Archives of Disease in Childhood 1995; 73: F181-F183 F181
Autologous umbilical cord blood transfusion
A Ballin, E Arbel, G Kenet, M Berar, D Kohelet, A Tanay, H Zakut, D Meytes
Abstract been published elsewhere.3 The salient
The purpose of this study was to examine features of the preliminary work done before the
some aspects of umbilical cord blood autologous umbilical cord blood transfusion
collection for autologous transfusion in (AUCBT) was performed are summarised
premature infants. All 120 microbacterial briefly below.
cultures (aerobic and anaerobic) of cord The collection of umbilical cord blood was
blood samples as well as 30 cultures of initiated immediately following delivery. After
mycoplasma were treated. Cord pro- clamping the cord a distal portion of the cord
thrombin fragment (F 1+2) concentra- was cleansed with betadine and alcohol 70%
tions were quantified at one and 10 for five seconds (the same procedure was per-
minutes after clamping of the cord. F 1+2 formed following caesarean sections). A 21
concentrations assessed on 25 newborn gauge needle of a quadruple blood collection
infants were similar and no linear associa- set (Travenol Co) was introduced into the
tion with time of clamping could be umbilical vein. The collecting bag contained
drawn. This means that cord blood CPDA-1 in the recommended ratio of about
thrombosis is not activated for at least 10 1:7 with blood. Cord blood filled the bag by
minutes following clamping of the cord. gravity and by uterine contractions.
As far as is known, the first newborn Sterility was tested by culturing the blood in
infant to benefit from this method of a pair of Bactec NR 730 bottles (aerobic and
transfusion is reported here. The prema- anaerobic), incubating it for seven days at
ture infant received two portions of autol- 35°C, and checking it daily.4 Other samples
ogous blood (on days 5 and 7). No were cultured for genital mycoplasma
untoward effects were noted. (Mycoplasma homonis and Ureaplasma
Blood, collected from the umbilical urealyticum). To investigate the coagulation
cord, is a safe source for autotransfusion, profile of the frozen plasma, fresh cord plasma
provided that bacteriological testing has and plasma stored at -29°C for 45 days were
been carried out. tested. Prothrombin time (PT), partial throm-
(Arch Dis Child 1995; 73: F181-F183) boplastin time (PTT), thrombin time (TT),
fibrinogen, and factors V, VII, VIII, IX, X, XI,
Keywords: anaemia of prematurity, cord blood, XII, and XIII were determined (reagents were
autologous transfusion. from Sigma Chemical Co, USA). All tests
were performed using the automated coagula-
tion timer (Electra 800 USA) and Dade
Studies have shown that very low birthweight (Switzerland). The coagulation screening tests
infants receive about 50 ml of blood per kg (PT, PTT, and TT) were performed using
from as many as eight to 10 adult donors routine microtechniques.5
during the first month of life.1 The use of adult Factor assays were based on mixing known
donor blood poses risks for these infants, human deficient plasma with 10 Rl to 30 ,ul of
including transmission of viral agents such as each infant's cord plasma. Von Willebrand
hepatitis B, cytomegalovirus, and HIV, as well factor and plasminogen activities were also
The Paediatric as graft versus host disease and other compli- determined.6 To verify that thrombin activity
Haematology Unit cations.' 2 Placental vessels contain a quarter had not been generated following clamping,
to a third of the newborn blood volume - blood concentrations of prothrombin fragments (F
The Departments of which is currently discarded. Following pre- 1+2) were determined as follows: at one and
Gynecology liminary studies on collection and storage of 10 minutes after clamping a 21 gauge needle
M Berar cord blood,1 3 we have further expanded the was inserted into the umbilical vein and a 2 ml
H Zakut experiments on sterility of the collected blood blood sample collected. The syringe contained
and its coagulation system. 0-4 ml anticoagulant solution consisting of 38
A Tanay To examine the possibility of thrombin mmol/l citric acid, 75 mmolIl sodium citrate,
activation caused by clamping of the cord, we 136 mmolIl dextrose, 6 mmol/l EDTA, 6
Newborn Infants quantified the concentration of prothrombin mmol/l adenosine and 25 ,u/ml heparin.7 The
E Arbel fragments (F 1+2) one and 10 minutes after
D Kohelet ratio of anticoagulant to blood was 0 2: 10
clamping the cord. We report the first prema- (vol/vol). The blood was centrifuged and the
Haematology ture infant who benefited from cord blood plasma frozen at - 20°C until batch analysis of
D Meytes transfusion following storage of the blood. the specimens. Concentrations of prothrombin
Edith Wolfson Medical F 1 + 2 were determined with the ELISA
Center, Holon, Israel Thrombonostica F 1+2 (Organon, Teknika
Correspondence to: Methods Corporation) using a spectrophotometer
Dr Ami Ballin, The Institute All the studies were carried out in the Edith (Molecular devices Co) at 450 nm.
of Haematology, Edith
Wolfson Medical Center, Wolfson Medical Center, Holon, Israel, with Comparison of the results of the coagulation
Holon 58100, Israel. the approval of the institutional committee of tests and of the cord blood and adult F 1+2 con-
Accepted 8 August 1995 ethics. Some of the preliminary results have centrations was determined using Student's t test.
F182 Ballin, Arbel, Kenet, Berar, Kohelet, Tanay, Zakut, Meytes
Results a 1.6
The following results are based on 120 CN4
collections: the mean volume of the cord blood
obtained was 86 ml (range 62-105) from full en 1.2
term and 44 ml (range 31-58) from premature CD
infants (31-34 weeks' gestation). 0.8
All microbacterial cultures (aerobic and m 0.6
anaerobic) were negative, as were all 30
cultures examined for the presence of : 0.4
All the blood coagulation results, performed 0
on 10 samples either before or after storage X 1 2 3 4 5 6 7 8 9 10 15
were within normal limits for this age group Minutes after clamping
(data not shown). There were no significant Figure 2 F 1+2 concentrations at one and 10 minutes
differences in all the paired variables tested. after clamping.
The mean (SD) concentrations of Von
Willebrand factor (n=26) and plasminogen mechanically ventilated. On day 5, the haemat-
(n=26) were 131-8 (32-9)% and 57 (10-7)%, ocrit dropped to 35% and the haemoglobin to
respectively - that is in the normal range for 1 7 g/l. The collected umbilical blood was com-
this age group. patible with the mother's serum, and the cul-
Thirty three samples of cord blood were tures were negative. Following approval by
examined for F 1+2 concentrations. The the committee of ethics and the mother's
samples were obtained from 25 cords of full written consent having been obtained, 10 ml of
term infants (14 boys and 11 girls) whose birth- the packed cells were transfused into the infant.
weights ranged from 3030 to 3970 g. Seventeen No untoward effects were observed. Blood gas
samples were collected at varying times after analysis, as well as concentrations of sodium,
clamping while another eight sample pairs were potassium, calcium and glucose were all within
withdrawn at one and 10 minutes after clamp- normal limits. The packed cell volume (PCV)
ing. The results of F 1+2 determinations are increased to 42%. Two days later, the PCV
shown in fig 1. The mean (SD) cord blood F dropped to 37%, so an additional 10 ml of
1 +2 concentration was 0 53 (0 35) nM. The F packed cells were transfused. Again, vital signs,
1+2 concentrations at one and 10 minutes after blood gas analygis, and all biochemical tests
clamping were statistically similar, and no linear were normal. Two days later, the patient was
association with time of clamping could be extubated. At the time of writing, the infant was
drawn (r=0.29) (fig 2). All these results encour- 3 months old and healthy.
aged us to apply AUCBT in the infant reported
Premature infants are frequently subjected to
CASE REPORT blood transfusions. Most premature infants
The third in a set of triplets born by caesarean born at less than 32 weeks of gestational age will
section at 31 weeks' gestation, the baby receive more than two transfusions after 2 weeks
weighed 1250 g. Following delivery, the pla- of age.' Serious medical complications may
centa was placed on a sterile surface and cord accompany the transfusions. They include
blood was collected from the umbilical vein, as infections such as AIDS,8 cytomegalovirus,9
described. Thirty five millilitres of venous and hepatitis'0; sensitisation to plasma, red cell,
blood were collected within two minutes. A or HIA antigensI 1; errors in blood group or
sample of the blood was sent for bacteriological patient identification and graft-versus-host
testing; the remainder was stored in the blood disease.12 Placental vessels contain 75 to 125 ml
bank in three separate bags. of blood at birth or nearly one quarter to one
Subsequently, the infant, whose haematocrit third of the fetal blood volume.'3 This study
and haemoglobin concentrations at delivery had shows that the fetal blood left in the
been 46% and 153 g/l, respectively, placental vessels may serve as a source of
developed hyaline membrane disease and was blood for autotransfusion. Horn et al found that
cord blood cells can be stored in a CPDA-1
2 2.5 medium for at least 35 days. 14 Bifano et al exam-
ined the feasibility of collecting and storing
+ placental blood.' The authors concluded that
placental blood can be used for autologous
transfusion for the sick neonate. Apart from
E 15 being 'self donation, thus avoiding the above
mentioned homologous transfusion associated
C 1-0 . complications, AUCBT has other advantages:
(a) immediate availability; (b) high levels of
E 0-5* .U haematopoietic progenitor cells, such as colony
forming units - granulocyte macrophage
& 1 2 345 67 89 10 15 16 (CFUJ-GM), as well as high levels of granulo-
Minutes after clamping cyte machrophage-colony stimulating factor
Figure 1 F 1+2 concentrations in 33 samples of cord (GM-CSF) and granulocyte-colony stimulating
blood. factor (G-GSF7). It has already been shown that
Autologous umbilical cord blood transfusion F183
cord blood stem and progenitor cells, may 1 Bifano EM, Dracker RA, Lorah K, Palit A. Collection and
28-day storage of human placental blood. Pediatr Res
reconstitute marrow haematopoiesis.15-19 1994; 36: 90-4.
Umbilical cord blood, enriched with self- 2 Camielli V, Montini G, Da Riol R, Dall'Amico R,
Cantarruti F. Effect of high dose of human recombinant
haematopoietic growth factors and progenitor erythropoietin on the need for blood transfusions in
cells, may be beneficial to premature infants preterm infants.JPediatr 1992; 121: 98-102.
3 Ballin A, Kenet G, Gutman R, Samra Z, Zakut H,
who frequently suffer from leucopenia and Meytes D. Autologous cord blood transfusion (ACBT).
thrombocytopenia caused by lack of marrow Acta Paediatrica 1994; 83: 700-3.
4 Reller LB, Murray PR, MacLowry JO. Blood cultures II.
reserves. Cumitech 1A. Washington, DC: ASM, 1982.
Our study has shown that umbilical cord 5 Johnson M, Zipursky A. Microtechnology for the study of
the blood coagulation system in newbom infants. Can Jf
blood can be withdrawn without bacterial con- Med Technol 1980; 42: 159.
tamination. The coagulation assays performed 6 Andrew M, Paes B, Milner R, Johnston M, Mitchell L,
Tollefsen DM, et al. Development of the human coagula-
in this study have proved the efficacy of stored tion system in the full-term infant. Blood 1987; 70:
cord plasma as a source of coagulation factors. 165-77.
7 Bauer KA, Brockmans AW, Bertina RM, Conard J,
We have also demonstrated that the clotting Horrelon M-H, Samama MM, et al. Hemostatic enzyme
system is not significantly activated for at least generation in the blood of patients with hereditary protein
C deficiency. Blood 1988; 71: 1418-26.
10 minutes after clamping. 8 Nicholas SW, Sondheimer DL, Willoughby AD, Yaffe SJ,
The premature infant described here is the Katz SL. Human immunodeficiency virus infection in
childhood, adolescence and pregnancy: A status report
first (as far as is known) reported newborn to and national research agenda. Pediatrics 1989; 83:
benefit from AUCBT following storage of the 293-308.
9 Kin HC. Red blood cell transfusion in the neonate. Semin
blood. The collected umbilical cord blood Perinatol 1983; 7: 159-74.
served for autologous transfusion twice. No side 10 Esteban JL, Gonzalez A, Hemandez JM, Viladomiu L,
Sanchez C, Lopez-Talavera JC, et al. Evaluation
effects were seen. The idea of using placental of antibodies to hepatitis C virus in a study of trans-
blood for transfusion was initially proposed by fusion-associated hepatitis. N Engl J Med 1990; 323:
Halbrecht.20 He reported 220 cord blood trans- 11 Class FMJ, Lagaay EL, Van Rood JJ. Immunological conse-
fusions (obtained from 520 placentas) given to quences of blood transfusion. Schweiz Med Wochenschr
1991; 121 (suppl 43): 70.
anaemic children and adults. In recent years 12 Anderson KC, Weinstein HJ. Transfusion-associated-
there has been renewed interest in the use of graft-versus-host disease. N Engl J Med 1990; 323:
cord blood as a source of stem and progenitor 13 Oski FA. Normal blood values in the newborn period. In:
cells for marrow reconstitution.9 Oski FA, Naiman JL, eds. Hematological problems in the
newborn. Philadelphia: WB Saunders Co, 1988: 8.
Recombinant human erythropoietin (EPO) 14 Horn S, Mazor D, Zmora E, et al. Storage induced changes
administration to premature infants is a rela- in human newborn erythrocytes. Transfusion 1982; 27:
tively new strategy to reduce the risks of blood 15 Laver J, Duncan E, Abboud M, Gasparetto C, Sahdev I,
transfusion in nurseries.21 This mode of treat- Warren D, et al. High levels of granulocyte and granulo-
cyte-macrophage-colony-stimulating factors in cord blood
ment has some disadvantages and side effects. of normal full-term neonates. J Pediatr 1990; 116:
Treatment with EPO, given subcutaneously, 627-32.
16 Broxmeyer HE, Douglas GW, Hanjoc G, Cooper S, Bard J,
may increase the risk of infections, it depletes English D, et al. Human umbilical cord blood as a poten-
iron stores, the infants gain less weight than tial course of transplantable hematopoietic stem/progeni-
tor cells. Proc NatlAcad Sci USA 1989; 86: 3828-32.
their controls and it is expensive. 17 Gluckman E, Broxmeyer HE, Auerbach AD, Friedman HS,
We believe that blood, collected from the Douglas GW, Devergie A, et al. Hematopoietic reconsti-
tution in a patient with Fanconi's anemia by means of
umbilical cord, is a safe source for auto- umbilical cord blood from an HLA-identical sibling. N
transfusion, provided that bacteriological test- EnglJ3Med 1989; 321: 1174-8.
18 Broxmeyer HE, Kurtzberg J, Gluckman E, Auerbach AD,
ing has been carried out. Furthermore, Douglas GW, Cooper S, et al. Umbilical cord blood
any autologous transfusion programme should hematopoietic stem and repopulating cells in human
clinical transplantation. Blood Cells 1991; 17: 313.
incorporate secure methods and safeguards in 19 Mayanai H, Landsdorp PM. Thy-i expression is linked to
identification for collecting, storing, and dis- functional properties of primitive hematopoietic progeni-
tor cells from human umbilical cord blood. Blood 1994;
pensing the harvested blood. More clinical 83: 2410-7.
experience is warranted to consolidate all these 20 Halbrecht J. Fresh and stored placental blood. Lancet 1939;
concepts. 21 Maler RF, Obladen M, Scigalla P, Lindercamp 0, Duc G,
Hieronimi G, et al. The effect of Epoetin beta (recombi-
We thank Dr S Bar-Shany and Dr E Shinar from the Israeli nant human erythropoietin) on the need for transfusion in
Blood Bank for their professional assistance and Mrs D Ben- very low-birth-weight infants. N Engl J Med 1994; 330:
Porath for her cooperation. 1173-8.