目的：通过对多所医院患者输血情况的调查和分析，为不同病种患者的输血评估提供循证医学依据，实现精准输血。方法：调研分析2021年7月—12月上海用血量最多的2所综合性三级医院和3所综合性二级医院共16 254例输血患者的病例资料，调查分析患者血型、科室、病种、输血量、输血成分等。结果：悬浮红细胞输注量在三级医院总量为37 813 U，人均输注量为2（2，4） U，二级医院总量为45 577 U，人均输注量为4（2，8） U，两者差异具有统计学意义（P<0.05）；冰冻血浆输注量在三级医院总量为33 689 U，人均输注量为2（0，4） U，二级医院总量为26 822 U，人均输注量为0（0，4） U，两者差异具有统计学意义（P<0.05）；单采血小板输注量在三级医院总量为3 586 U，人均输注量为0（0，0） U，二级医院总量为23 414 U，人均输注量为0（0，5） U，两者差异具有统计学意义（P<0.05）。血液科患者的悬浮红细胞和单采血小板输注量最高，分别为6（2，11） U和5（1，9） U，冰冻血浆输注量以神经内科患者[4（0，32） U]最高。结论：不同规模医院不同种类的血液制品在不同科室的输注量存在较大差异。各医院应根据自身医院和不同科室疾病患者的情况，以循证医学为依据，制定不同患者的输血评估方案，做到精准输血，保证临床科学合理用血。

Objective To provide evidence-based medical basis for evaluation of blood transfusion in patients with different diseases and achieve precise blood transfusion through the investigation and analysis of blood transfusion situation in several hospitals. Methods The data of blood transfusion patient in two general tertiary hospitals and three general secondary hospitals in Shanghai using the largest amount of blood from July to December 2021 was collected, and the information including age, blood type, department, disease, blood transfusion volume and blood transfusion components of patients were analyzed. Results The total red blood cell infusion was 37 813 U [median is 2(2, 4) U]in tertiary hospitals and 45 577 U [median is 4(2, 8) U] in secondary hospitals, and the difference was statistically significant (P<0.05). The total plasma infusion was 33 689 U [median was 2(0, 4) U] in tertiary hospitals and 26 822 U [median was 0(0, 4)U] in secondary hospitals., and there was significant difference between the two groups (P<0.05). The total platelet transfusion volume was 3 586 U [median was 0(0, 0) U] in the tertiary hospitals and 23 414 U [median was 0(0, 5) U] in the secondary hospitals, and the results showed statistically different (P<0.05). The per capita infusion volume of suspended red blood cells and platelets was the highest in patients with hematology diseases(respectively 6(2,11) U and 5(1, 9) U), while the per capita plasma infusion volume was the highest in patients with neurological diseases [4(0, 32) U]. Conclusions There is large variation in the transfusion volumes of different types of blood components in different departments in hospitals of different size. Each hospital should make blood transfusion evaluation plan based on evidence-based medicine according to its own conditions and the patient disease, so as to achieve precise blood transfusion and ensure scientific and rational use of blood in clinic.