Research www. AJOG.

org

OBSTETRICS
Pediatric outcome in Rhesus hemolytic disease treated
with and without intrauterine transfusion
Inge P. De Boer, MD; Eliane C. M. Zeestraten, MD; Enrico Lopriore, MD, PhD; Inge L. van Kamp, MD, PhD;
Humphrey H. H. Kanhai, MD, PhD; Frans J. Walther, MD, PhD

OBJECTIVE: To study the short-term morbidity in Rhesus hemolytic
disease of infants treated either with or without intrauterine transfu-
sions (IUT).
STUDY DESIGN: All term and near term infants (gestational age ⱖ36
weeks) with neonatal Rhesus hemolytic disease admitted to our center
between January 2000-March 2005 were retrospectively included in
the study. We recorded the duration of phototherapy, the need of ex-
change transfusions, and the need of top-up red blood cell transfu-
sions until 6 months of age.
RESULTS: A total of 89 infants were included, of whom 52 received at
least one IUT. Duration of phototherapy in the IUT and no-IUT group
was 3.8 and 5.1 days, respectively (P ⫽ .01). The percentage of infants
requiring an exchange transfusion in the IUT group was 71% com-
pared to 65% in the no-IUT group (P ⫽ .64). The percentage of infants
requiring a top-up transfusion in the IUT and no-IUT group was 77%
and 26.5%, respectively (P ⬍ .01).
CONCLUSION: Infants with Rhesus hemolytic disease treated with IUT
required less days of phototherapy and more top-up red blood cell
transfusions than neonates without IUT. However, the need for ex-
change transfusion was similar in both groups.
Key words: anemia, hyperbilirubinemia, intrauterine transfusion,
pediatric outcome, Rhesus hemolytic disease

Cite this article as: De Boer IP, Zeestraten ECM, Lopriore E, et al. Pediatric outcome in Rhesus hemolytic disease treated with and without intrauterine
transfusion. Am J Obstet Gynecol 2008;198:54.e1-54.e4.

S ince the introduction of Rhesus(D)

immunoprophylaxis in 1965, the inci-
dence of perinatal Rhesus hemolytic dis-
ease (RHD) has decreased dramatically.
Antenatal treatment with intrauterine
blood transfusions (IUTs) further reduced
perinatal mortality and morbidity in se-
verely anemic fetuses.1,2 However, fetal
anemia, hydrops, and neonatal icterus
caused by maternal red cell alloimmuniza-
tion still occur.3 Postnatal treatment of
From the Division of Neonatology,
Department of Pediatrics (Drs De Boer,
Zeestraten, Lopriore, and Walther), and the
Division of Fetal Medicine, Department of
Obstetrics (Drs van Kamp and Kanhai),
Leiden University Medical Center, Leiden,
the Netherlands.
Received Dec. 14, 2006; revised May 8,
2007; accepted May 22, 2007.
Reprints: Enrico Lopriore, MD, PhD, Division
of Neonatology, Department of Pediatrics,
Leiden University Medical Center, PO Box
9600, 2300 RC Leiden, The Netherlands;
e.lopriore@lumc.nl
0002-9378/$34.00
© 2008 Mosby, Inc. All rights reserved.
doi: 10.1016/j.ajog.2007.05.030
RHD includes phototherapy (PT), ex-
change transfusion (ET), and top-up red
blood cell transfusion. PT lowers bilirubin
through photooxidation, whereas ET re-
moves bilirubin and maternal red blood
cell antibodies. Although the mortality rate
associated with ET in term infants is now
⬍ 0.3%, the morbidity rate is at least
5%.4-7 Top-up red blood cell transfusions
correct late anemia and have, as all donor
transfusions, a minimal but theoretical risk
for anaphylactic reaction and transmission
of viral diseases.
Limited studies are available on the ef-
fect of IUT on postnatal outcome in in-
fants with RHD.8,9 Moreover, most
studies are based on small and nonho-
mogeneous groups and include infants
with comorbidity, such as prematurity,
perinatal asphyxia, and fetal hy-
drops.10,11 The aim of this study was to
evaluate the short-term clinical outcome
in a large, homogeneous group of infants
with RHD but without comorbidity,
treated with and without IUT.
M ATERIALS AND M ETHODS
All infants with a gestational age of ⱖ 36
weeks with Rhesus(D) hemolytic disease
(Rhesus D, Rhesus D in combination
with Rhesus C, or Rhesus E antigen), ad-
mitted between January 2000-March
2005 to the neonatal nursery of the Lei-
den University Medical Center (LUMC)
were included in the study. The LUMC is
the national referral center for the man-
agement and intrauterine treatment of
red cell alloimmunization in the Nether-
lands.1 Our methods for fetal monitor-
ing in pregnancies at risk for fetal hemo-
lytic disease and technical aspects of
intravascular intrauterine transfusion at
our institution have been described pre-
viously.1 Intrauterine transfusion ther-
apy was usually continued until 34-35
weeks, enabling pregnancy to progress to
term.1 For patients without IUT, neona-
tal hemolytic disease was defined as a
positive direct Coombs test in combina-
tion with either a maternal ADCC ⬎
30%12 or hyperbilirubinemia within the
first 4 hours of life.
We excluded neonates with perinatal
asphyxia (defined as an Apgar score ⬍7
at 5 minutes and evidence of multiorgan
dysfunction such as respiratory failure,
cardiac failure, anuria, or elevated liver
enzymes), sepsis (based on clinical

54.e1 American Journal of Obstetrics & Gynecology JANUARY 2008

www.AJOG.org
symptoms of infection or positive blood
culture), and cases with maternal red
blood cell alloimmunization other than
Rhesus(D).
The following obstetric data were re-
corded: fetal hemoglobin at first IUT and
number of IUTs. We recorded the fol-
lowing neonatal data: gestational age at
birth, birthweight, first available hemo-
globin level, reticulocyte count and bili-
rubin level at birth (umbilical cord or ve-
nous sample), maximum bilirubin level
during admission, duration of PT
(days), the number of ETs required,
and the number of top-up red blood
cell transfusions received during the
first 6 months of life. Data on the num-
ber of top-up red blood cell transfu-
sions and hemoglobin levels before
those transfusions, in infants receiving
follow-up outside the LUMC, were
gathered through correspondence with
the local pediatrician.
PT was administered with white light
PT devices (Air-Shields Phototherapy
System 7850, Healthdyne Company,
Hatboro, PA, and Ohmeda lamps, Ohm-
eda Medical, Columbia, MD) at a dis-
tance of 60 cm from the bed surface with
irradiance of 15-25 ␮W/cm/nm in com-
bination with biliblanket (biliblanket,
Ohmeda Medical) blue light, with irradi-
ation of 19-45 ␮W/cm/nm. ET was per-
formed with double-volume transfusion
(160 mL/kg) using irradiated and leuko-
cyte-depleted erythrocytes.
Criteria for ET were: bilirubin level at
birth ⬎ 3.5 mg/dL and bilirubin levels
above threshold13 in combination with
failure of PT (rise of bilirubin value
above threshold ⬎ 0.5 mg/dL/hr with in-
tensive PT). In the no-IUT group, a he-
moglobin level at birth ⬍ 12.9 g/dL (8
mmol/L) was an additional criterion for
ET. In the IUT group, a hemoglobin level
at birth ⬍ 9.7 g/dL (6 mmol/L) was an
indication for red cell blood transfusion
on day 1. Top-up transfusions were per-
formed after initial discharge from the
hospital when hemoglobin levels were ⬍
8.0 g/dL (5 mmol/L) or at higher levels if
clinical symptoms of anemia (lethargy,
feeding problems, or failure to thrive)
were present.
Primary outcome variables were dura-
tion of PT (days) and need for both ET
Obstetrics Research
TABLE 1
Baseline characteristics in neonates with RHD treated with and
without IUT
No-IUT group IUT group
(n ⴝ 37) (n ⴝ 52)
Male (n [%]) 18 (49) 27 (52)
..............................................................................................................................................................................................................................................
Gestational age at birth (wk) a
38 (36-40) 37 (36-39)
..............................................................................................................................................................................................................................................
Birthweight (g) b
2964 ⫾ 484 2997 ⫾ 390
..............................................................................................................................................................................................................................................
Neonates with Rhesus type D (n [%]) 21 (57) 21 (40)
..............................................................................................................................................................................................................................................
Neonates with Rhesus type D/C (n [%]) 13 (35) 25 (48)
..............................................................................................................................................................................................................................................
Neonates with Rhesus type D/E or D/C/E (n [%]) 2 (5) 6 (12)
..............................................................................................................................................................................................................................................
Number of IUTs a
– 3 (1-6)
..............................................................................................................................................................................................................................................
a
Gestational age at first IUT (wk) – 28 (17-35)
..............................................................................................................................................................................................................................................
a
Hemoglobin level at first IUT (g/dL) – 6.9 (1.4-11.7)
..............................................................................................................................................................................................................................................
a
Median (range).
b
Mean ⫾ SD.
De Boer. Pediatric outcomes in Rhesus hemolytic disease. Am J Obstet Gynecol 2008.
and top-up transfusions. Outcome was Duration of PT
compared between patients treated with Complete information on PT treatment
and without IUT. In addition, the effect was obtained for 67/89 (75%) of the pa-
of ET on the need for top-up red blood tients, having fully completed their PT
cell transfusions in the first 6 months of treatment in our hospital. Overall mean
life was studied. duration of PT during admission to our
Statistical analysis was carried out us- nursery was 4 days. Duration of PT in the
ing ␹2 test, Mann–Whitney test, and Stu- IUT and no-IUT group was 3.8 and 5.1
dent t test, as appropriate. To assess the days, respectively (P ⫽ .01; Table 2).
relationship between the number of There was no correlation between the
IUTs administered and the number of number of IUTs and the duration of PT
(Spearman correlation coefficient ⫽
ETs and top-up red blood cell transfu-
⫺0.21; P ⫽ .17). The mean maximum
sions, a Spearman correlation coefficient
bilirubin levels during admission were
was computed. P ⬍ .05 was considered
significantly lower in the IUT group
statistical significance. Statistical analysis
(12.5 mg/dL) than in the no-IUT group
was performed with SPSS 12.0 (SPSS Inc,
(15.4 mg/dL; P ⬍ .01).
Chicago, IL).
Use of ET
Information on ET was available for all
R ESULTS
included patients (100% follow-up) be-
During the study period, 141 neonates
cause all ETs were performed at our in-
with hemolytic disease were admitted to
stitution. Overall, at least 1 ET was re-
our neonatal nursery. Fifty-two patients
quired in 69% (61/89) of the patients.
were excluded for the following reasons: Thirty-nine patients (44%) only re-
gestational age ⬍36 weeks (n ⫽ 31), he- quired 1 ET, 17 patients (19%) required
molytic disease caused by Kell, Duffy, or 2 ETs, 4 patients (4%) needed 3 ETs, and
Rhesus (c, E, or C) immunization (n ⫽ 1 patient (1%) was treated with 5 ETs.
12), sepsis (n ⫽ 6), and perinatal as- The median number of ETs in the IUT
phyxia (n ⫽ 3). A total of 89 patients group and no-IUT group was similar
were included in this study, of which 52 (Table 2). No significant relationship
(58%) were treated with IUT and 37 was found between the number of IUTs
(42%) without IUT. Baseline character- and the number of ETs (Spearman cor-
istics are summarized in Table 1. relation coefficient ⫽ ⫺0.14; P ⫽ .30)
JANUARY 2008 American Journal of Obstetrics & Gynecology 54.e2
Research Obstetrics www.AJOG.org

TABLE 2
Clinical outcome in neonates with RHD treated with and without IUT
No-IUT group IUT group
(n ⴝ 37) (n ⴝ 52) P value
a
Hemoglobin level at birth (g/dL) 12.3 11.0 .02
................................................................................................................................................................................................................................................................................................................................................................................
a
Bilirubin level at birth (mg/dL) 5.8 5.8 .80
................................................................................................................................................................................................................................................................................................................................................................................
Reticulocyte count at birth (%) b
73 (8-294) 7 (0-108) ⬍ .01
................................................................................................................................................................................................................................................................................................................................................................................
a
Phototherapy (d) 5.1 3.8 .01
................................................................................................................................................................................................................................................................................................................................................................................
Neonates requiring ET (n [%]) 24 (65) 37 (71) .64
................................................................................................................................................................................................................................................................................................................................................................................
Median number of ETs (range) 1 (0-3) 1 (0-5) .70
................................................................................................................................................................................................................................................................................................................................................................................
Neonates requiring top-up transfusion (n [%]) 9/34 (26.5) 30/39 (77) ⬍ .01
................................................................................................................................................................................................................................................................................................................................................................................
Median number of top-up transfusions (range) 0 (0-2) 1 (0-4) ⬍ .01
................................................................................................................................................................................................................................................................................................................................................................................
Treatment with PT was assessed in 67 and need for top-up red cell transfusions in 73 neonates.
a
Mean.
b
Median (range).
De Boer. Pediatric outcomes in Rhesus hemolytic disease. Am J Obstet Gynecol 2008.

The first ET in the IUT group and no- was 55% (28/51) and 50% (11/22), re- reduce the need for neonatal intensive
IUT group was performed at an average spectively (P ⫽ .45). There was no rela- treatment. Our present finding that neo-
time after birth of 6 hours and 5.5 hours tion between the number of ETs and the nates without IUT need phototherapy
of life, respectively. Most ETs (88%, 78/ need for top-up blood transfusions. for a longer time than neonates with IUT
89) were performed within the first 12 matches this hypothesis and the results
hours after birth. The mean bilirubin of other studies.8
levels at first ET in the IUT group and C OMMENT However, the need for ET in this study
no-IUT group were 9.3 mg/dL and 11.6 This study shows that neonates with was comparable in neonates treated with
mg/dL, respectively. RHD treated with IUT still need inten- and without IUT, suggesting a similar
sive neonatal treatment, including ETs. degree of hemolysis in both groups.
Use of top-up red blood cell Neonates with RHD treated with IUTs Moreover, we found no relation between
transfusions had a different short-term clinical course the number of IUTs and the number of
Complete information on top-up red compared to neonates with RHD who ETs. Again, the low percentage of fetal
blood cell transfusions were obtained for did not receive IUTs, requiring fewer PT red blood cells in neonates treated with
73/89 (82%) of the patients because not days but more top-up red blood cell IUT should theoretically lead to a re-
all pediatricians responded to our letter. transfusions. However, the need for ET
duced hemolysis and less need for ET.
Top-up red blood cell transfusions were was similar in both groups.
This paradoxical result may be related to
administered at a mean hemoglobin Intrauterine transfusions are per-
the specific early ET criteria used in this
level of 8.2 g/dL. The percentage of neo- formed in fetuses with severe hemolysis
study and explain the higher incidence of
nates in the IUT and no-IUT group re- resulting in fetal anemia.1 In severely af-
ET compared to other studies.8,9 Most
quiring a top-up red blood cell transfu- fected fetuses, IUTs will adequately cor-
ETs in this study (88%) were performed
sion was 77% (30/39) and 26.5% (9/34), rect fetal anemia. Because IUTs are only
respectively (P ⬍ .01; Table 2). The me- performed in anemic or even hydropic within the first 12 hours after birth, sec-
dian number of top-up transfusions in fetuses, intrauterine treatment may the- ondary to an aggressive approach used at
the IUT and no-IUT group was 1 (range, oretically be associated with the need for our institution. In most other studies, in-
0-4) and 0 (range, 0-2), respectively (P ⬍ more intensive neonatal treatment. On dications for ET, phototherapy devices,
.01). We performed a linear regression the other hand, repeated IUTs are asso- and irradiance as well as accepted biliru-
analysis to assess the relation between ciated with a reduction in fetal hemolysis bin levels are not accurately described.
the reticulocyte count at birth and the by the replacement of fetal red blood This makes it difficult to compare the re-
number of top-up transfusions. A lower cells by Rhesus negative donor red blood sults and find an explanation for the rel-
reticulocyte count was correlated with a cells. We previously reported that most atively high ET incidence in our study.
higher number of top-up transfusions fetal red blood cells have disappeared at An alternative hypothesis for the high
(Spearman correlation coefficient: the second IUT.8,11,14 The replacement rate of ETs in the IUT group could be
⫺0.43; P ⫽ .05). The percentage of neo- of fetal cells by donor red cells should that elevated bilirubin levels persist in
nates in the ET and no-ET group requir- theoretically result in a less active neona- these fetuses despite replacement with
ing a top-up red blood cell transfusion tal alloimmune hemolysis and therefore adult hemoglobin-containing cells.15

54.e3 American Journal of Obstetrics & Gynecology JANUARY 2008

www.AJOG.org
Recently, the therapeutic and prophy-
lactic value of early ET to prevent ker-
nicterus has been questioned.16 Several
new guidelines suggest that the early cri-
teria for ET should be abandoned.6,16
The American Academy of Pediatrics
(AAP) recently published new bilirubin
thresholds for PT and ET indications in
RHD.6 These guidelines were instituted
in our nursery after completion of this
study. They suggest that except for inten-
sive PT, intravenous immunoglobulin
administration may also reduce the need
for ET.6 However, a recent Cochrane re-
view indicates that the evidence thus far
is not sufficient to recommend the rou-
tine use of immunoglobulin in RHD.16
Given the conflicting recommendations,
we recently started a randomized con-
trolled trial for the use of intravenous
immunoglobulins in RHD. Alterna-
tively, a recent study suggests that mater-
nal phenobarbital may also reduce the
need for ET by enhancing the ability of
the neonate to conjugate bilirubin.17
Our study also demonstrates that ne-
onates with RHD treated with IUT re-
quire more top-up transfusions during
the first 6 months of life. This result is
consistent with published reports and is
explained by a decrease in erythropoie-
sis.11,18 We indeed found a significantly
lower reticulocyte count at birth in in-
fants after IUT when compared with
those who received only neonatal ther-
apy. As shown in this study, low reticu-
locyte count at birth is associated with an
increased need for top-up transfusions.
Use of ET in this study was not associ-
ated with a reduction of duration of PT
and top-up transfusions. These findings
are in accordance with the results of
Ebbenson,19 but in contradiction with al
Alaiiyan,10 who reported a significant
decrease in the incidence of late anemia
in patients who were treated with ET.
Our findings do not support the theory
that ET prevents late anemia by washing
out red blood cell antibodies.10 How-
ever, level of maternal red cell antibodies
was not measured in our study.
In conclusion, although the need for
ET in term and near-term neonates with
RHD treated with and without IUT is
similar, neonates in the IUT group re-
quired less PT and more top-up red
blood cell transfusions. f cytotoxicity assay in the management of Rh D
REFERENCES
1. van Kamp IL, Klumper FJ, Meerman RH,
Oepkes D, Scherjon SA, Kanhai HH. Treatment
of fetal anemia due to red-cell alloimmunization
with intrauterine transfusions in the Nether-
lands, 1988-1999. Acta Obstet Gynecol Scand
2004;83:731-7.
2. van Kamp IL, Klumper FJ, Bakkum RS, et al.
The severity of immune fetal hydrops is predic-
tive of fetal outcome after intrauterine treatment.
Am J Obstet Gynecol 2001;185:668-73.
3. Bowman JM. The prevention of Rh immuni-
zation. Transfus Med Rev 1988;2:129-50.
4. Jackson JC. Adverse events associated with
exchange transfusion in healthy and ill new-
borns. Pediatrics 1997;99:E7.
5. Ip S, Chung M, Kulig J, et al. An evidence-
based review of important issues concerning
neonatal hyperbilirubinemia. Pediatrics
2004;114:e130-e153.
6. American Academy of Pediatrics Subcom-
mittee on Hyperbilirubinemia. Management of
hyperbilirubinemia in the newborn infant 35 or
more weeks of gestation. Pediatrics
2004;114:297-316.
7. Patra K, Storfer-Isser A, Siner B, Moore J,
Hack M. Adverse events associated with neo-
natal exchange transfusion in the 1990s. J Pe-
diatr 2004;144:626-31.
8. Weiner CP, Williamson RA, Wenstrom KD, et
al. Management of fetal hemolytic disease by
cordocentesis. II. Outcome of treatment. Am J
Obstet Gynecol 1991;165:1302-7.
JANUARY 2008 American Journal of Obstetrics & Gynecology
Obstetrics Research
9. Farrant B, Battin M, Roberts A. Outcome of
infants receiving in-utero transfusions for hae-
molytic disease. N Z Med J 2001;114:400-3.
10. al Alaiyan S, al Omran A. Late hyporegen-
erative anemia in neonates with rhesus
hemolytic disease. J Perinat Med 1999;27:
112-5.
11. Millard DD, Gidding SS, Socol ML, et al.
Effects of intravascular, intrauterine transfusion
on prenatal and postnatal hemolysis and eryth-
ropoiesis in severe fetal isoimmunization. J Pe-
diatr 1990;117:447-54.
12. Oepkes D, van Kamp IL, Simon MJ, Mes-
man J, Overbeeke MA, Kanhai HH. Clinical
value of an antibody-dependent cell-mediated
alloimmunization. Am J Obstet Gynecol
2001;184:1015-20.
13. Osborn LM, Lenarsky C, Oakes RC, Reiff
MI. Phototherapy in full-term infants with hemo-
lytic disease secondary to ABO incompatibility.
Pediatrics 1984;74:371-4.
14. Egberts J, van Kamp IL, Kanhai HH, Meer-
man RH, Giordano PC, Gravenhorst JB. The
disappearance of fetal and donor red blood
cells in alloimmunised pregnancies: a reap-
praisal. BJOG 1997;104:818-24.
15. Goodrum LA, Saade GR, Belfort MA, Car-
penter RJ Jr, Moise KJ Jr. The effect of intra-
uterine transfusion on fetal bilirubin in red cell
alloimmunization. Obstet Gynecol 1997;89:
57-60.
16. Alcock GS, Liley H. Immunoglobulin infu-
sion for isoimmune haemolytic jaundice in neo-
nates. Cochrane Database Syst Rev 2002;
CD003313.
17. Trevett TN Jr, Dorman K, Lamvu G, Moise
KJ Jr. Antenatal maternal administration of phe-
nobarbital for the prevention of exchange trans-
fusion in neonates with hemolytic disease of the
fetus and newborn. Am J Obstet Gynecol
2005;192:478-82.
18. Thorp JA, O’Connor T, Callenbach J, et al.
Hyporegenerative anemia associated with intra-
uterine transfusion in rhesus hemolytic disease.
Am J Obstet Gynecol 1991;165:79-81.
19. Ebbesen F. Late anaemia in infants with
rhesus haemolytic disease treated with inten-
sive phototherapy. Eur J Pediatr 1979;
130:285-90.
54.e4