Key points

Introduction

Emergency medical services (EMS) personnel frequently provide care for individuals with time-critical illnesses and injuries. One such condition is sepsis, which represents a broad spectrum of clinical presentations requiring early recognition and rapid intervention. Interventions such as the administration of antibiotics and intravenous (IV) fluids within the first few hours have been linked to lower mortalities. In the United States, EMS systems have a long-standing tradition of care coordination. For example, trauma victims are conveyed speedily to designated trauma centers and victims of ST-segment elevation myocardial infarction (STEMI) are delivered to waiting cardiac catheterization laboratories. Because sepsis represents a distinct medical entity that would benefit from timely medical intervention, it logically follows that a systematic approach to prehospital recognition and treatment would benefit this distinct group of patients. That said, the recognition of a sepsis syndrome during the prehospital phase of care is far more complex than teaching EMS providers to recognize varying degrees of hemodynamic instability.

Modern EMS systems incorporate a variety of professionals, and each level of EMS provider has been trained to a different level of understanding with respect to human anatomy and physiology. Basic emergency medical technicians (EMTs) can interpret abnormalities in vital signs but might not be familiar with the underlying physiology. Paramedics understand the physiologic implications of a septic state but might not appreciate subtle or occult presentations of sepsis in immunosuppressed or chronically ill individuals. Because early recognition and intervention are associated with a decreased mortality, it is imperative to engage EMS systems in a comprehensive approach to the treatment of sepsis. This article explores current practices and available medical decision-making tools, job aids, and point-of-care (POC) tests in order to articulate an evidence-based approach to the prehospital recognition of sepsis.

Introduction

Emergency medical services (EMS) personnel frequently provide care for individuals with time-critical illnesses and injuries. One such condition is sepsis, which represents a broad spectrum of clinical presentations requiring early recognition and rapid intervention. Interventions such as the administration of antibiotics and intravenous (IV) fluids within the first few hours have been linked to lower mortalities. In the United States, EMS systems have a long-standing tradition of care coordination. For example, trauma victims are conveyed speedily to designated trauma centers and victims of ST-segment elevation myocardial infarction (STEMI) are delivered to waiting cardiac catheterization laboratories. Because sepsis represents a distinct medical entity that would benefit from timely medical intervention, it logically follows that a systematic approach to prehospital recognition and treatment would benefit this distinct group of patients. That said, the recognition of a sepsis syndrome during the prehospital phase of care is far more complex than teaching EMS providers to recognize varying degrees of hemodynamic instability.

Modern EMS systems incorporate a variety of professionals, and each level of EMS provider has been trained to a different level of understanding with respect to human anatomy and physiology. Basic emergency medical technicians (EMTs) can interpret abnormalities in vital signs but might not be familiar with the underlying physiology. Paramedics understand the physiologic implications of a septic state but might not appreciate subtle or occult presentations of sepsis in immunosuppressed or chronically ill individuals. Because early recognition and intervention are associated with a decreased mortality, it is imperative to engage EMS systems in a comprehensive approach to the treatment of sepsis. This article explores current practices and available medical decision-making tools, job aids, and point-of-care (POC) tests in order to articulate an evidence-based approach to the prehospital recognition of sepsis.

Existing triage tools

In 2009 and 2010, researchers from Harborview Medical Center attempted to quantify EMS providers’ understanding of sepsis. Providers from 3 EMS agencies, at all levels of EMS education and training, participated in an online 10-question survey focused on sepsis recognition. The study population included firefighter (FF)//EMTs, other EMTs, and paramedics. Seven hundred eighty-six EMS providers completed the survey: 408 FF/EMTs (52%), 276 other EMTs (21%), and 102 paramedics (13%). Almost all (97%) of the participants had “heard of sepsis” and appreciated its association with increased in-hospital mortality. However, knowledge gaps were found when participants were asked about their understanding of sepsis. EMTs were less likely than paramedics to identify the correct definition of sepsis, and this finding persisted following logistic regression analysis. Importantly, 55% of respondents agreed that EMS providers could play a role in the early identification of patients at risk for sepsis. These results lend credence to the idea of a systemic, protocolized approach to sepsis care. Important limitations surfaced in the analysis of the study. Each EMS system has different training programs, and the results may not be readily generalizable. The survey did not incorporate clinical scenarios, and researchers were therefore unable to test a provider’s ability to engage in more complex medical decision making. However, paramedics’ broad and more refined familiarity with sepsis suggested that these providers could be integrated into more specific and widespread prehospital treatment strategies.

Guerra and colleagues examined the utility of an aggressive, goal-directed, prehospital sepsis protocol. Before this study’s results are examined, it is helpful to appreciate aspects of the regional EMS system that made such important research possible. First, physicians from a single group staffed all the emergency departments (EDs) involved in the study. The EMS system used an “all-advanced life support” (all-ALS) model of care, meaning that each ambulance crew included at least one credentialed paramedic. Finally, medical direction was provided by board-certified emergency physicians affiliated with area hospitals and EMS system. Although such a collaborative arrangement might seem logical, the reality of physician oversight in EMS is far less consistent. This study featured a rather fortuitous pairing of engaged medical direction and an all-ALS system. Although basic life support (BLS) providers have a pivotal role to play in terms of recognition, EMTs are not often authorized to insert IV lines or administer medication therapy. The authors used an evidence-based triage screening tool and incorporated POC lactic acid testing into the prehospital sepsis algorithm. Patients who met predefined prehospital triage parameters were directed into a “sepsis alert” protocol. Essentially, patients had to fulfill 4 criteria: (1) age greater than 18 years and not pregnant, (2) presence of 2 systemic inflammatory response criteria, (3) suspected infection, and (4) hypoperfusion manifested by prehospital systolic hypotension or an elevated lactic acid level (>4 mmol/L). The study population comprised 112 patients. Paramedics correctly identified 32 of 67 patients (48%) who were subsequently diagnosed as having severe sepsis. Patients dropped out of the “sepsis alert” protocol for reasons apart from provider familiarity with the sepsis definition. Of the 35 septic patients who were not identified by EMS personnel, 5 had vital signs that did not meet criteria. Thirteen additional septic patients were identified during their ED stay as a result of an elevated white blood cell count. In an unadjusted analysis that examined survival as a primary endpoint, patients classified by EMS personnel as having severe sepsis had an odds ratio of 3.19 in favor of survival (95% confidence interval [CI], 1.14–8.88). The authors also reported a lower mortality for EMS patients identified with the sepsis alert protocol as opposed to those not identified with the sepsis alert protocol. This study affirms that paramedic-level providers can identify severe sepsis and initiate appropriate treatment. Furthermore, the use of an “alert” protocol could have several positive downstream effects. Patients identified as having a time-sensitive condition are more likely to undergo timely interventions. Potentially negative effects of this specific protocol included false-positive sepsis activations and increased costs associated with paramedic education and training. It is logical to infer, however, that additional investment in training costs would be offset by the benefits gleaned from aggressive prehospital resuscitation of septic patients.

A recent retrospective study examined more than 66,000 EMS encounters for the purpose of developing a prehospital sepsis score (PRESS). This score can be used by a broad range of EMS providers to facilitate the early treatment and rapid transport of septic patients to definitive care. The investigators looked at dispatch and patient characteristics associated with adverse outcomes. In an urban, 2-tiered (BLS and ALS) EMS system, a few variables emerged that were persistently linked to the inpatient diagnosis of severe sepsis or septic shock: advanced age, hot tactile temperature, low systolic blood pressure, and low oxygen saturation. Compared with other currently available triage schemes, the PRESS relies mostly on clinical assessment and 9-1-1 triage. Indeed, prehospital POC lactate analysis is not readily available in most EMS agencies, and compliance with rigorous protocols may depend at least in part on the ambulance crew’s level of training. The sensitivity of the score was 91% in the derivation group and 78% in the validation group. When the predefined threshold of 2 or more points is met, the sensitivity of the PRESS increases to 86%.

This scoring technique has several important limitations. Patients in the derivation group had already met criteria for the systemic inflammatory response syndrome (SIRS). Patients deemed “at risk” for sepsis had to meet all the criteria in the prehospital setting: elevated heart rate, elevated respiratory rate, and systolic blood pressure less than 100 mm Hg. Therefore, the PRESS’s respectable sensitivity would suffer following validation in an external cohort.

Early warning scores

The concept of an early warning score (EWS) was developed in an effort to identify deteriorating hospitalized patients rapidly. These scores, most often a composite of physiologic and/or laboratory values, have become the standard of care in many parts of the world. However, the utility of an EWS in the prehospital environment is less well established. There is precedent for timely prehospital intervention directed at time-sensitive conditions such as stroke, STEMI, and penetrating trauma. It follows that sepsis, which is also a time-critical illness, would benefit from the same treatment strategy. Although early treatment of sepsis shows a greater morbidity and mortality benefit than early thrombolysis and balloon angioplasty for patients with acute coronary syndrome, until recently it received less attention and research funding. There is current interest in developing a prehospital EWS to identify patients with critical illness and facilitate their early access to appropriate, definitive care.

Developing an EWS for sepsis presents unique challenges. The first challenge is the lack of a consensus definition for sepsis. Sepsis is a clinical syndrome for which the inclusion and exclusion criteria have been changed over the years. The most recent definition by the International Committee on Sepsis was published in Journal of the American Medical Association in 2016. The authors of that article anticipate that, as one becomes more knowledgeable about the biology behind the clinical manifestations of sepsis, the definition of the disease will be refined. For example, the term “severe sepsis” is now considered redundant, so it has been removed from the clinical spectrum. Changing the nomenclature poses a problem when it comes to research and establishing evidence-based practices. Much of the data produced so far used “severe sepsis” criteria in determining whether an intervention was successful. The lack of a clear definition, as well as a lack of consensus on measurable outcomes, makes the existing body of literature heterogeneous and difficult to compare. Establishing a universal EWS under these circumstances becomes complex.

The first prehospital EWS was the Rapid Acute Physiology Score (RAPS), which is an abbreviation of the APACHE-II (Acute Physiology and Chronic Health Evaluation) score. RAPS was developed and tested for the aeromedical transport of critically ill patients. There are several in-hospital EWS currently in use. Attempts have been made to translate in-hospital EWSs into the prehospital environment via scoring systems such as the Modified EWS (MEWS), with varying success.

Guerra and colleagues developed a “sepsis alert” protocol that incorporates modified SIRS criteria along with prehospital lactate measurements to identify severe sepsis patients. The Robson screening tool takes into account temperature, heart rate, respiratory rate, mental status, plasma glucose concentrations, and a history suggestive of new infection. The BAS tool is based on the following vital signs: oxygen saturation less than 90%, respiratory rate greater than 30, and systolic blood pressure less than 90 mm Hg (BAS 90–30–90).

In 2015, German researchers developed the Prehospital Early Sepsis Detection score (PRESEP), which gives weighted values to abnormal vital signs ( Table 1 ). The cutoff indicating potential septic disease is greater than or equal to 4. The authors compared their scoring system with MEWS, BAS 90-30-90, and the Robson screening tool. The PRESEP score performed better than MEWS and BAS 90-30-90 in terms of sensitivity, specificity, as well as positive and negative predictive value. The Robson score, however, had a higher sensitivity and negative predictive value but a lower specificity and positive predictive value. It is important to note that the German model of prehospital care differs significantly from the American model. In Germany, ambulance crews typically include physicians. In addition, in the United States, body temperatures are not always measured by prehospital care providers, which could limit the generalizability of a scoring system that requires it.

Table 1

Only gold members can continue reading. Log In or Register to continue

Share this:

• Click to share on Twitter (Opens in new window)
• Click to share on Facebook (Opens in new window)

Related

Related posts:

Severe Sepsis Care in the Emergency Department An Introduction to the Most Complex Disease in Emergency Medicine End Points of Sepsis Resuscitation The Past, Present, and Future of the Centers for Medicare and Medicaid Services Quality Measure SEP-1 Appropriate Antibiotic Therapy Sepsis Resuscitation in Resource-Limited Settings