【Objective】 To investigate the factors influencing the production of platelet antibody and its effect on clinical platelet transfusion. 【Methods】 This is a single-center prospective observational study. The research subjects were patients with hematological diseases in West China Hospital of Sichuan University from October 1, 2018 to September 30, 2019, and their plasma were collected before platelet transfusion to detect platelet antibodies using solid-phase agglutination method. According to the results of platelet antibody screening, the patients were divided into platelet antibody positive group and negative group. The t test and nonparametric Mann-Whitney U test were used to compare the transfusion efficacy of two groups. Patients’ demographic and clinical information, including age, gender, diagnosis, the units of platelets and RBC transfused, were collected via HIS6.2.0 and whole process management system of blood in clinical (version 3.0) to analyze the influence of age, gender and the disease on the positive rate of platelet antibodies, as well as the profile of platelet antibodies in patients with different diseases, the correlation between the positive rate of platelet antibodies and the history of blood/platelets transfusion. In additional, the platelet transfusion process was observed on site. 【Results】 A total of 316 patients with hematologic diseases were included in this study, mainly with acute myeloid leukemia(188/316, 59.5%). All patients were transfused 1671 U platelet [1~17(5.3±3.1)U each person] and 1896 U RBC products [0~38(7.8±4.6)U each person] during the treatment. Out of the 316 patients, platelet antibodies were found in 85 (26.9%) of them. No significant differences in the positive rates of platelet antibody after transfusion were notice by genders or ages(P>0.05). The incidence of platelet antibody was related to diseases (P<0.05), with MDS as the highest (57.1%), followed by aplastic anemia (36.4%) and myeloid leukemia (27.7%). In additional, the positive rate of platelet antibody increased with the number of previous platelet transfusions(P<0.05). The 316 patients were divided into positive group and negative group according to the results of platelet antibody screening. The corrected count of increment (5.2×10⁹/L vs 11.5×10⁹/L, P<0.01) and absolute platelet increase(8×10⁹/L vs 17×10⁹/L, P<0.01)in positive group were lower than those in negative group. The positive group were transfused more units of platelets(1.7 U vs 1.2 U, P<0.01)and red blood cells(1.5 U vs 1.1 U, P<0.05)per week than negative group. The platelet transfusion interval was shorter in positive group than negative group (3.1 days vs 3.6 days, P<0.05), but there was no significant difference in red blood cell transfusion interval (3.1 days vs 3.8 days, P>0.05) between two groups. The minimum PLT count(5×10⁹/L vs 9×10⁹/L, P<0.01), average PLT count(27×10⁹/L vs 40×10⁹/L, P<0.01)and average Hb(71 g/L vs 77 g/L, P<0.05)in positive group were lower than those in negative group during hospitalization, but there was no significant difference in the minimum Hb(56 g/L vs 59 g/L, P>0.05)between two groups. According to transfusion events on site, the incidence of acute adverse reactions to transfusion was 13% (169/1 291). 【Conclusions】 The positive rates of platelet antibodies in patients with hematologic diseases were relatively high. In addition, the efficacy of platelet transfusion in positive group were worse than that in the negative group. It is recommended that platelet antibody testing should be routinely performed before transfusion in hematologic disease patients to select crossmatch-compatible platelets in order to improve the effectiveness of platelet transfusion.

【Objective】 To retrospectively analyze the situation of surgical blood ordering in our hospital and explore the value of optimizing preoperative blood ordering. 【Methods】 Surgical blood ordering and utilization data of West China Hospital of Sichuan University from 2012 to 2018 were gathered to evaluate the rationality of preoperative blood ordering by calculating the indicators including transfusion rate, transfusion probability, transfusion index etc. and recommend preoperative blood ordering guided by transfusion index ≥ 0.3, the transfusion rate ≥ 5%, and the transfusion index ≥ 0.5 respectively to calculate the cost saved. 【Results】 1) The preoperative blood ordering of Department of Cardiac Surgery and Burn Plastic Surgery were relatively rational, while other Surgery Departments was excessive, especially the Thoracic Surgery; 2) Among the top fifteen surgeries ranked by blood ordering rate, the blood ordering was rational for mitral valve replacement, ventricular septum (repair/occlusion), and aortic valve replacement, while excessive for other 12 surgeries, especially for lung resection surgery; 3) The surgical blood ordering guided by the three indicators can reduce 19% ~80% theoretically, saving 0.39~1.28 million yuan per year. 【Conclusion】 Preoperative blood ordering of the Department of Cardiac Surgery and Burn Plastic Surgery in our hospital is relatively rational. While excessive blood ordering exists in other surgical departments, especially for thoracic surgery. The establishment of Maximum Surgical Blood Order Schedule can reduce unnecessary blood ordering and improve blood utilization, and save manpower and material resources, and reduce the costs of patients.

【Objective】 To develop a prediction model of allogenic blood transfusion in elective patients based on machine learning, so as to guide clinicians to prepare blood for perioperative patients more reasonably. 【Methods】 Relevant data of all surgical patients from 2012 to 2018 were extracted from the big data integration platform of our hospital, to construct the surgical blood database based on Python V3.8.0. All data were analyzed using Excel and SAS, and the prediction model was developed based on SPSS Modeler 18.0. 【Results】 1) There was a negative correlation between preoperative Hb and BMI and intraoperative blood transfusion rate, with Pearson correlation coefficient (R) as -0.168 and -0.046, respectively. The transfusion rate of patients under 1 year old was the highest, up to 15.63%. The transfusion rate of female patients was higher than that of male patients (P>0.05), as cardiac surgery rated at the highest 11.38%, but their per capita blood transfusion was lower than that of males (P<0.01). 2) The AUC range corresponding to the prediction model for transfusion probability was 0.67~0.88, and when the AUC reached the highest, the hit ratio, coverage rate and specificity of Model 9 was 10.7%, 85.76% and 75.4%, respectively. 3) The main factors contributing to the prediction model for transfusion volume in surgery were weight, Hb, total protein(TP), etc. 【Conclusion】 The prediction efficiency of the successfully constructed prediction model for perioperative blood use was better than that of MSBOS.