Approach to Sepsis
Authors: Daniel Schneider, DO PGY-2, Christopher Burnsides, DO PGY-1, Lauren Gustafson, MD PGY-2 Rakesh Engineer, MD, CWRU/MetroHealth/Cleveland Clinic, Cleveland, OH
Editor: Michael Cole, MD University of Michigan, Chris Fowler, DO University of Arkansas for Medical Sciences
Updated: November 2019
An 84-year-old African American man presents to the local emergency department (ED) from a skilled nursing facility for fever, generalized malaise, and altered mentation for the past 12 hours. On arrival, the patient is answering questions appropriately but does appear increasingly somnolent. Pt has a history of hypertension, coronary artery disease and past CVA. Vital signs reveal a temperature of 39.4 C, HR of 125 bpm BP of 92/55 mm Hg, a respiratory rate of 20 per minute, and a pulse oximetry of 99% on room air. Physical exam is significant for dry mucous membranes, tachycardic regular rhythm, tachypnea but otherwise clear breath sounds. Pt has mild lower abdominal tenderness to palpation with an indwelling Foley catheter in place. Pt has brisk lower extremity pulses. The nurses ask “What are your initial orders doc?”
After completion of this module you should be able to:
- Define what sepsis is and its potential presentations
- Recognize the severity of illness of patients presenting with sepsis syndromes
- Describe the role and importance of early screening for at-risk patients presenting with symptoms concerning for sepsis syndromes
- Discuss briefly the evolution of sepsis management, from early goal directed therapy (EGDT), to lactate clearance, to CMS bundled quality-based measures, and their impact on sepsis care in the Emergency Department
- Demonstrate an understanding for the evaluation, treatment, and disposition of patients with sepsis syndromes
- Understand the various clinical decision tools available to the EM physician in the evaluation of septic patients
Sepsis is a disorder that often perplexes students and misleads clinicians. Fundamentally, sepsis is a clinical condition whereby the body enters into an intense and dysregulated inflammatory response to an infection. As with many disease processes, sepsis falls along of spectrum ranging from mild symptoms that are easily corrected and treated to severe sepsis that progresses rapidly to septic shock, which is heralded by end-organ hypoperfusion resulting in death. Of note SIRS (systemic inflammatory response syndrome) is no longer included in the definition of sepsis because it is not always caused by infection. Sepsis begins with infection leading to bacteremia (detection of viable bacteria in the blood stream) and places a patient at risk for developing sepsis. Once present, the bacteria lead to a systemic inflammatory response mediate by pro-inflammatory cytokines that further activates the immune system. This cascade feeds forward exacerbating the response further. This response causes metabolic derangements that lead to tissue hypoxia and hypoperfusion at the cellular level.
Exact numbers of sepsis can be difficult to obtain due to a multitude of factors. A large retrospective study in JAMA (Rhee, C et al, 2017) that included both emergency department (ED) and admitted patients, found that sepsis has a 6% incidence and a 15% in-hospital mortality (with an additional 6.2% of patients discharged to hospice). This data was used to estimate a national incidence of sepsis of ~1.7 million cases per year, with 270,000 sepsis mortalities. These statistics demonstrate the severity of illness of patients with sepsis syndromes. Despite the disease severity, patients can be difficult to diagnose as they often can present with atypical symptoms or delayed presentations. Therefore, it is important to have a high index of suspicion for sepsis and to understand the evaluation and reassessment of sepsis syndromes in our ED patients.
There has been an effort to simplify the definition of sepsis to make it easier for clinicians to identify. These have changed somewhat over time but below is a summary of these. Some outdated terms are included because of they are still used widely in a clinical setting.
Sepsis-1 (SEP-1- one hour) Definitions:
- These are still used as criteria for what is included in the publicly reported quality measure for sepsis (per CMS)
- Sepsis: two or more SIRS criteria# + Suspected infection
- #SIRS criteria: (SIRS = Systemic Inflammatory Response Syndrome)
- Fever of > 38 C (100.4°F) or < 36 C (96.8°F)
- Heart rate > 90 beats per minute
- Respiratory rate > 20 breaths per minute or arterial carbon dioxide tension (PaCO2) < 32 mm Hg
- Abnormal white blood cell count (< 4,000/µL or >12,000/µL or >10% immature [band] forms)
- Severe Sepsis: Sepsis (as above) + sepsis-induced organ dysfunction or tissue hypoperfusion
- Evidence of organ dysfunction or tissues hypoperfusion includes:
- Hypoxemia, oliguria, acute kidney injury, coagulopathy, ileus, thrombocytopenia, hyperbilirubinemia, altered mental status, lactate ≥ 2.0 mmol/L, or hypotension
- Septic Shock: Hypoperfusion despite fluid resuscitation or lactate ≥ 4.0 mmol/L
- Sepsis-3.0 (SEP-3) Definitions: This is the current recommendation but has had some difficulty in being adopted
- Updated by Surviving Sepsis Campaign (SSC) in 2016, but inconsistent adoption
- Sepsis: a life-threatening organ dysfunction caused by a dysregulated host response to infection.
- Clinically, a SOFA score+ (see below) increase of 2 or more points can be used to characterize organ dysfunction
- Severe sepsis - removed from the definition
- Septic Shock: a subset of sepsis in which particularly profound circulatory, cellular, and metabolic abnormalities are associated with a greater risk of mortality than with sepsis alone. Clinically identified by a serum lactate level greater than 2 mmol/L (>18 mg/dL) in the absence of hypovolemia OR vasopressor requirement to maintain MAP of 65 mm Hg or greater.
+SOFA score: (Sequential [Sepsis-related] Organ Failure Assessment (SOFA) score)
The SOFA score, which was validate and most accurate in an ICU population and not an ED population, is a tool to quantify end organ dysfunction in septic patients and correlates with mortality. It can be used to monitor treatment response over hospital course, as a research tool, and as a prognostic indicator. Points are assigned and tabulated for a total score between 0-24. Mortality risk is then stratified based on initial, mean, and highest values. For instance, an initial SOFA score of 2-3 carries a 5% mortality risk; whereas an initial score >11 carries a 95% mortality rate. (Please see the cited primary article if interested in further details.) The disadvantage of using the SOFA score as an initial screening tool is that it requires laboratory results and data such asurine output that are not available when patients initially present to the ED.
qSOFA (Quick SOFA) score
In adult, non-ICU patients, when an infection is suspected, a positive qSOFA score predicts that a patient is at higher risk for a poor outcome. A positive score is defined as the presence of at least 2 out of 3 of the following criteria:
- Systolic blood pressure (SBP) ≤ 100 mm Hg
- Respiratory rate ≥ 22 breaths per minute
- Glasgow Coma Score (GCS) < 15
The qSOFA score is a modification of the SOFA score that can be performed at bedside without laboratory results and can aid in determining appropriate disposition. Its predictive value is improved when point of care lactate is added (see below). Of the new evidence-based scoring systems, qSOFA is the most applicable to the ED diagnosis of sepsis and predicts a worse outcome from patients with sepsis.
Initial Actions and Primary Survey
As with all patients and initial patient evaluation begins with assessment of airway, breathing, and circulation with interventions aimed at stabilizing these essential systems. Beyond this, the initial screening for sepsis should be carried out during triage and repeated by the initial provider. Although screening protocols vary by institution, they typically involve an assessment of vital signs to identify the presence of at least 2 of the 4 criteria systemic inflammatory response (SIRS) criteria. These criteria are well known to have poor sensitivity and specificity for the presence of sepsis syndromes, with a number of other, unrelated diagnoses fulfilling SIRS criteria (e.g., hemorrhage, dehydration, cirrhosis, non-infectious pulmonary disorders, etc.). Although, SIRS criteria are still widely used as a screening modality, there has been increased utilization of qSOFA (per SEP-3.0). Either way, initial risk stratification begins in triage. Many institutions have begun using a “Code Sepsis” or similar terminology to notify providers that a patient has screened positive for potential sepsis. This can be somewhat misleading as septic patients can present with normal vital signs or only have mildly abnormal vital signs and inconsistent history. It is key that a provider avoid anchoring for these patients. A detailed history and physical is vital to identify potentially septic individuals during the primary survey.
As previously stated the presentation of sepsis can vary widely depending on the patient. In an elderly patient with multiple co-morbidities, sepsis can be readily identified and diagnostic workup and treatment can begin rapidly. However, a younger patient may present atypically and ultimately be diagnosed with sepsis. Classic presentation is going to be a patient with fever, tachycardia, tachypnea, and/or hypotension with altered mental status (delirium) as the metabolic derangements associated with an infection manifest. Patients may complain of a cough and shortness of breath, which would incline the practitioner to investigate pneumonia as a possible source of infection. Likewise, a patient complaining of lower abdominal pain radiating into the back with dysuria would spur diagnostic testing for a possible urinary tract infection. Common sources of infection in sepsis include pneumonia, urinary tract infection (UTI), central line infection, or skin and soft tissue infection, though there are many other potential sources.
Often, patients will present with one or maybe no abnormal vital signs. Symptoms may be nonspecific, such as malaise, fatigue, or weakness. Sometimes, patients will have only mild delirium, manifesting with inattention. At other times, patients may appear well, but give history of fever, rigors, or confusion occurring earlier in the day. Table 1 lists factors that increase a patients risk for having sepsis.
Table 1: Factors associated with a higher risk of having sepsis
Occasionally may present early in the course of a disease and will become increasingly ill while in the emergency department. Sepsis can occur at any point during presentation. Frequent re-evaluations of patients is key in recognizing changes in a patient’s clinical course. Failure to do adequately re-evaluate patients can lead to delayed treatment and worsened outcomes. Patients may also return to the emergency department for worsening of symptoms and should be evaluated similarly for a change in their clinical presentation that might indicate early sepsis.
Diagnostic testing for patient is two-fold. Not only are you attempting to uncover a possible source of sepsis, but also testing for the severity of a patient’s presentation. As with all patients, testing should complement a complete history and physical. What can complicate this is that often patients with severe infections and sepsis may not be able to give adequate history. In these cases, it is important that practitioners use their best clinical judgment to direct testing.
Laboratory testing will often include blood tests as well as cultures. A complete blood count (CBC) with differential, basic or complete metabolic panel (BMP/CMP), urinalysis (UA) and venous lactate are typically ordered in the initial evaluation. Blood cultures and urine cultures should also be obtained before the administration of antibiotics, however antibiotic administration should not be delayed if obtaining cultures is delayed. Blood cultures should be obtained from two different blood draw locations to decrease the chances of contamination being interpreted as a positive culture.
For cases of suspected meningitis or encephalitis CSF analysis and cultures should be obtained. Additional testing should be performed specific to the suspected causes of sepsis (ie: wound cultures from skin and soft tissue infections, body fluid cultures or joint aspirates for septic arthritis).
Lactic acid (lactate) is the primary method of early risk stratification for patients presenting with sepsis. It is a marker of anaerobic glycolysis during periods of insufficient oxygen delivery. Low lactate levels correlate with low mortality; conversely, elevated lactate levels correlate with high mortality.
Lactate measurements should be performed at triage or immediately after blood is drawn from the patient. Venous samples are adequate to perform lactate testing and there is no need to draw an arterial sample to measure lactate for screening purposes. Delays in measurement may lead to falsely elevated levels (e.g., EMS blood and other older samples should not be used to obtain lactate levels, as this may result in falsely elevated values). Ideal testing involves use of point of care assays or an arterial blood gas machine based in the ED.
Ultimatelty labs are going to help to establish the degree of a patients severity of illness, level of end organ dysfunction and help to determine response to treatment. The Centers for Medicare and Medicaid Services (CMS) define end-organ dysfunction by the following criteria:
- SBP < 90 mm Hg or mean arterial pressure < 70 mm Hg or a SBP decrease > 40 mm Hg or < 2 SD below normal for age or known baseline
- Creatinine > 2.0 mg/dl (176.8 mmol/L) or Urine Output < 0.5 ml/kg/hour for > 2 hours,
- Bilirubin > 2 mg/dl (34.2 mmol/L),
- Platelet count < 100,000,
- Coagulopathy (INR >1.5 or aPTT >60 secs),
- Lactate > 2 mmol/L (18.0 mg/dl)
Most patients during their evaluation will get a chest xray. Stable patients should ideally have PA and lateral views. Patients that are more critical should have bedside portable exams. A chest exam will allow for evaluation for possible infectious infiltrates, effusions, pneumothoraces, enlarged cardiac silhouette or free air under the diaphragm. This exam will help for diagnosis of other possible etiologies of a patient’s instability or presenting symptoms. Acute abdominal series, which consist of an upright and supine abdominal views, can be used to screen for bowel obstruction or possible bowel perforation. CT scans should be obtained if more evaluation is needed to evaluate for the etiology of a patient’s sepsis.
Treatment for sepsis focuses on management of the suspected infection and management of the inflammatory response related to it. Described below are various protocols that have been used to guide the management of sepsis for the past several years. These are included here to help you understand the history and complexity of sepsis care. Broadly speaking, based on current guidelines, evidence and best practice, the components that are most essential in the treatment of sepsis include:
- Broad spectrum antibiotics (choice should be guided by institutional antibiograms and suspect source of infection)
- IV Lactated Ringers infusion at 30 mL/kg (unless there pulmonary edema or cardiogenic shock is suspected)
- Vasopressors should be initiated for persistent hypotension despite adequate volume resuscitation (or hypotension with pulmonary edema)
Choice of vasopressor varies however, typically in:
- Adults: norepinephrine is first line
- Infants/Children: dopamine is first line
- Steroids: not routinely administered
Should be given to patients already on chronic steroids or (more controversial) persistent hypotension refractory to adequate volume and vasopressors
Early Goal-Directed Therapy (EGDT) was a protocol focusing on early, aggressive therapy for treatment of sepsis in patients with a lactate > 4.0 or fluid refractory hypotension. The protocol recommended fluid resuscitation to a central venous pressure (CVP) of 8-12 mm Hg, vasopressor support to a mean arterial pressure (MAP) ≥65 mm Hg, and blood transfusions and inotropic support to achieve a central venous oxygen saturation (ScvO2) ≥ 70% (representing oxygen extraction from blood returning to the heart). Think in terms of “fill the tank,” “pressure to perfuse,” and “fix the oxygen debt.” The initial study showed an absolute risk reduction in mortality of 16% (30% mortality in EGDT vs 46.5% mortality in control group). EGDT involved placement of arterial line and central venous catheter and was a time and resource intensive process requiring staffing that is often unavailable in the ED.
Lactate clearance has been shown to be an acceptable alternative to EGDT. It did not require placement of a CVL or arterial line. This is a less time and resource intensive method to monitor septic patients in the ED. Achieving a normal lactate (≤ 2 mmol/L) or reduction of initial lactate by 10% was found to be non-inferior to achieving an ScvO2 ≥ 70%.
Comparison of EGDT, lactate clearance, and standard care: The ProCESS, ARISE, and ProMISE trials compared EGDT, a standardized protocol of care, and routine care, finding no difference in mortality, resource utilization, or length of stay between the treatment arms. Monitoring of CVP and ScvO2 is no longer a significant component in the management of patients with sepsis syndromes.
Antibiotics: Administration of broad-spectrum antibiotics is an important step in treatment of sepsis and septic shock and one of the core measures of SEP-1 that hospitals are evaluated on. A retrospective, multicenter study by Kumar et al. (2006) showed a direct correlation between mortality and the amount of time elapsed from of time from the first measurement of measured hypotension and to administration of “effective” antibiotics and mortality. This study was a retrospective review of ICU patients and “effective” antibiotics were defined by culture and sensitivity data that returned long after treatment was initiated. This data is generally not available in the ED and is the reason why broad-spectrum antibiotics are started in the ED when sepsis is identified/suspected. Other data has shown no significant change in mortality when antibiotics were administered at one, two, three, and up to six hours from triage (EMSHOCKNET research network ), or between 0 and 5 hours to antibiotic administration (retrospective review of 17,990 patients in the Surviving Sepsis Campaign by Ferrar, et al). While these studies show conflicting data, it does suggest administering antibiotics early, but that it is also reasonable to obtain diagnostic data (imaging, UA, other data that can be collected in the SEP-1 3 hour time window prior to antibiotic administration) to determine the most likely source of infection, allowing for selection of the most appropriate antibiotics.
Hospital measures (SEP-1) and Sepsis Treatment Bundles
The Centers for Medicare and Medicaid Services (CMS) SEP-1 measure established in 2015 is a publicly reported quality measure for sepsis care. All U.S. hospitals are required to report compliance which may affect hospital reimbursement. Components of the measure include 3- hour and 6- hour performance bundles.
Three Hour Bundle
- Lactate measurement
- Blood cultures obtained before antimicrobial administration
- Administration of broad-spectrum antimicrobials
- Administration of 30 ml/kg IV crystalloid Lactated Ringers for hypotension or lactate ≥ 4 mmol/L.
Six Hour Bundle
- Administration of vasopressors for hypotension that does not respond to fluids in order to achieve a MAP ≥ 65 mm Hg
- Reassess and document volume status after fluid administration for patients with hypotension that does not respond to fluids or lactate ≥ 4 mmol/L
- Repeat lactate measurement for patients with initial lactate ≥ 4 mmol/L
Initial data presented by CMS indicated that nationally, the majority of SEP-1 cases fail to meet all components of this measure, resulting in higher mortality rates. A retrospective cohort study was completed in 2018 to evaluate the association between compliance rate and mortality. Of the cases reviewed, 33% (281 of 851) “passed” SEP-1. Failure cases resulted in overall higher mortality rates grossly. However, this was insignificant when adjusting for disease severity. When evaluating individual components, only a delay in antibiotic administration resulted in higher mortality rates. Failure for any other reason showed no significant difference. Despite the controversy, hospitals across the U.S. continue to devote time and resources to reporting and compliance due to the financial implications of nationally reported quality metrics. Providers though, must still treat patients as individuals, especially if comorbidities provide relative or absolute contraindications (i.e. avoiding aggressive fluid resuscitation for congestive heart failure and end stage renal disease.)
In 2018, the Surviving Sepsis Campaign published a new Hour-1 treatment bundle
- Hour-1 Bundle
- Measure lactate level*
- Obtain blood cultures before administering antibiotics
- Administer broad-spectrum antibiotics
- Begin rapid administration of 30 mL/kg crystalloid for hypotension or lactate level ≥ 4 mmol/L
- Apply vasopressors if hypotensive during or after fluid resuscitation to maintain MAP ≥ 65 mm Hg
* Re-measure lactate if initial lactate is elevated (> 2 mmol/L)
You may come across this in your reading and clinical education but at this time the 1-hour bundle update is controversial and was initially rejected by both the American College of Emergency Physicians (ACEP) and the Society of Critical Care medicine (SCCM).
Nearly all patients with sepsis syndromes will require hospitalization. Patients with septic shock (either lactate ≥ 4 mmol/L or vasopressor dependent hypotension) should be admitted to an intensive care unit (ICU). For less severely ill patients, the specific level of care will be based on local facility capabilities and practice patterns. Community hospitals with lower levels of physician coverage after hours may prefer these patients to be admitted to an ICU. Larger academic hospitals with greater physician coverage may be capable of managing these patients on step-down or regular nursing floors.
In our hospital system, disposition is based on post-intervention lactate and blood pressure. Normotensive patients with lactate ≤ 2 mmol/L are generally considered safe for regular nursing floor. Normotensive patients with lactate between 2.0 – 3.5 mmol/L receive a consult from the medical ICU. Hypotensive patients and patients with lactate ≥ 3.5 mmol/L are routinely admitted to an ICU.
Pearls and Pitfalls
Maintain a high index of suspicion for sepsis syndromes, especially in patients of advanced age, significant comorbidities, immunosuppressed, and/or those who appear to be clinically ill.
Screen all patients with suspected sepsis with a lactate, obtained from a fresh blood sample.
Utilize hospital screening tools and bundled order sets, whether on paper or within the electronic health record, to achieve desired consistency in process. Recognize the limitations of these tools and scoring criteria for the different categories of sepsis.
Reassess patients frequently - they’re critically ill! Often, patients may become hypotensive 30-60 minutes after antibiotic administration.
Hospital systems are under great pressure from CMS to implement very specific sepsis bundles, despite evidence showing lack of benefit. While this makes for great intellectual and political debate, students should follow hospital protocols.
With your concern for sepsis, and what you know about this topic, you ask the nurse to draw blood for a point of care lactate, CBC, and 2 sets of blood cultures, in addition to other standard blood work. You also order a PA and lateral chest X-ray and request a urine sample when he is able. You ask the patient about his symptoms, he tells you he is tired and does not feel well, but he is only oriented to self and is unable to provide much more useful information. Patient’s family arrive at bedside. They state he is normally fully oriented and a good historian. The lactate returns at 4.6. Astutely, you order a 30 ml/kg fluid bolus and broad-spectrum antibiotics. CXR shows no acute process. Urinalysis is positive for large leukocyte esterase, nitrites, and 50-100 WBCs. After the fluid bolus, the patient’s BP improves to 124/70, HR is 105 bpm, and repeat lactate is 2.2. Patient is given a dose of IV ceftriaxone and admitted to the hospital for further management of his sepsis.
- PMID: 25272316, The ARISE Investigators and the ANZICS Clinical Trials Group. Goal – directed resuscitation for patients with early septic shock. N Engl J Med 2014, 371:1496-1506
- PMID: 11594901, Ferreira, Flavio; et al. “Serial Evaluation of the SOFA Score to Predict Outcome in Critically Ill Patients”, Oct 10, 2001. JAMA. 2001;286(14):1754-1758. doi:10.1001/jama.286.14.1754
- PMID: 24717459, Ferrer R, et al. Empiric antibiotic treatment reduces mortality in severe sepsis and septic shock from the first hour: results from a guideline-based performance improvement program. CCM. 2014
- PMID: 25585332, Gomez, Hernando; Kellum, John. “Lactate in Sepsis”. Jan 13, 2015. JAMA. 2015;313(2):194-195. doi:10.1001/jama.2014.13811.
- PMID: 20179283, Jones AE, Shapiro NI, Trzeciak S, et al. Lactate Clearance vs Central Venous Oxygen Saturation as Goals of Early Sepsis Therapy: A Randomized Clinical Trial. JAMA. 2010;303(8): 739-746
- PMID: 16625125, Kumar A, Roberts D, Wood KE. Duration of hypotension before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006; 34:1589-1596
- PMID: 25776532, Mouncey PR, Osborn TM, Power GS, et al for the ProMISe trial investigators. Trial of early, goal – directed resuscitation for septic shock. N Engl J Med 2015: DOI: 10.1056/NEJMoa1500896
- PMID: 24635773, ProCess Investigators, Yealy DM, Kellum JA, Juang DT, et al. A randomized trial of protocol – based care for early septic shock. N Engl J Med 2014; 370(18):1683-1693
- PMID: 21572327, Puskarich MA, Trzeciak S, Shapiro NI. Association between timing of antibiotic administration and mortality from septic shock in patients treated with a quantitative resuscitation protocol. Crit Care Med 2011; 30:2066-71
- PMID: 30015667, Rhee C et al. Compliance with the national SEP-1 quality measure and association with sepsis outcomes: A multicenter retrospective cohort study. Crit Care Med 2018 Jun 28.
- PMID: 26903338, Singer M, Deutschman CS, Seymour CW, et al. The Third International Consensus Definitions for Sepsis and Septic Shock (Sepsis-3). JAMA. 2016;315(8):801–810. doi:10.1001/jama.2016.0287
- http://survivingsepsis.org/SiteCollectionDocuments/Surviving-Sepsis-Campaign-Hour-1-Bundle.pdf (accessed March 25, 2019)