Review Article

Transfus Med Hemother 2021;48:306–315 Received: April 21, 2021

Accepted: July 27, 2021
DOI: 10.1159/000518782 Published online: September 8, 2021

Laboratory Monitoring of Mother, Fetus,

and Newborn in Hemolytic Disease of
Fetus and Newborn
Morten Hanefeld Dziegiel a, b Grethe Risum Krog a Anne Todsen Hansen a
Marianne Olsen c Birgitte Lausen c Lone Nikoline Nørgaard d
Thomas Bergholt e Klaus Rieneck a Frederik Banch Clausen a
aDepartment
of Clinical Immunology, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark;
bDepartment
of Clinical Medicine, University of Copenhagen, Copenhagen, Denmark; cDepartment of Pediatrics
and Adolescent Medicine, Copenhagen University Hospital, Rigshospitalet, Copenhagen, Denmark; dDepartment of
Obstetrics, Center of Fetal Medicine and Ultrasound, Copenhagen University Hospital, Rigshospitalet, Copenhagen,
Denmark; eDepartment of Obstetrics, Rigshospitalet, Copenhagen, Denmark

Keywords type is predicted by non-invasive genotyping based on cell-

Hemolytic disease of fetus and newborn · free DNA (RhD, K, Rhc, RhC, RhE, ABO), and serial monitoring
Alloimmunization · Cell-free DNA · Middle cerebral artery, of titer commences. Based on titers and specificity, monitor-
Peak systolic velocity · Next-generation sequencing ing with serial peak systolic velocity measurements in the
fetal middle cerebral artery to detect anemia will take place.
Intrauterine transfusion is given when fetal anemia is sus-
Abstract pected. Monitoring of the newborn by titer and survival of
Background: Laboratory monitoring of mother, fetus, and fetal red blood cells by flow cytometry will help predict the
newborn in hemolytic disease of fetus and newborn (HDFN) length of the recovery of the newborn.
aims to guide clinicians and the immunized women to focus © 2021 The Author(s).
on the most serious problems of alloimmunization and thus Published by S. Karger AG, Basel

minimize the consequences of HDFN in general and of anti-

D in particular. Here, we present the current approach of lab- Introduction
oratory screening and testing for prevention and monitoring
of HDFN at the Copenhagen University Hospital in Denmark. Alloimmunization is the process where an individual
Summary: All pregnant women are typed and screened in lacking a specific antigen of a blood group is exposed to
the 1st trimester. This serves to identify the RhD-negative the antigen and responds by producing specific antibod-
pregnant women who at gestational age (GA) of 25 weeks ies. Exposure might occur by transfusion with donor
are offered a second screen test and a non-invasive fetal RhD blood, or by accidental transfer of fetal red blood cells
prediction. At GA 29 weeks, and again after delivery, non- (RBCs) to the pregnant woman, such as fetomaternal
immunized RhD-negative women carrying an RhD-positive hemorrhage [1–3].
fetus are offered Rh immunoglobulin. If the 1st trimester Active transplacental transfer of maternal antibodies
screen reveals an alloantibody, antenatal investigation is ini- via the neonatal Fc receptor [4] will take place when the
tiated. This also includes RhD-positive women with alloanti- antibody production has switched from the initial IgM
bodies. Specificity and titer are determined, the fetal pheno- response to IgG. Transfer will accelerate in the 2nd and

[email protected] © 2021 The Author(s). Correspondence to:

www.karger.com/tmh Published by S. Karger AG, Basel Morten Hanefeld Dziegiel, morten.dziegiel @ regionh.dk
This is an Open Access article licensed under the Creative Commons
Attribution-NonCommercial-4.0 International License (CC BY-NC)
(http://www.karger.com/Services/OpenAccessLicense), applicable to
the online version of the article only. Usage and distribution for com-
mercial purposes requires written permission.
3rd trimester and can lead to hemolytic disease of fetus non-immunized women in GA 25 weeks. Detection of the
and newborn (HDFN) [5]. RHD gene is based on selective amplification of fetal
The essential clinical manifestation of HDFN is fetal DNA encoding the RHD gene.
and neonatal anemia. This is observed as erythroblastosis Selective amplification is, however, not reliably achiev-
fetalis, hepatic dysfunction leading to hypoalbuminemia, able for other blood group polymorphisms [9]. We exam-
ascites, hydrops fetalis, congestive heart failure, intrauter- ine women alloimmunized to the other prevalent anti-
ine growth retardation, abdominal and pericardial ede- gens (K, RhC, Rhc, RhE, and ABO) by non-invasive an-
ma, antenatal asphyxia, acute bilirubin encephalopathy, tenatal molecular diagnostics that amplify single
and kernicterus spectrum disorder [6]. The concentra- nucleotide variants (SNVs) potentially present in the cell-
tion of unconjugated bilirubin might surpass albumin free DNA (cfDNA), maternal as well as fetal. We describe
bilirubin binding capacity and translocate across the our clinically implemented non-invasive methods based
brain-blood barrier with subsequent accumulation in on cfDNA for 1st/2nd trimester determination of the
basal ganglia resulting in neuronal-cell death. Bilirubin genes encoding the clinically most important targets of
encephalopathy, or kernicterus, may lead to minor neu- alloantibodies (see Non-Invasive Prediction of Fetal K,
rodevelopmental disabilities, nerve deafness, spastic cere- RhC, Rhc, RhE, and ABO Blood Group).
bral palsy, or even death [7]. Prediction of other phenotypes based on antenatal ge-
Laboratory monitoring is an important tool to predict notyping has not yet been implemented in our laboratory.
a potential risk; but it cannot with certainty forecast clin- Instead, we do a paternal phenotype if antibodies to the
ical severity for the fetus. However, crucially, the labora- relevant antigens are available and make a statistical risk
tory setup removes the urgency of the diagnosis of HDFN assessment based on that. Blood group antibodies anti-A
and allows timely implementation of tested diagnostic and anti-B are responsible for neonatal hemolysis and hy-
and therapeutic measures. Cases of unpredicted HDFN perbilirubinemia, which in rare cases necessitate treat-
still occur and they are consequently not included in the ment with transfusion (see Maternal ABO Antibodies).
antenatal fetal screening program [2, 3, 8]. Flow cytometry (FC) is a useful method for small pop-
It is important at an early gestational age (GA) to de- ulation detection and quantification, for example, after
termine if the woman is RhD negative and thus is at risk intrauterine transfusion (IUT) for detection of fetal RBCs
of producing the most frequent alloantibody, anti-D, and and donor RBCs. Also, minute samples of fetal blood can
whether or not she is amenable to preventive treatment, be examined for multiple parameters improving labora-
Rh prophylaxis, later in pregnancy. At the same time all tory guidance (see FC in HDFN, Fetus and Newborn). In
women are screened for the presence of any alloantibody. this paper, we present the procedures related to labora-
In Denmark, if the woman is typed RhD negative and tory screening and monitoring in HDFN as currently per-
has no alloantibodies in the 1st trimester antibody screen, formed at the Copenhagen University Hospital, Rigshos-
she will be offered a non-invasive prediction of the fetal pitalet, in Denmark.
RhD blood group at GA 25 weeks at routine consultation
with her general practitioner. At the same time, an anti- Routine Blood Group Typing and Screening for
body screen test is performed. If the fetus is RhD positive Irregular Antibodies, Rh Prophylaxis
and the women is non-immunized, anti-D immunoglob- Blood samples from the 1st trimester initial pregnancy
ulin (RhIg) is offered at GA 29 weeks and again after de- consultation with the general practitioner are typed for
livery. This is described below (see Antenatal RHD ABO and RhD blood groups and an antibody screen is
Screening). performed. We use automated equipment and Capture-
In contrast, if a potentially harmful maternal antibody R® Ready-Screen (I and II) for detection of IgG antibod-
is detected in the screen in the 1st trimester, it is essential ies to RBC antigens. Typing identifies the RhD-negative
to determine if the present fetus carries the allele for the women who can develop anti-D antibodies. Antibody
antigen targeted by the maternal antibody. Only in this screening identifies those who have already developed al-
case is the fetus at risk of developing HDFN. In cases loantibodies in the 1st trimester, whether RhD positive or
where the fetus is at risk, intensified pregnancy monitor- RhD negative.
ing and treatment by transfusion can be instituted, and in At GA 25 weeks, RhD-negative women with a negative
cases where risk of HDFN due to known alloantibodies 1st trimester antibody screen are offered routine non-in-
can be excluded, a less intensive and financially less bur- vasive fetal antenatal RHD screening, and the repeated
densome approach can be taken. Also, much anxiety routine antibody screening is offered only to RhD-nega-
from the prospective parents can be avoided. If the wom- tive women (see Antenatal RHD Screening). At GA 29
an is alloimmunized to the RhD antigen when tested in weeks, the nonimmunized pregnant woman is offered in-
the 1st trimester in the antibody screen, we immediately tramuscular injection of 250–300 μg RhIg by the midwife
use the same antenatal RHD screening assay as we use for if RHD is detected in the cell-free fetal DNA (cffDNA)

Laboratory Monitoring of HDFN: The Transfus Med Hemother 2021;48:306–315 307

Copenhagen Approach DOI: 10.1159/000518782
from plasma. The 250–300 μg RhIg injection is repeated heterozygous for the target antigen, and the same reagent
within 72 h after delivery. Investigation for fetomaternal in-house single donation is used throughout pregnancy
hemorrhage and quantification by flowcytometry is only to reduce variation. An increase of two dilution steps or
made if hemorrhage is suspected. more is considered a significant development that de-
serves special attention by the clinician and possibly re-
Investigating and Monitoring Irregular Antibodies ferral to the fetal medicine center. Any two-step or more
For alloimmunized women, identification of the target deviation is investigated in the laboratory by comparison
antigen of the alloantibody will be conducted to provide with a side-by-side analysis of previous samples. The
further information on the potential clinical impact of the Working Standard Anti-D is included every day as a con-
alloimmunization. Maternal antibodies targeting a dis- trol.
tinct blood group antigen often lead to a known distinct An antibody screen represents a “snapshot status” of
pattern of clinical manifestations. This knowledge guides the antibody content of the pregnant woman at the time
the planning of the laboratory monitoring and the fetal of sampling; the titer might increase rapidly because of
specialist surveillance. continuing exposure to fetal or donor RBCs. Within days
For the antibody identification we use 11 different re- additional antibodies may also develop. Therefore, a se-
agent in-house single donation glycerol frozen-thawed rial monitoring is important.
RBCs and anti-IgG column agglutination technique
(CAT). The woman’s own RBCs are included in the pan- Antenatal RHD Screening
el to distinguish allo- and autoantibodies. If Rh antibodies As part of a targeted RhIg prophylaxis program for
are suspected the examination is extended with a panel of non-immunized RhD-negative women, knowledge of the
papain-treated RBCs. Per definition, for an alloantibody fetal RhD type helps restrict prophylaxis to those women
to be present, a phenotype of the woman’s own RBCs only who carry an RhD-positive fetus [12, 13]. This re-
should demonstrate absence of the target antigen of the striction avoids superfluous exposure to prophylaxis in
alloantibodies. women carrying an RhD-negative fetus and reduces the
Some antibodies have empirically been found to be of overall use of RhIg, which is a limited resource [14, 15].
no clinical significance (anti-N, -Lea, -Leb, -A1, -IH, -I), The fetal RhD status is predicted by analysis of cffDNA
whereas others are known to be of potential dire conse- in the maternal plasma that also contains maternally de-
quences (anti-K, -c) and referral to a fetal medicine center rived cell-free DNA. Presence of the fetal RHD gene indi-
should be considered regardless of titer, and yet another cates that the fetus is RhD positive. Since the first reports
group (anti-D, -C, -E, -e, -Cw, -Kpa, -Kpb, -k, -Jka, -Jkb, of cell-free fetal RHD in maternal plasma [16, 17], non-
-Fya, -Fyb, -S, -s, -Wra, -M, -P1, -Lua, and -Lub) is referred invasive fetal RHD genotyping has become highly inte-
to a fetal medicine center if a titer above 16 is measured. grated into clinical medicine, and its accurate perfor-
At GA 25 and 32 weeks, the alloimmunized woman is mance has been covered comprehensively in the litera-
routinely examined with a measurement of titer and with ture [12, 13, 18–24].
a screening for additional antibodies. A titer above 16 for As an antenatal screening to guide RhIg prophylaxis,
the latter group of antibodies is empirically determined non-invasive prenatal testing of fetal RHD has been intro-
as the threshold value indicating increased risk of HDFN duced as a nationwide clinical service in several European
and warrants closer surveillance by a fetal medicine spe- countries [16–25]. Evaluations of national programs have
cialist with serial Doppler ultrasound measurements of demonstrated high test accuracy, with sensitivities of
the peak systolic velocity (PSV) in the fetal middle cere- >99.9% around 25 weeks of gestation and >99% from GA
bral artery (MCA) [10, 11]. 10 weeks [13]. Recent recommendations for validation
Semi-quantification of alloantibody is done by a serial and quality assurance of fetal RHD genotyping have been
2-step dilution of plasma in saline followed by examina- prepared [25].
tion with CAT. The titer is defined as the inverse of the The Copenhagen setup for antenatal RHD screening
highest dilution still demonstrating a positive reaction of has been described in detail [26–28]. Briefly, blood sam-
at least a weak reaction (w+). Determination of titer has ples are taken by the general practitioner at GA 25 weeks.
an intra-laboratory variation of ±1 titer step and a CV% Blood samples arrive at the laboratory after an average of
of 12.4 calculated based on 100 manually titrations con- 4 days in transport (up to 7 days are accepted). Plasma is
ducted in the routine laboratory during 6 months by the separated and DNA is extracted from 1 mL of plasma.
routine staff. The material used for this validation was the Eluted DNA is tested by real-time PCR targeting RHD ex-
Working Standard Anti-D for assuring operator and test ons 7 and 10 in a duplex manner with the same dye, which
performance (The National Institute for Biological Stan- increases the analytical sensitivity [29]. The RHD PCR is
dards and Control [NIBSC], Potters Bar, UK; Code No. sensitive enough to detect one genome equivalent (geq)
07/304). The reagent RBCs we use to determine titer are per PCR [30], and the overall detection limit of the setup

308 Transfus Med Hemother 2021;48:306–315 Dziegiel/Krog/Hansen/Olsen/Lausen/

DOI: 10.1159/000518782 Nørgaard/Bergholt/Rieneck/Clausen
is 6 geq per mL. Total DNA is targeted by GAPDH as qual- data analysis we currently employ is two-pronged: one
ity control for sample handling and DNA purification. analysis of the fastq sequences is done using FastQC soft-
Based on the amplification of fetal RHD exon targets, ware and another analysis is performed using simple
the sample is predicted to be positive or negative, or in- string searches with grep in a Linux formatted PC. We do
conclusive. If positive or inconclusive, the pregnant wom- not perform alignment-based analysis.
an is recommended to receive prophylaxis. Some mater- Even though the prediction of the different blood
nal RHD variants may give a positive result masking the groups: K, RhC, Rhc, RhE, and ABO is based on the same
detection of the fetal RHD. In such cases, the result is de- generic method, there are important differences in re-
termined to be inconclusive, and prophylaxis is offered to spect to primer design and data analysis. For instance, in
the woman. As an outcome of implementing nationwide the case of Rhc prediction, background reads from high-
non-invasive prenatal testing for fetal RHD, unnecessary ly homologous sequences of RHD may complicate the
antenatal prophylaxis is avoided in 97.3–99.6% of the prediction. The ABO prediction requires the combined
women carrying an RhD-negative fetus [28, 31, 32]. Fu- results of two primer sets for an antigen prediction.
ture applications may include expanding the targeted ap- After implementation of the two-pronged data analy-
proach to RhD-negative pregnant women with early sen- sis, we have not yet had any discordant results from a
sitizing events [33]. small cohort of samples. As a postnatal blood group was
not determined in many cases, a significant number of
Non-Invasive Prediction of Fetal K, RhC, Rhc, RhE samples have not been used for formally validating the
and ABO Blood Group results of the prenatal fetal blood group prediction. De-
We have recently reported a procedure based on next- velopment of laboratory methods, validation as well as
generation sequencing (NGS) analysis of PCR-amplified continuous quality control, is dependent on meticulous
cfDNA from maternal plasma for prediction of the fetal and continuous contribution from laboratories and clin-
blood group [34–37]. As some fetuses may die from ics. O’Brien et al. [38] have used digital PCR for fetal
HDFN as early as GA 18 weeks, it is necessary to be able blood group prediction, and Orzińska et al. [39] have also
to predict the fetal blood group early in pregnancy. We used NGS for fetal blood group prediction.
use this general approach to predict fetal K, RhC, Rhc,
RhE, and ABO blood groups in cases with a risk of HDFN
due to maternal production of the corresponding anti- Maternal ABO Antibodies
bodies [34–37].
The NGS based analysis can detect the presence or ab- ABO incompatibility is now the most prevalent cause
sence of alleles encoding incompatible antigens on the of HDFN with hyperbilirubinemia in developed coun-
fetal RBCs. NGS is a powerful technology that enables the tries due to the success of Rh prophylaxis [2, 6, 40]. A re-
parallel sequencing of many million DNA sequences. We cent Danish study found ABO incompatibility in 15 of 21
use this technology in a very simple approach: cfDNA is cases with total serum bilirubin ≥600 μmol/L, comprising
purified from 4 mL of maternal plasma and after PCR a significant risk of kernicterus spectrum disorder [41].
amplification of the genetic basis of the blood group, the Furthermore, rare cases of fetal hemolysis, anemia, and
PCR product is sequenced to great depth. The number of hydrops fetalis caused by ABO antibodies have been de-
times that the blood group SNV in question occurs is scribed [42, 43].
counted and the relative frequency of the SNV, exceeding Currently, we do not have a systematic screening pro-
the background threshold, will be the basis of the predic- cedure for maternal ABO antibodies harmful to the fetus
tion. As there are some background reads due to errors of and newborn [44–46]. Maternal anti-A and anti-B IgG
PCR amplification and sequencing, this threshold is im- titers are predictive of neonatal requirement for treat-
portant to determine empirically. The background is re- ment of hyperbilirubinemia [47, 48]. However, we found
markably low with an empirical threshold for a positive the positive predictive values both in the 1st trimester
sample of approximately 0.05% positive reads. (65%) and perinatally (73–76%) to be too low to be used
Preanalytical conditions are important to address as clinically for routine screening and we aim for enhance-
these NGS-based assays rely on maintaining the in vivo ment of predictive values [49] by on-going research.
ratio of fetal versus maternal SNVs. Thus, it is important We have described the use of two antibody screening
to ensure that maternal cells do not contribute DNA after methods: (i) solid phase red cell adherence assay (SPR-
blood sampling. Taking blood samples in Streck tubes is CA) only detecting IgG anti-A and anti-B and (ii) manu-
therefore highly recommended. Some factors are impor- al anti-IgG CAT detecting both IgG and IgM reacting at
tant, such as keeping the amplicons short and keeping 37°C. The two methods yielded comparable results. SPR-
spurious amplification to a minimum. The volume of CA is most suitable for batch analysis, whereas CAT is
plasma interrogated is approximately 1 mL. Finally, the amenable for single sample analysis.

Laboratory Monitoring of HDFN: The Transfus Med Hemother 2021;48:306–315 309

Copenhagen Approach DOI: 10.1159/000518782
 Standard infant transfusion practice in our health care anemia in the newborn. Initially, a hepatic cause, Alagille
region is ABO-identical RBCs. Therefore, in addition to syndrome, was suspected. However, only HDFN was
an antibody screening test for irregular antibodies, we found. To substantiate the diagnosis of HDFN, we used
also perform a determination of regular anti-ABO anti- FC to determine a series of percentages of newborn E-
bodies of the IgG class of the incompatible newborn to be positive RBC.
transfused. Detection of IgG anti-A and anti-B is followed Newborn E-positive RBCs were identified in FC by re-
by determination of the maternal IgG anti-A and anti-B acting RBCs with reagent anti-E followed by anti-human
titer. This is likely to lead to identification of more wom- IgG conjugated to a fluorophore, as previously described
en with high-titer IgG anti-A and anti-B. [52]. Total hemoglobin and reticulocytes were measured
with a hematology analyzer. To corroborate E-positive
Laboratory Monitoring, anti-A and anti-B results, we supplemented with measurement by FC of fe-
In pregnancies with identified maternal high-titer IgG tal hemoglobin (HbF) and obtained similar results (data
anti-A and anti-B or a history of a previous pregnancy not shown) [50].
where maternal anti-A/B was responsible for HDFN, the We observed a close correlation between the waning
anti-A and anti-B IgG titer is determined in the 1st tri- of the allo-anti-E and an increasing survival of fetal RBCs.
mester as well as at GA 32 weeks. For the methods de- Figure 2 demonstrates that the effects of the anti-E is re-
scribed a common cut-off value of 512 was initially found flected in the newborn RBCs for more than 72 days and
for anti-A/B [49]. However, additional studies (in prepa- that survival of the newborn is dependent on transfusion
ration) showed that distinct cut-off values for anti-A and therapy during the first 47 days.
anti-B ​​increased accuracy. Therefore, we now apply a Generally, if a fetus has received intrauterine transfu-
cut-off value of 512 for maternal anti-A and 256 for anti- sion it is possible to monitor the percentage of fetal versus
B. The cut-off value is used for recommendation of ante- donor RBCs. This can be done in several ways depending
natal non-invasive fetal ABO blood group prediction and on the specific situation, but typically we use as a marker
for fetal monitoring by ultrasound of MCA-PSV in case the antigen targeted by the maternal antibodies, positive
of incompatible antigens on fetal RBCs. The flow chart RBCs are fetal, antigen-negative RBCs are from the donor
presented in Figure 1 presents the complete laboratory due to the use of compatible blood. The percentage of fe-
monitoring of HDFN. tal RBCs is also measurable with the marker HbF [50].

FC in HDFN, Fetus and Newborn Determining the Actual Clinical Course, Doppler
Agglutination techniques are informative in most cas- Ultrasonography MCA-PSV for Non-Invasive
es, but by supplementing with FC more detailed and Prediction of Fetal Anemia
semiquantitative information [50] can be produced also Maternal alloantibodies and fetal expression of the
in unexpected urgent cases of suspected HDFN where di- corresponding RBC antigen is the prerequisite for HDFN.
agnosis is initially uncertain. Determination of fetal and However, a large variation in clinical impact is observed
newborn antigens and direct antiglobulin test (DAT)- with identical laboratory findings. Even in the same wom-
positive RBCs can be made impossible or inconclusive by an clinical variation occurs from one antigen-positive fe-
access to a limited volume of sample, small surviving pop- tus to another despite an unchanged alloantibody titer
ulations of fetal cells after multiple IUTs, and due to weak [53]. Supplementary modalities of monitoring are needed
fetal expression of antigens [51]. FC enables quantifica- to determine the actual clinical consequence of the allo-
tion of subpopulations, for example several populations immunization.
of distinct RBC phenotype in cases of mixed populations Measurement of MCA-PSV is the golden standard for
of donor and patient cells, enabling measurement of the non-invasive prediction of fetal anemia. Mari et al. [11]
survival of the infant’s own RBCs, as well as donor RBCs. showed that a cut-off of 1.5 multiples of median on Dop-
In Figure 2, we present an example of serial monitor- pler ultrasound measurement of MCA-PSV has 100%
ing of various parameters of a severely anemic newborn, sensitivity with a false-positive rate of 12% in the predic-
with hemoglobin (Hb) at birth of 6.3 g/dL (3.9 mmol/L). tion of moderate to severe anemia in the non-hydropic
The RhD-positive woman unexpectedly delivered an ane- fetus. Timely identification of significant fetal anemia is
mic infant in GA 38 weeks. Upon investigation after de- the basis for therapeutic intervention with intrauterine
livery, the mother had an allo-anti-E, titer of 2,048. The blood transfusion or delivery, depending on GA and
anti-E developed between the 1st trimester antibody thereby preventing fetal demise.
screening and delivery. The newborn was DAT positive.
Immediately after birth the newborn was given a trans- The Newborn in HDFN
fusion with compatible donor RBCs, and again on day 10 When the fetus becomes a newborn it might still be suf-
and day 26 in accordance with guidelines for treatment of fering from anemia and the other pathophysiologic conse-

310 Transfus Med Hemother 2021;48:306–315 Dziegiel/Krog/Hansen/Olsen/Lausen/

DOI: 10.1159/000518782 Nørgaard/Bergholt/Rieneck/Clausen
Fig. 1. The flow chart illustrates the use of individual components screened. Alternative timing of examination for both RhD-nega-
of the laboratory HDFN monitoring of all pregnant women. RhD- tive and positive women is followed if the clinician decides so.
positive and RhD-negative women are screened for irregular anti- * Shows a potential trigger for conducting antenatal antigen pre-
bodies against RBC antigens in the 1st trimester. Antibodies to diction. The specific criteria for antenatal antigen prediction are:
ABO blood group antigens are only included in the examination if C, c, K, or D at titer ≥1; E at titer >1; A at titer ≥512; B at titer ≥256.
supplementary information gives an indication to do so. RhD-pos- For blood groups A and B, also HDFN due to anti-A or anti-B in
itive women who test negative at this first antibody screening are a previous pregnancy gives an indication for antigen prediction.
not examined later. All RhD-negative women are re-tested for an- ** Shows a potential trigger for referral to the fetal medicine center.
tibodies at GA 25 weeks, and the antenatal RHD screening based The specific criteria for referral are: anti-D, -C, -E, -e, -Cw, -Kpa,
on cffDNA is performed. RHD screening showing an RHD-posi- -Kpb,- k, -Jka, -Jkb, -Fya, -Fyb,-S, -s, -Wra, -M, -P1, -Lua, -Lub
tive fetus leads to the administration of Rh prophylaxis at GA 29 titer >16, and anti-K, -c titer ≥1. *** Designates postnatal monitor-
weeks and subsequently also the postnatal prophylaxis. Immu- ing of the newborn with titer, serological antigen detection, and
nized RhD-negative women are screened for antibodies again at flow cytometric quantification of fetal and donor RBCs. A black
GA 32 weeks. At GA 25 weeks and GA 32 weeks, RhD-positive box designates an analysis, a grey box designates a result, and a
women with irregular antibodies detected in the first trimester are white box designates Rh prophylaxis.

Laboratory Monitoring of HDFN: The Transfus Med Hemother 2021;48:306–315 311

Copenhagen Approach DOI: 10.1159/000518782
Fig. 2 Serial determinations of parameters for the first 72 days after determined by 2-step dilutions and analysis in CAT; total Hb (g/
the birth of the newborn. On day 19, less than 0.6 g/dL (0.4 mmol/L) dL) from donor RBCs and newborn RBCs measured in a hematol-
of the total Hb 7.1 g/dL (4.3 mmol/L) was from endogenous E- ogy analyzer; E-positive Hb was calculated as the E-positive RBC
positive RBCs. Only after 47 days, at a titer of anti-E below 4, did fraction of total RBCs measured by FC multiplied with the total
the E-positive RBCs survive and total Hb started to increase in Hb; reticulocytes were measured in a hematology analyzer. In-
parallel. Reticulocytes transiently decreased as a response to trans- formed consent was obtained from the parents of the infant. The
fusions, and decreased again after day 61, when E-positive RBCs study followed the guidelines of the institutional review board of
were being normalized. The newborn parameters were: anti-E titer Copenhagen University Hospital.

quences of the persisting maternal antibody [54] present in supplying K-negative RBC components for premeno-
the newborn. In most cases the fetal RBCs will carry ma- pausal women in Denmark. Matching has been extended
ternal antibodies detectable by the DAT. We routinely de- to routinely encompass Rhc and E in some countries [3].
termine the titer of free alloantibody in the plasma of the A study on the effect of matching donor and recipient
newborn and determine the fetal blood group antigen tar- in IUT indicates that an efficient prevention of alloim-
geted by the maternal antibody. The latter is routinely done munization (64%) can be achieved by an extended phe-
to assess the quality of laboratory work. FC-based mea- notypic match: C, c, E, K, Fya, Jka, S [55]. Another study
surement of fetal versus donor cells is decided in each case. in ordinary transfusion recipients demonstrated that
The laboratory should be aware of the importance of matching for C, c, E, K, Jka could prevent 78% of immu-
information being shared with the team of neonatologists nizations, and enhanced matching for C, c, E, K, Fya, Jka,
providing postnatal care for the newborn. It should be Cw improved prevention to 83.4% of immunizations [56].
remembered that laboratory investigation of the mother We have implemented matching for IUT for C, c, E, K,
might still be relevant and can yield valuable information, Fya, Jka with a pragmatic view for the available supply.
for example examination for antibodies, phenotype, de- However, our extensive genotyping of donors helps mak-
termination of titers, fetomaternal hemorrhage, especial- ing matches possible by access to ample donor genotype
ly in the RhD-positive women who have not been tested information [57].
since the 1st trimester. Platelet transfusion seems to be a source of alloimmu-
nization that could be taken into consideration. Small
Further Preventive Measures to Avoid amounts of RBCs in the platelet component are enough
Alloimmunization to immunize. We administer RhIg if, for logistical rea-
Prevention of alloimmunization due to transfusion in sons, D-positive platelet or plasma components must be
girls and women of premenopausal age, or under the age given to a female RhD-negative recipient of premeno-
of 50 years, has been implemented in some countries by pausal age.
matching a limited number of RBC antigens. Basic match-
ing of ABO and RhD blood groups is supplemented by

312 Transfus Med Hemother 2021;48:306–315 Dziegiel/Krog/Hansen/Olsen/Lausen/

DOI: 10.1159/000518782 Nørgaard/Bergholt/Rieneck/Clausen
 Perspectives for Optimization and Future It has also been attempted to interfere with an estab-
Developments lished immune response by administration of peptides
Several studies have addressed the feasibility of screen- derived from the antigen to end the active production of
ing all pregnant women for irregular antibodies in the 3rd antibody [63, 64]. Another approach is administration of
trimester, not only RhD-negative women. First trimester non-destructive antibodies competing for the antigen to
screening of all pregnant women is already implemented the alloimmunized woman [65, 66].
in many health care systems. A 3rd trimester repeated
screening of Rhc-negative women has been proposed. Fo-
cusing on individuals with a high risk of immunization Conflict of Interest Statement
would enhance cost benefit in comparison with screening
The authors declare no conflicts of interest.
all women [3, 8].
Some individuals develop alloantibodies after alloan-
tigen exposure whereas others can be transfused repeat-
edly without being alloimmunized [58, 59]. Elucidation Funding Sources
of the genetic background for individual propensity to The authors have not received any funding relevant for this
develop alloantibodies as well as the genetic background manuscript.
for regulation of quantities of antibodies produced has
been attempted by several groups [60–62]. Access to this
information for pregnant women would potentially add Author Contributions
useful guidance to the clinical risk assessment of a spe-
cific woman. All authors contributed to writing of the text. All authors have
read and accepted the final version of the manuscript.

References
1 Bowman JM, Pollock JM, Penston LE. Feto- 9 Finning K, Martin P, Summers J, Daniels G. 17 Lo YM, Hjelm NM, Fidler C, Sargent IL, Mur-
maternal transplacental hemorrhage during Fetal genotyping for the K (Kell) and Rh C, c, phy MF, Chamberlain PF, et al. Prenatal diag-
pregnancy and after delivery. Vox Sang. 1986; and E blood groups on cell-free fetal DNA in nosis of fetal RhD status by molecular analysis
51(2):117–21. maternal plasma. Transfusion. 2007 Nov; of maternal plasma. N Engl J Med. 1998 Dec
2 de Haas M, Thurik FF, Koelewijn JM, van der 47(11):2126–33. 10;339(24):1734–8.
Schoot CE. Haemolytic disease of the fetus 10 Moise KJ. Red blood cell alloimmunization in 18 Alshehri AA, Jackson DE. Non-invasive pre-
and newborn. Vox Sang. 2015 Aug; 109(2): pregnancy. Semin Hematol. 2005 Jul; 42(3): natal fetal blood group genotype and its ap-
99–113. 169–78. plication in the management of hemolytic
3 Koelewijn JM, Vrijkotte TG, de Haas M, van 11 Mari G, Deter RL, Carpenter RL, Rahman F, disease of fetus and newborn: systematic re-
der Schoot CE, Bonsel GJ. Risk factors for the Zimmerman R, Moise KJ Jr., et al. Noninva- view and meta-analysis. Transfus Med Rev.
presence of non-rhesus D red blood cell anti- sive diagnosis by Doppler ultrasonography of 2021;35(2):85–94.
bodies in pregnancy. BJOG. 2009 Apr;116(5): fetal anemia due to maternal red-cell alloim- 19 Clausen FB. Integration of noninvasive pre-
655–64. munization. Collaborative Group for Dop- natal prediction of fetal blood group into clin-
4 Firan M, Bawdon R, Radu C, Ober RJ, Eaken pler Assessment of the Blood Velocity in Ane- ical prenatal care. Prenat Diagn. 2014 May;
D, Antohe F, et al. The MHC class I-related mic Fetuses. N Engl J Med. 2000 Jan 6;342(1): 34(5):409–15.
receptor, FcRn, plays an essential role in the 9–14. 20 Clausen FB, Damkjær MB, Dziegiel MH.
maternofetal transfer of gamma-globulin in 12 Clausen FB. Lessons learned from the imple- Noninvasive fetal RhD genotyping. Transfus
humans. Int Immunol. 2001 Aug; 13(8): 993– mentation of non-invasive fetal RHD screen- Apher Sci. 2014 Apr;50(2):154–62.
1002. ing. Expert Rev Mol Diagn. 2018 May; 18(5): 21 Daniels G, Finning K, Martin P, Massey E.
5 Simister NE, Mostov KE. An Fc receptor 423–31. Noninvasive prenatal diagnosis of fetal blood
structurally related to MHC class I antigens. 13 van der Schoot CE, de Haas M, Clausen FB. group phenotypes: current practice and fu-
Nature. 1989 Jan 12;337(6203):184–7. Genotyping to prevent Rh disease: has the ture prospects. Prenat Diagn. 2009 Feb;29(2):
6 Le Pichon JB, Riordan SM, Watchko J, Shap- time come? Curr Opin Hematol. 2017 Nov; 101–7.
iro SM. The neurological sequelae of neonatal 24(6):544–50. 22 Runkel B, Bein G, Sieben W, Sow D, Polus S,
hyperbilirubinemia: definitions, diagnosis 14 Bills VL, Soothill PW. Fetal blood grouping Fleer D. Targeted antenatal anti-D prophy-
and treatment of the kernicterus spectrum using cell free DNA – an improved service for laxis for RhD-negative pregnant women: a
disorders (KSDs). Curr Pediatr Rev. 2017; RhD negative pregnant women. Transfus systematic review. BMC Pregnancy Child-
13(3):199–209. Apher Sci. 2014 Apr;50(2):148–53. birth. 2020 Feb 7;20(1):83.
7 Luban NL. Hemolytic disease of the newborn: 15 Kent J, Farrell AM, Soothill P. Routine ad- 23 Yang H, Llewellyn A, Walker R, Harden M,
progenitor cells and late effects. N Engl J Med. ministration of Anti-D: the ethical case for of- Saramago P, Griffin S, et al. High-throughput,
1998 Mar 19;338(12):830–1. fering pregnant women fetal RHD genotyp- non-invasive prenatal testing for fetal rhesus
8 Koelewijn JM, Vrijkotte TG, van der Schoot ing and a review of policy and practice. BMC D status in RhD-negative women: a system-
CE, Bonsel GJ, de Haas M. Effect of screening Pregnancy Childbirth. 2014 Feb 25;14:87. atic review and meta-analysis. BMC Med.
for red cell antibodies, other than anti-D, to 16 Faas BH, Beuling EA, Christiaens GC, von 2019 Feb 14;17(1):37.
detect hemolytic disease of the fetus and new- dem Borne AE, van der Schoot CE. Detection
born: a population study in the Netherlands. of fetal RHD-specific sequences in maternal
Transfusion. 2008 May;48(5):941–52. plasma. Lancet. 1998 Oct 10;352(9135):1196.

Laboratory Monitoring of HDFN: The Transfus Med Hemother 2021;48:306–315 313

Copenhagen Approach DOI: 10.1159/000518782
24 Zhu YJ, Zheng YR, Li L, Zhou H, Liao X, Guo 35 Rieneck K, Clausen FB, Dziegiel MH. Next- 48 Chen JY, Ling UP. Prediction of the develop-
JX, et al. Diagnostic accuracy of non-invasive generation sequencing for antenatal predic- ment of neonatal hyperbilirubinemia in ABO
fetal RhD genotyping using cell-free fetal tion of KEL1 blood group status. Methods incompatibility. Zhonghua Yi Xue Za Zhi.
DNA: a meta analysis. J Matern Fetal Neona- Mol Biol. 2015;1310:115–21. 1994 Jan;53(1):13–8.
tal Med. 2014 Dec;27(18):1839–44. 36 Rieneck K, Clausen FB, Dziegiel MH. Nonin- 49 Krog GR, Donneborg ML, Hansen BM, Lo-
25 Clausen FB, Hellberg A, Bein G, Bugert P, vasive antenatal determination of fetal blood renzen H, Clausen FB, Jensen KV, et al. Pre-
Schwartz D, Drnovsek TD, et al. Recommen- group using next-generation sequencing. diction of ABO hemolytic disease of the new-
dation for validation and quality assurance of Cold Spring Harb Perspect Med. 2015 Oct 28; born using pre- and perinatal quantification
non-invasive prenatal testing for foetal blood 6(1):a023093. of maternal anti-A/anti-B IgG titer. Pediatr
groups and implications for IVD risk classifi- 37 Rieneck K, Egeberg Hother C, Clausen FB, Ja- Res. 2021;90(1):74–81.
cation according to EU regulations. Vox Sang. kobsen MA, Bergholt T, Hellmuth E, et al. 50 Dziegiel MH, Nielsen LK, Berkowicz A. De-
2021. doi: 10.1111/vox.13172. Next generation sequencing-based fetal abo tecting fetomaternal hemorrhage by flow cy-
26 Clausen FB, Christiansen M, Steffensen R, blood group prediction by analysis of cell-free tometry. Curr Opin Hematol. 2006 Nov;
Jørgensen S, Nielsen C, Jakobsen MA, et al. DNA from maternal plasma. Transfus Med 13(6):490–5.
Report of the first nationally implemented Hemother. 2020 Feb;47(1):45–53. 51 Reyes F, Gourdin MF, Lejonc JL, Cartron JP,
clinical routine screening for fetal RHD in D– 38 O'Brien H, Hyland C, Schoeman E, Flower R, Gorius JB, Dreyfus B. The heterogeneity of
pregnant women to ascertain the requirement Daly J, Gardener G. Non-invasive prenatal erythrocyte antigen distribution in human
for antenatal RhD prophylaxis. Transfusion. testing (NIPT) for fetal Kell, Duffy and Rh normal phenotypes: an immunoelectron mi-
2012 Apr;52(4):752–8. blood group antigen prediction in alloim- croscopy study. Br J Haematol. 1976 Dec;
27 Clausen FB, Rieneck K, Krog GR, Bundgaard munised pregnant women: power of droplet 34(4):613–21.
BS, Dziegiel MH. Noninvasive antenatal digital PCR. Br J Haematol. 2020 May;189(3): 52 Dziegiel MH, Hansen MH, Haedersdal S, Bar-
screening for fetal RHD in RhD negative e90–4. rett AN, Rieneck K, Main KM, et al. Blood
women to guide targeted anti-D prophylaxis. 39 Orzińska A, Guz K, Mikula M, Kluska A, Bal- chimerism in dizygotic monochorionic twins
Methods Mol Biol. 2019;1885:347–59. abas A, Ostrowski J, et al. Prediction of fetal during 5 years observation. Am J Transplant.
28 Clausen FB, Steffensen R, Christiansen M, blood group and platelet antigens from ma- 2017 Oct;17(10):2728–32.
Rudby M, Jakobsen MA, Jakobsen TR, et al. ternal plasma using next-generation sequenc- 53 Dooren MC, Kuijpers RW, Joekes EC, Huis-
Routine noninvasive prenatal screening for ing. Transfusion. 2019 Mar;59(3):1102–7. kes E, Goldschmeding R, Overbeeke MA, et
fetal RHD in plasma of RhD-negative preg- 40 Delaney M, Matthews DC. Hemolytic disease al. Protection against immune haemolytic
nant women – 2 years of screening experience of the fetus and newborn: managing the disease of newborn infants by maternal
from Denmark. Prenat Diagn. 2014 Oct; mother, fetus, and newborn. Hematology Am monocyte-reactive IgG alloantibodies (anti-
34(10):1000–5. Soc Hematol Educ Program. 2015;2015:146– HLA-DR). Lancet. 1992 May 2; 339(8801):
29 Clausen FB, Krog GR, Rieneck K, Råsmark 51. 1067–70.
EE, Dziegiel MH. Evaluation of two real-time 41 Donneborg ML, Hansen BM, Vandborg PK, 54 Gijtenbeek M, Lopriore E, Steggerda SJ, Te
multiplex PCR screening assays detecting fe- Rodrigo-Domingo M, Ebbesen F. Extreme Pas AB, Oepkes D, Haak MC. Persistent pul-
tal RHD in plasma from RhD negative women neonatal hyperbilirubinemia and kernicterus monary hypertension of the newborn after fe-
to ascertain the requirement for antenatal spectrum disorder in Denmark during the tomaternal hemorrhage. Transfusion. 2018
RhD prophylaxis. Fetal Diagn Ther. 2011; years 2000-2015. J Perinatol. 2020 Feb; 40(2): Dec;58(12):2819–24.
29(2):155–63. 194–202. 55 Schonewille H, Klumper FJ, van de Watering
30 Clausen FB, Urhammer E, Rieneck K, Krog 42 Ziprin JH, Payne E, Hamidi L, Roberts I, Re- LM, Kanhai HH, Brand A. High additional ma-
GR, Nielsen LK, Dziegiel MH. How to evalu- gan F. ABO incompatibility due to immuno- ternal red cell alloimmunization after Rhesus-
ate PCR assays for the detection of low-level globulin G anti-B antibodies presenting with and K-matched intrauterine intravascular
DNA. APMIS. 2015 Sep;123(9):731–9. severe fetal anaemia. Transfus Med. 2005 Feb; transfusions for hemolytic disease of the fetus.
31 de Haas M, Thurik FF, van der Ploeg CP, 15(1):57–60. Am J Obstet Gynecol. 2007 Feb;196(2):143–6.
Veldhuisen B, Hirschberg H, Soussan AA, et 43 Zonneveld R, van der Meer-Kapelle L, Sylva 56 Evers D, Middelburg RA, de Haas M, Zalpuri
al. Sensitivity of fetal RHD screening for safe M, Brand A, Zijlstra M, Schonewille H. Severe S, de Vooght KM, van de Kerkhof D, et al.
guidance of targeted anti-D immunoglobulin fetal hemolysis and cholestasis due to high- Red-blood-cell alloimmunisation in relation
prophylaxis: prospective cohort study of a na- titer maternal IgG anti-A antibodies. Pediat- to antigens' exposure and their immunoge-
tionwide programme in the Netherlands. rics. 2019 Apr;143(4):e20182859. nicity: a cohort study. Lancet Haematol. 2016
BMJ. 2016 Nov 7;355:i5789. 44 Kaplan M, Hammerman C, Vreman HJ, Jun;3(6):e284–92.
32 Haimila K, Sulin K, Kuosmanen M, Sareneva Wong RJ, Stevenson DK. Hemolysis and hy- 57 Krog GR, Rieneck K, Clausen FB, Steffensen
I, Korhonen A, Natunen S, et al. Targeted an- perbilirubinemia in antiglobulin positive, di- R, Dziegiel MH. Blood group genotyping of
tenatal anti-D prophylaxis program for RhD- rect ABO blood group heterospecific neo- blood donors: validation of a highly accurate
negative pregnant women – outcome of the nates. J Pediatr. 2010 Nov;157(5):772–7. routine method. Transfusion. 2019 Oct;
first two years of a national program in Fin- 45 McDonnell M, Hannam S, Devane SP. Hy- 59(10):3264–74.
land. Acta Obstet Gynecol Scand. 2017 Oct; drops fetalis due to ABO incompatibility. 58 Hendrickson JE, Delaney M. Hemolytic dis-
96(10):1228–33. Arch Dis Child Fetal Neonatal Ed. 1998 May; ease of the fetus and newborn: modern prac-
33 Jensen MPS, Damkjaer MB, Clausen FB, Ali 78(3):F220–1. tice and future investigations. Transfus Med
HA, Hare KJ, Dziegiel MH, et al. Targeted 46 Sarici SU, Yurdakök M, Serdar MA, Oran O, Rev. 2016 Oct;30(4):159–64.
Rhesus immunoglobulin for RhD-negative Erdem G, Tekinalp G, et al. An early (sixth- 59 Stack G, Tormey CA. Estimating the immu-
women undergoing an induced abortion: a hour) serum bilirubin measurement is useful nogenicity of blood group antigens: a modi-
clinical pilot study. Acta Obstet Gynecol in predicting the development of significant fied calculation that corrects for transfusion
Scand. 2019 Sep;98(9):1164–71. hyperbilirubinemia and severe ABO hemo- exposures. Br J Haematol. 2016 Oct; 175(1):
34 Rieneck K, Bak M, Jønson L, Clausen FB, lytic disease in a selective high-risk popula- 154–60.
Krog GR, Tommerup N, et al. Next-genera- tion of newborns with ABO incompatibility. 60 Jonsson S, Sveinbjornsson G, de Lapuente
tion sequencing: proof of concept for antena- Pediatrics. 2002 Apr;109(4):e53. Portilla AL, Swaminathan B, Plomp R,
tal prediction of the fetal Kell blood group 47 Bakkeheim E, Bergerud U, Schmidt-Melbye Dekkers G, et al. Identification of sequence
phenotype from cell-free fetal DNA in mater- AC, Akkok CA, Liestol K, Fugelseth D, et al. variants influencing immunoglobulin levels.
nal plasma. Transfusion. 2013 Nov; 53(11 Maternal IgG anti-A and anti-B titres predict Nat Genet. 2017 Aug;49(8):1182–91.
Suppl 2):2892–8. outcome in ABO-incompatibility in the neo-
nate. Acta Paediatr. 2009 Dec; 98(12): 1896–
901.

314 Transfus Med Hemother 2021;48:306–315 Dziegiel/Krog/Hansen/Olsen/Lausen/

DOI: 10.1159/000518782 Nørgaard/Bergholt/Rieneck/Clausen
61 Tan JCG, Yuan FF, Daley J, Marks K, Flower 63 Hall AM, Cairns LS, Altmann DM, Barker 65 Mathiesen L, Nielsen LK, Andersen JT,
RL, Dyer WB. D-immunized blood donors RN, Urbaniak SJ. Immune responses and tol- Grevys A, Sandlie I, Michaelsen TE, et al. Ma-
who are female and who possess at least one erance to the RhD blood group protein in ternofetal transplacental transport of recom-
HLA-DRB1*15 allele show a propensity for HLA-transgenic mice. Blood. 2005 Mar 1; binant IgG antibodies lacking effector func-
high serum RhIG production. Transfusion. 105(5):2175–9. tions. Blood. 2013 Aug 15;122(7):1174–81.
2018 May;58(5):1182–8. 64 Hall LS, Hall AM, Pickford W, Vickers MA, 66 Nielsen LK, Green TH, Sandlie I, Michaelsen
62 Verduin EP, Brand A, van de Watering LM, Urbaniak SJ, Barker RN. Combination pep- TE, Dziegiel MH. In vitro assessment of re-
Roelen DL, Kanhai HH, Doxiadis II, et al. The tide immunotherapy suppresses antibody and combinant, mutant immunoglobulin G anti-
HLA-DRB1*15 phenotype is associated with helper T-cell responses to the RhD protein in D devoid of hemolytic activity for treatment
multiple red blood cell and HLA antibody re- HLA-transgenic mice. Haematologica. 2014 of ongoing hemolytic disease of the fetus and
sponsiveness. Transfusion. 2016 Jul; 56(7): Mar;99(3):588–96. newborn. Transfusion. 2008 Jan;48(1):12–9.
1849–56.

Laboratory Monitoring of HDFN: The Transfus Med Hemother 2021;48:306–315 315

Copenhagen Approach DOI: 10.1159/000518782