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


Antibiotic Therapy

Three studies indicate that ED antibiotic therapy is administered 30–50 minutes sooner if EMS identify sepsis and inform the receiving clinician of their diagnosis.[5,11,26] However, this finding is not universal – Guerra et al.[27] failed to identify any significant reduction in time to antibiotic therapy (pre-alert: 72.6 minutes Standard Deviation (SD) 59.3 minutes) vs no pre-alert: 98.5 minutes (SD 89.9 minutes), p=0.07). None of the studies concerned with prehospital recognition of sepsis, without concomitant administration of antibiotics, were able to identify any significant improvement in length of stay[11,25,27] or mortality.[11,25–28]

Two studies[29,30] address prehospital administration of antibiotic therapy. Chamberlain[25] reported that antibiotics were delivered 3.4+-2.6 hours sooner while Bayer et al.[30] noted that among EMS sepsis patients median time to antibiotics was 19 minutes (IQR 18–24 minutes) from initial emergency call (time of administration was estimated to commence 10 minutes after arriving at scene). Bayer et al.[30] do not report interval to hospital nor report time to antibiotics in the ED. Chamberlain[29] suggests that prehospital antibiotic therapy leads to reduced intensive care unit (ICU) stay (Mean ICU stay: 6.8±2.1 days (intervention) vs 11.2±5.2 days (control), p=0.001) and reduced mortality (28-day mortality: 42.4% (intervention) vs 56.7% (control); odds ratio (OR) 0.56; 95% CI [0.32–1.00]). Bayer et al.[26] did not report mortality, ICU admission or length-of-stay data.

Intravascular Fluid Therapy

Band et al..[26] reported that arrival by EMS reduces time to initiation of intravascular fluid therapy when compared with those who arrive by privately owned vehicle (POV, EMS: 34 minutes [IQR 10–88 minutes] vs POV: 68 minutes, IQR 25–121 minutes, p≤0.001), but did not improve mortality (adjusted risk ratio [RR] 1.24; 95% CI [0.92–1.66]). Similarly Bayer et al.[30] noted that among EMS sepsis patients median time to initiation of Intravenous fluids was 19 minutes (IQR 18–24 minutes) from initial emergency call (time of administration was estimated to commence 10 minutes after arriving at scene), with patients receiving an average of 2.5l intravascular fluid (IQR 1.5–3.01) until admission to the ED. A third study by Guerra et al.[27] indicated that early identification of sepsis by EMS was not associated with improved six-hour fluid resuscitation targets in the ED (EMS pre-alert: 42.97 cc/kg (SD 33.23cc/kg) vs no EMS pre-alert: 35.17cc/kg (SD 26.81 cc/kg, p=0.30).

The only study to demonstrate a positive impact following prehospital fluid administration among sepsis patients indicated that prehospital fluids were associated with reduced likelihood of organ failures (adjusted OR 0.58; 95% CI [0.34–0.98]) and reduced hospital mortality (adjusted OR 0.46; 95% CI [0.23–0.88]), but not reduced ICU admission (adjusted OR 0.64; 95% CI [0.37–1.10]).[31] The median volume of prehospital fluid administered in this study was 500mL (IQR 200–1000mL).

Early Goal Directed Therapy (EGDT) Targets

Femling et al.[11] reported that patients who arrived at the ED via EMS had shorter time to central line placement (required for central venous pressure monitoring) than those who arrived by other means (EMS: 200 minutes [IQR 89–368 minutes] vs non-EMS: 275 minutes [IQR 122–470 minutes], difference 75 minutes, p<0.01), while Guerra et al.[27] noted that when EMS provided a sepsis pre-alert to the hospital the advance notification it did not impact the decision to place a central venous catheter (EMS pre-alert: 61% vs no EMS pre-alert: 68%, p=0.54). Although Seymour et al.[28] reported that higher proportion of patients achieved a SVCO2>70% within six hours when EMS initiated fluid therapy prior to arriving at the ED, the unadjusted risk ratio found no evidence of a difference (EMS IV fluids: 13/24 (54%) vs no IV fluids: 9/25 (36%), Unadjusted RR 1.5, 95% CI [0.8–2.9]). This same study also identified no improvement in time to MAP>65mmHg (EMS IV fluids: 17/24 (70%) vs no IV fluids: 12/26 (44%), unadjusted RR 1.53 (95% CI [0.9–2.65]), and time to CVP>8 mmH20 (EMS IV fluids: 15/25 (60%) vs no IV fluids: 17/24 (70%), unadjusted RR 1.2 (95% CI [0.8–1.8]).[28]

Studnek et al.[5] reported that if patients arrived by EMS they had shorter times to EGDT than if they arrived by other means (EMS: 119 minutes vs non-EMS: 160 minutes, SD/range not reported, p=0.005). Furthermore, among EMS-transported patients, if EMS documented suspicion of sepsis then time to EGDT was shorter than if they did not document suspicion of sepsis (documented suspicion: 69 minutes vs not documented: 131 minutes, SD/range not reported, p=0.001). McClelland et al.[25] similarly reported that time to delivery of the 'Sepsis 6' (administration of supplemental oxygen, intravenous fluids, antibiotics, measurement of venous lactate, urine output, and drawing blood to identify causative pathogen) was shorter if EMS identified sepsis prior to arrival at hospital (EMS identified: mean 205 minutes [SD 271 minutes, range 10–720 minutes] vs not identified: mean 120 minutes [SD 110, 17–450 minutes]). These data points include one outlier where the fluid balance chart was not started for 12 hours. Excluding this case, the mean time to delivery of the 'Sepsis 6' would be 76 minutes (SD 95 minutes, range 10–240 minutes).