsis have seemingly diminished mortality over the last

decade.35-37 In addition, despite the use of consistent criteria

for identifying patients with severe sepsis and septic shock,

variability in diagnosis, and chart abstraction may have led to

heterogeneity in historical controls. The designation of patients

into control and treatment groups was based on assumptions of

care or the treatment effect. The control group patients not

receiving or achieving the RB (Group IA and IB) could have

received partial RB completion. The treatment group or

patients receiving the RB (Group II A) could have received less

than full RB completion. Group IIA in particular was assumed

to have the same intervention as group IIB; however, despite

the lack of quantification of bundle compliance for group IIA,

it is likely they received less than 100% RB compliance, given

that the overall concurrent model averaged 67% compliance

between fully achieved and not fully achieved cohorts.

In either scenario regarding the control or treatment groups,

the overall treatment effect of the study and findings of this

study are conservative and understated. Although there is a

4-fold greater number of patients in the before-and-after group

compared to the concurrent group, the findings are comparable.

Baseline illness severity and mortality were similar across all 4

subgroups. The similar mortality reduction seen in both assess-

ment strategies indicate that the treatment effect of the RB is

very robust. Because of incomplete data capture, the post hoc

detail of which bundle element is most important remains

unconfirmed.

Despite these aforementioned issues, this is the first study to

show similar in-hospital mortality reductions in both a before-

and-after and concurrent implementation designs. The concur-

rent findings emphasize the importance of RB compliance and

CQI to realize improved outcomes.38 These findings also sup-

port the notion that implementation trials can be scientifically

acceptable alternatives to randomized controlled trials.39 From

the standpoint of equipoise in following sepsis guidelines, this

study design avoids the ethical issue of randomization to poten-

tially inferior care that is not in accordance with current best

practice recommendations.40

Conclusions

Patients with severe sepsis and septic shock receiving RB

within community and tertiary care settings experience signif-

icant reductions in in-hospital mortality, organ dysfunction,

and health care resource utilization. This results from early

detection of high-risk patients, a reduction in time to delivery

of critical elements of best practice care, and a CQI process.

Future emphasis should be directed to overcoming logistical,

institutional, and professional barriers to the implementation

of RB similar to the approaches for acute myocardial infarc-

tion, stroke, and trauma. In doing so, this may lead to at least

1 life being saved for every 7 patients treated for severe sepsis

and septic shock.

Cannon et al 9

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

Appendix A

Table A1. GENESIS Participating Members

Centers Authors and Contributors

Barnes-Jewish Hospital, Division of Emergency Medicine, Washington University School of Medi-cine, Campus Box 8072, 660 S. Euclid Avenue, St. Louis, MO 63110, USAIRB committee approval:Washington University in St. Louis Human Research Protection Office (07-1235)

Contact: Chris Holthaus, MDOffice: (314) 747-5994Fax: (314) 362-0419Email: holthausc@wusm.wustl.eduPI:(1) Chris Holthaus, MD(2) Brent Ruoff, MDSub-PI: NACollaborators:Jennifer Williams, MSN, RN, ACNS-BCJohn StanleyMac McMastersCourtney Harrison, RNBrian Wessman, MDCraig McCammon, PharmDLara Christy, RNDamon Vincent, MDKari Yount, RNShelby Rives, RNAmy Stiefel, RNMelissa Caldwell, RN

California Pacific Medical Center, 2333 Buchanan St., San Francisco, CA 94115, USAIRB committee approval:California Pacific Medical Center Institutional Review Board (FWA00000921)

Contact: Ron Elkin, MDOffice: (415) 749-5746Email: elkin.ron@gmail.comPI: Ron Elkin, MDSub-PI:Cathy Camenga, RNAlice Chang, RNGouri Chavan, RNMinh Nguyen, MDBing Tschai, RNCollaborators:Christopher Brown, MDThomas Knight, MDMartie Mattson, RNThomas Peitz, MDMichael Rokeach, MD

Christiana Care Health System, 4755 Ogletown-Stanton Rd., Newark, DE 19718, USAIRB Committee Approval:Christiana Care Health System Institutional Review Board (CCC#28136)

Contact: Marc T. Zubrow, MD, FCCPPhone: 302-733-1000Email: mzubrow@christianacare.orgPI: Marc T. Zubrow, MDSub-PI:Gerard J. Fulda, MDCollaborators:Thomas A. Sweeney, MDMaureen A. Seckel, APNAlison C. Ellicott, RNDonna D. MahoneyPaula M. Fasano-Piectrazak, RNMegan B. Farraj, PharmD

Henry Ford Hospital-Main Campus, 2799 W. Grand Blvd CFP-2, Detroit, MI 48202, USAIRB Committee Approval:Henry Ford Health System Institutional Review Board (#4693)

Contact: Arturo Suarez, MDEmail: asuarez1@hfhs.orgPI: Arturo Suarez, MDCollaborators:Victor Coba, MDKristine McGregor, RN

(continued)

10 Journal of Intensive Care Medicine 00(0)

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

Table A1. (continued)

Centers Authors and Contributors

Damon J. Goldsmith, BSBrandon Claxton, BSDante Figueroa, MDMelissa Whitmill, MDCraig Bailey, MDDaniel Singer, MDAnja Kathrin Jaehne, MDLaura Eichhorn-Wharry, MDSamantha Brown, BSGarrett Chapman, BSAaron Baugh, BSEmanuel Rivers, MD, MPH, FCCP

Henry Ford Hospital-Wyandotte (HFWH), Department of Emergency Medicine, 2333 Biddle,Wyandotte, MI 48192, USAIRB Committee Approval:Henry Ford Health System Institutional Review Board (#4693)

Contacts: Satheesh Gunaga, DOPhone: 734-246-7380Email: drgunaga@yahoo.comPI:(1) Satheesh Gunaga, DO(2) Vipul Kella, MDSub-PI: NACollaborators:Jedediah Walker, MS4Gage Dixon, MS4Christopher Nedzlek, DOJason Fletcher, MS4Ryan Siwiec, MS4Asad Mehboob, M4Ryan Buckley, MS3Robert McCurren, MDMark Pensler, MD, FCCPRavinder Gandhi, MDKevin Boehm, DO, MScLisa Simpson, RNKenda Calhoun, RNJames Machcinski, RNLisa Blanchette, RNCharles Arnold, RNLois Vandercook, RNTravis Esckridge, RN

Northwest Community Hospital, 800 West Central Road, Arlington Heights, IL 60005, USAIRB Committee Approval:Northwest Community Hospital Institutional Review Board (NCH11-10)

Contact: Nathan Lidsky, MDEmail: nlidsky@nch.orgPI: Nathan Lidsky, MDSub-PI: NACollaborators:Melanie Atkinson, MSN, APRN, BC,CCRNDiane Ryzner

Porter Hospital, 814 LaPorte Ave., Valparaiso, IN 46383, USAIRB Committee Approval:Porter Hospital Institutional Review Board (S-502010)

Contact: Bonnie Turman, MS, RNPhone: 219-263-7110Email:bonnie.turman@porterhealth.comPI: Bonnie Turman, MS, RNSub-PI:Douglas Mazurek, MDTim Whetsel, MDBaqhar Mohideen, MDCollaborators:Terrie Fontenot, BSN, RNLynn Kowert, BSN, RN

(continued)

Cannon et al 11

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

Table A1. (continued)

Centers Authors and Contributors

Valerie Johns, RNKitty Cavanaugh, BSN, RNEllen Rastovski, BS, RNJeff Chin, PharmDTracy Dabrowiak, PharmD

St. Cloud Hospital, Intensive Care Unit, 1406 Sixth Ave. North, St. Cloud, MN 56303-1901, USAIRB Committee Approval:St. Cloud Hospital Institutional Review Board (FWA00001162)

Contact: Scott Davis, MD, FCCP, FCCMWork (320)-251-2700Fax (320)-240-7850Email: Daviss@Centracare.comPI: Scott Davis, MDSub-PI:John Olsen, MD, FCCP, FACPCollaborators:Kirsten Skillings, RN, MA, CCNSBonnie Curtis, RNGail Jenson, RNJoe Sauer, RPhJulianne Heath, BA

St. Joseph Mercy Hospital, 5301 McAuley Drive, Ypsilanti, MI 48197, USAIRB Committee Approval:St. Joseph Mercy Health System Institutional Review Board (R-05-670)

Contact: Pat PosaPhone: 248-890-0044Email: posap@trinity-health.orgPI: Pat Posa, RN, BSN, MSASub-PI:Mary-Anne Purtill, MDCollaborators:Sheri Brown, RNJennifer Cirino, MS1Christine Curran, MDLisa Fetters, RN, MSNRegina Freeman, RNDenise Harrison, RN, MSNLaszlo Hoesel, MD (G3)Jeong Hyun, MD (G2)Brian Kurylo, RNNicolas Mouawad, MD (G3)Wendy Nieman, RNFran Rocheleau, NPLora Silverman, MD (G3)Cecila Sosonowski, RNCatherine Stewart, RN

The University of Kansas Hospital, Department of Emergency Medicine, 3901 Rainbow Boulevard,Kansas City, KS 66160, USAIRB Committee Approval:Kansas University Medical Center Human Subjects Committee (#11157)

Contact: Chad M. Cannon, MDOffice: 913-588-6504Fax: 913 588-9104Email: ccannon@kumc.eduPI: Chad M. Cannon, MDSub-PI:Scott T. Rawson, MDMichael Hastings, RN, BSN, CEN, EMTJeremy Strom, MDCarol Cleek, RN, MSN, APRN-BC,CNAA-BCMichael Moncure, MDSteven Q. Simpson, MD, FCCPCollaborators:Katherine Mann, RN, BSN

(continued)

12 Journal of Intensive Care Medicine 00(0)

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

Acknowledgments

The authors would also like to extend tremendous gratitude to the 113

additional participating members (Appendix A) that contributed to

making a study of this magnitude successful, without them this would

have otherwise been impossible. The collaborative is indebted to the

team’s tireless efforts in caring for patients with sepsis and daily dedi-

cation to quality improvement. The authors would like to acknowledge

Kimberly McFarland, PhD, who assisted in the initial preparation of this

manuscript for a total of 15 hours work compensated by the Department

of Emergency Medicine of Henry Ford Hospital.

Authors’ Notes

The authors Cannon, Holthaus, Posa, Gunaga, Turman, Weingarten,

Suarez, and Rivers were involved in the study conception and design,

coordination, data collection, statistical analyses, interpretation of the

data and results, manuscript preparation, approval and overall responsi-

bility for the contents of the manuscript. The authors Zubrow, Kella,

Elkin, Davis, Milling, Lidsky, Coba, and Yang were involved in the study

coordination, data collection, interpretation of the data and results, manu-

script preparation, and approval of the manuscript. This research has been

presented at the (1) Society of Critical Care Medicine Annual Congress

in Nashville, Tennessee and received Annual Scientific Award for Top

Ten Abstracts, 2009. (2) Mediterranean Emergency Medicine Congress

in Valencia, Spain and received the Best Abstract Award, 2009. (3)

American College of Emergency Physicians, Boston, Massachusetts in

2009. (4) International Conference on Emergency Medicine, Singapore

and received the Best Oral Presentation Award, 2010.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to

the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, author-

ship, and/or publication of this article.

References

1. Martin GS, Mannino DM, Eaton S, Moss M. The epidemiology of

sepsis in the United States from 1979 through 2000. N Engl J

Med. 2003;348(16):1546-1554.

2. Andrews R, Elixhauser A. The national hospital bill: growth

trends and 2005 update on the most expensive conditions by

payer. Healthcare Cost and Utilization Project 2007.

Retrieved July 23, 2012 from http://www.hcup-us.ahrq.gov/

reports/statbriefs/sb42.pdf.

3. Levy MM, Dellinger RP, Townsend SR, et al. The Surviving

Sepsis Campaign: results of an international guideline-based

performance improvement program targeting severe sepsis. Crit

Care Med. 2010;38(2):367-374.

4. Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy

in the treatment of severe sepsis and septic shock. N Engl J Med.

2001;345(19):1368-1377.

5. Dellinger RP, Levy MM, Carlet JM, et al. Surviving Sepsis Cam-

paign: international guidelines for management of severe sepsis

and septic shock: 2008. Crit Care Med. 2008;36(1):296-327.

6. Rivers EP, Katranji M, Jaehne KA, et al. Early interventions in

severe sepsis and septic shock: a review of the evidence one

decade later. Minerva Anestesiol. 2012;78(6):712-724.

7. Moore LJ, Jones SL, Kreiner LA, et al. Validation of a screening

tool for the early identification of sepsis. J Trauma. 2009;66(6):

1539-1546.

Table A1. (continued)

Centers Authors and Contributors

Ira Marsh, MDNiaman Nazir, MBBS, MPHAlison Pontious, RN, BSNChad Toney, MBANia Thompson, BAShannon M. Wimsett, RN, BSNRick Blevins, RN, BSN, CEN, EMT-PMichelle L. Bolen, RN, BSN, CCRNAmanda Gartner, RN, MSN, CCRNAkiko Kubo, RN, BSN, CCRNDoug Peterson, MS, RN

University Medical Center at Brackenridge, ED Administration Offices, 601 E. 15th Street, Austin,TX 78701, USAIRB Committee Approval:Brackenridge Hospital Institutional Review Board (#07-BHIRB-353)

Contact: Truman J. Milling Jr, MDPhone: 512-324-7023Email: tjmilling@yahoo.comPI:(1) Jordan Weingarten, MD, FCCP(2) Truman J. Milling Jr, MDSub-PI:Patrick Crocker, DOCollaborators:Ben KingOmid Zad, MDHassie CooperAnna Sicher, RN, MPA

Cannon et al 13

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from

8. Rivers E, Rubinfeld I, Mantueffel J, Abou Dagher G, McGregor

K, Mlynarek M. Implementing sepsis quality initiatives in

multiprofessional care model. ICU Director. 2011;2:147-157.

9. Otero RM, Nguyen HB, Huang DT, et al. Early goal-directed

therapy in severe sepsis and septic shock revisited: concepts,

controversies, and contemporary findings. Chest. 2006;130(5):

1579-1595.

10. Li X, Fan QX, Wang J. Clinical observation of early

goal-directed therapy in patients with septic shock [Article in

Chinese]. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue. 2009;

21(12):742-743.

11. Cardoso T, Carneiro AH, Ribeiro O, Teixeira-Pinto A, Costa-Per-

eira A. Reducing mortality in severe sepsis with the implementa-

tion of a core 6-hour bundle: results from the Portuguese

community-acquired sepsis study (SACiUCI study). Crit Care.

2010;14(3):R83.

12. Schramm GE, Kashyap R, Mullon JJ, Gajic O, Afessa B. Septic

shock: a multidisciplinary response team and weekly feedback

to clinicians improve the process of care and mortality. Crit Care

Med. 2011;39(2):252-258.

13. Coba V, Whitmill M, Mooney R, et al. Resuscitation bundle

compliance in severe sepsis and septic shock: improves survival,

is better late than never. J Intensive Care Med. 2011.

14. Reed K, May R. HealthGrades Emergency Medicine in American

Hospitals Study. Health Grades 2010; The First Annual Report: June

2010. Retrieved July 23, 2012 from http://www.healthgrades.com/

business/img/HealthGradesEmergencyMedicineStudy2010.pdf.

15. Lundberg JS, Perl TM, Wiblin T, et al. Septic shock: an analysis

of outcomes for patients with onset on hospital wards versus

intensive care units. Crit Care Med. 1998;26(6):1020-1024.

16. Gaieski DF, Mikkelsen ME, Band RA, et al. Impact of time to

antibiotics on survival in patients with severe sepsis or septic

shock in whom early goal-directed therapy was initiated in the

emergency department. Crit Care Med. 2010;38(4):1045-1053.

17. Dorresteijn MJ, van Eijk LT, Netea MG, Smits P, van der Hoeven

JG, Pickkers P. Iso-osmolar prehydration shifts the cytokine

response towards a more anti-inflammatory balance in human

endotoxemia. J Endotoxin Res. 2005;11(5):287-293.

18. National Heart Lung, and Blood Institute Acute Respiratory

Distress Syndrome Clinical Trials Network, Wiedemann HP,

Bernard GR, Thompson BT, et al. Comparison of two fluid-

management strategies in acute lung injury. N Engl J Med.

2006;354(24):2564-2575.

19. Rivers EP. Fluid-management strategies in acute lung injury—lib-

eral, conservative, or both? N Engl J Med. 2006;354(24):2598-2600.

20. Pope JV, Jones AE, Gaieski DF, Arnold RC, Trzeciak S, Shapiro

NI. Multicenter study of central venous oxygen saturation

(ScvO(2)) as a predictor of mortality in patients with sepsis. Ann

Emerg Med. 2010;55(1):40-46.e1.

21. Sprung CL, Annane D, Keh D, et al. Hydrocortisone therapy for

patients with septic shock. N Engl J Med. 2008;358(2):111-124.

22. Durairaj L, Schmidt GA. Fluid therapy in resuscitated sepsis: less

is more. Chest. 2008;133(1):252-263.

23. Varpula M, Tallgren M, Saukkonen K, Voipio-Pulkki LM, Pettila

V. Hemodynamic variables related to outcome in septic shock.

Intensive Care Med. 2005;31(8):1066-1071.

24. Chamberlain DJ, Willis EM, Bersten AB. The severe sepsis bun-

dles as processes of care: a meta-analysis. Aust Crit Care. 2011;

24(4):229-243.

25. Mikkelsen ME, Miltiades AN, Gaieski DF, et al. Serum lactate is

associated with mortality in severe sepsis independent of organ

failure and shock. Crit Care Med. 2009;37(5):1670-1677.

26. Aduen J, Bernstein WK, Khastgir T, et al. The use and clinical

importance of a substrate-specific electrode for rapid determina-

tion of blood lactate concentrations. JAMA. 1994;272(21):

1678-1685.

27. Estenssoro E, Gonzalez F, Laffaire E, et al. Shock on admission

day is the best predictor of prolonged mechanical ventilation in

the ICU. Chest. 2005;127(2):598-603.

28. Nguyen HB, Loomba M, Yang JJ, et al. Early lactate clearance is

associated with biomarkers of inflammation, coagulation, apopto-

sis, organ dysfunction and mortality in severe sepsis and septic

shock. J Inflamm (Lond). 2010;7:6.

29. Rivers EP, Elkin R, Cannon CM. Counterpoint: should lactate

clearance be substituted for central venous oxygen saturation as

goals of early severe sepsis and septic shock therapy? No. Chest.

2011;140(6):1408-1413.

30. Shorr AF, Micek ST, Jackson WL Jr, Kollef MH. Economic

implications of an evidence-based sepsis protocol: can we

improve outcomes and lower costs? Crit Care Med. 2007;35(5):

1257-1262.

31. Talmor D, Greenberg D, Howell MD, Lisbon A, Novack V, Sha-

piro N. The costs and cost-effectiveness of an integrated sepsis

treatment protocol. Crit Care Med. 2008;36(4):1168-1174.

32. Kiers HD, Griesdale DE, Litchfield A, et al. Effect of early

achievement of physiologic resuscitation goals in septic patients

admitted from the ward on the kidneys. J Crit Care. 2010;

25(4):563-569.

33. Ross JS, Normand SL, Wang Y, et al. Hospital volume and 30-day

mortality for three common medical conditions. N Engl J Med.

2010;362(12):1110-1118.

34. Dellinger RP, Carlet JM, Masur H, et al. Surviving Sepsis Cam-

paign guidelines for management of severe sepsis and septic

shock. Crit Care Med. 2004;32(3):858-873.

35. Lagu T, Rothberg MB, Nathanson BH, Pekow PS, Steingrub JS,

Lindenauer PK. The relationship between hospital spending and

mortality in patients with sepsis. Arch Intern Med. 2011;171(4):

292-299.

36. Lagu T, Rothberg MB, Shieh MS, Pekow PS, Steingrub JS, Linde-

nauer PK. Hospitalizations, costs, and outcomes of severe sepsis in

the United States 2003 to 2007. Crit Care Med. 2012;40(3):754-761.

37. Kumar G, Kumar N, Taneja A, et al. Nationwide trends of severe

sepsis in the 21st century (2000-2007). Chest. 2011;140(5):

1223-1231.

38. Nguyen HB, Corbett SW, Steele R, et al. Implementation of a

bundle of quality indicators for the early management of severe

sepsis and septic shock is associated with decreased mortality.

Crit Care Med. 2007;35(4):1105-1112.

39. Vincent JL. Toward the end of randomized, controlled trials in the

intensive care unit? Crit Care Med. 2011;39(4):921.

40. Miller FG, Joffe S. Equipoise and the dilemma of randomized

clinical trials. N Engl J Med. 2011;364(5):476-480.

14 Journal of Intensive Care Medicine 00(0)

at UNIV OF KANSAS MEDICAL CENTER on August 18, 2012jic.sagepub.comDownloaded from