Transfusion Medicine Reviews 28 (2014) 1–6

Contents lists available at ScienceDirect

Transfusion Medicine Reviews journal homepage: www.tmreviews.com

Original Articles

Hemolytic Disease of the Fetus and Newborn With Late-Onset Anemia due to Anti-M: A Case Report and Review of the Japanese Literature Hiroyasu Yasuda a,⁎, Hitoshi Ohto a, Kenneth E. Nollet a, b, Kinuyo Kawabata a, Shunnichi Saito a, Yoshihito Yagi c, Yutaka Negishi d, Atsushi Ishida d a

Department of Blood Transfusion and Transplantation Immunology, Fukushima Medical University, Fukushima, Japan Radiation Medical Science Center, Fukushima Medical University, Fukushima, Japan Division of Central Laboratories, Gifu Prefectural Tajimi Hospital, Tajimi Japan d Department of Pediatrics, Gifu Prefectural Tajimi Hospital, Tajimi Japan b c

a r t i c l e

i n f o

Article history: Available online 19 October 2013

a b s t r a c t Hemolytic disease of the fetus and newborn (HDFN) attributed to M/N-incompatibility varies from asymptomatic to lethally hydropic. Case reports are rare, and the clinical significance of anti-M is not completely understood. A challenging case of HDFN due to anti-M prompted an investigation of the Japanese literature, in order to characterize the clinical spectrum of M/N-incompatibility pregnancies in Japan and report results to English-language readers. Japanese reports of HDFN attributed to M/N incompatibility were compiled. Abstracted data include maternal antibody titers at delivery, fetal direct antiglobulin test, hemoglobin, total bilirubin, reticulocyte count at birth, and therapeutic interventions. We investigated characteristics of HDFN due to M/N-incompatible pregnancies in Japan after encountering a case of severe HDFN along with late-onset anemia in an infant born to a woman carrying IgG anti-M with a titer of 1. In total, thirty-three babies with HDFN due to anti-M and one due to anti-N have been reported in Japan since 1975. The median maternal antibody titer was 64 at delivery and was 16 or less in 10 of 34 women (29%). Five of 34 babies (15%) were stillborn or died as neonates. Twenty-one of 29 survivors (72%) had severe hemolytic anemia and/or hydrops fetalis. The reticulocyte count of neonates with anemia stayed below the reference interval. Sixteen (55%) developed late-onset anemia and 14 (48%) were transfused with M-negative RBCs. Significant positive correlation (P b .05) between the hemoglobin value and the reticulocyte count within 4 days of birth was obtained in 16 babies with anti-M HDFN. In the Japanese population, 21 of 34 cases of M/N-incompatible HDFN (72%) have manifested as severe hemolytic anemia and/or hydrops fetalis. Low reticulocyte count in neonates with late-onset anemia is consistent with suppressed erythropoiesis due to anti-M. © 2014 Elsevier Inc. All rights reserved.

Anti-M is usually a naturally occurring IgM of clinical insignificance, but alloimmunization through pregnancy or transfusion may stimulate IgG anti-M. The fetus or newborn may be asymptomatic [1] to hydropic [2] with hemolytic disease of the fetus and newborn (HDFN) severity independent of the anti-M titer. It is thought that the incidence of HDFN due to anti-M is very low [1], but more than 34 cases of mild to fatal HDFN have been reported in Japan since 1975; most required therapeutic interventions [3–34]. Here we report case of HDFN with low-titer anti-M at delivery and severe late-onset anemia accompanied by hemolysis. This case is discussed in the context of 32 previous reports from Japan.

Conflict of Interest Statement: The authors declare that they have no conflicts of interest pertaining to the content of this manuscript. ⁎ Corresponding author. Hiroyasu Yasuda, MT, PhD, Department of Blood Transfusion and Transplantation Immunology, Fukushima Medical University, Fukushima 960-1295 Japan. E-mail address: [email protected] (H. Yasuda). 0887-7963/$ – see front matter © 2014 Elsevier Inc. All rights reserved. http://dx.doi.org/10.1016/j.tmrv.2013.10.002

Case Report A 22-year-old Japanese woman sought prenatal care at 15 weeks, at which time screening revealed anti-M. She was gravida 2, para 1, transfusion 0 with a history of intrauterine fetal demise at 39 weeks. Relevant phenotypes were M−N + for the woman and M+N− for her husband. At delivery, the titer rose to 4 by saline-indirect antiglobulin test (Sal-IAT). Thereafter, IgG anti-M with a titer of 1 was detected in DTT-treated maternal plasma by Sal-IAT. A 3270-g male infant was delivered at 38 weeks and 2 days of gestation. Apgar scores at 1 and 5 minutes were 9 and 10, respectively. The infant's phenotype was M+N+ but the direct antiglobulin test (DAT) was negative. The indirect antiglobulin test (IAT) was positive for anti-M. Laboratory results included hemoglobin (Hb, 8.6 g/dL); lactate dehydrogenase (LDH, 764; reference interval, 119–229 IU/L); haptoglobin (Hp, b10 mg/dL; reference interval, 43–180 mg/dL) and complete blood count showing

2

H. Yasuda et al. / Transfusion Medicine Reviews 28 (2014) 1–6 IVIG

Materials and Methods

CS ( 2mg/kg/day )

Photo

Serology

16 14 12

120

10

100

8

80

6

60

4

40

2

20

0

0 40 Day

5

10

15

20

25

30

35

Hp (mg/dL), LDH(×10IU/L), Reticulocyte count (×103/µL)

T-Bil, D-Bil (mg/dL), Hb (g/dL)

18

Fig. 1. Neonatal course of severe anemia due to anti-M HDFN in the present case. □ = Total bilirubin (T-Bil); ■ = Direct bilirubin (D-Bil); ○ = Hemoglobin (Hb); ●= Reticulocyte count; ◊= LDH; ♦= Haptoglobin (Hp); ↓ = M-negative RBC transfusion; Photo = phototherapy.

peripheral leukocytes and platelets as normal on admission. He developed icterus with elevated levels of total bilirubin (TB, 14.8 mg/dL) and indirect bilirubin (IB, 14.3 mg/dL) 18 hours after birth (Fig. 1). High-dose intravenous immunoglobulin (IVIG) therapy and phototherapy were started immediately after admission to the neonatal intensive care unit. Physical examination revealed mildly edematous skin. There was no hepatosplenomegaly or neurologic abnormality. Hb decreased to 6.6 g/dL on day 2 and he received five M-negative red blood cell (RBC) transfusions through Day 12. TB decreased to 3.1 mg/dL by day 17 after birth, but his anemia progressed along with a decreasing reticulocyte count despite IVIG therapy. By Day 20, the reticulocyte count had fallen to 6000/μL (reference interval: 25 000–75 000/μL) but serum erythropoietin (EPO) was normal at 12.4 U/L (Reference interval: 5-20 U/L). A bone marrow aspirate was morphologically normal, but the nucleated cell count was 40 800/μL and M/E ratio was 2.0, indicating that no erythroblastosis accompanied by hemolytic anemia was observed. With that, impaired erythropoiesis was suspected. High-dose of corticosteroid (CS) therapy for acquired pure red cell aplasia (PRCA) was started on Day 25. A sixth M-negative RBC transfusion was given on Day 34 because the Hb reached a nadir of 6.3 g/dL, following severe reticulocytopenia. However, there was no serological evidence of hemolysis or viral infection, such as parvovirus B19 (IgM negative), cytomegalovirus (IgM negative) or herpes simplex virus-type 1 (IgM negative). Thereafter, reticulocytopenia and persistent anemia were resolved with CS therapy and the infant was discharged on Day 59.

ABO and D blood typing and MNS phenotyping were performed in tubes according to reagent manufacturer’s instructions (OrthoClinical Diagnostics, Inc, Raritan, NJ). A 3-cell antibody screen (Surgiscreen 1, 2 and 3, Ortho, Inc) and Diego A RBC reagents (Ortho, Inc) for antibody screening were used with 20% polyethylene glycol (in-house 20% w/v in phosphate-buffered saline) in tubes [35]. Subsequent antibody identification was performed by polyethylene glycol–IAT using an 11-cell antibody panel (Resolve Panel A; Ortho, Inc.) per manufacturer’s instructions. DAT was also performed in tubes with anti-IgG and anti-C3d (ImmucorGamma, Norcross, GA). Antibody titration was performed by a test tube method [35]. Serial 2fold dilutions of the plasma in saline were tested against selected group O, D +, M+N − RBCs at 37°C for 60 min by saline-IAT. At that time, group O, D +, M−N + RBCs were used as a negative control. The reciprocal of the highest dilution giving a 1 + reaction was recorded as the titer. To distinguish IgG from IgM antibodies, patient plasma samples and reagent control (32-fold-diluted Anti-A BioClone, Ortho, Inc) were added in an equal volume to 0.01 mol/L dithiothreitol (DTT, Wako Pure Chemical Industries, Osaka, Japan) and treated at 37°C for 15 minutes. One 0.05 mmol/L DTT aliquot was added to 9 aliquots of plasma when it was negative at 1:2, so that the final concentration of DTT and its titer became 0.005 mmol/L and 1:1, respectively. Dilution of untreated or DTT-treated plasma was performed in parallel. Results Parental Phenotypes The maternal red cell phenotype was group A, D+, M−N +, the paternal phenotype was group O, D +, M+N−, and the newborn boy was group A, D +, M+N+ (Table 1). Antibody Identification and Titration Anti-M antibody was detected in maternal plasma at 15 weeks gestation at a titer of 4 in saline and 1 in saline-IAT. At delivery, the titer rose to 4 by Sal-IAT. Untreated maternal plasma reacted 3 + with group O D+, M+N– RBCs in saline at room temperature at a titer of 4. DTT-treated maternal plasma was unreactive with M+N– RBCs in saline but was reactive 1 + in saline-IAT, suggesting predominately IgM and low-titer IgG isotypes (Table 1). There was no evidence of other clinically significant alloantibodies. At birth, the baby had anti-M titers of 16 and 2 in saline and in saline-IAT, respectively, but DAT was negative. Anti-M was also detected in the baby’s plasma collected on Day 24 by saline and salineIAT methods but not by saline-IAT on Day 42.

Table 1 Serological characteristics of anti-M in this case ABO

Rh

MNS

Irregular antibody at delivery Specificity

Saline

Mother Child Father

A A O

D+C-c+E+e+ D+C+c+E+e + D+C+c+E+e +

M-N+S-s+ M+N+S-s+ M+N-S-s+

anti-M anti-M

DAT

Untreated titera

DTT-treated titera Saline-IAT

Saline

Saline-IAT

4°C

RT

37°C

4°C

RT

37°C

32 16

4 2

4 2

0 0

0 0

1b 0

NT, not tested; RT, room temperature. a Twofold dilution. b One 0.05 mmol/L DTT aliquot was added to 9 aliquots of plasma when results were negative at 1:2.

NT 0

H. Yasuda et al. / Transfusion Medicine Reviews 28 (2014) 1–6

Literature Review We searched online for previous Japanese- and English-language literature on HDFN due to M/N incompatibility among Japanese from 1975 to 2012, using Ichushi-Web (Ver. 4, Japan Medical Abstracts Society, JAMAS) and PubMed (National Center for Biotechnology Information, US National Library of Medicine). Data abstracted from relevant cases included: (1) maternal anti-M or anti-N titers at delivery; (2) fetal direct antiglobulin test (DAT), hemoglobin, total bilirubin and reticulocyte count at birth; and (3) therapeutic interventions. To quantify relationships between Hb and reticulocyte count, authors were contacted if published data were lacking. Twenty of 34 Japanese pregnant women (59%) had previous histories of fetal demise, spontaneous abortion and/or HDFN due M/N incompatibility [3–34] (Table 2). The median AHG titer of maternal antibody was 64 including 10 women (29%) with a titer 16 or less at delivery or at the time of intervention. To decrease the titer of maternal anti-M, the pregnant woman in Case 3 [5] received multiple

3

plasma exchanges by apheresis and reinfusion of anti-M-depleted autologous plasma after antibody adsorptions using M-positive RBCs; the woman in Case 13 [14] also received 28 antenatal plasmaphereses to reduce maternal anti-M titers. Thirty-four Japanese cases of HDFN resulting from M/N incompatibility (33 anti-M and 1 anti-N) have been reported from 1975 to 2012. Twenty-nine live-birth infants (85%), 2 stillbirths and 3 neonatal deaths included 21 babies (72%) with severe hemolytic anemia (b 10 g/dL) [36] and/or hydrops fetalis due to anti-M. IgG subclasses have been investigated in 5 of 29 babies. IgG1 was found in 3 babies (cases 15, 21, 32), IgG3 was found in case 7, and both IgG1 and IgG3 were found in case 20. At least 23 babies (79%) were DATnegative, but 28 (97%) required interventions such as exchange transfusion (n= 18, 62%), phototherapy (n= 18, 62%), IVIG (n= 9, 31%), Epo (n=3, 10%) and/or CS (n= 2, 7%). Fourteen babies (48%) received M-negative RBC transfusions after birth. Of note, 16 babies (55%) developed late-onset anemia despite interventions and had delayed recovery of Hb to 9 g/dL. Of 17 babies (except neonatal death

Table 2 Clinical findings of pregnant women and babies in case reports of HDFN due to M/N incompatibility among Japanese from 1975 to 2012 Case

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34

First author [Ref.], year

Mimaya [3], 1975 Matsumoto [4], 1981 Aso [5], 1983 Ishihara [6], 1983 Wakatsuki [7], 1983 Yamamoto [8], 1985 Hiroshige [9], 1986 Ito [10], 1986 Kobayashi [11], 1987 Matsuoka [12] (2 siblings), 1988 Noda [13], 1990 Furukawa [14], 1993 Miyagi [15], 1993 Tada [16], 1994 Yamada [17], 1996 Hojo [18], 1998 Oshida [19], 1999 Honda [20], 1999 Serizawa [21], 1999 Fujimaru [22], 2002 Usuda [23], 2002 Yatsugi [24], 2004 Kawamura [25], 2005 Maruyama [26], (2 siblings), 2005 Asano [27], 2008 Yatomi [28], 2009 Yoshida [29], 2009 Toriumi [30], 2010 Kosaka [31], 2010 Ishimaru [32], 2011 Uemura [33], 2012 Present case

Pregnant woman Previous historiesa

Yes Yes Yes Yes Yes Yes Yes No Yes Yes Yes Yes Yes No No No No No Yes Yes Yes No Yes Yes No Yes No No No No Yes Yes Yes Yes

Infant

Antibody Titerb Saline

Saline-IAT

128 2,048 NR 1,024 NR NR 512 NR NR 64 8 NR NR NR 8 NR NR 32 NR 8 NR NR 64 NR NR NR NR 64 NR NR 32 NR NR 32

64 1,024 128-256c 64 NR NR 128 64 16 64 1 64 256d 256 8 256 128 8 16 32 256 64 2 64 32 64 128 8 128 64 8 32 16 4

DAT IgG Subclass

G3

G1

G1 G3 G1

G1

− NR − NR + − NR − NR − − + + − − NR − − − − − − − − − − + − + + − − − −

Highest Bil (mg/dL)

Lowest Hb (g/dL)

Major complaint at birth

Intervention

Ex-Tx

Tx

30.2 NR 11.0 NR 26.5 NR NR 23.0 10.8 28.0 25.4 8.2 NR 15.0 6.4 NR 17.7 20.1 5.5 5.2 15.0 4.4 4.1 NR 33.7 0.5 NR 11.7 17.5 15.7 16.9 11.2 8.7 17.0

6.5 NR 6.1 NR 12.6 NR NR NR 6.4 16.6 10.4 7.5 NR 7.3 4.4 10.7 1.8 17.6 9.0 6.0 4.9 3.3 5.0 4.7 4.9 2.4 3.1 6.9 6.4 6.7 6.2 3.4 3.9 6.7

SA, Icterus HF SA, HF HF Icterus Icterus HF Icterus SA, Icterus HF, Icterus Icterus SA, Heart failure Icterus Icterus SA, HF Icterus SA, Icterus Icterus SA SA, HF SA, Icterus SA SA, Fetal hypoxia SA SA, Icterus SA, HF SA, HF SA, Icterus SA, Icterus SA, Icterus SA, Icterus SA, Icterus SA, Icterus SA, Icterus

1 NR 2 NR 1 3 NR 3 1 3 1 3 0 0 10e 0 4 1 1 3f 1 0 0 1 1 1 0 0 0 0 0 0 1 0

2 NR 6 NR 0 0 NR 0 0 0 0 0 0 0 0 0 0 0 0 10g 2 0 1 1 2 0 1 2 0 1 2 8 5h 6

Number of

Outcome Other

Photo

Photo

Photo Photo Photo

IVIG, Photo Photo IVIG, Photo IVIG, Photo Photo IVIG, Photo, EPO, CS IVIG, Photo, EPO IVIG, Photo, EPO

Photo IVIGi IVIG, Photo Photo Photo IVIG, Photo, CS

Alive, LA (81)j Stillbirth Alive, LA (34)j Stillbirth Alive Alive Neonatal death Alive Alive Alive, Kernicterus Alive Neonatal death Alive Alive, LA (96)j Alive Alive Alive, LA (44)j Alive Alive Alive Alive, LA (55)j Alive, LA (47)j Alive, LA (90)j Alive, LA (58)j Alive, LA (42)j Neonatal death Alive, LA (60)j Alive, LA (47)j Alive Alive, LA (16)j Alive, LA (68)j Alive, LA (84)j Alive, LA (39)j Alive, LA (42)j

Bil, bilirubin; CS, corticosteroid; EPO, erythropoietin; Ex-Tx, exchange transfusion; Hb, hemoglobin; HF, hydrops fetalis; LA, late onset anemia; NR, not reported; SA, severe anemia (Hb b10g/dL); Tx, transfusion; Photo, phototherapy; Case 12 was a HDFN due to anti-N. a Uncertain fetal demise, spontaneous abortion and/or anti-M HDFN. b Titer at delivery or at the time of intervention. c Received 6 plasma exchanges and 22 antibody eliminations during pregnancy. d Received 28 plasma exchanges during pregnancy. e Including 4 intrauterine exchange transfusions. f Intrauterine exchange transfusion. g Including 8 intrauterine transfusion. h Including 3 intrauterine transfusions. i Immunoglobulin injected into the fetal abdominal cavity. j Figure in parentheses means days required for recovery of Hb to 9 g/dL after intervention.

4

H. Yasuda et al. / Transfusion Medicine Reviews 28 (2014) 1–6

in Cases 12 and 26) who received one or more exchange transfusions, only 7 (41%) developed late-onset anemia, vs 9 (82%) of 11 (except no intervention in Case 16) without exchange transfusion (P = .054). Interestingly, positive correlation between the Hb value and the reticulocyte count within 4 days of birth was statistically significant (P b .05) in 15 babies with HDFN due to anti-M (Fig. 2). Three neonates (10%) died of apparent heart failure or severe hydrops fetalis within 2 days of birth. One neonate suffered from kernicterus but the other 28 had no sequelae after medical intervention.

Discussion Hemolytic disease of the fetus and newborn generally arises from transport of maternal IgG through the placenta, with subsequent binding to cognate antigens on fetal RBC membranes. The ensuing extravascular hemolysis can result in severe anemia and hydrops fetalis. Rh antibodies (especially anti-D and anti-c) or anti-K are frequent causes of severe anemia in the fetus and newborn [37,38]. M alloimmunization per se is common but M/N-related HDFN is rarely reported. Meanwhile, HDFN caused by anti-N seems even less frequent than that caused by anti-M because M+N− individuals usually have RBCs with glycophorin B, which includes N-like epitope ‘N’ to inhibit N alloimmunization. Anti-N alloimmunization remains possible if ‘N’ is absent, as in the M+N−U − phenotype [39,40]. Recent Japanese data show that anti-M was encountered in about 9% of 375 Japanese pregnant women with positive antibody screens although the overall frequency of alloantibodies was 1.43% among 248,785 patients hospitalized in 29 institutions from 2008 to 2009 [41]. These data put the Japanese alloimmunization incidence of M/Nincompatible pregnancies on par with the nearly 10% incidence reported in a 26-year retrospective from Ohio State University [1]. HDFN arising from M/N incompatibility may be rare in USA. However, it is unclear why 34 cases of mild to fatal HDFN have been reported in Japan since 1975 [3–34]. Another report, from China, describes the consequences of anti-M HDFN in a series of 21 M/N-incompatible pregnancies spanning 16 years [42]. Chinese HDFN in varying degrees of severity due to anti-M also led to intrauterine deaths or medical interventions for treatment of hyperbilirubinemia and hemolytic anemia. Mild to severe anemia and icterus were found in all cases including 3 that resulted in death. Similar reports from non-Asian populations are lacking. It is reasonable to wonder if genetic factors linked to race affect the immune response to erythrocyte alloantigens. Solid organ transplant outcomes have been investigated along racial lines, with some indication of immune response differences [43], but specifically in regard to blood transfusion, the literature is lacking. This suggests an important frontier for international cooperation in our discipline.

Hemoglobin (g/dL)

20 15 10 5

0 50,000

100,000

150,000

200,000

Reticulocyte count (/µL) Fig. 2. Relationship between hemoglobin (Hb) and reticulocyte count within 4 days of birth in 16 Japanese case reports from 1975 through 2012 (r = 0.57; Pb.05). Superscripts refer to case numbers in Table 2; Death = †; Late-onset anemia ( Yes = ◊; No = □); Hydrops fetalis (○); Present case = *.

We summarized Japanese cases with anti-M HDFN that included 29 of 34 antibodies with titers (median 64: range 1-1024) in salineIAT (Table 2). According to our review of 29 infants with HDFN due to anti-M, 80% of 10 infants with maternal titers of 16 or less at delivery or at the beginning of intervention experienced severe HDFN, indicating anti-M antibody may lead to severe HDFN at a lower maternal antibody titer. Generally, anti-M antibodies are predominantly IgM type but usually include a component of IgG. Fifty to eighty percent of anti-M antibodies contain some IgG forms [39,44]. IgG antiM antibodies often agglutinate M+RBCs directly in saline because glycophorin A (GPA) is abundantly expressed on RBCs and M antigen epitopes are located at the terminal end of GPA molecules [44–46]. In the present case, anti-M detected in untreated maternal and baby’s plasma might be misinterpreted as only IgM if IgG anti-M had not been detected in DTT-treated maternal plasma. But the maternal AHG titer was 4 in untreated and 1 in DTT-treated plasma in Sal-IAT at delivery (Table 1). At first, no reaction was found in DTT-treated maternal plasma at 2-fold dilution but a 1 + reaction was detected in Sal-IAT with minimum dilution. These findings indicate that a very low IgG anti-M may coexist in the affected body’s circulation and cause a moderate HDFN. Wikman et al. [2] report three pregnancies in a Swedish family that all resulted in severe fetal anemia due to anti-M with a titer of 1 or less by IAT. In particular, hemolytic activity of antiM was demonstrated in maternal blood in the third pregnancy, although the anti-M was mainly IgM with only slight amounts of IgG1 and IgG3. The first fetus died in utero with hydrops fetalis during the 20th gestational week and the second child was delivered after 28 weeks of gestation with hydrops fetalis and a Hb of 1.6 g/dL at birth, whereas the third affected child was treated with intrauterine RBC transfusions at 28 weeks. In addition, severe anemia accompanied by reticulocytopenia developed in the second baby from 12 days to 7 weeks of age and in the third baby from 2 days before birth to 1 month of age. During this period, they required several RBC transfusions for treatment of late-onset anemia. Kell antibodies are especially significant for suppressing erythropoiesis at the progenitor-cell level because anti-K specifically inhibits the growth of Kell-positive erythroid progenitor cells [47]. Anti-M [48] and anti-Gerbich 3 [49–52] antibodies also appear to cause HDFN and subsequent late-onset anemia possibly by the same mechanism, because M/N and Ge antigens appear on membrane proteins such as GPA or glycophorin C, which are expressed, like Kell, on early erythroid precursors [53,54]. Interestingly, Wikman’s report indicates 2 important possibilities that an anti-M may rarely cause severe HDFN even when the titer is below 2 by IAT and may induce suppression of erythropoiesis along with late-onset anemia. Nolan et al. [48] also reported HDFN due to anti-M in an infant with a transient severe neonatal PRCA, accompanied by reticulocytopenia (reticulocyte count, 5000/μL). He demonstrated that maternal anti-M induced a dose-dependent reduction in proliferation of M-positive erythroid progenitor cells in culture. These findings indicated that anti-M antibodies could target erythroid precursors and contribute to neonatal hypogenerative anemia or a late-onset anemia after birth. In other words, M antigen on the M/N determinants of transmembrane GPA are abundantly expressed on the surface of erythroid progenitors, so M/N alloimmunization may suppress erythropoiesis as well as provoke hemolysis. M/N determinants are fully developed on fetal RBCs and can be detected as early as 9 weeks of gestation [55]. Probably, abundant M/N antigens may be also expressed on neonatal RBCs as approximately a million copies of GPA are expressed on mature RBCs [45]. In the present case, late-onset anemia was accompanied by severe peripheral reticulocytopenia from Day 9 till Day 32 and the absence of erythroblastosis in bone-marrow aspiration on Day 20. These findings suggest suppression of erythropoiesis by anti-M at the progenitor-cell level, exacerbating anemia, may be involved in such cases. According to our literature investigation, reticulocyte counts at birth tend to be low in babies with HDFN due to

H. Yasuda et al. / Transfusion Medicine Reviews 28 (2014) 1–6

anti-M, in contrast to being high in typical cases of hemolytic anemia. The significant positive correlation (P b .05, Fig. 2) between Hb value and reticulocyte count within four days of birth supports the idea that anti-M could inhibit erythroid precursor growth, as occurs with anti-K [47] and anti-Ge [49–52]. Severe late-onset anemia found in babies with HDFN due to M incompatibility may be exacerbated by decreasing erythropoiesis in the presence of anti-M. These phenomena were also found in present case as well as families reported by Wikman and Nolan. Further, 2 cases with severe neonatal hypogenerative anemia due to anti-Vw (Verwest, MNS9) have been described [56,57]. Vw is a low-incidence antigen in the MNS blood group system that arises from a transposition between genes encoding for GPA and glycophorin B [hybrid A(20-46)-B(47)-A(48-150)] but includes the same M/N determinants carried on the transmembrane GPA [58]. In both HDFN cases due to anti-Vw, late-onset anemia with a significant reticulocytopenia was found and additional transfusions were necessary for each infant around Day 30 of life. After delivery, the infant's M/N phenotype should be determined because a negative DAT may be found even when M antigen is present in the infant and hemolysis is occurring. Twenty-three babies (79%) including our case were DAT-negative in our review. On occasion, an elution test is also effective because causative antibodies of hemolytic anemia may exist in the eluate from an infant’s RBCs. Unfortunately, this opportunity was lost in the present case. In Japan, medical interventions for infants with HDFN have evolved recently, favoring phototherapy and high-dose IVIG for neonatal hyperbilirubinemia and diminishing the role of exchange transfusion in acute care. Results summarized in Table 2 suggest that late-onset anemia following anti-M HDN might be more common in neonates treated with IVIG or phototherapy (82%, 9 of 11) compared with exchange transfusion (41%, 7 of 17), P = .054. CS therapy comprised the first immunosuppressive regimen to treat PRCA and continues to be the treatment of first choice, especially in young individuals [59]. Increment of the reticulocyte count was found in the baby after administration of CS on Day 25. However, the effect of CS on the treatment against late-onset anemia with transient PRCA due to M alloimmunization remains controversial. Further investigation of CS as therapy for alloimmune hypogenerative anemia is warranted. In anti-K HDFN the lack of a consistent correlation between the degree of fetal anemia and the titer of anti-K, anti-Ge or anti-Vw has confounded the assessment of such pregnancies.[47,49,56] As previously mentioned, anti-M also defies prediction and serial antibody titers are not reliable. In Japan, detection of anti-M during pregnancy should arouse clinical vigilance regardless of maternal titer. Clinicians may choose to monitor fetal Hb using a non-invasive Doppler assessment of middle cerebral artery blood velocity (MCAPSV) [60]. After delivery, our observations suggest that the course of infants with maternal anti-M may need closer monitoring for hemoglobin and reticulocyte count for a few weeks regardless of initial hemoglobin. This may be especially valuable when there is clinical hemolysis at birth accompanied by a low reticulocyte count. Specific recommendations cannot be made and further studies are needed to determine the real clinical impact of anti-M in HDFN. Generally, the first important test, if anti-M is detected, is to determine if it is IgG or IgM. If no IgG is detected, then the antibody should not be of concern as it will not cross the placenta. However, close prenatal observation may be especially appropriate for the rare patient with suspected HDFN due to anti-M in conjunction with a previous unexplained fetal demise or spontaneous abortion. Possible interventions include the use of ultrasound and MCA-PSV for the fetal assessment regardless of maternal anti-M. Neonatal blood samples can be used to assess the severity of hemolysis and to monitor possible late-onset anemia by measurement of Hb, indirect bilirubin, LDH and reticulocyte count. For infants with evidence of moderate to severe HDFN at birth, continued follow-up in the early neonatal period will allow detection and prompt treatment of late-onset anemia.

5

Conclusion We report a neonate with maternal anti-M associated with early onset hemolytic anemia and hyperbilirubinemia, who subsequently developed transient PRCA manifesting as a late-onset anemia accompanied by reticulocytopenia. Several RBC transfusions were required through Day 35 after birth. It appears that a low-titer IgG type anti-M may lead to an unusual HDFN consisting of alloantibodydependent hemolysis and a late-onset anemia due to both immune destruction of fetal RBCs and suppression of erythropoiesis. Reticulocyte counts in conjunction with Hb measurements in the workup of anti-M HDFN are advisable. Acknowledgments Both parents of this newborn consented to all investigations and subsequent reporting. The relevant institutional review boards are guided by local policy, national law, and the World Medical Association Declaration of Helsinki. Some clinical aspects of the case were reported previously in Japanese [34]. References [1] Young-Owens AD, Kennedy M, Rose RL, Boyle J, O'Shaughnessy R. Anti-M isoimmunization: management and outcome at the Ohio State University from 1969 to 1995. Obstet Gynecol 1997;90:962–6. [2] Wikman A, Edner A, Gryfelt G, Jonsson B, Henter JI. Fetal hemolytic anemia and intrauterine death caused by anti-M immunization. Transfusion 2007;47:911–7. [3] Mimaya J, Nagao D, Komiya H. A case of hemolytic disease of the newborn due to M/N incompatibility] [in Japanese. Acta Pediatri Jpn 1975;79:340–4. [4] Matsumoto H, Tamaki Y, Sato S, Shibata K. A case of hemolytic disease of the newborn caused by anti-M: Serological study of maternal blood. Acta Obstet Gynaecol Jpn 1981;33:525–8. [5] Aso T, Tatsumi K, Yoshida H, Yoshida Y. A new method of treatment for intrauterine fetal hemolytic anemia caused by blood type incompatibility. Acta Obstet Gynaecol Jpn 1983;35:88–96. [6] Ishihara S, Kato K, Tsutsumi O, Nakagawa M, Sugo M. A case of recurrent intrauterine deaths due to M/N blood group incompatibility. Obstet Gynecol 1983;50:107–10 [in Japanese]. [7] Wakatsuki S, Hara M, Yoshimura K, Yabe K, Yanagida J. Anti-M produced by M/N incompatible pregnancy. J Saitama Med Sch 1983;10:112–3 [abstract] [in Japanese]. [8] Yamamoto Y, Ohara K, Ishiwata C. A case of hemolytic disease of the newborn due to M/N blood group incompatibility. Ibaraki Med J 1985;21:64 [abstract] [in Japanese]. [9] Hiroshige Y. A case of blood incompatible pregnancy due to anti-M. Jpn J Tran Med 1986;32:60–1 [abstract] [in Japanese]. [10] Ito S. A case of hemolytic disease of the newborn caused by anti-M. Kyosai Med J 1986;35:302 [abstract] [in Japanese]. [11] Kobayashi A, Yamazaki M, Deguchi M. A case of severe fatal anemia caused by M/N incompatible pregnancy. Acta Neonatol Jpn 1987;23:1 [abstract] [in Japanese]. [12] Matsuoka T, Fukamatsu Y, Mori A, Kasai S, Kamijo T, Yabuhara A, et al. A case of brother and sister with neonatal hemolytic disorder that was thought to be due to M/N blood group incompatibility. Perinatol Med 1988;18:1229–31 [in Japanese]. [13] Noda Y, Tatsumi H, Ueno N, Nishijima M, Shimada N, Oguchi H, et al. A case of hemolytic disease of the newborn caused by M/N blood group incompatibility. Jpn J Obstet Gynecol Soc (Kanagawa) 1990;26:57–60 [in Japanese]. [14] Fukuwara K, Nakajima T, Kogure T, Yazaki K, Yoshida M, Fukaishi T, et al. Example of a woman with multiple intrauterine deaths due to anti-M who delivered a live child after plasmapheresis. Exp Clin Immunogenet 1993;10:161–7. [15] Miyagi S, Hoshino K, Takeya H, Kawada N, Shigeta K, Tsukimoto I. A cases of hemolytic disease of the newborn caused by anti-M. Perinatol Med 1993;23: 1772–4 [in Japanese]. [16] Tada K, Hiramatsu Y, Masaoka H, Kishimoto Y, Kudo T, Naoki K. Intrauterine fetal exchange transfusion for hydrops fetalis caused by anti-M: A case report. Jpn J Obstet Gynecol Neon Hematol 1994;4:158–64. [17] Yamada N, Hamada Y, Watanabe H. A case of fatal anemia caused by anti-M derived from mother. Ibaraki Med J 1996;32:106 [abstract] [in Japanese]. [18] Hojo S, Ishikawa K, Ogawa K. A case of pregnancy complicated with“MNSs” system incompatibility. Acta Obstet Gynaecol Jpn 1998;50:963–6 [in Japanese]. [19] Oshida M, Nagamine K, Kiyokawa T, Aochi H, Hayashi S, Kurata Y. A case of hemolytic disease of the newborn caused by anti-M. Jpn J Tran Med 1999;45:263 [abstract] [in Japanese]. [20] Honda Y, Nishiyama C, Ohto H. A case of hemolytic disease of the newborn caused by anti-M. J Jpn Pediatr Soc 1999;103:489 [abstract] [in Japanese]. [21] Serizawa M, Matsui H, Oishi A, Maeda T, Sutoko K, Ohhashi T, et al. A case of hydrops fetalis caused by M/N blood group incompatibility. Jpn J Obstet Gynecol Neon Hematol 1999;8:145–52 [in Japanese].

6

H. Yasuda et al. / Transfusion Medicine Reviews 28 (2014) 1–6

[22] Fujimaru R, Ichiba H, Hirai C, Saito M, Shintaku H, Yamano K. A case of severe hemolytic anemia due to M/N blood group incompatibility. Jpn J Pediatr 2002;55: 28–30 [in Japanese]. [23] Usuda T, Ohishi M, Nagayama Y, Itano T, Yamazaki A. A case following a prolonged course of hemolytic anemia due to M/N blood group incompatibility. Niigata Med J 2004;118:145 [abstract] [in Japanese]. [24] Yatsugi S, Hirano H, Hosoya N, Hosoya N, Sato A, Ogawa M, et al. A case of neonatal anemia developed by M/N incompatible pregnancy. Jpn J Obstet Gynecol Soc (Kita-Nippon) 2005;53:72. [25] Kawamura M, Ishida A, Nakajima K, Takahashi M, Miura S, Takada G. A case of hemolytic disease of the fatal and newborn due to M/N blood group incompatibility. Jpn J Soc Premature Newborn Med 2005;17:146 [abstract] [in Japanese]. [26] Maruyama K, Sugiyama M, Koizumi T, Miyazaki M, Fjiu T. Two siblings with severe anemia due to M/N blood group incompatibility. Jpn J Soc Perinatol Neon Med 2005;41:603–7 [in Japanese]. [27] Asano Y, Takayama C, Kawasaki H, Kobayashi K, Sobashima H, Tamura M. A male infant with severe anemia due to M/N blood group incompatibility. J Jpn Pediatr Soci 2008;112:379 [in Japanese]. [28] Yatomi Y, Henmi N, Sato S, Kashikura M, Takano K, Tajima H, et al. A case of hemolytic anemia caused by M/N incompatibility. Jpn J Obstet Gynecol Neon Hematol 2009;19:S133–4 [abstract] [in Japanese]. [29] Yoshida M, Matsuda H, Wakamatsu F, Oeda M, Furuya K. Successful treatment of immunoglobulin injection into fetal abdominal cavity (IFAC) in a case of fetal hemolytic anemia caused by anti-M alloimmunization. Jpn J Soc Perinat Neon Med 2009;45:895–7 [in Japanese]. [30] Toriumi A, Uemura C, Matsuhashi H, Mochida N, Igarashi Y, Okuyama K, et al. A case of hemolytic disease of the newborn caused by anti-M. Jpn J Tran Med 2010;56:296 [abstract] [in Japanese]. [31] Kosaka M, Sano Y, Ueno A. A case of hemolytic anemia of the newborn due to MNS blood system incompatibility. Jpn J Chubu Med Technol 2010;49:157 [abstract] [in Japanese]. [32] Ishimaru K, Ohashi W, Sato S, Hayashi A, Kawaguchi S, Kohata N, et al. A case of hydrops fetalis born in a pregnant woman carrying anti-M antibody. Jpn J Tran Med 2011;57:246 [abstract] [in Japanese]. [33] Uemura M, Togashi K, Ohki N, Saito Y, Fuse I. A case of severe HDFN caused by a low-titer of anti-M antibody. Jpn J Tran Med 2012;58:358 [abstract] [in Japanese]. [34] Negishi Y, Ishida A, Nakano M, Kokubo Y. A case of hemolytic anemia occurring with pure red cell aplasia-like disorder due to anti-M. Jpn J Soc Premature Newborn Med 2008;20:352 [abstract] [in Japanese]. [35] Antibody Detection, Antibody Identification, and Serologic Compatibility. In: Brecker ME, editor. AABB Technical manual. 15th ed. Bethesda: American Association of Blood Banks; 2005. p. 751–69. [36] Roberts IA. The changing face of haemolytic disease of the newborn. Early Hum Dev 2008;84:515–23. [37] Nordvall M, Dziegiel M, Kristine H, Bidstrup M, Jonsbo F, Christensen B, et al. Red blood cell antibodies in pregnancy and their clinical consequences: synergistic effects of multiple specificities. Transfusion 2009;49:2070–5. [38] Koelewijn JM, Vrijkotte TGM, van der Schoot CE, Bonsel GJ, de Haas M. Effect of screening for red cell antibodies, other than anti-D, to detect hemolytic disease of the fetus and newborn: a population study in the Netherlands. Transfusion 2008;48:941–52. [39] Issitt PD, Anstee DJ. The MN Blood Group system. Applied Blood Group Serology. 4th ed. North Carolina: Montgomery Scientific, Publication; 1998 461–557.

[40] Reid ME. MNS blood group system: a review. Immunohematology 2009;25: 95–101. [41] Takeshita A, Watanabe H, Fijihara H, Oshida M, Yurugi K, Tomoda Y, et al. Collaborative study of irregular erythrocyte antibodies in Japan: results from the Japanese study group of allo-immunity and antigen diversity in Asian populations. Transfus Apher Sci 2010;43:3–8. [42] Gao XY, Huang H, Li LD. Hemolytic disease of neonates due to anti-M: report of one case and review of reports of 21 cases. Zhonghua Er Ke Za Zhi 2009;47:648–52. [43] Padiyar A, Hricik DE. Immune factors influencing ethnic disparities in kidney transplantation outcomes. Expert Rev Clin Immunol 2011;7:769–78. [44] Smith ML, Beck ML. The immunoglobulin structure of human anti-M agglutinins. Transfusion 1979;19:472–4. [45] Heathcote DJ, Carroll TE, Flower RL. Sixty years of antibodies to MNS system hybrid glycophorins: what have we learned? Transfus Med Rev 2011;25: 111–24. [46] Reid ME, Lomas-Francis C. Blood group antigen factsbook. 3rd ed. San Diego: Academic Press; 2012. 65–7. [47] Vaughan JI, Manning M, Warwick RM, Letsky EA, Murray NA, Roberts IAG. Inhibition of erythroid progenitor cells by anti-K antibodies in fetal alloimmune anemia. N Eng J Med 1998;338:798–803. [48] Nolan B, Hinchliffe R, Vora A, Letsky EA, Murray NA, Roberts IA. Neonatal pure red cell aplasia due to maternal anti-M. Blood 2000;96:8A [Abstract]. [49] Arndt PA, Garratty G, Daniels G, Green CA, Wilkes AM, Hunt P, et al. Late-onset neonatal anaemia due to maternal anti-Ge: possible association with destruction of eythroid progenitors. Transfus Med 2005;15:125–32. [50] Denomme GA, Shahcheraghi A, Blackall DP, Oza KK, Garratty G. Inhibition of erythroid progenitor cell growth by anti-Ge3. Br J Haematol 2006;133:443–4. [51] Blackall DP, Pesek GD, Montgomery MM, Oza KK, Arndt PA, Garratty G, et al. Hemolytic disease of the fetus and newborn due to anti-Ge3: combined antibodydependent hemolytic and erythroid precursor cell growth inhibition. Am J Perinatol 2008;25:541–5. [52] Pate LL, Myers JC, Palma JP, Viele M, Galel SA, Ferrer Z, et al. Anti-Ge3 causes lateonset hemolytic disease of the newborn: the fourth case in three Hispanic families. Transfusion 2013;53:2152–7. [53] Daniels G, Green C. Expression of red cell surface antigens during erythropoiesis. Vox Sang 2000;78:149–53. [54] Southcott MJG, Tanner MJA, Anstee DJ. The expression of human blood group antigens during erythropoiesis in a cell culture system. Blood 1999;93:4425–35. [55] Bony V, Gane P, Bailly P, Cartron JP. Time-course expression of polypeptides carrying blood group antigens during human erythroid differentiation. Br J Haematol 1999;107:263–74. [56] Gorlin JB, Beaton M, Poortenga M, Davidson S, Mullane S. Anti-Vw causing a case of clinically significant hemolytic disease of the newborn. Transfusion 1996;36: 938–9. [57] Waldvogel S, Natterer J, Canellini G, Truttmann A, Tissot JD. Severe neonatal hyporegenerative anemia due to anti-Vw (anti-MNS9) alloantibody. J Perinat Med 2009;37:422–4. [58] Reid ME, Lomas-Francis C. Blood group antigen factsbook. 3rd ed. San Diego: Academic Press; 2012. 81–2. [59] Sawada K, Fujishima N, Hirokawa M. Acquired pure red cell aplasia: updated review of treatment. Br J Haematol 2008;142:505–14. [60] Illanes S, Soothill P. Management of red cell alloimmunisation in pregnancy: the non-invasive monitoring of the disease. Prenat Diagn 2010;30:668–73.