Intravenous anti-D as a treatment for immune thrombocytopenic purpura (ITP) during pregnancy

Authors


Marc Michel, The New York Presbyterian Hospital, Department of Pediatrics, Division of Pediatric Hematology/Oncology, Weill Medical College of Cornell University, 525 East 68th street, New York, NY 10021, USA. E-mail: drmarcmichel@hotmail.com

Abstract

Summary. This pilot study assessed the safety and efficacy of intravenous anti-D in eight Rh(D)-positive women with immune thrombocytopenic purpura (ITP) during the second and third trimesters of pregnancy. The median pretreatment platelet count was 28 × 109/l. The patients received one to seven anti-D infusions at a mean dose of 62·7 µg/kg, and the response rate to anti-D was 75%. A haemoglobin decrease of > 2·0 g/dl occurred only once. Fetal hydrops was not identified by ultrasonography. The direct antiglobulin test was positive in three out of seven Rh+ newborns, none of whom was anaemic or jaundiced. Anti-D is effective and appears to be safe for both mother and fetus.

Immune thrombocytopenic purpura (ITP) occurs commonly in women of child-bearing age (Fredriksen & Schmidt, 1999). The management of ITP during pregnancy must take into account certain features: (1) a platelet count of 50 × 109/l is recommended for delivery and a count ≥ 80 × 109/l is usually required for epidural anaesthesia (Bromage, 1993; British Committee for Standards in Hematology General Haematology Task Force, 2003); and (2) its management primarily involves steroids and intravenous immunoglobulin (IVIG) (Gill & Kelton, 2000). Anti-Rh (anti-D) has been used for 40 years, including antepartum dosing, to prevent rhesus alloimmunization during pregnancy in Rh(D)-negative women (Moise, 2001). In Rh(D)-positive patients, with ITP and at doses 10- to 20-fold higher than those used for the prophylaxis of Rh disease, anti-D provides a significant platelet increase in > 70% of cases (Newman et al, 2001). At such doses, post-infusion reactions (i.e. fever, chills, headache) are common, and anti-D can lead to some degree of haemolysis (Scaradavou et al, 1997). Compared with IVIG, anti-D has the advantages of being less expensive, having a much shorter infusion time and being derived from a smaller donor pool (Blanchette & Carcao, 1998). Its use in ITP during pregnancy has been limited by the fear of fetal/neonatal haemolysis and the lack of data describing its efficacy in pregnant women. This prospective pilot study was performed to investigate the safety and efficacy of anti-D in pregnant women with ITP.

Study design

Eligible women had ITP (George et al, 1996) requiring treatment, a platelet count ≤ 55 × 109/l, were Rh(D) positive and in the second or third trimester of pregnancy. Exclusion criteria were previous splenectomy, ongoing haemorrhage, history of serious reactions to plasma products, haemoglobin < 9·0 g/dl and treatment with anti-D or IVIG within 1 week before entry to the study. Prednisone was allowed if at a stable dose of ≤ 20 mg/d for at least 2 weeks before study entry. Institutional review board-approved informed consent was obtained from all patients.

Anti-D (WinRhoSDF; Cangene, Winnipeg, Canada) was infused at 50–70 µg/kg over 3–15 min. To minimize reactions, paracetamol (650 mg) and prednisone (20 mg) were administered as premedications.

Fetal sonograms were performed at inclusion and repeated monthly until delivery. Complete blood counts (CBC) were initially performed weekly and then twice a month until delivery. Blood type, direct antiglobulin test (DAT), CBC, reticulocyte count and bilirubin level were measured in the neonates.

Partial response (PR) to anti-D was defined as a platelet level > 50 × 109/l and an increase of at least 20 × 109/l within 1 week of treatment; and a complete response (CR) was defined as a platelet count ≥ 150 × 109/l.

Results

Mothers

Eight women with a median age of 36 years and a median platelet count of 28 × 109/l were included. The median duration of ITP was 5 months, and the median gestational age was 34 weeks (Table I). They each received one to seven infusions of anti-D at a median dose of 4500 µg (62 µg/kg). Six patients (75%) responded (one CR, five PR) by d 7 after the first infusion. The median platelet increase was 54 × 109/l (25–125 × 109/l). During the pregnancy, two patients (nos 3 and 6) received only anti-D; three patients responded to anti-D while continuing 10–20 mg/d prednisone (nos 2, 4 and 8); and three, including the two non-responders (nos 1, 5 and 7), received additional therapy before delivery (Table I).

Table I.  Baseline characteristics of the mothers, efficacy and tolerance of anti-D.
Case no./
age (years)
Gestational age at
inclusion (weeks)
Duration of
ITP (months)
Previous treatment
for ITP
Anti-D infusion (no.), dose (µg)/ time between infusions (d)Platelet count (× 109/l)
before anti-D
Platelet count (× 109/l) on d 7 after each IV anti-DOther medication
for ITP
Related adverse events
and comments*
  1. MP, methylprednisolone; pred., prednisone; IVIG, intravenous immunoglobulins; Hb, haemoglobin; del., delivery. Patient 1 had previously been treated successfully with anti-D and wished to receive this same treatment during pregnancy. Patient 5 was treated with anti-D as she did not want to take prednisone during her pregnancy.

1/281524Pred, IVIG,
IV anti-D
(1) 3600
(2) 3600/58
(3) 3600/34
(4) 3600/27
(5) 4200/33
(6) 3000/10
(7) 4500/11
31
22
23
28
29
90*
66
 74
 78
 59
 64
 79
 74
 73




MP 1 g, IVIG 1 g/kg
MP 1 g, IVIG 1 g/kg
MP 1 g, IVIG 1 g/kg
Headache and nausea after
2nd and 3rd infusions,
dizziness and nausea
after 7th infusion.
*Received IVIG and MP in combination with anti-D
as she was supposed to deliver
a week later and was feared to relapse.
2/23384Pred.(1) 450039103Ongoing pred. 20 mg/dNone
3/36315Pred.(1) 450020136NoneNone
4/36346Pred.(1) 570017  57Ongoing pred. 20 mg/dNone
5/37141Pred, IVIG(1) 6000
(2) 6600/24
(3) 6600/77
(4) 6600/58
27
16
32
16
 43
 95
 66 (d 3)
 3 (d 4)



IVIG 1 g/kg
Chills, fever, headache and
fatigue 2–5 d after each
infusion, haemolytic anaemia
(Hb decrease of 3·2 g/dl)
6/373723IV anti-D(1) 450055*120NoneHeadache, eczema
*Was scheduled for a caesarean
section 7 d later
7/31361None(1) 600010  47 (d 3)IVIG 1 g/kg after
d 3 (just before del.)
Chills, myalgias
8/46204Pred.(1) 390046*171Pred. was tapered from
20 to 10 mg/d
None
*Was treated with anti-D
to allow the tapering of pred.
that was not well tolerated

Four patients (50%) complained of mild adverse events after anti-D (Table I). The median haemoglobin decrease was 1·1 g/dl; only patient 5 had a decrease > 2·0 g/dl (3·2 g/dl).

Fetuses

Hydrops fetalis was not detected on ultrasonography. The median gestational age at birth was 38 weeks, the median birthweight was 2·9 kg, and the Apgar scores were ≥ 8 (Table II). Seven babies were Rh(D) positive, and the DAT was positive in three; one case (no. 3) resulted from an ABO incompatibility as demonstrated by elution. Only one (DAT negative) newborn (no. 4) was anaemic (see Table II). The median bilirubin level was 40·2 μmol/l at birth (normal 30·8–50 μmol/l). Only one baby (no. 6) had jaundice, which resolved promptly with phototherapy. His DAT was negative, he was not anaemic, was breastfed, and his older sibling had also had jaundice requiring phototherapy after birth.

Table II.  Characteristics of newborns.


Infant
Gestational age
at birth
(weeks)
Last anti-D infusion
received by the mother
(d before del.)

Birthweight
(kg)
Apgar
score
(1 min/5 min)

Blood
type


DAT

Hb
(g/dl)

Retic.
(%)
Lowest platelet
count (× 109/l)
after birth (d)
Neonatal/
unconjugated
bilirubin (µmol/l)
Clinical examination
within 5 d
after birth
  • DAT, direct antiglobulin test; Hb, haemoglobin (normal value in newborns = 18–21·5 g/dl); retic., reticulocytes; NA, not available; del., delivery; +, positive; –, negative.

  • *

    Positive DAT because of ABO incompatibility (anti-A).

  • This Hb level was likely to be erroneous because the CBC was performed on cord blood; less than 24 h later, the Hb level was found to be 19 g/dl without transfusion.

  • Not related to IV anti-D.

  • §

    On d 3 after birth, neonatal/unconjugated bilirubin level rose up to a maximum of 323·2/314·6 µmol/l (normal neonatal bilirubin level = 0·3–180 µmol/l).

140 −22·99/9O++19·15·26140 (1)22·2/20·5Normal
240 −143·39/9A++16·04·03102 (2)41/35·9Normal
335 −263·29/9A++*17·64·7993 (8)53/49·6Normal
439 −422·98/8A+11·54·64250 (1)22·2/NANormal
536 −52·99/9O+23·0NA220 (1)NAExtra digit/toe on left hand and foot,
immature reflex system
638 −73·39/9O+20·03·151 (5)39·3/37·6Jaundice on d 3§
(resolved with
phototherapy)
738 −103·08/9O+16·14·96327 (1)27·4/25·7Normal
835 −34NA9/9O17·4NA200 (1)75·2/71·8Normal

Discussion

The management of ITP during pregnancy may be difficult for several reasons: IVIG and/or prednisone may be ineffective or toxic; it is best to avoid intrapartum splenectomy; and a platelet count ≥ 50 × 109/l is usually required for delivery (Letsky & Greaves, 1996). Therefore, it was logical to test anti-D treatment of ITP during pregnancy. The widespread, antepartum use of anti-D for Rh prophylaxis meant that there was less concern planning this pilot study than there would have been for another medication not previously used in pregnancy.

In this study, six out of eight patients responded after the first treatment, similar to the 72% response rate reported outside pregnancy (Scaradavou et al, 1997). However, four out of six responders received additional therapy to attain a platelet count adequate for delivery or for epidural anaesthesia. Therefore, although anti-D is effective, supplementary treatment with corticosteroids and IVIG may be required, especially at term.

As expected, the toxicity of anti-D in mothers was comparable to that observed outside pregnancy with mild reactions in half the patients and a single case of haemoglobin decrease > 2 g/dl.

Anti-D is a polyclonal antibody predominantly of IgG subclasses I and III that could cross the placenta. In this study, although four out of eight mothers received treatment at least 2 weeks before delivery and two received multiple infusions, no cases of hydrops fetalis or neonatal haemolysis were observed. The high avidity of anti-D for Rh(D)-positive maternal red cells presumably facilitated the prompt removal of ‘free’ anti-D from the maternal circulation, limiting its availability for transplacental passage. Infrequent cases of fetal and/or neonatal haemolysis might have occurred in a larger study.

This study provides evidence for both the efficacy and the safety of anti-D in ITP during the second and third trimesters of pregnancy. However, the number of patients included was relatively low so, pending additional data, we believe that anti-D during pregnancy should be limited to the following situations: (1) as a second-line therapy in patients who do not tolerate or respond to steroids and/or IVIG; and (2) if needed, in combination with IVIG and/or steroids to achieve a safe platelet count at term. In both situations, monitoring of the fetus and newborn is recommended given the risk of haemolysis.

Acknowledgment

Dr Michel was partially supported by the Laboratoire Français du Fractionnement et des Biotechnologies (LFB), Les Ulis, France.

Ancillary