Novak Robert, Powell Keith, Christopher Norman
Department of Pediatrics, Akron Children's Hospital, One Perkins Square, Akron, OH 44308, USA.
Pediatr Dev Pathol. 2004 May-Jun;7(3):226-30. doi: 10.1007/s10024-003-8091-2. Epub 2004 Mar 17.
Urinary tract infection (UTI) is a major concern in young febrile children. Current recommendations favor use of urine obtained by bladder catheterization or aspiration, but opinion varies as to the best ancillary tests to predict a positive culture and guide initial management. The utility of dipstick leukocyte esterase, blood, and nitrite; unspun urine leukocyte count; gram stain of cytocentrifuged urine; and standard spun sediment examination was evaluated in 142 febrile, < 5-year-old children seen in the Emergency Department, 25 of whom had culture-proven UTI. Using sensitivity and negative predictive values as criteria for performance, unspun urine leukocyte count and gram stain of cytocentrifuged urine used in parallel was the best approach but still failed to detect some UTI. Analysis of the nature of the specimens evaluated provided explanation of differences from previous observations. Adoption of this modified approach to prediction of urinary tract infection appears to improve prediction but has operational implications and creates potential problems for the laboratory.
尿路感染（UTI）是发热幼儿的一个主要问题。目前的建议倾向于使用通过膀胱导尿或穿刺获取的尿液，但对于预测培养阳性并指导初始治疗的最佳辅助检查，意见不一。在142名急诊就诊的5岁以下发热儿童中，评估了尿试纸白细胞酯酶、血液和亚硝酸盐；未离心尿液白细胞计数；细胞离心尿液革兰氏染色；以及标准离心沉淀物检查的效用，其中25名儿童经培养证实患有UTI。以敏感性和阴性预测值作为性能标准，并行使用未离心尿液白细胞计数和细胞离心尿液革兰氏染色是最佳方法，但仍未能检测出一些UTI。对所评估标本性质的分析解释了与先前观察结果的差异。采用这种改良方法预测尿路感染似乎可以改善预测，但具有操作上的影响，并给实验室带来潜在问题。
