Introduction

Maternal red-cell alloimmunisation (also referred to as isoimmunisation) occurs when a woman’s immune system is sensitised to foreign red-blood-cell surface antigens, leading to the production of immunoglobulin G (IgG) antibodies.

The most common causes of maternal sensitisation are blood transfusion or feto-maternal haemorrhage (i.e. transplacental passage of fetal red-blood cells) associated with childbirth, trauma, spontaneous or induced miscarriages, ectopic pregnancy or invasive procedures. The resulting antibodies often cross the placenta during pregnancies in sensitised women and, if the fetus is positive for the red-blood-cell surface antigens, this will lead to haemolysis of fetal red-blood cells and anaemia. This may result in potentially serious consequences for the fetus, such as hydrops fetalis and a high cardiac output failure syndrome.

The rhesus blood-group system is the most common cause of maternal alloimmunisation (among more than 300 recognised blood-group antigens). It comprises the c, C, D, e and E antigens. Alloimmunisation against rhesus antigens, leading to haemolytic disease of the fetus and newborn (HDFN) used to be a major cause of perinatal mortality, morbidity and long-term disability until the 1970s. Before the discovery of the rhesus system by Landsteiner and Weiner in 1940, the mechanism of the disease was little understood. After this discovery, however, Levine et al subsequently established that HDFN was usually caused by rhesus incompatibility. Furthermore, it was shown that rhesus alloimmunisation was caused by the passage of fetal rhesus D-positive red cells into the maternal circulation.

The monitoring of pregnancies in which alloimmunisation was known to have occurred historically involved invasive testing, such as serial measurement of amniotic fluid optical density, which was shown to predict the severity of the disease. Initial attempts to treat the condition in utero (in cases in which severe disease was predicted before fetal maturity) included intra-peritoneal transfusion and fetoscopy. These which were replaced by ultrasound-guided cordocentesis and intra-vascular transfusion in the 1980s. Attempts were also made to reduce the severity of the disease using techniques such as plasma exchange, but with limited success.

One of the most important breakthroughs occurred in 1960 when administration of rhesus D IgG (also known as anti-D immunoglobulin or RhoGam) was shown to prevent rhesus-D alloimmunisation, Ultimately, this led to the licensing of anti-D and widespread use for prophylaxis. Anti-D prophylaxis has greatly reduced the frequency of HDFN because of anti-D (which remains the most important cause of HDFN). Perinatal mortality resulting from rhesus disease has also decreased 100-fold in the past 3 decades. Unfortunately, new immunisations continue to occur, with over 500 fetuses in England and Wales developing HDFN annually.

Aetiology

Maternal alloimmunisation and fetal haemolytic disease is most commonly caused by the rhesus blood-group system. This system comprises the c, C, D, e, and E antigens. Importantly, there is no d antigen, and d refers to the absence of the D allele. The D antigen of the rhesus blood-group system remains the cause of the most severe cases of HDFN.

The development and implementation of antenatal rhesus D immune globulin prophylaxis has led to a significant reduction in the frequency of maternal anti-D antibodies, leading to a fall in the incidence of HDFN because of these antibodies. The different types of atypical antibodies found in maternal blood are presented in Table 1 . The most severe cases of haemolytic disease in the fetus and newborn baby are caused by anti-D, anti-c, anti-E and anti-K antibodies.

Sensitisation to non-D group continues to occur because of the lack of availability of antibody prophylaxis for other blood-group antigens. Consequently, the rate of HDFN related to these rarer antibodies is either stable or rising. In several Western countries with effective prophylaxis programmes, the combined frequency of these immunisations already exceeds the frequency of rhesus-D immunisation.

Aetiology

Maternal alloimmunisation and fetal haemolytic disease is most commonly caused by the rhesus blood-group system. This system comprises the c, C, D, e, and E antigens. Importantly, there is no d antigen, and d refers to the absence of the D allele. The D antigen of the rhesus blood-group system remains the cause of the most severe cases of HDFN.

The development and implementation of antenatal rhesus D immune globulin prophylaxis has led to a significant reduction in the frequency of maternal anti-D antibodies, leading to a fall in the incidence of HDFN because of these antibodies. The different types of atypical antibodies found in maternal blood are presented in Table 1 . The most severe cases of haemolytic disease in the fetus and newborn baby are caused by anti-D, anti-c, anti-E and anti-K antibodies.

Sensitisation to non-D group continues to occur because of the lack of availability of antibody prophylaxis for other blood-group antigens. Consequently, the rate of HDFN related to these rarer antibodies is either stable or rising. In several Western countries with effective prophylaxis programmes, the combined frequency of these immunisations already exceeds the frequency of rhesus-D immunisation.

Pathophysiology

Maternal rhesus-D alloimmunisation usually results from the exposure of the maternal immune system to rhesus-D-positive red-blood cells as previously stated. This leads to the formation of anti-D IgG antibodies in the maternal circulation, which can cross the placenta and subsequently sensitise fetal red-blood cells for destruction by macrophages in the reticuloendothelial system. The potential sensitising events are presented in Table 2 .

Transplacental feto–maternal haemorrhage occurs in over 75% of pregnancies, and accounts for most cases of maternal rhesus-D alloimmunisation. It has been reported that tiny (0.1 ml) quantities of fetal red-blood cells gain access to the maternal circulation in pregnancy as shown by studies using flow cytometry. The frequency and volume of these exchanges increases with increasing gestation, and is believed to be highest at the time of childbirth. Feto–maternal haemorrhage can also be associated with other antenatal and postnatal events.

Reports have been published of alloantibodies to rhesus-D antigen without any identifiable maternal exposure to red cells carrying the D antigen. It has been suggested that these cases may be the result of clinically unrecognised early pregnancy losses. A ‘grandmother theory’ has also been proposed to explain this phenomenon. It is suggested that rhesus-D-positive red cells from the patient’s mother gain access into fetal circulation at birth (maternal–fetal haemorrhage), leading to the formation of low levels of antibody to rhesus-D-positive cells in the neonate who is rhesus-D negative.

The risk of alloimmunisation in women who are rhesus-D negative is influenced by the frequency of feto–maternal transfusion and volume of transfused blood. Increasing frequency and higher volume increase the chance of alloimmunisation. ABO blood group status also influences risk of alloimmunisation in rhesus. In a fetus with ABO blood group compatible with its mother, the risk of alloimmunisation is about 16% in the absence of post-natal anti-D administration. This falls to 1.5–2% if the ABO group is incompatible. This protective effect conferred by the ABO incompatibility is believed to be caused by maternal destruction and subsequent clearance of the ABO-incompatible fetal red-blood cells before rhesus sensitisation can occur. Finally, although the rhesus-D antigen is highly immunogenic, the immune response can vary considerably among individuals. Two studies investigating the effect of the volume of fetal blood to which the mother is exposed reported that as many as 30% of individuals who are rhesus D negative were found not to become alloimmunised, even when challenged with large volumes of rhesus D-positive blood.

Individuals with AIDS may not form alloantibodies to the D antigen. The risk of sensitisation is greatest in the first pregnancy and decreases with each subsequent pregnancy. Once sensitisation has occurred, it is irreversible and lifelong.

Successful implementation of anti-D prophylaxis has led to a dramatic reduction in the risk of sensitisation in susceptible pregnancies. About 0.27% of susceptible women still become alloimmunised to rhesus-D antigen because of failure to follow recommended protocols ; spontaneous immunisation continues to occur despite the prophylaxis programme. Coupled with the incidence of immunisation to other red-cell antigens, complete prevention has not been achieved at this time.

Haemolytic disease of the fetus and newborn

Alloimmunisation does not adversely affect the health of the mother. If a woman is exposed to the red-cell antigen during a subsequent pregnancy, the immune response is quicker and much greater. Fetal anaemia results if the red blood cells are removed faster than they can be produced, leading to fetal anaemia. Severe anaemia can lead to fetal heart failure, fluid retention and swelling (hydrops), and intrauterine death.

During pregnancy, anaemia and hydrops can be treated with intrauterine transfusions, but this intervention carries a risk of fetal loss (around 2% per procedure). Haemolysis leads to excess production of bilirubin. In utero , this is cleared by the placenta and therefore is not harmful. After birth, however, the neonatal liver is unable to cope with conjugation of the excess production of bilirubin, leading to jaundice. Low levels of jaundice are not harmful. If left untreated, higher levels of unconjugated bilirubin can result in damage to the basal ganglia of the neonatal brain causing permanent damage (kernicterus). This can lead to a range of neurodevelopmental problems, such as cerebral palsy, deafness, and motor and speech delay. Postnatal jaundice can be treated with phototherapy and exchange transfusion.

The incidence of haemolytic disease of the newborn depends on the proportion of the population that is rhesus-D negative. The prevalence of rhesus-D-negative blood type varies with ethnicity, whites having the highest prevalence ( Table 3 ).

Care of unsensitised women who are rhesus negative

The primary aim of care in unsensitised women who are rhesus negative is the prevention of alloimmunisation through appropriate prophylaxis regimens. All women should have their ABO blood group and rhesus types determined, along with screening for atypical antibodies using the indirect Coombs test at booking during the first trimester. This should be repeated at around the 28 weeks.

Antenatal prophylaxis

Women identified to be rhesus-D negative without any antibodies after screening, should be offered anti-D prophylaxis. The anti-D immunoglobulin neutralises this fetal antigen on the red cells gaining entry into maternal circulation.

Currently, the standard UK practice, according to the National Institute for Health and Clinical Exellence, is to offer anti-D immunoglobulin routinely to all unsensitised women who are rhesus-D negative in the third trimester (28 and 34 weeks). This is for prophylaxis against small volumes of feto–maternal haemorrhage that can occur in the absence of specific sensitising events. This is known as routine antenatal anti-D prophylaxis (RAADP). Routine antenatal prophylaxis was introduced after reports that between 1 and 2% of susceptible women continue to have sensitisation in spite of postpartum anti-D administration, possibly as a result of feto–maternal haemorrhages before childbirth.

The doses for routine prophylaxis vary throughout the National Health Service. The most commonly used regimens are as follows: (1) two doses of anti-D immunoglobulin of 500 IU (one at 28 weeks’ and one at 34 weeks’ gestation); (2) two doses of anti-D immunoglobulin of 1000–1650 IU (one at 28 weeks’ and one at 34 weeks’ gestation); and (3) a single dose of 1500 IU between 28 and 30 weeks’ gestation.

Anti-D prophylaxis for unsensitised women who are rhesus-D negative is also recommended after potentially sensitising events ( Table 1 ). The incidence of feto–maternal haemorrhage during spontaneous miscarriage is reported to be between 6 and 7% in the first trimester of pregnancy, and 20% or more in the second trimester. In pregnancies less than 13 weeks old, the fetal blood volume is usually less than 5 ml; therefore, half the usual prophylactic dose (e.g. 250 IU) is usually sufficient to provide adequate protection; after 20 weeks, the standard dose should be given.

Routine antenatal anti-D prophylaxis is given independently to the routine anti-D administration in response to sensitising events. Moreover, use of RAADP is not affected by the administration of anti-D immunoglobulin for other indications earlier in the pregnancy. Prophylaxis programmes have dramatically reduced the prevalence of rhesus D alloimmunisation in non-immunised women who are rhesus-D-negative. Immunisation during pregnancy without clinical signs of feto–maternal haemorrhage, however, is the most common cause of new immunisations.

Postnatal prophylaxis

The most common occasion for fetal red cells to enter the maternal circulation is at the time of birth. It has been reported that without administration of anti-D immunoglobulin during this time, unsensitised women who are rhesus negative and give birth to a rhesus positive baby have a 7.2% risk of developing rhesus antibodies within 6 months of giving birth. Routine anti-D immunoglobulin prophylaxis should be given within 72 h after birth.

Post-natally, the procedure should be as follows: (1) collect cord blood for fetal blood group; (2) collect maternal blood for Kleihauer test (this is to identify large feto–maternal bleeding); and (3) if the baby is rhesus positive, give anti-D as early as is possible, but within 72 h.

Post partum anti-D immunoglobulin administration is not needed if the infant is rhesus-D negative. If there are any doubts about whether or not to administer, the best approach is to give anti-D unless contraindicated. Anti-D immunoglobulin should definitely be offered if the baby is rhesus-D positive.

Currently, limited evidence is available on the optimal dose of anti-D immunoglobulin to be given for routine prophylaxis postnatally. A total of 500 IU anti-D is sufficient to neutralise 4 ml of rhesus-positive red cells. Some of the dose regimens currently in use in different countries for postnatal prophylaxis include the following; UK 100 mg (500 IU); Canada 100–120 mg (500–600 IU); and USA and parts of Europe 200–300 mg (1000–1500 IU). A feto–maternal haemorrhage of 4 ml and above is seen in about 0.7–0.8% of births. The volume may be larger, especially with traumatic births, caesarean sections, manual removal of placenta and, in some cases, of intrauterine fetal death. In these situations, anti-D in doses larger than 500 IU may be needed. The amount of feto–maternal haemorrhage is routinely assessed at birth by counts of fetal red cells in the maternal circulation using the Kleihauer test. This allows for additional anti-D immunoglobulin to be given only when needed, based on the result of the test.

Care of a sensitised woman

Once sensitisation has occurred, the pregnancy is at risk, and the following investigations are required: (1) maternal antibody titre – most useful only in the first sensitised pregnancy; (2) partner’s genotype; and (3) fetal blood group (if the partner is heterozygous).

Maternal serum antibody levels

The initial investigation into alloimmunised pregnant women who are rhesus-D positive, or those for whom another antibody screen result is positive, is the assessment of the amount of antibody present in the maternal serum. The presence of antibodies indicates that the baby is potentially at risk of harm if it carries the antigen to which the antibodies were formed. The indirect antiglobulin (Coombs) test is used to detect the presence and the degree of alloimmunisation. Antibody titres should be seen as screening tests. For example, positive anti-D titres mean that the mother is sensitised and the rhesus-positive fetus is at risk of haemolytic disease. It does not mean that haemolytic disease has occurred or that it will necessarily develop.