自动化尿液干化学和有形成分分析复检规则的制定和应用

Development and clinical application of microscopy review criteria of automated urine chemistry and sedimental analysis

目的 结合尿液干化学分析（简称干化学）和有形成分分析（简称尿流式）,制定自动化尿常规分析的复检规则.方法 收集北京协和医院2008年11月至2010年10月1 714份中段尿标本,其中1 300份用于建立复检规则,214份用于验证所建立的规则,另外200份健康体检者的标本用于建立UF-1000i尿流式分析仪的参考范围.所有标本通过Siemens Bayer Clinitek 500型干化学分析仪、Sysmex UF-1000i尿流式分析仪及尿沉渣显微镜（简称尿沉渣）检测,进行RBC、WBC、PRO及CAST分析.针对自动化尿常规的不同检测方法,设计了4种复检方案：（1）方案1：单独使用干化学时RBC、WBC和PRO任何一项参数出现阳性者;（2）方案2：单独使用尿流式时RBC、WBC和CAST任何一项参数超出参考范围上限者;（3）方案3：联合使用2种方法检出的WBC结果不相符合,即同一项参数在一台仪器上检测为阳性或超出参考范围上限,在另一台仪器上检测为阴性或不超出参考范围上限,而RBC、PRO（对应尿流式的CAST）中任何一项参数在任何一台仪器上出现阳性或超出参考范围上限者;（4）方案4：联合使用2种方法检出的RBC、WBC、PRO（对应尿流式的CAST）结果不相符合者.通过Sysmex Laboman UriAccess Ver 3.0软件进行统计分析,建立不同检测方法的自动化尿常规分析的复检规则.结果 UF-1000i尿流式分析仪的参考范围：RBC为0～7.5个/μl（男）、0～15.9个/μl（女）;WBC为0～11.6个/μl（男）、0～12.7个/μl（女）;上皮细胞为0～6.5个/μl（男）、0～21.4个/μl（女）;CAST为0～1.3个/μl.尿沉渣显微镜镜检结果显示,1 300份用于建立复检规则的标本中,阳性标本占47.46%（617/1 300）,阴性标本占52.54%（683/1 300）.在阳性标本中,RBC阳性者最多,占60.13%（371/617）,其次为CAST阳性者,占8.43%（52/617）.4种方案的假阴性率（漏诊率）分别为8.38%（109/1 300）、4.69%（61/1 300）、0.62%（8/1 300）和0.54%（7/1 300）;复检率分别为47.85%（622/1 300）、59.38%（772/1 300）、72.85%（947/1 300）及52.23%（679/1 300）;方案4的漏诊病例中无1例血肌酐结果异常.采用214份临床标本对所建立的复检规则（方案4）进行验证,其漏诊率为0,复检率为53.74%（115/214）.结论 方案4的漏诊率最低,复检率也较低,且无严重肾功能异常者漏诊,是理想的复检方案.

Objective To integrate urine strip chemistry analysis with urine sedimental analysis and set up the criteria for urine microscopy review following automated urine analysis.Methods A total of 1 714 urine samples were collected from Peking Union Medical College Hospital from November 2008 to October 2010.Out of 1 714 samples, 1 300 samples were used for the establishment of review criteria, and 214 samples were used for criteria verification.The other 200 samples from healthy donors were used to set up the normal reference range of fully automated urine sedimental analyzer UF-1000i.RBC,WBC,PRO and CAST in all the samples were measured by Siemens Bayer Clinitek 500 urine strip chemistry analyzer, Sysmex UF-1000i urine sedimental analyzer and microscopic examination.Based on the different laboratory automation in urine analysis, four microscopic review protocols were defined： （1） Protocol 1： based on chemistry results only, microscopy review was performed when any of WBC, RBC and PRO was positive; （2） Protocol 2： based on fully automated sedimental analyzer only,microscopy review was performed when any of WBC, RBC and CAST was over the upper limit of the reference range; （3） Protocol 3： All the results of urine chemistry analyzer and sedimental analyzer were integrated.If two WBC results were different between two systems （in one system WBC was positive or over the upper limit of the reference range but in another system WBC was negative or within the reference range）, and any of RBC, PRO/CAST was positive or over the upper limit, microscopic review was performed; （4） Protocol 4： if any of WBC, RBC, PRO/CAST was different between two systems, microscopic review was performed.Review criteria were performed with Sysmex Laboman UriAccess 3.0 software.Results The reference ranges of UF-1000i parameters were RBC 0-7.5/μl （male）, 0-15.9/μl （female）; WBC 0-11.6/μl （male）, 0-12.7/μl （female）; Epithelial cell were 0-6.5/μl （male）, 0-21.4/μl （female）; CAST 0-1.3/μl.The results of microscopic examination revealed that positive samples were 47.46% （617/1 300） and negative samples were 52.54% （683/1 300）. Among positive samples, majority showed the presence of RBC （60.13%,371/617）, followed by CAST （8.43%,52/617）.The false negative rates of four protocols were 8.38% （109/1 300）, 4.69% （61/1 300）, 0.62% （8/1 300） and 0.54% （7/1 300）, respectively.The review rates were 47.85% （622/1 300）, 59.38% （772/1 300）, 72.85% （947/1 300） and 52.23% （679/1 300）, respectively.Although there were false negative cases in protocol 4, all the patients had normal serum creatine level.In those 214 patients for verification, the false negative rate using protocol 4 was zero, the review rates were 53.74% （115/214）.Conclusions Protocol 4 shows lest false negative rate and lower review rate.Importantly, there was no patients with serious renal function abnormality missed using protocol 4.Therefore, protocol 4 is an ideal criteria for microscopy review following automated urine analysis.