血清尿调蛋白评估血液透析患者残余肾功能
在151例血液透析患者中,血清尿调蛋白识别残余尿素清除率≥2 mL/min/35L的AUC为0.891,与β2微球蛋白相当;两者联合模型AUC达0.923,优于单一指标。采用双阈值灰区策略,尿调蛋白单独可使76%患者免于尿液收集,联合模型达85%。
要点
- 血清尿调蛋白可独立识别血液透析患者残余肾功能,AUC 0.891。
- 尿调蛋白与β2微球蛋白联合模型AUC 0.923,优于单一指标。
- 灰区策略下联合模型使85%患者免于 interdialytic 尿液收集。
核心结论
- 血清尿调蛋白提供与传统滤过标志物正交且非冗余的残余肾功能信息,联合β2微球蛋白可可靠识别KRU≥2 mL/min/35L阈值,使多数患者免于尿液收集。
研究设计与样本
- 前瞻性观察研究,纳入瑞士日内瓦两个中心的151例血液透析患者。
- 透析前测量血清尿调蛋白和β2微球蛋白,通过透析间期定时尿液收集测定残余尿素清除率(KRU)。
- 采用ROC分析评估各标志物识别KRU≥2 mL/min/35L的 discriminative 性能,并用逻辑回归联合两标志物。
- 应用双阈值灰区策略,预先设定排除和纳入截点以达到≥90%灵敏度和特异度。
主要发现
- 151例患者中92例(61%)KRU<2 mL/min/35L,59例(39%)KRU≥2 mL/min/35L。
- 血清尿调蛋白识别KRU≥2 mL/min/35L的AUC为0.891(95%CI 0.839-0.944),与β2微球蛋白(AUC 0.877;P=0.66)统计学等效。
- 联合模型AUC为0.923(95%CI 0.883-0.964),优于尿调蛋白单独(P=0.03)和β2微球蛋白单独(P=0.01)。
- 灰区策略下,尿调蛋白单独分类76%患者免于尿液收集(阴性预测值97%,阳性预测值84%),联合模型分类85%。
- 限制于非无尿患者分析, discriminative 性能轻度下降(AUC 0.833),阳性预测值保留(84%),80%该亚组免于尿液收集。
关键数值与推荐
- KRU≥2 mL/min/35L为临床显著残余肾功能阈值。
- 灰区策略预先设定排除和纳入截点以达到≥90%灵敏度和特异度。
- 尿调蛋白单独:阴性预测值97%,阳性预测值84%,76%患者免于尿液收集。
- 联合模型:85%患者免于尿液收集。
- 非无尿亚组:AUC 0.833,阳性预测值84%,80%免于尿液收集。
临床怎么用
- 血清尿调蛋白可单独或联合β2微球蛋白用于识别血液透析患者残余肾功能,减少透析间期尿液收集需求。
- 联合模型在灰区策略下可使85%患者免于尿液收集,支持 incremental 血液透析处方。
- 非无尿患者中联合模型仍保持较高阳性预测值,80%可避免尿液收集。
局限性
- 研究为观察性设计,样本来自两个中心,结论需外部验证。
- 原文未报告其他局限性。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: Residual kidney function (RKF) monitoring is essential to safe incremental hemodialysis (HD) prescription. Yet the cumbersome nature of interdialytic urine collections represents a major barrier to routine implementation. We investigated serum uromodulin, as a marker of functional renal mass, in identifying HD patients with clinically significant RKF when used alone, or in combination with serum β2-microglobulin, a traditional marker of glomerular filtration. METHODS: We conducted a prospective observational study of 151 HD patients at two centers in Geneva, Switzerland. Serum uromodulin and β2-microglobulin were measured on pre-dialysis samples, and residual urea clearance (KRU) was determined by interdialytic timed urine collection. The discriminative performance of each biomarker for identifying patients with KRU ≥2 mL/min/35L was assessed by receiver operating characteristic (ROC) analysis, and the two markers were then combined using logistic regression. We applied a two-threshold grey-zone strategy, defining a priori rule-out and rule-in cutoffs to achieve ≥90% sensitivity and specificity, respectively. RESULTS: Of the 151 patients, 92 (61%) had KRU <2 mL/min/35L and 59 (39%) had KRU ≥2 mL/min/35L. Serum uromodulin discriminated KRU ≥2 mL/min/35L with an area under the curve (AUC) of 0.891 (95% CI 0.839-0.944), statistically equivalent to β2-microglobulin (AUC 0.877; P=0.66). Each marker contributed independent information: the combined model achieved an AUC of 0.923 (95% CI 0.883-0.964), superior to either alone (vs uromodulin P=0.03; vs β2-microglobulin P=0.01). Using the grey-zone strategy, uromodulin alone classified 76% of patients without urine collection (negative predictive value (NPV) 97%, positive predictive value (PPV) 84%), while the combined model classified 85%. Restricting analysis to non-anuric patients modestly decreased discrimination (AUC 0.833) with preserved PPV (84%), while avoiding the need for urine collection in 80% of this sub-group. CONCLUSIONS: Serum uromodulin provides RKF information orthogonal to and non-redundant with conventional filtration markers. Their combination reliably identifies the KRU ≥2 mL/min/35L threshold relevant to incremental hemodialysis, sparing most patients from urine collection.
引用
- Jaques DA, Jotterand Drepper V, Marangon N, et al. Estimating Residual Kidney Function in Hemodialysis Patients Using Serum Uromodulin. Clin J Am Soc Nephrol. 2026. PMID: 42879202. doi:10.2215/CJN.0000001247
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。