1. Introduction
Hemolytic disease of the newborn (HDN) is still a matter of concern, due to the morbidity and mortality that can produce in the fetuses and newborns affected. HDN leads to severe fetal anemia and consequently to increased cardiac output, tissue hypoxia, fetal hydrops and in the most severe cases, in utero death. Many different red blood cell (RBC) alloantibodies can produce the disease, being anti-D the most frequently implicated followed by anti-c, anti-K and anti-E
[1,
2].
Overall prevalence of red blood cell alloantibodies in pregnant women ranges between 1 % and 2 % as published in different studies of many different developed and developing countries
[3-
5]. Introduction of routine antenatal and postnatal prophylaxis in Rh (D) negative women with anti-D immunoglobulin from the 1970s dramatically reduced the incidence of anti-D alloimmunization and consequently of HDN. Before the practise of passive administration of anti-D IgG, the risk of immunization was calculated as high as 16 %6. Postpartum immunoprophylaxis decreased the incidence of post-pregnancy anti-D immunization to 1–2% and a further reduction in the alloimmunization rate to 0.1 % was achieved by introducing routine antenatal prophylaxis
[7,
8]. However and despite anti-D prophylaxis administration, a variable percentage of Rh(D) negative pregnant women still sensitizes due to different reasons. In addition, there is no prophylaxis for the other alloantibodies.
Our objective was to analyse the prevalence of RBC significant alloantibodies in pregnant women of our area. In addition, follow-up of deliveries and newborns was performed in order to evaluate the hemolytic potential of the different alloantibodies.
2. Materials And Methods
We retrospectively reviewed pregnancies that were positive screened for RBC antibodies for a five-year period. Data was collected at la Fe University Hospital in Valencia, Valencian Community, Spanish Mediterranean area. Anti-D prophylaxis was administered to D negative women at 28 gestation weeks and after delivery if newborn was D positive. Obstetric and immunohematologic data from pregnant women controlled in the Obstetrics Department from January 2011 to January 2016 were reviewed. ABO and Rh status of mothers and newborns, RBC alloantibodies specificity and transfusions including intrauterine transfusions of affected newborns were collected from Blood Bank database. Newborn data was also reviewed from the medical notes and included: gestational age, hemoglobin and bilirrubin levels at birth, cord blood direct antiglbulin test and red blood cell antigen status, and treatment received before and after birth. Not all data were available for each mother or newborn. Women with anti-D as a result of Rh Immunoglobulin administration were excluded. Only clinically significant RBC alloantibodies were considered.
ABO, Rh typing and the Indirect Antiglobulin Test (IAT) were determined in the automated system ORTHO Autovue Innova (Ortho Clinical Diagnostics, England). Sera of patients with positive results were tested for specificity using panels of RBC with known antigens (Ortho Clinical and Diagnostics, Diamed and Makropanel).
ABO group and DAT were systematically performed in newborns. The DAT was performed on RBC from ethylenediaminetetra acetic acid samples according to standard methods. RBC eluates were prepared by acid elution (Elu-Kit II, Gamma Biologicals, Houston, TX, USA). Eluates and sera of patients with positive results were tested for specificity using panels of RBC with known antigens (Ortho Clinical and Diagnostics, Diamed and Makropanel).
Computer software (SPSS, version 13, SPSS Inc., Chicago, IL) was used to perform the statistical analysis. Descriptive statistics are presented for variables. Results are expressed as median and range for continuous variables and as numbers with percentages for categorical variables. Categorical variables were compared by means of Chi square test or Fisher exact test.
3. Results
During the study period, 27609 pregnant women (23215 Rh (D) positive and 4394 Rh (D) negative) were controlled in the obstetric service of the hospital. Of these, 176 (0.63 %) had some clinically significant RBC alloantibody and 40 women (22.7 %) had more than 1 alloantibody. Parity median and range of patients with alloantibodies was 2 (1-8). Tables 1 and 2 show the specificity of alloantibodies according to the D status of women. RBC alloantibodies were detected in 1.3 % and 0.5 % of Rh (D) negative and Rh (D) positive women, respectively. The antigens most often found were anti-D for Rh (D) negative women, and anti-E for Rh (D) positive women. Alloantibodies were detected in third trimester in seventy-four cases.
The percentage of pregnant women who had clinically significant alloantibodies by blood group O, A, B and AB were 1.6 %, 1.8 %, 1.5 % and 3.2 % respectively (p=ns).
Of all sensitized women, 128 patients were followed and delivered at our hospital (37 with anti-D and 91 with other antibodies). In 33 cases (89.1 %) of the 37 women carrying an anti-D, the newborns were D positive and 25 (67.5 %) had a positive DAT and anti-D in eluate. Of 91 patients with alloantibodies different to D, 46 newborns had a positive DAT (50.5 %), and the alloantibody was detected in eluate in 18 cases that were positive for the antigen.
In thirteen patients requiring intrauterine transfusion (IUT), the antibody pattern was as follows: 7 patients with anti-D, 4 anti-D combined with anti-C, 1 anti-K, 1 anti-c combined with anti-E. Characteristics of patients requiring ITU are showed in table 3.
One fetal death occurred after a second ITU. Plasma exchange was also performed in the patient number 10 in an attempt to reduce the ITU requirements. Anti-D and anti-C titres ascended to 131072 and 16384 respectively after the fifth ITU. Six procedures were
performed between 26 and 28 weeks gestation, allowing a reduction of anti-D and anti-C titers to 16384 and 128 respectively. After, a 6th ITU was performed and baby was delivered at 31 gestation week. A healthy newborn was obtained who required transfusion of 5 antigen negative RBC aliquots and phototherapy, being discharged 30 days after. Hemoglobin and bilirubin at birth were 15.3 g/dl and 3.92 mg/dl, respectively.
Among patients carrying anti-D who did not require any treatment during pregnancy (n=26), 6 newborns received phototherapy + antigen negative RBC transfusion, 6 were treated with phototherapy, 1 received antigen negative RBC transfusion and 1 received an exchange transfusion after delivery. Overall, 25 patients with anti-D required any treatment during pregnancy or after delivery (67.5 % of patients with anti-D who delivered at la Fe Hospital). Among 91 patients carrying and alloantibody with specificity different to D and who did not require treatment during pregnancy, 5 newborns received phototherapy, and 1 received phototherapy + antigen negative RBC transfusion.
4. Discussion
To our knowledge, this is the first study analysing the presence of maternal red blood cell alloantibodies, and their clinical consequences on newborns in our area. The prevalence of RBC alloantibodies in pregnant women of our hospital is less than 1%, which is low as compared to other reports in which prevalence is systematically higher than 1 %
[3,
4,
9]. On the contrary, Pal et al
[10] have published a prevalence of maternal red blood cell alloimmunization of 0.73 % in Queensland, Australia, while Zwingerman R et al
[5] in an analysis performed in Canadian prenatal.
population revealed a prevalence of alloimmunization as low as 0.36 %. In the last study, cases of anti-K as sole antibody were excluded. In most studies the most frequent antibody detected was anti-E
[3],
[5],
[10]. We should note that in our study, anti-D (23.8 %) was the most frequent alloantibody detected in pregnant women, followed by anti-E (20.4 %) and anti-M (13.0 %). Anti-E was the antibody most frequented detected in Rh (D) positive women. Despite the routine prophylaxis with anti-D immunoglobulin that is well established, anti-D was developed in almost 1% of Rh (D) negative pregnant women at our hospital. In fact, anti-D contributed to 23.8 % of total alloimmunizations and is the first cause of severe HDN due to RBC alloantibodies. Of 13 ITU performed during the study period, 11 were due to HDN caused by anti-D. The reasons for persistence of anti-D immunization in pregnant women, is beyond the reach of our study. Besides, our hospital is a tertiary-care facility that receives the high risk patients from other institutions and the alloimmunization data cannot be an accurate reflection of the prevalence in the general pregnant population.
In our study, 17 % of alloimmunized women had a combination of antibodies, being the most frequent association of anti-D + anti-C. These results are different of previous studies in which the combination of antibodies was detected in 27 %
[11] and 13 %
[12]. The presence of more than 1 alloantibody has been associated to a higher risk to develop significant of HDN, than those with a single antibody, especially if the anti-D is present
[11],
[12]. In this sense, it has to be noted that in our study 38 % of ITU were performed to women carrying more than 1 alloantibody, suggesting a high risk for these patients.
ITU is an effective and safe treatment that has contributed to perinatal survival rates near or more than 95 % in experienced centres
[13],
[14]. However, treatment of HDN before 20 weeks gestation with ITU is a technical challenge because of the difficulty to gain the fetal vascular access. A combination of plasmapheresis and intravenous immunoglobulin has been reported in the literature for treatment of HDN in first trimester high-risk pregnancies
[15],
[17]. In these reports, plasmapheresis was used in addition to the IG with the aim of reducing the start of fetal anemia, and consequently the ITU requirements. Other authors18 have used plasmapheresis alone for the same purpose. We also initiated plasmapheresis to the case number 10 of the table 3, in order to decrease the alloantibody titers and consequently the ITU needs. Our case has some differences when compared to other cases of anti-D treated with plasma exchange: the procedure was initiated late in 26 gestation week, after detecting an important increase in alloantibody titers and when the patient had received 5 ITUs. However we thought that fetus could benefit of this procedure by reducing the maternal alloantibody titters. In fact, after 6 procedures, a significant reduction of antibodies was achieved. Newborn outcome was good. We have only found 1 similar case in the literature,in which the antibody implicated was anti-K18. A combination of ITU and plasmapheresis was employed from 27 gestation week´s with good newborn outcome.
In conclusion, anti-D remains the most frequent antibody in pregnant women and produces the most severe HDN in our area. Since it is the only alloimmunization that can be prevented, an effort to administrate prophylaxis to all RhD negative pregnant women population is mandatory. Plasmapheresis can be useful, as adjuvant treatment in those cases of severe HDN to reduce the alloantibody titer and consequently ITU needs.