A prospective randomized controlled trial to investigate the effect of early administration of antibiotics for patients with suspected sepsis

INTRODUCTION AND RATIONALE
Introduction
Sepsis is one of the most common and life-threatening diseases in the world,
causing more deaths than AIDS, breast cancer and prostate cancer put together.
Despite the fact that the mortality of sepsis is ten times higher than
myocardial infarction and to five times higher than stroke, relatively little
attention is given to sepsis. In recent years successful clinical care
management pathways have been developed for patients suffering from a
myocardial infarction, stroke or a trauma. Even though there is strong
evidence in scientific literature to support the need for a series of
time-dependent actions, for sepsis this is still not the case.
Prompt recognition and treatment are extremely important for improving
survival, while patients who survive sepsis can still continue to suffer from
physical or psychological symptoms. The likelihood and severity of these
complications depends on a number of factors including the severity of sepsis
and the length of stay in hospital stay and in ICU.

Definition
Sepsis is defined as a proven or strongly suspected infection that is
associated with a *systemic inflammatory response syndrome ' (SIRS). SIRS
exists if at least two of the four criteria are met: abnormal body temperature,
increased heart rate (over 90 beats per minute), increased respiratory rate
(more than 20 per minute) and an abnormal white blood cell (WBC) count. There
are different degrees of sepsis on the basis of severity. Sepsis may develop to
severe sepsis or septic shock, if treatment is not timely initiated. Severe
sepsis is defined as sepsis with failure of one or more organ systems and
septic shock and severe sepsis with persistent low blood pressures despite
adequate resuscitation. In particular organ failure and shock cause high
mortality.

In the Netherlands, more than 10,000 patients with sepsis are admitted to a
hospital annually, with an average stay of 15 days. The medical costs being
approximately $ 20,000 per person, the total cost of severe sepsis is estimated
at nearly 170 million per year. Mortality can be very high if sepsis is not
timely or adequately treated, especially among the vulnerable elderly
population. The mortality rates vary from 20 to 60 percent, depending on namely
age and other underlying diseases such as diabetes and cancer. Multi-organ
failure due to sepsis and septic shock is the leading cause of death in the
ICU. The incidence of sepsis has increased in recent years and it is expected
that this trend will continue, partly due to the aging population and partly
because of increasing numbers of immune-compromised patiënts who are highly
susceptible to all kinds of (opportunistic) infections.

Early Goal Directed Therapy (EGDT)
The advent of antibiotics was a major step forward in the treatment of sepsis,
causing a mortality decrease by approximately 25 percent. It is noteworthy that
in the decades hereafter very little progress in the treatment of sepsis was
made, until the introduction of the 'Early goal directed therapy (EGDT). A
study by Rivers and colleagues (4) shows that by applying EGDT during the first
six hours after detection of sepsis, an absolute mortality reduction of almost
16% is achieved. This EGDT consists of a number of interventions, which have
the purpose to optimize hemodynamics as quickly as possible by means of tight
monitoring of arterial / venous pressures and oxygen saturation. The
cornerstones of this treatment include aggressive fluid resuscitation,
administration of vasopressors, giving protective ventilation and
administration of broad-spectrum antibiotics. Several large clinical trials
have confirmed the value of EGDT with sometimes even greater mortality
reduction.

Survival Sepsis Campaign Timely recognition and rapid treatment of sepsis
appears crucial, but recognizing sepsis still remains a challenge: the symptoms
are often non-specific and various other diseases might fit as well. Therefore
the "Surviving Sepsis Campaign" (SSC) was launched in 2003, with the aim of
creating awareness for sepsis for better recognition and treatment of sepsis to
improve the prognosis. Through this campaign a directive was developed wherein
a somewhat modified form of the EGDT was incorporated. This directive also
states that broad-spectrum antibiotics should be administered as soon as
possible, preferably within one hour after arrival in the emergency room.
Despite extensive attention in the last few years ( major campaigns of VMS) in
shortening time to administration of antibiotics (the so called "onset to
needle time"), there are still delays in the start up of antibiotic therapy in
the emergency department (ED). EDs are still not functioning optimally, with
waiting times sometimes exceeding 6 hours. This is also the conclusion of the
report: *Haastige spoed niet overal goed* from 2004 (Inspectie der
Volksgezondheid). Herein EDs in the Netherlands are described as the weakest
link in the emergency care, and in addition according to this report, little
progress in the quality of care in the emergency department was made from 1994
to 2004. After much effort and recommendations, progress was made in the last
few years but not sufficiently enough (see report ** Ziekenhuizen goed op weg
met implementatie normen voor afdelingen spoedeisende hulp* ").Our study will
therefore be able to contribute to the improvement of both the emergency and
strengthening of the acute care chain. Not only will we save costly time in
the trip till reaching the hospital, but moreover we will also overcome
(potential) delays in the emergency department by starting therapy in the
ambulance. Delays which can amount from one to even six hours.

Why antibiotics should be administered early? The first hour of presentation in
the emergency room, also known as the 'Golden Hour' seems to be the most
critical one in the treatment of a septic patient. Rapid antibiotic
administration means better chance of survival as well as a reduction in the
chance of lasting physical problems. Moreover, rapid intervention can shorten
hospital stay and even prevent the need for ICU admission. In daily practice
however, implementation of the SSC directives is not always easy, and there
may be several reasons to delay the start of treatment . The so-called 'onset
to needle time' can be as high as several hours. A recent pilot study in the
VUMC showed that 25% of patients had to wait longer than three hours at the
emergency department before treatment was initiated with antibiotic therapy.
Not only in the pilot study of our university but also in a retrospective study
conducted by Kumar and colleagues only 32.5% of the patients received the first
gift within the first 3 hours. Any delay in the administration of antibiotics,
causes an increase in mortality rate with almost 8 percent per hour! The later
the treatment is initiated, the greater the chance of multi-organ failure.
Besides higher mortality-rates, multi-organ failure is directly correlated with
more complications, longer hospital stay and higher use of costly healthcare
facilities. So it is very important that the onset to needle time is as short
as possible.
Moreover in the long term sepsis can cause much damage. Patients who survive
sepsis often suffer for months of complications that arise during or after a
prolonged hospitalization in intensive care (eg critical illness neuropathy,
problems with speech or swallowing by prolonged ventilation). The quality of
life can also sharply deteriorate after experiencing sepsis.
An important point to note is that all the studies which state that early
antibiotic administration is associated with improved survival, were
retrospective and uncontrolled studies, making occurrence of selection bias
probable. In order to investigate the optimal timing of antibiotic
administration, prospective randomized controlled studies should be performed
at the emergency department. However, it would be completely ethically
unjustified to randomize patients and delay initiation of antibiotic therapy.
An alternative and perhaps a better option is to perform a prospective
randomized trial in the pre-hospital setting, i.e in the ambulances. In current
practice, initiation of antibiotic therapy starts at the emergency department
(ED) and not in the ambulances. Pre-hospital antibiotic administration may be a
solution to avoid delays in treatment at the ED and contribute to improved
clinical outcomes such as improved survival, shorter hospital stay and better
quality of life.

Prehospital care, even in sepsis?
Ambulance personnel have already made a significant contribution in improving
care for patients with acute coronary syndrome, stroke and multiple trauma.
Patients with severe sepsis or septic shock can also benefit from early
pre-hospital care. Pre-hospital care is the first medical care, which is given
by ambulance personnel at reaching the patient. Since time plays a crucial role
in the treatment of sepsis, early recognition and initiation of treatment by
the ambulance personnel may help to reduce mortality. The provision of
pre-hospital care is associated with a shorter start-up time of EGDT and
antibiotic therapy in the hospital. In addition, it appears that this
pre-hospital care leads to quicker achievement of an optimal blood pressure,
and oxygen saturation. Therefore, it can be expected that by the administration
of broad-spectrum antibiotics in the ambulance, the survival of sepsis can be
improved by greatly reducing the time to the administration of the necessary
antibiotics.
To date no randomized controlled trials on the effect of antibiotics in the
pre-hospital settings on adults have been conducted. In children with
meningitis some uncontrolled studies have been done. In the studies by Strang
and Cartwright. a clear beneficial effect on survival is seen after
pre-hospital administration of antibiotics by general practitioners. Hamden and
Sorensen on the contrary showed that administration of antibiotics in the
pre-hospital setting was associated with worse outcomes. A possible explanation
for these divergent results is that there occurred a strong selection bias.
The group of children receiving pre-hospital antibiotics could be in a more
critical stage of illness.
A definitive answer to the question whether administration of antibiotics in
pre-hospital setting is effective, can only be obtained by a prospective
randomized controlled trial. In this form of study selection bias can be
avoided.
However recognition of sepsis is difficult, Suffoletto et al investigated how
accurately the ambulance personnel in Pennsylvania were able to recognize a
serious infection, the negative predictive value was 93%. In contrast, 69% of
the patients with a severe infection were missed (8). A pilot study in
Maastricht showed, that this situation can probably be extrapolated to the
Netherlands as well: a large proportion of patients with septic shock was in
fact not recognized as such and transported with a B-ride (low priority). The
information transfer is hereby often incomplete or not entirely accurate
causing substantial delays in initiating treatment. Therefore, much can be
gained by training ambulance personnel and getting them skilled enough in
recognizing sepsis and improving the transfer of information.

Importance of this project
Sepsis is a major global health problem with an increasing incidence and high
mortality rate. While in the past decades, modern medicine has become
increasingly sophisticated, and the treatment and care around diseases, such as
heart failure and cancer is greatly improved, treatment of sepsis remains a
problem with a high mortality. Early administration of antibiotics in
critically ill patients with sepsis could possibly change that. It is not clear
what the best time of administration should be, because how early is early?
Moreover, in practice it often happens that the time limits are not, or not
sufficiently monitored closely, whilst increasing the risk of death.
Moreover, the treatment of an acutely ill patient requires teamwork and an
optimally functioning acute-care chain. Through this project it will be
possible to investigate whether pre-hospital antibiotic therapy leads to better
outcomes for this category of seriously ill and vulnerable patients.

Primary Objectives:
To evaluate whether early, pre-hospital administration of antibiotics reduces
28-day mortality in patients referred to the ED with suspected severe sepsis or
septic shock.


Secondary Objective(s):
1. To compare whether there is a difference in the length of hospital stay and
admission to intensive or medium care unit in the standard treatment group
versus the pre-hospital EGDT group.
2. To evaluate any adverse events in the group of patients who receive
pre-hospital antibiotics compared to those who receive regular treatment.
3. To evaluate whether early antibiotic administration has a beneficial effect
on the quality of life after discharge from hospital. This will be measured one
month after discharge using validated questionnaires (SF 36).

The design is a non-blinded randomized clinical trial for assessing effect of
pre-hospital administration of antibiotics nested within a stepped-wedge design
to .assess effect of training on initiation of treatment. The training of
ambulance is implemented sequentially in different regions (i.e. phased
implementation). The training is first implemented in the region of Amsterdam
because of logistics regions. The order in which training occurs in the other
regions is randomized. Every 3 months a new region is trained.
Patients included after training of ambulance personnel in their region is
completed will at an individual level be randomized to the intervention
(pre-hospital administration of antibiotics) or the control condition (current
standard procedure). Interim analyses are planned for the randomized clinical
trial as soon as 28-day mortality of the first 25, 50 and 75 percent of
patients included after training has been observed.

the prehospital treatment group will receive 2000 mg ceftriaxone intravenously.
After samples of bloodcultures are taken. The control group will receive no
intervention

All patients suspected of sepsis are treated with broad-spectrum antibiotics in
the hospital according to the specific antibiotic guidelines. As prompt
initiation of antibiotic therapy is important, we hypothesize that starting
treatment in the ambulance can save us costly time and enhance the chances for
survival. Thus the patients treated in our intervention group will be given
intravenously broad- spectrum antibiotics (ceftriaxone) in the ambulance after
blood-cultures are sampled. Further treatment of both randomised groups will
be according to the national ambulance protocol for septic shock.
Treatment with ceftriaxone may be associated with a slightly higher
complication rate, such as allergic reactions. Possible benefits for study
subjects are a reduction in mortality rates, shorter length of stay in the
hospital or intensive care unit and a better quality of life.