Nephrometry scores and perioperative outcomes following robotic partial nephrectomy

ABSTRACT Objectives: Based on imaging features, nephrometry scoring systems have been conceived to create a standardized and reproducible way to characterize renal tumor anatomy. However, less is known about which of these individual measures are important with regard to clinically relevant perioperative outcomes such as ischemia time (IT), estimated blood loss (EBL), length of hospital stay (LOS), and change in estimated glomerular filtration rate (eGFR) after robotic partial nephrectomy (PN). We aimed to assess the utility of the RENAL and PADUA scores, their subscales, and C-index for predicting these outcomes. Materials and Methods: We analyzed imaging studies from 283 patients who underwent robotic PN between 2008 and 2014 to assign nephrometry scores (NS): PADUA, RENAL and C-index. Univariate linear regression was used to assess whether the NS or any of their subscales were associated with EBL or IT. Multivariable linear regression and linear regression models were created to assess LOS and eGFR. Results: The three NS were significantly associated with EBL, IT, LOS, and eGFR at 12 months after surgery. All subscales with the exception of anterior/posterior were significantly associated with EBL and IT. Collecting system, renal rim location, renal sinus, exophytic/endophytic, and nearness to collecting system were significant predictors for LOS. Only renal rim location, renal sinus invasion and polar location were significantly associated with eGFR at 12 months. Conclusions: Tumor size and depth are important characteristics for predicting robotic PN outcomes and thus could be used individually as a simplified way to report tumors features for research and patient counseling purposes.

摘要 目的：基于影像学特征，学界已构建肾测量评分系统（nephrometry scoring systems），旨在建立标准化且可重复的肾脏肿瘤解剖特征表征方式。但目前针对各评分体系中哪些单项指标与机器人辅助部分肾切除术（robotic partial nephrectomy, PN）后临床相关围手术期结局——包括缺血时间（ischemia time, IT）、估计失血量（estimated blood loss, EBL）、住院时长（length of hospital stay, LOS）及术后估算肾小球滤过率（estimated glomerular filtration rate, eGFR）变化——存在关联的研究仍相对较少。本研究旨在评估RENAL评分、PADUA评分及其子量表与C指数（C-index）对上述结局的预测效能。

材料与方法：本研究分析了2008年至2014年间接受机器人辅助PN的283例患者的影像学资料，为其赋予肾测量评分（nephrometry scores, NS），具体包括PADUA评分、RENAL评分及C指数。采用单因素线性回归分析，评估各NS评分及其子量表是否与EBL或IT存在关联；构建多因素线性回归与线性回归模型，分别对LOS及eGFR相关结局进行评估。

结果：三种NS评分均与术后12个月的EBL、IT、LOS及eGFR显著相关。除前/后位置亚项外，其余所有子量表均与EBL和IT显著相关。集合系统、肾缘位置、肾窦、外生性/内生性特征以及距集合系统的距离均为LOS的显著预测因子。仅肾缘位置、肾窦侵犯及极性位置与术后12个月的eGFR显著相关。

结论：肿瘤大小与深度是预测机器人辅助PN结局的关键特征，因此可单独作为简化方案用于报告肿瘤特征，以满足科研与患者咨询的需求。