Flahault Adrien, Chassé Jean-François, Thervet Eric, Karras Alexandre, Pallet Nicolas
Université Paris Descartes, Paris, France, Service de néphrologie, Hôpital Européen Georges Pompidou, Paris.
Université Paris Descartes, Paris, France, Service de biochimie, Hôpital Européen Gorges Pompidou, AP-HP, Paris, France.
Ann Biol Clin (Paris). 2018 Jun 1;76(3):259-269. doi: 10.1684/abc.2018.1343.
The analysis of urinary protein composition is an important step in the evaluation and monitoring of kidney diseases. Among the various approaches, the determination of urinary-specific proteins makes it possible to non-invasively detect a preferentially tubular or glomerular injury, to orientate towards a pathological process, to guide the indication of a kidney biopsy, and to follow the evolution of the disease and the effectiveness of a therapy. No study systematically evaluated the performance of urinary-specific proteins for the diagnosis of a renal disease. We conducted this retrospective study to perform an exhaustive analysis of the correlations that may exist between histologically proven kidney disease and the corresponding specific urinary protein composition it in order to evaluate the diagnostic value of each of its components. Urinary concentrations of total protein, albumin, transferrin, alpha1microglobulin, beta2microglobulin, retinol binding protein, and immunoglobulin G were analyzed in more than 500 patients who underwent renal biopsy and concomitant urine specific protein analysis. Our analysis shows that these markers have a limited positive predictive value in this cohort of complex and unselected kidney diseases. In particular, low molecular weight proteins, and especially alpha1microglobulin, are frequently associated with glomerular diseases. We identified transferrin as an independent predictor of minimal changes disese and renal amyloidosis, and beta2microglobulin as an independent predictor of acute tubulointerstitial nephropathy and myelomatous tubulopathy. Finally, we defined the thresholds at which these parameters had excellent negative predictive values.
尿蛋白成分分析是评估和监测肾脏疾病的重要步骤。在各种方法中,测定尿特异性蛋白能够非侵入性地检测优先出现的肾小管或肾小球损伤,确定病理过程的方向,指导肾活检的指征,并跟踪疾病的进展和治疗效果。尚无研究系统评估尿特异性蛋白对肾脏疾病的诊断性能。我们进行了这项回顾性研究,以详尽分析经组织学证实的肾脏疾病与其相应的特异性尿蛋白成分之间可能存在的相关性,从而评估其各成分的诊断价值。对500多名接受肾活检并同时进行尿特异性蛋白分析的患者的尿总蛋白、白蛋白、转铁蛋白、α1微球蛋白、β2微球蛋白、视黄醇结合蛋白和免疫球蛋白G的浓度进行了分析。我们的分析表明,在这组复杂且未经选择的肾脏疾病患者中,这些标志物的阳性预测价值有限。特别是低分子量蛋白,尤其是α1微球蛋白,经常与肾小球疾病相关。我们确定转铁蛋白是微小病变疾病和肾淀粉样变性的独立预测因子,β2微球蛋白是急性肾小管间质性肾炎和骨髓瘤性肾小管病的独立预测因子。最后,我们确定了这些参数具有极佳阴性预测价值的阈值。