Not as “D”eadly as once thought – the risk of D-alloimmunization and hemolytic disease of the fetus and newborn following RhD-positive transfusion in trauma

ABSTRACT

The use of blood products to resuscitate injured and massively bleeding patients in the prehospital and early in-hospital phase of the resuscitation is increasing. Using group O red blood cells (RBC) and low titer group O whole blood (LTOWB) avoids an immediate hemolytic reaction from recipient’s naturally occurring anti-A and – B, but choosing the RhD type for these products is more nuanced and requires the balancing of product availability and survival benefit against the risk of D-alloimmunization, especially in females of childbearing potential (FCP) due to the possible future occurrence of hemolytic disease of the fetus and newborn (HDFN). Recent models have estimated the risk of fetal/neonatal death from HDFN resulting from D-alloimmunization of an FCP during her trauma resuscitation at between 0-6.5% depending on her age at the time of the transfusion and other societal factors including trauma mortality, her age when she becomes pregnant, frequency of different RHD genotypes in the population, and the probability that the woman will have children with different fathers; this is counterbalanced by an approximately 24% risk of death from hemorrhagic shock. This review will discuss the different models of HDFN outcomes following RhD-positive transfusion as well as the results of recent surveys where the public was asked about their preferences for urgent transfusion in light of the risks of fetal/neonatal adverse events.

KEYWORDS:

• Hemolytic disease of the fetus and newborn (HDFN)
• trauma
• emergency
• transfusion
• RBC
• LTOWB
• alloimmunization
• RhD

Introduction

In the right setting, prehospital and early in-hospital transfusion saves lives in trauma. Several civilian and military observational studies have shown approximately 15% absolute risk reductions in early mortality when transfusions were administered early in the prehospital phase of the resuscitation [1,2]. Improved survival amongst injured prehospital blood product recipients has also been found in the primary [3] and secondary [4] outcomes of several randomized clinical trials (RCT), and several observational studies [1,2], that compared prehospital transfusion to the standard of prehospital care. Furthermore, low titer group O whole blood (LTOWB) is increasingly being used early in the in-hospital resuscitation of trauma patients [5], and its use has also been associated with equivalent or improved survival compared to injured patients who receive conventional component therapy in observational studies [6–9]. In both the pre – and early in-hospital phases of the resuscitation, the recipient’s ABO and RhD groups will not be known. Using group O red blood cells (RBC) or LTOWB avoids causing an immediate hemolytic reaction, because group O RBCs lack A and B antigens and are therefore compatible with recipients of any ABO group. However, a more nuanced question about whether to provide RhD-positive or RhD-negative RBCs or LTOWB (herein, the red blood cell contents of both RBC and LTOWB units are collectively referred to as RBCs) remains unanswered. RhD-negative RBCs will not lead to D-alloimmunization with the potential for hemolytic disease of the fetus and newborn (HDFN) to occur, but they are very scarce products compared to RhD-positive RBCs. This review will discuss the latest evidence on the rates of D-alloimmunization and the potential for adverse fetal/neonatal outcomes from HDFN if an RhD-negative female of childbearing potential (FCP) is transfused with RhD-positive RBCs during trauma resuscitation.

What’s “D” problem?

In the 1970s and 1980s, the textbook answer to the question of what is the rate of D-alloimmunization (i.e. producing the anti-D antibody) following the transfusion of RhD-positive RBCs to an RhD-negative patients was approximately 80% [10–12]. This answer is true but it is misleading because it was generated using healthy volunteers with presumably intact immune systems, and so it represents a worst case scenario when it comes to determining the D-alloimmunization rate amongst sick, hospitalized patients. The classic teaching was to avoid transfusing RhD-positive blood to RhD-negative women. Myriad studies of the D-alloimmunization rate amongst hospitalized patients have since been performed [13,14]; many featured relatively small numbers of patients with heterogeneous diseases or reasons for requiring a transfusion, but germane to this discussion, they did not include healthy volunteers. Some of these studies included trauma patients. In general, the rate of D-alloimmunization in the hospitalized patients in these studies ranged from 20-30% [13]. Interestingly, there were two recent observational studies that focused only on trauma patients – both single institution studies from the USA – with similar study designs and patient inclusion criteria that found highly divergent rates of D-alloimmunization, 7.8% and 42.7%, respectively [13,15]. The former study included all patients ≥18 years of age while the latter included patients who were between 13–50 years of age in order to focus on the patients with immune systems representative of females in the childbearing years. Males were also included in the latter study due to the paucity of injured and transfused females of childbearing potential (FCP) at that center. Thus, while it is not clear why these two similarly designed studies found such markedly different D-alloimmunization rates, even if the actual rate of D-alloimmunization in trauma patients is 42.7%, it is still only approximately half of the rate in healthy immunocompetent individuals.

Furthermore, the consequences of D-alloimmunization are more manageable than had been thought in previous decades. There are four possible consequences of D-alloimmunization:

1. Future routine transfusions could be delayed

If the patient becomes alloimmunized, owing to the time required to perform pretransfusion testing to identify the antibody (perhaps amongst other allo – or autoantibodies) and then locate and crossmatch antigen negative units, the provision of future routine transfusions could be delayed. While non-trivial when these delays occur, they can generally be avoided by having the pretransfusion testing performed in advance of planned surgical procedures [16].

2.	If the patient’s anti-D is active at the time of the uncrossmatched RhD-positive RBC transfusion in an emergency situation, the patient could have an immediate hemolytic reaction or a delayed reaction if the concentration of the anti-D has decreased to below the detection level of the laboratory’s testing methodology
3.	The patient could have a hemolytic reaction to a future RhD-positive emergency RBC transfusion if the index RhD-positive emergency RBC transfusion caused them to become D-alloimmunized

These two complications are in theory quite serious but in practice appear to be of minor clinical consequence. A review of 11 studies that included 2,906 recipients of at least one unit of uncrossmatched RBCs found an overall rate of hemolysis of 4/2,096 (0.1%) [17]. One of these studies in particular was very specific about the immunohematological status of their patients [18]: in this study, 17 patients had active clinically significant antibodies (i.e. antibodies capable of causing RBC destruction) at the time of their uncrossmatched RBC transfusion and of them, 7/17 patients received a total of 15 RBC units that were incompatible with their antibody. In 6/7 patients, hemolytic reactions were not reported to the blood bank nor were hemolytic reactions detected upon chart review following the transfusions. In the remaining patient, the biochemical markers of hemolysis were consistent with a hemolytic reaction but there was also evidence that he was experiencing a delayed hemolytic reaction due to the transfusion of incompatible RBCs 4 days before the index uncrossmatched RBC. In any case, the overall rate of hemolytic reactions following uncrossmatched RBC transfusion is very low and should not preclude their administration in a patient with hemorrhagic shock where mortality can be approximately 24% [19].

4.	The patient could experience HDFN during a subsequent pregnancy

This is perhaps the most clinically severe and contentious issue surrounding the use of uncrossmatched RhD-positive RBCs for FCPs of unknown RhD-type.

What is the risk of HDFN following the transfusion of RhD-positive RBCs to an RhD-negative FCP?

HDFN is a disease caused by maternal anti-D that crosses the placenta and then binds to fetal RhD-positive RBCs leading to their destruction. The range of outcomes from HDFN can vary widely from none or mild symptoms, to severe anemia that can cause either long-term neurologic disability or fetal/neonatal death. In this review, “overall HDFN” will refer to the entire spectrum of potential outcomes from this illness, while individual HDFN outcomes, such as fetal/neonatal death, will be referred to specifically when appropriate.

Recently, the cumulative risk of a FCP experiencing fetal harm or loss due to anti-D that was stimulated by the transfusion of RhD-positive RBCs during her trauma resuscitation was calculated [20]. The authors assumed that 76% of patients survived their trauma, 21% became D-alloimmunized, 86% of American women become pregnant, 60% of fetuses are RhD-positive due to the homozygous/hemizygous nature of the RHD genotype, and that 4% of fetuses affected by HDFN die. These probabilities yielded a cumulative fatal HDFN rate of 0.3% (Figure 1). If other adverse fetal/neonatal events, such as requiring an intra-uterine transfusion or neonatal exchange transfusion, are considered along with fetal demise, the cumulative risk of these events rises to 2%.

Figure 1. Graphic representation of the risk of hemolytic disease of the fetus and newborn (HDFN) following the transfusion of RhD-positive RBCs to an injured RhD-negative female of childbearing age considering five critical events that must take place for HDFN to occur following trauma [20]. The percentages in brackets are the risks of each discreet event occurring; the dark shaded icons represent the cumulative risk of each event occurring as a percentage. For example, 76% of injured adults survive the trauma and 21% become D-alloimmunized, therefore the cumulative risk of fetal death from HDFN at this stage is approximately 16%, assuming the three additional events also occur. The overall cumulative risk of fetal demise from HDFN was calculated to be 0.3%.

In fact, these estimates are slightly high if an even broader perspective on trauma resuscitation is considered. These two models assumed that the patient was an RhD-negative FCP, however, only 15% of Caucasian individuals are RhD-negative [21] and only 55% of American women between the ages of 15-85 + years were actually in their childbearing years (15-49 years) in 2019 [22]. Thus, the 0.3% and 2% risks of adverse HDFN events would be even lower if the woman’s RhD status and probability of being in her childbearing years are considered. Furthermore, the frequency of pregnancy following injury also seems to depend on the nature of the trauma. A Finnish study found that women with some serious injuries like pelvic or hip/thigh fractures were significantly less likely than age-matched women with wrist fractures to become pregnant in the ensuing 5-years following the injury [23]. Though the cause for this reduced birth rate is unclear, it appears that some women are less likely than others to become pregnant following a serious trauma, which would also lower the overall HDFN rate for the overall population.

The above model did not account for other societal factors that could influence the HDFN rate. Some of these factors include the age at which the woman becomes pregnant and the number of times she is likely to become pregnant during her lifetime. The relatively younger the age of the woman at the time of her last pregnancy, the less likely she will be affected by HDFN because her younger age means that she has a lower probability of having been in a trauma before she became pregnant compared to a woman who became pregnant later in life (assuming the rate of trauma is the same for the duration of the childbearing years). Other factors to consider also include the RHD gene frequency and zygosity in the population because this will determine the probability that an RhD-negative FCP’s partner could transmit an RHD gene to the fetus, as well as the number of different fathers of her future children than she might have. The rate of death from trauma is also an important factor to consider. When these factors were simulated in a computer model that included one million FCPs using data from eight different countries, the rate of overall HDFN varied from nearly 0% to approximately 0.6% when both RhD-positive and RhD-negative FCPs were included to simulate the real-life situation where an FCP of unknown RhD-type is injured and requires a transfusion. In this model, when only RhD-negative FCPs were considered, the overall HDFN rate increased to between approximately 2% to 6% as expected [24]. When different severities of HDFN (mild, moderate, severe, intrauterine fetal death) were factored into a model based on the demographics of American FCPs, the average occurrence of HDFN was calculated to be 0.97 cases per 100 transfused RhD-negative FCPs between the ages of 18–49 years. Overall, the highest risk of HDFN was approximately 3.4% when the RhD-negative FCP was 18 years old using a D-alloimmunization rate of 21%, and the overall HDFN rate decreased to nearly 0% as her age approached 43 years. Thus, the rates of HDFN are still very low when a comprehensive evaluation of the variables that can affect its occurrence was performed.

The above described models included only adult women, however, young girls might also become injured and require an emergency transfusion thereby putting them at risk for future HDFN. Using the same detailed model described above, the risk of HDFN amongst RhD-negative girls age 0–17 years who receive an RhD-positive RBC transfusion was estimated [25]. The main limitation in this model was that the D-alloimmunization risk amongst children was unknown because so few RhD-negative children are transfused with RhD-positive RBCs. In fact, an attempt to calculate the pediatric D-alloimmunization rate failed when a survey of at least a dozen large pediatric specialty hospitals in the USA identified only a handful of such children (M. Yazer, personal communication). However, a Japanese study of the overall RBC alloimmunization rate in children in various age groups between <1 month to 15-<20 years informed what the D-specific alloimmunization rate might be in children [26]. By correcting the pediatric age-group RBC alloimmunization rates for the overall adult alloimmunization rate (2.2%) [27], then multiplying by both 7.8% (the lowest published D-alloimmunization rate in injured adults) and 42.7% (the highest published D-alloimmunization rate in injured adults), an estimate of the HDFN risk for RhD-negative pediatric patients transfused with RhD-positive RBCs could be produced [25]. The risk was nearly 0% for neonates and rose to approximately 1.0% or 6.5% for females between 18–20 years of age using the lowest and the highest adult D-alloimmunization rates, respectively. Once again, the HDFN risk decreased to approximately 0% when the patient was at least 40 years old. Of particular interest in the model was the nearly identical risk of future HDFN (approximately 2% using the highest D-alloimmunization rate) if either a 5-year old or a 30-year old was transfused with RhD-positive RBCs; the 5 year old has an immature immune system and a lower predicted D-alloimmunization rate but a much longer time to become pregnant compared to the 30 year old. Conversely, the 30-year old has a mature immune system but a much shorter time to become pregnant. These data can help inform the design of future trials of potentially RhD-positive RBCs and injured children. Figure 2 demonstrates the extremes of the risk estimates of overall HDFN for females of all ages based on the highest (42.7%) and lowest (7.8%) reported D-alloimmunization rates in injured patients.

Figure 2. Estimated risks of overall HDFN for RhD-negative females based on the highest (42.7%) and lowest (7.8%) reported D-alloimmuniztion risks. See text for further details of the assumptions made in this model. Reprinted from reference [25] with the kind permission of John Wiley and Sons, Inc.

Recently, a Monte Carlo simulation of adverse events following RhD-positive transfusion in the prehospital phase of the resuscitation based on data from the London, UK air ambulance service was created [28]. The model was based on similar assumptions and probability distributions as those described in the models above for survival following trauma, RhD-alloimmunization, HDFN mortality rates etc. and was calculated over 1,000 iterations. This model predicted that if the London air ambulance transfused only RhD-positive RBCs to the 5,561 patients who annually receive a prehospital transfusion while in their care, it would take approximately 5 years for there to be one HDFN death or disability event. This frequency is higher than the predicted 1 case of HDFN in 20 years reported by one large academic medical center in the USA that administered RhD-positive RBCs to all injured FCPs who required a transfusion because the London air ambulance has a larger catchment area than this U.S facility [24]. A different US Level 1 trauma center estimated that if they were to transfuse RhD-positive RBCs to injured FCPs who require a transfusion then it would take approximately 250 years for between 3–30 (depending on the D-alloimmunization rate) FCPs to become D-alloimmunized, and the rate of HDFN would be expected to be even lower [29]. Once again, this simulation predicted a very low rate of HDFN adverse events when compared to the benefits of prehospital transfusion. A follow up model showed that if the London air ambulance service did not provide prehospital RBCs while en route to the hospital, the loss of quality adjusted life years (QALY) due to death by hemorrhage would be valued at just over £1 billion compared to providing RBCs of any RhD type over the lifetime of these patients. This model also predicted that the receipt of RhD-positive RBCs would be expected to cause 0.13 fetal/neonatal major morbidity events and 0.18 fetal/neonatal mortality events in the future due to HDFN caused by D-alloimmunization during the resuscitation. The very low incidence of adverse HDFN events reduces the total number of QALYs amongst the recipients of prehospital RhD-positive RBCs from 141,899.0–141,879.8, a difference of only 19.2 QALYs or approximately £600,000. While any loss of life is tragic, this model and those described above show that there is a clear benefit from transfusing RBCs to injured people who need them.

Lastly, recent data has demonstrated that the risk of D-alloimmunization does not increase with an increased volume of transfused RhD-positive RBCs. Two studies in adults, one of RhD-negative general hospitalized patients and another that focused on RhD-negative trauma patients, both of which included patients who received 10 or more RhD-positive units, demonstrated that the risk of D-alloimmunization was not statistically significantly higher after receipt of >1 RhD-positive RBC/LTOWB unit compared to receipt of one unit. These data suggest that once an RhD-negative patient has received one unit of RhD-positive RBCs, the rest of their transfusion needs can be met with RhD-positive products without increasing their risk for D-alloimmunization, with the caveat that information bias remains a limitation of these two studies given the retrospective design. For those patients who have received only a few RhD-positive RBCs, D-alloimmunization risk mitigation strategies have been discussed elsewhere [30,31].

How are these risks interpreted by medical and lay people?

The overall rates of HDFN have been modeled to range between 0% for the youngest patients to a maximum of approximately 6.5% for those who are between 18–20 years using the highest published D-alloimmunization rate. However, the implementation of transfusion policies based on these low HDFN probabilities balanced against the benefits of prehospital transfusion, and use of LTOWB that is likely to be RhD-positive owing to the scarcity of suitable donors, is what really matters. While the HDFN probabilities have only recently been calculated and published, some surveys of medical professionals and lay people reveal some interesting trends in the tolerance of risk as it relates to early transfusion and the potential for HDFN. Of course, if RhD-negative RBCs are available in the prehospital or early in-hospital phase of the resuscitation, they should be used especially for FCPs of unknown RhD type. However, it is often the case, especially in the prehospital phase, that only RhD-positive RBCs are available because, for example, at one large U.S. blood collector 7% of all donors can donate RhD-negative RBCs while only 3% of donors are eligible to donate RhD-negative LTOWB. Thus, due to the scarcity of RhD-negative blood products, both societal values and medical judgement based on data need to be considered when selecting the RhD-type of the blood products to be transfused.

A recent survey of adult American Level 1 trauma centers found that 22/43 (51%) of the centers that transfuse LTOWB to their injured patients would administer RhD-positive LTOWB to FCPs who are either RhD-negative or RhD type unknown [5]. When pediatric Level 1 hospitals were surveyed, 9/55 (16.4%) indicated that they would provide RhD-positive RBCs to injured boys of any age [32], while 11/55 (20%) indicated that the choice of RhD type would be made based upon the boy’s age. Perhaps not surprisingly, 51/55 (92.7)% of these centers indicated that they would only provide RhD-negative RBCs for injured girls, regardless of their age.

Interestingly, this survey revealed that 30.4% of the road ambulances or helicopters that bring patients to these pediatric hospitals carry only RhD-positive RBCs and 66.7% carry only RhD-positive LTOWB; while not specifically asked in the survey, presumably these pediatric hospitals permit the transfusion of RhD-positive products to their incoming patients if that is all that is available during transport.

Apropos of these RhD-transfusion policies, when the transfusion and trauma medical directors at 30 large children’s hospitals in the USA were asked if they would be comfortable enrolling pediatric patients in an RCT where the patient could potentially be randomized to receive RhD-positive LTOWB [33], 80% of the transfusion directors and 71.0% of the trauma responded in the affirmative when asked about enrolling boys. However, only 20.0% and 37.5%, respectively, responded that they were comfortable enrolling girls in such a trial. Thus, the pediatric medical community appears to be quite risk averse when it comes to the potential for D-alloimmunization and HDFN. It would have been interesting to have asked these centers if they would be willing to take a nuanced, risk-based approach to resuscitating girls by using RhD-positive LTOWB for those under approximately 10 years of age when their calculated risk of overall HDFN is relatively low due to their immature immune systems and reserve any RhD-negative LTOWB in their inventory for those who are between 18–40 years age and are thereby are at the highest risk of a future pregnancy complicated by HDFN.

However, both the medical and general populations seem to take quite a nuanced view of the balance between the life-saving benefits of early transfusion and the small potential future consequences of an RhD-mismatch. In a survey of members of the University of Alabama’s Department of Surgery and the School of Nursing, 87% of males responding on behalf of hypothetical female partners and 89% of females indicated that they would accept a life-saving transfusion that could cause some harm to a future fetus without being more specific about the frequency of the harm to the fetus [34]. The positive response rates remained similar when the risks to the future fetus were quantified at 1:100 and 1:1000. A different survey of women in and around the St. Louis, MO metropolitan area presented the participants with two different sets of probabilities following an emergency transfusion: a fixed rate of fetal harm of 0.3%−4.0% and an absolute risk reduction in her mortality should she receive a transfusion during her trauma resuscitation [35]. Interestingly, when the absolute risk reduction in her mortality was given as 10% and 8%, > 95% of the women claimed that they would accept the transfusion. When the absolute risk reduction of her mortality was given as 4%, which was equivalent to the upper limit of a risk occurring to a future fetus, the transfusion acceptance rate decreased to 90%. When the absolute risk reduction in maternal mortality was reduced to 2%, approximately 80% of respondents replied that they would accept the transfusion and the acceptance rate dropped to approximately 65% when there was a 1% risk reduction in maternal mortality. Thus, the respondents were clearly balancing their perception of their own benefit from receiving a transfusion in an emergency situation with the risk to future pregnancies.

Summary

It is worth repeating that if RhD-negative RBCs or LTOWB are available, they should be used for injured FCPs of unknown RhD-type who are in need of an emergency transfusion. However, the supply of RhD-negative blood products is severely limited and a life-saving transfusion should never be withheld for fear of future alloimmunization events even if the unit is RhD-positive. Prehospital and early in-hospital transfusion are a vital part of modern damage control resuscitation, and it is clear that the general public values their own survival over the highly manageable risks that might occur following an RhD-positive RBC transfusion. Health care professionals should listen to the community and make blood products widely available.

Disclosure statement

No potential conflict of interest was reported by the author(s).