Objective
The ABO system modulates the inflammatory response, and it has been involved in SARS-CoV-2 infection. There is increasing evidence of underlying immune-inflammatory mechanisms in post-COVID-19 syndrome (PCS). Blood group O seems to protect against COVID-19 infection, but there is no data on the relationship between blood group O and PCS. This study aimed to assess this potential association.
Subjects and methods
A case-control study including subjects who had suffered from mild COVID-19 in a community setting was designed. Cases were PCS+ patients, controls were PCS- subjects and the exposure variable was blood group O. Baseline epidemiological data (age, sex, BMI, smoking, comorbidities), and clinical and laboratory parameters (inflammatory markers and IgG anti-N antibodies) 3 months after the acute episode, were obtained. Five composite indices of inflammation were built, combining the upper ranges of the distributions of inflammatory markers. Blood group and Rh factor were obtained from the patient's medical history or capillary blood samples. Bivariate and multivariate analyses were performed.
Results
One-hundred and twenty-one subjects were analyzed (56.2% women). The mean age was 45.7 (16) years (range, 18-88 years). Blood group frequencies were 43.3% (group A), 7.7% (group B), 5.7% (group AB), and 43.3% (group O). Thirty-six patients were classified as PCS+ (25 women, 11 men; p=0.07). The most frequent symptom was fatigue (42.8%). There were no significant differences between PCS+ and PCS- subjects regarding age, BMI, smoking, or previous comorbidity. The prevalence of PCS was 43.2% (19/44) in the blood group O and 23.7% in non-O subjects (14/60) (p=0.036). The mean number of PCS symptoms was 0.82 (1) in the blood group O and 0.43 (0.9) in the non-O group (p=0.017). There were no significant differences between A, B, and AB groups (PCS+ and PCS-) regarding inflammatory markers. In contrast, blood group O PCS+ patients had significantly higher lymphocyte count, plasma CRP, fibrinogen levels, and higher percentages of C2, C3, and C4 composite indices than PCS- subjects. The blood group O had an increased risk of developing PCS compared to non-O subjects (adjusted OR=4.20 [95%CI, 1.2-14]; p=0.023). The variables that contributed the most to the predictive model were blood group O, lymphocyte count, neutrophil count, and female sex. R-squared was 0.46, and the area under the ROC curve was 0.807 [95% CI, 0.70-0.90] (p=0.0001).
Conclusion
An increased risk of PCS associated with blood group O has been observed. Slightly, albeit significant, raised levels of fibrinogen, CRP, and neutrophil and lymphocyte counts, not observed in the other ABO blood groups, have been demonstrated in blood group O. Further investigations are needed to confirm these results.