Marín Mercedes, Muñoz Patricia, Sánchez Mónica, Del Rosal Marina, Alcalá Luis, Rodríguez-Créixems Marta, Bouza Emilio
Servicio de Microbiología Clínica y Enfermedades Infecciosas del Hospital General Universitario "Gregorio Marañón," Universidad Complutense de Madrid, Madrid, Spain.
Medicine (Baltimore). 2007 Jul;86(4):195-202. doi: 10.1097/MD.0b013e31811f44ec.
Traditionally, infective endocarditis (IE) has been microbiologically diagnosed by blood cultures or serology. However, conventional microbiologic methods do not always provide an etiologic diagnosis. We conducted the current study to evaluate the usefulness of a universal real-time polymerase chain reaction (PCR) of the 16S rRNA gene followed by sequencing for the diagnosis of IE in explanted heart valve tissue (HV) as part of the routine of a clinical microbiology laboratory, and to compare it with conventional culture of blood or HV. We prospectively analyzed 177 HV samples by universal PCR and sequencing: 48 were from 35 patients with definite IE and 129 were from 120 patients without IE. Specific PCR tests were used when necessary to confirm broad-range PCR results. For the 35 patients with IE, all of the HV samples except for 2 from the same patient gave positive PCR results. The microorganisms identified matched those isolated by blood culture in 31 cases. The other 3 patients had negative blood culture IE, but PCR made possible the detection of Tropheryma whipplei, Bartonella quintana, and Streptococcus gallolyticus. For the negative control group, universal PCR was completely negative in 123 of the 129 samples. Sensitivity, specificity, and negative and positive predictive values of this real-time PCR method were 96%, 95.3%, 98.4%, and 88.5%, respectively, for the diagnosis of IE, using the Duke criteria to define IE and using blood culture results to identify etiologic microorganisms. Conventional HV culture correlated poorly with blood cultures and molecular techniques, and frequently represented tissue contamination resulting from valve handling. Our universal PCR method has proved to be more sensitive, specific, and rapid than conventional culture methods, and should therefore be included as a new major Duke criterion for the diagnosis of IE. According to the results of the current study, this technique should be used to supplement blood and HV culture. Conventional HV cultures are frequently responsible for false-positive and false-negative results, and are not always useful to establish the etiology of IE.
传统上，感染性心内膜炎（IE）通过血培养或血清学进行微生物学诊断。然而，传统的微生物学方法并不总能提供病因诊断。我们开展了本研究，以评估16S rRNA基因通用实时聚合酶链反应（PCR）随后测序在诊断植入心脏瓣膜组织（HV）中的IE的实用性，作为临床微生物学实验室常规工作的一部分，并将其与血液或HV的传统培养进行比较。我们对177份HV样本进行了前瞻性分析，采用通用PCR和测序：48份来自35例确诊IE患者，129份来自120例无IE患者。必要时使用特异性PCR检测来确认广谱PCR结果。对于35例IE患者，除同一患者的2份样本外，所有HV样本的PCR结果均为阳性。鉴定出的微生物在31例中与血培养分离出的微生物相匹配。另外3例患者血培养IE为阴性，但PCR使得检测到了惠普尔嗜组织细胞菌、五日热巴尔通体和解脲链球菌成为可能。对于阴性对照组，129份样本中的123份通用PCR结果完全为阴性。使用杜克标准定义IE并使用血培养结果鉴定病原微生物，这种实时PCR方法诊断IE的敏感性、特异性、阴性和阳性预测值分别为96%、95.3%、98.4%和88.5%。传统的HV培养与血培养和分子技术相关性较差，且常常代表瓣膜处理导致的组织污染。我们的通用PCR方法已被证明比传统培养方法更敏感、特异和快速，因此应作为诊断IE的一项新的主要杜克标准纳入。根据本研究结果，该技术应用于补充血液和HV培养。传统的HV培养常常导致假阳性和假阴性结果，对于确定IE的病因并不总是有用。
