Objective
The objective of the study was to determine whether women with combinations of red blood cell antibodies are more likely to develop significant hemolytic disease of the fetus and newborn than those with single antibodies.
Study Design
A retrospective exposure cohort study was conducted of pregnant women with red blood cell antibodies. The development of significant hemolytic disease of the fetus and newborn was then compared between patients with single antibodies and those with multiple antibodies. Data analysis was limited to pregnancies delivering since the year 2000.
Results
Thirteen percent of the patients referred to our program had multiple red blood cell antibodies. Odds of developing significant hemolytic disease of the fetus and newborn for patients with anti-Rh(D) combined with at least 1 additional red blood cell antibody were 3.65 times the odds for women with anti-Rh(D) antibodies in isolation (95% confidence interval, 1.84–7.33). In the setting of multiple antibodies including anti-Rh(D), Rh-positive fetuses/neonates have an increased odds of developing significant hemolytic disease even if the fetus is negative for the other corresponding red blood cell antigen.
Conclusion
Women with multiple red blood cell antibodies are more likely to develop significant hemolytic disease of the fetus and newborn than those with a single antibody especially in the presence of anti-(Rh)D. This pathophysiology may suggest a more aggressive immune response in women who develop more than 1 red blood cell antibody.
Maternal red blood cell alloimmunization is an important cause of morbidity and mortality in the antepartum and neonatal periods. In the United States, 35 per 10,000 live births are at risk for hemolytic disease of the fetus and newborn (HDFN) because of red blood cell alloimmunization, 20% of which may become severely affected. Of those who are severely affected by anti-D, approximately half are sufficiently mature to be delivered and receive neonatal care, whereas the other half require antenatal intervention for survival.
With the introduction of Rh immune globulin, the incidence of Rh(D) alloimmunization has decreased, leading to a relatively increased proportion of red blood cell alloimmunization because of other antibodies. A significant number of these red blood cell antibodies have well-recognized associations with HDFN, including anti-K, anti-c, anti-E, anti-Fya, and anti-Jka.
Further complicating matters, a significant number of patients produce more than 1 red blood cell antibody during pregnancy. Filbey et al demonstrated that 8.2% of pregnancies complicated by HDFN in Sweden had multiple red blood cell antibodies. It has been proposed that such pregnancies affected by multiple red blood cell antibodies are at greater risk for HDFN. Such an effect was seen by Spong et al in 2001, who found an increased need for intrauterine fetal transfusions (IUTs) in pregnancies affected by anti-D in combination with other red blood cell antibodies.
The objective of our study was to further evaluate the effect of multiple red blood cell antibodies on the development of HDFN. To do so, we have compared the fetal and neonatal outcomes of pregnancies with multiple maternal red blood cell antibodies with those with single antibodies. We chose to limit our data analysis to pregnancies delivered since the year 2000 to reflect modern techniques for monitoring women with elevated red blood cell titers via middle cerebral artery Doppler assessment.
Materials and Methods
The Ohio State University Maternal Alloimmunization Program has maintained a computerized database of pregnancies complicated by alloimmunization since 1959. This database includes patients from our institution as well as referrals from central and southeastern Ohio and neighboring regions and was used to obtain patient data for this study.
All laboratory testing was performed at The Ohio State University Medical Center using guidelines established by the American Association of Blood Banks. These guidelines were updated frequently to remain consistent with medical care over the years. Techniques were routinely used to rule out anti-G as a mimicker of anti-D in combination with anti-C. Permission to retain and review patients’ data was obtained from our local institutional review board before proceeding with this study.
This was a retrospective exposure cohort study including women managed in our alloimmunization program from January 2000 through May 2013. Data collected included, but were not limited to, maternal pregnancy history, paternal antigen testing, maternal indirect antiglobulin tests (antibody identification), fetal direct antiglobulin test, fetal antigen type, and fetal hematocrit. Fetal antigen information was obtained prior to the performance of the first IUT of the pregnancy. Neonatal data included gestational age at delivery, delivery hemoglobin, cord blood direct antiglobulin test results and red blood cell antigen status, and necessary treatment for HDFN. Not all data were available for each case.
We identified all women with relevant red blood cell antibodies (those known to place patients at risk for HDFN), analyzing only the first documented pregnancy in our system for each individual patient. Patients were categorized into the following 4 groups according to their antibodies: (1) anti-Rh(D) only, (2) anti-Rh(D) in combination with another red blood cell antibody, (3) other relevant red blood cell antibodies in isolation, and (4) other relevant red blood cell antibody combinations without anti-Rh(D).
For the purpose of this study, clinically significant HDFN was defined as the following: (1) fetal demise, (2) the development of hydrops fetalis, (3) a requirement for intrauterine transfusion, (4) a birth or umbilical cord blood hemoglobin of 10 g/dL or less, or (5) the need for neonatal blood transfusions.
Patient demographics and clinical characteristics are described and compared between the 4 groups. Comparisons of the categorical data were made using the likelihood ratio χ 2 test, and trends were assessed by the Cochran-Armitage trend test. Maternal mean ages were compared using a 1-way analysis of variance. Logistic regression models with multiple predictors were constructed in a step-wise process starting with only the individually significant predictors. Only the significant predictors were retained in the final multiple predictor model. Confidence intervals for odds ratios were estimated using this logistic regression model, and 95% confidence intervals are reported. All P values are 2 sided and considered significant if the value is P = .05. All analyses were performed in SAS JMP, version 10 (SAS Institute, Cary, NC).
Results
A total of 1014 patients with pregnancies complicated by red blood cell alloimmunization were managed at The Ohio State University Wexner Medical Center between January of 2000 and May 2013, a period of more than 13 years. Of these, 132 (13.0%) had more than 1 red blood cell antibody. Demographic data comparing these patients are depicted in Table 1 , whereas Table 2 depicts the HDFN-defining categories for each group.
| Variable | Anti-Rh(D) alone (n = 138) (13.6%) | Anti-Rh(D) in combination (n = 54) (5.3%) | Single other antibody (n = 744) (73.4%) | Other antibody combinations (n = 78) (7.7%) | P value |
|---|---|---|---|---|---|
| Maternal age, y a | Mean (SD) 28.7 (5.0) Median (range) 28 (17–39) Unknown: 15 | Mean (SD) 29.0 (5.4) Median (range) 29 (20–44) Unknown: 6 | Mean 28.7 (6.1) Median (range) 28 (15–49) Unknown: 100 | Mean 28.7 (5.1) Median (range) 28 (20–40) Unknown: 6 | .9836 a |
| Maternal race | |||||
| White | 47 (34.1) | 30 (55.6) | 121 (16.3) | 20 (25.6) | < .0001 |
| Black | 16 (11.6) | 2 (3.7) | 53 (7.1) | 8 (10.3) | |
| Other | 8 (5.8) | 1 (1.85) | 25 (3.4) | 4 (5.1) | |
| Unknown | 67 (48.6) | 21 (38.9) | 545 (73.3) | 46 (59.0) | |
| Parity (>20 wks) | |||||
| 0 | 26 (18.8) | 12 (22.2) | 241 (32.4) | 16 (20.5) | .0054 |
| 1 | 50 (36.2) | 25 (46.3) | 217 (29.2) | 26 (33.3) | |
| 2 | 36 (26.1) | 12 (22.2) | 150 (20.2) | 22 (28.2) | |
| >3 | 26 (18.8) | 5 (9.3) | 136 (13.3) | 14 (18.0) | |
| Fetal sex | |||||
| Female | 37 (26.8) | 18 (33.3) | 110 (14.8) | 16 (20.5) | < .0001 |
| Male | 36 (26.1) | 12 (22.2) | 139 (18.7) | 13 (16.7) | |
| Unknown | 65 (47.1) | 24 (44.4) | 495 (66.5) | 49 (62.8) |
a P value is from 1-way analysis of variance (ages normally distributed); rest of P values are based on the Pearson χ 2 test.
| Category | Other antibody, n, % | Other antibody combination, n, % | Anti-Rh(D) alone, n, % | Anti-Rh(D) in combination, n, % | P value a |
|---|---|---|---|---|---|
| Pregnancies, n | 744 | 78 | 138 | 54 | |
| No HDFN | 729 (98.0) | 74 (94.9) | 108 (78.3) | 27 (50.0) | < .0001 |
| HDFN total b | 15 (2.0) | 4 (5.1) | 30 (21.7) | 27 (50.0) | < .0001 |
| Fetal demise | 1 (0.1) | 0 (0) | 0 (0) | 1 (1.9) | .2924 |
| Hydrops fetalis | 1 (0.1) | 0 (0) | 3 (2.2) | 6 (11.1) | < .0001 |
| IUT required | 7 (0.9) | 1 (1.3) | 11 (8.0) | 16 (29.6) | < .0001 |
| Neonatal anemia | 3 (0.4) | 0 (0) | 4 (2.9) | 6 (11.1) | < .0001 |
| Neonatal transfusion | 10 (1.3) | 4 (5.1) | 24 (17.4) | 19 (35.2) | < .0001 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
