American Journal of Emergency Medicine xxx (2015) xxxxxx

Contents lists available at ScienceDirect

American Journal of Emergency Medicine

journal homepage: www.elsevier.com/locate/ajem

Controversies

Do not drown the patient: appropriate uid management in critical illness

Kees H. Polderman, MD, PhD, Joseph Varon, MD
Department of Critical Care Medicine, University of Pittsburgh Medical Center, Pittsburgh, PA, USA
The University of Texas Health Science Center at Houston, Houston, TX, USA
The University of Texas Medical Branch at Galveston, Galveston, TX, USA
University General Hospital, Houston, TX, USA

a r t i c l e

i n f o

Article history:
Received 29 November 2014
Received in revised form 28 January 2015
Accepted 29 January 2015
Available online xxxx

a b s t r a c t
Administering intravenous uids to support the circulation in critically ill patients has been a mainstay of
emergency medicine and critical care for decades, especially (but not exclusively) in patients with distributive
or hypovolemic shock. However, in recent years, this automatic use of large uid volumes is beginning to be
questioned. Analysis from several large trials in severe sepsis and/or acute respiratory distress syndrome have
shown independent links between volumes of uid administered and outcome; conservative uid strategies
have also been associated with lower mortality in trauma patients. In addition, it is becoming ever more clear
that central venous pressure, which is often used to guide uid administration, is a completely unreliable parameter of volume status or uid responsiveness. Furthermore, 2 recently published large multicenter trials (ARISE
and ProCESS) have discredited the early goal-directed therapy approach, which used prespecied targets of
central venous pressure and venous saturation to guide uid and vasopressor administration. This article discusses the risks of iatrogenic submersion and strategies to avoid this risk while still giving our patients the uids
they need. The key lies in combining good clinical judgement, awareness of the potential harm from excessive
uid use, restraint in reexive administration of uids, and use of data from sophisticated monitoring tools
such as echocardiography and transpulmonary thermodilution. Use of smaller volumes to perform uid
challenges, monitoring of extravascular lung water, earlier use of norepinephrine, and other strategies can help
further reduce morbidity and mortality from severe sepsis.
2015 Elsevier Inc. All rights reserved.

One of the most challenging and controversial areas in the care of

emergent and critically ill patients is the administration of intravenous
uids to support the circulation. This does not only apply to hemodynamically unstable patients. In clinical conditions such as subarachnoid
hemorrhage, large volumes of uid are often administered over
prolonged periods to reduce the risk of vasospasm, often targeting a
positive uid balance or a specic central venous pressure (CVP) [1].
However, especially when a patient presents with a distributive or hypovolemic shock, rapid administration of uids is one of the mainstays
of treatment, one that has been recommended for decades. This applies
to both the initial and later phases of treatment, especially in distributive shock. The 2012 Surviving Sepsis Campaign guidelines recom-

Disclosures: Neither of the authors has a relevant conict of interest to declare.

Corresponding author at: Department of Critical Care Medicine, University of Pittsburgh Medical Center, 3550 Terrace St, Scaife Hall/6th Floor, Pittsburgh, PA 15261.
E-mail addresses: k.polderman@tip.nl, PoldermanKH@upmc.edu (K.H. Polderman).

mend an initial uid challenge, followed by continued uid

administration if hypotension persists or blood lactate concentration
exceeds 4 mmol/L [2]. Again, CVP is often used to guide uid volume;
this, in spite of abundant evidence showing that CVP is completely unreliable as a parameter of volume status or uid responsiveness [2-5].
In 2001, a highly inuential single-center study reported that uid
and vasopressor administration using prespecied targets including a
CVP of 8 to 12 and venous saturation greater than 65% in the rst
6 hours of sepsis could reduce mortality by 15.8% [6]. This approach
was termed early goal-directed therapy (EGDT) [6]. Recently, 2 large
multicentered studies (the ProCESS trial and the ARISE trial) failed to
demonstrate any benets of the EGDT approach [7,8]; in spite of this,
current guidelines still recommend EGDT, and a recent statement on
behalf of the Surviving Sepsis Campaign panel put out after publication of the ProCESS trial suggests that no change in guidelines will be
forthcoming because the ProCESS trial used protocolized care in all
study groups, and thus, its negative ndings do not invalidate the
EGDT approach [9].

Please cite this article as: Polderman KH, Varon J, Donot drownthepatient: appropriateuidmanagement in critical illness, Am J Emerg Med
(2015), http://dx.doi.org/10.1016/j.ajem.2015.01.051

K.H. Polderman, J. Varon / American Journal of Emergency Medicine xxx (2015) xxxxxx

Table
Diagnostic tools and methods to determine volume status and predict uid responsiveness
Clinical assessment

Devices needed

Comments

Blood pressure
Heart rate
Urinary output
Capillary rell
Peripheral temperature
Neurological examination

Blood pressure cuff/arterial line

Electrocardiogram/arterial line
Urinary catheter
N/A
Temperature probe
N/A

Cheap, easy to obtain; should be the basis of our

assessments. However, supplemental information
is often needed in more severely ill patients,
especially in cases of shock with multiple causes.

Laboratory equipment (in laboratory or

as point-of-care equipment)

Allows assessment of changes over time periods

of several hours. Changes in lactate levels (or lack
thereof) have been shown to correlate with
outcome. Chlorine can be used to assess metabolic
acidosis/chlorine overload.

Central venous catheter

PA catheter, PiCCO, LidCO, FloTrac,
echocardiography, USCOM

Poor prediction of volume status

Fair prediction of volume status. Some devices
(eg, FloTrac) are less reliable in patients with
more severe critical illness.
Fair to good prediction of volume status
Fair prediction of volume status
Good prediction of volume status Invasive; catheter
must be removed within 96 h
Fair to good prediction of volume status
Good safety parameter for volume overload
Fair to good prediction of volume status
Good prediction of volume status

Biochemical parameters
Base excess
Serial lactate
Serum creatinine/urea
Serum chlorine

Hemodynamic monitoring
CVP
Cardiac output

Stroke volume variation

Venous saturation
Mixed venous saturation

Arterial line, PiCCO, LidCO, FloTrac

Central venous and PA catheters
PA catheter

Blood volume
EVLW
Intrathoracic blood volume
Systolic/diastolic function

PiCCO, LidCO
PiCCO
PiCCO
TEE/TTE, PiCCO, LidCO

FloTrac is a proprietary arterial waveform analysis and cardiac output monitoring system. Abbreviations: PA, pulmonary artery; PiCCO, pulse contour cardiac output; LidCO, lithium dilution
cardiac output; TEE, transesophageal echocardiography; TTE, transthoracic echocardiography; USCOM, ultrasound cardiac output monitoring.

However, in recent years, a number of studies in emergency medicine and critical care have raised concerns over the practice of unrestrained uid administration that has become so ingrained in daily
practice. One of the rst studies to address this issue was a clinical
trial in children performed in Africa, designed to compare uid bolus
of albumin to normal saline with the hypothesis that albumin administration might improve outcome in sepsis [10]. Children receiving normal saline and albumin uid bolus had similar outcomes, but
mortality was signicantly lower in children who had not received
any uid bolus [10]. As the study had been performed in poor countries
in Africa, there were questions regarding the applicability of these ndings to industrial countries with modern health care systems. The
causes of infection, disease course, time to medical treatment, health
care delivery systems, preexisting conditions, and many other factors
are very different in Africa compared to Western countries. Although
these criticisms are valid, other reports support the initial observations
and lend credence to the hypothesis that excessive uid administration
can be detrimental. A post hoc analysis of the acute respiratory distress
syndrome network (ARDS-NET) showed that a negative cumulative
uid balance at day 4 was associated with signicantly lower mortality,
independent of other measures of severity of illness including a diagnosis of sepsis [11]. This observation was conrmed in another study of patients with acute respiratory distress syndrome (ARDS) secondary to
septic shock [12]. In fact, in this study, patients receiving uid management considered inadequate but conservative had better outcomes
than uid administration considered adequate but liberal [12]. Similarly, in the Vasopressin vs. Norepinephrine Infusion in Patients with
Septic Shock trial, a more positive uid balance both early in resuscitation and cumulatively over 4 days was associated with increased risk
of mortality in septic shock, corrected for other factors [13].
This does not just apply to sepsis. A recent meta-analysis of randomized controlled trials and cohort studies and cohort studies found that
conservative uid strategies were associated with lower mortality in
trauma patients [14]. In addition, it is not only unrestrained crystalloid
infusion that is being called into question; a recent study reported that
transfusion of red blood cells was associated with signicantly worse
outcomes in patients with traumatic brain injury and no evidence of

shock if the initial hemoglobin was greater than 10 g/dL [15]. Twenty
years ago, Bickell et al [16] challenged the practice of early uid resuscitation in patients with penetrating injuries, suggesting that this practice
might be linked to increased bleeding and adverse outcomes. This issue
still remains controversial [17].
In this issue of The American Journal of Emergency Medicine, Sirvent
[18] reports on the effects of uid administration at the onset of severe
sepsis and septic shock. The author found that the accumulated positive
uid balance in the rst 48, 72, and 96 hours was signicantly associated
with increased mortality [18]. These results are in keeping with the results discussed above and remind us again of the risks of iatrogenic
submersion. How can we avoid this risk, while still giving our patients
the uids that they may need?
In our view, the key lies in a multipronged approach, using clinical
judgment along with sophisticated monitoring tools to guide our treatment. The rst step is awareness and restraint: awareness that excessive uid administration could be harmful, and restraint in the
restraint in the reexive administration of uids. If a patient does not respond to a bolus of uid, we should think twice before giving yet more
uids or trying yet another uid challenge; instead, we might consider
earlier initiation of pressors or perhaps accepting less ambitious target
values. Especially, we should not target specic CVPs to guide treatment; rather, a combination of clinical and biochemical parameters
and more sophisticated hemodynamic monitoring (echocardiography,
cardiac output, extravascular lung water [EVLW], and stroke volume
variation) can be used to better tailor our therapeutic approach (see
Table). Fluid responsiveness can be assessed with smaller volumes
(100-250 mL administered rapidly, rather than 500-1000 mL as is common practice). Monitoring of EVLW may be a valuable safety parameter
to prevent uid overload. A recent study in ARDS patients suggests that
high EVLW is an independent risk factor for mortality in ARDS [19]. This
approach may also apply to less sick patients; to take the earlier example of subarachnoid hemorrhage, a recent randomized controlled trial
reported signicantly improved outcomes using preload volume and
cardiac output (monitored by transpulmonary thermodilution) to
guide treatments compared to patients where traditional parameters
such as uid balance and CVP were used [20]. Noninvasive devices such

Please cite this article as: Polderman KH, Varon J, Donot drownthepatient: appropriateuidmanagement in critical illness, Am J Emerg Med
(2015), http://dx.doi.org/10.1016/j.ajem.2015.01.051

K.H. Polderman, J. Varon / American Journal of Emergency Medicine xxx (2015) xxxxxx

as continuous wave Doppler ultrasound cardiac output monitoring may

also have a useful role in cardiac output monitoring [21,22].
Mortality was signicantly lower in all arms of the ProCESS trial than
in the initial EGDT study [6,7]. In the EGDT arm of the ProCESS trial, the
use of vasopressors in the rst 6 hours was double (54.9% vs 27.4%), and
the volume infused in the rst 72 hours half (7.22 vs 13.44 L) compared
to the EGDT study. This suggests that more restrictive uid management
is feasible even when using a judicious EGDT approach. We urge the
readers to take to heart the important lessons of Sirvent and from previous trials and not to use too much of a good thing.

References
[1] Meyer R, Deem S, Yanez ND, Souter M, Lam A, Treggiari MM. Current practices of
triple-H prophylaxis and therapy in patients with subarachnoid hemorrhage.
Neurocrit Care 2011;14:2436.
[2] Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign: international
guidelines for management of severe sepsis and septic shock: 2012. Crit Care Med
2013;41:580637.
[3] Shippy CR, Appel PL, Shoemaker WC. Reliability of clinical monitoring to assess
blood volume in critically ill patients. Crit Care Med 1984;12:10712.
[4] Osman D, Ridel C, Ray P, Monnet X, Anguel N, Richard C, et al. Cardiac lling pressures are not appropriate to predict hemodynamic response to volume challenge.
Crit Care Med 2007;35:648.
[5] Marik PE, Cavallazzi R. Does the central venous pressure predict uid responsiveness? An updated meta-analysis and a plea for some common sense. Crit Care
Med 2013;41:177481.
[6] Rivers E, Nguyen B, Havstad S, et al. Early goal-directed therapy in the treatment of
severe sepsis and septic shock. N Engl J Med 2001;345:136877.
[7] The ProCESS Investigators. A randomized trial of protocol-based care for early septic
shock. N Engl J Med 2014;370:168393.
[8] ARISE Investigators, ANZICS Clinical Trials Group, Peake SL, Delaney A, Bailey M,
Bellomo R, et al. Goal-directed resuscitation for patients with early septic shock. N
Engl J Med 2014;371:1496506.

[9] Surviving Sepsis Campaign. Surviving Sepsis Campaign responds to ProCESS trial.
Updated May 19, 2014 http://www.survivingsepsis.org/SiteCollectionDocuments/
SSC-RespondsProcess-Trial.pdf. [Accessed November 28, 2014].
[10] Maitland K, Kiguli S, Opoka RO, et al. Mortality after uid bolus in African children
with severe infection. N Engl J Med 2011;364:248395.
[11] Rosenberg AL, Dechert RE, Park PK, Bartlett RH, NIH NHLBI ARDS Network. Review of
a large clinical series: association of cumulative uid balance on outcome in acute
lung injury: a retrospective review of the ARDSnet tidal volume study cohort. J Intensive Care Med 2009;24:3546.
[12] Murphy CV, Schramm GE, Doherty JA, Reichley RM, Gajic O, Afessa B, et al. The importance
of uid management in acute lung injury secondary to septic shock. Chest 2009;136:1029.
[13] Boyd JH, Forbes J, Nakada TA, Walley KR, Russell JA. Fluid resuscitation in septic
shock: a positive uid balance and elevated central venous pressure are associated
with increased mortality. Crit Care Med 2011;39:25965.
[14] Wang CH, Hsieh WH, Chou HC, Huang YS, Shen JH, Yeo YH, et al. Liberal versus restricted uid resuscitation strategies in trauma patients: a systematic review and
meta-analysis of randomized controlled trials and observational studies. Crit Care
Med 2014;42:95462.
[15] Elterman J, Brasel K, Brown S, et al. Transfusion of red blood cells in patients with a
prehospital Glasgow Coma Scale score of 8 or less and no evidence of shock is associated with worse outcomes. J Trauma Acute Care Surg 2013;75:814.
[16] Bickell WH, Wall Jr MJ, Pepe PE, et al. Immediate versus delayed uid resuscitation for
hypotensive patients with penetrating torso injuries. N Engl J Med 1994;331:11059.
[17] Kwan I, Bunn F, Chinnock P, Roberts I. Timing and volume of uid administration for
patients with bleeding. Cochrane Database Syst Rev 2014;3:CD002245.
[18] Sirvent JM. Fluid balance in sepsis and septic shock as a determining factor of mortality. Am J Emerg Med 2015.
[19] Jozwiak M, Silva S, Persichini R, et al. Extravascular lung water is an independent
prognostic factor in patients with acute respiratory distress syndrome. Crit Care
Med 2013;41:47280.
[20] Mutoh T, Kazumata K, Terasaka S, Taki Y, Suzuki A, Ishikawa T. Early intensive versus
minimally invasive approach to postoperative hemodynamic management after
subarachnoid hemorrhage. Stroke 2014;45:12804.
[21] Udy AA, Altukroni M, Jarrett P, Roberts JA, Lipman J. A comparison of pulse contour
wave analysis and ultrasonic cardiac output monitoring in the critically ill. Anaesth
Intensive Care 2012;40:6317.
[22] Chong SW, Peyton PJ. A meta-analysis of the accuracy and precision of the ultrasonic
cardiac output monitor (USCOM). Udy AA1, Altukroni M, Jarrett P, Roberts JA,
Lipman J Anaesthesia 2012;67:126671.

Please cite this article as: Polderman KH, Varon J, Donot drownthepatient: appropriateuidmanagement in critical illness, Am J Emerg Med
(2015), http://dx.doi.org/10.1016/j.ajem.2015.01.051