Bone Marrow Transplantation (2003) 31, 269–273
& 2003 Nature Publishing Group All rights reserved 0268-3369/03 $25.00
www.nature.com/bmt
Cord blood stem cells
Comparison between two strategies for umbilical cord blood collection
P Solves1, R Moraga2, E Saucedo2, A Perales3, MA Soler1, L Larrea1, V Mirabet1, D Planelles1,
F Carbonell-Uberos1, J Monleón2, T Planells1, M Guillén1, A Andrés1 and E Franco1
1
Valencia Cord Blood Bank, Processing and Cryopreservation Service, Valencia Transfusion Centre, Valencia, Spain; 2La Fe
University Hospital, Valencia, Spain; and 3Dr Peset Hospital, Valencia, Spain
Summary:
The use of cord blood (CB) for transplantation has
increased greatly in recent years. The collection strategy
is the first step in collecting good-quality CB units. There
are two main techniques for collecting CB from the
umbilical vein: in the delivery room while the placenta is
still in the uterus by midwives and obstetricians or in an
adjacent room after placental delivery by CB bank trained
personnel. In this study, the benefits and disadvantages
between the two different CB collection strategies were
evaluated, in order to improve CB bank methodology.
Valencia CB bank maintains the two different collection
strategies. CB was obtained from 569 vaginal and 70
caesarean deliveries and obstetrical and clinical charts
were reviewed. Before processing CB units, volume was
calculated and samples were drawn for cell counts. After
processing and before cryopreservation samples were
drawn for cell counts, CD34+cell analysis, viability,
clonogenic assays and microbiology were drawn directly
from the bags. We compared the efficiency of the two
collection techniques. Obstetric data and umbilical CB
were obtained from 569 vaginal (264 collected in utero and
305 collected ex utero) and 70 caesarean deliveries. The
proportion of excluded CB units before processing was
33% for vaginal ex utero, 25% for vaginal in utero and
46% for caesarean deliveries. Differences were statistically significant. For vaginal deliveries a larger volume
and a higher number of nucleated cells, percentage of
CD34+ cells and colony-forming units (CFUs) were
harvested in the in utero collection group. There was no
statistical difference between CB collected after placental
expulsion from vaginal and caesarean deliveries. Comparison between all vaginal and caesarean deliveries did not
show any difference.
We conclude that the mode of collection influences the
haematopoietic content of CB donations. Collection
before placental delivery is the best approach to CB
collection and allows optimisation of CB bank methodology. Caesarean deliveries seem to contain similar
progenitor content to vaginal deliveries.
Correspondence: Dr P Solves, Valencia Transfusion Centre, Avda del
Cid 65-A, 46014 Valencia, Spain
Received 23 May 2002; accepted 02 October 2002
Bone Marrow Transplantation (2003) 31, 269–273.
doi:10.1038/sj.bmt.1703809
Keywords: cord blood; obstetric factors; cord blood
collection
Cord blood (CB) is recognised as a rich source of
haematopoietic stem and progenitor cells and its use has
increased greatly in recent years. Since the first CB
transplantation was performed in Paris in 1988, knowledge
of the biologic characteristics of umbilical CB has
improved, and the benefits of using CB stem cells have
become apparent.1 In fact, CB has been shown to have
several advantages as compared to bone marrow: easy
availability, lower risk of infectious disease transmission
such as CMV and EBV and lower risk of graft-versus-host
disease. However, the progenitor content of CB units is
usually only sufficient for small recipients, usually children
weighing less than 40 kg. Gluckman et al 2 showed grafting
occurring in 85% of patients receiving 37 million or more
nucleated cells per kilogram of body weight2. In the last few
years, use of CB in larger recipients is increasing and an
effort to improve CB quality is mandatory.
The collection strategy is the first step for collecting
good-quality CB units and varies among banks and among
collection sites for the same CB bank.3 In this sense,
different collection methods have been proposed to
optimise volume and total nucleated cells content of CB
units. There are two main techniques for collecting CB
from the umbilical vein: in the delivery room while the
placenta is still in the uterus by midwives and obstetricians
or in an adjacent room after placental delivery by CB bank
trained personnel. Both methods have some advantages
and some disadvantages. The advantage of the first method
is that the volume of cells collected is usually higher if the
cord is clamped early and collection is started immediately.
However, this can disrupt the normal process of delivery.
After delivery, CB collection is easier and it can be
performed by trained personnel. However, fewer cells
may be collected and there may be an increase in the risk
of bacterial contamination. Feasibility of obstetricianbased collection has been shown by many CB banks.4 On
the other hand, open systems, which were initially used for
CB collection, have been replaced by closed systems with
blood collection bags to minimise the risk of microbiological contamination.5
Strategies for umbilical cord blood collection
P Solves et al
270
In this study, the benefits and disadvantages between two
different CB collection strategies were evaluated, in order
to improve CB bank methodology.
Material and methods
Collection strategies
Valencia CB bank maintains two different collection strategies: before placental delivery by Maternal Hospital staff and
after placental expulsion by CB bank trained personnel.
Before placental delivery (in utero) collection
Donors must sign informed consent before delivery, during
the last months of pregnancy. After delivery, the umbilical
cord is clamped immediately and cleaned with 70% alcohol
and an iodine swab. Umbilical CB is collected from the
umbilical vein by gravity in a closed sterile collection 350 ml
double bag (R 1315, Baxter) containing 23 ml CPDA
anticoagulant. Previously, 26 ml of CPDA has been
removed from the collection bag. This process is performed
by trained midwives or the obstetric staffs. CB collection is
stored at 41C until processing.
After placental delivery (ex utero) collection
After informed consent is obtained from the mother, CB
collection is performed by trained umbilical CB bank
personnel. Once the placenta is delivered, it is transported
quickly to the collection area, which contains necessary
supplies and equipment. There is no time limit set between
placental delivery and the CB collection. The placenta is
placed on a specially designed tray with a central hole that
allows the umbilical cord to hang down. After stringently
cleaning the umbilical cord with 70% alcohol and an iodine
swab, CB is collected from the umbilical vein by gravity in a
sterile collection 350 ml double bag (R 1315, Baxter)
containing 23 ml CPDA anticoagulant. CB collection is
stored at 41C until processing. All caesareans were
processed in this way.
Obstetric data
For both methods, obstetric and clinical maternal chart
was reviewed by CB bank medical staff. Delivery information (route of delivery, gestation length, gestation number,
placental weight, amniotic fluid characteristics and newborn characteristics), sex and weight were reviewed. In all
cases, placental weight was calculated by adding the CB
collection weight to the after collection placental weight.
Red cell depletion and cryopreservation
Within 48 h after collection, CB is processed. More than
5 108 total nucleated cells (TNC) were required to accept
CB units for the bank. Volume reduction is performed by
the method developed by Rubinstein et al.6 Briefly, a 6%
hydroxy-ethyl-starch (HES, Grifols) solution is mixed with
the umbilical CB in the ratio of one volume of HES to five
volumes of blood in the collection bag (1.2% HES final
Bone Marrow Transplantation
concentration). Then, the bag is centrifugated at 45 g for
7 min at 101C (Varifuge 3.ORS, Heraeus Instruments). The
removed supernatant leucocyte-rich plasma and 12 g of red
blood cells are centrifuged again at 600 g for 15 min at
101C. After discarding the supernatant, CB is separated
into two 50 ml cryopreservation bags (Cryocyte R4R9951,
Nexwell) containing 20 ml each. Cryopreservation of CB is
performed by adding 50% dimethylsulphoxide (DMSO)
solution with dextran (Rheomacrodex, Ibyss) for 15 min to
reach a 10% DMSO final concentration. The cryocyte bags
are placed in aluminium cassettes for immediate freezing.
Biological studies
Before processing CB units, volume was calculated and
samples were drawn for cell counts. After processing and
before cryopreservation, samples were drawn for cell
counts, CD34 analysis, viability, HLA typing, clonogenic
assays and microbiology directly from the bags. Cells
counts were made with an autoanalyser Sysmex K800 (Toa
Medical Electronics, Kobe, Japan). TNC was calculated.
CD34+ cells were quantified by flow cytometry. FACScalibur cytometer was used with Procount software (Becton
Dickinson). Cell viability was assessed by ethidium
bromide and acridine orange. Clonogenic assays were
performed using a commercially prepared complete methylcellulose medium (Methocult GF H4434), supporting
growth of CFU-GM, BFU-E and CFU-GEMM. Cultures
were plated at 1 104 viable cells in duplicate 35 mm
diameter Petri dishes and incubated for 14 days at 371C
with 5% CO2 in a humidified atmosphere. Colonies defined
as aggregates of more than 40 cells were counted under an
inverted microscope. Sterility controls were performed
using an automated blood culture system (BacT/Alert:
Organon Teknica) at 351C for 14 days.
Statistical analysis
Groups were compared regarding maternal, neonatal and
CB units parameters. Statistical Package for Social Sciences
(SPSS) v. 10 was used to perform the statistical analysis.
The groups were compared by the w2 test or the Fisher test
for categorical variables, the Kruskal–Wallis and Mann–
Whitney U-test for continuous variables. A Po0.05 was
considered to be significant. The correlations between the
TNC, CD34+ cells and CFUs of CB collected were
analysed by means of Pearson’s correlation coefficient.
Results
Obstetric data and umbilical CB were obtained from 569
vaginal and 70 caesarean deliveries. In Table 1, we
summarise the obstetric data according to collection by
vaginal and caesarean deliveries. There was no statistical
difference between the groups for obstetric variables except
for the infant weight that was greater for group 3 (Po0.05).
The proportion of CB units excluded before processing
was 33% for vaginal ex utero, 25% for vaginal in utero and
46% for caesarean deliveries. Differences were statistically
Strategies for umbilical cord blood collection
P Solves et al
271
Table 1
Obstetric characteristics of deliveries according to the umbilical CB harvested before or after placental
delivery and route of delivery
Route of delivery
Mode of collection
Maternal age
Gestational age (weeks)
Gestation number
Infant weight (g)
Placental weight (g)
Newborn sex
Male
Female
Amniotic fluid
Normal
Meconium stained
Vaginal deliveries
Caesarean
Group 1
In utero (n=264)
Group 2
Ex utero (n=305)
Group 3
Ex utero (n=70)
30.774.69
3971.49
1.971.13
3332.847437.52
649.797118.37
31.6574.85
39.1971.46
1.7671.03
3295.497420.08
705.067123.25
32.0474.79
39.1671.45
1.5671.31
34597497.97
701.777136.76
143
121
148
157
37
33
224
40
274
31
61
9
Continuous variables are expressed as mean and s.d. There were no statistical differences among groups except for the variable
infant weight that was greater for group 3 (Po0.05).
Table 2
Umbilical CB units data of vaginal and caesarean deliveries
Route of delivery
Vaginal deliveries
Mode of collection
Group 1
In utero (n=264)
Volume (ml)
WBC 106/ml
TNCap 108
TNC 108
TNC recovery (%)
CD34+cells%
CD34+cells 106
CFUsb 104
Viability (%)
108.82728.6
10.8873.23
12.5175.39
10.5474.15
77.0878.27
0.3170.2
3.6573.38
154.147141.61
89.1976.75
Caesarean
Group 2
Ex utero (n=309)
Group 3
Ex utero (n=70)
98728.47
10.875.83
11.1474.79
8.5573.52
73.95710.72
0.3270.16
2.9672.25
121.137113.98
89.0878.99
102.5721.65
9.9273.01
10.3174.38
8.3472.87
73.26710.45
0.3570.2
3.0871.97
152.927153.31
90.5275.72
a
Total nucleated cells.
Colony-forming units.
Volume and pTNC were calculated from CB units before processing. The other variables were calculated after volume reduction
of CB units. There was a statistical difference for all variables except for total CD34+ cells (P: 0.057) and viability between groups
1 and 2. There was no statistical difference between groups 2 and 3.
b
significant. The main reasons for discarding these units
were low volume and low total cell count.
Microbiological studies were positive for three cases in
before placental delivery collections and for 19 cases in
after placental delivery collection (Po0.05).
In ex utero collections, clots were present in five cases
that were not discarded (Po0.05). CB units data are shown
in Table 2.
In utero vs ex utero CB collections from vaginal deliveries.
A larger volume and a higher number of nucleated cells
(before and after volume reduction), CD34+ cells,
percentages and CFUs were harvested in the in utero
collection group. There was also a trend for a higher
number of absolute CD34+ cells, although the difference
did not reach statistical significance (P ¼ 0.06). The WBC
concentration was similar for both groups. The difference
also was significant for TNC recovery after volume
reduction (Po0.05).
Ex utero CB collection from vaginal deliveries vs
caesareans. There was no statistical difference between
CB collected after placental expulsion from vaginal and
caesarean deliveries.
In utero CB collections from vaginal deliveries vs
caesareans. A higher number of nucleated cells (before
and after volume reduction) and percentage of CD34+
cells percentages were harvested in the in utero collection
group (Po0.05). A better TNC recovery after volume
reduction was also observed (Po0.05). There was also a
trend for a larger volume (P ¼ 0.074).
All vaginal deliveries versus caesareans. Comparison
between all vaginal deliveries and caesarean did not show
any differences.
Volume and TNC correlate with haematopoietic content.
For vaginal deliveries, a significant correlation was found
between TNC and CD34+ total cells (r: 0.693; Po0.01),
between TNC and total CFUs (r: 0.537; Po0.01), and
between CD34+cells and CFUs (r: 0.641; Po0.01). These
correlations were maintained for caesarean deliveries.
Discussion
Since umbilical CB banking requires important financial
investment and organisational effort, banking efficiency
Bone Marrow Transplantation
Strategies for umbilical cord blood collection
P Solves et al
272
should be optimised. Many CB banks have a minimum
volume limit as the first selection criterion for CB
donations in order to store clinically useful CB units. Some
banks have reported more than one-half of CB collections
are discarded for banking, mainly because of their low
volume and/or cell counts.7 CB banks must develop
strategies to increase the TNC content of stored units. CB
collection is the first step in this. In this sense, it is of major
concern to know which collection strategy allows the bestquality CB units.
The impact of obstetric factors on quality of CB units
has been investigated by some authors. Donaldson et al8
concluded that significant factors affecting volume and
total nucleated cells count were length of gestation, time
from delivery of the infant to cord clamping and total
length of labour. Although these obstetric variables
have not been analysed, in our maternity ward, cord
clamping is always performed immediately after delivery.
Other authors have shown that babies of longer gestational age had higher cell counts, but lower CD34+ cell
counts and CFU-GM. Bigger babies had higher cell
counts, more CD34+ cells and more CFU-GM.9 Longer
duration stress during delivery increases the number of
nucleated cells, granulocytes, CD34+ cells and haematopoietic progenitors.10 In our study, no differences in
maternal or neonatal characteristics were found between
groups.
Obstetric factors are beyond the control of the CB
collector, but collection mode can be controlled. At
present, there is no international consensus on the
procedure of UCB collection in the maternity ward.11
Surbek et al12 compared both strategies in vaginal deliveries
and concluded that in utero collection yielded a significant
higher volume and total number of mononuclear cells.12
Results from a recent study demonstrated that the median
concentrations of nucleated cells and total CFUs were
significantly lower in CB collected after placental delivery
when compared to their counterparts collected before
placental delivery. The incidence of macroscopic clots was
also higher in ex utero collections. The reduction of stem
and progenitor cells is possibly because of clotting activities
developed with time.13 Our results are in agreement with
those presented by Surbeck, and show also higher numbers
of total nucleated cells and CFUs for in utero CB
collections. The presence of haemorrhage in the maternal
and fetal areas of delivered placenta and the clots formed in
fetal placental vessels could explain the loss of haematopoietic progenitor cells in ex utero collections. In our cases,
total nucleated cells correlate with both CFUs and the
number of CD34+ cells, indicating the TNC reflects
haematopoietic content of CB units as shown by Lim
et al.14 Besides, a smaller number of CB units was discarded
for banking compared to ex utero vaginal and caesarean
deliveries, and the only five cases with macroscopic clots
belonged to the ex utero collection group. Time from cord
clamping to collection also could contribute to differences
between in utero and ex utero methods; while in the first
case collection is performed immediately after CB clamping, in the second case it is necessary to wait until placental
expulsion. Other authors have confirmed a larger CB
volume and more total nucleated cells were obtained in the
Bone Marrow Transplantation
samples collected before placental expulsion for caesarean
deliveries.15
However, our conclusions differ from those recently
reported by Sparrow et al who did not find differences
between the two ways of collection for vaginal deliveries, or
between vaginal and caesarean deliveries, although there
was a trend towards a higher cell concentration and
collected volume for in utero collections16. The limited
number of cases (99 ex utero vs 58 in utero) could account
for a lack of statistical power.
Yamada et al17 compared 29 caesarean deliveries with
126 vaginal deliveries and found that umbilical CB volume
and CD34+ cells following caesarean sections and
collected after placenta expulsion were greater, although
the gestational age and the weight of the newborn were
lower.17 The difference was thought to be because of the
position of the umbilical cord and the infant prior to
clamping. Placing the newborn on the maternal abdomen
after delivery increases the volume and CD34+ cell content
in the umbilical CB.18 In our study, obstetric practice was
not modified in any case of CB collection.
We performed all CB collections with a closed bag
system. Open systems have mostly been discarded because
they result in excessive microbiological contamination, as
shown by some authors.5 Our results show microbiological
contaminations were higher for ex utero collections. This
could be explained by manipulation of placenta after
delivery as compared to in utero collections.
For both groups, neither a second puncture nor other
additional collection procedure was used. However, some
authors have tried different systems in order to maximise
the yield of haematopoietic cells in UCB units. The syringeassisted sodium chloride solution flush collection method
with a blood bag has been found to be most effective,
although it resulted in high bacterial contamination rate
and a lower CD34+ cell percentage. Also, the potential
risk of maternal cell contamination could be increased.5
Based on our findings, we conclude that the mode of
collection influences the haematopoietic content of CB
donations. Collection before placental delivery is the best
approach for CB collection to allow optimisations of CB
bank methodology. These results show the feasibility of an
obstetrician-based collection network that has other
advantages. The close relation that the delivering physician
has with the prospective donors facilitates the process of
obtaining quality consent for the banking. Obstetricians
can easily decide not to collect the CB if there are some
maternal/infant problems. This strategy avoids the financial
investment that generates the presence of CB banking
personnel in the maternity ward.
Taking into account our results, caesarean deliveries
seem to contain similar progenitor content to vaginal
deliveries.
References
1 Gluckman E, Broxmeyer HE, Auerbach AD et al. Haematopoietic reconstitution in a patient with Fanconi’s anemia by
means of umbilical-cord blood from an HLA-identical sibling.
N Engl J Med 1989; 321: 1174–1178.
Strategies for umbilical cord blood collection
P Solves et al
273
2 Gluckman E. Current status of umbilical cord blood haematopoietic stem cell transplantation. Exp Hematol 2000; 28:
1197–1205.
3 Stanworth S, Warwick R, Fehily D et al. An international
survey of unrelated umbilical cord blood banking. Vox Sang
2001; 80: 236–243.
4 Gluckman E. Cellular characteristics of cord blood and cord
blood transplantation. In Broxmeyer HE (ed). Cord Blood
Banking and Transplantation in Europe. AABB Press: Bethesda, 1998, pp 147–164.
5 Elchalal U, Fasouliotis SJ, Shtockheim D et al. Postpartum
umbilical cord blood collection for transplantation: a comparison of three methods. Am J Obstet Gynecol 2000; 182: 227–
232.
6 Rubinstein P, Taylor PE, Scaradavou A et al. Unrelated
placental blood for bone marrow reconstitution: organization
of the placental blood program. Blood Cells 1994; 20: 587–596.
7 Jefferies LC, Albertus M, Morgan MA, Moolten D. High
deferral rate for maternal–neonatal donor pairs for an
allogeneic umbilical cord blood bank. Transfusion 1999; 40:
122–123.
8 Donaldson C, Armitage WJ, Laundy V et al. Impact of
obstetric factors on cord blood donation for transplantation.
Br J Haematol 1999; 106: 128–132.
9 Ballen KK, Wilson M, Wuu J et al. Bigger is better: maternal
and neonatal predictors of haematopoietic potential
of umbilical cord blood units. Bone Marrow Transplant 2001;
27: 7–14.
10 Lim FTH, Scherjon SA, van Beckhoven JM et al. Association
of stress during delivery with increased numbers of nucleated
cells and haematopoietic progenitor cells in umbilical cord
blood. Am J Obstet Gynecol 2000; 183: 1144–1151.
11 Surbek DV, Aufderhaar U, Holzgreve W. Umbilical cord
blood collection for transplantation: which technique should
be preferred? Am J Obstet Gynecol 2000; 183: 1587–1588.
12 Surbek DV, Schonfeld B, Tichelli A et al. Optimizing cord
blood mononuclear cell yield: a randomized comparison of
collection before vs after placenta delivery. Bone Marrow
Transplant 1998; 22: 311–312.
13 Wong A, Yuen PM, Li K et al. Cord blood collection before
and after placental delivery: levels of nucleated cells, haematopoietic progenitor cells, leukocyte subpopulations and
macroscopic clots. Bone Marrow Transplant 2001; 27: 133–138.
14 Lim F, Beckhoven J, Brand A et al. The number of nucleated
cells reflects the haematopoietic content of umbilical cord
blood for transplantation. Bone Marrow Transplant 1999; 24:
965–970.
15 Surbek DV, Visca E, Steinmann C et al. Umbilical cord blood
collection before placental delivery during cesarean delivery
increases cord blood volume and nucleated cell number
available for transplantation. Am J Obstet Gynecol 2000;
183: 218–221.
16 Sparrow RL, Cauchi JA, Ramadi LT et al. Influence of mode
of birth and collection on WBC yields of umbilical cord blood
units. Transfusion 2002; 42: 210–215.
17 Yamada T, Okamoto Y, Kasamatu H et al. Factors affecting
the volume of umbilical cord blood collections. Acta Obstet
Gynecol Scand 2000; 79: 830–833.
18 Grisaru D, Deutsch V, Pick M. Placing the newborn on the
maternal abdomen after delivery increases the volume and
CD34 cell content in the umbilical CB collected: an old
maneuver with new applications. Am J Obstet Gynecol 1999;
180: 1240–1243.
Bone Marrow Transplantation