Research shows that early goal-directed therapy is improving survival rates for sepsis patients.
There is an estimated 750,000 cases of severe sepsis per year in the U.S. With our aging population and strengthening pathogens, these numbers are expected to double in the next 25-30 years, making sepsis an even more widely prevalent and devastating disease. Of all these cases, two-thirds present through the ED with an average ED LOS of 5 hours. Prior to Early Goal-Directed Therapy (EGDT), mortality estimates for sepsis have ranged as high as 56%. Sepsis occupies the largest growth in health care costs of the national bill from 1997-2005 and represents the 4th largest recipient of Medicare and Medicaid expenditures for hospitalized patients.
A landmark randomized controlled trial published in 2001 showed that EGDT had a significant absolute mortality reduction vs. “standard therapy” for treatment of patients with severe sepsis or septic shock. Patients that presented to the ED qualified for the protocol in one of two ways: 1) sepsis [infectious source + ≥2 SIRS criteria] and persistent hypotension [systolic blood pressure < 90 mm/Hg after a 20-30mL/kg fluid bolus] or 2) sepsis with a serum lactate ≥ 4.0 mmol/L.
“Standard” therapy in the study was left to the discretion of the emergency physician but included monitoring of urine output, central venous pressure (CVP), and mean arterial pressures (MAP) to preset goals. The EGDT arm included the same monitoring as the “standard” group with the addition of continuous central venous oxygen saturation (ScvO2) monitoring and “3 goal-directed endpoints” to maximize (CVP, MAP, ScvO2) [Fig.1]. In the end, the main difference in treatments between the two groups was that the EGDT group in the first 6 hours of presentation received significantly more fluids (1.5L; however approximately the same amount over 72hrs @ 13L), blood transfusions (102 ml more over 72 hours), and inotropic support (14% more). The EGDT group also had less use of vasopressors and mechanical ventilation over 72 hours. These simple changes resulted in an absolute mortality reduction of 16% (46.5% to 30.5%) vs. “standard therapy” for treatment.
But EGDT is more than “early antibiotics and fluids.” EGDT simplistically is a resuscitation algorithm for optimizing hemodynamic support by following three indices: 1) CVP 8-12 mmHg: optimized with repeated IVF bolusing 2) MAP 65-90 mmHg: optimized with vasopressors or vasodilators 3) ScvO2 ≥70%: optimized with oxygen, CVP 8-12, vasopressors, blood transfusions, and dobutamine for inotropic support if needed. In practice, the ScvO2 can be measured by either a spot blood gas (co-oximetry hemoglobin saturation) from the distal port of the central line or by using a continuous monitoring system. The protocol may seem intimidating or difficult to remember at first but becomes easier to recall with repeated performance. Here’s how it works out schematically. (see figure 1).
Figure 1. Protocol for treating patients in Severe Sepsis and Septic Shock with Early Goal Directed Therapy. All endpoints must be optimized within 6 hours to achieve protocol compliance. (Used with kind permission from NEJM). (Click on the image to view a high-res version)
As the science currently stands, EGDT via an “invasive” approach has been proven and externally validated with similar mortality reductions (15-20%) in multiple studies comprising over 6,000 patients. The socioeconomic benefit of EGDT has repeatedly shown an average of a 20% reduction in sepsis related costs. This benefit exceeds that of other acute interventions such as thrombolytics for ST elevation MI (3-5% absolute mortality reduction). Given that nearly two-thirds of severely septic patients present through the ED, can EPs as resuscitation experts apply EGDT early in the patient’s presentation? Will EDs support their nursing staff and physicians in this effort? The answers are yes, and maybe.
Overcoming Practice Challenges
A 2004 national survey found that only 10% of EDs were routinely performing EGDT. By 2006 that number was increased to 45%. EGDT represents an enormous breakthrough in sepsis management that significantly impacts patient mortality and has been endorsed nationally and internationally via the Surviving Sepsis Campaign. Third party payors and insurance companies are incentivising hospitals with pay for performance for high risk diseases such as sepsis. Despite the push for practice change, why aren’t EDs and EPs performing it as much?
Moving from Awareness to Action
Despite well documented benefits and endorsement by international awareness campaigns, many EDs are still unaware of the benefits of this aggressive approach to sepsis and related shock. Why? Information overload and inadequate time may preclude practicing clinicians from in-depth familiarization with current research findings. What to do? Some of the strategies for staying abreast of EGDT literature include becoming familiar with appraisals, such as ACEP Clinical Guidelines, Cochrane, etc… Even more up to date information can be obtained by setting up automatic email notifications of clinically relevant evidence via topic specific parameters through individual journals or literature search engines. The Surviving Sepsis Campaign (survivingsepsis.org) and Institute for Healthcare Improvement (www.ihi.org) websites both contain useful references and resources for developing an EGDT protocol at both the institutional and provider level.
Despite EGDT being adopted as standard of care by multiple authorities, a 2006 survey found 32% of EDs were planning an EGDT protocol while 23% reported neither using nor planning an EGDT protocol. The reason given by some for this reluctance to acceptance is the traditional practice of delaying invasive monitoring to ICU admission. Others resist adoption of EGDT due to contradictory personal experience or skepticism of study results. A recent national survey found that 16% of EPs report a “lack of agreement with the EGDT protocol.” Despite this resistance from a minority the greater barrier to acceptance seems to be in whether the protocol is practical.
Despite clear enrollment criteria for severely septic and septic shock patients, some EPs resist applying EGDT to their patients thinking that the study patients were somehow different from their patients. Are my patients older or sicker and less likely to benefit? Are my patients less sick and thus, less in need of aggressive monitoring? What about the patient that “looks good” with a lactate of >4? This notion has been disproven as others have validated similar mortality and illness severity with this screening criteria.
The general answer is that EGDT trials with clear enrollment criteria have consistently shown a significant mortality benefit across a wide spectrum of practice locations whether community hospitals or large tertiary care centers. Barring a patient preference for noninvasive care, all patients meeting criteria should be enrolled. Other additional exclusion criteria from the original Rivers trial were age <18 years, pregnancy, acute cerebral vascular event, acute coronary syndrome, acute pulmonary edema, status asthmaticus, primary diagnosis of cardiac dysrhythmia, contraindication to central line, active GI hemorrhage, seizure, uncured cancer (during chemotherapy), and immunosuppression. Even when the indications for treatment have been expanded the mortality reduction is still the same.
The scenario of “looking good/normotensive” despite a lactate >4 and being uncertain if they may benefit from a central line and EGDT is slightly more difficult. The majority of studies showing a mean 20% +/- 13% absolute mortality reduction from EGDT have used the same inclusion criteria as the original trial and combined results of septic patients with lactate ≥4.0 mmol/L and patients with persistent hypotension making it harder to tease out the severe sepsis (lactate >4) groups. Despite this, the lactate>4 group still showed a trend toward mortality reduction with EGDT.
Interestingly, in Rivers’ study the mean initial MAP in the “standard” and EGDT groups was 76 and 74 mmHg. Some investigations have shown that untreated septic patients with lactate ≥ 4.0 mmol/L will have a significantly higher short term mortality than patients with a normal lactate and would benefit from EGDT. Future possible alternatives to invasive monitoring may be geared toward resuscitating to a 10% lactate clearance every 6 hours or addressing other biomarkers of shock and illness severity.
Overall most institutions that have successfully instituted EGDT can point to a physician champion that advocated for the resources needed to follow this protocol. Granted it is impossible to do without measuring a CVP or ScvO2 and these obviously need to be acquired or else EGDT will not happen. But once the ED acquires the appropriate equipment, department-wide training initiatives and nursing support are needed to educate physicians and nurses on EGDT endpoints. Some EDs train their nurses to monitor CVP and ScvO2 and to alert physicians when goals are not being met thereby freeing up physicians from hands-on bedside care.
There may be an up-front capital expense for implementing an EGDT protocol due to additional staffing and equipment; however, cost analysis studies have shown an overall cost savings when performing the protocol, mostly due to shorter ICU and hospital length of stays.
Larger academic centers with resident support may have adequate resources to perform the protocol without hiring additional staff. At smaller community centers, rapid response teams (RRT’s) capable of providing the staffing and expertise for rapid resuscitation may be a feasible approach. These teams, modeled on the traditional code team, are an effective solution to high acuity presentations such as acute myocardial infarction, cerebrovascular accident and sepsis resuscitation. They have been found to be a cost-effective technique for providing the significant but fairly brief staffing and resources required to resuscitate high risk patients in the ED, in-patient floors and ICUs of community hospitals. With practice EP’s and nurses quickly develop familiarity and confidence with EGDT and require less time at the bedside for protocol optimization. And since resources used in the ED for sepsis resuscitation provide an overall cost savings to the institution due to decreased ICU and overall length of stay, EP’s should advocate for hospital budget restructuring to reflect these savings.
Some EP’s may fear that they will have to provide critical care for which they lack advanced training. However, following therapeutic protocols can simplify care for prolonged ED stays and allow for a seamless transition between care teams.
Standing order sets for severe sepsis and septic shock can be tailored for specific individual institutions (i.e. formulary medication, labs, monitoring, responsibilities). For instance, some patients use continuous ScvO2 monitoring while others test spot blood gasses from the central line for oxygen saturation. Regardless of the protocol, the core EGDT focus should remain on optimizing the MAP, CVP, and ScvO2.
“We have difficulty identifying septic patients. We can’t do arterial lactate levels.” Studies have shown that a venous lactate and finger stick can be used as a surrogate for arterial lactate. To simplify the process many institutions have encouraged nurses to draw a lactate tube every time they draw blood cultures. Bedside point of care lactate measurement is also an option that allows for rapid identification (<5 minutes) of EGDT candidates. Once providers adjust to drawing a lactate with blood cultures or screening for ≥2 SIRS criteria and suspected infection, the process of identifying EGDT patients becomes routine.
Some EDs (5-38%) report difficulty or inability to place a central line in the ED. Placing the central line (IJ or Subclavian) is usually the most time consuming task. With bedside ultrasound the line can be placed quickly and safely. Once the line is placed, the physician can monitor the resuscitation while the nurse follows the endpoints with minimal time at the bedside.
ScvO2 can be examined continuously or with intermittent sampling from the central line. With experience, nurses can optimize therapeutic outcomes with minimal physician oversight. Patients already with a PowerPICC or Port-a-cath catheter can also be used to trend CVPs and obtain spot ScvO2 measurements.
PDAs, pocket cards, or EGDT posters in resuscitation rooms are a useful tool allowing physicians and nurses to follow the protocol. The learning curve for EGDT can be steep for some, but treatment and therapeutic goals remain constant. Patient care rapidly becomes predictable for both provider and nurse, with all parties on the same page and able to anticipate further treatment decisions.
The majority of severely septic and septic shock patients present through the ED with a short term mortality of 30-60% if treated without EGDT. Sepsis is an expensive disease process that already occupies a sizable portion of national healthcare costs, and it is anticipated to worsen in coming years. Multiple studies have shown that the survival rate can be improved (7-33% absolute mortality reduction) with a feasible and cost-effective means of treatment in the form of EGDT (Fig.1). EPs as resuscitation experts have the background and skill set to initiate management and stabilization of this critically ill patient population. However, individual provider and institutional barriers should be anticipated when attempting to set up or maintain an EGDT pathway.
Chris Holthaus, M.D.
Chief Resident 2007-2008
Washington University Emergency Medicine Residency
Barnes-Jewish/St. Louis Children’s Hospital
Jeffrey Green, M.D.
Assistant Residency Director
Department of Emergency Medicine
New York Hospital Queens
Emanuel P. Rivers, MD, MPH, IOM
Vice Chairman and Director of Research,
Department of Emergency Medicine
Senior Staff Attending, Emergency Medicine and Critical Care
Henry Ford Hospital, Detroit Michigan
Institute of Medicine, National Academies