对于复杂肾癌（CRM）而言，局部肾切除（PN）的效用存在争议。我们确定了手术方式对CRM术后肾功能影响的重要性。我们回顾性分析了ROSULA多中心登记簿的数据。CRM的定义为RENAL Score 10-12。我们将队列分为PN和根治性肾切除（RN）进行分析。主要结果是新患估计肾小球滤过率（eGFR）<45 mL/min/1.73 m2的发生。次要结果是新患eGFR<60和诊断与最后随访之间的ΔeGFR。我们采用Cox比例风险法来确定de-novo eGFR<60和<45的预测因子。线性回归用于分析ΔeGFR。我们用Kaplan-Meier分析（KMA）来分析新患eGFR<60和<45的5年无疾病自由率。我们分析了969名患者（RN=429/PN=540；中位随访24.0个月）。RN患者的BMI较低（P<0.001），肿瘤大小较大（P<0.001）。PN的总术后并发症率较高（P<0.001），但重大并发症（Clavien III-IV级）没有差异（P=0.702）。多元回归分析显示年龄（HR=1.05，P<0.001），肿瘤大小（HR=1.05，P=0.046），RN（HR=2.57，P<0.001）和BMI（HR=1.04，P=0.001）与估计eGFR<60 mL/min/1.73 m2的风险有关。年龄（HR=1.03，P<0.001），BMI（HR=1.06，P<0.001），基线eGFR（HR=0.99，P=0.002），肿瘤大小（HR=1.07，P=0.007）和RN（HR=2.39，P<0.001）是估计eGFR<45 mL/min/1.73 m2的风险因素。RN（B=-10.89，P<0.001）与更大的ΔeGFR相关。KMA显示RN的5年新患eGFR<60（71% vs. 33%，P<0.001）和新患eGFR<45（79% vs. 65%，P<0.001）较差。PN在具备技术可行性的选择的CRM患者中提供功能性益处，且与RN相比，不会显著增加重大并发症的发生率，因此应当作为一种可行选择。

Utility of partial nephrectomy (PN) for complex renal mass (CRM) is controversial. We determined the impact of surgical modality on postoperative renal functional outcomes for CRM.We retrospectively analyzed a multicenter registry (ROSULA). CRM was defined as RENAL Score 10-12. The cohort was divided into PN and radical nephrectomy (RN) for analyses. Primary outcome was development of de-novo estimated glomerular filtration rate (eGFR)<45 mL/min/1.73 m2. Secondary outcomes were de-novo eGFR<60 and ΔeGFR between diagnosis and last follow-up. Cox proportional hazards was used to elucidate predictors for de-novo eGFR<60 and <45. Linear regression was utilized to analyze ΔeGFR. Kaplan-Meier Analysis (KMA) was performed to analyze 5-year freedom from de-novo eGFR<60 and <45.We analyzed 969 patients (RN=429/PN=540; median follow-up 24.0 months). RN patients had lower BMI (P<0.001) and larger tumor size (P<0.001). Overall postoperative complication rate was higher for PN (P<0.001), but there was no difference in major complications (Clavien III-IV; P=0.702). MVA demonstrated age (HR=1.05, P<0.001), tumor-size (HR=1.05, P=0.046), RN (HR=2.57, P<0.001), and BMI (HR=1.04, P=0.001) to be associated with risk for de-novo eGFR<60 mL/min/1.73 m2. Age (HR=1.03, P<0.001), BMI (HR=1.06, P<0.001), baseline eGFR (HR=0.99, P=0.002), tumor size (HR=1.07, P=0.007) and RN (HR=2.39, P<0.001) were risk factors for de-novo eGFR<45 mL/min/1.73 m2. RN (B=-10.89, P<0.001) was associated with greater ΔeGFR. KMA revealed worse 5-year freedom from de-novo eGFR<60 (71% vs. 33%, P<0.001) and de-novo eGFR<45 (79% vs. 65%, P<0.001) for RN.PN provides functional benefit in selected patients with CRM without significant increase in major complications compared to RN, and should be considered when technically feasible.