Vijaya Gopalan, Cartwright Rufus, Bhide Alka, Derpapas Alexandros, Fernando Ruwan, Khullar Vik
St. Mary's Hospital-Department of Urogynaecology, London, United Kingdom.
Imperial College London-Institute of Reproductive and Developmental Biology, London, United Kingdom.
Neurourol Urodyn. 2016 Nov;35(8):944-948. doi: 10.1002/nau.22832. Epub 2015 Jul 30.
The validity and reliability of measurement of urinary NGF as a diagnostic biomarker in women with lower urinary tract dysfunction (LUTD) is uncertain. We aimed to evaluate both the diagnostic and discriminant validity, and the test-retest reliability of urinary NGF measurement in women with LUTD.
Urinary NGF was measured in women with LUTD (n = 205) and asymptomatic subjects (n = 31). Urinary NGF was assayed using an ELISA method and normalized against urinary creatinine. NGF/creatinine ratios were compared between symptom subgroups using Mann-Whitney U test, and between different urodynamic diagnoses using the Kruskal-Wallis test. Receiver Operator Characteristic (ROC) analysis was employed to evaluate the diagnostic performance of urinary NGF. Test-retest reliability of NGF measurement was assessed using intra-class correlation (ICC).
Urinary NGF was significantly but non-specifically increased in symptomatic patients when compared to controls (13.33 vs. 2.05 ng NGF/g Cr, P < 0.001). On multivariate logistic regression NGF was a good predictor of patients having OAB or not, however, the adjusted odds ratio only 1.006. ROC analysis demonstrated poor discriminant ability between different symptomatic groups and urodynamic groups. Using a cut off of 13.0 ng NGF/g creatinine the test provides a sensitivity of 81%, but a specificity of only 39% for overactive bladder. The assays demonstrated good test-retest reliability with ICC of 0.889.
Although urinary NGF can be reliably assayed, and is increased in various LUTDs, it discriminates poorly between these disorders therefore has very limited potential as a biomarker. Neurourol. Urodynam. 35:944-948, 2016. © 2015 Wiley Periodicals, Inc.
作为下尿路功能障碍（LUTD）女性诊断生物标志物的尿神经生长因子（NGF）测量的有效性和可靠性尚不确定。我们旨在评估尿NGF测量在LUTD女性中的诊断和鉴别效度以及重测信度。
对LUTD女性（n = 205）和无症状受试者（n = 31）测量尿NGF。采用酶联免疫吸附测定（ELISA）法检测尿NGF，并根据尿肌酐进行标准化。使用曼-惠特尼U检验比较症状亚组之间以及使用克鲁斯卡尔-沃利斯检验比较不同尿动力学诊断之间的NGF/肌酐比值。采用受试者操作特征（ROC）分析评估尿NGF的诊断性能。使用组内相关系数（ICC）评估NGF测量的重测信度。
与对照组相比，有症状患者的尿NGF显著但非特异性升高（13.33对2.05 ng NGF/g Cr，P < 0.001）。多因素逻辑回归显示，NGF是患者是否患有膀胱过度活动症（OAB）的良好预测指标，然而，调整后的优势比仅为1.006。ROC分析表明，不同症状组和尿动力学组之间的鉴别能力较差。以13.0 ng NGF/g肌酐为临界值，该检测对膀胱过度活动症的敏感性为81%，但特异性仅为39%。检测显示出良好的重测信度，ICC为0.889。
尽管尿NGF能够可靠检测，且在各种LUTD中升高，但它在这些疾病之间的鉴别能力较差，因此作为生物标志物的潜力非常有限。《神经泌尿学与尿动力学》35:944 - 948，2016年。© 2015威利期刊公司
