See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/257596621

Crystalloids, colloids, blood, blood products and blood substitutes

Article  in  Anaesthesia & intensive care medicine · June 2013

DOI: 10.1016/j.mpaic.2013.03.003

CITATIONS READS

3 5,446

2 authors, including:

Roop Kishen
The University of Manchester
48 PUBLICATIONS   463 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

A review of Stewart approach View project

A review article View project

All content following this page was uploaded by Roop Kishen on 16 October 2017.

The user has requested enhancement of the downloaded file.

 CLINICAL ANAESTHESIA
Crystalloids, colloids, blood,
blood products and blood
substitutes
Hugo Buckley
Roop Kishen
Abstract
Understanding the physiology of fluid distribution within the human body
is fundamental to the practice of anaesthetists and intensivists of all
grades. There is a necessity to recognize the range of actions and conse-
quences of the commonly infused intravenous fluids if safe patient care is
to be provided. There are many historical and on-going trials surrounding
fluid therapy and it is important for the physician to keep up to date with
current guidelines.
There is a continued drive to improve the safety of donor blood and
prevent transfusion errors. Knowledge of how blood products are
collected separated and stored is essential to prevent harm to patients
through transfusions. Work in producing blood substitutes is progressing,
but to date, trials have failed to market a product in Europe and the USA
with an acceptable risk profile.
Keywords Blood; blood substitutes; coagulation; colloid; crystalloid;
plasma; platelets
Royal College of Anaesthetists CPD Matrix: 1A01, 1A02, 1E04, 2A05
Physiology
A solution is a product of a chemical reaction that completely
incorporates a solute into a solvent. The size of the dissolved
molecules determines their ability to cross the capillary mem-
brane. The capillary membrane acts as a semi-permeable mem-
brane and by measuring the amount of solute present, in terms of
osmolarity and osmolality, we can describe fluids as hypotonic,
isotonic and hypertonic.
Fluid is distributed across the intracellular and extracellular
compartments. Total body water accounts for 55e60% of a 75-kg
male’s body weight, which is the equivalent of 45 litres of water.
The intracellular compartment contains 30 litres, with the
extracellular compartment holding the remaining 15 litres. This
extracellular compartment is further subdivided into the inter-
stitial and transcellular space comprising of 11.5 litres and the
Hugo Buckley FRCA is a ST7 in Anaesthesia and Intensive Care Medicine
in the North Western Deanery, UK. Conflicts of interest: none declared.
Roop Kishen FRCA is a Retired Consultant Intensivist and Anaesthetist at
the Salford Royal Foundation Trust, UK. Conflicts of interest: none
declared.
ANAESTHESIA AND INTENSIVE CARE MEDICINE 14:6
Learning objectives
After reading this article you should be able to:
C
C
describe the physiology of the body’s fluid compartments
recognize the effects of the commonly used non-blood fluids
C list the advantages and disadvantages of non-blood fluid
therapy
C explain the methodology of blood collection and storage
C summarize current and prospective methods utilized to reduce
the risk of donor blood transfusion
plasma only accounting for 3.5 litres, which is 8% of total body
water.
Solutions will ultimately follow the same pattern of distri-
bution as body water. Forces work across the membranes to
control water balance. Osmosis, osmolality and tonicity deter-
mine water flow across the capillary bed’s semi-permeable
membrane. Sodiumepotassium adenosine triphosphatase
(ATPase) pumps help control the balance of electrolytes. These
processes are described through the GibbseDonnan equation
and an understanding of Starling’s forces. GibbseDonnan de-
scribes the demand for electrical neutrality across a membrane,
whereas Starling explains the balance of oncotic and hydrostatic
pressures in the arteriole and venoule capillary bed.
In practice therefore the distribution of a fluid across the
body’s compartments is dependent on the size, type and con-
centration of its solute. A solution containing small, rapidly
metabolized molecules will quickly re-distribute as free water
whereas one containing large, difficult to metabolize molecules
will remain in the intravascular space for longer.
Crystalloids
Crystalloids contain solutes of a low molecular weight dissolved
in water. One that contains ions or molecules that are difficult to
metabolize will initially remain in the extracellular space. The
subsequent rise in osmotic pressure will draw water out of the
intracellular space increasing intravascular volume further. This
will last until the ions re-distribute into the intra-cellular space, a
process dependent on the permeability of the cell membrane and
activity of the ATPase pumps. The end result is a temporary
expansion of the intravascular volume but ultimately a redistri-
bution of free water into all compartments. Small sugar solutions
will quickly undergo this process rapidly leaving behind free
water to be re-distributed.
Salt solutions
Most salt solutions are isotonic and isosmotic and can be further
divided into non-balanced and balanced solutions. There is an
increasingly recognized role for hypertonic solutions.
0.9% Sodium chloride solution: is an isotonic and isosmotic
solution when compared to plasma. However the solution con-
tains a far higher chloride concentration compared to that of
plasma (Table 1) that has led to criticism of its use. The concern
is that an infusion of large volumes will lead to a hyper-
chloraemic metabolic acidosis (HCMA). There is little evidence
255 Ó 2013 Elsevier Ltd. All rights reserved.
 CLINICAL ANAESTHESIA

Electrolyte content of crystalloids versus plasma

Type of fluid Sodium Chloride Potassium Calcium Others Osmolality pH
(mmol/l) (mmol/l) (mmol/l) (mmol/l) (mosmol/kg)

Plasma 135e145 98e105 3.5e5.0 2.2e2.6 280e300 7.35e7.45

5% Dextrose e e e e D-glucose 50 g/l 278 4
10% Dextrose e e e e D-glucose 100 g/l 278 4
0.9% Saline 154 154 e e e 308 5
1.8% Saline 308 308 e e e 616
0.18% Saline/4% 30 30 e e D-glucose 40 g/l 283 4-5
glucose
Compound sodium 131 111 5 2 Lactate 29 mmol/l 278 5
lactate (Hartmann’s)

Table 1

that this is of clinical consequence, most likely in part due to the
kidney’s ability to excrete the excess chloride. However the
British Consensus Guidelines on Intravenous Fluid Therapy for
Adult Surgical Patients (GIFTASUP) recommend the use of
balanced salt solutions over 0.9% sodium chloride when crys-
talloid resuscitation and replacement is indicated in surgical
patients.1 They quote the risk of HCMA as the reason for this
with level 1b evidence.
The mechanism behind the resultant HCMA is best under-
stood using Stewart’s approach to acidebase balance.2 This
approach relies on the theory that electrical neutrality must al-
ways be maintained and therefore the addition of an anion needs
to be balanced by the production of cations.
The strong ion difference (SID), total quantity of weak acids
(mainly albumin) and carbon dioxide are recognized by Stewart
as the three major components or independent variables that
influence plasma pH. In this instance carbon dioxide quantity
can be considered stable. Large infusions of crystalloid will
cause a dilution in albumin levels and a resultant alkalosis.
However this effect is overwhelmed by the change in SID, which
becomes the predominant influence on plasma pH. This can be
calculated as:



SID ¼ Na1 þ Kþ þ Ca2þ þ Mg2þ  Cl þ Lactate :
An increase in chloride levels leads to a decrease in strong ion
difference that needs to be balanced. This causes dissociation of
water into hydrogen and hydroxide ions. The increase in plasma
hydrogen ions leads to the resultant acidosis.
However, CHEST (Crystalloid versus Hydroxyethyl Starch
Trial) investigators have recently published in favour of using
0.9% sodium chloride over starches during the resuscitation of
intensive care patients.3
resuscitation of trauma and critically ill patients.4 However there
is growing evidence that hypertonic saline solutions may be
beneficial in the management of raised intra-cranial pressure
over mannitol. They are also used to correct hyponatraemia.
When being administered caution must be exercised. The
higher concentrations are irritant to smaller vessels and large
bore intravenous access, often via a central line, may be
required. In correcting hypernatraemia, care must be taken to not
cause a rapid rise in plasma sodium concentration that can lead
to disruption of osmotic gradients across the blood–brain barrier
and potentially result in osmotic demyelination syndrome.
Balanced salt solutions: the two most commonly used balanced
solutions are Hartmann’s solution (compound sodium lactate
solution) and Lactated Ringer’s solution. The electrolyte
composition of both solutions closely resembles that of plasma.
For this reason they are often perceived as ‘physiological’ solu-
tions when compared to other crystalloids.
In both instances lactate has been added in order to reduce the
chloride concentration found in 0.9% saline. The lactate acts as
an anion substitute for chloride and maintains the solutions’
electro neutrality. Benefits from this include a reduction in the
incidence of HCMA, which forms the foundation of GIFTASUP’s
recommendation discussed earlier.
These solutions have a buffering capacity through the meta-
bolism of lactate in the liver. Gluconeogenesis removes lactate
form the solution leaving sodium behind which directly in-
fluences the SID (an increased SID will reduce plasma pH).
Oxidation metabolizes a proportion of lactate to bicarbonate also
providing some buffering capacity. The solutions are not rec-
ommended for use in patients with diabetes mellitus for this
reason, but there is very little evidence of any clinical harm from
their use in this patient cohort.5

Hypertonic sodium chloride solution: sodium chloride con- Sugar solutions

centrations greater than 0.9% are available as 1.8, 3 and 30%. In
these concentrations saline expands the circulating volume by
drawing water from the extravascular space when administered
intravenously. This is a direct result from its high tonicity.
A Cochrane systematic review in 2004 showed no benefit in
using hypertonic solutions against near isotonic solutions for the
Sugar solutions utilize the D-isomer of glucose, dextrose, as the
solute and water as the solvent. They are formulated as a per-
centage, the most common clinically used being 5 and 10%.
On intravenous administration dextrose is rapidly metabo-
lized leaving behind free water that redistributes across the
bodies compartments. This short plasma half-life and inability to

ANAESTHESIA AND INTENSIVE CARE MEDICINE 14:6 256 Ó 2013 Elsevier Ltd. All rights reserved.
 CLINICAL ANAESTHESIA
maintain an osmotic pressure leaves sugar solutions little role as
plasma expanders.
Sugar solutions do however have an important role as a
maintenance fluid, helping fulfil the body’s free water require-
ment. They are often used as part of the management of hyper-
natraemia where the cause is often due to water depletion.
Their use is more common in paediatric setting where
dextrose is often combined with a saline solution.
Colloids
Colloids rely on a suspension of synthetic molecules of a large
enough molecular weight to exert an oncotic pressure across the
capillary basement membrane. In clinical practice gelatins,
starches or dextrans are suspended in a salt solution, commonly
0.9% sodium chloride. In recognition of the potential detrimental
effects of HCMA balanced colloid solutions are now produced
with a reduced chloride concentration.
Gelatins
Gelatins use polypeptide molecules derived primarily from
bovine collagen. They undergo modification to reduce their vis-
cosity in one of two ways:
 hydroxylation and succinylation, as in Gelofusine, Isoplex
and Volplex
 degradation and modification with nitrogen, as in
Haemaccel.
Gelatins have a mean molecular weight of about 35 kDa.
Compared to crystalloids this enables them to exert a greater
oncotic pressure across the basement membrane resulting, with a
half-life of approximately 4 hours, in greater plasma volume
expansion.
In conjunction with other colloidal plasma expanders,
gelatins can cause mild urticarial reactions. Severe anaphylactic
reactions have a reported incidence of 1:6000 to 1:13,000 units
administered.
Dextrans
Dextran is formed from lactic acid-producing bacteria, most
commonly Leuconostoc mesenteroides. It is a high-molecular-
weight branched polysaccharide and preparations average be-
tween 40 and 70 kDa, commonly suspended in 0.9% or 7.5%
saline.
In clinical practice dextran is used for its colloidal, anticoag-
ulant and hypo-viscosity effects on the plasma. As with the gel-
atins, the high molecular weight of dextran exerts a large oncotic
pressure thus expanding intravascular volume. Dextran mediates
its anticoagulant effects by reducing platelet and erythrocyte
aggregation. It also reduces levels of von WiIlebrand factor
further decreasing platelet function. Furthermore dextran acts
as a plasminogen activator providing thrombolytic properties.
Anticoagulant and antithrombotic effects reduce clot production
and sustainability and dextran’s oncotic effects reduce plasma
haematocrit and thus viscosity. The combination of these two
effects achieves an improvement in blood flow.
Dextran use has however been linked to direct nephrotoxicity
causing acute kidney injury (AKI). There is also a recognized
incidence of anaphylaxis and it has been shown to interfere with
blood cross-matching.
ANAESTHESIA AND INTENSIVE CARE MEDICINE 14:6
RescueflowÔ, 6% dextran 70/7.5% saline, has been pro-
moted in the management of acute hypovolaemia associated
with trauma and in the management of raised intra-cranial
pressure.6 There are no strong data to promote its use over
mannitol or hypertonic saline in the management of raised intra-
cranial pressure, both of which do not have the side effect profile
of dextran.
Starches
Hydroxyethyl starches (HES) are synthetic polymers of glucose
derived from the breakdown of amyloid peptin by amylase.
There are multiple preparations that can be differentiated by
three characteristics, which influence the length of time taken to
metabolize the starch molecules and hence a solution’s effect on
plasma expansion.
Molecular weight (MW): Within a given solution starch molecules
vary in weight. A solution is therefore defined by its average mo-
lecular weight and can be subdivided as low (70e130 kDa), me-
dium (200e260 kDa) and high MW (>450 kDa) groups. After
intravenous administration the molecules are metabolized and at a
MW of less than 50 kDa the molecules can be excreted in the urine.
Degree of substitution: within a starch solution there are a
mixture of glucose polymers with hydroxyl group substitutions
and those without. The ratio of the two within a solution is called
the substitution ratio (SR). A fluid with higher SR takes longer to
metabolize and increases the duration of plasma expansion.
C2/C6 ratio: the hydroxyl group substitutions occur at the C2, C3
and C6 positions of the glucose polymer. Substitution at C2 in-
creases the time taken to metabolise the molecule compared to
C6 substitution. Hence a high C2/C6 ration (>8) has an increased
duration of plasma expansion.
Initial starches contained high MW (e.g. 480) and high SR
(0.6), written as 480/0.6. These solutions were shown to provide
a plasma expansion of up to 24 hours. However concern devel-
oped over the development of coagulopathies secondary to a
reduction in factor VIII and von Willebrand factor as well as
increased incidence of AKI. Production was switched to medium
starch solutions (200/0.5) but not only did these demonstrate the
same anti-coagulopathic properties and AKI requiring dialysis
they also showed a trend towards increases mortality.7
On this basis most physicians switched to lowweight starches
(130/0.4). However two large trials have demonstrated direct
harm from their infusion when compared to crystalloid admin-
istration. The Scandinavian Starch for Severe Sepsis/Septic shock
trial demonstrated an increased mortality and requirement for
dialysis in those patients with sepsis who received low molecular
weight starch.8 Most recently the Australian and New Zealand
Intensive Care Society Clinical Trial Group demonstrated no
mortality difference at 90 days but an increased need for dialysis
with starch administration.3
Albumin
Albumin is a purified single polypeptide derived from pooled
donor plasma. It has an average MW of 65e69 kDa. It comes in
isotonic and concentrated formulations and being purified can be
considered free from the risk of transmitting infection.
257 Ó 2013 Elsevier Ltd. All rights reserved.
 CLINICAL ANAESTHESIA
Plenty of controversy has surrounded its use in critical care. In
1998 the Cochrane Injuries Group Albumin Reviewers recom-
mended the cessation of albumin administration to critically
ill patients after a meta-analysis demonstrated an absolute
increased risk of death of 6%.9 The Saline versus Albumin Fluid
Evaluation study investigators (SAFE) demonstrated no differ-
ence in outcome or harm between 4% albumin and normal saline
within critical care. Subgroup analysis showed a possible
exception in the management of trauma patients with head in-
juries, where albumin may increase harm.10
Blood products
Almost 17 million units of blood are donated in Europe each year
and are now grouped and typed to prevent incompatibility re-
actions. In 1901 Karl Landsteiner discovered the most well
known of the grouping systems, the ABO groups. Prior to this,
blood transfusions, which date back to the 17th century, were
associated with mortality in the region of 50%. Since the devel-
opment of the ABO system, identification of blood’s rhesus status
and new antigen recognition has further driven down the
complication rates.
New challenges surrounding blood safety have developed with
the need to avoid bacterial, viral and prion disease transmission.
White cells, recognized as pathogens, have been reduced in
donated blood in order to reduce graft versus host disease and
transfusion-related lung injury (TRALI). The UK monitors trans-
fusion adverse advents through the independent advisory group:
Serious Hazards of Transfusion (SHOT). Data from 2011 demon-
strate that the most frequent adverse event is the transfusion of the
incorrect blood component.11 This adverse event is documented
throughout developed nations, and where documented, in devel-
oping countries as well. The main safety challenge now is deliv-
ering the right product to the right patient at the right time.
NHS Blood and Transplant (NHSBT) is responsible for the
collection and distribution of blood products in the UK. Whole
blood is collected and currently separated into red cells, platelets
and plasma (Figure 1).
Whole blood
Whole blood has rarely been used in the developed world and
many developing nations since the 1980s due to an increased
demand for specialist clotting factors. However it still has its
place in many developing countries and the military theatre.
In theory whole blood has a shelf life of about 35 days; blood
can confer a lot of benefits if transfused fresh (under 5 days).
Fresh blood is rich in clotting factors, has good oxygen-carrying
capabilities and acts as an efficient volume expander; however it
is impractical to store and can rarely be used within this time
frame. With time levels of 2, 3 DPG, factor V and VII reduce and
platelets stop functioning after 3 days.
Whole blood is collected from volunteers who undergo a health
and travel questionnaire. The collection bags contain citrate (which
acts as an anticoagulant by binding calcium and phosphate),
dextrose and adenine to support cell metabolism (CPDA).12
Packed red cells
The collected blood (suspended in CPDA) is leucodepleted by
passing it through a negatively charged filter after which the
ANAESTHESIA AND INTENSIVE CARE MEDICINE 14:6
plasma is removed. In the UK, the remaining red blood cells are
re-suspended in a saline, adenine, glucose and mannitol (SAGM)
solution. Each bag has a shelf life of 35 days.
Packed red cells (PRC) sometimes undergo further screening
to prevent the transmission of cytomegalovirus (CMV) to at-risk
groups, namely allogeneic stem cell transplant patients and the
CMV-negative parturient. PRC can also undergo gamma irradia-
tion to ensure complete leucodepletion, particularly important
for patients suffering with an immunodeficiency syndrome or
haematological malignancy.
PRC are stored between 2 and 6 C and, ideally after being
warmed, transfused through a 170e200-mm filter, designed to
remove larger clots and other aggregates.
Platelets
Platelets are derived in two forms. Random donor platelet concen-
trate (RDP) is comprised of units of blood from several donors being
pooled into a single pool. Alternatively a single donor can be used
and the amount required achieved through plateletphoresis.13 This
process isolates the platelets immediately from the donated blood
before returning the remaining products back to the donor.
Platelets are suspended in plasma and are therefore ABO and
Rhesus typed. As the volume of plasma is small the need for ABO
matching in the clinical setting is desirable rather than necessary;
however Rhesus status must be considered especially in rhesus
negative women of childbearing age.
Platelets are stored at 22 C and are continually agitated to
prevent clumping. Shelf life is only 5 days due to the risk of
bacterial growth.
Plasma
Plasma is removed almost immediately from whole blood after
donation. It can be frozen and delivered as fresh frozen plasma
(FFP) or further processed into its constituents:14
 cryoprecipitate
 human albumin solution
 human immunoglobulin
 specific clotting factors.
FFP: although traditionally derived from a single donor, FFP can
also be obtained from pooled plasma after undergoing solvent-
detergent treatment (SDFP). Each 150-ml bag is rich in clotting
factors and stored at 20 to 30 C and can be kept like this for
2 years. It is ABO matched and once thawed needs to be used
within 4 hours. As with all blood products it carries an infection
risk and paediatric donations should be sourced from bovine
spongiform encephalopathy-free areas, namely the USA. The
usual dose is 10e15 ml/kg.
Cryoprecipitate: As FFP thaws the high-molecular-weight mole-
cules precipitate first which are separated to form cryoprecipitate.
This concentrated solution is rich in factor VIII, fibrinogen and von
Willebrand factor. It is stored at 30 C for up to 12 months in
aliquots of 40e50 ml. A normal dose is 10 units but five-pooled
donations can be used. ABO compatibility is desired.
Specific clotting factors: Individual clotting factors can now be
produced through recombinant processes that negate the concern
of viral transmission.
258 Ó 2013 Elsevier Ltd. All rights reserved.
 CLINICAL ANAESTHESIA

Production of blood components and plasma derivatives

Education
Recruitment
Selection
Donation

Test for: HIV;

hepatitis B;
Plateletpheresis
hepatitis C; HTLV;
syphilis; ABO + RhD;
other phenotypes;
red cell antibodies Process into blood components
(CMV, HbS, malaria)
Filter to remove leucocytes

Plasma from
Red cells Pooled Fresh-frozen non-UK source
platelets plasma

Plasma derivatives
e.g. albumin,
immunoglobulin

4°C 22°C –30°C

35 days 35 days 24 months
Confirm compatibility (Pool) (Thaw)

Patient

CMV, cytomegalovirus; Hb, haemoglobin; HTLV, human T-lymphotropic virus.

Figure 1 Reproduced, with permission from The Stationery Office.12

Red cell substitutes to side effect profiles, namely myocardial events, none have been
approved. New-generation products are attempting to address
Red cell substitutes have been in serious development for over
these issues.16
30 years.15 Some products are involved in on-going clinical trials
There remains continued interest in haemoglobin-based oxy-
but none are commercially available in Europe or the USA. The
gen carriers (HBOC). One product showing promise is the poly-
ideal blood substitute should have little antigenicity, be unable to
merized bovine haemoglobin marketed as Hemopure. It is
transmit infection, be readily available, have a long half-life and
currently available in South Africa and undergoing further
be stored at room temperature. Furthermore its oxygen carrying
investigation in Europe and the USA. Hemopure can be stored at
ability should at least match that of human blood.
room temperature, has a prolonged half-life compared to
Success in this field has remained elusive. First-generation
native haemoglobin and demonstrates high oxygen delivery
development focused around the use of perfluorocarbons. Initial
capacity.15
results demonstrated a marked increase in oxygen carrying ca-
In Europe MP4 is in on-going clinical trials. This product uses
pacity independent of body temperature and pH; however stor-
polyethylene glycol (PEG) to prevent toxicity and prolong half-
age and safety issues have seen most trials stopped early.
life. It is based on ‘old’ human donor blood and displays a
Since 1940 stroma-free blood has been under investigation.
very high oxygen affinity. Its chemical structure allows it to
This cell-free haemoglobin is developed from bovine or recom-
dissociate from oxygen only in tissue areas of low oxygen con-
binant sources. Several products reached phase III trials but due
centration delivering oxygen to the tissues with the most need.15

ANAESTHESIA AND INTENSIVE CARE MEDICINE 14:6 259 Ó 2013 Elsevier Ltd. All rights reserved.
 CLINICAL ANAESTHESIA
Alternative trials are trying to further the functionality of
blood substitutes. The addition of enzyme cross-links aim to
reciprocate additional native while blood functions and possibly
act synergistically with chemotherapeutic agents.16
Increased attention is being placed on blood ‘pharming’. This
technique grows human red blood cells from cells extracted from
umbilical cords. The technique can in theory produce universal
donor group blood. Researches envisage the technique being
available at the point-of-need in the next 5e10 years.15,17 A
REFERENCES
1 Powell-Tuck J, Gosling P, Lobo DN, et al. British consensus guidelines
on intravenous fluid therapy for adult surgical patients, GIFTASUP.
http://www.bapen.org.uk/pdfs/bapen_pubs/giftasup.pdf (accessed 28
August 2012).
2 Stewart PA. How to understand acid-base a quantitative acid-base
primer for biology and medicine. New York: Elsevier, 1981.
3 Myburgh John A, Finfer Simon, Bellomo Rinaldo, et al. Hydroxyethyl
starch or saline for fluid resuscitation in intensive care. N Engl J Med
2012; 367: 1901e11.
4 McAlister V, Burns KEA, Znajda T, Church B. Hypertonic saline for peri-
operative fluid management. Cochrane Database Syst Rev 2010;
http://dx.doi.org/10.1002/14651858.CD005576.pub2. Issue 1. Art.
No.: CD005576.
5 Thomas DJB, Alberti KGMM. Hyperglycaemic effects of Hartmann’s
solution during surgery in patients with maturity onset diabetes. Br J
Anaesth 1978; 50: 185e8.
ANAESTHESIA AND INTENSIVE CARE MEDICINE 14:6
View publication stats
6 Wigginton JG, Roppolo LP, Pepe PE. Advances in Resuscitative trauma
care. Minerva Anestesiol 2011; 77: 993e1002.
7 Brunkhorst FM, Engel C, Bloos F, et al. Intensive insulin therapy and
pentastarch resuscitation in severe sepsis. VISEP Trial. N Engl J Med
2008; 358: 125e39.
8 Perner A, Haase N, Guttormsen AB, et al. the 6S Trial Group and the
Scandinavian Critical Care Trials Group. Hydroxyethyl starch 130/0.42
versus Ringer’s acetate in severe sepsis. N Engl J Med July 12, 2012;
367: 124e34. http://dx.doi.org/10.1056/NEJMoa1204242.
9 Cochrane injuries group albumin Reviewers. Why albumin may not
work. Br Med J 1998; 317: 235.
10 The SAFE Study Investigators. A comparison of albumin and saline
for fluid resuscitation in the intensive care Unit. N Engl J Med 2004;
350: 2247e56.
11 Cohen H. Annual SHOT Report 2011. Serious hazards of transfusion.
Steering group chair. www.shotuk.org.
12 McClelland DBL, ed. Handbook of transfusion medicine. 4th edn.
United Kingdom: Blood Services. The Stationery Office, 2007.
13 Stroncek D, Rebulla P. Transfusion medicine 2, platelet transfusions.
Lancet 2007; 370: 427e38.
14 Key N, Negrier C. Transfusion medicine 3, coagulation factor
concentrates: past, present, and future. Lancet 2007; 370:
439e48.
15 Notman N. Artificial blood. Chemistry World, www.chemistryworld.
org; 2010.
16 Grethlein S. Medscape Research. Blood substitutes. http://emedicine.
medscape.com/article/207801-overview#a1.
17 Luban N. Advances in transfusion medicine. 50 Years in Hematology.
American Society of Hematology, 2008.
260 Ó 2013 Elsevier Ltd. All rights reserved.