Management of Immune Haemolytic Disease in the Fetus

Y. Tannirandorn,

C. H. Rodeck

SUMMA R Y. Although Rh aUoimmmdzation has been successfully reduced in frequency and severity since the implementation of Rh immune globulin, cases still occur. The management of affected pregnancies requires the efforts of a team which includes obstetrics/fetal medicine, the blood transfusion service, haematological support, musing assistance and neonatology. The aim of antenatal mauagement is to predict whether or not the fetus is severely affected, to correct the fetal anaemia and to deliver the baby at tbe optimal time. The management has improved markedly with the introduction of high-resohrtion real-time ultrasod, fetal blood sampling, intravascular fetal blood transfusion and/or intraperitoneal transfusion and meticulous fetal surveiIlance. With appropriate and timely management in severely alloimmmdxed patient, the survival rate of affected fetuses in some centres is now about 90%. There is still a need for research into new methods of treatment such as high dose intravenous immuuogMmRn, which might non-invasively diminish fetal red cell destruction. Due to the reduced frequency of severe disease, regionalized treatment centres are essential in order to maximize the experience and etticiency of the management teams.

The incidence of severe haemolytic disease of the fetus and newborn due to blood group alloimmunization has been significantly reduced by the use of prophylactic anti-D immunoglobulin given to the mother.’ However, there are still a small number of patients in whom anti-D immunoglobulin prophylaxis fails, or who become sensitized to the other ‘atypical antigens, such as Kell. Once sensitization has occured and the fetus has developed severe erythroblastosis fetalis, prolonging life in utero is dependent on intrauterine management. In recent years the number of doctors with sufkient experience to be confident in the management of the problem has diminished. In addition, there have been great technical advances Y. Tamimmlon.~ RSc, MD, C. H. Rodeck BSc, MR RS, FRCOC, Fetal Medicine Unit, Royal Postgraduate Medical School, Institute of Obstetrics and Gynaecology, Queen Charlotte’s and Chelsea Hospital, Goldhawk Road, London, W6 OXG, UK. Correspondence to: C. H. R&k, Department of Obstetrics and Gynaecology, University College and Middlesex School of Medicine, London, WCIE 6HX, UK. Blood Reviews (1991) 5, 00-014 Q 1991 Longman Group UK Ltd

which have led to more effective management. Of central importance has been access to the fetal circulation which has contributed to a better understanding of pathophysiology of the disease and has influenced both assessment and therapy. Pathological Changes in Haemolytic Disease in the Fetus Maternal IgG anti-D antibodies gain access to the fetal circulation. In D-positive fetuses, such antibodies are both adsorbed to the D-positive erythrocytes and exist unbound in fetal serum. The adsorbed antibodies act as a haemolysin, leading to an accelerated rate of red cell destruction. This process may be so mild that the infant is normal at birth and requires no treatment, or it may be so severe that the fetus becomes hydropic and dies in utero.2 The earlier this process begins in utero and the greater its intensity, the more severe will be the effects upon the fetus.

2

MANAGEMENT

OF IMMUNE HAEMOLYTIC DISEASE IN THE FETUS

Hydrops fetalis

Table 1 Prerequisites for fetal blood sampling

The pathological changes in the organs of the fetus vary with the severity of the process. The severely affected fetus may show signs of hydrops fetalis (ascites, pericardial and pleural effusions, subcutaneous oedema, polyhydramnios and placentomegaly) which can be diagnosed in utero by ultrasonography. Excessive and prolonged haemolysis serves to stimulate marked erythroid hyperplasia of the bone marrow as well as large areas of extramedullary haematopoiesis,3 particularly in the liver and spleen, There may be cardiac enlargement and pulmonary haemorrhages. Pleural effusions may be so severe as to compromise respiration after birth. The pathophysiology of hydrops remains obscure. Theories of its causation include heart failure and/or capillary leakage caused by hypoxia, due to severe anaemia; portal and umbilical venous hypertension from hepatic parenchymal disruption by extramedullary haematopoiesis; and decreased colloid oncotic pressure from hypoproteinaemia caused by liver dysfunction or capillary endothelial leakage. The degree and duration of anaemia influence the severity of hydrops. Second trimester hydropic fetuses have a haematocrit < 15% or haemoglobin ) Antibodies The titre of Rh (D) antibody (indirect Coombs’ titre) often does not correlate very well with the severity of the disease.23 It seems that not only the quantity but also the quality of antibodies may be important in predicting the severity i.e. IgG subclasses, Gm allotypes and the number of IgG bound molecules.24s25 Recently, observations in vitro and in vivo have shown that assays based on the interaction of sensitized red cells with Fc receptor bearing cells cellular cytotoxicity) with (antibody-dependent monocytes and lymphocytes and the phagocytosis assay known also as monocyte-monolayer assay Table 2

Fetal blood tests in alloimmunized pregnancies

Haematological indices: Haematocrit, haemoglobin, mean cell volume, red cell count, white cell count, platelet count Peripheral smear: reticulocyte count, nucleated red cell count, differential white cells, red blood cell and platelet morphology Kleihauer Betke test: %fetal cells Blood group serology: Precise blood grouping Direct Coombs’ test Biochemistry: Liver function: aspartate transaminase, alanine transaminase, gamma glutamyl transpeptidase, alkaline phosphatase, albumin and total protein. pH and blood gases

Amniocentesis Bilirubin is the end product of fetal red cell destruction. Although it is cleared placentally via the maternal circulation, small amounts enter the fetal enterohepatic circulation and pass out into the amniotic fluid via fetal pulmonary fluid and fetal urine. When amniotic fluid containing bilirubin is analyzed spectrophotometrically, a rise in absorption occurs, the delta OD 450, which is the change in optical density at 450 nm. Although the test is not difficult, interpretation is not always easy and experience is required. Blood pigments may produce peaks at 405 and 420 nm, which contaminate the reading; meconium may raise the delta OD diffusely from 550-580, obscuring the 450 peak.27 Once a pure delta OD 450 peak has been obtained, it is plotted on a Liley chart which has three zones: Zone I, mildly or not at all affected; Zone II, mildly to moderately affected and Zone III, severely affected, treatment or delivery indicated.2* It must be pointed out that Liley’s original work focused on fetuses that were 27 weeks’ pregnancy or greater, therefore Liley’s lines should not be extrapolated backward, since it has been found that delta OD 450 is an unreliable indicator of the fetal condition before 24-26 weeks.29 Timing. The first amniocentesis should be performed 10 weeks before the time of the earliest previous fetal

BLOOD REVIEWS

or neonatal death, fetal transfusion, or birth of a severely affected baby. 2,23 In the absence of such a history, amniocentesis is undertaken when a level of maternal Rh (D) antibody is > 4 iu/ml and pregnancy is >26 weeks. Patients at risk of severe disease in the early and mid-second trimester should have FBS. The timing of repeat amniocentesis is dictated by the history, the value of the previous delta OD 450 and the trend of the changes in delta OD 450. High delta OD 450 levels are an indication for FBS and/or transfusion. Technique. When performing amniocentesis, it is important to avoid traversing the placenta since this can provoke feto-maternal haemorrhage.30 Singleoperator amniocentesis, one hand inserting the needle (20 or 22-gauge spinal needle), the other hand holding the transducer, with continuous visualisation is mandatory. Amniotic fluid is put into a light-proof container and immediately sent for analysis. The risk of premature rupture of membranes, infection and fetal injury from the procedure is very low. Ultrasound

Ultrasonographic evaluation of the fetus is an integral part of the modern management of Rh alloimmuniz-

Fig. 4 Ultrasound picture showing fetal ascitis

5

ation. Weekly ultrasound scans may detect fetal ascites which is an early sign of fetal hydrops3’ (Fig. 4) thereby facilitating timely intervention. However, in the absence of hydrops, ultrasound measurements of placental thickness, extra-and intra-hepatic vein diameters, abdominal circumference, head circumference, head/abdominal circumference ratio and intraperitoneal volume do not appear to be reliable in distinguishing mild from severe haemolytic disease.32-34

Fetal Doppler Measurements

Doppler ultrasound has been used as a test for prediction of fetal compromise, but the use of fetal Doppler flow velocity waveforms (Fig. 5) has also been proposed as an indirect means of assessing the fetal haematocrit. Kirkinen et al first reported a close correlation between umbilical venous blood velocities and postpartum haematocrits.3J Griffin et al reported a weak relationship of uncertain clinical significance between the time-average (maximal) velocity in the fetal descending aorta and the haematocrit.36 Rightmire et al similarly found in a retrospective study that a relationship was present between Doppler waveform indices and haematocrit.37 However, a prospective study failed to support the ability of the

6

MANAGEMENT

OF IMMUNE

HAEMOLYTIC

DISEASE IN THE FETUS

Fig. 5 Normal Doppler umbilical artery waveforms at 25 weeks’ gestation.

retrospective study derived formulas to predict haematocrit.38*39 Therefore at present Doppler studies appear to be of limited usefulness in the prediction of which alloimmunized pregnancies will require fetal transfusion, or in the timing of transfusions. Fetal Behaviour Fetal behaviour is assessed using the biophysical profile scores as a framework. Parameters observed are fetal tone, gross body and limb movements, breathing movements, amniotic fluid volume, and fetal heart rate variability and reactivity.40 Study of fetal behaviour in alloimmunized pregnancy is helpful for evaluation of fetal status, and in timing of transfusions. Lack of spontaneous fetal movements, muscle tone or breathing movements on ultrasound and abnormal fetal heart rates (FHR) are a common finding in moribund hydropic fetuses and indicate the need for immediate intravascular fetal transfusion.41 Fetal Heart Rate Patterns When fetal anaemia is severe enough to compromise the oxygen carrying capacity of blood, a cardiotocograph may show abnormal FHR patterns, such as decreased reactivity and baseline variability, tachycardias, spontaneous decelerations or sinusoidal patterns.42y43 However, the present criteria of normality

or abnormality of FHR patterns do not allow for an accurate prediction of the degree of fetal anaemia.44 Antibodies Other Than Anti-D Alloimmunization may develop to antigens other than D. Most, but not all atypical alloimmunization is produced by blood transfusion, since cross-matched blood is only compatible in the ABO blood system and for D in the Rh blood group system. Atypical alloimmunization develops in l-2% of individuals following blood transfusion.45 Many of the antibodies developed are of little or no clinical significance. However, c and Kell antibodies are capable of causing severe haemolysis in the fetus46 and are second only to anti-D as a cause of fetal death.’ Even though the principles of management are similar to those in Rh (D) haemolytic disease, several points must be stressed. With anti-c, confusion arises because this is a Rhesus antibody developing in a Rh (D) positive woman. Haemolytic disease due to anti-c tends to be less severe than that due to anti-D. However, the anaemia in the fetus may be severe enough to cause hydrops in utero if appropriate measures are not taken. Anti-Kell can also cause severe haemolysis in the fetus and in some cases the percentage of reticulocytes in the fetal blood is much lower than expected. 47 Furthermore, hydrops fetalis may develop very rapidly in cases in which amniotic fluid bihrubin measurement has indicated mild to

BLOOD REVIEWS

moderate disease.48 This evidence suggests that the antibody causes suppression of effective erythropoiesis in the fetus rather than haemolysis, so that the bilirubin concentration does not rise significantly in the amniotic fluid even in the presence of very severe anaemia. Management of the fetus at risk should be by a combination of early FBS for Kell grouping and assessment of anaemia, followed by frequent ultrasound screening if immediate treatment with transfusion is not indicated.

Mauagement The management of haemolytic disease in the fetus is similar regardless of the inciting antigen with the exception of Kell. Since D alloimmunization is most common, general management for this clinical situation is discussed. Since the introduction of intrauterine fetal transfusion for the management of Rh alloimmunization, conservative management of a severely affected fetus has become inappropriate. Expectant management is indicated only in the presence of stable serial maternal antibody titres, falling delta OD 450, normal antepartum fetal surveillance and in the absence of sonographic signs of fetal hydrops. The aim of management is to reach a safe gestational age for the fetus to be delivered, in a nonhydropic state. Conservative management can be pursued with more precision by determination of the severity of anaemia by FBS. Mackenzie et al suggested that if the anaemia is found to be mild (haematocrit >30%) serial FBS is the optimal management.49 However FBS is not a procedure without risk and our aim is to perform intrauterine transfusion at the same time, if the fetal haematocrit is more than 2SD below the mean for gestational age, and the fetus is too immature for delivery.

Intravascular Transfusion (IVT) The therapeutic advances in Rh alloimmunization are a direct result of access to the fetal circulation which enables intravascular transfusion (IVT). This technique was pioneered by Rodeck et a150s51using a fetoscopic approach but later, ultrasound guidance was used. Since then the technique has been introduced in many centres around the world. The advantages of transfusing directly into the fetal intravascular space are that (1) fetal grouping can be confirmed; (2) pre- and post-transfusion haematocrit can be measured; (3) the transfused red cells are not dependent on lymphatic transport from the fetal peritoneal cavity for entry into the intravascular space; (4) IVT can be performed before 20 weeks; (5) reversal of hydrops in utero can be achieved regardless of gestational age, thereby reducing complications in the neonatal period; (6) fetuses can be delivered as late as 37-38 weeks’ gestation, reducing the need for

7

neonatal intensive care, and (7) timing of transfusions and delivery can be performed rationally. There are two methods of IVT: straight and exchange. Straight (‘top up’) transfusion refers to the intravenous administration of blood into the fetus without removing any blood. An exchange transfusion consists of the administration and removal of approximately equal amounts of blood. The debate whether to perform exchange rather than straight transfusion continues, especially in the hydropic fetuses. The rationale for exchange transfusion is to avoid fluid overload in the severely anaemic fetus since fetal loss due to a ruptured spleen has been described after straight transfusion.” It was speculated that this had occurred as a result of overexpansion of the vascular space. However, this risk is very small since the fetoplacental unit has a relatively massive vascular capacity. Indeed, the stability of the fetoplacental circulation is best demonstrated by the effects of transfusion in the hydropic fetus, most of whom have a haematocrit < 15%. Even under these circumstances, the fetus can withstand a straight transfusion and no significant changes in heart rate,23v”3 ECG waveforms (unpublished observations), or umbilical venous pressures4 with transfusion of up to lOO-150% of fetoplacental blood volume. Straight transfusions also have the advantage of being completed in less than half an hour whereas exchanges often take 2 or 3 hours, which increases the likelihood of complications especially bacterial contamination and needle dislodgement. Therefore we always perform straight transfusions.

First Transfusion The timing of the first transfusion depends on the gestation at referral and the predicted severity of disease. Ideally, the first FBS followed by IVT should be obtained when the fetus is already anaemic but before hydrops has developed. In general, for patients with previous red cell alloimmunized pregnancy, it is aimed to perform the first FBS and IVT at approximately 10 weeks before the time of the earliest previous fetal or neonatal death, fetal transfusion, or birth of a severely affected baby, but not before 18 weeks’ gestation unless signs of hydrops occur.23 In our experience fetal death or hydrops do not occur before this gestation presumably due to the immaturity of the fetal reticuloendothelial system. The technique and preprocedure preparation for IVT vary from one institution to another.

Premeditation The purpose of premeditation is to minimize maternal discomfort and anxiety. The drug commonly used is diazepam 5-10 mg IV immediately prior to the procedure. However, appropriate counselling before the procedure often allays maternal anxiety.

8

MANAGEMENT

OF IMMUNE

HAEMOLYTIC

DISEASE IN THE FETUS

Antibiotics

Even though the role of prophylactic antibiotic coverage is unproven, several documented cases of serious infection have occurred after a straightforward IVT.55 Based on our past experience with fetoscopy, and our present experience with invasive procedures, we give cefuroxime 750 mg IV to the mother immediately prior to the procedure as a prophylactic antibiotic. Site of IVT

The most common site of IVT is the umbilical cord, usually at its placental insertion, but occasionally at its fetal insertion or free portion.2g+55v56 The technique is similar to that of FBS. However when access is difficult, or sampling failure occurs at the placental cord insertion, the intrahepatic vein is an alternative approach. ‘OJ~ Cardiac puncture is the least used site due to potentially dangerous complications such as bradycardia, arrhythmia, asystole, haemopericardium and cardiac tamponade.57 Fetal Paralysis

Prevention of fetal movement during IVTs is important in order to minimize fetal trauma and facilitate the procedure. To achieve this aim, either maternal administration of sedative medication or fetal neuromuscular blockage can be used. However, the latter offers several advantages and complete cessation of fetal movement is achieved rapidly and uniformly. Various agents have been used to induce temporary fetal paralysis, 58*5gbut it seems that pancuronium bromide offers several advantages over other agents.60 Pediatric experience with this agent is widespread. In addition, the drug is concentrated and its duration of action (30-40 min) is of medium length compare with other agents such as atracurium or d-tubocurarine. The dose of pancuronium bromide depends on fetal weight, gestational age and fetal haematocrit.60 The fetus is more capable of metabolising pancuronium as gestational age increases but is less able to do so if anaemic. Hence the dose varies from 0.1-0.3 mg/kg of estimated fetal weight. In addition we attach a sterile length of extension tubing that is connected to a three-way stopcock during IVT (Fig. 6). This allows for minimal movement of the needle after it is correctly placed in the fetal umbilical vein. Donor Blood

Adult group 0 Rh negative blood crossmatched with maternal blood, collected within 24 hours, washed and packed to a haematocrit of 70-85% is used for IVT. The high haematocrit minimizes the volume of blood required to raise the fetal haematocrit into the normal range. All blood is also screened for HIV, hepatitis and cytomegalovirus. We have found that this unbuffered donor blood

Fig. 6 Ultrasound

guided intravascular transfusion.

transfused into the fetus can produce an acute fall in fetal pH and base excess and a rise in pCO2 without significant change in ~02. 61 These changes cause no harm in non-acidotic fetuses before transfusion because of the buffering capacity of the placental circulation. However, they may be dangerous if there is pre-existing acidosis. Therefore we recommend the routine measurement of fetal pH, pCO2, pO2 and base excess before and after IVT. Volume and Rate of Blood Transfused

The amount of blood transfused is determined by consideration of the estimated fetoplacental blood volume 62 the pretransfusion fetal haematocrit and the hadmatocrit of the donor blood as shown in Figure 7.23 The blood can be transfused at a rate of 5- 15 ml/ min without obvious adverse effects. During the procedure, the fetal heart and the flow of transfused blood are visually monitored by ultrasound. We have measured umbilical venous pressure immediately before and after IVT and found a mean increase of 4.6 mmHg, which is well tolerated.53 In addition we have recently monitored fetal ECG during IVT and found that T/QRS ratios did not change significantly either during or or after uncomplicated IVT (unpublished observations). After transfusion of the calculated volume of donor blood, the fetal haematocrit is assessed, and further blood is given as necessary to bring the final haematocrit to 40-45%. Then a final sample is aspirated for analysis. A Kleihauer-Betke test is also performed on the final aspirate in order to determine the proportion of adult cells in the fetal circulation. Subsequent Transfusion

The second transfusion is usually performed not later than 2 weeks after the first because the rate of fall of haematocrit of each fetus is unpredictable during

BLOOD REVIEWS

250g

9

90

225-

;; 4 200-

$

5:

175-

c

:: ISO-

? 5

2 : e z

125. IOO-

8 =60

Y $

50 -

e

25.

75-

8o 70

8 x

if

p

O~l,,,,,,,,1 18 20 22 24 26

50

40 28 30

32

34 36

OESTATION t WW

0

20 FETAL

HEMATOCRfT

30

40

(X)

Fig. 7 TO calculate the volume of donor blood necessary to achieve a posttransfusion fetal haematocrit of 40%, the estimated fetoplacental blood volume (left, e.g. 100 ml at 27 weeks) is multiplied by F (right, e.g., 0.8 for a pretransfusion fetal haematocrit of 10% and a donor haematocrit of 80%). (From Nicolaides et alz3; with permission of the publishers).

the first interval, although it is on average 1% of haematocrit per day (range 0.2 O-2).63 The subsequent intervals can then be planned more rationally when the fetal haematocrit achieved at the end of the previous IVT, the rate of fall in the first interval, and the degree of suppression of fetal erythropoiesis have been established. The degree of suppression of fetal erythropoeisis can be assessed by measuring the percentage of fetal red cells in the fetal blood (Kleihauer Betke test) at the end of the previous IVT and at the start of the next IVT. Subsequent IVTs are timed to prevent hydrops by maintaining the fetal haematocrit above the critical level of one third of the normal mean for gestational age (haematocrit 20-25%).

cental passage, and, the most important, the skill and experience of the operator. Estimated fetal losses are about 2% per transfusion.43*63

Complications of IVT

Timing of delivery is based on multiple factors including gestational age, estimated fetal weight, fetal lung maturity, fetal response to transfusion, technical ease of transfusion, and the results of antenatal fetal monitoring. Transfusions are normally continued up to 33-34 weeks,7*23*52*63*64 which often allows the delivery of a baby > 36 weeks’ gestation. In general, we have not been guided by tests of pulmonary maturity; the baby is delivered when the haematocrit is sufficiently high to tolerate labour and when the fetus is of a gestational age at which prematurity should be a minor problem.

Although transfusing directly into the fetal circulation clearly has several advantages, it is technically a more demanding procedure than intraperitoneal transfusion (IPT). The possible complications of IVT are summarised in Table 3. All these complications depend on the severity of the disease, the gestation, the size of the needle, the site of the procedure, transplaTable 3 Complications of IVT Maternal supine hypotension Maternal infection Rupture of membranes Chorioamnionitis Premature labour Fetomaternal haemorrhage Fetal bradycardia Cord haematoma or laceration Umbilical artery spasm Fetal death

Fetal Surveillance after ZVT In general after IVT, the patient is monitored by cardiotocography for 1 h. If the tracings are normal, she is allowed to go home. Then daily fetal movement count, weekly ultrasound, biophysical profile or cardiotocography and Doppler ultrasound are recommended. Timing of Delivery

Route of delivery With the availability of various methods for cervical ripening and induction of labour together with the advances in fetal monitoring during labour and the ability to intervene rapidly in cases of fetal distress, many of the Rh patients are now undergoing a trial

10

MANAGEMENT

OF IMMUNE HAEMOLYTIC DISEASE IN THE FETUS

of vaginal delivery. However, delivery should take place in a hospital where neonatologists are experienced in the management of Rh alloimmunization. The mature Rh babies, whose own erythropoiesis has been suppressed and who have only Rh negative donor blood in their circulation, rarely need exchange transfusions after birth. However, it may take 3-4 months before reticulocytes appear and these infants may need several top-up transfusions during this time.

Intraperitoneal Transfusion (II?) Intraperitoneal transfusion (IPT) alone has been performed with less frequency since the advent of the intravascular procedure. A major disadvantage of the intraperitoneal route is its poor efficacy when the fetus has ascites, when the absorption of blood from the peritoneal cavity through the subdiaphragmatic lymphatics and thoracic duct appears to be reduced. Therefore in the severely hydropic fetus, IPT offers little hope.41

be increased safely, reducing the total number of transfusions required and the overall procedure related risk for each fetus. However, several factors should be considered before performing IPT after IVT. These factors are: ease of approach, duration of the procedure of less than 45 min, and absence of fetal bradycardia (i.e. normal FHR). Technique. The technique for IVT is the same as previously described with the aim of raising fetal haematocrit to about 40%. After confirming that the fetal haematocrit at the end of IVT is in the region of 40%, IPT is performed using the same 20-gauge spinal needle as for IVT. It is particularly easy if the intrahepatic vein was used for the IVT. The amount of blood transfused intraperitoneally is calculated with the aim of achieving a final theoretical fetal haematocrit ranging from 50-60%.53*63 A simple proportion is used to estimate this final theoretical haematocrit, based on the volume transfused intravascularly and the measured change in haematocrit, according to the formula63: X= [A(B-C)/D] + B, where X = final theoretical fetal Hct, A = ml of donor blood transfused intraperitoneally, B = fetal Hct at the end of IVT, C = initial fetal Hct, D = ml of donor blood transfused intravascularly. Criteria for timing repeat transfusion are the same as previously described.

Technique. The technique for IPT varies between institutions but is typically as follows: the preparations before the procedure are the same as for IVT, and then a 20-gauge spinal needle is inserted into the fetal abdominal cavity under ultrasound guidance. The ideal site of entry into the fetus is below the umbilical vein but above the bladder, as this minimizes the risk of traumatizing an enlarged liver. Proper needle placement is verified by aspirating ascitic fluid in the hydropic fetus or by injecting a small amount of saline solution and observing layering of fluid in the peritoneal cavity in the nonhydropic fetus. Transfusion is performed directly through the needle, although in some centres a catheter is threaded down for a slow transfusion. The type of blood for transfusion is the same as for IVT and the amount of blood transfused is usually estimated by the following formula: gestation (in weeks) -20 x 10 m1.65 If a large amount of ascites is present, most of it should be removed before transfusing. The fetal heart rate should be monitored with ultrasound visualisation throughout the procedure. The transfusion should be discontinued if persistent fetal bradycardia is observed, as this is probably due to an increased intraperitoneal pressure compromising the venous return to the fetal heart.54

In alloimmunized pregnancies, several recent studies have reported an increased survival following IVTs, which in some centres approximates 90%; survival rate is also improved in hydropic fetuses (Table 4). 23,43,51,52,56,58,63,66-69 With IPT alone, the overall survival rate in more recent years has been about 50%. However, the team in Winnipeg has been much more successful. In 1983, they published impressive outcome data following the institution of real-time ultrasound surveillance during and after the transfusions. Overall survival for the interval from 1980-1982 was 92% (22 of 24 fetuses) with 75% survival of hydropic fetuses.70 They are by far the best survival rates reported for fetuses treated with IPTs alone.

Combined Intravascular and Intraperitoneal Transfusion

Reversal of Intrauterine Hydrops Following Intrauterine Transfusion

At Queen Charlotte’s and Chelsea Hospital, a combination of IVT, followed by IPT during the same procedure has been employed since 1986. It is usually performed after the first transfusion. This combination allows a greater total quantity of blood to be administered without overloading the fetal circulation. It enables the interval between transfusions to

The reversal of hydrops in utero in several recent studies is about 60% and the survival rate in this

Outcome Overall Survival Rate Following Intrauterine Transfusion

group is about 86% (Table 5).23,51,52,56,58,63,66-69,71

Initial improvement of hydrops is variable, occurring as early as 48 h after the first IVT or not until after several IVTs. This probably depends on many factors, including the duration and severity of the hydrops,

BLOOD REVIEWS Table 4

The survival of Rh alloimmunized fetuses in several recent studies No. of fetuses Non Hyd Hyd

Authors

Average no. of TX/fetus

GA at TX (wk)

Technique (site)

3.1

19-32

5

25-31

IVT (fetoscopy) IVTjIPT (IHV) IVT (PCl) IVT (fetoscopy) IVT (IC/PCI) IVT/IPT (PCI) IVT/IPT (PCI/IHV) PETx/IVT (PCI) IVT/IPT (PCI) PETx/IPT (PCI/IHV) IVT + IPT (PCI/IHV)

15

10

2

2

Nicolaides et alz3

11

10

6.9

18-36

MacKenzie et al66

5

5

1.7

18-30

16

21

2.5

18-30

Barss et a16s

6

8

4.4

22-34

Berkowitz et al”

5

12

2.3

21-34

Grannum et al56

20

6

2.8

17-34

Pare+

12

33

2.1

_

Poissonier et aF9

47

60

1.9

17-32

4

21

3.4

18-35

143

194

Rodeck et a15i de Crespigny et a15’

Westgren et a16’

Nicolini et al63 Total

11

Survival Hyd

Table 5

TX = Transfusion IPT = Intraperitoneal transfusion PC1 = Placental cord insertion

Survival (%)

11

7

72

1

2

15

10

10

95

1

1

20

10

19

78

7

85

3

10

76

16

5

81

6

30

80

29

55

79

3

23

84

;:6.4%) GA = Gestation age IVT = Intravascular transfusion IHV = Intrahepatic vein

Non Hyd

169 (87.1%)

78.3

Hyd = Hydrops PETx = Partial exchange transfusion IC = Intracardiac

In utero reversal of hydrops and fetal survival No. of fetuses

Authors Rodeck et a15r de Crespigny et a15s Nicolaides et alz3 MacKenzie et al66 Socol et al’i Barss et a16s Berkowitz et aIs2 Grannum et als6 Westgren et a16’ Poissennier et aF9 Nicolini et alh3 Total no Percent

Hydropic 15 2 11 5 3 6 5 20 16 41 4 134 100%

No. of surviving Rev.

Non-rev.

Rev.

6 1 11 0 3 4 0 14 10 33 3

9 1 0 5 0 2 5 6 6 14

9 1 10 0 3 4 0 13 8 29 3

85 63.4%

49 36.6%

1

77 90.6%

Non-rev. 5 0 0 1 0 1 3 3 2 0 0 15 30.6%

Rev. = Reversal of hydrops Non-rev. = Non-reversal of hydrops

the efficacy of treatment of anaemia, and other as yet unknown aspects of pathophysiology. High Dose Immunoglobulin Therapy A favorable outcome of pregnancy complicated by severe Rh disease has been reported in 5 cases in whom high-dose intravenous immunoglobulin (WIG) was administered.72-74 In 3 cases IVIG alone was given to the mothers, in one, maternal IVIG was given after intensive plasmapheresis followed by 5 IPTs, and in the other case a single dose of IVIG was given to the fetus followed by 4 IVTs. These reports may reflect a bias of positive reporting; however, there are distinct theoretical advantages to the

use of high dose immunoglobulin in the treatment of this condition. The mechanism of action of IVIG is unclear, but possibilities during pregnancy include feedback inhibition of antibody synthesis, competition for macrophage or Fc receptors of target cells, and blockade of Fc-mediated antibody placental transport.75 There is not yet enough experience with this method of treatment. It is hoped that the results of larger, preferably randomised, studies will become available. Other Treatment Protocols In an attempt to prevent D antibody formation, to remove antibody already formed, or to block the

12 MANAGEMENT OF IMMUNE HAEMOLYTIC DISEASE IN THE FETUS action of the antibody on the red cell, a number of techniques have been tried without consistent success.65

12.

Plasmapheresis

To reduce maternal antibody levels does not appear to provide benefits that outweigh the risks and the Costs.76 Promethazine

In large doses beneficial,” as phagocytosis of cell membranes, however, this is

13.

14.

has been cited by some as being this drug was claimed to decrease affected fetal red cells, stabilise red and decrease antibody production; unproven.

D-positive erythrocyte

Membrane in enteric coated capsules has been administered orally to sensitized women throughout pregnancy on the basis that such treatment might induce T-suppressor cell formation that would, in turn, reduce antibody response to challenges by the antigen ” but it does not appear to provide any benefit. Neiiher do attempts at immunosuppression with corticosteroids.79

15.

16.

17.

18.

Acknowledgement

19.

Dr Yuen Tannirandorn is financially supported by the Anandamahidol Foundation and Chulalongkorn University, Thailand.

20.

References 1. Clarke C A, Mollison P L 1989 Deaths from Rh haemolytic disease of the fetus and newborn, 1977-87. Journal of the Royal College of Physicians of London 23: 181-184 2. Nicolaides K H, Rodeck C H 1985 Rhesus disease: the model for fetal therapy. British Journal of Hospital Medicine 34: 141-148 3. Nicolaides K H. Thilaeanathan B. Rodeck C H. Mibashan R S 1988 Erythroblast&is and reticulocytosis in’ anemic fetuses. American Journal of Obstetrics and Gynecology 159: 1063-1065 4. Nicolaides K H, Rodeck C H, Millar D S, Mibashan R S 1985 Fetal haematology in rhesus isoimmunisation. British Medical Journal 2%: 661-663 5. Nicolaides K H, Soothill P W, Clewell W H, Rodeck C H, Mibashan R S, Campbell S 1988 Fetal haemoglobin measurement in the assessment of red cell isoimmunisation. Lancet i: 1073-1075 6. Nicolaides K H, Warenski J C, Rodeck C H 1985 The relationship of fetal plasma protein concentration and haemoglobin level to the development of hydrops in rhesus isoimmunization. American Journal of Obstetrics and Gynecology 152: 341-344 7. Grannum P A T, Cope1 J A 1988 Prevention of Rh isoimmunization and-treatment of the compromised fetus. Seminars in Perinatology 12: 324-335 8. Rodeck C H, CampbelieS 1978 Sampling pure fetal blood by fetoscopy in second trimester of pregnancy. British Medical Journal 42: 728-730 9. Daffos F, Capella-Pavlovsky M, Forestier F 1985 Fetal blood sampling during pregnancy with use of a needle guided by ultrasound: a study of 606 consecutive cases. American Journal of Obstetrics and Gynecology 153: 6.5-660 10. Nicolini U, Santolaya J, Ojo 0 E et al 1988 The fetal intrahepatic umbilical vein as an alternative to cord needling for prenatal diagnosis and therapy. Prenatal Diagnosis 8: 665-671 II. Nicolini U, Nicolaidis P, Fisk N M, Tannirandorn Y,

21.

22.

23.

24.

25.

26.

27.

28.

29.

30.

31.

32.

Rodeck C H 1990 Fetal blood sampling from the intrahepatic vein: analysis of safety and clinical experience with 214 procedures. Obstetrics and Gynecology (in press) Fisk N M, Tannirandorn Y, Santolaya J, Nicolini U, Letsky E A, Rodeck C H 1989 Fetal macrocytosis in association with chromosomal abnormalities. Obstetrics and Gynecology 74: 61 I-616 Nicolini U, Kochenour N K, Greco P et al 1988 Consequences of fetomatemal haemorrhage after intrauterine transfusion. British Medical Journal 297: 13791381 Bowel1 P J, Selinger M, Ferguson J, Giles J, MacKenzie I Z 1988 Antenatal fetal blood sampling for the management of alloimmunized pregnancies: effect upon maternal anti-D potency levels. British Journal of Obstetrics and Gynaecology 95: 759-764 Gemke R J B J, Kanhai H H H, Overbeeke M A M et al 1986 ABO and Rhesus phenotyping of fetal erythrocytes in the first trimester of pregnancy. British Journal of Haematology 64: 689-697 Furhman H C, Klink F, Grzejszczyk G et al 1987 First trimester diagnosis of RHO (D) with an immunofluorescence technique after chorionic villus sampling. Prenatal Diagnosis 7: 17-21 Kanhai H H, Gravenhorst J B, Gemke R J, Overbeeke M A, Bernini L F, Beverstock G C 1987 Fetal blood group determination in first-trimester pregnancy for the management of severe immunization. American Journal of Obstetrics and Gynecology 156 120-123 Rodesch F, Lambermont M, Donner C et al 1987 Chorionic biopsy in management of severe Kell alloimmunization. American Journal of Obstetrics and Gynecology 156: 124-125 Warren R C, Butler J, Morsman J M, McKenzie C, Rodeck C H 1985 Does chorionic villus sampling cause fetomaternal haemorrhage? Lancet i: 691 Rodeck C H, Letsky L 1989 How the management of erythroblastosis fetalis has changed. British Journal of Obstetrics and Gynecology %: 759-763 Nicolaides K H 1989 Studies on fetal physiology and pathophysiology in Rhesus disease. Seminars in Perinatology 13: 328-337 Weiner C P 1989 Nonhematologic effects of intravascular transfusion on the human fetus. Seminars in Perinatology 13: 338-341 Nicolaides K H, Soothill P W, Rodeck C H, Clewell W 1986 Rh disease: intravascular fetal blood transfusion by cordocentesis. Fetal Therapy 1: 185-192 Eklund J, Jouppila P, Seppala I J T 1988 IgG subclasses of anti Rh(D) and haemolytic disease of the newborn. VOX Sangiunis 55: 51-52 Zupanska B, Brojer E, Richards Y, Lenkiewicz B, Seyfried H, Howell P 1989 Serological and immunological characteristics of maternal anti-Rh(D) antibodies in predicting the severity of haemolytic disease of the newborn. VOXSangiunis 56: 247-253 Bowell, P J, Waincoat J S, Peto T E A, Gunson H H 1982 Maternal antiD-concentrations and outcome in rhesus haemolytic disease of the newborn. British Medical Journal 285: 327-329 Harman C R 1989 Fetal monitoring in the alloimmunized pregnancy. In: Manning F A (ed) Clinics in Perinatology. W B Saunders, London, p 691-733 Liley A W 1961 Liquor amnii analysis in the management of the pregnancy complicated by rhesus sensitisation. American Journal of Obstetrics and Gynecology 82: 1359-1370 Nicolaides K H, Rodeck C H, Mibashan R S, Kemp J R 1986 Have Liley charts outlived their usefulness? American Journal of Obstetrics and Gynecology 155: 90-94 Golbus M S, Stephens J D, Cann H M, Mann J, Hensleigh P A 1982 RH isoimmunization following genetic amniocentesis. Prenatal Diagnosis 2: 149-156 Benacerraf B R, Frigoletto F D 1985 Sonographic sign for the detection of early fetal ascites in the management of severe isoimmune disease without intrauterine transfusion. American Journal of Obstetrics and Gynecology 152: 10391041 Chitkara U, Wilkins I, Lynch L, Mehalek K, Berkowitz R L 1988 The role of sonography in assessing severity of fetal

BLOOD REVIEWS

33.

34.

35.

36.

37.

38.

39.

40.

41.

42.

43.

44.

45.

46. 41.

48.

49.

50.

51.

52.

anemia in Rh- and Kell-isoimmunized pregnancies. Obstetrics and Gynecology 71: 393-398 Nicolaides K H, Fontanarosa M, Gabbe S G, Rodeck C H 1988 Failure of ultrasonographic parameters to predict the severity of fetal anemia in rhesus isoimmunization. American Journal of Obstetrics and Gynecology 158: 920-926 Reece E A, Gabrielli S, Abdalla M, O’Connor T Z, Hobbins J C 1988 Reassessment of utility of fetal umbilical vein diameter in the management of isoimmunization. American Journal of Obstetrics and Gynecology 159: 937-938 Kirkinen P, Jouppila P 1983 Umbilical vein blood flow in rhesus isoimmunization. British Journal of Obstetrics and Gynecology 90: MO-643 Griffin D, Coven-Overbeek T, Campbell S 1983 Fetal and uteroplacental blood flow. Clinical Obstetrics and Gynecology 10: 565-602 Rightmire D A, Nicolaides K H. Rodeck C H, Campbell S 1986 Fetal blood velocities in Rh isoimmunization: relationship to gestational age and to fetal hematocrit. Obstetrics and Gynecology 68: 233-236 Cope1 J A, Grannum P A, Belanger K, Green J, Hobbins J C 1988 Pulsed doppler flow-velocity waveforms before and after intrauterine intravascular transfusion for severe erythroblastosis fetalis. American Journal of Obstetrics and Gynecology 158: 768-774 Cope1 J A, Grannum P A, Green J J, Belanger K, Hobbins J C 1989 Pulsed Doppler flow-velocity waveforms in the prediction of fetal hematocrit of the severely isoimmunized pregnancy. American Journal of Obstetrics and Gynecology 161: 341-344 Manning F A, Platt L D, Sipos L 1980 Antepartum evaluation: development of a fetal biophysical profile score. American Journal of Obstetrics and Gynecology 136: 787-790 Harman C R, Manning F A, Bowman J M, Lange I R, Menticoglou S M 1988 Use of intravascular transfusion to treat hydrops fetalis in a moribund fetus. Canadian Medical Association Journal 138: 827-830 Visser G 1982 Antepartum sinusoidal and decelerative heart rate patterns in Rh disease. American Journal of Obstetrics and Gynecology 143: 538-544 Parer J T 1988 Severe Rh isoimmunization-Current methods of in utero diagnosis and treatment. American Journal of Obstetrics and Gynecology 158: 1323- 1329 Nicolaides K H, Sadovsky G. Cetin E 1989 Fetal heart rate patterns on red blood cell isoimmunized pregnancies. American Journal of Obstetrics and Gynecology 161: 351-356 Bowman J M 1986 Haemolytic disease of the newborn (Erythroblastosis fetalis). In: Robertson N R C (ed) Textbook of neonatology, Churchill Livingstone, New York, p 469-483 Tovey L A D 1990 Haemolytic disease of the newborn and its prevention. British Medical Journal 300: 3 13-3 16 Mollison P L. Engelfreit C P, Contreras M 1987 Haemolytic disease of the newborn. In: Mollison P L. Engelfreit C P, Contr’eras M (eds) Blood Transfusion in Clinical Medicine, Blackwell Scientific Publications, London, pp 637-687 Berkowitz R L, Beyth Y, Sadovsky E 1982 Death in utero due to Kell sensitization without excessive elevation of the OD 450 value in amniotic fluid. Obstetrics and Gynecology 60: 146-749 MacKenzie I Z, Bowel1 P J, Castle B M, Selinger M, Ferguson J F 1988 Serial fetal blood sampling for the management of pregnancies complicated by severe rhesus (D) isoimmunization. British Journal of Obstetrics and Gynaecology 95: 753-758 Rodeck C H, Holman C A, Karnicki J, Kemp J R. Whitmore D N. Austin M A 1981 Direct intravascular fetal blood transfusion by fetoscopy in severe rhesus isoimmunisation. Lancet i: 625-627 Rodeck C H, Nicolaides K H, Warsof S L, Fysh W J, Gamsu H R, Kemp J R 1984 The management of severe rhesus isoimmunization by fetoscopic intravascular transfusions. American Journal of Obstetrics and Gynecology 150: 769-774 Berkowitz R , Chitkara U, Wilkins I A, Lynch L, Plosker H, Berstein H H 1988 Intravascular monitoring and management of erythroblastosis fetalis. American Journal of Obstetrics and Gynecology 158: 783-795

13

53. Nicolini U, Rodeck C H 1988 A proposed scheme for planning intrauterine transfusion in patients with severe Rhimmunisation. Journal of Obstetrics and Gynaecology 9: 162-163 54. Nicolini U, Talbert D G, Fisk N M, Rodeck C H 1989 Pathophysiology of pressure changes during intrauterine transfusion. American Journal of Obstetrics and Gynecology -_ 160: 1131-1145 55. Berkowitz R L, Chitkara U, Wilkins I, Lynch L, Mehalek K E 1987 Technical asoects of intravascular intrauterine transfusions: lesson iearned from thirty-three procedures. American Journal of Obstetrics and Gynecology 157: 4-9 56. Grannum P A T, Cope1 J A, Moya F R et al 1988 The reversal of hydrops fetalis by intravascular intrauterine transfusion in severe isoimmune fetal anemia. American Journal of Obstetrics and Gynecology 158: 914-919 57. Westgren M, Selbing A, Stangenberg M 1988 Fetal intracardiac transfusions in patients with severe rhesus isoimmunisation. British Medical Journal 2%: 885-886 58. de Crespigny L C, Robinson H P, Quinn M, Doyle L, Ross A, Eauchi M 1985 Ultrasound-guided fetal blood transfusion for severe rhesus isoimmunization. Obstetrics and Gynecology 66: 529-532 59. Bernstein H W. Chitkara U. Plosker H. Gettes M. Berkowitz R L’l988 Use of’atracurium’besylate td arrest fetal activity during intrauterine intravascular transfusions. Obstetrics and Gynecology 72: 8 13-815 60. Moise K J, Deter R L, Kirshon B, Karolina A, Patton D E, Carpenter R J 1989 Intravenous pancuronium bromide for fetal neuromuscular blockage during intrauterine transfusion for red-cell alloimmunization. Obstetrics and Gynecology 74: 905-908 61. Nicolini U, Santolaya J, Fisk N M et al 1988 Changes in fetal acid base status during intravascular transfusion. Archives of Disease in Childhood 63: 710-714 62. Nicolaides K H, Clewell W H, Rodeck C H 1987 Measurement of human fetoplacental blood volume in erythroblastosis fetalis. American Journal of Obstetrics and Gynecology 157: 50-53 63. Nicolini U, Kochenour N K, Greco P, Letsky E, Rodeck C H 1990 When to perform the next intra-uterine transfusion in patients with Rh allo-immunization: Combined intravascular and intraperitoneal transfusion allows longer intervals. Fetal Therapy (in press) 64. Ronkin S, Chayen B, Wapner R J et al 1989 Intrauterine exchange and bolus transfusion in the severely isoimmunized fetus. American Journal Obstetrics and Gynecology 160: 407-409 65. Bowman J M 1989 Hemolytic disease (Erythroblastosis fetalis). In: Creasy R K. Resnik R (eds) Maternal-Fetal Medicine; Principles and Practice, 2nd edn. W B Saunders, London, pp 613-655 66. MacKenzie 1 Z, Bowel1 P J, Ferguson J, Castle B M, Entwistle C C 1987 In-utero intravascular transfusion of the fetus for the management of severe Rhesus isoimmunizationa reappraisal. British Journal of Obstetrics and Gynaecology 94: 1068-1073 67. Westgren M, Jabbar F, Larsen J F, Rahman F, Selbing A. Stangenberg 1988 Introduction of a programme for intravascular transfusions at severe rhesus isoimmunization. Journal of Perinatal Medicine 16: 417-422 68. Barss V A, Benacerraf B R, Frigoletto F D et al 1988 Management of isoimmunized pregnancy by use of intravascular techniques. American Journal of Obstetrics and Gynecology 159: 932-937 69. Poissonnier M H. Brossard Y, Demedeiros N et al 1989 Two hundred intrauterine exchange transfusion in severe blood in compatibilities. American Journal of Obstetrics and Gynecology 161: 709-713 70. Bowman J M, Manning F A 1983 Intrauterine fetal transfusions: Winnipeg 1982. Obstetrics and Gynecology 61: 203-208 71. Socol M L. MacGregor S N, Pielet B W, Tamura R K, Sabbagha R E 1987 Percutaneous umbilical transfusion in severe rhesus isoimmunization resolution of fetal hydrops. American Journal of Obstetrics and Gynecology 157: 13691375 72. Berlin G, Selbing A, Ryden G 1985 Rhesus haemolytic disease treated with high-dose intravenous immunoglobulin. Lancet i: I1 53

14

MANAGEMENT

OF IMMUNE HAEMOLYTIC DISEASE IN THE FETUS

73. de la Camara C, Arrieta R, Gonzalez A, Iglesias E, Omenaca F 1988 High-dose intravenous immunoglobuhn as the sole prenatal treatment for severe Rh immunization. The New England Journal of Medicine 318: 519-520 74. Scott J R, Branch W, Kochenour N K, Ward K 1988 Intravenous immunoglobulin treatment of pregnant patients with recurrent pregnancy loss caused by antiphospholipid antibodies and Rh immunization. American Journal of Obstetrics and Gynecology 159: 1055-1056 75. Sacher R A, King J C 1988 Intravenous gamma-globulin in pregnancy a review. Obstetrical and Gynecological Survey 44: 25-34 16. Erkkola R, Ekblad U, Piiroinen 0, Kero P, Rajamaki A,

Karanko M, Katka K 1989 Plasma exchange and intrauterine transfusion in the management of severe rhesus isoimmunization. International Journal of Feto-Maternal Medicine 2: 11-14 11. Charles A G, Blumenthal L S 1982 Promethazine hydrochloride therapy in severely Rh-sensitized pregnancies. Obstetrics and Gynecology 60: 627-630 78. Bierme S J, Blanc M, Abbal M, Foumie A 1979 Oral Rh treatment for severely immunized mothers. Lancet i: 604-605 19. Caudle M R, Scott J C 1982 The potential role of immunosuppression, plasmapheresis, and desensitization as treatment modalities for Rh immunization. Clinical Obstetrics and Gynecology 25: 313-319