The Journal of International Medical Research

2008; 36: 171 – 177

A Clinical Pilot Study of Fresh Frozen

Plasma versus Human Albumin in
Paediatric Craniofacial Repair
T KERNER1, A MACHOTTA5, S KERNER2, O AHLERS1, H HABERL3, H RIESS4 AND
B HILDEBRANDT4
1
Clinic for Anaesthesiology and Intensive Care Medicine (Charité-Centrum 07 for
Anaesthesiology, OP-Management and Intensive Care Medicine), 2Clinic for Paediatric
Surgery (Charité-Centrum 17 for Women’s, Children’s and Youth Medicine), 3Department of
Paediatric Neurosurgery (Charité-Centrum 15 for Neurology, Neurosurgery and Psychiatry),
and 4Medical Clinic for Haematology and Oncology (Charité-Centrum 14 for Tumour
Medicine), Campus Virchow-Klinikum, Charité-University Hospital, Berlin, Germany;
5Department of Anaesthesiology, Sophia Children’ s Hospital, Erasmus MC, University

Medical Centre, Rotterdam, The Netherlands

Paediatric craniofacial surgery (pCFS)
regularly requires transfusion of packed
red blood cells (pRBC). In this clinical pilot
study two different transfusion regimens
were prospectively compared concerning
pRBC transfusions, postoperative bleeding
and other clinical parameters. Thirty
infants (aged < 12 months) scheduled for
pCFS were assigned to receive fresh frozen
plasma (FFP-group, n = 15) or 5% human
albumin (HA-group, n = 15) during the
entire surgical procedure. Perioperative
amounts of pRBC, postoperative bleeding,
major complications, duration of stay in
the intensive care unit and overall
hospital stay were compared. Differences
in pRBC transfusions, postoperative
bleeding, and duration of intensive care
unit stay were not significant and no
major complications occurred in either
group. A significantly shorter overall
hospital stay was observed in favour of
the FFP-group. Volume replacement
during pCFS can be safely performed
with both applied protocols. Our data do
not demonstrate a major advantage for
FFP use, but further evaluation is
necessary.

KEY WORDS: CRANIOSYNOSTOSIS; PAEDIATRIC ANAESTHESIA; VOLUME REPLACEMENT; TRANSFUSION

MANAGEMENT; FRESH FROZEN PLASMA; HUMAN ALBUMIN; PACKED RED BLOOD CELLS

Introduction operative repair of craniosynostosis in

Primary non-syndromic craniosynostosis is
usually diagnosed in infancy and early
childhood, and is a relatively common
disorder that occurs in 1:2000 births. It
affects the child’s morphology and may lead
to functional impairments.1,2 Thus, primary
infants and young children is generally
recommended.
Paediatric craniofacial surgery (pCFS) is
associated with high perioperative blood loss
of 20 – 500% of patients´ estimated blood
volume, with average values in the range

171
 T Kerner, A Machotta, S Kerner, et al.
Fresh frozen plasma versus human albumin in craniofacial surgery
of 60 – 100%. Today, pCFS is considered to be
safe in experienced hands but, historically,
the occurrence of massive haemorrhage and
a perioperative mortality up to 0.5% have
been reported.3 One major problem in pCFS
is that allogenic transfusions of packed red
blood cells (pRBC) are required in virtually
all children treated.4,5 In addition, excess
volume replacement during major surgery
may seriously alter haemostasis by causing a
dilution of clotting factors and, thus, further
increasing the need for transfusions of
pRBC.6 – 8 One strategy to decrease blood loss
during major surgery is the primary
substitution of clotting factors by the
administration of fresh-frozen plasma (FFP)
instead of sole colloid volume replacement.
As yet, a general benefit of (prophylactic) FFP
application has not been clearly proven, but
the approach has been evaluated in certain
types of paediatric cardiac surgery.9 However,
no prospective data on the primary
substitution of FFPs in the conduct of pCFS
are available yet.
We report here the results of a prospective
clinical pilot study in infants undergoing
elective surgical repair of craniosynostosis.
The study was designed to compare the
consequences of primary FFP application
with those of sole colloid volume
replacement by the administration of
human albumin (HA). Assessments were
made with regards to the amount of pRBC
transfusion, perioperative blood loss and
other clinical endpoints.
Patients and methods
STUDY DESIGN
This pilot study was designed to compare
different strategies of intraoperative volume
replacement in infants undergoing
paediatric craniofacial repair on an
exploratory level. Patients for which FFP was
available from the same donor as for the
172
pRBC were scheduled to receive primary
administration of FFP (FFP-group). In all
other patients, 5% human albumin was
administered as a colloid volume
replacement instead (HA-group). Groups
were compared with regards to the amounts
of intra- and postoperative transfusions of
pRBC, postoperative blood losses, major
complications, durations of stay in the
intensive care unit and overall in the
hospital. The aim of the study was to provide
an informational basis for the hypothesis
that prophylactic administration of FFP is
safe and may be suitable to decrease the
amount of pRBC transfusion in infants
undergoing pCFS. A sample size of n = 30 (15
patients/group) was chosen, because this
magnitude is generally accepted to detect
major differences with regard to predefined
endpoints on an ordinal data level.10
As data available so far on the use of FFP
in pCFS do not justify a possible assignment
of infants to multiple-donor transfusions, no
randomization was performed. Instead a
‘biological randomization’ was used that
depended on the availability of FFP from the
same donor as pRBCs.
The study was approved by the local
ethics committee and was performed in
accordance with the ethical standards as
described in the Declaration of Helsinki. A
detailed written, informed parental consent
was obtained in every case at least 24 h
before surgery.
INCLUSION AND EXCLUSION
CRITERIA
Infants with primary, non-syndromic,
frontal, sagittal, coronal or multiple
craniosynostosis who were electively
scheduled for craniofacial repair were
included in the study if they were born
between the 37th and 42nd week of
pregnancy, were aged ≤ 365 days, had a
 T Kerner, A Machotta, S Kerner, et al.
Fresh frozen plasma versus human albumin in craniofacial surgery
recent body weight of ≥ 5 kg and an
American Society of Anesthesiologists (ASA)
physical status of I or II. Exclusion criteria
consisted of the presence of syndromic
synostosis or concomitant diseases, as well as
intake of regular medications and
participation in another clinical trial.
CLINICAL PROCEDURES
The surgical procedure, as described by
Haberl et al.,3 was performed on all infants
by the same team of experienced
neurosurgeons. After monitor adjustment
and mask induction using sevoflurane (3 – 5
vol%) in an air/oxygen mixture, a venous
access was established. Muscle relaxation
with 0.1 mg/kg cisatracurium was performed
and, after tracheal intubation, controlled
mechanical ventilation was initialized. A
continuous infusion of remifentanil at a rate
of 0.1 µg/kg per min was started, and
anaesthesia was maintained with desflurane
(3 – 6 vol%) in an air/oxygen mixture. Two
22 – 20 G peripheral venous cannulas were
inserted and a 22 G arterial cannula was
inserted in the left radial artery of each
infant for invasive blood pressure
monitoring. For central venous pressure
(CVP) measurement, a dual-lumen central
venous catheter was placed into the right
internal jugular vein by ultrasound
guidance. In order to maintain a mean
arterial pressure of ≥ 55 mmHg and a CVP of
5 – 10 mmHg, volume replacement was
given by primary infusion of crystalloid
solutions including 2.5% glucose (10 – 20
ml/kg per h). Noradrenaline was infused if
the mean arterial pressure fell to < 55
mmHg. After crystalloid infusion reached
20% of the estimated blood volume, 5%
human albumin was infused in the HA-
group, and FFP was transfused in the FFP-
group before homologous transfusion of
pRBC was performed. The pRBC were
173
transfused at haemoglobin levels < 4 mmol/l
(6.5 g/dl) in order to preserve the levels
between 5 mmol/l (8 g/dl) and 6.2 mmol/l
(10 g/dl). Platelet transfusions were applied
at a count of < 50 000 platelets/µl. Following
surgery, all patients were extubated and
transferred to a paediatric intensive care
unit. Postoperative intervention criteria (e.g.
volume substitution, transfusions, use of
noradrenaline) were the same as for the
intraoperative period.
EVALUATIONS
Duration of surgery as well as the total
intraoperative quantity of crystalloids,
colloids and pRBC were assessed and
compared between the groups. Postoperative
measurements included blood loss drainage
volumes and the amounts of crystalloids and
pRBC used during the 0 – 12 h and 12 – 48 h
postoperative periods. Fluid quantities were
expressed as ml/kg body weight. Further
assessments over the entire treatment period
included: administration of noradrenaline
and platelet transfusions, signs of systemic
infection (temperature > 38.0 °C or < 36.0 °C,
leucocyte count > 12 000 or < 4000 cells/µl),
complications leading to surgical
interventions, durations of stay in the
intensive care unit and overall in the hospital.
STATISTICAL ANALYSIS
Differences between the groups were
analysed using the χ2 test on a nominal level
and the Mann–Whitney U-test on an ordinal
level. All analyses were performed using the
Statistical Package for Social Sciences (SPSS®
version 12.0; SPSS Inc., Chicago, IL, USA). A
two-sided P-value < 0.05 was considered to be
statistically significant.
Results
Both groups were balanced in terms of the
relevant patients´ characteristics, such as
 T Kerner, A Machotta, S Kerner, et al.
Fresh frozen plasma versus human albumin in craniofacial surgery

TABLE 1:
Characteristics of infants undergoing paediatric craniofacial surgery entered into the
study to compare intraoperative volume replacement using fresh frozen plasma (FFP-
group) with administration of 5% human albumin (HA-group)
Patient characteristic FFP-group (n = 15) HA-group (n = 15)
Age (months) 5.3 (4 – 12) 7.0 (4 – 12)
Sex (female /male) 4 / 11 6/9
Body weight (kg) 7.5 (5.2 – 11.5) 7.8 (6.0 – 10.8)
Height (cm) 66 (59 – 75) 68 (64 – 78)
Type of craniosynostosis
Frontal 4 3
Sagittal 9 8
Coronal 1 2
Multiple 1 2
Data are number or mean (range).
No statistically significant differences between the groups.

TABLE 2:
Intraoperative, postoperative and total treatment period characteristics of infants entered
into the study to compare intraoperative volume replacement using fresh frozen plasma
(FFP-group) with administration of 5% human albumin (HA-group)

Period Characteristics FFP-group HA-group P-value

Intraoperative Duration of surgery (min) 190 ± 12 174 ± 12 NS
Transfusion of pRBC (ml/kg) 23 ± 3 27 ± 5 NS
Amount of HA / FFP (ml/kg) 45 ± 6 35 ± 9 NS
Crystalloids (ml/kg) 52 ± 12 45 ± 7 NS
Postoperative Time in intensive care unit (days) 3.0 ± 0.3 3.4 ± 0.4 NS
Transfusion of pRBC (ml/kg) 8 ± 2 12 ± 3 NS
0 – 12 h after surgery 6 ± 2 7 ± 3 NS
12 – 48 h after surgery 2 ± 1 5 ± 2 NS
Crystalloids (ml/kg) 189 ± 16 183 ± 18 NS
0 – 12 h after surgery 62 ± 3 60 ± 4 NS
12 – 48 h after surgery 127 ± 14 123 ± 16 NS
Blood loss (ml/kg) 30 ± 4 33 ± 4 NS
0 – 12 h after surgery 20 ± 3 18 ± 2 NS
12 – 48 h after surgery 10 ± 2 15 ± 3 NS
Total treatment Overall hospital stay (days) 8.0 ± 0.3 9.9 ± 0.7 0.03
period Transfusion of pRBC (ml/kg)a 31 ± 4 39 ± 7 NS
Platelet transfusion (ml/kg)a 0.0 0.0 NS
Crystalloids (ml/kg)a 241 ± 17 228 ± 18 NS
Administration of noradrenaline 0 0 NS
Operative revision 0 0 NS
Infection 0 1 NS
Data are number or mean ± SE.
pRBCs, packed red blood cells; NS, not significant.
aFrom the start of surgery until 48 h postoperative.

174
 T Kerner, A Machotta, S Kerner, et al.
Fresh frozen plasma versus human albumin in craniofacial surgery
gender, age, weight, height, and type of
craniosynostosis (Table 1). The
intraoperative, postoperative and total
treatment evaluations for both groups are
given in Table 2.
Analyses of the intraoperative period
revealed a trend towards higher volumes of
FFP than human albumin infused in the
respective groups, whereas the amounts of
transfused pRBC and crystalloids were
similar, as was the duration of surgery. The
amount of pRBCs transfused in the
postoperative period, the amount of
postoperative blood loss (particularly in the
period 12 – 48 h after surgery) and the
duration of stay in the intensive care unit
were slightly lower in the FFP-group
compared with the HA-group, but these
differences were far from being statistically
significant. The amounts of postoperative
crystalloids administered also revealed no
statistical differences.
All treatments were performed without
surgical complications leading to re-
intervention. Regarding the entire treatment
period, the amounts of crystalloid infusions
and pRBC transfusions were similar between
the FFP-group and the HA-group. Platelet
transfusions, administration of
noradrenaline or operative revisions were
not required in either group. Signs of
infection were recorded in one infant
assigned to human albumin treatment, but
symptoms disappeared under empirical
antibiotic therapy. The overall time of
hospital stay was significantly shorter in the
FFP-group compared with the HA-group (8.0
days vs 9.9 days, P = 0.03).
Discussion
The aim of this pilot study was to compare
two different transfusion regimens,
prophylactic administration of FFP versus
sole colloid volume replacement using
175
human albumin, in infants undergoing
craniosynostosis repair.
The ‘biological randomization’ used in
this study resulted in groups of 15 patients
per treatment that were balanced for the
most important baseline characteristics.
With regards to the treatment characteristics,
no significant differences were observed
between the groups. Indeed, overall amounts
of transfusions of pRBC, postoperative blood
losses as well as intra- and postoperative
pRBC quantities were quite similar.
Intraoperative estimations of blood losses
were not recorded, due to difficulty in their
quantification in small infants. pRBC
transfusions and bleeding in the late
postoperative period (i.e. 12 – 48 h after
surgery) were slightly lower in the FFP-group
compared with the HA-group, but did not
reach statistical significance.
One interesting point to mention is the
significantly shorter overall hospital stay in
patients treated with FFP. Hospital stay may
be regarded as an imprecise endpoint in the
given context, which is not appropriate for
the objective of this study in the light of
published data. On the other hand, the
difference is approximately 2 days and, thus,
far from negligible. Against the background
of our other results it remains speculative
which other factors than the intraoperative
volume replacement protocol actually
contributed to the favourable result of
patients treated with FFP.
The procedure of pCFS has evolved as
being safe with limited risks. Perioperative
complications, such as severe electrolyte
imbalances, respiratory distress, as well as
epidural abscess and frontal bone necrosis
have been reported, but massive
haemorrhage mainly determines patient
mortality rate from the procedure which,
based on previous literature, is estimated in
the range of 0 – 0.5%.3,5,11 – 14 Skull
 T Kerner, A Machotta, S Kerner, et al.
Fresh frozen plasma versus human albumin in craniofacial surgery

deformation, type of surgical procedure, as
well as the experience of neurosurgeons and
anaesthesiologists have been identified as
the most important risk factors for excess
bleeding in pCFS.15,16 To minimize further
the risk of the procedure, strategies to reduce
intraoperative bleeding (and the need for
allogenic blood transfusions) in pCFS are of
the utmost importance. These include the
implementation of innovative planning
systems, endoscopic techniques, microneedle
electrocautery, or special approaches like the
rigid external distraction system on the one
hand, as well as autotransfusion procedures,
normovolaemic haemodilution, and
perioperative administration of
3,14,17 – 22
erythropoietin on the other hand. In
addition, the occurrence of dilutional
coagulopathy (DC) has been identified as
one reason for massive bleeding in the
context of major haemorrhage. DC
represents a complex disorder that is
characterized by dilution of coagulation
factors. It is most pronounced with the use of
high-molecular weight starch solution, but
also occurs with the use of human albumin
or crystalloids.6 – 8,23,24 Although the
evaluation of coagulation parameters was
not a subject of this clinical study, one has to
keep in mind that the substitution of clotting
factors by FFP application is suggested to
minimize DC and consecutive occurrence of
major bleeding episodes. It thus builds an
important rationale for primary FFP
application, although results of a recent
systematic review have not revealed
persuasive evidence for this approach so
far.25
Regarding the situation in patients
undergoing pCFS, only few data on the
administration of FFPs and its consequences
on perioperative haemorrhage are available.
Williams et al.5 performed a prospective
study in which they performed analyses on
various coagulation parameters during
pCFS. They used a transfusion protocol in
which FFPs were administered on demand,
e.g. if the prothrombin time given by the
international normalized ratio or the
activated partial thromboplastin time
increased to > 1.5 of the upper limit of
normal. A higher amount of postoperative
bleeding was observed in infants who
experienced an intraoperative blood loss of
> 100 ml/kg during pCFS, whereby this
correlation was drawn independent from the
transfusion protocol. The authors
hypothesized that major haemorrhage
during pCFS is at least partially due to a
depletion of soluble coagulation factors.
Overall, in principle it has been suggested
that primary application of FFP is valuable
to decrease perioperative bleeding in major
paediatric surgery. However, the results of
our pilot study, which prospectively
compared two different transfusion regimens
in small infants undergoing paediatric
craniofacial surgery for the first time, did not
demonstrate a major advantage for FFP over
volume replacement with human albumin
in these patients. Both FFP and colloid
volume replacement using human albumin
can be safely applied to infants undergoing
paediatric craniofacial surgery.
Conflicts of interest
No conflicts of interest were declared in
relation to this article.

• Received for publication 13 September 2007 • Accepted subject to revision 18 September 2007
• Revised accepted 21 November 2007
Copyright © 2008 Field House Publishing LLP

176
 T Kerner, A Machotta, S Kerner, et al.
Fresh frozen plasma versus human albumin in craniofacial surgery

References 14 Nishimoto S, Oyama T, Shimizu F, et al: Fronto-

1 Kabbani H, Raghuveer TS: Craniosynostosis.
Am Fam Physician 2004; 69: 2863 – 2870.
2 Panchal J, Uttchin V: Management of
craniosynostosis. Plast Reconstr Surg 2003; 111:
2032 – 2048.
3 Haberl H, Hell B, Zockler MJ, et al: Technical
aspects and results of surgery for
craniosynostosis. Zentralbl Neurochir 2004; 65:
65 – 74.
4 Faberowski LW, Black S, Mickle JP: Blood loss
and transfusion practice in the perioperative
management of craniosynostosis repair. J
Neurosurg Anesthesiol 1999; 11: 167 – 172.
5 Williams GD, Ellenbogen RG, Gruss JS:
Abnormal coagulation during pediatric
craniofacial surgery. Pediatr Neurosurg 2001; 35:
5 – 12.
6 Hobisch-Hagen P: Hemodilution and
coagulation. An overview. Minerva Anestesiol
2002; 68: 178 – 181.
7 Ng KF, Lam CC, Chan LC: In vivo effect of
haemodilution with saline on coagulation: a
randomized controlled trial. Br J Anaesth 2002;
88: 475 – 480.
8 Ruttmann TG: Haemodilution enhances
coagulation. Br J Anaesth 2002; 88: 470 – 472.
9 Casbard AC, Williamson LM, Murphy MF, et al:
The role of prophylactic fresh frozen plasma in
decreasing blood loss and correcting
coagulopathy in cardiac surgery. A systematic
review. Anaesthesia 2004; 59: 550 – 558.
10 Wernecke KD: Operational Practical Statistics.
Bonn/ Paris: Addison-Wesley, 1995 [in
German].
11 Buntain SG, Pabari M: Massive transfusion and
hyperkalaemic cardiac arrest in craniofacial
surgery in a child. Anaesth Intensive Care 1999;
27: 530 – 533.
12 Kang JK, Lee SW, Baik MW, et al: Perioperative
specific management of blood volume loss in
craniosynostosis surgery. Childs Nerv Syst 1998;
14: 297 – 301.
13 Levine JP, Stelnicki E, Weiner HL et al:
Hyponatremia in the postoperative
craniofacial pediatric patient population: a
connection to cerebral salt wasting syndrome
and management of the disorder. Plast Reconstr
Surg 2001; 108: 1501 – 1508.
facial monobloc advancement with rigid
external distraction (RED-II) system. J Craniofac
Surg 2004; 15: 54 – 59.
15 Eaton AC, Marsh JL, Pilgram TK: Transfusion
requirements for craniosynostosis surgery in
infants. Plast Reconstr Surg 1995; 95: 277 – 283.
16 Meyer P, Renier D, Arnaud E, et al: Blood loss
during repair of craniosynostosis. Br J Anaesth
1993; 71: 854 – 857.
17 Di Rocco C, Tamburrini G, Pietrini D: Blood
sparing in craniosynostosis surgery. Semin
Pediatr Neurol 2004; 11: 278 – 287.
18 Fearon JA, Weinthal J: The use of recombinant
erythropoietin in the reduction of blood
transfusion rates in craniosynostosis repair in
infants and children. Plast Reconstr Surg 2002;
109: 2190 – 2196.
19 Jimenez DF, Barone CM: Intraoperative
autologous blood transfusion in the surgical
correction of craniosynostosis. Neurosurgery
1995; 37: 1075 – 1079.
20 Jimenez DF, Barone CM, Cartwright CC, et al:
Early management of craniosynostosis using
endoscopic-assisted strip craniectomies and
cranial orthotic molding therapy. Pediatrics
2002; 110: 97 – 104.
21 Meneghini L, Zadra N, Aneloni V, et al:
Erythropoietin therapy and acute preoperative
normovolaemic haemodilution in infants
undergoing craniosynostosis surgery. Paediatr
Anaesth 2003; 13: 392 – 396.
22 Ririe DG, David LR, Glazier SS, et al: Surgical
advancement influences perioperative care: a
comparison of two surgical techniques for
sagittal craniosynostosis repair. Anesth Analg
2003; 97: 699 – 703.
23 Innerhofer P, Fries D, Klingler A, et al: In vivo
effect of haemodilution with saline on
coagulation. Br J Anaesth 2002; 89: 934 – 936.
24 Ruttmann TG, James MF: Haemodilution and
coagulation: a caveat. Anasthesiol Intensivmed
Notfallmed Schmerzther 2000; 35: 707 – 709.
25 Oliver WC, Jr., Beynen FM, Nuttall GA, et al:
Blood loss in infants and children for open
heart operations: albumin 5% versus fresh-
frozen plasma in the prime. Ann Thorac Surg
2003; 75: 1506 – 1512.

Author’s address for correspondence

Dr T Kerner
Charité Centrum for Anaesthesiology, OP Management and Intensive Care Medicine, Clinic
for Anaesthesiology and Intensive Care Medicine, Campus Virchow-Klinikum and Charité
Campus Mitte, Charité University Hospital Berlin, Augustenburger Platz 1, D-13344 Berlin,
Germany.
E-mail: thoralf.kerner@charite.de

177