From the Centers for Disease Control
From the Centers for Disease Control
复制标题
来自疾病控制中心
DOI：
10.1017/s0899823x00087821
复制
发表时间：
1992
期刊：
影响因子：
--
通讯作者：
E. Larson
中科院分区：
文献类型：
--
作者：
E. Larson
1991年6月27日至7月30日，俄亥俄州一家医院发生了4起血小板池细菌污染事件，医院通过食品药品监督管理局（FDA）向疾病控制中心（CDC）报告。本报告总结了这些事件的流行病学调查结果。接受血小板池的4例患者中有3例因恶性肿瘤住院，并在开始输血后1 - 40分钟内发生寒战和寒战;其中1例发生低血压。其中两次发作后，医院开始输血前革兰氏染色和培养所有血小板，但没有延迟血小板输注等待革兰氏染色或培养结果。这一增加的监测检测到第四个污染的血小板池，但它已经被输注给了一个血液透析患者与骨髓增生异常相关的特发性血小板减少症;该患者发展为无症状菌血症。FDA和CDC发起了一项调查，试图检测这些血小板池污染的来源和风险因素。受污染的血小板池定义为细菌培养阳性的血小板池，经革兰氏染色阳性、作为血小板池一部分的单个单位中的阳性培养物或6月27日至7月31日从输血受者处获得的血液中的阳性培养物确认。四个被污染的血小板池，每个来自五个单独的血小板单位，通过革兰氏染色和蜡样芽孢杆菌培养呈阳性（两个血小板池）和表皮葡萄球菌和铜绿假单胞菌（各一个血小板池）;血小板池中的菌落计数范围为10 - 108个菌落形成单位（CFU）/ml，并且在构成库的20个个体血小板单位中的11个中为10至1010 CFU/ml。平均而言，在输血前2.5小时（范围= 15分钟-3.2小时）合并构成污染血小板池的单个血小板单位。研究人员检查了四个受污染的血小板池的数据，以确定其与采血师、采血地点或日期、成分分离技术人员、合并技术人员、合并日期和时间以及血小板年龄的相关性。然而，任何两个库中的单位都没有这些因素。从6月27日到7月30日，医院血小板池的细菌污染率（1，063个血小板池中的4个[0.4%]输血）显著高于前21个月（7，350个血小板池中的1个[0.01%]输血，p = 0.001）。在1991年6月27日之前的21个月内（当时医院制定了更严格的血小板输注反应检测和报告指南），医院的护理人员向医院血库报告的血小板输注反应越来越多（其中99%导致了相关血小板池的培养）。按月报告的反应发生率从1989年9月的每1,000个血小板池输注2.3例增加到1991年7月的每1,000个血小板池输注72.4例（9<0.001，线性趋势卡方）。医院继续输血前革兰氏染色和培养所有血小板，现在要求在释放血小板进行输血前进行阴性革兰氏染色。
From June 27 through July 30, 1991, four episodes of bacterial contamination of platelet pools occurred in an Ohio hospital and were reported by the hospital through the Food and Drug Administration (FDA) to the Centers for Disease Control (CDC). This report summarizes the results of the epidemiologic investigation of these episodes. Three of the four patients who received the platelet pools were hospitalized for malignancies and developed chills and rigors within one to 40 minutes after initiation of the transfusion; one of the three developed hypotension. After two of these episodes, the hospital began pretransfusion gram staining and culturing of all platelets but did not delay platelet transfusion awaiting gram stain or culture results. This increased surveillance detected a fourth contaminated platelet pool, but it had already been transfused to a hemodialysis patient with idiopathic thrombocytopenia associated with myelodysplasia; this patient developed asymptomatic bacteremia. The FDA and CDC initiated an investigation to attempt to detect the source of and risk factors for contamination of these platelet pools. A contaminated platelet pool was defined as a platelet pool with a positive bacterial culture confirmed by either a positive gram stain of the pool, a positive culture in an individual unit that was part of the platelet pool, or a positive culture from blood obtained from the transfusion recipient from June 27 through July 31. The four contaminated platelet pools, each derived from five individual platelet units, were positive by gram stain and culture for Bacillus cereus (two platelet pools) and Staphylococcus epidermidis and Pseudomonas aeruginosa (one platelet pool each); colony counts ranged from 10” to lo8 colony-forming units (CFU) per ml in the platelet pools, and from 10 to lOlo CFU/ml among 11 of the 20 individual platelet units constituting the pools. On average, the individual platelet units constituting the contaminated platelet pools were pooled 2.5 hours (range = 15 minutes-3.2 hours) before transfusion. Investigators examined data on the four contaminated platelet pools for associations with the phlebotomists, bloodmobile sites or dates, component separation technicians, pooling technician, pooling date and time, and platelet age. However, no units from any two pools shared any of these factors. From June 27 through July 30, the rate of bacterial contamination of platelet pools at the hospital (4 [0.4%] of 1,063 platelet pools transfused) was significantly greater than for the previous 21 months (1 [O.Ol%] of 7,350 platelet pools transfused, p = .OOl). During the 21 months before June 27,199l (when the hospital instituted more stringent guidelines for detection and reporting of platelet-transfusion reactions), the nursing staff at the hospital increasingly reported platelet-transfusion reactions to the hospital blood bank (99% of which resulted in culturing of the involved platelet pool). The rate of reported reactions, by month, increased from 2.3 per 1,000 platelet pools transfused during September 1989 to 72.4 reported reactions per 1,000 platelet pools transfused during July 1991 (9<.001, chi square for linear trend). The hospital continues pretransfusion gram staining and culturing of all platelets and now requires a negative gram stain before releasing the platelets for transfusion.