文献速递小儿泌尿与泌尿修复

排尿性膀胱尿道造影定量特征可预测膀胱输尿管反流患儿核素显像肾实质缺损

Quantitative Voiding Cystourethrogram Features Predict Detecting Photopenic Renal Parenchyma Defects on Radionuclide Imaging in Patients With Vesicoureteral Reflux.
PubMed 文献 · Urology · 2025-05-21

本研究纳入130例原发性膀胱输尿管反流患儿(163个肾单位),评估排尿性膀胱尿道造影(VCUG)定量特征预测光缺损性肾实质缺损(PRD)的价值。结果显示,联合模型预测性能最高(AUROC 0.84,AUPRC 0.87),优于单独使用反流分级或定量特征。定量VCUG特征有望优化PRD风险预测。

要点

  • VCUG定量特征(输尿管宽度、迂曲度)可提升PRD预测模型性能。
  • 联合反流分级与定量特征的模型预测效能最高(AUROC 0.84)。
  • PRD阳性与阴性组在年龄、高级别反流及复发性尿路感染史方面差异显著。

核心结论

  • 定量VCUG特征在优化预测模型以识别VUR患儿中具有前景,其预测PRD的性能优于传统指标。

研究设计与样本

  • 回顾性研究,纳入2016至2021年间在本机构接受VCUG及核素肾扫描(间隔6个月内)的130例原发性VUR患儿(163个肾单位),中位VCUG年龄20个月。

主要发现

  • PRD阳性与阴性组在年龄(47 vs 7个月,P <.001)、高级别反流(54% vs 29%,P = .001)及复发性尿路感染史(57% vs 26%,P <.001)方面差异显著。
  • 联合模型预测性能最高(AUROC 0.84,AUPRC 0.87),其次为定量VCUG模型(AUROC 0.80,AUPRC 0.85)和分级模型(AUROC 0.81,AUPRC 0.82)。

临床怎么用

  • 临床可利用VCUG定量特征(如输尿管宽度和迂曲度)辅助识别VUR患儿中PRD高风险个体,优化管理决策。

局限性

  • 模型开发未纳入慢性肾脏病和高血压等临床终点。

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

OBJECTIVE: To evaluate the utility of quantitative features on voiding cystourethrogram (VCUG) to predict the presence of photopenic renal defects (PRDs) consistent with scarring. PRDs remain an indicator of kidney damage in patients with vesicoureteral reflux (VUR), and their presence influences management and long-term outcomes. METHODS: We identified children with primary VUR who underwent a VCUG and a nuclear renal scan within 6 months at our institution between 2016 and 2021. Quantitative VUR features (qVUR) were extracted, including ureteral width (proximal, distal, and maximum) and ureteral tortuosity. Logistic regression models were developed to predict PRDs employing clinical features, VCUG indication, and VUR laterality (baseline model); the addition of VUR grade (grade model); the addition of qVUR (qVUR model); and the addition of both grade and qVUR (combined model). RESULTS: In total, 130 children (163 renal units) were included, with median age at VCUG of 20 months. PRD+ and PRD- groups differed significantly in age (47 vs 7 months, P <.001), high-grade reflux (54% vs 29%, P = .001), and recurrent urinary tract infection history (57% vs 26%, P <.001). Model performance was highest for the combined model (AUROC 0.84, AUPRC 0.87), followed by qVUR (AUROC 0.80, AUPRC 0.85) and grade (AUROC 0.81, AUPRC 0.82). This analysis was limited by the absence of clinical endpoints incorporated into model development, including the development of chronic kidney disease and hypertension. CONCLUSION: Quantitative VCUG features show promise in refining predictive models to identify children with VUR who are at risk for PRDs suggestive of renal scarring, outperforming traditional metrics.

来源:Urology,查看原文。版权归原作者所有。

引用

  1. Ahmad I, Khondker A, Palchik S, et al. Quantitative Voiding Cystourethrogram Features Predict Detecting Photopenic Renal Parenchyma Defects on Radionuclide Imaging in Patients With Vesicoureteral Reflux. Urology. 2025. PMID: 40409415. doi:10.1016/j.urology.2025.05.033

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。