The advancement of personalized medicine and the use of precise matching to a patient with specific antigen profile, particularly for chronically transfused populations, such as those with sickle cell disease and thalassemia, has increased the request for rare blood.

The objective of this study is to present data aimed at diagnosing patients and donors with rare blood types as a basis for the development of a computer system enabling risk management related to rare blood.

A cross-sectional study analyzing data from 3,647 patients with hemoglobinopathies at the Rio de Janeiro Blood Center, Brazil, was conducted as a partial diagnosis to identify patients with rare phenotypes. The institution's rare donor registry was also accessed to assess blood availability for these patients.

The absence of multiple common antigens and other rare phenotypes such as Kp(b-), Jk(a-b-), Lu(b-), K+k-, U-, Js(b)-, Hy-, Uvar Jo(a-) were identified in these patients. Rare phenotypes resulting from RH variants in sickle cell patients were also identified, such as DAR, DIIIa, DAU5, hrB-. In twenty (32.3%) patients, 28 antibodies were identified. The mean red blood cell transfusion was 14 units. One hundred sixteen donors with phenotypic profiles corresponding to these patients were identified. However, we found that there are more patients with rare phenotypes than blood donors and that the number of compatible donors is insufficient to meet the transfusion need of the patients.

Based on the assistential diagnosis of compatible transfusions conducted in these patients with hemoglobinopathies, which demonstrates the shortage of donors with rare phenotypes for performing transfusions with precise matching, we propose a modeling for the development of a computer system that allows for better identification and monitoring of patients and donors with rare phenotypes, preventing associated complications, and enhancing blood transfusion processes in the most optimized way that best suits the reality of blood centers.