Anaesthesiology Intensive Therapy

2015, vol. 47, s44–s55

ISSN 0209–1712
10.5603/AIT.a2015.0075
REVIEWS www.ait.viamedica.pl

Initial resuscitation from severe sepsis:

one size does not fit all
Stefanie Vandervelden1, Manu L.N.G. Malbrain2

1ICU Department, Ziekenhuis Netwerk Antwerpen, ZNA Stuivenberg, Antwerpen, Belgium

2ICU Department and High Care Burn Unit, Ziekenhuis Netwerk Antwerpen, ZNA Stuivenberg, Antwerpen, Belgium

Abstract
Over recent decades many recommendations for the management of patients with sepsis and septic shock have been
published, mainly as the Surviving Sepsis Campaign (SSC) guidelines. In order to use these recommendations at the
bedside one must fully understand their limitations, especially with regard to preload assessment, fluid responsive-
ness and cardiac output. In this review we will discuss the evidence behind the bundles presented by the Surviving
Sepsis Campaign and will try to explain why some recommendations may need to be updated. Barometric preload
indicators, such as central venous pressure (CVP) or pulmonary artery occlusion pressure, can be persistently low or
erroneously increased, as is the case in situations of increased intrathoracic pressure, as seen with the application
of high positive end-expiratory pressure, or in situations with increased intra-abdominal pressure. Chasing a CVP of
8 to 12 mm Hg may lead to under-resuscitation in these situations. On the other hand, a low CVP does not always
correspond to fluid responsiveness and may lead to over-resuscitation and all the deleterious effects on end-organ
function associated with fluid overload. We will suggest the introduction of new variables and more dynamic meas-
urements. During the initial resuscitation phase, it is equally important to assess fluid responsiveness, either with
a passive leg raising manoeuvre or an end-expiratory occlusion test. The use of functional hemodynamics with stroke
volume variation or pulse pressure variation may further help to identify patients who will respond to fluid administra-
tion or not. Furthermore, ongoing fluid resuscitation beyond the first 24 hours guided by CVP may lead to futile fluid
loading. In patients that do not transgress spontaneously from the Ebb to Flow phase of shock, one should consider
(active) de-resuscitation guided by extravascular lung water index measurements.

Key words: sepsis guidelines, bundle care, resuscitation

Anaesthesiology Intensive Therapy 2015, vol. 47, s44–s55

During recent decades many important studies have Due to the complexity of hemodynamics in sepsis, the
been published which provide recommendations regarding goals of treatment are much more difficult to define with
resuscitation, including guidelines for the clinician caring for certainty than in other forms of shock. The limitations of care
a patient with severe sepsis or septic shock [1−13]. The use bundles include a lack of agreement on hemodynamic goals
of a combination of individual strategies to facilitate rapid for management of patients with sepsis, proposing that
adoption of (un)proven therapies, to benchmark perfor- this lack of consistency may contribute to heterogeneity in
mance, and to improve patient outcomes is called bundled treatment effects for clinical trials of novel sepsis therapies.
care. Although care bundles are simple, uniform and have Moreover, the relative contributions of each element of the
universal practical applicability, the bar needs to be raised bundle are not known [15, 16].
[14]. A 2001 study by Rivers et al. (referred to hereafter as The Surviving Sepsis Campaign (SSC) was launched as
‘the Rivers study’) was the first to show that the institution a collaboration of three professional organizations at the
of early goal-directed therapy (EGDT) upon admission to the European Society of Intensive Care Medicine’s annual con-
emergency room (ER) can significantly reduce the mortality gress in Barcelona in 2002 (www.survivingsepsis.org) and
of patients in severe sepsis and septic shock [1]. the SSC guidelines were first published in 2004 and revised

44
 Stefanie Vandervelden, Manu L.N.G. Malbrain, Care bundles in sepsis

twice afterwards in 2008 and in 2012 [17−19]. The algorithm above this range, increase the risk for developing pulmonary
used in these guidelines was adopted from the Rivers’ study. oedema.”The thresholds used by Hollenberg originate from
They provide a 6-hour bundle aimed at achieving the initial an article written in 1983, entitled: “Optimum left heart
resuscitation of sepsis-induced hypoperfusion. Therefore, filling pressure during fluid resuscitation of patients with
during the first 6 hours of resuscitation, the goals of initial re- hypovolemic and septic shock”. The total population of this
suscitation of sepsis-induced hypoperfusion should include study was only 20 patients while the analysis was limited
all of the following as a part of a treatment protocol [19]: to 15 patients. The effect of increasing filling pressures on
• CVP 8–12 mm Hg cardiac performance was examined in those 15 patients
• MAP ≥ 65 mm Hg undergoing fluid resuscitation for hypovolemic and septic
• Urine output ≥ 0.5 mL kg-1 h-1 shock. Moreover, in 2 patients the protocol was terminated
• Superior vena cava oxygenation saturation (ScvO2 ) or early because of the inability to increase the wedge pressure
mixed venous oxygen saturation (SvO2) 70% or 65%, by 10 mm Hg, despite administration of 5 and 8 litres of nor-
respectively. mal saline solution, respectively [23]. This is where the chain
Despite the fact that this initiative is a great step for- of evidence leading to the CVP threshold of 8−12 mm Hg
ward in the standardisation of the initial management of used in the SSC guidelines stops, as was nicely investigated,
patients with sepsis and septic shock, and the authors and eloquently discussed and concluded by Perel [22].
co-workers on this project have to be congratulated, some Using pressures to measure preload has been found
recommendations may have limitations when applied at to be inaccurate, particularly in patients ventilated with
the bedside. intermittent positive pressure ventilation (IPPV), (auto) posi-
tive end expiratory pressure (PEEP), post-cardiac surgery,
THE (LACK OF) EVIDENCE BEHIND THE SSC obesity and those with intra-abdominal hypertension (IAH)
TARGETS or abdominal compartment syndrome (ACS) [24−29]. It is
A recent multi-Society Statement clearly states that [20]: a step forward that the latest version of the SSC guidelines
“The results of clinical research, pathophysiologic reasoning, does mention the possible effects of increased intrathoracic
and clinical experience represent different kinds of medical pressure (ITP) and intra-abdominal pressure (IAP) on CVP.
knowledge crucial for effective clinical decision making […] However, Perel continued his analysis of the evidence be-
Each kind of medical knowledge has various strengths and hind the SSC and found that in one of the references to justify
weaknesses when utilized in the care of individual patients this statement, the SSC guidelines refer to a review that lacks
[…] No single source of medical knowledge is sufficient to evidence to support the statement that a higher CVP should
guide clinical decisions […] No kind of medical knowledge be aimed for in a patient on mechanical ventilation [30]. This
always takes precedence over the others.” The importance review clearly states that filling pressures have a low predictive
of this statement at this point and time for medicine in value in estimating fluid responsiveness during mechanical
general cannot be underestimated and some summarized it ventilation and that using them to guide fluid therapy can lead
as follows: “It reflects the swing of the pendulum away from to inappropriate therapeutic decisions. Thus, chasing the 8−12
rigidly adhering to evidence based medicine principles and mm Hg CVP target may institute aggressive fluid resuscitation
expresses the growing disappointment from randomized in a certain group of patients with low CVP values, which may
controlled clinical trials as a guide to clinical decisions.” lead to fluid overload, and may aggravate pulmonary oedema.
Furthermore “bundling” therapies may result in unintended The opposite will happen in patients with high CVP that are in
side effects, particularly if the patient population is not the fact responders. Here fluids are withheld with a possible risk
same as the one that was originally studied. For example, of hypoperfusion and, thus, under-resuscitation (Fig. 1) [29].
some sepsis treatments have been studied in sepsis, others A recent analysis of the SSC results, based on 15,022 voluntary
in severe sepsis, and others in septic shock, yet we bundle submitted data, showed that attainment of a CVP of 8 mm Hg
them all together in the SSC guidelines [21, 22]. and ScvO2 > 70% did not influence survival in patients with
septic shock [31]. Another trial even showed that patients
THE FIRST TARGET IS TO REACH A CVP OF 8−12 MM HG with a CVP less than 8 mm Hg who had received less fluid,
This target became part of the bundle, having come had a better survival than those who had a CVP of about
from the Rivers’ study. It is noteworthy that in this RCT, 12 mm Hg [32]. Moreover, Marik’s meta-analysis, incorporat-
both the standard group and the GDT group were treated ing recent studies that investigated indices predictive of fluid
with a CVP to from 8 to 12 mm Hg. This recommendation is responsiveness, showed that there are no data to support the
based on the previously stated practice parameters [15]: “In widespread practice of using CVP to guide fluid therapy [33].
most patients with septic shock, cardiac output will be opti- Indeed, Marik nicely elaborates on this issue with his tale of
mized at filling pressures between 12−15 mm Hg. Increases seven mares [34].

45
Anaesthesiol Intensive Ther 2015, vol. 47, s44–s55

Figure 1. Inability of central venous pressure (CVP) to discriminate between fluid responders and non-responders. Adapted from Osman et al. [29]

The leadership of the SSC has believed since its incep- • Chasing a static CVP target of 8 to 12 mm Hg as a resusci-
tion that both the SSC Guidelines and the SSC performance tation endpoint may lead to over- or under-resuscitation
improvement indicators will evolve as new evidence that and should be abandoned.
improves our understanding of how best to care for patients
with severe sepsis and septic shock becomes available [19]. THE SECOND TARGET IS TO MAINTAIN
The May 2015 update of the SSC (http://www.survivingsepsis. A MAP > 65 MM HG
org/bundles) explains that there are indeed limitations to This recommendation is based on the findings of small
ventricular filling pressure estimates as surrogate for fluid studies, which showed no significant differences in lactate
resuscitation and that we possibly should use a dynamic levels or regional blood flow when the MAP was elevated to
measure of fluid responsiveness, including cardiac output more than 65 mm Hg in patients with septic shock, indicat-
(CO) in combination with volumetric preload indices. How- ing that a target of 65 mm Hg should be sufficient in most
ever, the latter have yet to be included into the SSC guidelines. cases [37, 38]. In the past, some authors suggested that
Moreover, the River’s group recently published an up- a higher blood-pressure target might be better, for example
date on early sepsis management where they in fact admit to maintain kidney function [39, 40].
that aggressive fluid resuscitation in the late stages of the Recently, the SEPSISPAM investigators showed in a RCT
sepsis spectrum may increase morbidity [35]. As such, the that targeting a MAP of 80 to 85 mm Hg, as compared with
global clinical picture should be given greater weight than 65 to 70 mm Hg, in patients with septic shock undergo-
an isolated value. However, they remain convinced that ing resuscitation, did not result in significant differences in
CVP-guided fluid administration in the early stages of sepsis mortality [2]. Moreover, analogous with cerebral perfusion
might decrease mortality. They refer to Walkey’s study on pressure defined as CPP = MAP – IAP, one could also calculate
early central venous catheter (CVC) introduction in sepsis abdominal perfusion pressure APP = MAP – IAP, which seems
as evidence to support this statement [36]. Unfortunately, a better resuscitation endpoint in patients with abdominal
this study clearly states that an increased use of early CVC hypertension [41]. Therefore, the ideal MAP target should
placement is not correlated with an increased quality of care be based on pre-existing hypertension and co-morbidities.
in general, including concurrent implementation of other Recommendations:
elements of the Surviving Sepsis bundle (e.g., early antibiotic • Chasing a static MAP target of 65 mm Hg may be too
administration) besides CVP measurement. low or too high and, as such, MAP should be tailored
Recommendations: individually.
• Barometric preload indicators, such as central venous • In patients with abdominal hypertension, abdominal
pressure (CVP) or pulmonary artery occlusion pressure perfusion pressure (APP), calculated as MAP minus
(PAOP), should not be used to guide fluid resuscitation intra-abdominal pressure, may be a better resuscita-
in septic patients. tion endpoint.

46
 Stefanie Vandervelden, Manu L.N.G. Malbrain, Care bundles in sepsis

THE THIRD TARGET IS A URINE OUTPUT OF 0.5 ML KG-1H-1 tissue hypoxia (lactate of greater than or equal to 2 mmol
Urine output has classically been adopted as the primary L-1 and ScvO2 of less than 70%), and finally, a group with
endpoint to guide resuscitation in burn care. The prevail- resolved global tissue hypoxia (lactate of less than or equal
ing view has deemed it appropriate to target a diuresis to 4 mmol L-1 and ScvO2 of greater than or equal to 70%) [49].
of greater than 0.5 mL kg-1 h-1 in adults and 1 mL kg-1 h-1 However, one group of patients was still missing: patients
in the paediatric population. This endpoint, however, has with high lactate and high ScvO2 or, thus, those with severe
been brought into question by various studies that have global tissue hypoxia and low O2 extraction. In a multicenter
shown no correlation between urine output and invasively European study [39], it was found, however, that out of 44
derived physiologic variables. Moreover, urine output is un- septic patients, 10 (23%) had lactate levels of greater than
able to identify fluid responders after a fluid challenge and or equal to 2 mmol L-1 and ScvO2 values above 70% [50].
it is inaccurate as a resuscitation target. Decreased urinary Perel performed a highly impressive bench-to-bedside
output can easily mislead the clinician as, while it may be the analysis of the Rivers study showing that the extremely low
result of intravascular hypovolemia, it equally could also be ScvO2 values seen in Rivers’ patients on admission to the ER
caused by IAH and ACS [42]. In the latter situation, a vicious indicate that these patients must have had very low CO’s
cycle is established with further fluid loading. This will cause [22]. He stated that the most likely cause for these low CO’s
even more intestinal oedema and visceral swelling, leading was probably a combination of pre-existing co-morbidities
to increasing IAP, venous hypertension and deteriorating and profound hypovolemia, which may have developed
renal function. due to a late arrival to the hospital (ethnic group, low so-
Recommendation: cioeconomic status, no insurance) [22]. The very significant
• Urine output is a poor endpoint that may lead to over- hypovolemic element of their shock was successfully cor-
or under estimation of fluid resuscitation: needs and, as rected by aggressive fluid loading which was guided by
such, can no longer be recommended. a simple protocol that may be unsuitable for many ICU
• However, in situations with limited monitoring tech- septic patients [22]. Interestingly, Perel also compared the
niques, it may still be used to guide fluid resuscitation co-morbidity of the Rivers’ patients to those included in the
CORTICUS (Corticosteroid Therapy of Septic Shock) study
THE FOURTH TARGET TO REACH IS AN SCVO2 OF 70% and concluded that Rivers’ patients had significantly more
Rivers showed that during fluid resuscitation, ScvO2 in- severe co-morbidities [22, 51]. Whether or not ScvO2 should
creases, suggesting a concomitant increase in CO. As such, be used is not a problem of evidence-based medicine but
ScvO2 can be used as a surrogate for CO [1, 43]. In the Rivers rather a problem of generalizability and extrapolation of the
study the baseline ScvO2 value is around 50%. These ob- Rivers' study results to other patient populations.. Therefore,
served ScvO2 values are extremely low compared to the nor- it was very disappointing that even in the third revision of
mal ScvO2 value of about 75%. Moreover, in septic patients, the SSCG guidelines the Rivers protocol was still perceived
the ScvO2 is usually normal or even supranormal due to as high-grade evidence [19]. The Rivers single-center study
a reduced oxygen extraction ratio, which is characteristic of dates back to 2001 and has never been repeated, until re-
septic shock [44, 45]. This can easily be calculated using the cently. The results have been recommended for all hypo-
Fick Formula (the oxygen extraction ratio is approximately tensive and/or hyperlactatemic septic patients, both in and
equal to (1 – ScvO2)). Only Rivers found such a low ScvO2 of outside the ER regardless of the fact that this study was, so
50%. Recent studies have indeed found much higher ScvO2 far up to then, the only evidence for the effectiveness of the
values in septic shock patients, either in the emergency hemodynamic protocol suggested in the SSC guidelines.
department or on admission to the ICU [13, 46]. In two of Recommendation:
these studies, the mean ScvO2 was 72% to 74% [47, 48]. • As chasing an ScvO2 target of 70% in isolation does not
A normal/high ScvO2 may be due to reduced O2 extraction and make sense, ScvO2 should always be seen in relation to
does not necessarily indicate adequate tissue oxygenation. previous history, co-morbidities and actual lactate levels.
Further evidence of the fact that the ScvO2 values of
Rivers’ patients are not characteristic for all septic patients DO NOT IGNORE THE NEW EVIDENCE
can be found in a later study of Rivers himself and his col- First came the ProCESS trial, which concluded that pro-
leagues, in which patients of both the usual treatment and tocol-based resuscitation of patients in whom septic shock
the EGDT groups of their original study were combined was diagnosed in the ER did not improve outcomes [3]. The
and then divided into three resuscitation groups. These patients (in total 1,341) were randomly assigned to protocol-
included firstly, a group with severe global tissue hypoxia based EGDT, a protocol-based standard treatment, or to
(lactate of greater than or equal to 4 mmol L-1 and ScvO2 usual care. There were no significant differences in 90-day
of less than 70%), secondly, a group with moderate global mortality, 1-year mortality, or the need for organ support.

47
Anaesthesiol Intensive Ther 2015, vol. 47, s44–s55

Table 1. Overview of studies on GDT

Author Year Ref n Setting Mortality EGDT Mortality Control
Rivers 2001 [1] 263 ER 38/130 (29.2%) 59/133 (44.4%)
Wang 2006 [8] 33 NA 4/16 (25%) 7/17 (41.2%)
De Oliveira* 2008 [7] 102 mixed 6/51 (11.8%) 20/51 (39.2%)
EGDT 2010 [9] 314 NA 41/163 (25.2%) 64/151 (42.4%)
LACTATE 2010 [13] 348 ICU 58/171 (33.9%) 77/177 (43.5%)
Jones 2010 [6] 300 ER 34/150 (22.7%) 25/150 (16.7%)
Tian 2012 [10] 71 NA 12/19 (63.2%) 12/34 (35.3%)
Yu 2013 [11] 50 NA 6/23 (26.1%) 5/25 (20%)
Lu 2014 [12] 82 NA 7/40 (17.5%) 7/42 (16.7%)
PROCESS 2014 [3] 1341 ER 92/439 (21%) 167/902 (18.5%)
SEPSISPAM 2014 [2] 776 ICU 142/388 (36.6%) 132/388 (34%)
ARISE 2014 [4] 1600 ER 147/792 (18.6%) 150/796 (18.8%)
PROMISE 2015 [5] 1260 ER 184/623 (29.5%) 181/620 (29.2%)
Total 6491 771/3005 (25.7%) 906/3486 (26%)
EGDT — early goal directed therapy; ER — emergency room; ICU — intensive care unit; NA — not available, n: number of patients included; *paediatric patients from ER,
ward and ICU

This was soon followed by a second RCT, the ARISE study, The mortality in the GDT group was 771/3005 (25.7%) com-
showing that EGDT did not reduce all-cause mortality at pared to 906/3486 (26%) in the control group. EGDT did
90 days in critically ill patients presenting to the emergency not confer a reduction in overall mortality (pooled OR 0.94
department with early septic shock [4]. In total, 1600 pa- [95 % CI 0.84–1.05]; P = NS) (Fig. 2). There was evidence of
tients were randomly assigned in a 1:1 ratio to receive either heterogeneity (I2 = 62%; P = 0.27).
EGDT bundle care or usual care for 6 hours.
Finally, in the latest RCT, the ProMISE trial, patients were ARE WE COMPLIANT WITH THE BUNDLES WE PRETEND
randomly assigned to receive either EGDT (a 6-hour re- TO USE?
suscitation protocol) or usual care [5]. The primary clinical The SSC guidelines attempt to include nearly every as-
outcome was all-cause mortality at 90 days. The investiga- pect of critical care potentially related to sepsis, perhaps
tors enrolled 1,260 patients, with 630 assigned to EGDT losing focus in the process. As discussed previously, the
and 630 to usual care. In patients with septic shock that evidence behind some of the elements of the bundles is
were identified early and received intravenous antibiotics not strong (e.g. CVP) and the bundles are turned into quality
and adequate fluid resuscitation, hemodynamic manage- measures on which providers will be benchmarked, even
ment according to a strict EGDT protocol did not lead to an though clinicians may correctly disagree with some of the
improvement in outcome. recommendations [53]. It seems that other factors may also
The SSC explains that a large number of observational play a role: pharmaceutical, financial, political, legal etc.
studies have shown significant mortality reduction com- A Chinese study showed that only 47% of surveyed intensiv-
pared to historical controls. Although this may be the case, ists believed that CVP should be used to guide resuscitation,
closer analysis reveals that the beneficial effects may solely while 86% used it because of the SSC guidelines [54, 55].
depend on the proper use of antibiotics. The early admin- Despite the hype and pressure, full compliance with all
istration of antibiotics, and the right antibiotic, may be the applicable elements of the sepsis resuscitation bundle was
secret to success of the Surviving Sepsis Campaign [52]. Al- only 21.6% in the USA and 18.4% in Europe [56]. Finally the
though the use of bundles in order to ensure timely delivery SSC leadership concluded: “The strong recommendation
of treatments with recognized benefits may be important in for achieving a CVP of 8 to 12 mm Hg and an ScvO2 of 70%
the ER and ICU, on the other hand, the institution of current in the first 6 hours of resuscitation of sepsis-induced tissue
sepsis bundles may force physicians to provide unproven hypoperfusion, although deemed desirable, are not yet the
or even harmful care, particularly if the patient population standard of care as verified by practice data. The publication
is not the same as the one originally studied [21]. of the initial results of the international SSC performance
Table 1 summarizes the different RCT's on GDT in the improvement program demonstrated that adherence to
critically ill. In total, around 6,491 patients have been studied CVP and ScvO2 targets for initial resuscitation was low” [57].
in 13 trials. All patients were available for mortality analysis As discussed above, these reservations on the Rivers proto-
while overall mortality was 25.8 % (1677 of 6491 patients). col have already been raised by others and are based on its

48
 Stefanie Vandervelden, Manu L.N.G. Malbrain, Care bundles in sepsis

Figure 2. Forrest plot. Effect of EGDT on mortality in patients presenting to the emergency room or ICU with septic shock. Primary mortality
outcome is given for each study. The control was usual care or another non- EGDT resuscitation strategy. Fixed-effect model: the individual points
denote the OR of each study and the lines either side, the 95 % confidence intervals. OR: odds ratio, CI: confidence interval.

perceived physiological flaws (having the same targets of chart annotation consistent with all elements of severe
CVP in both arms) and on the possibility that the patients sepsis or septic shock ascertained through chart review):
of the Rivers study do not represent all septic patients [22]. 1. Measure lactate level
After the release of the 3rd SSC guidelines in 2013, the 2. Obtain blood cultures prior to administration of
authors anticipated the growing international criticism [19]. antibiotics
In response to comments and questions, the SSC leader- 3. Administer broad spectrum antibiotics
ship has provided additional background regarding the 4. Administer 30 mL kg-1 crystalloid for hypotension
guideline recommendation regarding measurement of CVP, or lactate ≥ 4 mmol L-1
ScvO2 and lactate. The performance indicators for bundle • To be completed within 6 hours of time of presentation:
compliance now call for measuring CVP and ScvO2, and re- 5. Apply vasopressors (for hypotension that does not
measuring lactate if the initial lactate was elevated. The respond to initial fluid resuscitation) to maintain
rationale for the indicators’ being measurement, and not a mean arterial pressure (MAP) ≥ 65 mm Hg
target achievement, is that the decision to give more fluid 6. In the event of persistent hypotension after initial
or add inotropes to the resuscitation should be based on fluid administration (MAP < 65 mm Hg) or if initial
the entire clinical picture (www.survivingsepsis.org/SiteCol- lactate was ≥ 4 mmol L-1, volume status and tissue
lectionDocuments/Guidelines-Statement-Leadership-CVP- perfusion needs to be re-assessed and the findings
ScvO2-Lactate-Measurements.pdf ). documented with:
Only recently, after 10 years of lively discussions and a. either:
debate, CVP and ScvO2 were removed from the 6-hour bun- i. repeat focused exam (after initial fluid resus-
dle in April 2015 (http://www.survivingsepsis.org/News/ citation) including vital signs, cardiopulmo-
Pages/SSC-Six-Hour-Bundle-Revised.aspx). However, the nary, capillary refill, pulse, and skin findings.
SSC leadership still keeps recommending measuring these b. or two of the following:
parameters. With the publication of 3 trials that do not dem- ii. measure CVP
onstrate the superiority of required use of a central venous iii. measure ScvO2
catheter (CVC) to monitor central venous pressure (CVP) iv. bedside cardiovascular ultrasound
and central venous oxygen saturation (ScvO2) in all patients v. dynamic assessment of fluid responsiveness
with septic shock who have received timely antibiotics and with passive leg raising or fluid challenge
fluid resuscitation compared with controls or in all patients 7. Re-measure lactate if initial lactate elevated
with lactate > 4 mmol L-1, the SSC Executive Committee has
revised the improvement bundles [3−5]. Therefore, finally, ALTERNATIVE SOLUTIONS TO SSC TARGETS
the 6-hour bundle has been updated as follows: IMPROVING BAROMETRIC PRELOAD INDICATORS
• To be completed within 3 hours of time of presentation As stated above, the CVP and PAOP may be errone-
(defined as the time of triage in the emergency department ously increased in patients with increased ITP [58]. The
or, if presenting from another care venue, from the earliest latest revision of the SSC guidelines still advocates initial

49
Anaesthesiol Intensive Ther 2015, vol. 47, s44–s55

fluid management based on CVP measurements with the CVPtm = CVPee – IT x PEEP
usual targets of 8 to 12 mm Hg. However, using pressures
to measure preload has been found to be inaccurate time Moreover, we have previously suggested a correction formu-
and time again, particularly, as discussed above, in patients la based on ITP or IAP (with an average index of transmission
ventilated with intermittent positive pressure ventilation between abdomen and thorax of 50% [26,58]):
(IPPV), (auto) PEEP, post-cardiac surgery, obesity and those
with IAH and ACS [24, 27−29, 58]. Although it is re-assuring CVPtm = CVPee – ITP = CVPee – IAP/2
and noteworthy that the latest version of the SSC guide-
lines does mention the possible effects of increased ITP Albeit far from perfect, these correction formulas for
and IAP on CVP, it advises that: “In mechanically ventilated PEEP and IAP may better reflect the true preload status
patients or those with known pre-existing decreased ven- and thus may improve the value of the barometric preload
tricular compliance, a higher target CVP of 12 to 15 mm Hg indices. Teboul et al. demonstrated the biggest risk in these
should be achieved to account for the impediment in filling. recommendations again, in a study showing that CVP is not
Similar consideration may be warranted in circumstances a reliable predictor of volume responsiveness. They found no
of increased abdominal pressure. Elevated CVP may also difference in CVP values of septic patients who are respond-
be seen with pre-existing clinically significant pulmonary ers or non-responders (responder = cardiac index increase
artery hypertension, making use of this variable untenable after fluid challenges) [29].
for judging intravascular volume status”. Moreover, they Recommendation:
properly refer to previous publications on this topic [59−61]. • Transmural filling pressures, or their estimates, may bet-
Within this respect, the compliance of the thorax and the ter reflect the true preload status (especially in patients
abdomen are key elements in order to explain the index of with high PEEP and IAP) and thus could be a better
transmission of a given pressure from one compartment to resuscitation endpoint.
another: “The use of lung-protective strategies for patients
with ARDS […] has been widely accepted, but the precise VOLUMETRIC PRELOAD INDICATORS
choice of tidal volume […] may require adjustment for such Volumetric estimates of preload status, such as global
factors as the plateau pressure achieved, the level of posi- end-diastolic volume index (GEDVI) and right ventricular
tive end-expiratory pressure chosen, the compliance of the end-diastolic volume index (RVEDVI), are of significant value
thoraco-abdominal compartment […]”[19]. This recently in the assessment of traumatically injured patients. This
led to the recognition of the polycompartment syndrome volumetric assessment is especially useful in patients with
[28, 62]. Instituting aggressive fluid resuscitation in patients increased IAP or patients with changing ventricular com-
with low CVP values may lead to fluid overload, which may pliance and elevated ITP in whom traditional barometric
aggravate pulmonary oedema, especially in those patients preload indicators are elevated and difficult to interpret,
in whom sepsis is associated with acute respiratory distress since they are zero-referenced against atmospheric pres-
syndrome (ARDS) and severe pulmonary dysfunction [22]. sure [26, 58, 66−68].
We therefore disagree with the SSC guidelines’ statement Reliance on such pressures to guide resuscitation can
that “a low CVP is still a good indicator of someone needing lead to inappropriate therapeutic decisions, under- or over-
fluid resuscitation”. Many patients with a low CVP are in fact -resuscitation, and organ failure [28]. Correction of the GEDVI
non-responders [29, 63, 64]. for the corresponding global ejection fraction can further
The SSC leadership could have referred to the excellent improve its predictive value [69]. One must, however, take
paper by Teboul et al. with a calculation of the index of trans- into account that no good normal values exist for GEDVI in
mission that is dependent on dynamic lung compliance [65]. different patient populations [70]. The same static volumet-
ric targets, although better than those which are barometric,
IT = (CVPei – CVPee)/(Pplat – PEEP) may not apply to all patients [71]. A recent meta-analysis
showed that baseline values for GEDVI are around 694 mL m-2
The higher the compliance (e.g. emphysema), the higher the in surgical and 788 mL m-2 in septic patients [70] and thus
IT and vice versa, the lower the compliance (lung fibrosis, below the upper limit of normal of 850 mL m-2, as was
ARDS), the lower IT. This is easy to understand, as the deno- recently used as target for initiating a fluid challenge [72].
minator is the same in the formula to calculate Cdyn and IT. We must remember that no single parameter can improve
outcome. This can only be achieved by a good protocol [73].
Cdyn = TV/(Pplat – PEEP) Recommendation:
• Volumetric preload indicators (like right ventricular or
Hence, the transmural CVP can be estimated as follows: global end diastolic volume) are superior compared to

50
 Stefanie Vandervelden, Manu L.N.G. Malbrain, Care bundles in sepsis

Table 2. Overview of recommendations regarding the initial resuscitation and resuscitation endpoint in patients with sepsis and septic shock
Resuscitation endpoints
1. Monitoring Every patient with septic shock should be adequately monitored with regard to cardiac output, fluid status, fluid
responsiveness and organ perfusion.
2. Cardiac output When treating shock patients, by definition, CO should be monitored to identify patients with low or high CO and to
assess the response to treatment.
3. Barometric preload Barometric preload indicators, such as central venous pressure (CVP) or pulmonary artery occlusion pressure (PAOP),
should not be used to guide fluid resuscitation in patients with septic shock.
Chasing a static CVP target of 8 to 12 mm Hg as resuscitation endpoint may lead to over- or under resuscitation and
should be abandoned.
Transmural filling pressures, or their estimates, may better reflect the true preload status (especially in patients with
high PEEP and IAP) and thus could be a better resuscitation endpoint.
4. Perfusion pressure Chasing a static mean arterial pressure (MAP) target of 65 mm Hg may be too low or too high and, as such, MAP should
be tailored individually.
In patients with abdominal hypertension, abdominal perfusion pressure (APP), calculated as MAP minus intra-
abdominal pressure (IAP), may be a better resuscitation endpoint.
5. Urine output Urine output is a poor endpoint that may lead to over- or under estimation of fluid resuscitation and, as such, can no
longer be recommended.
However, in situations with limited monitoring techniques, urine output can still be used to guide fluid resuscitation.
6. Mixed venous As chasing an ScvO2 target of 70% in isolation does not make sense, ScvO2 should always be seen in relation to previous
saturation history, co-morbidities and actual lactate levels.
7. Volumetric preload Volumetric preload indicators (such as right ventricular or global end diastolic volume) are superior compared to those
which are barometric and are recommended to guide fluid resuscitation, especially in septic patients with increased
intrathoracic pressure or IAP.
If the GEDVI is high, the measurement needs to be corrected for the global ejection fraction, as this leads to a more
accurate estimation of preload.
8. Fluid responsiveness Fluid resuscitation in septic patients should be guided by physiological parameters (SVV or PPV) or tests that are able
to predict fluid responsiveness (passive leg raising or endexpiratory occlusion test).
9. Fluid balance An excessive positive daily and cumulative fluid balance should be avoided.
10. Lung water The use of the extravascular lung water index (EVLWI) is recommended to guide de-resuscitation in septic patients not
transgressing spontaneously from the Ebb to Flow phase
11. Perfusion Fluid resuscitation should only be given/increased in case of evidence of tissue hypoperfusion (base deficit, lactate etc.).
APP — abdominal perfusion pressure; CO — cardiac output; CVP — central venous pressure; EVLWI — extravascular lung water index; GEDVI — global end-diastolic
volume index; IAP — intra-abdominal pressure; MAP — mean arterial pressure; PAOP — pulmonary artery occlusion pressure; PEEP — positive end expiratory pressure;
PPV — pulse pressure variation; SVV — stroke volume variation

those which are barometric and are recommended to spontaneous breathing, while tidal volumes must be above
guide fluid resuscitation, especially in septic patients 6 mL kg-1 [77, 78]. The presence of right heart failure and
with increased IAP. conditions related to increased ITP or IAP will increase the
• If the GEDVI is high, the measurement needs to be cor- baseline values of the functional hemodynamic parameters
rected for the global ejection fraction as this leads to a making them less reliable, unless we define new thresholds
more accurate estimation of preload. [79, 80]. In such situations (or thus in patients with dimin-
ished respiratory compliance) other techniques are available
FLUID RESPONSIVENESS in order to assess fluid responsiveness, such as the use of
A significant relationship between values of CVP has not a passive leg raise (PLR) or end-expiratory occlusion (EEO)
been found to identify responders from non-responders. test [81−84]. However, the PLR may result in a false negative
Different techniques are available to assess fluid respon- response in conditions of increased IAP due to diminished
siveness [74, 75]. However, there are certain limitations to venous return [85, 86]. The administration of repeated fluid
the use of functional hemodynamic monitoring, such as boluses until the patient is no longer fluid responsive cannot
stroke volume variation (SVV) or pulse pressure variation be advocated [72, 73, 87].
(PPV). The patient needs to be in regular sinus rhythm, while Recommendation:
the presence of atrial fibrillation, along with ventricular or • Fluid resuscitation in septic patients should be guided
supraventricular extra systoles, limit their use [76]. The pa- by physiological parameters or tests that are able to
tient also needs to be fully mechanically ventilated without predict fluid responsiveness.

51
Anaesthesiol Intensive Ther 2015, vol. 47, s44–s55

CARDIAC OUTPUT MONITORING lar fluid resulting in negative fluid balances. Recent studies
Cardiac output (CO) is the main determinant of oxygen have shown that conservative late fluid management (CLFM)
delivery and shock is defined by an imbalance between oxy- with 2 consecutive days of negative fluid balance within the
gen delivery and oxygen consumption. Physical examina- first week of stay is a strong and independent predictor of
tion and vital signs alone often fail to reflect significant alter- survival [97]. In this context, the global increased perme-
ations in CO [50]. Because of the complexity of assessment ability syndrome (GIPS) has been introduced, characterized
of clinical variables in septic patients, direct measurement by high capillary leak index (CLI, expressed as CRP over
of CO by invasive hemodynamic monitoring is advisable as albumin ration), excess interstitial fluid and persistent high
it is, therefore, very useful for proper decision-making in the extravascular lung water (EVLWI), no CLFM achievement and
critically ill [88]. Furthermore, perioperative optimisation has progressive organ failure [98]. GIPS represents a ‘third hit’
resulted in better or altered outcomes [89−92]. following acute injury with progression to MODS [99]. The
The main two reasons to measure CO are firstly, the dual response to acute inflammatory insult is characterized
identification of patients who have low (or high) CO values by a crucial turning point on day 2 to 3. Lower EVLWI and
that are not evident clinically (in order to stratify patients pulmonary vascular permeability indices (PVPI) [100] at day
between those having cardiac vs. septic shock) and sec- 3 of shock were shown to correlate with better survival. As
ondly, to assess the response to diagnostic (eg. passive leg adverse effects of fluid overload in states of capillary leak-
raising test) and therapeutic (eg. fluid bolus) intervention. age are particularly pronounced in the lungs, monitoring of
Based on the available evidence, we cannot agree with the EVLWI may offer a valuable tool to guide fluid management
SSC guidelines statement that: “The efficacy of these (CO) in the critically ill. It must be stated that EVLWI can never be
monitoring techniques to influence clinical outcomes from a trigger to start fluids but it is rather a safety parameter in
early sepsis resuscitation remains incomplete and requires order to define the extent of capillary leak and to guide de-
further study before endorsement” [19]. As repeatedly stated resuscitation [73, 101]. In this hypothesis (change in) EVLWI
previously by Perel and others, physiological examination, has a prognostic value as a reflexion of the extent of capillary
i.e. observing multiple parameters on the monitor in real leakage, rather than as a quantification of lung function im-
time should be considered to be (at least) as important as pairment by lung water [98]. The proposed Berlin definition
the classic physical examination [76, 92−94]. for ARDS, therefore, has no real added value compared to
Recommendation: the previous AECC definition [102]. Thus, the value of EVLWI
• By definition, when treating shock patients, CO should in combination with PVPI should “by definition” by part of
be monitored to identify patients with low or high CO a future ARDS definition [101, 103].
and to assess the response to treatment. Recommendation:
• An excessive positive cumulative fluid balance should
FLUID BALANCE AND DE-RESUSCITATION be avoided.
As early as 1942, the concept of a dual metabolic re- • The use of extravascular lung water is recommended to
sponse to bodily injury was introduced. In direct response guide de-resuscitation in septic patients not transgress-
to initial pro-inflammatory cytokines and stress hormones, ing spontaneously from the Ebb to Flow phase.
the Ebb phase represents a distributive shock characterised
by arterial vasodilatation and transcapillary albumin leakage CONCLUSION
abating plasma oncotic pressure [95]. Arterial underfill- One could come to the erroneous conclusion that pro-
ing, microcirculatory dysfunction and secondary intersti- tocols may not have a role in the treatment of septic shock
tial oedema lead to systemic hypoperfusion and regional as suggested by some and as was the conclusion in a recent
impaired tissue use of oxygen. In this early stage of shock, meta-analysis showing that EGDT is not superior to usual
adequate fluid therapy comprises goal directed filling to care for emergency department patients with septic shock
prevent development into multiple organ dysfunction syn- but is associated with increased utilisation of ICU resources
drome (MODS). Patients with a higher severity of illness need [104, 105]. However, Rivers et al. have started one of the most
more fluids to reach cardiovascular optimization. Therefore, important change processes in modern critical care and the
at this point fluid balance may be considered a biomarker SSC has probably saved many lives. The methodology that
of critical illness, as proposed by Bagshaw et al. [96]. Clas- was part of Rivers, ProCESS, ARISE and ProMISE studies can
sically, patients overcoming shock attain homeostasis of be applied in clinical practice to ensure early diagnosis and
pro-inflammatory and anti-inflammatory mediators within treatment for all patients with septic shock.
three days. Subsequent hemodynamic stabilization and As we know that time is of the essence, the elements we
restoration of plasma oncotic pressure set off the Flow phase, should focus on in order to save lives, are the early recogni-
with resumption of diuresis and mobilization of extravascu- tion of sepsis, early source control, early administration of

52
 Stefanie Vandervelden, Manu L.N.G. Malbrain, Care bundles in sepsis

antibiotics, early adequate volume resuscitation, and clinical ACKNOWLEDGEMENTS

assessment of the adequacy of circulation. However, the 1. The authors declare no financial disclosure.
thresholds and targets suggested by the SSC to guide initial 2. Manu L.N.G. Malbrain is founding president and current
resuscitation cannot be extrapolated to all septic patients Treasurer of the World Society of Abdominal Compart-
and may be potentially harmful in selected patients. Take ment Syndrome (WSACS, www.wsacs.org) and member
home messages for the reader can therefore be summarized of the medical advisory board of Pulsion Medical Sys-
as follows: With regard to EGDT it is not advisable to guide tems (Maquet Getinge group). Stefanie Vandervelden
the initial fluid resuscitation based on CVP measurements declares no potential conflict of interest with regard to
since they expose the patient to possible over-resuscitation the content of this paper. Parts of this review were based
along with all the deleterious effects of fluid overload, and, on the proceedings of the International Fluid Academy
in some situations with increased ITP, also to under-resus- Days (held between 2011–2014)(www.fluid-academy.
citation. No single parameter has ever improved survival, org). The authors are especially grateful and indebted
as only a good protocol or algorithm can. However, each to Prof Paul Marik and Prof Azriel Perel, whose lectures
patient is unique and, as such, also merits individualized are always very inspiring.
personalized care. As the best fluid is the one that has not
been given to the patient (that is, one which is unnecessary), References:
1. Rivers E, Nguyen B, Havstad S et al.: Early goal-directed therapy in the
it is advisable not to perform fluid bolus administrations treatment of severe sepsis and septic shock. N Engl J Med. 2001; 345:
until the patient is no longer fluid responsive but rather 1368−1377.
2. Asfar P, Meziani F, Hamel JF et al.: High versus low blood-pressure target
to assess responsiveness with passive leg raising or end- in patients with septic shock. N Engl J Med 2014; 370: 1583−1593 doi:
expiratory occlusion tests instead. Despite its limitations, 10.1056/NEJMoa1312173.
3. The ProCESS Investigators; Yealy DM, Kellum JA, Huang DT: A randomized
dynamic functional hemodynamic monitoring can provide trial of protocol-based care for early septic shock. N Engl J Med 2014;
further insights towards the identification of fluid respond- 370: 1683−1693. doi: 10.1056/NEJMoa1401602..
4. Arise Investigators, Anzics Clinical Trials Group; Peake SL, Delaney A, Bailey
ers. In many situations, volumetric preload indicators have M et al.: Goal-directed resuscitation for patients with early septic shock.
been proven superior over those which are barometric. Tak- N Engl J Med 2014; 371: 1496−506. doi: 10.1056/NEJMoa1404380.
5. Mouncey PR, Osborn TM, Power GS et al.: Trial of early, goal-directed
ing this into account the global ejection fraction can further resuscitation for septic shock. New Engl J Med 2015; 372: 1301−1311.
improve them. By definition, CO should be measured in all doi: 10.1056/NEJMoa1500896.
6. Jones AE, Shapiro NI, Trzeciak S, Arnold RC, Claremont HA, Kline JA: Lactate
patients with septic shock (especially when ScvO2 is low), clearance vs central venous oxygen saturation as goals of early sepsis
as this is the gold standard in order to assess the effect of therapy: a randomized clinical trial. JAMA 2010; 303: 739−746. doi:
10.1001/jama.2010.158.
fluids (e.g. a 15% increase in baseline CO). Ongoing fluid 7. de Oliveira CF, de Oliveira DS, Gottschald AF et al.: ACCM/PALS haemodyna-
resuscitation beyond the initial 24-48 hours cannot be rec- mic support guidelines for paediatric septic shock: an outcomes compa-
rison with and without monitoring central venous oxygen saturation. In-
ommended, unless a safety parameter, such as extravascular tensive Care Med 2008; 34: 1065−1075. doi: 10.1007/s00134-008-1085-9.
lung water, is taken into account. De-resuscitation needs 8. Wang XZ, Lu CJ, Gao FQ, Li XH, Yan WF, Ning FY: Efficacy of goal-directed
therapy in the treatment of septic shock. Zhongguo Wei Zhong Bing Ji
to be considered in those patients who do not transgress Jiu Yi Xue 2006; 18: 661−664.
spontaneously from the Ebb to Flow phase. Finally, a future 9. Early Goal−Directed Therapy Collaborative Group of Zhejiang Province:
The effect of early goal-directed therapy on treatment of critical
ARDS definition should “by definition” take into account the patients with severe sepsis/septic shock: a multi-center, prospective,
value of EVLWI in combination with the PVPI. Only recently, randomized, controlled study. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue
2010; 22: 331−334.
the American National Quality Forum (NQF) formulated 10. Tian HH, Han SS, Lv CJ et al.: The effect of early goal lactate clearance
a new interpretation with regard to adherence to the Severe rate on the outcome of septic shock patients with severe pneumonia.
Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2012; 24: 42−45.
Sepsis and Septic Shock Management Bundle quidelines 11. Yu B, Tian HY, Hu ZJ et al.: Comparison of the effect of fluid resuscitation
(NQF#0500) and state that measuring central venous pres- as guided either by lactate clearance rate or by central venous oxygen
saturation in patients with sepsis. Zhonghua Wei Zhong Bing Ji Jiu Yi
sure and central venous oxygen saturation is still preferred Xue 2013; 25: 578−583.
(http://www.qualityforum.org) [106]. All patients also need 12. Lu N, Zheng R, Lin H, Shao J, Yu J: Clinical studies of surviving sepsis
bundles according to PiCCO on septic shock patients. Zhonghua Wei
to receive a fluid bolus of 30 mLs kg-1 regardless of weight Zhong Bing Ji Jiu Yi Xue 2014; 26: 23−27.
and fluid responsiveness and regardless of the fact that they 13. Jansen TC, van Bommel J, Schoonderbeek J et al.: Early lactate-guided
therapy in icu patients: a multicenter, open-label, randomized,
may have pulmonary edema, renal failure or end-stage heart controlled trial. Am J Respir Crit Care Med 2010; 182: 752−761. doi:
failure. The NQF also announced that hospitals are required 10.1164/rccm.200912-1918OC.
14. Raising the bar with bundles: Improving the quality of care by impro-
to submit data on the SSC bundle adherence starting from ving the work environment. Joint Commission Perspectives on Patient
October 1, 2015 with the risk for penalisation if they fail to do Safety 2006; 6: 5−6.
15. Hollenberg SM, Ahrens TS, Annane D et al.: Practice parameters for
so. This is really freightening as it means that the American hemodynamic support of sepsis in adult patients: 2004 update. Crit
Federal Government seems to support overzealous futile Care Med. 2004; 32: 1928−1948.
16. Sevransky JE, Nour S, Susla GM, Needham DM, Hollenberg S, Pronovost
fluid resuscitation. At least some food for further thought P: Hemodynamic goals in randomized clinical trials in patients with
[106]. sepsis: a systematic review of the literature. Crit Care 2007; 11: R67.

53
Anaesthesiol Intensive Ther 2015, vol. 47, s44–s55

17. Dellinger RP, Carlet JM, Masur H et al.: Surviving Sepsis Campaign gu- 43. Rivers EP, Ander DS, Powell D: Central venous oxygen saturation
idelines for management of severe sepsis and septic shock. Crit Care monitoring in the critically ill patient. Curr Opin Crit Care 2001; 7:
Med 2004; 32: 858−873. 204−211.
18. Dellinger RP, Levy MM, Carlet JM et al.: Surviving Sepsis Campaign: 44. Krafft P, Steltzer H, Hiesmayr M, Klimscha W, Hammerle AF: Mixed venous
International guidelines for management of severe sepsis and septic oxygen saturation in critically ill septic shock patients. The role of
shock: 2008. Intensive Care Med. 2008; 34: 17−60. defined events. Chest 1993; 103=: 900−906.
19. Dellinger RP, Levy MM, Rhodes A et al.: Surviving sepsis campaign: inter- 45. Vincent JL, Gerlach H: Fluid resuscitation in severe sepsis and septic
national guidelines for management of severe sepsis and septic shock: shock: an evidence-based review. Crit Care Med 2004; 32(11 Suppl):
2012. Crit Care Med 2013; 41: 580−637. S451−4.
20. Tonelli MR, Curtis JR, Guntupalli KK et al.: An official multi-society sta- 46. Pope JV, Jones AE, Gaieski DF, Arnold RC, Trzeciak S, Shapiro NI: Multicenter
tement: the role of clinical research results in the practice of critical study of central venous oxygen saturation (ScvO(2)) as a predictor of
care medicine. Am J Respir Crit Care Med 2012; 185: 1117−11124. doi: mortality in patients with sepsis. Ann Emerg Med 2010; 55: 40−6 e1.
10.1164/rccm.201204-0638ST. doi: 10.1016/j.annemergmed.2009.08.014.
21. Moreno R, Rhodes A: Evidence should not be viewed in isolation. Crit Care 47. Shapiro NI, Howell MD, Talmor D et al.: Implementation and outcomes
Med 2010; 38 (10 Suppl): S528−33. doi: 10.1097/CCM.0b013e3181f1cd02. of the Multiple Urgent Sepsis Therapies (MUST) protocol. Crit Care Med
22. Perel A: Bench-to-bedside review: The initial hemodynamic resuscitation 2006; 34: 1025−1032.
of the septic patient according to Surviving Sepsis Campaign guidelines 48. van Beest PA, Hofstra JJ, Schultz MJ, Boerma EC, Spronk PE, Kuiper MA: The
— does one size fit all? Crit Care 2008; 12: 223. doi: 10.1186/cc6979. incidence of low venous oxygen saturation on admission in the ICU:
23. Packman MI, Rackow EC: Optimum left heart filling pressure during a multicenter observational study in the Netherlands. Crit Care 2008;
fluid resuscitation of patients with hypovolemic and septic shock. Crit 12: R33. doi: 10.1186/cc6811.
Care Med 1983; 11: 165−169. 49. Rivers EP, Kruse JA, Jacobsen G et al.: The influence of early hemodynamic
24. Bendjelid K: Right atrial pressure: determinant or result of change in optimization on biomarker patterns of severe sepsis and septic shock.
venous return? Chest 2005; 128: 3639−3640. Crit Care Med 2007; 35: 2016.
25. Cheatham ML, Malbrain ML: Cardiovascular implications of abdominal com- 50. Perel A, Maggiorini M, Malbrain MLNG et al.: Optimal hemodynamic
partment syndrome. Acta Clin Belg Suppl 2007: add volume (62) 98−112. management according to the Surviving Sepsis Guidelines is not
26. Malbrain ML, De Waele JJ, De Keulenaer BL: What every ICU clinician ne- applicable to all ICU patients. Crit Care (London, England) 2008;
eds to know about the cardiovascular effects caused by abdominal 12(Suppl 2): S156.
hypertension. Anaesthesiol Intensive Ther 2015; 47: 388−399. doi: 51. Sprung CL, Annane D, Keh D et al.: Hydrocortisone therapy for patients
10.5603/AIT.a2015.0028. with septic shock. N Engl J Med 2008; 358: 111−124. doi: 10.1056/NEJ-
27. Magder S: Central venous pressure: a useful but not so simple measu- Moa071366.
rement. Crit Care Med 2006; 34: 1523.5. 52. Barochia AV, Vitberg D, Cui X et al.: Bundled care for septic shock:
28. Malbrain ML, Wilmer A: The polycompartment syndrome: towards An analysis of clinical trials. Crit Care Med 2010; 38: 668−678. doi:
an understanding of the interactions between different compartments! 10.1097/CCM.0b013e3181cb0ddf.
Intensive Care Med 2007; 33: 1869−1872. 53. Vo M, Kahn JM. Making the GRADE: how useful are the new Survi-
29. Osman D, Ridel C, Ray P et al.: Cardiac filling pressures are not appropriate ving Sepsis Campaign guidelines? Crit Care 2013; 17: 328. doi:
to predict hemodynamic response to volume challenge. Crit Care Med 10.1186/cc13113.
2007; 35: 64−69. 54. Li J, Xi XM, Luo X: Analysis of a survey of SSC guideline implemented
30. Bendjelid K, Romand JA: Fluid responsiveness in mechanically ventilated among Chinese intensivists. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue
patients: a review of indices used in intensive care. Intensive Care Med 2008; 20: 155−158.
2003; 29: 352−360. 55. Liu Y, Wang DF, Shen F: Monitoring of intra-abdominal pressure in ma-
31. Nee PA, Rivers EP: The end of the line for the Surviving Sepsis Campaign, nagement of abdominal compartment syndrome complicating severe
but not for early goal-directed therapy. Emerg Med J 2011; 28: 3−4. doi: acute pancreatitis. Zhongguo Wei Zhong Bing Ji Jiu Yi Xue 2008; 20: 574.
56. Levy MM, Artigas A, Phillips GS et al.: Outcomes of the Surviving Sepsis
10.1136/emj.2010.097147.
Campaign in intensive care units in the USA and Europe: a prospective
32. Boyd JH, Forbes J, Nakada TA, Walley KR, Russell JA: Fluid resuscitation
cohort study. The Lancet Infectious Diseases. 2012; 12: 919−924. doi:
in septic shock: a positive fluid balance and elevated central venous
10.1016/S1473-3099(12)70239-6.
pressure are associated with increased mortality. Crit Care Med 2011; 57. Levy MM, Dellinger RP, Townsend SR et al.: The Surviving Sepsis Campaign:
39: 259−265. doi: 10.1097/CCM.0b013e3181feeb15. results of an international guideline-based performance improvement
33. Marik PE, Cavallazzi R: Does the central venous pressure predict fluid re- program targeting severe sepsis. Intensive care medicine. 2010; 36:
sponsiveness? An updated meta-analysis and a plea for some common 222−231. doi: 10.1007/s00134-009-1738-3.
sense. Crit Care Med 2013; 41: 1774−1781. doi: 10.1097/CCM.0b013e- 58. Cheatham ML, Malbrain ML: Cardiovascular implications of abdominal
31828a25fd. compartment syndrome. Acta Clin Belg Suppl 2007; 62: 98−112. doi:
34. Marik PE, Baram M, Vahid B: Does central venous pressure predict fluid 10.1179/acb.2007.62.s1.013.
responsiveness?: a systematic review of the literature and the tale of 59. Malbrain ML, Deeren D, De Potter TJ: Intra-abdominal hypertension in
seven mares. Chest 2008; 134: 172−178. doi: 10.1378/chest.07-2331. the critically ill: it is time to pay attention. Curr Opin Crit Care 2005;
35. Pflaum-Carison J, Gardner-Gray J, Hurst G, Rivers EP: Early Sepsis Mana- 11: 156−171.
gement. ICU Management. 2015; 15: 68−71. 60. Malbrain ML, De laet I, Cheatham M: Consensus conference definitions
36. Walkey AJ, Wiener RS, Lindenauer PK: Utilization patterns and out- and recommendations on intra-abdominal hypertension (IAH) and
comes associated with central venous catheter in septic shock: the abdominal compartment syndrome (ACS) — the long road to the
a population-based study. Crit Care Med 2013; 41: 1450−1457. doi: final publications, how did we get there? Acta Clin Belg Suppl 2007;
10.1097/CCM.0b013e31827caa89. 62: 44−59.
37. LeDoux D, Astiz ME, Carpati CM, Rackow EC: Effects of perfusion pressure 61. Cheatham ML, Malbrain ML, Kirkpatrick A et al.: Results from the inter-
on tissue perfusion in septic shock. Crit Care Med 2000; 28: 2729−2732. national conference of experts on intra-abdominal hypertension and
38. Bourgoin A, Leone M, Delmas A, Garnier F, Albanese J, Martin C: Increasing abdominal compartment syndrome. II. Recommendations. Intensive
mean arterial pressure in patients with septic shock: effects on oxygen Care Med 2007; 33: 951−962.
62. Malbrain MLNG, Roberts DJ, Sugrue M et al.: The polycompartment
variables and renal function. Crit Care Med 2005; 33: 780−786.
syndrome: a concise state-of-the-art review. Anaesthesiol Intensive
39. Dunser MW, Takala J, Ulmer H et al.: Arterial blood pressure during early
Ther 2014; 46: 433−450. doi: 10.5603/AIT.2014.0064.
sepsis and outcome. Intensive Care Med 2009; 35: 1225−1233. doi:
63. Michard F, Teboul JL: Predicting fluid responsiveness in ICU patients:
10.1007/s00134-009-1427-2. a critical analysis of the evidence. Chest 2002; 121: 2000−2008.
40. Badin J, Boulain T, Ehrmann S et al.: Relation between mean arterial 64. Parker MM: Goals for fluid resuscitation: a real challenge. Crit Care Med
pressure and renal function in the early phase of shock: a prospective, 2007; 35: 295−296.
explorative cohort study. Crit Care 2011; 15: R135. 65. Teboul JL, Pinsky MR, Mercat A et al.: Estimating cardiac filling pressure
41. Cheatham ML, White MW, Sagraves SG, Johnson JL, Block EF: Abdominal in mechanically ventilated patients with hyperinflation. Crit Care Med
perfusion pressure: a superior parameter in the assessment of intra- 2000; 28: 3631−3636.
-abdominal hypertension. J Trauma 2000; 49: 621−626; discussion 6−7. 66. Malbrain ML, Ameloot K, Gillebert C, Cheatham ML: Cardiopulmonary
42. De laet I, Malbrain ML, Jadoul JL, Rogiers P, Sugrue M: Renal implications monitoring in intra-abdominal hypertension. Am Surg 2011; 77 Suppl
of increased intra-abdominal pressure: are the kidneys the canary for 1: S23−30.
abdominal hypertension? Acta Clin Belg Suppl 2007; 62: 119−130. doi: 67. Malbrain ML, Cheatham ML: Cardiovascular effects and optimal pre-
10.1179/acb.2007.62.s1.015. load markers in intra-abdominal hypertension. In: Vincent J-L (ed.):

54
 Stefanie Vandervelden, Manu L.N.G. Malbrain, Care bundles in sepsis

Yearbook of intensive care and emergency medicine. Springer-Verlag; 89. Gurgel ST, do Nascimento P, Jr.: Maintaining tissue perfusion in high-risk
2004: 519−543. surgical patients: a systematic review of randomized clinical trials. Ane-
68. Wauters J, Claus P, Brosens N et al.: Relationship between abdominal sth Analg 2011; 112: 1384−1391. doi: 10.1213/ANE.0b013e3182055384.
pressure, pulmonary compliance, and cardiac preload in a porcine mo- 90. Hamilton MA, Cecconi M, Rhodes A: A systematic review and meta-
del. Crit Care Res Pract 2012; 2012: 763181. doi: 10.1155/2012/763181. -analysis on the use of preemptive hemodynamic intervention to
69. Malbrain ML, De Potter TJ, Dits H, Reuter DA: Global and right ventricular improve postoperative outcomes in moderate and high−risk surgical
end-diastolic volumes correlate better with preload after correction patients. Anesth Analg 2011; 112: 1392−1402. doi: 10.1213/ANE
for ejection fraction. Acta Anaesthesiol Scand 2010; 54: 622−631. doi: .0b013e3181eeaae5.
10.1111/j.1399-6576.2009.02202.x. 91. Miller TE, Roche AM, Gan TJ: Poor adoption of hemodynamic optimiza-
70. Eichhorn V, Goepfert MS, Eulenburg C, Malbrain ML, Reuter DA: Compa- tion during major surgery: are we practicing substandard care? Anesth
rison of values in critically ill patients for global end-diastolic volume Analg 2011; 112: 1274−1276. doi: 10.1213/ANE.0b013e318218cc4f.
and extravascular lung water measured by transcardiopulmonary 92. Perel A, Pizov R, Cotev S: Respiratory variations in the arterial pressure
thermodilution: a metaanalysis of the literature. Med Intensiva 2012; during mechanical ventilation reflect volume status and fluid respon-
36: 467−474. doi: 10.1016/j.medin.2011.11.014. siveness. Intensive Care Med 2014; 40: 798−807. doi: 10.1007/s00134-
71. Michard F, Alaya S, Zarka V, Bahloul M, Richard C, Teboul JL: Global end- 014-3285-9.
-diastolic volume as an indicator of cardiac preload in patients with 93. Navarro LH, Bloomstone JA, Auler JO et al.: Perioperative fluid therapy:
septic shock. Chest 2003; 124: 1900−1908. a statement from the international Fluid Optimization Group. Perioperat
72. Trof RJ, Beishuizen A, Cornet AD, de Wit RJ, Girbes AR, Groeneveld AB: Med (London, England). 2015; 4: 3. doi: 10.1186/s13741-015-0014-z.
Volume-limited versus pressure-limited hemodynamic management 94. Sakka SG, Reuter DA, Perel A: The transpulmonary thermodilution tech-
in septic and nonseptic shock. Crit Care Med 2012; 40: 1177−1185. doi: nique. J Clin Monit Comput 2012; 26: 347−353. doi: 10.1007/s10877-
10.1097/CCM.0b013e31823bc5f9. 012-9378-5.
73. Malbrain ML, Reuter DA: Hemodynamic treatment algorithms should 95. Malbrain ML, Marik PE, Witters I et al.: Fluid overload, de-resuscitation,
follow physiology or they fail to improve outcome. Crit Care Med 2012; and outcomes in critically ill or injured patients: a systematic review
40: 2923−2924. doi: 10.1097/CCM.0b013e31825f6cd0. with suggestions for clinical practice. Anaesthesiol Intensive Ther 2014;
74. Marik PE, Cavallazzi R, Vasu T, Hirani A: Dynamic changes in arterial 46: 361−380. doi: 10.5603/AIT.2014.0060.
waveform derived variables and fluid responsiveness in mechanically 96. Bagshaw SM, Bellomo R: The influence of volume management on
ventilated patients: a systematic review of the literature. Crit Care Med outcome. Curr Opin Crit Care 2007; 13: 541−548.
2009; 37: 2642−2647. doi: 10.1097/CCM.0b013e3181a590da. 97. Murphy CV, Schramm GE, Doherty JA et al.: The importance of fluid
75. Perel A, Habicher M, Sander M: Bench-to-bedside review: functional management in acute lung injury secondary to septic shock. Chest
hemodynamics during surgery — should it be used for all high-risk 2009; 136: 102−109. doi: 10.1378/chest.08-2706.
cases? Crit Care 2013; 17: 203. doi: 10.1186/cc11448. 98. Cordemans C, De laet I, Van Regenmortel N et al.: Fluid management in
76. Hofkens PJ, Verrijcken A, Merveille K et al.: Common pitfalls and tips and critically ill patients: The role of extravascular lung water, abdominal
tricks to get the most out of your transpulmonary thermodilution hypertension, capillary leak and fluid balance. Annals Intensive Care
device: results of a survey and state-of-the-art review. Anaesthesiol 2012; 2(Suppl. 1): S1. doi: 10.1186/2110-5820-2-S1-S1.
Intensive Ther 2015; 47: 89−116. 10.5603/AIT.a2014.0068. 99. Malbrain ML, De Laet I: AIDS is coming to your ICU: be prepared for acute
77. De Backer D, Heenen S, Piagnerelli M, Koch M, Vincent JL: Pulse pressure bowel injury and acute intestinal distress syndrome. Intensive Care Med
variations to predict fluid responsiveness: influence of tidal volume. 2008; 34: 1565−1569. doi: 10.1007/s00134-008-1135-3.
Intensive Care Med 2005; 31: 517−523. 100. Kuzkov VV, Kirov MY, Sovershaev MA et al.: Extravascular lung water
78. Reuter DA, Bayerlein J, Goepfert MS et al.: Influence of tidal volume on determined with single transpulmonary thermodilution correlates
left ventricular stroke volume variation measured by pulse contour with the severity of sepsis-induced acute lung injury. Crit Care Med
analysis in mechanically ventilated patients. Intensive Care Med 2003; 2006; 34: 1647−1653.
29: 476−480. 101. Michard F, Fernandez-Mondejar E, Kirov MY, Malbrain M, Tagami T: A new
79. Mahjoub Y, Pila C, Friggeri A et al.: Assessing fluid responsiveness in cri- and simple definition for acute lung injury. Crit Care Med 2012; 40:
tically ill patients: false-positive pulse pressure variation is detected by 1004−1006. doi: 10.1097/CCM.0b013e31823b97fd.
Doppler echocardiographic evaluation of right ventricle. Crit Care Med 102. Ranieri M, Rubenfeld GD, Force TADT: Acute respiratory distress
2009; 37: 2570-2575. doi: 10.1097/CCM.0b013e3181a380a3. syndrome the berlin definition. JAMA 2012; 307: E1−8. doi:
80. Malbrain ML, de Laet I: Functional hemodynamics and increased intra- 10.1001/jama.2012.5669.
-abdominal pressure: same thresholds for different conditions ...? Crit 103. Perel A: Extravascular lung water and the pulmonary vascular perme-
Care Med 2009; 37: 781−783. doi: 10.1097/CCM.0b013e318194c397. ability index may improve the definition of ARDS. Crit Care 2013; 17:
81. Cavallaro F, Sandroni C, Marano C et al.: Diagnostic accuracy of passive 108. doi: 10.1186/cc11918.
leg raising for prediction of fluid responsiveness in adults: systematic 104. Cabrera JL, Pinsky MR: Management of septic shock: a protocol-
review and meta-analysis of clinical studies. Intensive Care Med 2010; -less approach. Crit Care 2015; 19: 260. doi: 10.1186/s13054-015-
36: 1475−1483. doi: 10.1007/s00134-010-1929-y. -0968-8.
82. Monnet X, Bleibtreu A, Ferre A et al.: Passive leg-raising and end-expi- 105. Angus DC, Barnato AE, Bell D et al.: A systematic review and meta-ana-
ratory occlusion tests perform better than pulse pressure variation in lysis of early goal-directed therapy for septic shock: the ARISE, ProCESS
patients with low respiratory system compliance. Crit Care Med 2012; and ProMISe Investigators. Intensive Care Med 2015; 41: 1549−1560.
40: 152−157. doi: 10.1097/CCM.0b013e31822f08d7. doi: 10.1007/s00134-015-3822-1.
83. Monnet X, Rienzo M, Osman D et al.: Passive leg raising predicts fluid 106. Marik P, Bellomo R: A rational approach to fluid therapy in sepsis.
responsiveness in the critically ill. Crit Care Med 2006; 34: 1402−1407. Br J Anaesth. 2015 pii: aev349. [Epub ahead of print].
84. Monnet X, Teboul JL: Passive leg raising. Intensive Care Med 2008; 34:
659−663. doi: 10.1007/s00134-008-0994-y.
85. Mahjoub Y, Touzeau J, Airapetian N et al.: The passive leg-raising ma- Corresponding author:
neuver cannot accurately predict fluid responsiveness in patients with
intra-abdominal hypertension. Crit Care Med 2010; 38: 1824−1829. doi: Manu L.N.G. Malbrain, MD, PhD
10.1097/CCM.0b013e3181eb3c21. Intensive Care Unit and High Care Burn Unit Director
86. Malbrain ML, Reuter DA: Assessing fluid responsiveness with the pas- Ziekenhuis Netwerk Antwerpen, ZNA Stuivenberg
sive leg raising maneuver in patients with increased intra-abdominal
pressure: be aware that not all blood returns! Crit Care Med 2010; 38: Lange Beeldekensstraat 267
1912−1915. doi: 10.1097/CCM.0b013e3181f1b6a2. B-2060 Antwerpen 6, Belgium
87. Maitland K, Kiguli S, Opoka RO et al.: Mortality after fluid bolus in Afri- e-mail: [email protected]
can children with severe infection. N Engl J Med 2011; 364: 2483−2495.
doi: 10.1056/NEJMoa1101549.
88. Beale RJ, Hollenberg SM, Vincent JL, Parrillo JE: Vasopressor and inotropic
support in septic shock: an evidence-based review. Crit Care Med 2004; Received: 23.09.2015
32(11 Suppl): S455−65. Accepted: 15.11.2015

55