Impact of Prehospital Care on Outcomes in Sepsis

A Systematic Review

Michael A Smyth, MSc; Samantha J Brace-McDonnell, MSc; Gavin D Perkins, MD


Western J Emerg Med. 2016;17(4):427-437. 

In This Article


Very few, if any, EMS systems are capable of delivering the entire initial resuscitation bundle advocated by the Surviving Sepsis Campaign guidelines.[1] Most EMS systems lack the capability to draw blood and analyze the required parameters; in addition some of the technical skills required, such as central line placement, will be beyond the scope of many non-physician providers. It is therefore unreasonable to expect EMS systems to be able to deliver all elements of the initial resuscitation bundle. However, key interventions, such as oxygen therapy, antibiotic administration, fluid resuscitation and measuring venous lactate are possible. Despite the ability of EMS to deliver the aforementioned, recent hospital trials[32–34] have brought into question several of the EGDT objectives. We therefore need to examine carefully the need to extend EMS scope of practice to deliver those elements not routinely practiced, such as measuring venous lactate and administering antibiotics.

Prehospital recognition of sepsis is challenging.[8,27,35] The limited evidence identified suggests the initiation of treatment by EMS may lead to improved process outcomes, i.e. reduces time taken to achieve initial resuscitation targets but is not necessarily associated with improved clinical outcomes.

There is currently no evidence addressing impact of prehospital oxygen therapy in sepsis. The ARISE,[33] ProCESS[32] and ProMISe[34] trials have all suggested that the need to rigidly adhere to EGDT may be overstated. Furthermore, a systematic review by Sterling et al.[36] indicates that antibiotic administration within the first three hours is not associated with improved patient outcomes.

One study[29] identified during this review suggests that prehospital antibiotics may reduce mortality (OR 0.56 (95% CI [0.32–1.00]), p=0.049); however, this study was published in abstract only and enrolled a limited number of patients (n=198). We cannot therefore be confident that prehospital antibiotics would improve outcomes. The PHANTASi trial (NCT01988428) will hopefully provide further evidence to determine if EMS systems should extend clinical practice to deliver prehospital antibiotic therapy in cases of suspected sepsis.

Fluid therapy is an established clinical practice in many EMS systems. Seymour et al.[31] identified that prehospital fluid therapy was associated with both reduced organ failures (OR 0.58, 95% CI [0.34–0.98]) and mortality (OR 0.46, 95% CI [0.23–0.88]); however, the mean volume of fluid administered was only 500ml, considerably below what would normally be administered as part of the initial resuscitation bundle (30mL/kg).[1] This led the authors to question if the reduced mortality was due to the small volume of fluid or indeed if it was associated with process improvements secondary to prehospital recognition of sepsis. The latter argument is strengthened by their finding that placement of an intravenous catheter, without any fluid being administered, was also associated with reduced hospital mortality (OR 0.31, 95% CI [0.17–0.57]).[31]

One further aspect that has not been examined is the influence of EMS system design. Internationally, two distinct EMS systems, the EMT/paramedic (Anglo-American) model and physician (Franco-German) model are observed. Typically physician responders might be expected to have higher clinical acumen than paramedics/EMTs as a result of their longer, more in-depth education and training. In addition they may have greater scope to initiate a broader range of interventions, as well as direct admission to specialist services. These factors could improve recognition and indeed treatment of sepsis before arriving at hospital.

Eight of the included studies were conducted in EMT/paramedic EMS systems[5,11,25–29,31] with a single study, published in abstract only, conducted in a physician-based EMS system.[30] Studies conducted in both system designs suggested reduced times to interventions; however, Bayer et al.[30] did not publish data addressing mortality, ICU admission nor length of stay in their EMS physician-based study. Although Bayer et al.[30] reported a high proportion of suspected prehospital sepsis cases were later confirmed in the hospital, they did not report data concerning missed cases making it impossible to determine if EMS physicians are able to accurately identify sepsis patients out of the hospital. Bayer et al.[30] did however report a larger mean fluid volume (2.5l intravascular fluid (IQR 1.5–3.0l)),[30] than in the paramedic-based study (mean volume 500mL (IQR 200–1000mL)) reporting this outcome,[31] which may reflect greater understanding of beneficial treatments. With such limited data it is not possible to draw any meaningful conclusions concerning the impact of EMS physicians on outcomes in sepsis.