Article

The early chain of care in bacteraemia patients: Early suspicion, treatment and survival in prehospital emergency care

a b s t r a c t

Introduction: Bacteraemia is a first stage for patients risking conditions such as septic shock. The primary aim of this study is to describe factors in the early chain of care in bacteraemia, factors associated with increased chance of survival during the subsequent 28 days after admission to hospital. Furthermore, the long-term outcome was assessed.

Methods: This study has a quantitative design based on data from Emergency Medical Services (EMS) and hospi- tal records.

Results: In all, 961 patients were included in the study. Of these patients, 13.5% died during the first 28 days. The EMS was more frequently used by non-survivors. Among patients who used the EMS, the suspicion of sepsis al- ready on scene was more frequent in survivors. Similarly, EMS personnel noted the ESS code “fever, infection” more frequently for survivors upon arriving on scene. The delay time from call to the EMS and admission to hos- pital until start of antibiotics was similar in survivors and non-survivors. The five-year mortality rate was 50.8%. Five-year mortality was 62.6% among those who used the EMS and 29.5% among those who did not (p b 0.0001). Conclusion: This study shows that among patients with bacteraemia who used the EMS, an early suspicion of sep- sis or fever/infection was associated with improved early survival whereas the delay time from call to the EMS and admission to hospital until start of treatment with antibiotics was not. 50.8% of all patients were dead after five years.

(C) 2018

Introduction

The term Bacteraemia encompasses different types of underlying se- vere infections. Bacteraemia is a first phase for patients risking sepsis or progress into more severe conditions – such as septic shock- associated with a rapid course, high mortality and patient suffering [1-3]. Research indicates that personnel in the Emergency Medical Services (EMS) en- counter a significant number of patients suffering from sepsis [2,4,5]. It is therefore essential for EMS personnel to make a proper assessment of patients with bacteraemia as early as possible, and to interpret pa- tients’ vital signs and symptoms correctly to identify those at risk of organ dysfunction [6].

* Corresponding author at: University of Boras, PreHospen – Centre for Prehospital Research, SE-501 90 Boras, Sweden.

E-mail address: [email protected]. (H. Andersson).

Early suspicion and prevention of the progression of bacteraemia involve assessing vital signs, for example blood pressure, heart rate, re- spiratory rate, oxygen saturation, Degree of consciousness and body temperature, as well as vague symptoms such as deteriorated general condition, altered mental status, dyspnoea, muscle weakness, pain and Gastrointestinal symptoms. Patients suffering from one or several of the above-mentioned symptoms should be assessed for possible infec- tion by EMS personnel [7]. Simultaneously, normal vital signs and vague symptoms may make it difficult to identify patients with severe infections. In addition to early suspicion, timely treatment is also of great importance, including High-flow oxygen, intravenous fluids and an early start with effective antibiotics [8-10]. This is essential since de- lays have been reported to be associated with an increased mortality [11-13].

Few studies have been conducted on suspicion and treatment of life- threatening infections in the prehospital setting. Thus, evidence is

https://doi.org/10.1016/j.ajem.2018.04.004

0735-6757/(C) 2018

lacking on how prehospital emergency care could influence the progno- sis, for example in patients with sepsis [14]. However, studies show that EMS personnel do not always suspect life-threatening infections such as sepsis and their assessment and treatment may therefore be incomplete or delayed [15-18].

The emphasis on EMS personnel’s ability to identify and assess vital signs and vague symptoms of bacteraemia is a result of the develop- ment of the EMS system and the focus on advanced care and treatment [19]. In Sweden, the EMS system is staffed with Ambulance nurses, a rel- atively new profession in the prehospital field [20]. This is a result of Swedish regulations and legislation stipulating that personnel in the EMS must have the competence to administer drugs [21]. EMS person- nel need to have an advanced medical education to be able to recognize the symptoms of bacteraemia and other conditions.

EMS personnel thus have an important function in the early chain of care for patients suffering from bacteraemia. They may also play an es- sential role in advancing sepsis care [22]. The aim of this study was to describe factors in the early chain of care in patients with bacteraemia that were associated with an increased chance of survival during the subsequent 28 days after admission to hospital. The hypothesis was that EMS personnel’s early suspicion of bacteraemia and no delay in treatment with antibiotics both are associated with increased chances of survival.

Therefore, patients who died during the 28 days after arrival in hospital will be compared with patients who survived the first 28 days. As a secondary endpoint, the risk of death during five years of follow up was assessed.

Methods

Study design

This study is a quantitative retrospective record audit of EMS and hospital medical records. A trained chart abstractor conducted the data collection. Data from EMS and hospital records was registered by using a standardized study protocol including the following factors: gender, age, use of EMS, growth of bacteria and other microorganisms in blood cultures, organs affected, vital signs, symptoms, delays, levels of priority, suspicion of fever/infection and/or sepsis, treatment and complications.

Inclusion criteria

The inclusion criteria were: 1) positive blood culture for bacteria;

2) patient admitted by the EMS or by own application for help to one of the nine Emergency Departments (EDs) in the region; and 3) blood culture and analysis between February 1, 2012 and April 30, 2012. Patients were thus included in the study regardless of whether they were transported by the EMS or not.

Exclusion criteria

The exclusion criteria were: 1) age b 18 years and 2) cases in which neither the EMS nor the ED were involved in the care of the patient.

The EMS organization and setting

The study was carried out in a region of Western Sweden with an EMS catchment population of 1.6 million inhabitants where five EMS systems with 60 ambulances are responsible for prehospital emergency care. All ambulances were staffed with an emergency medical techni- cian and at least one registered nurse with or without specialist educa- tion in prehospital emergency care. This study also includes the EDs at the nine hospitals: three university, three central, and three district hospitals.

Triage system

All patients, pre- and in-hospital, were assessed according to the Rapid emergency triage and Treatment System (RETTS) [23,24]. The RETTS is an assessment tool that determines the patient’s need of emergency care, classified with a colour that indicates priority. Red and orange mean immediate care; yellow means urgent care and green means care within a reasonable time but not urgent. Prioritization according to the RETTS is based on Emergency Signs and Symptoms (ESS) and six vital signs (blood pressure, heart rate, respiratory rate, oxygen saturation, degree of consciousness and body temperature).

Definition of primary and secondary endpoints

Death within 28 days after arrival in hospital was defined as the primary endpoint. This was determined according to the Swedish Register of Deaths [25]. The primary analyses in this study involved the relationships between death and early suspicion of bacteraemia, and between death and delay of treatment with antibiotics. The sec- ondary endpoint, five-year mortality, was defined as death during five years from the time of arrival in hospital at the index event.

Data analysis

Fisher’s exact test was used to evaluate proportions. Wilcoxon’s two-sample test was used to evaluate continuous variables. Tests with a p-value b 0.05 were considered statistically significant for the primary endpoint. The remaining analyses were regarded as sig- nificant if there was a p-value b 0.01. Two-tailed tests were applied. All analysis was carried out using the SAS software programme (SAS Institute, Cary, North Carolina, USA).

Data is reported for all patients and also separately for patients with true pathogens. However, this does not apply to non-EMS pa- tients in EDs. For this group, only the results for all patients are re- ported. In all analyses, patients who died within 28 days will be compared with those who survived. These comparisons will first be reported for all patients with bacteraemia and then for patients with true pathogens, with the exception, as stated above, of patients who did not use EMS.

In this study, “all patients” were defined as patients with a positive blood culture containing bacteria, fungus or mould. Bacteria were classified as being “true pathogens” or “commensals”. True pathogens are bacteria of high virulence that are not part of the normal human skin flora and known to highly correlate to clinical disease when identi- fied in blood cultures. Commensals are bacteria known to be part of the normal human skin flora, of low virulence, and mostly regarded as contaminants when found in blood cultures.

Ethical considerations

The Regional Ethical Committee in Gothenburg (Dnr 383-12) approved the study.

Results

In all, 961 patients fulfilled the inclusion criteria. Of these, 582 (60.6%) had true pathogens (Table 1). Among all patients, the 28- day mortality was 13.5% while it was 15.2% for patients with true pathogens.

Symptoms, organ affected, use of EMS, gender and age

Symptoms of fever such as shivering were significantly more fre- quent in survivors (p b 0.0001). This was found among all patients as well as among patients with true pathogens. Decreased conscious- ness was significantly more frequent in non-survivors. This was

Table 1

The most frequent true pathogens and commensals found in blood cultures.

Bacteria N

Coagulasenegative Staphylococcib 306

Escherichia colia 195

Streptococcus spp.a 99

Staphylococcus aureusa 94

Streptococcus pneumoniaea 50

Klebsiella spp.a 29

Enterococcus spp.a 23

Staphylococcus epidermisb 18

Other enterobacteriaceae spp.a 15

Proteus spp.a 8

Pseudomonas aeruginosaa 4

Bacteroides fragilisa 4

Citrobacter spp.a 4

Serratia marcescensa 4

Aerococcus urinaea 2

Eggerthella lentaa 2

Staphylococcus spp.a 2

Streptococcus anginosusa 2

a True pathogens.

b Commensals.

found among all patients (p b 0.0001) as well as among patients with true pathogens (p b 0.01). Among patients with true pathogens, the airway was the most frequent localization of infection. Infections originating in the urinary tract were significantly more frequent in survivors (p b 0.0001). This was found among all patients as well as among patients with true pathogens. The use of the EMS was signif- icantly more frequent among non-survivors. This was found among all patients (p b 0.0001) and among patients with true pathogens (p b 0.004). Survivors were on average about 10 years younger than non-survivors. This was found among all patients and among patients with true pathogens (Table 2).

Symptoms, signs, treatment and complications for EMS patients in the prehospital setting

A suspicion of sepsis by the EMS personnel, mostly the EMS nurse, already on scene was significantly more frequent among survivors. This was found among all patients (p b 0.02) as well as among those with true pathogens (p b 0.01). Similarly, EMS personnel noted the ESS code “fever, infection” already on scene significantly more frequently among survivors. This was found among all patients (p b 0.0001) as well as among those with true pathogens (p = 0.001). The delay time from call to the EMS as well as from arrival in hospital until start of antibiotics did not differ between survivors and non-survivors. This was found among all patients as well as among those with true pathogens. No other delay times, nor priority at the dispatch centre nor evaluation according to the RETTS differed between survivors and non-survivors. At the time of the EMS’ arrival on scene, the degree of consciousness and the median systolic blood pressure were both significantly higher in survivors among all pa- tients (p b 0.0001 respectively p = 0.001). The median body temper- ature was also significantly higher (p b 0.0001) in survivors among all patients and among those with true pathogens. Neither treatment nor complications before arrival at the hospital differed between survivors and non-survivors (Table 3).

Symptoms, signs and treatment for non-EMS patients in the ED

The suspicion of sepsis by the ED staff in the ED did not differ be- tween the two groups. Neither did the use of the ESS code “fever” differ significantly. The delay time from symptom onset to arrival at the hos- pital was similar in survivors and non-survivors. In contrast, the delay time from arrival in hospital until start of antibiotics was significantly longer among non-survivors (p = 0.04). In terms of clinical findings on arrival at the ED, the body temperature was significantly higher among survivors (p = 0.004). Otherwise, no difference was found between the two groups. In terms of treatment and action, the use of

Table 2

Symptoms, organ affected, use of EMS, gender and age (%).

All patients

p

True pathogens

p

28-day survival Yes

N = 846

No

N = 115

28-day survival Yes

N = 505

No

N = 77

Initial symptom

Dyspnoea (154,28)a

27.3

36.8

0.08

25.1

31.0

0.34

Fever, shivering

59.4

40.0

b0.0001

64.2

41.3

b0.0001

(94,25)

Nausea, vomiting, diarrhoea

26.6

30.3

0.45

30.9

30.5

1.00

(159,26)

Muscle weakness

11.2

17.9

0.11

14.1

20.7

0.23

(196,31)

Decreased consciousness

13.5

31.0

b0.0001

14.2

28.6

0.01

(187,31)

Severe pain

28.2

22.0

0.29

32.0

25.0

0.36

(171,33)

Organ affected

Airways

33.1

44.6

0.02

23.5

40.3

0.004

(27,5)a

Urinary tract

27.4

7.3

b0.0001

34.1

9.7

b0.0001

Abdomen

13.1

12.7

1.00

13.1

11.1

0.85

Brain

0.8

0.9

1.00

0.8

1.4

0.50

Heart

3.7

5.4

0.42

4.9

4.2

1.00

Skin

11.7

11.8

1.00

12.9

15.3

0.58

Use of EMS

(1,0)a

Yes

62.8

79.1

b0.0001

65.5

81.8

0.004

Age (years, median)

(0,0)a

73.0

82.0

b0.0001

74.0

84.0

b0.0001

Sex (%) (0,1)a

Women

47.2

39.1

0.11

46.1

33.8

0.049

a Number of patients with missing information including all patients in relation to 28-day survival (the two left columns).

Table 3

Symptoms, signs, treatment and complications for

EMS patients in the prehospital setting (%).

All patients

p

True pathogens

p

28-day survival

28-day survival

Yes

No

Yes

No

N = 531

N = 91

N = 331

N = 63

Delay (median, min)

Call-alert (9,1)a

5.0

4.0

0.91

4.0

5.0

0.54

Call alert-arrival place

16.0

14.5

0.08

16.0

15.0

0.43

(9,1)

Arrival place-start of transport

15.0

16.5

0.02

15.0

16.0

0.053

(18,1)

Start of transport-arrival hospital

19.0

17.0

0.39

21.0

19.0

0.71

(14,0)

Call alert-arrival hospital

64.0

60.0

0.82

64.0

62.0

0.73

(15,2)

Symptom onset-arrival hospital

3 h

2h

0.06

3 h

2 h

0.31

(461,78)

39 min

0 min

31.5 min

15.5 min

Arrival hospital-start of antibiotics

2 h

2h

0.55

2 h

2 h

0.67

(100,22)

49 min

55 min

45 min

54 min

Call to EMS-start of antibiotics

3 h

4h

0.49

3 h

4 h

0.62

(100,22)

45 min

6 min

43 min

2 min

Priority

0.32

0.78

(%)

Dispatch centre

1

26.9

31.9

26.5

23.8

(6,0)a

2

53.1

50.6

52.7

55.6

3

19.4

17.6

20.4

20.6

4

0.6

0

0.3

0.0

Suspicion of sepsis by EMS

(128,26)a

Yes

15.1

4.6

0.02

19.5

4.4

0.01

RETTS

(145,26)a

Red

21.5

27.7

0.94

20.2

25.5

0.54

Orange

42.2

30.8

44.6

27.7

Yellow

26.4

32.3

26.0

36.2

Green

9.8

9.2

9.1

10.6

ESS codeb (361,63)a

Fever/infection

42.9

7.1

b0.0001

47.0

9.5

0.001

Status on EMS arrival

Consciousness (%)

(25,11)a

Awake

90.7

75.0

b0.0001

90.8

82.1

0.046

reduced consciousness

8.1

13.8

8.5

14.3

Unconscious

1.2

11.2

0.6

3.6

Vital signs (median)

Heart rate

(beats/min) (17,3)a

100.0

100.0

0.22

102.0

100.0

0.37

Systolic blood pressure

134.5

126.0

0.001

134.0

125.0

0.002

(mm Hg) (25,3)

Oxygen saturation

93.0

91.0

0.01

93.0

92.0

0.02

(%) (19,9)

Respiration rate

24.0

24.0

0.94

22.0

23.0

0.98

(breaths/min) (71,13)

Body temperature

38.4

37.3

b0.0001

38.6

37.2

b0.0001

(?C) (84,13)

Treatment before arrival in hospital

(%)

Venous access

(33,6)a

52.4

62.4

0.10

54.6

66.1

0.12

Oxygen

55.4

64.2

0.15

53.5

58.2

0.56

(51,10)

ECG monitoring

33.3

41.2

0.20

32.0

46.4

0.045

(83,11)

Intravenous infusion

24.2

26.8

0.66

26.9

28.0

0.8

(92,20)

Complications before arrival in hospital

(%)

Cardiac arrest

(120,24)a

0.0

1.5

0.14

0.0

2.1

0.15

Hypotension

5.5

11.6

0.07

5.8

14.6

0.06

(116,22)

Heart failure

2.7

4.5

0.43

1.5

4.4

0.22

(117,24)

Hypoglycaemia

0.2

0.0

1.0

0.4

0.0

1.00

(121,24)

Pain

6.4

8.6

0.45

7.2

8.2

0.77

(112,21)

High fever

11.0

3.0

0.045

12.9

2.1

0.04

(113,24)

a Number of patients with missing information among all patients in relation to 28 day survival (the two left columns).

b ESS = emergency symptoms and signs.

Table 4

Symptoms, signs and treatment for non-EMS patients in EDa (%).

All patients

p

28-day

survival

Yes

No

N = 315

N = 24

Delay (median, min)

Symptom onset-arrival hospital (305,23)b

9h 44 min

9h 50 min

1.0

Arrival hospital-start of antibiotics

3h

5h

0.04

(74,9)

20 min

16 min

Priority in ED

(%)

Suspicion of sepsis in ED (118,14)b

Yes

20.4

30.0

0.44

RETTS (108,8)b

0.18

Red

8.7

0

Orange

47.6

43.8

Yellow

37.9

43.8

Green

5.8

12.5

ESS codec (119,8)b

Fever/infection

36.9

18.8

0.18

Status on ED arrival

Consciousness (%) (11,0)b

0.38

Awake

98.4

95.8

Reduced consciousness

1.6

4.2

Unconscious

0

0

Vital signs (median)

Heart rate

(beats/min) (10,2)b

97.0

99.5

0.93

Systolic blood pressure

136.0

134.0

0.14

(mm Hg) (29,0)

Oxygen saturation

96.0

96.0

0.12

(%) (16,0)

Respiration rate

20.0

24.5

0.49

(breaths/min) (84,6)

Body temperature

38.0

37.3

0.004

(?C) (10,1)

Treatment and action

(%)

Venous access

(18,2)b

87.8

90.9

1.0

Oxygen

15.7

25.0

0.34

(21,4)

ECG monitoring

14.4

41.7

0.003

(92,12)

Intravenous infusion

59.0

54.6

0.82

(19,2)

Antibiotics

62.7

50.0

0.26

(19,2)

a Patients with true pathogens are not reported due to a small number of patients.

b Number of patients with missing information in relation to 28-day survival.

c ESS = emergency symptoms and signs.

ECG monitoring was significantly more frequent among non-survivors (p = 0.003). Otherwise, there was no difference between the two groups (Table 4).

Five-year mortality

Information on five-year mortality was missing in eight patients (0.8%). The overall five-year mortality rate was 50.8%. Five-year mortal- ity was much higher among patients who used the EMS than in those who did not (62.6% versus 29.5%; p b 0.0001; Fig. 1) and it was much higher among patients aged N74 years (median) than in those who were younger (72.1% versus 30.5%; p b 0.0001; Fig. 2). There was no difference between men and women (52.8% versus 48.5%; p = 0.19) and there was no difference between patients with and without true pathogens (51.9% versus 49.0%; p = 0.42).

Discussion

The main finding in this study was that our first hypothesis was con- firmed. Thus, a suspicion of sepsis and a medical record note of “fever, infection” by the EMS personnel before arrival in hospital were both clearly associated with a decreased risk of death. However, the

hypothesis that the delay time from call to the EMS until start of antibi- otics would be associated with the early risk of death could not be con- firmed. A relatively small number of evaluated patients may partly contribute to this conclusion.

Bacteraemia is a collective term for different bacteria with different pathogenicity and different mortality rates, affecting people with differ- ent comorbidities and at different ages, with or without organ dysfunc- tion. High age in itself is a risk factor for case fatality, and the average age in this cohort was high. Organ dysfunction is another risk factor. Time to effective antibiotic treatment may be important for those having severe conditions such as sepsis with organ dysfunction and bacteraemia with agents with known high case fatality rates, such as infections with Staphylococcus aureus, Streptococcus pyogenes and Streptococcus pneumoniae [26].

The results show that patients who died used the EMS more fre- quently and this finding is in line with previous research [27]. The EMS has the assignment of providing immediate care and treatment to patients suffering from an acute illness. Therefore, the most reason- able explanation for this finding is that not only the sicker but also the older patients used the EMS [27,28]. The association between increasing age and increased risk of death is not controversial. Research has dem- onstrated that in most diseases increasing age is associated with an

Mortality (%) 70

60

50

40

30

20

10

0

0 1 2 3 4 5

Years

EMS patients Non-EMS patients

Fig. 1. Cumulative mortality during five years after arrival in hospital among patients with bacteraemia who used EMS and those who did not use EMS.

increased risk of death. This is probably best explained by the fact that there is an increased risk of organ dysfunction with increasing age, as reflected in more Severe comorbidities [29-31].

Apart from age, there were four observations that were associated

with chance of survival among patients who used the EMS. They were the level of: 1) consciousness, 2) systolic blood pressure, 3) oxygen sat- uration and 4) body temperature. Higher measured values for all these vital signs were associated with a better chance of survival. The results also indicated an increased survival rate if patients demonstrated initial symptoms of infection such as fever, shivering and severe pain or if the patient had a urinary tract infection. The association between a decrease in consciousness, blood pressure and oxygen saturation and an in- creased risk of death was not a surprising finding.

Mortality (%) 80

70

60

50

40

30

20

10

0

0.9 1.9 2.9 3.9 4.9

Years

Age < 74 years Age > 74 years

Fig. 2. Cumulative mortality during five years after arrival in hospital among patients with bacteraemia who were aged below or equal to 74 years (median) and those who were above 74 years of age.

It is more difficult to explain why a decreasing body temperature should be associated with an increased risk of death. One possible ex- planation is that patients with a decreased body temperature are unable to generate a febrile response to bacteraemia [32]. Another potential ex- planation is that the quality of care given to patients with increased body temperature is better [33]. Previous research has shown that ab- normal vital signs, for example altered mental status, hypotension and hypoxaemia are significant for diagnosing and treating patients with se- vere infections [2,28]. However, research also indicates that a severe in- fection can be described as manifesting a variety of physical symptoms, deterioration and communication difficulties [34]. Therefore, patients’ symptoms should be given attention together with vital signs [7,35]. This underlines the value of the EMS personnel’s ability to identify relevant conditions in the care of patients with bacteraemia.

The results imply that EMS personnel did not always suspect sepsis in cases that were later diagnosed as sepsis and where there were strong reasons to believe that the patient was suffering from a severe in- fection in the prehospital phase. These results are in line with earlier studies [10,15,16,18] and might be interpreted differently. Firstly, the EMS personnel may have suspected other conditions than severe infec- tions since the prehospital diagnosis of sepsis can be difficult to establish [36]. Another explanation is that suspicions of sepsis by the EMS person- nel were not documented in the medical records. This might be ex- plained by the work situation, for example, that the EMS personnel chose to take care of the patient rather than to document observations and care measures. However, the problem with incomplete documenta- tion is not unique for this study; it has been described in other prehospital studies as well [15-17]. Secondly, the results may be related to the RETTS assessment tool that is not used in countries other than Sweden and Norway. All the patients were assessed according to the RETTS for rapid identification of patients with emergent, life-threaten- ing conditions, based on vital signs and symptoms. Vital signs are essen- tial for the assessment of the patient. Research indicates that abnormal Vital parameters are the most common signs for the identification of pa- tients with sepsis [37]. On the other hand, the vital signs in this study were not within the frame of reference for immediate care except for the level of consciousness [24]. Simultaneously, Altered consciousness may result from other Serious conditions such as stroke [38]. This means that it may be difficult to rely only on measuring vital signs in identifying patients with sepsis. One reason for this is that nothing trig- gers suspicion of sepsis if the measurements are near normal or normal. This suggests the need for a new or an improved assessment tool de- signed to detect patients with severe infection such as sepsis. Finally, the results could be interpreted as showing the EMS personnel’s uncer- tainty when identifying patients with sepsis. This uncertainty may be influenced by factors such as limited medical knowledge, uncritical atti- tudes or lack of ability to analyse the patient’s condition correctly [6,39]. Consequently, this can lead to an incorrect assessment of patients with bacteraemia and thus serious consequences such as organ dysfunction and death.

The results show that the majority of the patients in the EMS were triaged as red or orange, which means immediate care. This result is somewhat surprising due to the lack of association between the RETTS colour and the risk of death. This finding can be interpreted in different ways. One interpretation is that there was no association between the RETTS colour and death. The second interpretation is that patients need- ing immediate care were given early and effective treatment, thus elim- inating the association between the RETTS colour and risk of death. Simultaneously, the results indicate that patients needing immediate care did not always receive the required treatment before arrival in hos- pital, for example fluid resuscitation. This is in line with other studies in- dicating that only a minority of all patients with sepsis are given fluid resuscitation by EMS personnel [2,15,17]. The reason why adequate treatment was not initiated can only be speculated upon. One possible reason is that the EMS personnel assessed the vital signs such as heart rate and systolic blood pressure as normal and as a result decided not

to initiate fluid resuscitation. Another explanation is that current treat- ment guidelines do not support treatments such as fluid resuscitation when the patient has normal vital signs [40].

The results indicate that it took about 4 h from the first contact with the EMS until the start of treatment with antibiotics. However, no signif- icant association was shown between survival and the interval from first contact with EMS until the start of treatment with antibiotics. The results could not confirm any association between outcome and various aspects of delay time until delivery of treatment. This finding stands in contrast to our findings that early detection of sepsis by the EMS is associated with a more favourable outcome. Therefore, the interrelation between the early detection of bacteraemia and the delay until the de- livery of treatment as well as the subsequent outcome needs further exploration.

Assessment and decision-making are critical components in the early chain of care in patients with bacteraemia. It is utterly essential to provide EMS personnel with education and training in how to iden- tify patients who have or are at risk of developing severe infections in order to assess and meet patients’ need for care and treatment. There- fore, this type of education and Training needs to be further reinforced. The results indicate that the long-term prognosis among patients with bacteraemia is poor. Thus, more than half of them were dead five years after arrival in hospital. This can most likely be explained by high age and severe comorbidity in many of the patients, among other things. Thus, five-year mortality was strongly related to age. The obser- vation that five-year mortality was much higher among the patients who used the EMS may be explained by the higher age and comorbidity of these patients. However, the long-term outcome did not seem to be

related to the presence of true pathogens.

Strengths and limitations

A major limitation is that a number of variables involved a substan- tial proportion of missing data due to the retrospective nature of the study. A similar problem has been seen in other research [27,41]. This limitation may influence the validity and generalizability since the esti- mates might be biased. Another limitation concerns the patient’s co- morbidity in relation to her/his survival. It is possible that the results would have looked different if factors that could predispose for bacteraemia or factors besides bacteraemia that could influence the chance of survival had been included in this study. A third limitation is the timing of blood cultures. These were taken during a season when the risk of infection was at its highest. More blood cultures were proba- bly taken because of the time of year and consequently there were more patients detected with bacteraemia. It is possible that the representa- tiveness of the population would have been different if the blood cul- tures had been taken evenly over the year. However, positive blood cultures might also be contaminated which would reduce their clinical significance [32]. Therefore, the number of patients who had true path- ogens in their blood culture was reported in this study for all except non-EMS patients. Finally, patients with clinical signs of severe infection do not always have positive blood cultures [42]. It is possible that the findings in this study would have been different if the inclusion criteria also comprised patients with clinical signs of severe infection. A major strength, on the other hand, is that this study included a representative population with bacteraemia from a well-defined region with both urban and rural environments.

Conclusion

Among patients with bacteraemia who were admitted to hospital via the EMS or the ED, the 28-day mortality was 13.5-15.2%. Among pa- tients with bacteraemia who used the EMS, early suspicion of sepsis or fever/infection was associated with improved survival. However, it could not be confirmed that the delay time from the call to the EMS until the start of antibiotics is associated with the early risk of death.

This study underlines the importance of proper assessment, both of the patient’s vital signs and symptoms and of the EMS personnel’s po- tential to initiate adequate treatment in the prehospital setting. There- fore, since EMS personnel in their work come close to patients with bacteraemia, the results should encourage the EMS to make further in- vestments in the education and training of their personnel. Additional studies are required for further verification of the results as well as to explore more extensively the different aspects associated with the role of the EMS in the early chain of care in patients with bacteraemia.

Conflict of interest

The authors declare no conflict of interest with regard to the author- ship and/or publication of this article.

Ethical statement

The research has been performed according to correct ethical prac- tice. Ethical approval was obtained from the Regional Ethics Committee in Gothenburg (Ref. 383-12).

Funding

This research did not receive any specific grant from funding agen- cies in the public, commercial, or not-for-profit sectors.

Availability of data and materials

For ethical reasons, the raw data cannot be shared.

Acknowledgements

The authors thank Petter Major, Ambulance Department at North Alvsborg’s Hospital, Trollhattan, Lars Jonsson, Division of Bacteriology South Alvsborgs Hospital, Boras and Andreas Wandahl Ambulance De- partment at Skaraborg, Sweden for contributions to the current study. The authors also thank Margaret Myers for providing language editing.

References

  1. Bone RC, Balk RA, Cerra FB, Dellinger RP, Fein AM, Knaus WA. Definitions for sepsis and organ failure and guidelines for the use of innovative therapies in sepsis. The ACCP/SCCM Consensus Conference Committee. Chest 1992;101:1644-55.
  2. Seymour WC, Rea TD, Kahn JM, Walkey AJ, Yealy DM, Angus DC. Severe sepsis in Pre-hospital emergency care. Analysis of incidence, care, and outcome. Am J Respir Crit Care Med 2012;186(12):1264-71.
  3. Singer M, Deutschman CS, Seymour CW, Shankar-Hari M, Annane D, Bauer M, et al. The third international consensus definitions for sepsis and septic shock (sepsis-3). JAMA 2016;315(8):801-10.
  4. Wang HE, Weaverb MD, Shapiroc NI, Yealy DM. Opportunities for emergency med- ical services care of sepsis. Resuscitation 2010;81(2):193-7.
  5. Studnek JR, Artho MR, Garner CL, Jones AE. The impact of emergency medical ser- vices on the ED care of severe sepsis. Am J Emerg Med 2012;30:51-6.
  6. Andersson Hagiwara M. Development and evaluation of computerised decision sup- port system for use in pre-hospital care. (Dissertation) Jonkoping: Jonkoping Uni- versity; 2014.
  7. Edman-Waller J, Ljungstrom L, Jacobsson G, Andersson R, Werner M. Systemic symptoms predict presence or development of severe sepsis and septic shock. Infect Dis 2016;48(3):209-14.
  8. Robson W, Nutbeam T, Daniels R. Sepsis: a need for Prehospital intervention? Emerg Med J 2009;26:535-8.
  9. Gardlund B, Cronqvist J, Follin P, Furebring M, Gille-Johnson P, Sjolin J. Severe sepsis and septic shock requires immediate care (in Swedish). Lakartidningen 2011;108 (6):271-5.
  10. Guerra WF, Mayfield TR, Meyers MS, Clouatre AE, Riccio JC. Early detection and treatment of patients with severe sepsis by Prehospital personnel. J Emerg Med 2013;44(6):1116-25.
  11. Ferrer R, Martin-Loeches I, Phillips G, Osborn TM, Townsend S, Dellinger RP, et al. Empiric antibiotic treatment reduces mortality in severe sepsis and septic shock from the first hour: results from a guideline-based performance improvement pro- gram. Crit Care Med 2014;42:1749-55.
  12. Kumar A, Roberts D, Wood K, Light B, Parrillo J, Sharma S, et al. Duration of hypoten- sion before initiation of effective antimicrobial therapy is the critical determinant of survival in human septic shock. Crit Care Med 2006;34(6):1589-96.
  13. Seymour CW, Gesten F, Prescott HC, Friedrich ME, Iwashyna TJ, Phillips GS, et al. Time to treatment and mortality during mandated emergency care for sepsis. N Engl J Med 2017;376:2235-44.
  14. Herlitz J, Bang A, Wireklint Sundstrom B, Axelsson C, Bremer A, Hagiwara M, et al. Suspicion and treatment of severe sepsis. An overview of the prehospital chain of care. Scand J Trauma Resusc Emerg Med 2012;20:42.
  15. Groenewoudt M, Roest AA, Leijten F, Stassen PM. Septic patients arriving with emer- gency medical services: a seriously ill population. Eur J Emerg Med 2014;21(5):330.
  16. Axelsson C, Herlitz J, Karlsson A, Sjoberg H, Jimenez-Herrera M, Bang A, et al. The early chain of care in patients with bacteraemia with the emphasis on the prehospital setting. Prehosp Disaster Med 2016;31(3):272-7.
  17. Seymour CW, Band RB, Cooke CR, Mikkelsen ME, Hylton J, Rea TD, et al. Out-of hos- pital characteristics and care of patients with severe sepsis: a cohort study. J Crit Care 2010;25(4):553-62.
  18. Seymour CW, Carlbom D, Engelberg RA, Larsen J, Bulger EM, Copass MK, et al. Un- derstanding of sepsis among emergency medical services: a survey study. J Emerg Med 2012;42(6):666-77.
  19. Sjolin H, Lindstrom V, Hult H, Kurland L. What an ambulance nurse needs to know: a content analysis of curricula in the specialist nursing programme in prehospital emergency care. Int Emerg Nurs 2015;23:127-32.
  20. Suserud B-J. A new profession in the pre-hospital care field- the ambulance nurse. Nurs Crit Care 2005;10(6):269-71.
  21. The National Board of Health and Welfare. Regulations on ambulance care (SOSFS 2009:10). (In Swedish) Stockholm: The National Board of Health and Welfare; 2009.
  22. Polito CC, Isakov A, Yancey IIAH, Wilson DK, Anderson BA, Bloom I, et al. Prehospital recognition of severe sepsis: development and validation of a novel EMS screening tool. Am J Emerg Med 2015;33(9):1119-25.
  23. Widgren B. RETTS – emergency care directly. (in Swedish) Studentlitteratur: Lund; 2012.
  24. Widgren BR, Jourak M. Medical emergency triage and treatment system (METTS): a new protocol in primary triage and secondary priority decision in emergency med- icine. J Emerg Med 2011;40:623-8.
  25. The National Archives. The Swedish register of deaths. Stockholm: The National Archives; 2017.
  26. Mehl A, Asvold BO, Lydersen S, Paulsen J, Solligard E, Damas JK, et al. Burden of bloodstream infection in an area of mid-Norway 2002-2013: a prospective popula- tion-based observational study. BMC Infect Dis 2017;17:205.
  27. Roest AA, Stoffersa J, Pijpersb E, Jansenc J, Stassena PM. Ambulance patients with nondocumented sepsis have a high mortality risk: a retrospective study. Eur J Emerg Med 2017;24(1):36-43.
  28. Band RA, Gaieski DF, Hylton JH, Shofer FS, Goyal M, Meisel ZF. Arriving by emergency medical services improves time to treatment endpoints for patients with severe sep- sis or septic shock. Acad Emerg Med 2011;18:934-40.
  29. Naghavi M, Wang H, Lozano R, Davis A, Liang X, Zhou M, et al. Global, regional, and national age sex specific all-cause and cause-specific mortality for 240 causes of death, 1990 2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet 2015;385:117-71.
  30. Gotland N, Uhre ML, Mejer N, Skov R, Petersen A, Larsen AR, et al. Long-term mortal- ity and causes of death associated with Staphylococcus aureus bacteremia. A matched cohort study. J Inf Secur 2006;73:346-57.
  31. Nielsen SL, Lassen AT, G KO, Jensen TG, Kolmos HJ, Hallas J, et al. Bacteremia is asso- ciated with excess long-term mortality: a 12-year population-based cohort study. J Inf Secur 2015;70:111-26.
  32. Weinstein MP, Towns ML, Quartey SM, Mirrett S, Reimer LG, Parmigiani G, et al. The clinical significance of positive blood cultures in the 1990s: a prospective compre- hensive evaluation of the microbiology, epidemiology, and outcome of bacteremia and fungemia in adults. Clin Infect Dis 1997;24:584-602.
  33. Sunden-Cullberg J, Rylance R, Svefors J, Norrby-Teglund A, Bjork J, Inghammar M. Fever in the emergency department predicts survival of patients with severe sepsis and septic shock admitted to the ICU. Crit Care Med 2017;45:591-9.
  34. Bohm K, Kurland L, Bartholdson S, Castren M. Descriptions and presentations of sep- sis – a qualitative content analysis of emergency calls. Int Emerg Nurs 2015;23: 294-8.
  35. Wallgren UM, Bohm K, Kurland L. Presentations of adult septic patients in the prehospital setting as recorded by emergency medical services: a mixed methods analysis. Scand J Trauma Resusc Emerg Med 2017;25:23.
  36. Bayer O, Schwarzkopf D, Stumme C, Stacke A, Hartog CS, Hohenstein C, et al. An early warning scoring system to identify septic patients in the prehospital setting: the PRESEP score. Acad Emerg Med 2015;22(7):868-71.
  37. Lane D, Ichelson R, Drennan I, Scales D. prehospital management and identification of sepsis by emergency medical services: a systematic review. Emerg Med J 2016; 33(6):408-13.
  38. Chen N-C, Hsieh M-J, Tang S-C, Chiang W-C, Huang K-Y, Tsai L-K, et al. Factors asso- ciated with use of emergency medical services in patients with acute stroke. Am J Emerg Med 2013;31:788-91.
  39. Robson W, Newell J. Assessing, treating and managin patients with sepsis. Nurs Stand 2005;19(50):56-64.
  40. Swedish EMS managers’ organization. National treatment guidelines for EMS services. (in Swedish) Stockholm: Swedish EMS managers’ organization; 2011.
  41. Gilbert EH, Lowenstein SR, KozioI-McLain J, Barta DC, Steiner J. Chart reviews in emergency medicine research: where are the methods? Ann Emerg Med 1996;27 (3):305-8.
  42. Opal SM, Garber GE, LaRosa SP, Maki DG, Freebairn RC, Kinasewitz GT, et al. Systemic host responses in severe sepsis analyzed by causative microorganism and treatment effects of drotrecogin alfa (activated). Clin Infect Dis 2003;37:50-8.

Leave a Reply

Your email address will not be published. Required fields are marked *