Platelets, Frozen Plasma, and

Cryoprecipitate: What is the Clinical Evidence

for Their Use in the Neonatal Intensive Care Unit?
Brandon S. Poterjoy, DO,* and Cassandra D. Josephson, MD†

Transfusion of blood components such as platelets, frozen plasma, and cryoprecipitate is

a common practice in the neonatal intensive care unit. Although it is intuitive that these
components would be transfused in the context of bleeding, their use in neonatology has
often been on a prophylactic basis. Due to a lack of consensus guidelines regarding
indications for transfusion, however, the neonatologist is left to his/her opinion as to when
to transfuse. This article seeks to review the available evidence regarding the use of
platelets, frozen plasma, and cryoprecipitate in neonates, as well as the risks associated
with the administration of these products.
Semin Perinatol 33:66-74 © 2009 Elsevier Inc. All rights reserved.

KEYWORDS platelets, fresh frozen plasma, cryoprecipitate, transfusion, neonate

T he transfusion of plasma-containing products, such as

platelets, frozen plasma, and cryoprecipitate, is common
practice in the neonatal intensive care unit (NICU), especially
mostasis revealed that the result of these mostly low levels of
coagulants and anticoagulants was a delicately balanced neo-
natal hemostatic system, with overall reaction and coagula-
among extremely ill neonates. In this setting, neonatologists tion times similar to those of adults. Within the first 6 months
face complex decisions regarding what product(s) and what of life, the infant’s coagulation system matures and all factor
dose of a particular product to administer, when to order levels (except protein C) approach healthy adult values.
these products (ie, for therapy versus prophylaxis), and what To aid clinicians with the transfusion of platelet and
thresholds should dictate the transfusion. plasma products,3-8 numerous guidelines were published in
To add to the complexity of these decisions, the neonatal the past decade. However, the scientific sources for these
coagulation system is strikingly different from that of adults. guidelines were sparse, which led to recommendations that
Andrew and coworkers examined the development of the were mostly based on experience and expert opinion, rather
coagulation system in full-term and premature infants.1,2 than solid evidence. During the same period, the untoward
They found that many individual clotting factor levels in effects of transfusing plasma-containing products were better
healthy full-term and premature infants were approximately recognized and reported, which accentuated the need to op-
50% of those in healthy adults (ie, the vitamin K-dependent timize transfusion therapy in neonates.4,9-13 The purpose of
factors II, VII, IX, X, and contact factors XI and XII), whereas this manuscript was to review the types of platelets, frozen
others were within normal adult ranges (fibrinogen, factor V, plasma, and cryoprecipitate products available to the clini-
factor VIII, factor XIII, and von Willebrand factor). These cian, as well as the basis for current practices (ie, indications,
investigators also found that neonates had similarly low lev- triggers, and dosing of these products) in preterm and term
els of natural anticoagulants, including antithrombin-III, infants. A second goal was to apply a ranking system by
proteins C & S, and heparin cofactor-II. Tests of global he- Siwek and coworkers14 to rate each reference to neonatal
transfusion practices found in the literature in terms of the
level of evidence.
*Division of Neonatal/Perinatal Medicine, Drexel University College of Med-
icine, St. Christopher’s Hospital for Children, Philadelphia, PA.
†Department of Pathology, Emory School of Medicine, Children’s Health-
care of Atlanta, Blood & Tissue Services, Atlanta, GA. Platelets
Address reprint requests to Cassandra D. Josephson, MD, Emory School of
Medicine, Department of Pathology, 1405 Clifton Road NE, Atlanta, GA In the U.S., platelet products are available as platelet concen-
30322. E-mail: [email protected] trates (also known as whole-blood-derived platelets or ran-

66 0146-0005/09/$-see front matter © 2009 Elsevier Inc. All rights reserved.

doi:10.1053/j.semperi.2008.10.004
Platelets, frozen plasma, and cryoprecipitate 67

dom donor platelets) and apheresis platelets (also known as neonates transfused at a platelet count between 30 and 50 ⫻
single-donor platelets). Platelet concentrates are derived 109/L because of clinical instability or a previous IVH. In this
from a single-donor, whole-blood unit, whereas apheresis retrospective review, 51% of neonates with platelet counts
platelets are collected via an apheresis machine that returns ⬍50 ⫻ 109/L were transfused and no major hemorrhage was
the remaining constituents of whole blood to the donor. Five observed, irrespective of whether platelets were transfused or
to eight platelet concentrates (⬃7 ⫻ 1010 platelets per con- withheld. These investigators concluded that a prophylactic
centrate) must be pooled together to equate a single apheresis platelet threshold of ⬍30 ⫻ 109/L most likely represents a
platelet unit dose (containing 3-6 ⫻ 1011 platelets). The pro- safe practice for clinically stable NICU patients, particularly
cess of leukoreduction differentiates the two products as after the first week of life.27
well. The platelet concentrate unit must undergo post-collec- Other studies have revealed the degree of disparity that
tion leukofiltration for it to be leukoreduced, whereas the exists worldwide in the use of platelet transfusions to treat
process of apheresis itself renders the apheresis unit leukore- thrombocytopenic neonates.27-29 Specifically, a recent study
duced. Lastly, red blood cell contamination occurs more of- of transfusion practices in 10 NICUs in the U.S. reported a
ten in platelet concentrates than in apheresis platelets. Thus, 10-fold difference in platelet transfusion usage among VLBW
Rh sensitization in an Rh-negative recipient may occur more infants in the first week of life.30 Perhaps more importantly,
often when using the former than the latter. Both types of this study found no correlation between the platelet transfu-
platelets must be stored at room temperature with constant, sion usage and the patients’ severity of illness or the incidence
gentle agitation, and their shelf life is limited to 5 days.7 of thrombocytopenia in the different NICUs, suggesting that
the observed differences were mostly due to practice variabil-
ity. Further supporting these findings, our group recently
Indications performed a survey among neonatologists in the U.S. and
Thrombocytopenia is a common abnormality in sick neo- Canada on platelet transfusion practices. This survey re-
nates, affecting 20% to 35% of those admitted to NICUs.15,16 vealed significant diversity in platelet transfusion thresholds,
Approximately 75% of these neonates have mild to moderate product preferences, and component preparation. We spec-
thrombocytopenia, defined as a platelet count of 50 to 150 ⫻ ulated that the lack of uniformity in practice is multifactorial,
109/L, and 25% have platelet counts ⬍50 ⫻ 109/L, a level but mostly due to the lack of scientific evidence to inform
thought to significantly increase the bleeding risk. clinical practice.31 In the absence of clinical studies and sci-
The indications for transfusion of platelets can be divided entific data to allow for standardization of practice parame-
into prophylactic and therapeutic. Therapeutic transfusions ters, numerous experts and consensus groups have pub-
are logical in that they are ordered to treat bleeding, irrespec- lished guidelines for platelet administration in neonates
tive of the preexisting level of thrombocytopenia. However, (Table 1).27,32-35 Not surprisingly, the recommendations are
most platelet transfusions administered to neonates are given as diverse as the transfusion practices reported in the litera-
prophylactically, mostly to prevent intraventricular hemor- ture (Level C, expert opinion).
rhage (IVH). In those cases, the threshold to administer plate-
let transfusions and the appropriate dosing remain contro-
versial. IVH is a major morbidity in preterm neonates, with Dosing
an incidence of 25% to 31% among very-low-birth-weight A dose of 5 to 10 mL/kg of a platelet concentrate unit has
(VLBW) infants. A total of 74% of all intraventricular hem- been demonstrated to raise the platelet count in an average
orrhages occur in the first 48 hours of life.17-19 Several studies full-term newborn by 50,000 to 100,000/␮L.26,36 Based on
have linked hemostatic abnormalities in neonates to the de- these observations, most practitioners order 10 mL/kg for
velopment of IVH.20-24 However, although it may seem intu- each platelet transfusion. Despite lack of corroboration from
itive to administer prophylactic platelet transfusions to pre- rigorous testing, the same dosing regimen is implemented for
vent IVH or diminish its extension, such intervention has not apheresis platelets.
been shown to reduce the incidence of IVH in preterm in-
fants.25,26 Specifically, only one randomized controlled trial
(RCT) to date has compared two platelet transfusion triggers Rate the Evidence
(50 versus 150 ⫻ 109/L) in neonates. This trial, which was There is an absence of Level A studies to inform the practice
limited to VLBW infants in the first week of life and excluded of transfusing platelets in neonates (Table 2). The decision to
those with platelet counts ⬍50 ⫻ 109/L,25 revealed no signif- transfuse or not to transfuse platelets to infants with platelet
icant differences in the frequency or severity of IVH between counts below 50,000 remains unclear, especially in the con-
the two groups. Thus, the investigators concluded that non- text of risk of IVH. Based on the worldwide disparity in
bleeding premature infants with platelet counts ⬎50 ⫻ 109/L transfusion triggers, there is a critical need for randomized,
should not receive prophylactic platelet transfusions. Be- multicenter trials to evaluate the safety of a given platelet
cause all neonates with platelet counts ⬍50 ⫻ 109/L were count. Until that occurs, based on the recent survey by our
transfused, this study did not address whether lower platelet group, a platelet count ⬎50,000/␮L seems to be considered
counts could be safely tolerated. In response to this question, “safe” by the majority of U.S. and Canadian neonatologists,
Murray and coworkers evaluated the outcomes of stable ne- and can be used as a “consensus” level, particularly for criti-
onates transfused for platelet counts ⬍30 ⫻ 109/L, and of cally ill infants and for VLBW infants in the first week of life.
68 B.S. Poterjoy and C.D. Josephson

For stable neonates outside of the high risk period for IVH, a

<100 if major organ bleeding

<50 if failure of production
platelet transfusion trigger of 30,000/␮L is used by many

<50 if minor bleeding

functional disorder
Any platelet count if
Active Bleeding
neonatologists and is supported by Murray’s study.27 How-

<50 in all cases

Not addressed

<50 if stable
<100 if DIC
<100 if DIC

<100 if sick
ever, this would be considered Level B evidence.

<50
<100
<100

<100
<50
Frozen Plasma
Plasma consists of many proteins, with albumin being the most
abundant. Other plasma proteins include complement, trans-
port molecules, immunoglobulins (gamma-globulins), and co-
agulation factors, such as fibrinogen, factor XIII, vWF, Factor
VIII, and vitamin K-dependent coagulation factors (II, VII, IX,
<50 if failure of production

<50 if failure of production

<50 for minor procedure

X).7 Plasma products can be produced from whole blood (WB)

<100 for major surgery
Prior to Invasive

or through plasmapheresis, but are primarily produced from

Not addressed

Not addressed
WB. The “time-after-collection to time-of-freezing” deter-
<100 if DIC

<100 if DIC
Procedure

mines whether the product becomes fresh frozen plasma

<50
<100

<50
<50

<50
(FFP), ie, frozen within 6 to 8 hours of collection, or F24
plasma, frozen within the first 24 hours after collection.37
Both FFP and F24 can be stored at ⫺18°C or colder, and both
products are virtually equivalent, albeit slightly lower factor
VIII and factor V levels in F24.7 The term “frozen plasma”
(FP) encompasses both of these products.
Same as preterm

Same as preterm
Non-Bleeding

Not addressed
Not addressed

Indications
<20 if stable
<30 if sick

The practice of plasma transfusions in the NICU is not well

Term
<20

<30

<20
<30
<30

established. Several previous studies evaluated the poten-

tial usefulness of plasma transfusions in neonates to de-
crease mortality, to improve long-term neurodevelopmen-
tal outcomes, to prevent IVH, as cardiovascular support,
to reconstitute blood for extracorporeal bypass/exchange
Stable Preterm

transfusions, and to treat infections and/or support the

Non-Bleeding
Table 1 Summary of Platelet Transfusion Triggers Recommended for Neonates

immune system. Most of these studies, however, did not

ultimately provide scientific support for these indications
<50

<30

<50

<25
<20
<30
<50

<20

<30
<20

(Table 2).
The Northern Neonatal Nursing Initiative (NNNI) Trial38
was designed to test the hypothesis that early volume expan-
sion, and particularly FFP administration, would reduce
morbidity and mortality in low-birth-weight (LBW) infants.
<100 if falling rapidly

In that trial, infants born prematurely (⬍32 weeks gestation)

>100 if unstable
Maintain >50 to
Non-Bleeding
Sick Preterm

were randomized 2 hours after birth to receive one of three

<50 if DIC

therapies: (a) prophylactic FFP (n ⫽ 257), (b) an equal vol-

<100

<50

<50
<100
<50
<100

<30

<50

ume of inert gelatin plasma substitute (n ⫽ 261), or (c) an

infusion of 10% dextrose (control group, n ⫽ 258). Mortality
related to respiratory distress syndrome was similar among
all three groups, as was the number of infants with IVH. In
the 2-year follow-up study, no significant differences be-
tween groups in mortality or severe disability were reported
among survivors.39 The results of this large RCT disproved
Roberts and Murray, 199934

Roberts and Murray, 200880

the findings of an earlier, much smaller study,40 which

Blanchette et al., 199177

Blanchette et al., 199526

showed a lower incidence of IVH among infants transfused

Calhoun et al., 200032

Murray, et al., 200227

Gibson et al., 200433

with FFP (14%, n ⫽ 36) compared with controls (41%, n ⫽

Roseff et al., 200235
Author

37). To further settle this controversy, in 2004 Osborn and

Strauss, 200078

Strauss, 200879

Evans41 performed a meta-analysis of randomized trials of

early volume expansion in neonates using different volume
expanders (including FFP). This meta-analysis concluded
that there were no benefits associated with the early admin-
istration of FFP to preterm neonates (Table 2), not in terms of
Platelets, frozen plasma, and cryoprecipitate 69

Table 2 Summary of Selected Studies of Platelet, FP, and Cryoprecipitate

No.
Study Design Component Intervention Patients Outcome Level

Andrew et al., 199325 SCRCT Platelets Transfuse to Plts >150 k vs. 152 Increased plt count; shortened B
conventional therapy bleeding time; no change in
rate of IVH
Murray et al., 200227 RCR Platelets Plts <30 k vs. history of 53 No difference in rate of new IVH B
IVH ⴙ Plts <50 k
Malan and de V Heese, SCRCT FFP FFP vs. no therapy for 49 Similar rates of survival; no B
198081 polycythemia long-term disability
Gross et al, 198282 SCRCT FFP FFP ⴙ plts vs. ET vs. control 33 No difference in resolution of B
for pts with DIC DIC or survival
Beverley et al., 198540 SCRCT FFP FFP @ admission & 24 hrs of 73 Decreased rate of IVH in FFP B
life vs. control group
Emery et al., 199283 SCRCT FFP FFP vs. 4.5% and 20% 60 No sustained improvement in B
albumin for hypotension blood pressure in any group
NNNI Trial, 199838 MCRCT FFP FFP @ admission & 24 hrs vs. 776 No difference in mortality or A
Gelofusine vs. Control cranial ultrasound
abnormality
Supapannachart et al., SCRCT FFP FFP vs. haemaccel for 26 Similar rates in decrease of B
199984 polycythemia hematocrit
Mou et al., 200485 SCRCT FFP Whole blood vs. reconstituted 200 Reconstituted blood group has B
blood (FFP ⴙ pRBCs) for shorter length of stay and
ECMO circuit priming less peri-operative fluid
overload
Osborn and Evans, MA FFP Early volume expansion No impact on mortality, cerebral A
200441 with FFP palsy, hypotension, IVH
Branson et al., 198345 CR FFP & Cryo Acute episodes of DIC 1 Pt was stabilized B
controlled with FFP & Cryo
Yuen et al., 198644 CR FFP & Cryo FFP & Cryo for DIC due to 1 Pt was stabilized B
congenital protein C
deficiency
Bell et al., 198646 CR FFP & Cryo Transient improvement in pt 1 Pt went to surgery after addition B
with Kasabach-Merritt of transexamic acid
syndrome
Stammers et al., 200043 CR FFP, Cryo, Plts FFP, Cryo, Plts in pts with 1 Pt was stabilized with all 3 B
DIC products
Larsen et al., 198747 CS Plts & Cryo Multiple therapies for 2 Reversal of coagulopathy in all B
consumptive coagulopathy pts
in 6 pts with Kasabach-
Merritt Syndrome
Mba et al., 199549 CS pRBC & Cryo Pts with hemophilia managed 13 B
with cryo & pRBCs
Pomper et al., 200348 CR Cryo Cryo prepared from apheresis 1 Repeated plasma exchange B
of 1 donor to treat bleeding provided cryo with excellent
in child with vWD hemostatic function, even
after storage intervals of >1
year
Tuner et al., 198152 SCT Prothrombin complex concentrate, Component-directed therapy Corrected hemostatic defects B
cryoprecipitate, or platelet based on hemostatic but did not affect mortality
concentrate defects

Abbreviations: ET, exchange transfusion; MCRT, multicenter, randomized controlled trial; SCRT, single-center, randomized controlled trial; SCT, single center trial;
MA, meta-analysis; RCR, retrospective chart review; CR, case report; CS, case series.
Level A evidence: high-quality randomized, controlled trial (RCT) that considers all outcomes and meta-analysis (quantitative systematic review) using compre-
hensive search strategies. Level B evidence: (a) well-designed, nonrandomized trials (a nonquantitative systematic review with appropriate search strategies
and well-substantiated conclusions); (b) lower quality RCTs, clinical cohort studies, and case-controlled studies with nonbiased selection of study participants
and consistent findings; and (c) other evidence such as high-quality, historical, uncontrolled studies, or well-designed epidemiologic studies with compelling
findings. Level C evidence is consensus or expert opinion.

improving blood pressure, decreasing rates or severity of

Rate the Evidence
IVH, decreasing mortality, or improving neurodevelopmen-
tal outcomes. With the exceptions of the Osborn and Evans meta-analysis41
and the NNNI trial,38 all FP studies were rated as Level B
evidence. Interestingly, a small survey in 2001 on clinical
Dosing practices in NICUs in Ireland revealed that nearly half of the
Each milliliter (mL) of undiluted plasma contains 1 interna- surveyed practitioners used FP in LBW infants to treat hypo-
tional unit (IU) of each coagulation factor. Thus, a dose of 10 tension, despite the high level of evidence to the contrary.42
to 20 mL/kg of FP is expected to sufficiently correct a factor- At this point, the studies addressing potential indications for
deficient patient by approximately 30%.7,8 FP only support its use to reconstitute blood for ECMO cir-
70 B.S. Poterjoy and C.D. Josephson

cuit priming,82 and to treat coagulopathy associated with the due to low plasminogen concentrations.51 One clinical trial
disease processes listed in Table 3. noted correction of hematologic abnormalities in high-risk
neonates with laboratory evidence of coagulopathy with the
use of FP, cryoprecipitate, and/or platelets, yet had no impact
Cryoprecipitate on overall mortality in either preterm or term infants.52
Cryoprecipitate is an insoluble precipitate that forms from the
thawing of FFP and is refrozen in plasma within 1 hour. Among Dosing
all the plasma-based products, cryoprecipitate contains the Cryoprecipitate (a single unit has 15-20 mL) is prepared in
highest concentrations of Factor VIII, von Willebrand Factor the blood bank by thawing and pooling several individual
(vWF), fibrinogen, factor XIII, and fibronectin. Cryoprecipitate units together and then issuing the product. It is generally
is stored at temperatures ⱕ⫺18°C for up to 1 year. As with all known that, to increase the fibrinogen level by 60 to 100
FP products, sufficient time for component preparation must be mg/dL in children and adults, fibrinogen replacement is
allowed to thaw and relabel the product.7 dosed at 1 U/10 kg. Yet in neonates, 1 unit may increase the
fibrinogen level by more than 100 mg/dL. Several publica-
Indications tions have recommended dosing regimens in neonates that
Cryoprecipitate is indicated as a second-line therapy for von range from 2 mL/kg to 1 U/7 kg.53,54 The half-life of fibrino-
Willebrand’s disease (vWD) and factor VIII deficiency, when gen (3-5 days), along with the underlying cause for the defi-
specific coagulation factors or recombinant factors are un- ciency (hemorrhage/DIC versus congenital hypofibrinogen-
available. Other indications include factor XIII deficiency emia), will determine the dosing frequency. Thus, the
and control of bleeding in patients with congenital fibrinogen administration of cryoprecipitate may vary from every 8 to 12
deficiency or dysfunction.5,8 Twelve manuscripts consisting hours for up to every several days.
of case series or case reports were found describing the use of
cryoprecipitate in newborn infants. Cryoprecipitate has been Rate the Evidence
used in cardiac disease complicated by sepsis,43 congenital Most of the evidence in the literature regarding the use of
protein C deficiency,44,45 Kasabach-Merritt Syndrome,46,47 cryoprecipitate in neonates is Level B (Table 2). As standard
Type 3 vWD,48 and in infants with hemophilia A/B.49 Two practice, cryoprecipitate has been used to treat congenital
publications reported no benefit in a preterm infant with a hypofibrinogenemia and factor XIII deficiency. In addition,
ruptured umbilical artery,50 and in an 11-day-old with aortic the use of cryoprecipitate has been extrapolated from the
occlusion unresponsive to fibrinolytic therapy, presumably adult literature to treat neonates with acquired hypofibrino-
genemia during DIC or liver failure. This is now considered
Table 3 Uses of FFP/F24 Plasma6 standard therapy, despite the lack of evidence specifically in
the neonatal population. Similarly, patients with vWD or
Indicated
Multiple acquired coagulation factor deficiency
FVIII deficiency who are bleeding warrant a transfusion of
Liver disease cryoprecipitate when a vWF-containing concentrate or a re-
Massive transfusion combinant FVIII product is unavailable. These recommenda-
Disseminated intravascular coagulation tions are Level C evidence, because they are based on expert
Rapid reversal of warfarin effect opinion rather than scientific evidence.
Plasma infusion or exchange for thrombotic
thrombocytic purpura, hemolytic uremic syndrome
Congenital coagulation defects* Transfusion-Related
C1-esterase inhibitor deficiency*—acute episodes and Morbidity and Mortality
prophylaxis of angioedema
Reconstitution of packed red blood cells Several studies have consistently reported an association be-
Investigational tween platelet transfusions and increased mortality in neo-
Meningococcal sepsis nates. Del Vecchio and coworkers28 evaluated 114 neonates
Acute renal failure in the context of multiorgan failure who received platelet transfusions in a NICU, and found that
Not Indicated those who received a single transfusion during their hospital
Immunodeficiency stay had a relative risk of death 10.4 times greater than those
Infection
who received no platelets. Furthermore, the risk of death of
Prevention of intraventricular hemorrhage in preterm
infants who received more than 4 platelet transfusions was
neonate
Improvement of preterm neonatal neurodevelopmental 29.9 times greater than that of nontransfused neonates. A
outcomes subsequent study by Garcia and coworkers29 also found a
Burns significantly higher mortality among neonates who received
Wound healing platelet transfusions compared with those who did not
Volume expansion (hypotension) (24.5% versus 3.7%). More recently, Baer and coworkers55
Source of nutrients reported a near-linear correlation between the number of
*Fresh-frozen plasma is used in the absence of specific factor con- platelet transfusions and the mortality among 1600 throm-
centrate. bocytopenic NICU patients. Specifically, the mortality rate
74 B.S. Poterjoy and C.D. Josephson

67. Gajic O, Gropper MA, Hubmayr RD: Pulmonary edema after transfu- 79. Strauss RG: How I transfuse red blood cells and platelets to infants with
sion: how to differentiate transfusion-associated circulatory overload the anemia and thrombocytopenia of prematurity. Transfusion 48:209-
from transfusion-related acute lung injury. Crit Care Med 34: 217, 2008
S109-S113, 2006 (suppl) 80. Roberts I, Murray NA: Neonatal thrombocytopenia. Semin Fetal Neo-
68. Skeate RC, Eastlund T: Distinguishing between transfusion related natal Med 13:256-264, 2008
acute lung injury and transfusion associated circulatory overload. Curr 81. Malan AF, de V Heese H. The management of polycythaemia in the
Opin Hematol 14:682-687, 2007 newborn infant. Early Hum Dev 4:393-403, 1980
69. Shorr AF, Jackson WL, Kelly KM, et al: Transfusion practice and 82. Gross SJ, Filston HC, Anderson JC: Controlled study of treatment for
blood stream infections in critically ill patients. Chest 127:1722- disseminated intravascular coagulation in the neonate. J Pediatr 100:
1728, 2005 445-448, 1982
70. Elward AM, Fraser VJ: Risk factors for nosocomial primary bloodstream 83. Emery EF, Greenough A, Gamsu HR: Randomised controlled trial of
infection in pediatric intensive care unit patients: a 2-year prospective colloid infusions in hypotensive preterm infants. Arch Dis Child
cohort study. Infect Control Hosp Epidemiol 27:553-560, 2006 7:1185-1188, 1992
71. Richards C, Alonso-Echanove J, Caicedo Y, et al: Klebsiella pneumoniae 84. Supapannachart S, Siripoonya P, Boonwattanasoontorn W, et al: Neo-
bloodstream infections among neonates in a high-risk nursery in Cali, natal polycythemia: effects of partial exchange transfusion using fresh
Colombia. Infect Control Hosp Epidemiol 25:221-225, 2004 frozen plasma, Haemaccel and normal saline. J Med Assoc Thai 82:S82-
72. Pessoa-Silva CL, Miyasaki CH, de Almeida MF, et al: Neonatal late- S86, 1999 (suppl 1)
onset bloodstream infection: attributable mortality, excess of length of 85. Mou SS, Giroir BP, Molitor-Kirsch EA, et al: Fresh whole blood versus
stay and risk factors. Eur J Epidemiol 17:715-720, 2001 reconstituted blood for pump priming in heart surgery in infants.
73. Norfolk DR, Glaser A, Kinsey S: American fresh frozen plasma for N Engl J Med 351:1635-1644, 2004
neonates and children. Arch Dis Child 90:89-91, 2005 86. Adler SP, Chandrika T, Lawrence L, et al: Cytomegalovirus infections in
74. Centers for Disease Control and Prevention (CDC): Blood donor neonates acquired by blood transfusions. Pediatr Infect Dis 2:114-118,
screening for chagas disease–United States, 2006-2007. MMWR Morb 1983
Mortal Wkly Rep 56:141-143, 2007 87. Gilbert GL, Hayes K, Hudson IL, et al: Prevention of transfusion-ac-
75. Diaz JH: Chagas disease in the United States: a cause for concern in quired cytomegalovirus infection in infants by blood filtration to re-
Louisiana? J La State Med Soc 159:21-23, 25-29, 2007 move leucocytes. Neonatal Cytomegalovirus Infection Study Group.
76. Leiby DA, Herron RM Jr, Read EJ, et al: Trypanosoma cruzi in Los Lancet 1:1228-1231, 1989
Angeles and Miami blood donors: impact of evolving donor demo- 88. Brady MT, Milam JD, Anderson DC, et al: Use of deglycerolized red
graphics on seroprevalence and implications for transfusion transmis- blood cells to prevent posttransfusion infection with cytomegalovirus
sion. Transfusion 42:549-555, 2002 in neonates. J Infect Dis 150:334-339, 1984
77. Blanchette VS, Hume HA, Levy GJ, et al: Guidelines for auditing pediatric 89. Bowden RA, Slichter SJ, Sayers M, et al: A comparison of filtered leu-
blood transfusion practices. Am J Dis Child 145:787-796, 1991 kocyte-reduced and cytomegalovirus (CMV) seronegative blood prod-
78. Strauss R: Blood Banking and Transfusion Issues in Perinatal Medicine. ucts for the prevention of transfusion-associated CMV infection after
Philadelphia, PA, W.B. Saunders, 2000 marrow transplant. Blood 86:3598-3603, 1995