Development and validation of a scoring system to predict vasovagal reaction upon whole-blood donation
Tatsuma Hashizume,Gaku Kondo,Fumihiko Ishimaru,Yoko Ikeda,Keiko Kagawa,Noriko Kunii,Noriko Namba,Kiichi Aoki,Yoshihiro Sawamura,Shigeyoshi Makino
DOI: https://doi.org/10.1111/vox.13579
Abstract:Background and objectives: Risk factors for vasovagal reaction (VVR) have been extensively studied. With knowledge of the relative importance of these risk factors for VVR, collection staff could take care of blood donors from the same standpoint, leading to improved donor safety. We therefore developed a scoring system to predict VVR, which incorporates registration information. Materials and methods: Pre-syncopal and syncopal symptoms, as well as on- and off-site reactions, are included in this analysis as VVR. We defined the donor status as follows: first-time donors, repeat donors with no history of reaction and repeat donors with a history of reaction. We prepared two datasets: whole-blood donations at a blood donation site in Tokyo between January 2019 and December 2019 were included in training data (n = 361,114), and whole-blood donations between January 2020 and August 2020 were included in testing data (n = 216,211). Results: The most important variable was the donor status, followed by age, estimated blood volume and height. We integrated them into a scoring system. Training and testing datasets were combined (n = 577,325), and VVR rates in groups with scores of 0, 1, 2, 3, 4 and 5 or more were 0.09% (95% CI: 0.081%-0.10%), 0.33% (95% CI: 0.31%-0.36%), 0.87% (95% CI: 0.78%-0.96%), 1.17% (95% CI: 1.05%-1.30%), 2.15% (95% CI: 1.98%-2.32%) and 3.11% (95% CI: 2.90%-3.34%), respectively. Conclusion: The scoring system enables staff to significantly predict VVR and may help them to identify donors at increased risk of experiencing syncope, thereby mitigating the negative impact of VVR on donor safety and return by paying close attention to high-risk donors.