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


This systematic review addresses the impact of prehospital care on outcomes among patients with sepsis. The review adopted the Grading of Recommendations, Assessment, Development and Evaluation (GRADE) methodology[17] and is reported consistent with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations.[18]

Inclusion Criteria

Studies were eligible for inclusion if they reported the impact of prehospital care among adult patients with suspected sepsis (including severe sepsis and septic shock). Outcomes of interest include time to early goal-directed therapy (EGDT) related targets, admission to intensive care unit (ICU), length of stay and mortality. We included conference proceedings/meeting abstracts to capture gray literature.

Search Strategy

Electronic Searches. We systematically searched MEDLINE, EMBASE, CINAHL, the Cochrane Library and PubMed. No language restrictions were employed.

Search Terms/Search Strategy

Search strategies were based upon the terms below: (Sepsis OR septic OR septic?emia OR systemic adj inflammatory adj response adj syndrome OR SIRS OR septic adj shock OR hypotension adj induced adj hypoperfusion OR cryptic adj shock OR bacterial adj infection) AND (emergency adj medical adj service OR EMS OR HEMS OR emergency adj medical adj technician OR EMT OR paramedic OR pre-hospital OR prehospital OR pre adj hospital OR out-of-hospital OR out adj of adj hospital OR OOH OR Ambulance).

The initial MEDLINE search was conducted in July 2014 and adapted for each subsequent database. The searches were repeated in June 2015 to identify recent publications.


We contacted subject experts and scrutinized reference lists of included manuscripts in order to identify any missed studies.

Data Collection And Analysis

Study Selection. Study selection occurred in two stages. First, two reviewers (MAS and SJBM) independently reviewed each citation and abstract against the inclusion criteria. Citations rated as 'include' by either reviewer were retained; citations rated as 'exclude' by both reviewers were rejected. Second, full manuscripts of retained citations were independently screened by two reviewers (MAS and SJBM) who rated each manuscript as 'include,' 'maybe,' or 'exclude' against the inclusion criteria. If both reviewers rated a manuscript as 'include' it was included for critical appraisal. If both reviewers rated a manuscript as 'exclude' it was automatically rejected. If the two reviewers had differing opinions, the reviewers discussed the manuscript in order to achieve consensus. If the reviewers were unable to agree following discussion, a third independent reviewer (GDP) was available to adjudicate.

Risk Of Bias

For randomized controlled trials, we assessed risk of bias across the following domains: lack of allocation concealment, lack of blinding, incomplete accounting of patients and outcome events, selective outcome reporting bias and other limitations such as stopping a trial early for benefit. For observational studies, bias was assessed across the domains of failure to develop and apply appropriate eligibility criteria (inclusion of control population), flawed measurement of exposure and outcome, failure to adequately control confounding and incomplete follow up.

All papers were assessed across their respective domains with each being categorized as either high risk, low risk or level of risk unclear as per GRADE recommendations.[19] We considered studies categorized as high risk in any domain to be at high risk of bias overall. Studies categorized as low risk across all domains were considered to be at low risk of bias overall. Studies with a combination of low and unclear risk across domains were considered to have an unclear risk of bias overall.

Quality Of Evidence

We determined quality of evidence according to the GRADE framework. Study design informed initial quality presumptions; randomized controlled trials were initially presumed to be 'high quality,' while observational studies (non-randomized studies) were initially presumed to be 'low quality.' Two reviewers (MAS and SJBM) appraised each paper across the five core GRADE domains of risk of bias,[19] inconsistency,[20] indirectness,[21] imprecision[22] and other considerations (including publication bias).[23] If any concerns were identified quality of evidence was adjusted downward. Similarly, quality could be adjusted upward if, for example, a large treatment effect or dose response was noted, that subsequently raised confidence in the estimate of effect.[24] Ultimately each study is rated as follows:

  • High quality: We are very confident that the true effect lies close to that of the estimate of effect.

  • Moderate quality: We are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of effect, but there is a possibility that it is substantially different.

  • Low quality: Our confidence in the effect is limited: the true effect may be substantially different from the estimate of the effect.

  • Very low quality: We have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.