Sickle Cell Disease (SCD) and Blood Transfusion: Prevalence of Alloantibodies and Red Cell Antigens among Multiply Transfused Individuals with SCD in Selected Tertiary Health Institutions in Abuja, Nigeria

INTRODUCTION

Sickle cell disease (SCD) is the most common severe inherited disease in the world and one of the 50 most common causes of non-communicable death globally.^[1] SCD is an inherited hemoglobinopathy associated with significant morbidity, reduced quality of life, and premature mortality.^[2,3] It affects an estimated 20–25 million individuals globally, with the highest prevalence observed in tropical regions, particularly sub-Saharan Africa, India, and the Middle East.^[4] Nigeria plays a significant role in the global burden of SCD, with an estimated 100,000–150,000 newborns affected each year— representing roughly one-third of the global total.^[5]

The prevalence of SCD within Nigeria ranges from 2% to 3%.^[6] Some SCD individuals are relatively stable and rarely require a blood transfusion;^[7] others often require multiple blood transfusions, which may result in red blood cell (RBC) alloimmunization. Alloimmunization in SCD has several adverse effects, including a markedly elevated risk of hemolytic transfusion reactions, which may include hyperhemolytic reactions and iron overload.^[8] The occurrence rate of alloimmunization to red cell antigens is a serious complication of transfusion. Alloimmunization is a well-known risk factor associated with repeated transfusions, and the use of extended RBC antigen matching has been well documented to reduce alloimmunization.^[9,10] The most serious consequence of alloimmunization is the risk of developing a delayed hemolytic transfusion reaction (DHTR), which may be grievous. In several DHTR individuals, the hemoglobin level falls below the pre-transfusion level, suggesting that hemolysis of the transfused RBCs and the patient’s RBCs occurred, a condition called hyperhemolysis. In cases of DHTRs, additional transfusions may aggravate hemolysis, further worsening anemia and complicating patient management.^[11] This underscores Nigeria’s pivotal role in understanding and mitigating the global burden of SCD. In cases of DHTRs, the destruction of the patient’s RBCs is partly attributed to the development of autoantibodies, often precipitated by prior alloimmunization. It is partially explained by the presence of autoantibodies because alloimmunization is thought to trigger antibody production. However, DHTR/ hyperhemolysis cases also occur due to the absence of detectable alloantibodies or autoantibodies.^[8] In African populations, especially in Nigeria, the antigen frequency data are sparse, leading to mismatched transfusions and higher rates of alloimmunization. Furthermore, SCD patients are at high risk of developing alloantibodies due to repeated transfusions. These alloantibodies can complicate future transfusions, making it difficult to find compatible blood, and may lead to hemolytic^[12] transfusion reactions, both immediate and delayed. It can impact clinical outcomes, increasing morbidity and hospital stays. Identifying the types of alloantibodies prevalent in SCD patients helps blood banks prepare antigen-negative units and adopt preventive strategies like extended phenotype matching.

The study objectives include determining the distribution of minor red cell antigens in sickle cell individuals and control subjects, and determining the presence and type of alloantibodies in sickle cell individuals and control subjects.

METHODS

This was a cross-sectional study using a purposive sampling method. The study was conducted among individuals with SCD attending the National Hospital Central Area, Abuja, Federal Medical Center Jabi, Abuja, and University of Abuja Teaching Hospital, Gwagwalada, Abuja. The study population included 100 SCD individuals of all ages who had undergone at least three transfusions and 100 healthy non-SCD individuals who had not been transfused before as controls. Quantitative variables were grouped into categories to facilitate comparison between subgroups.

Both qualitative and quantitative methods were used for the data collection and laboratory analysis of the samples. The data were collected over a period of 9 months from October 2022 to July 2023.

This study adhered to the Standards for Reporting Qualitative Research to ensure transparency and completeness in presenting qualitative aspects such as problem formulation, research questions, data collection, and analysis. For components involving numerical data, the Strengthening the Reporting of Observational Studies in Epidemiology guidelines were followed to maintain rigor in methodology, statistical analysis, and presentation of results.^[12]

RESULTS

Most of the participants were aged ≤1–10 years (38%), while the least common age groups were aged 31 and 46 years (3%). There is an increase in the 11–20 years (44%) category, and the 21–30 years category has the lowest percentage (10%) of those experiencing crises. The steady state shows a greater concentration in the younger age groups, particularly in the 0–10 years (38%) category, as shown in Table 1 below.

Most of the study populations had the HB (SS) variant (95% of the total). The CC variant, which is a very rare variant, was 1%, while the SC variant represented 4% of the total population, as shown in Figure 1. Table 2 shows the presence and type of alloantibodies used in SCD individuals. Sixteen percent (4) developed anti-D antibodies, 32% (8) developed anti-E, and 24% (6) developed anti-Fya. Only 4% each developed JKb, Lua, M, Lea, and C antibodies, and 2% developed anti-S antibodies. Most sickle cell individuals in this study were O+ blood cells, constituting 55% of the total SCD population studied. This finding suggested a greater incidence of SCD among individuals with O+ blood. The A+ and B+ blood groups represented 15% and 19%, respectively. Rh-negative blood groups (O− and B−) accounted for a smaller percentage of the bacteria, with O− (7%) and B− (2%). Rh-negative blood groups are less common in the SCD population studied. The AB and A blood groups had the lowest representation, accounting for 1% of each SCD population. The high percentage of O+ blood groups among the SCD population may be a significant factor in the prevalence of this disease within this blood group, as shown in Table 3. Approximately 84% of the test subjects were negative for antigen A, 84% were negative for B, 45% were negative for O, 99% were negative for A− and AB, 98% were negative for B−, and 93% were negative for O−. Figure 2 shows the presence and type of minor RBC antigens in the study population. For the RH group, 92% of the study population was D positive, while 86% of the control group was D positive, 50% and 38% were D positive for the subject, and the control was D positive for the C. For the E group, 41% and 42% were D positive for the subject and control, respectively. For c, 63% and 40% of the samples were positive for the subject and control, respectively. For e, 87% and 8% of the samples were positive for the subject and control, respectively. Among the MNSs in the blood group, 58% were positive for the subjects, while 8% were positive for the control group. For N, 11% and 12% of the samples were positive for the subjects and controls, respectively. For S, 14% and 20% of the samples were positive for the subjects and controls, respectively. For s, 14% and 22% of the samples were positive for the subjects and controls, respectively [Table 4].

Figure 1:

Genotype variants in individuals with sickle cells. SS: Homozygous sickle cell anemia 95%, SC: Heterozygous sickle cell-hemoglobin C disease 4%, CC: Homozygous hemoglobin C disease 1%.

Figure 2:

Frequency, presence and type of minor red cell antigens present in sickle cell disease (SCD) individuals and controls.

DISCUSSION

The age distribution of SCD individuals revealed that individuals in the steady state aged <1 and 10 years had the highest frequency of SCD, although there was no association between age and SCD incidence. SCD across different age groups can be influenced by disease onset and progression, treatment and management considerations, mortality, and survival rates. Usman et al. indicated that most of the participants were aged five;^[13] this is consistent with the findings of the present study, possibly because SCD symptoms manifest early in life, typically within the first few months after birth. Infants with SCD may experience complications such as painful vaso-occlusive crises, hand and foot swellings (dactylitis), acute chest disease, and infections, which can lead to hospitalization and medical intervention.^[14] 95% of the study population with SCD were homozygous for sickle-hemoglobin (SS), and 1% of the population had CC variants, whereas 4% of the study population were heterozygous for sickle-hemoglobin (SC) variants. Similarly, Nwobuko observed trends in southeastern Nigeria: 98.55% and 1.45% for homozygous sickle-hemoglobin (SS) and heterozygous sickle-hemoglobin C (SC) variants, respectively.^[5] There was a noticeable change in sex distribution compared to that of the control and steady-state groups. The dominance of females in the control group and in the steady state may suggest variations in the increase in mortality and morbidity in male SCD individuals, even though SCD is not sex linked.^[15] Furthermore, Masese et al. and Fitzmaurice reported that higher female survival may be due to higher fetal hemoglobin levels.^[16,17] Understanding the reasons behind the observed shifts could provide insights into how different genders respond during crises. The increase in male representation during the crisis may be indicative of a specific response or vulnerability that was more pronounced in males during crisis scenarios, and this aligns with the work done by Ugwu in Ozalla, the eastern part of Nigeria.^[7] Assessing the gender distribution in each group can contribute to a broader understanding of diversity and inclusion dynamics within the community or context under study. The distribution of males and females within each group is likely due to random variation and is not reflective of a systematic association or effect. Further exploration of the factors contributing to these sex distributions could provide a more comprehensive understanding of the observed patterns. In his study in Sokoto, Usman et al. found gender to be equal.^[13]

In this study, the prevalence of red cell alloimmunization among multiply transfused SCD individuals in Abuja was 25%, while in Ghana, Boateng et al. reported an alloimmunization prevalence of 16%.^[18] Similarly, Natukunda et al. in Uganda documented a lower rate of 6.1%.^[19] The prevalence in Cameroon was 55%, while Morocco recorded 90.4% in 2013, and Cote d’ivoire recorded a prevalence of 28.6%, New Delhi (India) recorded a prevalence of 5.2%. By comparison, studies in developed countries such as the United States have reported alloimmunization rates ranging from 18% to 36% in SCD populations, with anti-E, anti-C, and anti-K as the leading specificities.

The results of the alloantibodies observed in this study are suggestive that a subgroup of individuals within the population has developed antibodies against the D antigen, which is a common target for immunization due to its high immunogenicity.^[20] The presence of anti-E antibodies in 32% of the population indicates that a larger proportion of individuals have developed antibodies against the E antigen. Anti-E antibodies can pose challenges in transfusion compatibility, as they can lead to adverse reactions if individuals are mismatched with inappropriate blood types. Rh antibodies contributed to approximately 52% of the total antibodies identified, and a total of 24% of the studied population developed anti-Fya antibodies. This antibody development is less common than that of anti-D and anti-E antibodies. Anti-Fya antibodies are specific to the Duffy blood group system and can influence the compatibility of blood transfusions. The development of antibodies against the JKb, Lua, M, Lea, and C antigens in only 4% of the population suggests a decreased prevalence of these specific antibodies. These findings indicate that these antigens may be less immunogenic or less likely to stimulate the production of antibodies in the studied population. The presence of anti-S antibodies was 2% in the population, suggesting a relatively low prevalence. Anti-S antibodies are associated with the S antigen, which is part of the MNSs blood group system. Awareness of the presence of anti-S antibodies is crucial for safe blood transfusion practices. No antibodies were detected in the control group. The most frequent antibodies reported by Adewoyin et al. were anti-E, followed by anti-C before anti-D, which is consistent with the findings of the present study.^[21] Anti-D was followed by anti-C, whereas in our study, we observed that anti-Fya was the next most common after anti-D. Wilson et al. reported that the most frequent antibodies were against the E antigen, which is consistent with what we observed from this work.^[22] Similarly, Samarah et al. reported the frequency of RBC alloimmunization among SCD individuals to be 7.76%, with anti-K showing the highest frequency.^[23] This finding aligns with our findings, but we did not observe anti-K. Coleman et al. reported that 26% of the delay in the hemolytic transfusion reaction was associated with Rh antibodies, while 38% of the non-Rh antibodies were associated with Rh antibodies.^[24] In contrast, we identified 52% of the Rh antibodies and 48% of the non-Rh antibodies. Our findings do not agree with those of Coleman et al., who found 26% Rh antibodies and 38% non-Rh antibodies. This could be due to discrepancies in population differences, antigen distribution, reagent sensitivity, or transfusion practices. Kangiwa et al. reported a frequency of alloimmunization among individuals with multiple transfusions of SCD that was consistent with our findings in this study, except for anti-KEL, which was not observed in this study.^[25]

These findings in this study align with previous African and Middle Eastern studies, where Rh and Duffy system antibodies are more common due to mismatches in transfused units. However, the absence of anti-K or anti-KEL may be because of the low frequency of the Kell antigen in the donor population, limited antigen exposure in this cohort, or insufficient panel sensitivity for detecting Kell-group antibody antigens.

Ugwu reported that the prevalence of RBC alloantibodies among multiple transfusion individuals with SCD was 9.3%. The antibodies they identified were slightly different from those recorded in the present study, and no antibodies were detected among the control group.^[7] Understanding the prevalence of these antibodies within a population of SCD individuals who have been transfused multiple times is vital for proper blood typing and cross-matching to ensure safe and compatible blood transfusions. These findings emphasize the importance of regular antibody screening and careful selection of compatible blood units for transfusion recipients.

The O+ blood group made up a greater percentage of the total number of SCD individuals studied. This finding suggested a greater incidence of SCD among individuals with O+ blood. The A+ and B+ blood groups represented 15% and 19%, respectively. No AB blood group was found in this study. Rh-negative blood groups (O− and B−) accounted for a smaller percentage of the population, with O− (7%) and B− (2%) exhibiting Rh negativity (8%), and they were less common in the SCD population studied. This finding is in line with previous records showing that blood group A had the lowest percentage, group B was a bit higher than group A, group O had the highest prevalence, and the AB blood group was not present.^[26] They also recorded a very high Rh positivity and very low percentage of Rh negativity,^[27-29] and all the authors also recorded low Rh negativity frequencies in Nigeria. Other researchers have also recorded low frequencies.^[30-32] In addition, Ab et al. recorded a high frequency of blood group O in Abuja.^[33] The authors equally observed that Rh-positive individuals had the same frequency, and AB- and Rh-negative individuals had the lowest frequency. A high prevalence of blood group O+ was also observed by Olayanju et al.^[34] Olayanju et al.^[34] and Abbas et al.^[35] recorded ABO blood group incidences of O > A > B > AB, while the prevalence of Rh+ was high, which could be influenced by genetic diversity,^[36] evolutionary factors, population mixing, and historical dynamics. The specific combination of these factors may have contributed to the high prevalence of both group O and Rh+ blood types within the Nigerian population and Abuja in particular. The frequencies of minor red cell antigen phenotypes in the study population align with the findings of Boateng et al. and Yu et al.^[37,38] Lamba et al. found Rhe to be the most common, followed by C, c, and E.^[39] This study recorded a frequency of Lea antigen of 18%, which was consistent with the work of Yu et al.^[38] For Kpa, 9% was recorded in this study, which was in line with the work of Tariq et al., who reported a low frequency of Lua, which aligns with our work.^[27] This could be because certain blood group antigens may exhibit lower frequencies in populations that have been historically isolated. If the Lua antigen is less common in ancestral populations, limited genetic interchange with other populations may have contributed to its low frequency within specific ethnic or regional groups. Within the MNS blood group system, the antigens identified in this study were M (58%), N (11%), S (14%), and S (14%), whereas the occurrence rate of alloimmunization to red cell antigens is a serious complication of transfusion. SCD individuals are prone to developing alloantibodies despite being of the same race as the donor.

Buhari et al. reported a higher frequency among pregnant women in Sokoto, northern Nigeria; however, these antigens were not the most common antigens observed in the population. Within the Duffy blood group system, the antigen frequencies were 19% for Fya and 12% for Fyb, which are higher than those observed by Buhari et al. The Kidd blood group system was found to be moderately low at 14% and 15%, respectively, which is consistent with what Buhari et al. reported and disagrees with what Gwaram and Yusuf reported.^[40,41]

These findings underscore several critical considerations for transfusion policy and immunohematologic research in SCD management.

Limitations of the Study

1. A longitudinal (cohort study) approach would provide a better understanding of how these parameters change over time and with repeated transfusions, offering more robust insights into the long-term risks and immune responses of SCD individuals, this may be considered as further work.
2. Variability in laboratory techniques or limited access to advanced testing facilities could affect the accuracy and reproducibility of the findings.

Cautious Interpretation of Results

The findings highlight a moderate prevalence of clinically significant alloantibodies, with anti-E, anti-FYa, anti-D, anti-S, anti-C, anti-Lea, anti-M, anti-Lua, and anti-JKb emerging as the most commonly encountered specificities. These results align with prior studies in similar settings, reinforcing the need for more antigen-matched transfusion practices for individuals with SCD. The data also show that alloimmunization risk increases with the number of transfusions, supporting a dose-dependent relationship between antigenic exposure and immune sensitization. While trends suggesting higher alloimmunization rates among older and female patients were noted, these did not consistently reach statistical significance and should be interpreted with caution. These subgroup findings, though suggestive, would benefit from further exploration in larger cohorts. Importantly, the interpretation of our results must take into account several limitations. First, the use of purposive sampling and the relatively small sample size may limit the generalizability of the findings to the broader SCD population in Nigeria. Second, some transfusion histories were based on patient recall or incomplete records, introducing the potential for recall bias or misclassification. Third, the cross-sectional design precludes any inference of causality between transfusion burden and alloantibody formation. The multiplicity of analyses, including subgroup and sensitivity analyses, raises the possibility of type I error, especially in the absence of adjustment for multiple comparisons. Nonetheless, these analyses were exploratory in nature and aimed to guide future hypothesis generation. When compared with evidence from other low-resource settings, our findings are consistent with reported alloantibody prevalence rates and specificities. However, variations in donor pool composition, antigenic exposure, and laboratory methods across studies underscore the importance of local data to inform transfusion policy.

Generalizability (External Validity) of the Study Results

The generalizability of our findings is subject to certain constraints due to the study design and setting. The use of purposive sampling and the selection of tertiary healthcare institutions within Abuja means that the sample may not fully represent all individuals living with SCD across Nigeria, particularly those in rural or underserved regions. Differences in transfusion practices, access to healthcare, and the composition of the donor pool in other parts of the country may affect the applicability of our results elsewhere.

Moreover, since the study population consisted of patients receiving care at well-resourced tertiary centers, the observed prevalence of alloantibodies and red cell antigen distribution might differ from those in primary or secondary healthcare settings where transfusion monitoring may be less robust. Therefore, extrapolation of the findings to the national level or to other low-resource settings should be done cautiously.

Nonetheless, the consistency of our results with those from similar studies conducted in other sub-Saharan African countries supports a degree of external validity. The patterns of alloantibody specificity, particularly the predominance of anti-E, anti-FYa, anti-D, anti-S, anti-C, anti-Lea, anti-M, anti-Lua, and anti-JKb, align with existing literature and suggest that the immunologic risks associated with transfusion in SCD patients are broadly similar across comparable settings.

CONCLUSION

The findings of this work showed that 25% of SCD individuals developed alloantibodies, and anti-E was the most common, followed by anti-Fya and anti-D, while all the controls were non-reactive. The distribution of minor RBC antigens in sickle cell individuals showed that the D antigen was more frequent than antigen e and C in SCD individuals attending tertiary health care centers in Abuja. Identifying alloantibodies in SCD individuals is crucial, as it can have implications for future transfusions and impact the overall management of the disease.