Constantina Politis1,2, Myrsini Parara1,2, Jenny Kremastinou2, Eleni Hasapopoulou3, Aliki Iniotaki4, Alexandra Siorenta4, Clive Richardson1,5, Anna Papa6, Lilian Kavallierou1,2, Marina Asariotou1,2, Olga Katsarou7, Athina Mougiou8, Lukas Dadiotis9, Zafeiria Alexandropoulou10, Angelica Megalou11, Evangelia Magoula12, Margarita Papadopoulou13, Danai Pervanidou2, Agoritsa Baka2, Christos Hadjichristodoulou14. 1. Coordinating Haemovigilance Centre. 2. Hellenic Centre for Disease Control and Prevention, Athens, Greece. 3. AHEPA University Hospital Blood Centre Thessaloniki, Thessaloniki, Greece. 4. National Histocompatibility Centre. 5. Panteion University of Social and Political Sciences, Athens, Greece. 6. Arboviruses National Reference Laboratory, Aristoteleio University, Thessaloniki, Greece. 7. Laiko General Hospital Blood Establishment, Athens, Greece. 8. Rio University Hospital Blood Centre, Patras, Greece. 9. Tzanio General Hospital Blood Establishment, Piraeus, Greece. 10. Asclepieio General Hospital Blood Establishment, Voula, Greece. 11. Evangelismos General Hospital Blood Establishment, Athens, Greece. 12. Veria General Hospital Blood Establishment, Imathia, Greece. 13. Katerini General Hospital Blood Establishment, Pieria, Greece. 14. Department of Hygiene and Epidemiology, University of Thessaly, Larissa, Greece.
Abstract
BACKGROUND: West Nile virus (WNV) infection, commonly asymptomatic, may cause mild West Nile fever (WNF) or potentially fatal neuroinvasive disease (WNND). An outbreak of 262 cases of the new Lineage 2 strain in Greece in 2010 continued with high mortality (17%) in WNND. The objective was to investigate ABO, D, and Lewis blood groups, as well as HLA Class I and Class II alleles, in relation to WNV Lineage 2 disease morbidity. STUDY DESIGN AND METHODS: A cohort of 132 Greek WNV cases in 2010 to 2013 (65% male; mean age 64 years; 41% WNF, 59% WNND) was compared to 51,339 healthy WNV-negative blood donors and 246 healthy subjects. RESULTS: Blood group A was more common in WNV cases (51%) than blood donors (39%) and group O less common (32% vs. 42%). D negativity within group A was higher in WNV than in blood donors (18% vs. 10%, p = 0.044). The frequency of secretors (Lewis(a-b+)) was 60% in WNV and 68% in donors (p = 0.16). HLA alleles C*08, DRB1*O4:O5, and DQB1*O2 occurred significantly less frequently in WNV than controls (p < 0.05 unadjusted for multiple testing) and DRB1*10:O1 more frequently (p = 0.039). CONCLUSION: This first study of symptomatic WNV Lineage 2 suggests A/D negativity as a new risk factor associated with WNV infection and level of morbidity. Further studies are required of the possibility that HLA C*08, DRB1*O4:O5, and DQB1*O2 are protective alleles and DRB1*10:O1 a "susceptible" allele to WNV infection and the role of secretor status in relation to WNV infection.
BACKGROUND: West Nile virus (WNV) infection, commonly asymptomatic, may cause mild West Nile fever (WNF) or potentially fatal neuroinvasive disease (WNND). An outbreak of 262 cases of the new Lineage 2 strain in Greece in 2010 continued with high mortality (17%) in WNND. The objective was to investigate ABO, D, and Lewis blood groups, as well as HLA Class I and Class II alleles, in relation to WNV Lineage 2 disease morbidity. STUDY DESIGN AND METHODS: A cohort of 132 Greek WNV cases in 2010 to 2013 (65% male; mean age 64 years; 41% WNF, 59% WNND) was compared to 51,339 healthy WNV-negative blood donors and 246 healthy subjects. RESULTS: Blood group A was more common in WNV cases (51%) than blood donors (39%) and group O less common (32% vs. 42%). D negativity within group A was higher in WNV than in blood donors (18% vs. 10%, p = 0.044). The frequency of secretors (Lewis(a-b+)) was 60% in WNV and 68% in donors (p = 0.16). HLA alleles C*08, DRB1*O4:O5, and DQB1*O2 occurred significantly less frequently in WNV than controls (p < 0.05 unadjusted for multiple testing) and DRB1*10:O1 more frequently (p = 0.039). CONCLUSION: This first study of symptomatic WNV Lineage 2 suggests A/D negativity as a new risk factor associated with WNV infection and level of morbidity. Further studies are required of the possibility that HLA C*08, DRB1*O4:O5, and DQB1*O2 are protective alleles and DRB1*10:O1 a "susceptible" allele to WNV infection and the role of secretor status in relation to WNV infection.