Brigham and Women's Hospital Neurosurgery Jamaica Plain, Massachusetts, United States
Background/Case Studies: There is a shortage of 114 million units of blood in low- and middle-income country (LMIC) blood banks every year, which is exacerbated in non-urban settings. This contributes to preventable patient outcomes because blood transfusion is unavailable to treat exsanguinating or severely anemic patients. An emerging strategy to address this dilemma is a walking blood bank (WBB), a strategy where blood donors are urgently summoned in times of dire transfusion need and blood is immediately collected, tested, and transfused to prevent near-certain morbidity. This approach depends on an accurate point-of-care rapid diagnostic testing (RDT) for transfusion transmissible infections (TTIs). We sought to determine the level of agreement of RDTs with Standard of Care (SOC) testing and quantify the levels of detection of RDTs.
Study
Design/Methods: This prospective cohort study occurred between August 2022 and March 2023 at a Hospital in Turkana County, a rural region of Kenya. The performance of locally available RDTs for TTIs on blood donor specimens was compared pairwise to the SOC testing conducted by the Kenyan National Blood Transfusion Service (KNBTS). We calculated the RDT sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV). Serial dilutions of known positive specimens were tested by RDTs to ascertain the level of detection.
Results/Findings: A total of 686 blood donors underwent routine blood donation where SOC and RDT tests were compared. The prevalence of TTIs was low (4.5%) consisting of 0.7% Human Immunodeficiency Virus (HIV), 0.9% Hepatitis C Virus (HCV), 2.2% Hepatitis B Virus (HBV), and 1.9% syphilis. The level of agreement (see Table) was high for RDTs with the PPV >87% (87.5-100.0%) and NPV >99% (range 99.7-100.0%). The sensitivity and specific values varied, due in part to the low frequency of positive test results. Several serial dilution tests were conducted to identify the maximum dilution levels to detect the presence of TTIs in RDTs as well as minimum dilution levels to avoid false positive tests. The level of detection dilutions ranged from 1:2 (HCV) to 1:256 (HIV). Conclusions: This study demonstrated that RDTs for TTI demonstrated excellent PPV and NPV for SOC TTIs. The limited sensitivity and specificity detected does not preclude acceptable accuracy given the low prevalence of TTIs in the study population. These data inform the safety of a WBB using RDTs compared to the SOC testing for TTIs in the study setting.
Importance of research: There is an evidence gap surrounding the performance of Rapid Diagnostic Tests (RDTs) to detect transfusion transmissible infection in a Walking Blood Bank model.
