Materials and methods

The study group included all babies born to Rh negative mothers at the Rabin Medical Center, Beilinson Campus, Petah Tiqva, Israel within a three month period (1 January to 31 March 1999). For each mother, number of previous pregnancies and births and number of doses of anti-D immunoglobulin received during the present pregnancy were recorded. For all babies, in addition to the routine blood group typing and Coombs' test, blood smear was performed at age 12–36 hours, and haemoglobin, packed cell volume, mean corpuscular volume, percentage reticulocytes, and bilirubin level were recorded, as were gestational age and birth weight. The study was approved by the ethics committee at Rabin Medical Center, and all parents gave informed consent.

The infants were grouped according to the number of maternal doses of anti-D immunoglobulin: one (routine prophylaxis at 28 weeks gestation) or two (standard prophylaxis plus extra dose after amniocentesis). Each of these groups was further divided by Rh status (positive or negative). The Rh negative subgroup later served as the control group for comparison of the haematological variables, as anti-D immunoglobulin cannot cause haemolysis in Rh negative babies. The results were also compared with a second control group of babies born to O-Rh positive mothers who underwent blood grouping and Coombs' test as routine procedures to detect possible ABO incompatibility. Blood smear was performed in this group as well.

Babies born to mothers with perinatal fever, premature rupture of the membranes of more than 24 hours duration, diabetes, or any other chronic disease that could influence neonatal haematological variables were excluded, as were babies with major congenital malformations, fetal distress, or respiratory distress after birth. In addition, we excluded babies with type A or B blood born to mothers with type O blood in order to avoid double incompatibility. In the O-Rh positive control group, we excluded babies with positive Coombs' test or with early jaundice.

Results

A total of 138 full term infants were included in the study, divided into four groups: group 1, 41 Rh positive infants born to Rh negative mothers who had received one dose of anti-D immunoglobulin during pregnancy; group 2, 25 Rh positive infants born to Rh negative mothers who had received two doses of anti-D immunoglobulin during pregnancy; group 3, 35 Rh negative infants born to Rh negative mothers who had received one or two doses of anti-D immunoglobulin during pregnancy (first control group); group 4, 37 infants born to O-Rh positive mothers with no ABO incompatibility (second control group).

Table 1 presents the characteristics of the four groups. No statistically significant differences were found among the groups with regard to number of gestations and parity of the mothers and birth weight and sex of the newborns (p = 0.8, 0.6, 0.07, 0.25 respectively). However, mean gestational age was slightly lower in group 2 than in groups 1, 3 and 4, which were similar for this variable (p = 0.01).

Table 2 contains the haematological findings for the four groups. There were no statistically significant differences among the groups with regard to mean haemoglobin, mean packed cell volume, or percentage reticulocytes (p = 0.51, 0.47, 0.10, respectively). Mean corpuscular volume was higher in group 2 than in group 3 (p = 0.02).

Six babies in groups 1, 2, and 3 (two in each group) had neonatal jaundice, with bilirubin levels higher than 12 mg/dl. Three of these babies had polycythaemia (venous packed cell volume higher than 65%).

Coombs' test, performed in the 101 babies born to Rh negative mothers, was positive in five of the 25 infants (20%) whose mothers had received two doses of anti-D immunoglobulin (group 2) compared with only one of the 41 infants (2%) whose mothers had received only one dose (group 1). In all infants with positive Coombs' test, antibody identification disclosed only anti-D immunoglobulin. Although mean gestational age (38.5 weeks) and birth weight (2882 g) were lower in the babies with a positive Coombs test, mean packed cell volume (60.25%), mean haemoglobin (19.8 g/dl), percentage reticulocytes (4.4), and presence of neonatal jaundice were similar to the findings for the babies with a negative Coombs' test.

Discussion

To the best of our knowledge, the fetal risk of haemolysis due to high maternal doses of anti-D immunoglobulin has hardly been explored. However, the current dose was extensively studied before clinical use. Hermann and Kjellman6 found no changes in cord packed cell volume or bilirubin concentration in 510 babies of mothers who had received a single dose of anti-D immunoglobulin during pregnancy. No prospective studies have evaluated haematological variables as signs of haemolysis in newborns of women treated with multiple doses.

Our study showed no significant differences, compared with controls, in haemoglobin, packed cell volume, or percentage reticulocytes in Rh positive babies born to Rh negative mothers who had received one or two doses of anti-D immunoglobulin. This was also true when the six babies of the 66 Rh negative mothers (9%) who had a positive direct Coombs' test were analysed separately. Furthermore, even though there were significantly more positive direct Coombs' tests among the babies of mothers who had received two doses of anti-D immunoglobulin (20%) than among babies whose mothers had received only one dose (2%), their haematological variables were similar. These findings indicate that the placental passage of anti-D immunoglobulin does not produce haematological changes relevant to haemolysis.

We therefore conclude that the prevention of Rh isoimmunisation by the injection of anti-D immunoglobulin, whether one or two doses, does not jeopardise newborns, even in the event of transplacental passage of anti-D immunoglobulin (9% of cases).

Our study supports the routine practice of blood group typing and direct Coombs' testing in every baby of an Rh negative mother immediately after birth to establish the necessity to administer anti-D immunoglobulin. It does not support the practice of taking a blood count in these babies as a routine procedure. In the presence of a positive direct Coombs' test, the antibodies need to be identified as Rh-D type to exclude the possibility of other minor blood group incompatibilities; of course, in this special situation, early jaundice and haematological signs of haemolysis will be present.