Bone Marrow Transplantation (2006) 38, 175–181
& 2006 Nature Publishing Group All rights reserved 0268-3369/06 $30.00
www.nature.com/bmt
ORIGINAL ARTICLE
Factors affecting volume reduction and red blood cell depletion of bone
marrow on the COBE Spectra cell separator before haematopoietic
stem cell transplantation
MG Guttridge1, C Sidders1, E Booth-Davey2, D Pamphilon1 and SM Watt2
1
Stem Cells and Immunotherapies Laboratories, National Blood Service – Bristol, NHS Blood and Transplant, Bristol, UK and
Stem Cells and Immunotherapies Laboratories, National Blood Service, NHS Blood and Transplant, Oxford, UK
2
The COBE Spectrat is used to volume/red blood cell
(RBC) deplete BM before transplantation or cryopreservation. We have audited our results to identify the
effect of transit time, refrigerated storage, age and
cellular composition on mononuclear cells (MNC) and
CD34 þ cell recoveries, volume/RBC depletion and neutrophil engraftment. In total, 88 consecutive collections
from autologous (n ¼ 25) and allogeneic (n ¼ 63) donors
were included. The mean collection volume was 12507
398 ml with RBC content of 3417113 ml. The MNC and
CD34 þ cell recoveries were 83.3718.5 and 88.1718.9%,
respectively, volume depletion was 88.274.4% and RBC
depletion 98.371.8%. Neutrophil engraftment was
achieved in 20.176.4 days. Factors affecting MNC and
CD34 þ cell recoveries were transit time (P ¼ 0.0060),
overall age (Po0.0210) and MNC/CD34 þ cell concentrations (Po0.0313). The presence of crenated RBC also
reduced CD34 þ cell recovery (P ¼ 0.0028). Refrigerated
storage did not adversely affect cell recovery (P40.8161)
or neutrophil engraftment (P ¼ 0.8959). This study
demonstrates that time in transit, overall age, MNC and
CD34 þ cell concentrations and RBC condition were
important factors affecting processing. RBCs show
artefacts soon after collection at ambient temperatures
and these may interfere with the separation and collection of MNC/CD34 þ cells. Refrigeration at 4–61C
during transit and storage may reduce formation of
RBC artefacts and maximize MNC and CD34 þ cell
recoveries.
Bone Marrow Transplantation (2006) 38, 175–181.
doi:10.1038/sj.bmt.1705420; published online 12 June 2006
Keywords: bone marrow; red blood cells; volume reduction; cell separator; CD34 recovery; engraftment
Correspondence: Dr MG Guttridge, Stem Cells and Immunotherapies,
National Blood Service – Bristol, NHS Blood and Transplant, Bristol
BS10 5ND, UK.
E-mail: [email protected]
Received 28 February 2006; revised 17 May 2006; accepted 20 May 2006;
published online 12 June 2006
Introduction
Bone marrow collections for allogeneic and autologous
haematopoietic stem cell transplantation are typically over
a litre in volume. In order to reduce the cryogenic storage
space required, it is important to volume reduce bone
marrow collections before cryopreservation and long-term
storage. In addition, allogeneic grafts that are ABO
incompatible may be red blood cell (RBC) depleted before
infusion to minimize the risk of ABO-induced transfusion
reactions.1 In the procedure, it is important to maximize
CD34 þ haematopoietic progenitor cell recovery, as
engraftment of haematopoietic stem cells correlates with
the dose of CD34 þ cells infused.2,3
The COBE Spectra cell separator has been used for
volume reduction and RBC depletion of bone marrow for
over 14 years. It is a continuous flow cell separator that
isolates cells initially through centrifugation and the
formation of a buffy layer. Mononuclear cells (MNCs)
and CD34 þ progenitor cells are collected from the buffy
layer through an MNC channel that is maintained in
position by pressure differentials in the collection chamber
through adjustment of the plasma flow rates. Using the
COBE Spectrat, Baker et al.4 reported in 1991 MNC and
colony-forming units-granulocyte–macrophage progenitor
cell recoveries of 76.4 and 86.5%, respectively, and volume
depletion of 87.0%. In a separate study, Koristek and
Mayer5 reported 78.8% MNC and 77.2% CD34 þ progenitor cell recoveries. These studies showed considerable
variation in the recoveries of MNC and CD34 þ cells,
ranging from 33 to 260%. Despite this, the factors affecting
the recovery of these cells were not identified.
We have undertaken an audit of our COBE Spectrat
processing at our Stem Cells and Immunotherapies (SCI)
Department of the National Blood Service (NBS) in Bristol
on all allogeneic bone marrow harvests collected between
1 January 2003 and 31 October 2005 and, because there
were many more, on all autologous collections between 1
April 2004 and 31 December 2004. The objectives were (1)
to compare the performance against published results and
(2) to investigate factors that might influence MNC and
CD34 þ cell recovery, including RBC, MNC and CD34 þ
cellular composition, and environmental factors such as
transit and storage conditions. Here, we describe the
outcome of this audit and make recommendations for
Bone marrow volume and erythrocyte reduction
MG Guttridge et al
176
optimizing MNC and CD34 þ cell recoveries during COBE
Spectrat processing for volume reduction and RBC
depletion.
Materials and methods
Bone marrow collections
All bone marrow harvests were collected with written, fully
informed consent for processing, testing and subsequent
storage of their stem cells. In total, 88 bone marrow
collections were included in the study, of which 25 were
autologous and 63 were allogeneic donations. Of these, 47
were collected locally in Bristol with assistance of NBS SCI
laboratory staff. The NBS collections were taken from
the posterior iliac crest under general anaesthesia. Anticoagulant, comprising ACD-A (1:7) and heparin (10 IU/ml),
was added gradually to avoid exposure to high ACD-A
concentrations. In most instances, the white blood count
(WBC) in the bone marrow was determined approximately
halfway through the collection procedure to allow an
estimate of the end point volume required to give a
nucleated cell dose of 3–4 108/kg recipient body weight.
These bone marrow collections were transported to the
processing laboratory at ambient temperature in insulated
transit containers validated for transport of blood donations. In addition, 41 allogeneic donations were imported
from other collection centres, of which 26 were within the
UK, 10 from other European countries, four from the USA
and one from Australia. In these cases, the collections were
transported to the processing laboratory by courier in
insulated transit containers according to World Marrow
Donor Association (WMDA) recommendations.6 The
temperature during transit was recorded for one collection
only. This was from Australia and ranged between 6 and
161C. A further nine of the 41 imported collections were
transported chilled, although the temperature on receipt
was recorded for only four and averaged 9.51C. Two
collections were supplemented with RPMI-1640 medium.
On receipt in the laboratory, the collections were evaluated
to determine (1) the collection volume (ml), (2) the
collection haematocrit (HCT), (3) the total nucleated cell
count, (4) the MNC count and (5) the specific CD45 þ ,
CD34 þ and CD3 þ cell counts. The transit time was
calculated as the time taken from the end of the harvest
procedure to receipt in the processing laboratory. If
processing could not be commenced immediately, collections were stored at 2–61C until processing could begin.
Processing using the COBE Spectra
The COBE Spectra (COBE BCT Inc., Lakewood, CO,
USA) was used with a single-stage WBC tubing set on the
WBC programme using the bone marrow processing
setting as described by the manufacturer. During the
MNC collection, manual adjustment of pump rates was
used to maintain the RBC/plasma interface within the
WBC collection channel. The bone marrow processing was
completed when (a) the MNC fraction reached 130 ml
volume, (b) the bone marrow had re-circulated through the
collection process five times or (c) WBC were no longer
apparent in the collection chamber.
Bone Marrow Transplantation
Cell counts, ABO typing and CD34 þ cell analysis
Cell counts were measured using an ethylenediaminetetraacetic acid-anticoagulated sample with an LH750
Haematology Analyser (Beckman Coulter, High Wycombe, Buckinghamshire, UK) and by single-platform
three-colour flow cytometry using an International Society
of Hematotherapy and Graft Engineering gating strategy.7
The CD45, CD34 þ and MNC cell counts were determined
by flow cytometry.8 The vital dye, 7-aminioactinomycin D
(7-AAD) was used to exclude non-viable cells and an
internal positive control8 was included to ensure the assay
was working optimally. Cell doses were calculated based on
recipient body weight (kg) at the time of processing. MNC
and CD34 þ recoveries in the final product were determined
as a percentage of the initial starting dose.
Measurement of neutrophil engraftment
Neutrophil engraftment was measured as the number of
days after infusion until the circulating peripheral blood
neutrophil count reached 0.5 109/l for 3 consecutive days.
Statistical analyses
Results are expressed as mean7s.d. Statistical analyses
were carried out using an unpaired two-tailed t-test to
compare two means (http://home.clara.net/sisa/t-test.htm).
Where the variances were significantly different (determined using the f test), the Welch’s t-test was used. A
P-value of o0.05 was considered significant, o0.01 very
significant and o0.001 extremely significant. Calculations
are presented to four significant figures.
Results
Characteristics of the bone marrow harvests before
and after processing
The overall characteristics of all bone marrow harvests
audited are summarized in Table 1. The initial bone
marrow collections had a mean volume of 12507398 ml
(range 396–2309 ml) and RBC content of 3417113 ml
(range 99–607 ml). Following processing on the COBE
Spectrat, the volume of the MNC fraction was 135732 ml
(88.2 þ 4.4% volume depletion) with RBC volume of
5.073.8 ml (98.371.8% RBC depletion). The mean
MNC and CD34 þ cell recoveries were 83.3718.5% (range
10–115%) and 88.1718.9% (range 10–116%), respectively.
In total, 60 of the 63 allogeneic grafts and three of the 25
autografts were infused with the remaining 25 cryopreserved for future use. The mean CD34 þ cell dose infused
was 4.774.0 106/kg and neutrophil engraftment at
0.5 109/l averaged 20.176.4 days. The overall CD45 þ
cell viability was 97.871.19% as determined by flow
cytometry using 7-AAD staining.
When the bone marrow harvests were divided into
autologous versus allogeneic and on the basis of local or
distant collection centre (Table 2), the highest MNC
(94.078.5%) and CD34 þ cell (99.777.6%) recoveries
were obtained for allogeneic donations collected locally.
Although the autologous donations were also all from local
Bristol centres, when compared to the allogeneic collections
from Bristol, they had significantly lower MNC recoveries
Bone marrow volume and erythrocyte reduction
MG Guttridge et al
177
Table 1
Collection and processing outcomes for all bone marrow harvests audited
Minimum
Maximum
Median
Mean
s.d.
n
Bone marrow collection
Volume (ml)
RBC volume (ml)
HCT
Transit time (h)
Storage time
Overall age (h)
396
99
15.9
0.5
0
0.87
2309
607
35.7
35.5
22.4
36.25
1223
339
27.5
2.3
1.4
5.63
1250
341
27.7
4.5
5.4
9.92
398
113
4.4
6.1
7.3
9.52
88
87a
87a
88
88
88
MNC fraction
Volume (ml)
RBC volume (ml)
MNC conc ( 108/l)
CD34 conc ( 106/l)
Total MNC ( 108)
Total CD34 ( 106)
87
1.3
3.2
1.9
6.4
3.9
262
23.5
84.8
595.0
87.0
508.4
123
3.8
35.4
141.3
41.4
172.7
135
5.0
36.9
169.9
42.8
187.4
32
3.8
15.0
117.5
15.4
106.0
88
87a
88
88
88
88
Recoveries/outcomes
CD34+ cell recovery (%)
MNC recovery (%)
RBC depletion (%)
Volume depletion (%)
Neutrohil engraftment (days)
CD34+ cell dose infused ( 106/kg)
10.0
10.0
86.4
70.7
7
0.24
115.7
115.0
99.6
94.7
51
18.38
93.2
86.6
98.8
89.1
20
3.53
88.1
83.3
98.3
88.2
20.1
4.65
18.9
18.5
1.8
4.4
6.4
4.00
88
88
87a
88
63b
63b
Abbreviations: HCT ¼ haematocrit; MNC ¼ mononuclear cells; RBC ¼ red blood cell; s.d. ¼ standard deviation.
a
One collection from an HBsAg-positive donor was not tested for RBC content using the LH750 haematology analyser.
b
Of the 88 harvests analysed, 63 were subsequently transplanted.
Table 2
Environmental factors, cellular composition and processing outcomes that differed for donor type (allogeneic/autologous) and
geographical distribution of the collection centres
Collection
centre
Environmental factors
Transit
time (h)
Pre-process cellular composition
Overall
age (h)
Bristol (allogeneic)
Mean
1.04
s.d.
0.42
N
22
3.42
6.36
22
4.46
6.45
22
28.4
2.7
22
45.0
14.8
22
230.6
140.7
22
46.6
14.8
22
99.7
7.6
22
94.0
8.5
22
Bristol (autologous)
Mean
1.33
s.d.
0.69
N
25
P-value
0.0867
7.49
7.56
25
0.0536
8.82
7.95
25
0.0466
24.3
4.5
25
0.0004
28.4
15.8
25
0.0005
140.1
136.7
25
0.0307
35.5
18.8
25
0.0310
86.1
16.2
25
0.0006
80.4
16.9
25
0.0012
Other UK centres
Mean
4.60
s.d.
1.86
N
26
P-value
o0.0001
3.18
5.74
26
0.8913
7.79
6.51
26
0.0828
29.5
3.7
26
0.2661
39.3
13.0
26
0.1626
149.9
65.0
26
0.0206
47.2
12.4
26
0.8771
89.7
16.7
26
0.0094
85.9
14.3
26
0.0194
Mainland Europe
Mean
9.46
s.d.
2.80
N
10
P-value
o0.0001
12.70
8.67
10
0.0019
22.16
8.68
10
o0.0001
29.3
5.6
10
0.6376
34.5
11.2
10
0.0555
157.2
96.6
10
0.1462
41.2
6.2
10
0.1529
70.0
28.0
10
0.0071
63.3
27.7
10
0.0060
0.57
0.22
5
0.0442
26.08
6.10
5
o0.0001
27.4
3.9
5
0.4580
37.1
7.3
5
0.2613
180.8
86.9
5
0.4587
43.3
16.6
5
0.6790
74.0
23.5
5
0.0776
76.8
23.1
5
0.1860
25.51
5.99
5
0.0008
MNC conc
( 108/l)
CD34+ cell
recovery (%)
Storage
time (h)
USA/
Australia
Mean
s.d.
N
P-value
HCT
Post-processing outcomes
CD34 conc
( 106/l)
Total
MNC
( 108)
MNC
recovery (%)
Abbreviations: HCT ¼ haematocrit; MNC ¼ mononuclear cells; s.d. ¼ standard deviation.
Bone Marrow Transplantation
Bone marrow volume and erythrocyte reduction
MG Guttridge et al
CD34 count (×106) recovered
178
500
y = 0.9681x
400
R2 = 0.9784
300
200
100
0
100
200
300
400
CD34 count (×106) in collection
500
Scatter graphs showing correlation between CD34 þ cell count
in the initial collection with that recovered after processing for allogeneic
collections segregated to show geographical distribution of the collection
centres. The ‘best-fit’ trend line assumes an intercept at the origin and
shows a linear relationship for local allogeneic donations (slope,
y ¼ 0.9681x, R2 ¼ 0.9784). Note that the allogeneic collections with poor
CD34 þ cell recovery are all imported from outside the Bristol region. E,
Bristol; &, other UK centres; W, mainland Europe; J, USA/Australian
collection centres. Results are expressed as CD34 þ cell count 106.
Figure 1
(80.4716.9%, P ¼ 0.0012) and lower CD34 þ cell recoveries (86.1716.2%, P ¼ 0.0006). These differences may be
attributable to the longer storage times (7.49 versus 3.42 h,
P ¼ 0.0536) and greater average age (8.82 versus 4.46 h,
P ¼ 0.0466) of the autologous collections, although other
differences identified in the composition of cells collected
from patients with haematological disorders (e.g. HCT,
MNC and CD34 þ cell content) may also be important
contributors.
To analyse which of these factors were important, we
compared similar harvests (namely the allogeneic harvests)
from healthy donors that were collected locally (in Bristol)
with those collected at other UK centres or from abroad
and then processed at the Bristol laboratory. The lowest
MNC and CD34 þ cell recoveries occurred with collections
from mainland Europe (MNC, 63.3727.7%; CD34 þ cells,
70.0728.0%) and USA/Australia (MNC, 76.8723.1%;
CD34 þ cells, 74.0723.5%). Figure 1 shows the relationship between the initial and recovered CD34 þ cell counts
for allogeneic collections based on their geographical distribution. Allogeneic collections from the local Bristol centre
showed a strong linear relationship (y ¼ 0.9681) with
excellent data correlation R2 ¼ 0.9784. The donations from
other UK centres generally show CD34 þ cell recoveries
close to the linear trend, whereas those from abroad more
frequently fell below the trend line. Although we cannot
exclude a collection centre effect, there were other environmental factors identified in the study, including transit and
storage times, and overall age, that were significantly
different (Table 2) and are likely to account for the results.
Effect of time after harvest for allogeneic donations
The time in transit varied between 0.5 and 35.5 h with a
median time over the 63 collections of 3.42 h. The mean cell
recoveries for collections in transit over 3.42 h (n ¼ 31) were
77.7723.6% for MNC and 81.8724.3% for CD34 þ cells
and were significantly lower (P ¼ 0.0060) than those with
shorter (o3.42 h) transit times (MNC: 91.079.9% and
CD34 þ cells: 95.7711.2%) (Table 3). Further to this, those
Bone Marrow Transplantation
in transit less than 2 h (n ¼ 21) achieved CD34 þ cell
recoveries of 99.777.8% and recoveries fell progressively
as transit time increased. Those in transit between 2 and 6 h
(n ¼ 20), between 6 and 12 h (n ¼ 15) and over 12 h (n ¼ 7)
achieved respective CD34 þ cell recoveries of 91.4711.5,
77.4728.6 and 73.2723%, and were all significantly
different from those in transit less than 2 h (Po0.0280).
The time in 2–61C storage ranged between 0 and 22.4 h
with a median for the allogeneic donations of 1.0 h. The
majority of collections were processed soon after receipt in
the laboratory with 50 stored for 3.5 h or less. The
remaining 13 were stored overnight (413 h) for processing
the following day. Those collections processed within 1 h of
receipt (n ¼ 32) had cell recoveries of 85.0717.9% for
MNC and 88.8720.1% for CD34 þ cells, whereas those in
storage over 1 h (n ¼ 31) had MNC and CD34 þ cell
recoveries only slightly lower at 83.9720.5% (P ¼ 0.8161)
and 88.8720.1% (P ¼ 0.9970), respectively. To further
support the observation that refrigerated storage did not
significantly affect cell recovery, and to exclude the
significant effect of transit times, the CD34 þ cell recoveries
were compared for all collections that were in transit less
than 2.25 h and were in storage for (a) less than 1 h (n ¼ 15),
(b) between 1 and 16 h (n ¼ 20) and (c) over 16 h (n ¼ 9)
(Figure 2). The CD34 þ cell recoveries were 96.3710,
94.4712.9 and 90.3712.5%, respectively, and were not
significantly different (P40.2082).
The overall age of the collections at the time of
processing had a significant effect on both MNC
(P ¼ 0.0175) and CD34 þ cell (P ¼ 0.0217) recoveries. Those
collections less than 5.7 h old gave MNC and CD34 þ cell
recoveries of 90.1710.1 and 94.6711.6%, respectively,
whereas the recoveries fell to 78.6724.0% for MNC and
82.9724.7% for CD34 þ cells with older collections. Those
between 0 and 5 h old (n ¼ 24) generated CD34 þ cell
recoveries of 97.578.7%, whereas CD34 cell recoveries fell
progressively to 86.3721.4% (n ¼ 23, 5–20 h old), and
79.4725.1% (n ¼ 16, 20–40 h old) as the collections aged
(Po0.0264).
Effect of graft composition on cell recoveries
for allogeneic collections
The concentration of MNC and CD34 þ cells (Table 3), in
the pre-processed bone marrows, had significant effects on
both CD34 þ cell (Po0.0269) recoveries. Bone marrow
collections with MNC concentrations above the median
(36.6 108/l) gave better CD34 cell recoveries (95.1713.4
versus 82.72723.3%) than those with MNC concentrations below the median. Similarly, collections with CD34 þ
cell concentration above the median (151.5 106/l) had
recoveries 11.2% higher (95.5714.2 versus 84.3723.6)
than those below the median. There was no effect of
collection volume, RBC content, total MNC or total
CD34 þ cell counts on CD34 þ or MNC cell recoveries
(data not shown).
RBC crenation is linked to poor CD34 þ cell recoveries
The reduced MNC and progenitor cell recoveries observed
may reflect RBC degenerative changes, such as sphering
and crenation, which can occur within a few hours of
storage at room temperature.10,11 These effects are likely to
Bone marrow volume and erythrocyte reduction
MG Guttridge et al
179
Table 3
Factors affecting processing on the Cobe Spectra
Outcomes
% CD34
Time in transit (median:
Less than median
Mean
s.d.
N
Greater than median
Mean
s.d.
N
P-value
Time in storage (median:
Less than median
Mean
s.d.
N
Greater than median
Mean
s.d.
N
P-value
+
cell recovery
% MNC recovery
91.0
9.9
32
98.4
1.5
32
87.0
4.8
32
20.4
6.8
31
81.8
24.3
31
0.0059
77.7
23.6
31
0.0060
98.1
2.5
31
0.6160
88.5
4.2
31
0.2070
18.3
4.1
29
0.1467
88.8
20.1
32
85.0
17.9
32
98.0
2.6
32
86.7
5.5
32
19.3
6.8
32
88.8
20.1
31
0.9970
83.9
20.5
31
0.8161
98.5
1.1
31
0.3142
88.8
3.1
31
0.0679
19.5
4.2
28
0.8959
90.1
10.1
32
98.4
1.6
32
87.4
4.8
32
19.8
6.8
31
78.6
24.0
31
0.0175
98.2
2.4
31
0.6846
88.1
4.3
31
0.5326
18.9
4.4
29
0.5338
79.4
23.1
32
98.6
2.4
32
88.9
4.3
32
18.9
3.6
29
89.6
12.1
31
0.0313
97.9
1.5
31
0.2267
86.5
4.5
31
0.0322
19.9
7.1
31
0.4774
81.1
23.4
32
98.7
1.12
32
89.3
3.1
32
19.2
3.7
29
87.9
12.7
31
0.1586
97.8
2.6
31
0.0641
86.1
5.2
31
0.0055
19.6
7.2
31
0.8310
1.0 h)
82.9
24.7
31
0.0217
95.1
13.4
31
0.0120
CD34+ cell concentration (median: 151.5 106/l)
Less than median
Mean
84.3
s.d.
23.6
N
32
Greater than median
Mean
s.d.
N
P-value
Neutrophil engraftment (days)
95.7
11.2
32
MNC concentration (median: 36.6 108/l)
Less than median
Mean
82.7
s.d.
23.3
N
32
Greater than median
Mean
s.d.
N
P-value
% Volume depletion
3.4 h)
Overall age (median 5.7 h)
Less than median
Mean
94.6
s.d.
11.6
N
32
Greater than median
Mean
s.d.
N
P-value
% RBC depletion
95.5
14.2
31
0.0269
Abbreviations: MNC ¼ mononuclear cells; RBC ¼ red blood cell; s.d. ¼ standard deviation.
Note: CD34+ cell recoveries were significantly affected by all parameters shown except time in storage.
interfere with the formation of the buffy layer with the
affected RBCs likely to segregate with WBCs owing to their
modified shape and density. The normal range for mean
cell volume (MCV) is 9279 fl,12 which was similar in our
study (92.675.9 fl). After processing on the COBE Spectra,
those collections with MCV less than 92 fl (n ¼ 33) had
CD34 þ cell recoveries only 1.5% higher (89.4722.3 versus
87.9717.5%) than those collections with MCV less than
Bone Marrow Transplantation
Bone marrow volume and erythrocyte reduction
MG Guttridge et al
180
110
CD34+ cell recovery (%)
105
100
95
90
85
80
75
70
65
60
<1
1 to 16
>16
Time in storage (h)
A bar chart comparing CD34 þ progenitor cell recovery with
storage between 2 and 61C. The mean CD34 þ cell recoveries (error
bars ¼ s.d.) are shown for collections in storage for less than 1 h (n ¼ 15,
four autologous, 11 allogeneic), between 1 and 16 h (n ¼ 20, 12 autologous,
eight allogeneic) and over 16 h (n ¼ 9, six autologous, three allogeneic).
The collections were all from the Bristol centre with transit times less
than 2.25 h. Note, CD34 recoveries fall by 1.9% (P ¼ 0.6265) between 1 and
16 h in storage and by 6% (P ¼ 0.2082) after overnight storage (416 h).
Figure 2
92 fl (n ¼ 29). Six collections had MCV above the normal
range (101–108 fI) and these had CD34 recoveries 12.3%
lower at 77.1723.8%. Further to this, collections with
MCV above 92 fl had increased initial collection volumes
(13437311 ml (n ¼ 29) versus 10817362 ml (n ¼ 33),
P ¼ 0.0036). The reason for this correlation is not clear,
but suggests that RBCs may be damaged as the bone
marrow collections progress. Stained films of allogeneic
bone marrow donations were reviewed to determine the
extent of RBC crenation and 11 collections were shown to
have 45% crenated RBC, 36 o5% crenated RBC and
16 were not tested. Those collections with crenated RBC
had CD34 þ cell recoveries of 64.9732% compared to
93.8712.4% (P ¼ 0.0028) for those where crenated RBC
were absent.
Discussion
This study supports the previous reports from Baker et al.4
and Koristek and Mayer5 that the COBE Spectrat can be
used effectively to volume reduce and red cell deplete bone
marrow collections. The final volume of the MNC fraction
was 135732 ml and was equivalent to a mean volume
depletion of 88.274.4%, which is sufficient to allow cryostorage of donations when necessary. Volume depletion
was not affected by the environmental factors studied, but
was significantly influenced by the initial harvest volume
and the MNC and CD34 þ progenitor cell concentrations
(data not shown). In addition, the mean RBC depletion of
98.371.8% equates to an average RBC volume of
5.073.8 ml and was sufficient to remove the risk of ABOinduced transfusion reactions in most cases.9 The MNC
and CD34 þ cell recoveries achieved for autologous and
allogeneic collections combined were 83.3718.5% (range
10–115%) and 88.1718.9% (range 10–116%), respectively,
and these compare favourably with those from the previous
studies.4,5
Bone Marrow Transplantation
As reported previously,4,5 there was considerable variation
in both MNC and progenitor cell recoveries. In the present
study, MNC and CD34 þ progenitor cell recoveries ranged
between 10 and 116% for both MNC and CD34 þ cells. The
influence of environmental factors, such as time in transit,
time in refrigerated storage and overall age, and the cellular
composition of the donations was investigated to identify
possible explanations for the wide range of cell recoveries.
The most significant environmental factor affecting both
MNC (P ¼ 0.0060) and CD34 þ cell (P ¼ 0.0059) recoveries
was time in transit. Cell recoveries for allogeneic collections
in transit over the median time of 3.42 h were on average
13.3% lower for MNC and 13.9% lower for CD34 þ
progenitor cells than those collections in transit less than
the median. Further to this, the CD34 þ cell recoveries
became progressively lower as the transit time increased
declining from 99.777.8% for 2 h transit times to
73.2723% for over 12 h transit times.
We hypothesized that reduced MNC and progenitor cell
recoveries observed as the harvests aged might reflect RBC
degenerative changes, such as sphering and crenation,
which can occur within a few hours of storage at room
temperature.10,11 An examination of the extent of RBC
crenation revealed that those collections with crenated
RBC had CD34 þ cell recoveries of 64.9732% compared
to 93.8712.4% (P ¼ 0.0028) for those where crenated RBC
were absent. These degenerative changes to RBCs are
known to be reduced by refrigerated storage between 2
and 61C,10,11,12,13 and this may explain the beneficial effect
of storage observed in our study, where both MNC and
CD34 þ progenitor cell recoveries were protected by
refrigeration. In line with previous findings,14 there was
no apparent effect of refrigerated storage on CD34 þ
viability of the allogeneic bone marrow grafts. Neutrophil
engraftment was not significantly different (P ¼ 0.8959) for
collections stored for less than 1 h when compared with
those stored for more than 1 h (19.574.2 versus 19.376.8
days). Further to this, a total of 11 allogeneic collections
were stored refrigerated overnight before processing and
neutrophil engraftment at 0.5 109/l was still reached
within 19.974.7 days.
The overall age of the collections also had a significant
effect on MNC and CD34 þ progenitor cell recoveries,
although to a lesser extent than the time in transit.
Collections that were less than the median age of 5.7 h
before processing had MNC and CD34 þ progenitor cell
recoveries 11.5 and 11.7% higher, respectively, when
compared to collections over 5.7 h old. The overall age of
the collections resulted from a combination of both transit
and storage times and, as refrigerated storage on receipt of
the harvest did not significantly reduce MNC or CD34 þ
cell recoveries, this most likely accounts for the effect of age
being smaller than transit time.
There was no correlation between neutrophil engraftment with either MNC or CD34 þ cell recovery or
with refrigerated storage. Only six of the 88 collections
gave CD34 þ cell recoveries below 50% ranging between
10 and 48.5%. In order to recover sufficient cells for
transplantation, the COBE Spectrat processing residues
from five of these six harvests were processed again using
the COBE 2991 according to the manufacturer’s instruc-
Bone marrow volume and erythrocyte reduction
MG Guttridge et al
181
tions and overall CD34 þ cell recoveries of 88.2711.9%
were achieved (data not shown). As such, the low CD34
recoveries obtained in the first processing run for these six
patients did not directly impact on the CD34 þ progenitor
cell dose at the time of infusion. The RBC content was
higher after processing on the COBE 2991 at 49.5713.2 ml
and, despite ABO incompatibility in two of the transplants,
no adverse effects occurred during the infusions. The
incidence of severe ABO-induced transfusion reactions
after incompatible RBC transfusion is o32%,15 and
factors such as the recipient’s antibody titre, and the
volume and rate of the infusion are likely to affect the
incidence of adverse reactions.9
It is common practice to transport bone marrow
collections between the collection and transplant centres
at ambient temperatures. Our studies suggest that transit
from other UK centres (4.671.9 h transit time) can reduce
MNC and CD34 þ progenitor cell recoveries when processed on the COBE Spectrat by an average of 10%.
Collections transported from other European countries
(9.572.8 h transit time) had average MNC and CD34 þ cell
recoveries that were 29.7% lower. As storage of the
collections at 2–61C did not appear to significantly reduce
MNC or CD34 þ cell recoveries, refrigerated transit of
collections may be beneficial. During the audit period, all
but 10 of the 41 imported products were transported at
ambient temperatures, although refrigerated transport at
2–61C was requested in all cases. Of the 10 collections
transported chilled, the temperature was measured during
transit for only one, and on receipt in only four cases with
only one being between 2 and 61C on receipt. There was
no significant difference (P ¼ 0.7614) in the CD34 þ cell
recovery between the 10 collections chilled during transit
and the remaining 31 collected outside the Bristol area,
although further studies using collections where temperatures are recorded during transit are required.
In conclusion, the COBE Spectra can be used to reduce
both the volume and RBC content of bone marrow
collections with mean RBC depletion and CD34 þ cell
recoveries of 98.371.8 and 88.1718.9%, respectively.
Time in storage and time to processing following harvest
affected CD34 þ cell recoveries and this may reflect
metabolic damage to RBCs associated with room temperature storage. In line with existing WMDA recommendations,6 this study confirms that the time in transit between
the collection and transplant centres should be minimized
and, once received in the processing centre laboratory,
collections should be processed as soon as possible.
Refrigeration between 2 and 61C during transit is likely
to be beneficial and, where processing cannot be performed
immediately, collections should be stored refrigerated
overnight to preserve RBC and maximize MNC and
CD34 þ cell recovery.
Acknowledgements
This work is supported by the NHS Blood and Transplant
Authority and the Department of Health, UK.
References
1 Klumpp TR. Immunohematologic complications of bone
marrow transplantation. Bone Marrow Transplantation 1991;
8: 159–170.
2 Korbling M, Huh YO, Durett A, Mirza N, Miller P, Engel H
et al. Allogeneic blood stem cell transplantation: peripheralization and yield of donor-derived primitive hematopoietic
progenitor cells (CD34+ Thy-1dim) and lymphoid subsets,
and possible predictors of engraftment and graft-versus-host
disease. Blood 1995; 86: 2842–2848.
3 Bensinger WI, Longin K, Appelbaum F, Rowley S, Weaver C,
Lilleby K et al. Peripheral blood stem cells (PBSCs) collected
after recombinant granulocyte colony stimulating factor (rhGCSF): an analysis of factors correlating with the tempo of
engraftment after transplantation. Br J Haematol 1995; 87:
825–831.
4 Baker PK, Rhodes EG, Duguid JK. Continuous flow cell
separator use for bone marrow processing. Transfus Sci 1991;
12: 183–187.
5 Koristek Z, Mayer J. Bone marrow processing for transplantation using the COBE Spectra cell separator. J Hematother
Stem Cell Res 1999; 8: 443–448.
6 Cleaver SA, Warren P, Kern M, Hurley CK, Raffoux C, Keller
J et al. Donor work-up and transport of bone marrow –
recommendations and requirements for a standardized practice throughout the world from donor Registries and Quality
Assurance Working Groups of the World Marrow Donor
Association (WMDA). Bone Marrow Transplant 1997; 20:
621–629.
7 Keeney M, Chin-Yee I, Weir K, Popma J, Nayar R,
Sutherland DR. Single platform flow cytometric absolute
CD34+ cell counts based on the ISHAGE guidelines.
International Society of Hematotherapy and Graft Engineering. Cytometry 1998; 34: 61–70.
8 Hollyman D, Sidders C, Lankester A, Watt S, Pamphilon D,
Guttridge MG. A novel lyse-no-wash 4-colour flow cytometry
assay with an internal positive control for the simultaneous
enumeration of CD34+ and CD3+ cells prior to allogeneic
transplantation. (ISCT abstract 230 http:www.celltherapy.org/
Abstracts ).
9 Pamphilon DH. Transfusion policy. In: Apperley J, Carreras
E, Gluckman E, Gratwohl A, Masszi T (eds). The EBMT
Handbook – Haematopoietic Stem Cell Transplantation. Forum
Service Editore: Genoa, 2004, pp 107–117.
10 Lewis SM. Collection and handling of blood. In: Lewis SM,
Bain BJ, Bates I (eds). Dacie and Lewis Practical Haematology.
Churchill Livingstone: London, 2001, pp 1–8.
11 Brittin GM, Brecher G, Johnson CA, Elashoff RM. Stability
of blood in commonly used anticoagulants. Am J Clin Pathol
1969; 52: 690–694.
12 Lewis SM. Reference ranges and normal values. In: Lewis SM,
Bain BJ, Bates I (eds). Dacie and Lewis Practical Haematology.
Churchill Livingstone: London, 2001, pp 9–18.
13 Lawrence AC, Bevington JM, Young M. Storage of blood
and the mean corpuscular volume. J Clin Pathol 1975; 28:
345–349.
14 Lioznov MV, Freiberger P, Kröger N, Zander AR, Fehse B.
Aldehyde dehydrogenase activity as a marker for the quality of
hematopoietic stem cell transplants. Bone Marrow Transplant
2005; 35: 909–914.
15 Stainsby D, Cohen H, Jones H, Boncinelli A, Knowles S,
Birchall J et al. Serious Hazards of Transfusion Annual
Report. 2004. http://www.shot-uk.org.
Bone Marrow Transplantation