Excluding Anti-cytomegalovirus Immunoglobulin M-Positive Cord Blood Units Has a Minimal Impact on the Korean Public Cord Blood Bank Inventory

Abstract

Cord blood units (CBUs) for transplantation should be free of communicable disease and must contain a specific amount of total nucleated cells and CD34⁺ cells. Although posttransplantation cytomegalovirus (CMV) infections are from latent infection in patients, ensuring CMV-free CBUs by performing CMV-specific IgM and nucleic acid amplification testing (NAT) is one of the mandatory procedures for the safety of CBUs. However, the exclusion policies (based on these test results) vary among nations and institutions. We tested 28,000 processed CBUs between May 2006 and June 2014. The cord blood leukocytes from CMV IgM-positive samples were then subjected to NAT. The total nucleated cell and CD34⁺ cell counts were measured for each CBU, and the results were compared to the CMV IgM and IgG results. The seroprevalence of CMV among pregnant women was 98.1% (18,459/18,818) for IgG and 1.7% (441/25,293) for IgM. The concentration and the total number of CD34⁺ cells were significantly higher in CBUs from IgM-negative mothers compared to those from IgM-positive mothers (72.4/μl vs. 57.2/μl, respectively, p < 0.0001; 1.45 × 10⁶/unit vs. 1.15 × 10⁶/unit, respectively, p < 0.0001). Among CBUs with positive CMV IgM in their mothers' plasma or cord blood plasma, only 0.58% of the samples (3/517) had a positive NAT. The number of excluded CBUs from inventory due to positive CMV IgM in the cord blood was 54 of 18,326 (0.3%). For inventory purposes, it is appropriate to remove CBUs with positive cord blood CMV IgM findings irrespective of the NAT status as well as positive maternal CMV IgM in South Korea.

Keywords

Cytomegalovirus (CMV)Cord blood Banking Inventory

Introduction

Umbilical cord blood units (CBUs) for transplantation must be tested to ensure that they are free of transmissible infectious disease, to ensure that they contain the appropriate amount of total nucleated cells (TNCs), and to identify their human leukocyte antigen (HLA) status (4). On the basis of these test results, CBUs can then be selected for transplantation according to their HLA status and anticipated potency, which is represented by their TNC count and number of CD34⁺ cells (17). In South Korea, public cord blood banks (pCBBs) store CBUs with a TNC count of 7 × 10⁸ (12,16,22), and a centralized Korean Network for Organ Sharing (KONOS) website (http://is.cdc.go.kr) has been developed to list characteristics of CBUs, including CD34⁺ cell counts, TNC count, sex of the baby, ABO and Rh blood groups, and HLA status. Therefore, transplantation physicians can use this database to identify CBUs that are appropriate for their patient.

To ensure patient safety, pCBBs perform testing for surface antigen of hepatitis B virus and antibodies to the human immunodeficiency virus (HIV), hepatitis C virus, human T-lymphotropic virus, syphilis, and cytomegalovirus (CMV) in both the maternal and cord blood plasma in South Korea (12,16,22). CBUs with positive CMV IgM results in cord blood plasma could only be transferred to the inventory after confirming that the cord blood leukocytes tested negative for CMV using a CMV nucleic acid amplification test (NAT). CBUs with positive CMV IgM in the mother's plasma should be excluded without performing CMV NAT. In Western countries, several endemic diseases are also included in the routine tests for managing CBUs, and the CMV NAT is performed at the time when the CBU is administered (1). The standard operating procedures of ALLCORD (Seoul Metropolitan Government Public Cord Blood Bank, Seoul, South Korea) also stipulate that the CMV NAT should be performed as a verification test before the CBU is issued, although no governmental regulatory bodies currently require CMV NAT for CBUs with negative CMV IgM results.

In the transplantation setting, CMV reactivation has been reported to affect treatment outcomes, resulting in morbidity and low survival rates, although it does not appear to affect graft outcomes (2). Although the posttransplantation CMV infection rate is higher for CBUs compared to alternative hematopoietic stem cell sources, the infection rate is comparable for donor-negative and recipient-positive procedures in the alternative transplantation setting (11). Therefore, it is important to confirm that CBUs are free from CMV and to test the mother's blood for anti-CMV IgM when screening for acute infection. Acute maternal infection increases the likelihood of congenital infection (10) and CMV DNAemia in the cord blood. However, testing the mother's blood for CMV IgM can be challenging, given the frequency of false-positive and false-negative test results (14,19). Furthermore, CMV acute infection only persists for 4–6 months, which may result in it being overlooked during the first trimester, and latent infection or reinfection is not frequently associated with positive IgM findings (6,19). Nevertheless, maternal CMV IgM testing is currently used to determine whether CBUs should be included in the pCBB inventory. In addition, CMV IgM-positive cord blood is subsequently subjected to CMV NAT, and only CBU with positive CMV NAT findings is excluded from the pCBB inventory (12,16). Because of some error, the KONOS database program does not import information of CBUs, which show CMV IgM positivity in the cord blood plasma and CMV NAT negativity in the cord blood leukocytes. Thus, such CBUs that conform to the Act on Cord Blood Management and Research (Cord Blood Act) (12) are neither searchable nor usable in the registry.

Recently, CBUs from IgM-positive mothers have been reported to have lower CD34⁺ counts compared to CBUs from IgM-negative mothers (6). Since this study included all donated cord blood without considering possible conforming units for inventory, the present study focused on units that passed the initial screening and reached the processing step. The study was designed to evaluate the association between CD34⁺ cell counts and maternal and cord blood CMV IgM status. This study would suggest robust principles for managing cord blood bank inventories.

Materials and Methods

Between May 11, 2006, and June 24, 2014, 49,730 CBUs were donated to ALLCORD. All mothers signed a document stating that the donation was voluntary; they agreed that the donated cord blood could be processed for allogeneic transplantation therapy and that units not suited for transplantation could be used for research. The initial screening criteria to select CBUs for subsequent processing were as follows: informed consent from the mother, free of medical illness, no perinatal issues in either the baby or the mother, and a TNC number of ≥8 × 10⁸ (8,16,22). On the basis of this screening, only 55% (28,000/49,730) of the donated CBUs were selected for processing. The processing procedure includes volume reduction and isolation of nucleated white blood cells to make CBUs, with simultaneous potency and safety tests using the retained sample. During the procedure, some CBUs cannot be cryopreserved because of damage to the bags, positive test findings for transmissible diseases in the maternal or cord blood plasma, and the absence of maternal blood for testing within 7 days of delivery, etc. Furthermore, exclusion from the pCBB inventory can occur after the cryopreservation stage if the baby's health is found to be poor due to a condition that would contraindicate transplantation.

The institutional review board (IRB) of Seoul's National University Boramae Hospital approved this study. All collection and processing of the CBUs were performed according to standard operating procedures (SOPs), as previously described (8).

Complete Blood and CD34⁺ Counts

The numbers of white blood cells (including neutrophils, lymphocytes, monocytes, eosinophils, and basophils), nucleated red blood cells, and platelets, as well as hemoglobin levels, were measured using the XE-2100 automated cell counter (Sysmex, Kobe, Japan). After CBU processing, the retained samples were used for counting the CD34⁺ cells. The CD34⁺ cell enumeration was performed using a commercially available kit (No. IM3630; BD Biosciences, San Jose, CA, USA) based on the single-platform International Society of Hematotherapy and Graft Engineering method (8). The results were expressed as cells/μl, and the total CD34⁺ cell count per CBU was calculated by multiplying the number of CD34⁺ cells by 20 (expressed as cells/unit) (the final volume is 25 ml, including 5 ml of a cryopreservant).

CMV Testing

The mothers' plasma and cord blood plasma were used for CMV IgM and or IgG testing, which was performed using the ARCHITECT CMV IgM assay and/or CMV IgG (Abbott Laboratories, Abbott Park, IL, USA), as previously described (7). To detect CMV DNA, we used the CMV Early Complete Kit (NEK015-02 and NEK015-05; Nanogen Advanced Diagnostics S.p.A., Buttigliera Alta, Italy) according to the manufacturer's instructions. After extracting DNA from the concentrated reference sample, nested polymerase chain reaction (PCR) was used to detect the CMV major immediate early antigen gene (HCMVUL123), with the human β globin gene serving as the internal control for inhibition. The limits of detection of 50% positivity were 1.2 gEq/reaction (95% confidence interval: 0.9–1.7 gEq/reaction) and 2.6 gEq/reaction (95% confidence interval: 2.0–3.5 gEq/reaction) for cellular samples and noncellular samples, respectively.

Statistical Analysis

Prior to analyses, exploratory data analyses were performed to identify and remove extreme outlier values. All data were expressed as median and interquartile range (IQR) for continuous variables or as a frequency and proportion for categorical variables. Intergroup comparisons of continuous variables were performed using the Mann–Whitney U test, whereas the Pearson's chi-square test or Fisher's exact test was used for categorical variables. Spearman's rank correlation analysis was applied to examine the relationship between two continuous variables. Additionally, a multivariable linear regression model was conducted to assess which parameters were related with the concentrations of log-transformed CD34⁺ cells. Values of p < 0.05 were considered statistically significant. All statistical analyses were performed using R software (version 3.1.2, http://www.r-project.org).

Results

CD34⁺ Cell Counts According to the Maternal CMV IgM Status

A total of 25,293 maternal plasma samples from the 28,000 processed CBUs were tested for CMV IgM, and the prevalence of CMV IgM in the mothers' plasma was 1.7% (441/25,293) (Table 1). When we analyzed the CBUs' CD34⁺ cell content, the median concentration of CD34⁺ cells was 72.0/μl (range: 1.20–435.0/μl; IQR = 95.7), and the total CD34⁺ count per unit was 1.44 × 10⁶ cells/unit (range: 0.02–8.70 × 10⁶ cells/U; IQR = 1.92). A significantly higher concentration and total number of CD34⁺ cells were observed in the CBUs from IgM-negative mothers compared to those in the CBUs of IgM-positive mothers (72.4/μl vs. 57.2/μl, respectively; p < 0.0001; 1.45 × 10⁶ cells/unit vs. 1.15 × 10⁶ cells/unit, respectively; p < 0.0001) (Fig. 1). There were no other significant differences among cord blood parameters with the maternal CMV IgM status, except the platelet counts, which were significantly higher among IgM-negative mothers (p = 0.0022). The CD34⁺ cell count of CBUs of CMV IgM positivity showed no difference from those of CMV IgM negativity in the cord blood plasma.

Figure 1.

Box plots of the concentrations (left): the median (interquartile range) CD34⁺ cells/μl for the cord blood of maternal CMV negative and positive status are 72.4 (96.0) and 57.2 (73.0), respectively. Box plots of the CD34 cells per unit (right): the median (interquartile range) CD34⁺ cells × 10⁶/unit for the cord blood of maternal CMV negative and positive status are 1.45 (1.92) and 1.15 (1.46), respectively.

Table 1.

Comparisons of Cord Blood Parameters According to the Maternal Cytomegalovirus IgM Status

	Maternal CMV IgM Status		*p Value
	Negative (n = 24,852)	Positive (n = 441)	*p Value
CD34⁺ cells/μl in reference sample	72.4 (96.0)	57.2 (73.0)	<0.0001
Total CD34⁺ cells (×10⁶/unit)	1.45 (1.92)	1.15 (1.46)	<0.0001
Volume of collected cord blood (ml)	108.19 (25.14)	108.67 (25.23)	0.9373
White blood cells/μl	10.7 (3.6)	10.5 (3.6)	0.0877
Hemoglobin (g/dl)	11.7 (1.7)	11.6 (1.6)	0.0563
Platelets/μl	209 (51)	202 (55)	0.0022

All data are presented as median (interquartile range). IgM, immunoglobulin M; CMV, cytomegalovirus. Intergroup comparisons of continuous variables were performed using the Mann–Whitney U test, whereas Pearson's chi-square test or Fisher's exact test was used for categorical variables. *The Mann–Whitney U test was performed.

Cord Blood-Related Characteristics Affecting CD34⁺ Cell Concentration

Not only the maternal CMV IgM status but also the babies' sex, body weight, the gestational age, and maternal age affect log-transformed CD34⁺ cell concentration (Table 2). Babies of female sex, heavy birth weight, and short gestational age as well as older maternal age showed a higher number of CD34⁺ cells.

Table 2.

Multiple Linear Regression Analysis of Cord Blood Characteristics on Log CD34⁺ Cells

Characteristics	Log(CD34⁺)
Characteristics	Slope	Standard Error	p Value
Maternal CMV IgM status
Positive versus negative	-0.194	0.044	<0.001
Sex
Male versus female	-0.069	0.011	<0.001
Neonatal weight (kg)	0.393	0.016	<0.001
Age of the mother (years)	0.010	0.002	<0.001
Gestational age (weeks)	-0.120	0.005	<0.001

Comparing the CMV IgM Findings From Maternal Plasma and Cord Blood Plasma

For comparison, we started with the 25,260 CBUs, which have CMV IgM results from the maternal plasma and cord blood plasma. The overall prevalence of maternal CMV IgM positivity was 1.7% (441/25,260), and the overall prevalence of cord blood CMV IgM positivity was 0.4% (112/25,260). However, when we compared the test results for the paired samples, the maternal CMV IgM status did not appear to fully reflect that of the cord blood (chi-square p < 0.001, κ value = 0.22) (Table 3). All the maternal CMV IgM⁺ samples were positive for CMV IgG.

Table 3.

Paired Analysis of Cytomegalovirus IgM and IgG in Maternal and Cord Blood Plasma

N = 25,260	Maternal CMV IgM Status		p Value
N = 25,260	Negative	Positive	p Value
Cord blood CMV IgM status
Negative	24,740	408	<0.001
Positive	79	33
Maternal CMV IgG status
Negative	359	0	0.0016
Positive	18,134	372

IgM, immunoglobulin M; CMV, cytomegalovirus.

Sensitivity and Specificity of the CMV IgM Status for Predicting Cord Blood CMV DNAemia

Among the 517 concentrated CBUs that tested positive for CMV IgM (maternal or cord blood plasma), only 0.58% CBUs (3/517) had positive CMV NAT findings. Neither the maternal CMV IgM positivity nor the cord blood CMV IgM positivity had sufficient sensitivity or specificity to predict cord blood CMV DNAemia (Table 4).

Table 4.

Sensitivity and Specificity of CMV IgM for Predicting Cord Blood CMV DNAemia

	CMV PCR in Cord Blood Leukocytes		Sensitivity (%) (95% CI)	Specificity (%) (95% CI)
	Negative	Positive	Sensitivity (%) (95% CI)	Specificity (%) (95% CI)
CMV IgM (cord blood)
Negative	404	2	33.333	78.599
Positive	110	1	(0–86.677)	(75.054–82.145)
CMV IgM (maternal)
Negative	78	1	66.667	15.175
Positive	436	2	(13.323–100)	(12.073–18.277)

IgM, immunoglobulin M; CMV, cytomegalovirus; PCR, polymerase chain reaction; CI, confidence interval.

Maternal CMV IgG Status

In the beginning, we did both CMV IgG and CMV IgM tests on maternal blood. Since nearly all the tested mothers were positive for CMV IgG, we changed our policy not to test CMV IgG for mothers after July 2011. The total number of CMV IgG tests done was 18,818. The CMV IgG seroprevalence among pregnant women was 98.1% (18,459/18,818), and the results according to the mothers' ages are listed in Table 5.

Table 5.

The Serologic Status of Pregnant Women Who Donated Cord Blood Between May 2006 and June 2014

Age (Years)	CMV IgM (Mother)			CMV IgG (Mother)
Age (Years)	Total Patients	Positive	Patients	Total Patients	Positive	Patients
13	1	0	0.0%	1	1	100.0%
14	1	0	0.0%	1	1	100.0%
15	3	0	0.0%	1	0	0.0%
16	7	1	14.3%	4	3	75.0%
17	10	2	20.0%	7	7	100.0%
18	31	0	0.0%	24	23	95.8%
19	44	0	0.0%	30	29	96.7%
20	58	0	0.0%	34	33	97.1%
21	95	4	4.2%	65	64	98.5%
22	142	3	2.1%	105	104	99.0%
23	194	5	2.6%	129	116	89.9%
24	310	6	1.9%	233	227	97.4%
25	488	11	2.3%	387	374	96.6%
26	941	14	1.5%	747	725	97.1%
27	1,549	34	2.2%	1,245	1,210	97.2%
28	2,302	40	1.7%	1,886	1,838	97.5%
29	2,724	45	1.7%	2,147	2,108	98.2%
30	3,052	53	1.7%	2,328	2,283	98.1%
31	2,963	51	1.7%	2,193	2,155	98.3%
32	2,622	55	2.1%	1,838	1,818	98.9%
33	2,191	40	1.8%	1,559	1,537	98.6%
34	1,762	23	1.3%	1,273	1,246	97.9%
35	1,269	24	1.9%	894	881	98.5%
36	911	15	1.6%	638	635	99.5%
37	608	4	0.7%	414	412	99.5%
38	389	2	0.5%	265	261	98.5%
39	252	3	1.2%	172	172	100.0%
40	115	3	2.6%	78	76	97.4%
41	89	2	2.2%	55	55	100.0%
42	49	0	0.0%	37	37	100.0%
43	26	0	0.0%	17	17	100.0%
44	8	0	0.0%	4	4	100.0%
45	5	0	0.0%	3	3	100.0%
46	1	0	0.0%	1	1	100.0%
48	2	0	0.0%	2	2	100.0%
49	1	0	0.0%	1	1	100.0%
Total	25,215	440	1.7%	18,818	18,459	98.1%

Discussion

The multivariate linear regression results between some cord blood characteristics and CD34⁺ cells in this report robustly support our previous report with 11,098 CBUs (8). The CD34⁺ cell concentration was positively correlated with female sex, heavy birth weight, and short gestational period (Table 2). The maternal age is the newly added parameter that positively affects CD34⁺ cell concentration.

When managing pCBBs, it is equally important to exclude CBUs that may contain transmissible disease and to store as many appropriate CBUs as possible. However, after infection, CMV is chronically present in the body, irrespective of whether active disease is observed. Furthermore, CMV resides in CD34⁺ bone marrow cells (3,21) and is known to cause myelosuppression (13). Interestingly, our findings regarding the lower number of CD34⁺ cells in CBUs from CMV IgM-positive mothers are in agreement with previously reported findings (6).

Therefore, the CMV IgM test is important in determining the safety and potency of CBUs, as the CD34⁺ cell count is the second most important factor (following TNC) when selecting CBUs for transplantation between HLA-matched patients.

In our experience, 65.5% (18,326/28,000) of the CBUs passed the initial safety screening. They were “the candidates” in the inventory irrespective of CMV-related tests (Table 6). Only 1.6% (307/18,326) of the candidates were subsequently excluded from the inventory because of the maternal CMV IgM status (12). This exclusion would create additional storage space for better CBUs such as those with higher CD34⁺ cell counts, which are expected to provide good clinical outcomes.

Table 6.

The Conditions of 9,674 (34.5%) Nonconforming Units for Storage Among 28,000 Donated Cord Blood

Reasons	A	B	C	D	E	F	G	H	I	J	K	L	M	N	O	P	Q	R	S	T	No.
A	V																				6,671
B		V																			844
C			V																		752
L												V									402
D				V																	384
N														V							209
O															V						93
P																V					72
Q																	V				40
M													V								38
R																		V			37
I									V												29
A, L	VV											V									24
S																			V		16
F, Q						VV											VV				14
T																				V	11
E					V																11
G, S							VV												VV		5
B, L		VV										VV									5
A, B	VV	VV																			3
A, P	VV															VV					2
A, C	VV		VV																		1
A, R	VV																	VV			1
A, S	VV																		VV		1
B, G, S		VV					VV												VV		1
	V					V												V
B, H, P, R		VV						VV								VV		VV			1
	VV						VV								VV		VV
B, K, P		VV									VV					VV					1
	V									V					V
B, R		VV																VV			1
I, T									VV										VV		1
J, L										VV		VV									1
L, O												VV			VV						1
L, P												VV				VV					1
Q, R																	VV	VV			1
Total																					9,674

The numbers of nonconforming conditions are presented with the V mark: V for 1, VV for 2, VVV for 3, and VVVV for 4. A, low total nucleated cells; B, microorganism growth; C, unable to separate the nucleated cell layer; D, other medical history of baby's relatives; E, low CD34 count; F, positive anti-hepatitis C virus (HCV) antibody in cord blood; G, positive syphilis test in cord blood; H, positive hepatitis B surface antigen (HBsAg) in cord blood; I, low cell viability; J, time elapsed from collection >36 h; K, usage for internal quality control; L, no maternal blood for testing; M, no informed consent for donation; N, used for transplantation; O, unacceptable maternal medical history; P, positive HBsAg in maternal sample; Q, positive anti-HCV antibody in maternal sample; R, alanine transaminase >65 U/L in the maternal sample; S, positive syphilis test in the maternal sample; T, positive anti-human T-lymphotropic virus antibody in the maternal sample.

If we exclude CBUs with CMV IgM positivity in cord blood plasma and CMV IgM negativity in the mother's plasma from the inventory irrespective of CMV NAT in cord blood leukocytes, 0.3% (54/18,326) of CBUs were excluded, which minimally affects the inventory pool in South Korea. The policy that excludes CBUs only after CMV NAT test in cord blood leukocytes could not be justified because of low sensitivity and specificity (Table 4). Thus, excluding CBUs with positive CMV IgM in the cord blood would be a better option in terms of ensuring the quality of CBUs. Consequently, regulations in the Cord Blood Act should be revised.

In this study, among the CBUs with a positive maternal or cord blood plasma CMV IgM status, positive CMV NAT findings in cord blood leukocytes were only observed in three cases (0.58%, 3/517) (Table 5). Among these three cases, only two (67%) were positive for CMV IgM in the maternal plasma, whereas the other case was positive for IgM in the cord blood plasma. None of the three babies exhibited symptoms until the final 6-month follow-up. Some researchers use CMV NAT in the cord blood for the second screening step for CMV in inventory (15). On the basis of our data, CMV NAT in cord blood leukocytes cannot substitute CMV serology of mother's plasma in the second screening step of CBU storage because of low sensitivity and specificity. CMV serology testing is the cheaper option, as the cost of CMV PCR is 1.6-fold higher than that of CMV IgM testing under the Korean health insurance coverage.

CMV NAT in cord blood leukocytes may still be appropriate as confirmatory testing when CBUs are issued for transplantation. Theiler et al. reported that two CBUs with positive CMV NAT findings came from mothers who were negative for CMV IgM (20). CMV DNAemia can be detected in CBUs from IgM-negative mothers; performing CMV NAT immediately before a CBU's release can provide evidence (per the European directive) that the CBU is free from CMV (4).

We have issued 192 CBUs for transplantation before December 2014, and none of these CBUs have exhibited positive CMV NAT findings. This may be explained by the fact that CMV IgG seroconversion only during pregnancy has been reported to increase the risk of congenital CMV infection by threefold compared to seroconversion before conception (3% vs. 1%) (5). Therefore, pregnant women with high seroprevalence and low primary infection during pregnancy have a lower risk of congenital CMV infection and a lower risk of CMV DNAemia in their cord blood.

In our institution, before the Cord Blood Act was instituted in 2011 (12,16), both CMV IgM and IgG testings were performed for both the maternal and cord blood plasma samples. When we analyzed that data, the CMV IgG seroprevalence among pregnant women was 98.1% (18,459/18,818), and the results according to the mothers' age are listed in Table 6. This seroprevalence was slightly higher than the reported seroprevalence of 91% in a Japanese study (18) and 93% in an American study (20). In terms of the primary infection, only 1.7% of our donors were positive for CMV IgM, which is similar to the reported rates of 2% (10/486) in the US (20) and South Korea (18), and lower than the reported rate of 5% (43/857) in Mexico (6,15,18). Therefore, a high seroprevalence and low primary infection in women of childbearing age may provide more suitable CBUs for storage (i.e., those negative for CMV) compared to CBUs from women with a low seroprevalence and acute primary infection. Additional experience is needed to determine the usefulness of CMV NAT at CBU release, particularly in the Korean population, where there is high maternal seroprevalence and a relatively low primary infection (18).

Unfortunately, we did not perform CMV IgG avidity testing to discriminate between recent and old infections (9,18). However, given the reported high avidity of CMV antibodies in the Korean population (18), a recent study has questioned the utility of the CMV IgM avidity test (9), which may reduce the usefulness of this test in the Korean population.

In conclusion, our data indicated that it is appropriate to use CMV serology and to exclude CBUs with CMV IgM positivity in cord blood plasma as well as CBUs with CMV IgM positivity in maternal plasma in South Korea, where women of childbearing age are highly immunized with CMV and sensitive enough CMV serology tests can be widely used. However, when releasing stored CBUs, using the CMV NAT to screen for CMV DNAemia may be appropriate until sufficient experience has been accumulated regarding the use of CBUs from mothers with a high CMV seroprevalence and low primary infection.

Footnotes

Acknowledgments

The authors would like to thank all of the staff of ALLCORD for their technical assistance and the Korean mothers who voluntarily donated cord blood to the public cord blood bank. This study was supported by a grant from the Korea Healthcare Technology R&D Project, Ministry of Health and Welfare, Republic of Korea (HI06C0874; A062260). S.S., E.Y.R., and J.H.Y. conceived and designed the experiments; S.S. and E.Y.R. performed the experiments; S.O. and E.Y.S. analyzed the data; S.S. and E.Y.R. contributed to the writing of the manuscript; and E.C.K. was the design and intellectual advisor on transmissible diseases. The authors declare no conflicts of interests.

References

Advisory Committee on the Microbiological Safety of Blood Tissues and Organs for Transplantation. Guidance on the microbiological safety of human organs, tissues and cells used in transplantation. United Kingdom; 2011 [accessed 2016 Aug 31]. https://www.gov.uk/government/uploads/system/uploads/attachment_data/file/215959/dh_130515.pdf

Beck

J. C.

; Wagner

J. E.

; DeFor

T. E.

; Brunstein

C. G.

; Schleiss

M. R.

; Young

J. A.

; Weisdorf

D. H.

; Cooley

; Miller

J. S.

; Verneris

M. R.

. Impact of cytomegalovirus (CMV) reactivation after umbilical cord blood transplantation. Biol. Blood Marrow Transplant. 16: 215–222; 2010.

Bego

M. G.

; St. Jeor

. Human cytomegalovirus infection of cells of hematopoietic origin: HCMV-induced immunosuppression, immune evasion, and latency. Exp. Hematol. 34: 555–570; 2006.

FACT and International Netcord Foundation. International standards for cord blood collection, banking, and release for administrations. 5th edition. Omaha, NE: FACT Accreditation Office; 2013 [accessed 2016 Aug 31]. https://www.factweb.org/forms/store/ProductFormPublic/fifth-edition-netcord-fact-international-standards-for-cord-blood-collection-banking-and-release-for-administration-free-download

Fowler

K. B.

; Stagno

; Pass

R. F.

. Maternal immunity and prevention of congenital cytomegalovirus infection. JAMA 289: 1008–1011; 2003.

Jaime-Perez

J. C.

; Colunga-Pedraza

J. E.

; Monreal-Robles

; Colunga-Pedraza

P. R.

; Mendez-Ramirez

; Salazar-Riojas

; Gomez-Almaguer

. Acute maternal cytomegalovirus infection is associated with significantly decreased numbers of CD34+ cells in umbilical cord blood. Blood Cells Mol. Dis. 49: 166–169; 2012.

Lagrou

; Bodeus

; Van Ranst

; Goubau

. Evaluation of the new architect cytomegalovirus immunoglobulin M (IgM), IgG, and IgG avidity assays. J. Clin. Microbiol. 47: 1695–1699; 2009.

Lee

H. R.

; Park

J. S.

; Shin

; Roh

E. Y.

; Yoon

J. H.

; Han

K. S.

; Kim

B. J.

; Storms

R. W.

; Chao

N. J.

. Increased numbers of total nucleated and CD34+ cells in blood group O cord blood: An analysis of neonatal innate factors in the Korean population. Transfusion 52: 76–81; 2012.

Lumley

; Patel

; Griffiths

P. D.

. The combination of specific IgM antibodies and IgG antibodies of low avidity does not always indicate primary infection with cytomegalovirus. J. Med. Virol. 86: 834–837; 2014.

10.

Manicklal

; Emery

V. C.

; Lazzarotto

; Boppana

S. B.

; Gupta

R. K.

. The “silent” global burden of congenital cytomegalovirus. Clin. Microbiol. Rev. 26: 86–102; 2013.

11.

Mikulska

; Raiola

A. M.

; Bruzzi

; Varaldo

; Annunziata

; Lamparelli

; Frassoni

; Tedone

; Galano

; Bacigalupo

; Viscoli

. CMV infection after transplant from cord blood compared to other alternative donors: The importance of donor-negative CMV serostatus. Biol. Blood Marrow Transplant. 18: 92–99; 2012.

12.

Ministry for Health, Welfare and Family Affairs.

Act on cord blood management and research.

Seoul, Korea; 2010 [accessed 2016 Aug 31]. http://elaw.klri.re.kr/eng_service/lawViewContent.do?hseq=23527

13.

Randolph-Habecker

; Iwata

; Torok-Storb

. Cytomegalovirus mediated myelosuppression J. Med. Virol.25(Suppl 2): S51–S56; 2002.

14.

Revello

M. G.

; Gerna

. Diagnosis and management of human cytomegalovirus infection in the mother, fetus, and newborn infant. Clin. Microbiol. Rev. 15: 680–715; 2002.

15.

Roback

J. D.

; Caliendo

A. M.

; Newman

J. L.

; Sgan

S. L.

; Saakadze

; Gillespie

T. W.

; Lane

T. A.

; Kurtzberg

; Hillyer

C. D.

. Comparison of cytomegalovirus polymerase chain reaction and serology for screening umbilical cord blood components. Transfusion 45: 1722–1728; 2005.

16.

Roh

E. Y.

; Lee

J. L.

; Yoon

J. H.

; Kim

S. Y.

; Lee

K. E.

; Song

D. Y.

; Kim

T. G.

; Shin

. Current status of cord blood banking during first two years of ‘National Government-Assigned Public Cord Blood Banks Operation’ in Korea. Stem Cell Rev. 10: 627–632; 2014.

17.

Schmitz

; Barrett

. Optimizing engraftment—Source and dose of stem cells. Semin. Hematol. 39: 3–14; 2002.

18.

Seo

; Cho

; Park

. Serologic screening of pregnant Korean women for primary human cytomegalovirus infection using IgG avidity test. Korean. J. Lab. Med. 29: 557–562; 2009.

19.

Stagno

; Tinker

M. K.

; Elrod

; Fuccillo

D. A.

; Cloud

; O'Beirne

A. J.

. Immunoglobulin M antibodies detected by enzyme-linked immunosorbent assay and radioimmunoassay in the diagnosis of cytomegalovirus infections in pregnant women and newborn infants. J. Clin. Microbiol. 21: 930–935; 1985.

20.

Theiler

R. N.

; Caliendo

A. M.

; Pargman

; Raynor

B. D.

; Berga

; McPheeters

; Jamieson

D. J.

. Umbilical cord blood screening for cytomegalovirus DNA by quantitative PCR. J. Clin. Virol. 37: 313–316; 2006.

21.

Wills

M. R.

; Poole

; Lau

; Krishna

; Sinclair

J. H.

. The immunology of human cytomegalovirus latency: Could latent infection be cleared by novel immunotherapeutic strategies? Cell. Mol. Immunol. 12: 128–138; 2015.

22.

Yoon

J. H.

; Oh

; Shin

; Park

J. S.

; Roh

E. Y.

; Song

E. Y.

; Park

M. H.

; Han

K. S.

. The minimum number of cord blood units needed for Koreans is 51,000. Transfusion 54: 504–508; 2014.

Excluding Anti-cytomegalovirus Immunoglobulin M-Positive Cord Blood Units Has a Minimal Impact on the Korean Public Cord Blood Bank Inventory

Abstract

Keywords

Introduction

Materials and Methods

Complete Blood and CD34+ Counts

CMV Testing

Statistical Analysis

Results

CD34+ Cell Counts According to the Maternal CMV IgM Status

Cord Blood-Related Characteristics Affecting CD34+ Cell Concentration

Comparing the CMV IgM Findings From Maternal Plasma and Cord Blood Plasma

Sensitivity and Specificity of the CMV IgM Status for Predicting Cord Blood CMV DNAemia

Maternal CMV IgG Status

Discussion

Footnotes

Acknowledgments

References

Complete Blood and CD34⁺ Counts

CD34⁺ Cell Counts According to the Maternal CMV IgM Status

Cord Blood-Related Characteristics Affecting CD34⁺ Cell Concentration