INTRODUCTION

Blood group phenotypes have been reported to be associated with increased susceptibility to some infectious diseases [¹, ²]. Blood group B is associated with a lower risk of Hepatitis B virus (HBV) infection while blood group O is at higher risk of HBV infection in endemic areas [³, ⁴]. Blood group O is associated with disease due to Norwalk virus infection while blood group B shows resistance to the infection [⁵]. The P/Gb3 blood group antigen is correlated with susceptibility to Human Immunodeficiency Virus (HIV) infection [⁶]. Blood groups A and B were over represented among patients with influenza (H1A1) compared with groups O and AB [⁷], and the higher susceptibility of influenza A and B were observed with blood group AB [⁸, ⁹].

COVID-19 is a new infectious disease caused by the novel corona virus (SARS-CoV-2) that has spread rapidly around the world [¹⁰, ¹¹]. Corona viruses are RNA viruses, two-thirds of RNA genome encode viral polymerase (RNA dependent RNA polymerase), RNA synthesis materials, and two non-structural polyproteins that are not involved in host response modulation (ORF1a-ORF1b). The other one-third of the genome encodes four structural proteins spike (S), envelope (E), membrane (M) ve-nucleocapsid (N), and other helper proteins [¹², ¹³].

The first step of corona virus infection is the interaction between human cells with S Protein [¹³]. The receptor binding domain (RBD) in SARS-CoV-2 S protein attaches to Angiotensin-converting enzyme 2 (ACE2) as host receptor [¹⁴]. It needs to be cleaved in two sites called S protein priming by transmembrane protease serine 2 (TMPRSS2) to enter the cell and replicate [¹⁵, ¹⁶, ¹⁷].

It has been reported that severe acute respiratory syndrome coronavirus (SARS-CoV) is less likely to infect people with blood group O compared with other blood groups [¹⁸], and a recent study of ABO blood group susceptibility of infection by SARS-CoV-2 from three hospitals in China (Wuhan and Shenzhen) indicated individuals of blood group A are more susceptible to COVID-19 compared with non-A blood groups. In contrast, people of blood group O appeared to be less susceptible to the COVID-19 disease [¹⁹].

The aim of this study was to determine if there was an association between group A and COVID-19 among other countries similar to the data reported from China.

MATERIALS AND METHODS

The Worldmeter website (www.worldmeter.info) was used to collect data on COVID-19 cases for 105 countries on the 13th of April 2020. The website listed numbers of cases in five categories. Total cases were the figures reported for the total cumulative counts of laboratory confirmed infections. Active cases were the total number minus deaths and those who recovered. Recovery was defined based on various criteria: resolution of symptoms and two negative tests for the virus within 24 hours; resolution of symptoms with no recurrence within 14 days; discharge from hospital. Deaths were the cumulative number of confirmed cases. A subset of cases defined as critical was analysed separately [²⁰].

Data on ABO blood groups for the countries was obtained from (https://en.m.wikipedia.org/wiki/Blood_type_distribution_by_country). Data for these countries in six geographical regions are listed in Tables 1, 2, 3, 4, 5 and 6 [²¹].

As there were reports from three hospitals in China indicating individuals of blood group A are more susceptible to COVID-19 compared with non-A blood groups [¹⁹], this retrospective study was carried out to determine if there was a similar pattern in other populations. The statistical analyses were performed with SPSS software version 16.0. Spearman correlation coefficient was used for the relationship of ABO blood group with COVID-19 disease outcome and Kruskal Wallis test for non-parametric data to compare the distribution of ABO and COVID-19 disease outcome in the six geographical areas.

RESULTS

Tables 1, 2, 3, 4, 5 and 6 summarise the data on COVID-19 cases and ABO blood groups by geographical region. The mean percentage of each ABO blood group distribution by region included in the analyses. Latin America has a significantly higher proportion of blood group O (59.4%) compared with Africa (49%), Australasia (47%), USA and Canada (45%), Asia (40%), and Europe (40%) (X2 = 47.36, df = 5, p < 0.0001). The USA and Canada have a higher proportion of blood group A (42%) compared with Europe (41%), Australasia (37%), Africa (29%), Asia (29%) and Latin America (27%) (X2 = 57.19, df = 5, p < 0.0001). Asia has a significantly higher proportion of blood group B (24%). Compared with Africa (18%), Europe (14%), Latin America (11%), Australasia (12%) then USA and Canada (10%) (X2= 49.91, df = 5, p < 0.0001). Asia had the highest proportion of AB (7.1%) compared with Europe (6%), Africa (4%), Australasia (4%), USA and Canada (4%) and Latin America (3%) (X2 = 46.38, df = 5, p < 0.0001).

While analyses of the ABO blood group distribution indicated there were significant differences among the six regions, analysis of their association with COVID-19 in 105 countries indicated there were significant correlations between group A and infections (Table 7).

The analyses of the data for ABO blood group distributions and numbers of COVID-19 cases in 105 countries on the 13th of April 2020 indicated there was a significant positive correlation between blood group A and the total number of COVID-19 cases (rs = 0.329, p < 0.001). Groups B (rs = -0.252, p < 0.010) and O (rs= - 0.231, p < 0.018) were significantly negatively correlated with the total number of COVID-19 cases; there was no significant correlation with AB (rs = 0.089, p < 0.366).

There was a significant positive correlation between the active cases and group A (r s= 0.331, p < 0.001). In contrast, groups B (rs = -0.229, p < 0.019) and O (rs = -0.266, p < 0.006) were significantly negatively correlated. There was no significant correlation with AB (rs = 0.103, p< 0.298).

Critical cases, defined as requiring ICU facilities, were positively correlated with blood group A (rs= 0.365, p < 0.0001); however, they were negatively correlated with blood group B (rs= -0.357, p < 0.0001). There was no significant correlation with O (rs = -0.133, p < 0.176) or AB (rs = 0.003, p < 0.977).Deaths due to COVID-19 in 105 countries were negatively correlated with blood group B (rs = -0.242 p < 0.013), but not with the other groups: A, rs = 0.094 p < 0.338; O, rs = 0.105, p < 0.285; AB, rs = -0.142 p < 0.150. Among individuals who recovered, there were insignificant positive correlations with blood group O (rs = 0.089 p < 0.368) and B (rs = 0.041 p < 0.676) and insignificant negative correlations with blood group A (rs = - 0.090 p < 0.362) and AB (rs = -0.044 p < 0.657).

Table 8 summarizes the mean value of total and active cases for each geographical area. The USA and Canada have a significantly higher proprotion of total cases with the mean total cases (306430) compared with other areas: Europe (25528); Asia (9754); Latin America (3816); Australasia (1946); Africa (629) (total cases; X2 = 27.26, df = 5, p < 0.0001; active cases; X2 = 29.53, df = 5, p < 0.0001).The statistical analysis on the mean percentage of critical cases indicated that Europe has a significantly higher percentage compared with other areas, followed by the USA and Canada, Latin America, Asia, Australasia and Africa (X2 = 30.92, df = 5, p < 0.0001).

DISCUSSION

Individuals of blood group A were over represented among those infected with the new coronavirus SARS-CoV-2 compared with non-A blood group. There were significantly fewer individuals of blood group O among the infected patients. [¹⁹], and in another study the odds of COVID-19 positive versus negative test results were increased in blood groups A and decreased in blood groups O [²²]. The current study indicated that blood group A was correlated with infection by SARS-CoV-2 and more among those with severe COVID-19 disease. Blood groups B and O had negative correlations with infections due to SARS-CoV-2. Group B was negatively correlated with deaths from COVID-19 disease.

The USA, Canada and Europe have the highest numbers of SARS-CoV-2 infections compared with other regions. This observation might reflect in part the higher proportion of blood group A.

SARS-CoV-2 shares a highly similar gene sequence and behaviour pattern with SARS-CoV [²³]. Both interact with Angiotensin-converting enzyme2 (ACE2) to enter the host cell [¹⁴, ²⁴]. A previous study with both monoclonal and human natural anti-A antibodies blocked the interaction between the S protein and ACE2 receptor providing [²⁵]. Recent observation indicate that subjects with anti-A in serum are significantly less represented in the COVID-19 group than those lacking anti-A [²⁶]. If anti-A inhibits binding of the virus to its receptor, it might reduce density of colonisation.

These observations require further studies to define epidemiological factors linked with COVID-19 disease such as nutrition, age, population density, personal hygiene and strength of health care system which influence the spread of the disease [²⁷], assessing the association of ABO blood group among COVID-19 cases and disease severity with respect to ethnic groups in each country, while study in the effects of anti-A on interactions between the SARS-CoV-2 virus and host cells in vitro.

CONCLUSIONS

This study supports the previous observation and indicted that blood group A contributes to the risk of COVID-19 disease. Groups B and blood group O which have the anti-A isohemagglutinin were negatively associated with SARS-CoV-2 virus infection. More detailed studies of ABO groups among COVID-19 patients for each country and in vitro studies on the effects of Anti-A on interactions between the SARS-CoV-2 virus and host cells are needed.

ACKNOWLEDGEMENTS

We wish to express our sincere appreciation and thankful to Prof. Caroline Blackwell for a provided wise expertise advice that greatly assisted this paper, significantly improving the manuscript.