The Duffy (Fy) blood group system is important in clinical medicine due to transfusion incompatibilities and haemolytic disease of the foetus and newborn (HDFN).1 However, many causes of acute and delayed haemolytc transfusion reaction and haemolytic disease of the foetus and neonate are underreported due to poor obstetric and neonatal screening of clinically important rare blood group system.2 The Duffy (Fy) antigen is a glycoprotein found on red cell membrane, endothelium, and epithelial cells of alveoli and collecting tubules of the kidneys.2 The antigen act as receptor for chemokines and plasmodium vivax.2 Globally, Haemolytic disease of the foetus and newborn affects an estimated 80 in 100,000 patients annually. 3 It could lead to bilirubin encephalopathy & late anaemia of infancy.4Though more commonly associated with Rhesus and ABO blood group. Severe haemolytic disease of foetus and newborn has been reported with minor blood groups like the Duffy and Kell blood group system.4Approximately 4% of pregnant women with anti-Fya have been reported to have foetus with severe anaemia, some requiring intrauterine transfusion.4Knowing the distribution of blood group antigens in a specific population, helps to determine the risk of alloimmunization, and among women, the risk of haemolytic disease of foetus and newborn. The distribution of Duffy phenotype among pregnant women in Gombe state is not known, hence the need to determine the primary data.

The study took place at Federal Teaching Hospital Gombe (F.T.H.G). This was a cross sectional study. Gombe state is located in northeastern part of Nigeria. It lies in the wooded savanna lands of the Gongola River basin. It is mainly inhabited by the Fulani, Bolewa, Tera (Terawa), Tangale, Hausa, Kanuri, Waja (Wajawa), and Tula peoples.5 A total of 259 pregnant women attending the antenatal clinic at the Federal Teaching Hospital Gombe, who consented, were recruited. A simple random sampling (balloting) technique was used to recruit indigenous pregnant women. Those who picked “yes” were included, whereas those who picked “no” were excluded. Socio-demographic information such as age, religion; tribe, marital status, educational level, occupation and previous obstetric history were obtained from consented participants. Three 3ml of venous blood was aseptically collected into EDTA sample bottle from each participants and transported to the lab for processing. Samples were analysed in batches for Duffy antigen using standard tube technique (LORNE LABORATORY UK),6 ABO, Rhesus blood grouping 7 and screening for Plasmodium species using thin and thick blood film by WHO MM-SOP-08.8 Data was compiled into an excel spread sheet and analysed using IBM SPSS Version 25.9 Analyzed variables were presented inform of frequency and percentage. Level of significance was set at p< 0.05. Ethical approval (NHREC/25/10/2013) was obtained from Research and Ethics Committee of F.T.H.G.

All the women were married, 96.5% had secondary school education and above. Majority (91.5%) were between the ages of 18 and 35 years, with a mean age of 26.71 + 4.58years. Two- hundred and forty-eight (95.7%) where rhesus positive and 114(44.1%) were blood group O positive. (Table 1)

The major indigenous tribes in this study were Fulani’s (45.6%), Tangale’s (17.8%), and Hausa’s (11.2%). Tera’s, Waja’s, Bolewa’s and Tula’s make up the rest (23.5%) of the population. Tula’s were only 5 out of 259 (1.9%) participants. (Table 2)

The most predominant Duffy phenotype among the study population was Fy(a-b-). This was observed in 84.6% of the women (Table3). The Fy(a-b-) was seen in 82.2% of Fulani’s, 89.7% of Hausa’s and 84.8% of Tangale’s, 81.0% of Tera’s. 84.2% of Waja’s. The least proportion (68.75%) was seen among the Bolewa’s (Table 4). The Fy (a+b+) was the least common phenotype seen in only 1.9% of the study population. It was not seen in the Hausas’, Tera’s and the Waja’s. The highest proportion for the Fy(a+ b+) was seen among the Bolewa’s. (Table 4). Thirty-four percent (34%) had a history of miscarriage, 23.6% had history of neonatal jaundice and 8.1% had a history of still births. However the proportion with this bad obstetric history did not vary across the different Duffy phenotypes (Table 5). A hundred and thirty-five (51.1%) pregnant women had malaria parasite by microscopy; 65.2% had P. falciparum infestation, 12.6% had P.vivax, 11.2% had P.malariae and P.ovale (Table 6). P falciparum was the predominant malaria specie seen among participant with Fy (a-b-) (74.52%) and Fy(a+b-) phenotypes (60%), while P.vivax was the predominant specie among those with the Fy(a-b+) (71.43%) and Fy(a+b+) (80%) phenotypes. None of the participants with the null Duffy phenotype has P vivax infestation. (Table 7)

The Duffy blood group antibodies are IgG antibodies and have the ability to cross the placental barrier and hence are associated with risk of haemolytic disease of foetus and new born. We investigated the prevalence and distribution of the Duffy blood group antigen among pregnant women in Gombe State, North - East Nigeria.

In this study, the null phenotype was the most prevalent phenotype seen in 80- 86% of pregnant women from all the major ethnic groups in Gombe, except among the Bolewa, where it accounted for 69% of the pregnant women. This agrees with previous study among blacks in sub Saharan Africa.10In a related study in Sokoto, a state in North West Nigeria, all the pregnant Fulani’s women (20/162) in the study lacked both the Fya and Fyb antigen, implying they also had a null phenotype. Kulkarni et al.,11 Erhabor et al., 12 also reported 98.8% of Duffy negativity among Hausa’s in North West Nigeria. The prevalence rate of Fy (a+b-), Fy (a-b+) and Fy (a+b+) phenotype disagree with the research carried out in north western Nigeria12, where prevalence rate of 4.3%, 5.6% and 0.61% were reported respectively.

In another research carried out at donor clinic of Aminu Kano teaching hospital, using potent antisera, Duffy antigens were not detected among the blood donors. 13

The Duffy antibody is an immune antibody produced in response to previous exposure and has the ability to cause immediate and delayed transfusion reaction and haemolytic disease of the foetus and newborn. In this study, out of 259 participants, 88 (33.98%) had history of miscarriage, 21 (8.11%) still births and 61 (23.55%) had neonatal jaundice. Contrary to expectations, the proportion of pregnant women with bad obstetric history was higher among the pregnant women with null Fy(a-b-) than those with Fy(a-b+). Table 5. The pregnant women with null Fy (a-b-) phenotype had the highest history of miscarriages, still births, and neonatal jaundice. The antibody against the Fya antigen (anti-Fya) is 20 times more potent than the anti-Fyb and is associated with mild-to-severe HDFN, than the anti-Fyb. The anti-Fyb is rarely associated with mild HDFN.14 Moreover, studies have shown that blacks with the F (a-b-) phenotype who develop Duffy antibodies, do not produce anti-Fyb but usually produce anti-Fya and anti Fya3 or Fy5.15This is because the Fy(a–b–) phenotype in blacks is associated with a mutation (a single amino acid substitution at position 46) which impairs the erythroid promoter, GATA-1 binding motif. The mutation prevents the transcription of the FY gene in the RBCs only while the Duffy Fyb antigen is expressed on the other tissues.16

These and the dosage effect observed in the Duffy blood group though not statistically significant, may explain the differences observed in the obstetric history across the various phenotypes.

The Duffy antigens serve as receptor for diverse group of chemokines and P. vivax. The Duffy antigen receptor plays a vital role as an immune regulator where it served as a chemokine sink during inflammatory processes. In this study, out of 259 participants, 135 had plasmodiasis, out of which 11.1% were infested with P. malariae and P. ovale respectively, 12.6% P. vivax, and 65.2% P. falciparum. The prevalence rates of plasmodiasis among participants with null Duffy phenotype Fy (a-b-) was 0.0% for P. vivax, 12.26 % P. malaria, 74.53 % P. falciparum and 13.21 % P. ovale. These findings supported previous studies which stated that red cells of null Duffy phenotype individuals resist invasion by P. vivax which was due to non-expression of Duffy antigen on the red cell surface, hence the resistivity.17This study did not screen for allo-antibodies among the pregnant women with the null phenotype. Doing this will have provided more data on the risk of alloimmunization among pregnant women with the null phenotype.

The prevalence of the null Duffy phenotype is high among pregnant women in Gombe and hence there may be a high risk of alloimmunization and obstetric complications. The least seen phenotype is the F (a+b+) phenotype.