The Egyptian Journal of Haematology

: 2017  |  Volume : 42  |  Issue : 1  |  Page : 9--13

Alloimmunization in Egyptian children with transfusion-dependent B-thalassaemia: a major challenge

Nahla F Osman1, Seham A Ragab2, Mohamed A Soliman3,  
1 Department of Clinical Pathology, Haematology Unit, Menoufia University Hospitals, Shebin El Kom, Egypt
2 Department of Pediatric Medicine, Menoufia University Hospitals, Shebin El Kom, Egypt
3 Department of Clinical Pathology, Menoufia University Hospitals, Shebin El Kom, Egypt

Correspondence Address:
Nahla F Osman
Department of Clinical Pathology, Faculty of Medicine, Menoufia University, Shebin El Kom


Purpose B-thalassaemia is a common health issue in Egypt. However, few national studies were carried out to address the problem of alloimmunization and common alloantibodies in multitransfused thalassaemia patients. This study was designed to address those issues with the aim of optimizing the management in those patients. Patients and methods The study included 281 multitransfused B-thalassaemia Egyptian children from Delta region. Antibody screening and identification were carried out using column agglutination technology. Results Sixty-seven (23.8%) patients were found to have alloantibodies. Anti-Kell and anti-E antibodies were the most commonly encountered antibodies seen in 37.3 and 34.3% of patients, respectively. There was no significant difference in alloimmunization rate between boys and girls or between patients with thalassaemia major and intermediate. Conclusion Alloimmunization is seen in nearly a quarter of multitransfused B-thalassaemia patients. Development of red blood cell antibodies is multifactorial; however, a significant proportion of those can be prevented if pretransfusion testing involved cross-matching for the most immunogenic minor red blood cell antigens. Although this would increase the upfront cost, the long-term cost is likely to fall and is likely to improve patients’ management.

How to cite this article:
Osman NF, Ragab SA, Soliman MA. Alloimmunization in Egyptian children with transfusion-dependent B-thalassaemia: a major challenge.Egypt J Haematol 2017;42:9-13

How to cite this URL:
Osman NF, Ragab SA, Soliman MA. Alloimmunization in Egyptian children with transfusion-dependent B-thalassaemia: a major challenge. Egypt J Haematol [serial online] 2017 [cited 2017 Jun 24 ];42:9-13
Available from:

Full Text


Haemoglobinopathies are the most common genetic disorders with a gene frequency of 0.03% and, each year, a thousand new cases per 1.5 million live births suffer from B-thalassaemia in Egypt. The mainstay in management in patients with B-thalassaemia disease is lifelong transfusion of red blood cells (RBCs) and iron chelation [1],[2]. In our organization, haemoglobinopathy patients are among the highest annual recipients of blood products. It constitutes a significant socioeconomic challenge and creates a major burden to health service. Alloimmunization, formation of alloantibodies to the transfused RBC antigens, is a common complication of chronic transfusion, with recorded frequencies ranging from 4 to 37% in different studies carried out on haemoglobinopathy patients [3].

Alloimmunization itself complicates the medical management in these patients further, as it increases the cost and complexity of the cross-matching, shortens the RBC survival, delays the provision of safe transfusion and may accelerate iron loading [3],[4]. Few studies were carried out to determine the size of the problem and to identify the common alloantibodies in our thalassaemia cohort. This study aimed at identification of the frequency of alloimmunization and the alloantibody pattern in Egyptian children with repeatedly transfused B-thalassaemia, with the aim of improving the process of blood/donor selection in those children and optimizing their management.

 Patients and methods

A cross-sectional study was carried out on 281 children with B-thalassaemia who presented to the Paediatric Haematology Unit, Menoufia University Hospital between June 2015 and February 2016, after consenting their parents. These patients required repeated transfusion and received a minimum of three transfusions to qualify for inclusion in our study. Clinical data, including age, sex, blood group and history of splenectomy, were collected for those patients. Data of 73 children and 1237 adults, who had no history of transfusion, were used as historical control to identify the alloimmunization rate in the population. Pretransfusion testing was carried out at the Transfusion Centre, Menoufia University, Egypt. The protocol of the study was approved by the local ethics committee before starting.

All patients were examined for the presence of alloantibodies and autoantibodies. Patients with positive screen were further tested to identify the antibody specificity.

Local transfusion policy

ABO-matched and RhD-matched RBCs are the main products used for transfusion in these patients. Blood samples for RBC antibody screening, antibody identification, direct antiglobulin test (DAT) and indirect antiglobulin test and compatibility testing were collected and tests were performed as appropriate. Patients’ samples were sent to the National Transfusion Centre, Cairo, for alloantibody identification in difficult cases including those with autoantibodies (DAT positive) for elution studies.

Patients received their transfusion at the paediatric day unit usually as day case, and they received ABO-compatible and D-compatible packed RBCs.

Laboratory tests

Antibody screening and identification were carried out using column agglutination technology by microtyping system reagent (Identisera Diana P Grifols; Spain and Grifols DG gel card, Diana; Spain). Antibody identification was done for patients with positive screening test. Commercial RBC panel (Identisera Diana P Grifols, spain), composed of 11 vials containing papainized human RBCs of group (O)blood group in low ionic strength saline, was used to cover Rh, Kell, duffy, Kidd, MNS and Lewis systems. The principle of the test is based on the gel technique described by Lapierre et al. [5].

DAT was performed using a polyspecific antihuman globulin (Grifols DG gel card, Diana; Spain).

Statistical analysis

We used descriptive statistics and bivariant analysis that was carried out using SPSS, version 15 (IBM analytics, Chicago, Illinois, USA).


The study included 281 children aged between 1.4 and 14.2 years (148 boys and 133 girls). The studied group comprised 208 thalassaemia major patients and 73 thalassaemia intermediate patients.

Sixty-seven of the 281 patients developed one or more alloantibodies (23.8%) to the transfused RBC antigens; of those, 65 (23.1%) had clinically significant IgG antibodies. The total number of alloantibodies identified in these patients was 77. Of these, 72 were considered of potential clinical significance. Fifty-eight (20.6%) patients had one antibody, eight (2.8%) had two antibodies and one patient had three antibodies. None of the children involved in the study had more than three alloantibodies. The distribution and specificity of RBC alloantibodies in all patients are presented in [Table 1]. Historical control showed an alloimmunization rate of 0.5%.{Table 1}

The immunization rate was not significantly different in boys (36/148, 24.3%) compared with girls (31/132, 23.3%, P>0.05). The mean age of the alloimmunized group was significantly higher than that of those with no alloantibodies (mean: 7.93±2.62 and 4.97±1.42 years, respectively, P<0.001) ([Table 2]).{Table 2}

Children with thalassaemia major received significantly more blood transfusion (159.45±46.08 ml/kg/year) compared with those with thalassaemia intermediate (122.5±39.24 ml/kg/year) (P=0.037) ([Table 3]). However, there was no significant difference in the rate of alloimmunization between thalassaemia major (52/208, 25%) and thalassaemia intermediate group (15/73, 20.5%) (P>0.05).{Table 3}

At the time of the study, 89 (31.7%) patients had splenectomy. Of those, 31 (34.8%) had alloantibodies compared with 36 (18.8%) out of 192 with intact spleen (P<0.01) ([Figure 1]).{Figure 1}

Anti-Kell was the most frequently encountered antibody in this multitransfused group, being detected in 25 (37.3%) patients. Anti-E was also common and seen in 23 (34.3%) patients. Anti-D antibodies were detected in five (7.4%) children, of whom four were typed as RhD negative. Four out of the eight children who developed two antibodies had anti-E and anti-Kell, two had anti-E and anti-S, one had anti-M and anti-C and one patient had anti-Lea and anti-Fya. Alloantibodies to M, Kell and S antigen specificities were seen in a 14-year-old boy.


Limited data are available regarding alloimmunization in frequently transfused thalassaemia patients in our area. In the present study, we investigated the prevalence of the alloimmunization and antibody specificity profile in frequently transfused Egyptian children with B-thalassaemia.

The incidence of alloimmunization in our study was 23.8%. Similar rates of 22 and 22.8% were reported by Singer et al. [6] and Hessein et al. [7], respectively. However, in the former study there was discrepancy in RBC phenotype between donors and recipients who were mainly of Asian descent. El-Danasoury et al. [8] reported a slightly lower overall alloimmunization rate of 19.5% in their study that was carried out on 235 Egyptian patients with thalassaemia. This can be, at least partially, due to involvement, in their study, of a proportion of patients with low donor exposure who showed a clear, but not statistically significant, trend of lower alloimmunization.

In a North American and UK multicentre study of 697 thalassaemia patients with diverse racial background, an alloimmunization rate of 16.5% was reported. This lower rate is likely due to the difference in transfusion practice, as some of the centres involved in the study transfused exclusively phenotypically matched Kell and Rh RBCs. In addition, using filtered RBCs for these patients and introducing the universal leucodepletion of blood products, which were shown in a number of studies to reduce alloimmunization, seems to have contributed and had offset the effect of the racial diversity between donors and recipients in the studied group [4],[9],[10].

Alloimmunization rate of 5–10% was reported in Italy and Greece. The clear lower rate of alloimmunization in these countries is attributed to the homogeneity of the population, as well as implementation of strict cross-matching policy since mid-1980s and the extended matching strategy subsequently, which has led to a drop in the number of new cases of alloimmunization [11],[12].

Higher incidence of alloantibody formation of 30% was reported by Ameen et al. [4] and was explained by the heterogeneity and the racial diversity in donors and recipients.

In our study, alloimmunized children were significantly older compared with those with no alloantibodies. This agrees with the results that reported higher incidence of alloimmunization with age logically because of increased exposure to foreign red cell antigen [7],[8],[10].

In our studied group, the rate of splenectomy, which is performed in thalassaemia patients to alleviate hypersplenism or improve the transfusion requirements, was comparable to that published by other groups [7],[8]. We demonstrated a higher frequency of alloimmunization in splenectomized patients. Similar results were obtained by Singer et al. [6], Thompson et al. [10] and El-Danasoury et al. [8]. This is thought to be due to the alteration of the immune response following splenectomy and postsplenectomy RBC membrane conformational changes, which are thought to be associated with an increased immunogenicity in addition to the absence of the role of the spleen in removing misshaped RBCs [3].

However, we have to be careful regarding interpretation of these data, as some of these studies, as well as in ours, have not reported the onset of alloimmunization in relation to the splenectomy, which might have been done merely because of high transfusion requirement as a consequence of existing alloimmunization.

On the other hand, Ho et al. [13] did not demonstrate correlation between splenectomy and RBC alloantibody formation, which may be because of the use of phenotypically matched RBCs and hence decreased alloimmunization regardless of other patient characteristics.

In spite of the fact that female sex has been shown as a risk factor for alloimmunization in general population [14], we, among others [4],[8],[10], did not demonstrate a significant correlation between sex and alloimmunization in thalassaemia patients. On the contrary, Jain et al. [15] found significantly higher alloimmunization frequency in female compared with male individuals. However, the age group involved in their study and whether multiparous female individuals were included is not clear.

The most frequently encountered alloantibodies in our study were anti-Kell and anti-E. This predominance is not unexpected given the strong immunogenicity of these antigens. Similarly, anti-Kell and anti-E were the commonly seen alloantibodies in other studies. In Western countries, anti-C is also commonly seen [7],[8],[10].

Anti-D alloantibodies were seen in our patients, although ABO–RhD-matched RBCs were used. This is likely because of transfusion of RBCs with weak D antigens that was not detected during pretransfusion testing, as retest of RhD-negative donors for weak D is not part of routine pretransfusion laboratory workup in Egypt. In addition, the epitopic diversity of RhD variants can mount a sufficiently strong immune response in recipients with different variants and is responsible for the formation of anti-D alloantibodies in RhD-positive individuals. Allosensitization to RhD antigen has been reported in other studies as well [4],[7],[8],[10].

Alloimmunization increases the cost and burden of treating patients on chronic transfusion program not only for the health service but for patients and their families as well. Several factors are known to affect the development of alloantibodies in those patients including the recipients’ immune system, the discrepancy between the RBCs antigens in donors and recipients, the duration of transfusion, the spleen status and, last but not least, the transfusion practice − for example, the use of leucodepleted blood products and cross-matching for minor red cell antigens. Universal prestorage leucodepletion has been adopted earlier this year in our centre, and it is likely to improve the rate of alloimmunization. Extended cross-matching is another effective step in preventing alloimmunization, and although implementation of such policy would increase the upfront cost subsequent decrease of the cost of pretransfusion testing is likely to decrease overall cost of patient care. Cross-matching for frequently seen antibodies, anti-E and anti-Kell, is a reasonable compromise that is likely to decrease the alloimmunization with modest increase of the cost.

Avoiding alloimmunization in patients on lifelong transfusion therapy should decrease the cost of treatment on the long run and improve the patients’ quality of life, as it will avoid the rise in transfusion requirements and time spent in the hospital. If the work/school absence is taken in consideration, the cost/benefit will be in favour of prevention of alloimmunization. Thus, we recommend transfusion of Kell-matched and Rh (or at least E)-matched blood particularly in immunocompetent patients who are likely to receive long-term transfusion.

Financial support and sponsorship


Conflicts of interest

There are no conflicts of interest.


1El-Beshlawy A, El-Shekha A, Momtaz A, Said F, Hamdy M, Osman O et al. Prenatal diagnosis for thalassaemia in Egypt: what changed parents’ attitude?. Prenat Diagn 2012; 32:1–6.
2El-Beshlawy A, Youssry I. Prevention of hemoglobinopathies in Egypt. Hemoglobin. 2009; 33(Suppl 1):S14–S20.
3Chou S, Liem R, Thompson A. Challenges of alloimmunization in patients with haemoglobinopathies. Br J Haematol 2012; 159:394–404.
4Ameen R, Al-Shemmari S, Al-Humood S, Chowdhury R, Al-Eyaadi O, Al-Bashir A. RBC alloimmunization and autoimmunization among transfusion-dependent Arab thalassemia patients. Transfusion 2003; 43:1604–1610.
5Lapierre Y, Rigal D, Adam J, Josef D, Meyer F, Greber S, Drot C. The gel test: a new way to detect red cells antigen-antibody reactions. Transfusion 1990; 30:109–113.
6Singer S, Wu V, Mignacca R, Kuypers F, Morel P, Vichinsky E. Alloimmunization and erythrocyte autoimmunization in transfusion-dependent thalassemia patients of predominantly Asian descent. Blood 2000; 96:3369–3373.
7Hussein E, Desooky N, Rihan, Kamal . Predictors of red cell alloimmunization in multitransfused Egyptian patients with B-thalassemia. Arch Pathol Lab Med 2014; 138:684–688.
8El-Danasoury A, Eissa D, Abdo R, Elalfy M. Red blood cell alloimmunization in transfusion-dependent Egyptian patients with thalassemia in a limited donor exposure program. Transfusion 2012; 52:43–47.
9Wang L, Liang D, Liu H, Chang F, Wang C, Chan Y, Lin M. Alloimmunization among patients with transfusion-dependent thalassemia in Taiwan. Transfusion Medicine 2006; 16:200–203.
10Thompson A, Cunningham M, Singer S, Neufeld E, Vichinsky E, Yamashita R et al. Red cell alloimmunization in a diverse population of transfused patients with thalassaemia. BJH 2011; 153:121–128.
11Datta S, Mukherjee S, Talukder B, Bhattacharya P, Mukherjee K. Frequency of red cell alloimmunization and autoimmunization in thalassemia patients: a report from Eastern India. Adv Hematol. 2015; 2015:1–6.
12Politis C, Hassapopoulou E, Halkia P, Kourakli A, Mougiou A, Zervou E et al. Managing the patient with haemoglobinopathy and multiple red cell antibodies. ISBT Sci Ser 2016; 11(Suppl S1):54–61.
13Ho HK, Ha SY, Lam CK, Chan GC, Lee TL, Chiang AK, Lau YL. Alloimmunisation in Hong Kong southern Chinese transfusion dependent thalassemia patients. Blood 2001; 97:3999–4000.
14Bauer MP, Wiersum-Osselton J, Schipperus M, Vandenbroucke JP, Briët E. Clinical predictors of alloimmunization after red blood cell transfusion. Transfusion 2007; 47:2066–2071.
15Jain R, Choudhury N, Chudgar U, Harimoorthy V, Desai P, Perkins J, Johnson ST. Detection and identification of red cell alloantibodies in multiply. Indian J Hematol Blood Transfus 2014; 30:291–296.