摘要: 目的：基于菏泽市立医院输血不良反应院内监测数据，系统分析2024~2025年院内输血不良反应的类型构成、发生时间分布及血液成分关联特征，在统一可比口径下测算各类制品的不良反应粗发生率，为临床输血安全精细化管控提供数据支撑。方法：采用回顾性描述性研究方法，纳入本院2024年1月~2025年12月上报的154例次输血不良反应完整记录，提取反应类型、发生时机、输注血液成分、患者性别年龄、输血/妊娠/不良反应既往史、不规则抗体筛查结果等核心变量。依据名称关键词将所有血液制品统一归并为红细胞类、血小板类、血浆类、冷沉淀类4个大类，匹配同期月度输血量(血袋数)数据，测算每1万血袋的不良反应粗发生率及Poisson 95%置信区间(CI)。采用Pearson卡方检验分析反应类型与血液成分大类的关联性，报告
χ2、自由度(df)、P值与Cramér’s V效应量；2 × 2列联表根据数据特征采用Fisher确切检验，报告比值比(OR)及95% CI。结果：154例次输血不良反应中，过敏性输血反应128例次(83.1%)，非溶血性发热性输血反应26例次(16.9%)；发生时机以输血过程中为主，共104例次(67.5%)。患者年龄中位数55岁(四分位间距IQR 29~69岁)，男性占比51.9%。不良反应的血液制品来源以血小板类(50.6%)与红细胞类(31.8%)为主。反应类型与血液成分大类存在显著关联性(
χ2 = 34.61, df = 3, P = 1.47 × 10
−7, V = 0.47)，其中非溶血性发热反应主要集中于红细胞类制品(21/26)。全院输血不良反应总体粗发生率为20.00/万血袋(95% CI: 16.97~23.42)；分品类来看，血小板类113.13/万血袋(95% CI: 89.42~141.19)，红细胞类18.28/万血袋(95% CI: 13.53~24.17)，血浆类10.09/万血袋(95% CI: 6.46~15.01)，冷沉淀类1.54/万血袋(95% CI: 0.32~4.49)。在已发生不良反应的病例中，抗体筛查阳性者更易表现为发热反应(OR = 12.50，95% CI：2.89~54.04；Fisher确切检验P = 0.0008)，该结果为探索性线索。结论：本院监测周期内，输血不良反应以过敏性反应为主，且多数发生于输血操作过程中；不同品类血液制品的不良反应类型构成存在显著差异，血小板类制品的不良反应粗发生率最高。受限于制品命名口径差异与患者层级关联数据缺失，本研究测算的发生率均为描述性粗率，后续需基于统一命名规范与事件级联数据，进一步开展不良反应风险因素与防控干预效果研究。本研究结果可为院内输血安全分层质控、重点环节风险防控提供基础数据依据。
Abstract: Objective: To systematically analyze the composition, onset timing distribution, and blood component-related characteristics of transfusion adverse reactions (TARs) in Heze Municipal Hospital from 2024 to 2025 based on in-hospital surveillance data, and to estimate the crude incidence of TARs under a unified and comparable caliber, so as to provide data support for the refined management of clinical transfusion safety. Methods: A retrospective descriptive study was conducted, and a total of 154 complete TAR episodes reported in the hospital from January 2024 to December 2025 were included. Core variables were extracted, including reaction type, onset timing, transfused blood component, patient’s sex and age, past history of transfusion/pregnancy/adverse reaction, and irregular antibody screening results. All blood products were uniformly classified into 4 categories (red blood cell, platelet, plasma, cryoprecipitate) according to the keywords in product names. By matching the monthly transfusion volume (bag count) data in the same period, the crude incidence of TARs per 10,000 bags and the Poisson 95% confidence interval (CI) were calculated. Pearson’s chi-square test was used to analyze the correlation between reaction type and blood component category, and χ2, degree of freedom (df), P value and Cramér’s V effect size were reported. Fisher’s exact test was used for 2 × 2 contingency tables according to data characteristics, and odds ratio (OR) and 95% CI were reported. Results: Among the 154 TAR episodes, 128 (83.1%) were allergic transfusion reactions, and 26 (16.9%) were non-hemolytic febrile transfusion reactions. Most reactions occurred during transfusion, accounting for 104 episodes (67.5%). The median age of patients was 55 years (interquartile range, IQR 29-69 years), and 51.9% were male. The blood components related to TARs were mainly platelet products (50.6%) and red blood cell products (31.8%). There was a significant correlation between reaction type and blood component category (χ² = 34.61, df = 3, P = 1.47 × 10−7, V = 0.47). Among them, non-hemolytic febrile reactions were mainly concentrated in red blood cell products (21/26). The overall crude incidence of TARs in the hospital was 20.00 per 10,000 bags (95% CI: 16.97~23.42). By component category, the crude incidence was 113.13 per 10,000 bags for platelets (95% CI: 89.42~141.19), 18.28 per 10,000 bags for red blood cells (95% CI: 13.53~24.17), 10.09 per 10,000 bags for plasma (95% CI: 6.46~15.01), and 1.54 per 10,000 bags for cryoprecipitate (95% CI: 0.32~4.49). Among the cases with TARs, patients with positive antibody screening were more likely to present febrile reactions (OR = 12.50, 95% CI: 2.89~54.04; Fisher’s exact test P = 0.0008), which was an exploratory finding. Conclusion: During the surveillance period in our hospital, TARs were dominated by allergic reactions, and most occurred during the transfusion procedure. There were significant differences in the composition of reaction types among different categories of blood products, and the crude incidence of TARs was the highest in platelet products. Limited by the differences in product naming caliber and the lack of patient-level linked data, the incidence calculated in this study is all descriptive crude rates. In the follow-up study, further research on risk factors of TARs and the effect of prevention and control interventions should be carried out based on unified naming specifications and event-level cascade data. The results of this study can provide basic data basis for hierarchical quality control of in-hospital transfusion safety and risk prevention and control of key links.