Background: Despite numerous studies assessing LCs in urological surgical oncology, high-quality evidence and a fully structured curriculum are missing. We aimed to systematically search and review the available literature on the LCs of robot-assisted surgery in urological cancers. Methods: Medline was systematically searched up to December 2024 to retrieve studies following the Preferred Reporting Items reporting on LC in robot-assisted radical prostatectomy (RARP), robot-assisted radical cystectomy (RARC), robot-assisted radical and partial nephrectomy (RARN, RAPN), and robot-assisted radical nephroureterectomy (RANU). The results of the last five years were then compared to those of the previous years. Results: In total, 82 studies were identified, 47 of which were for prostatectomy, 9 of which were for the last 5 years. Eighteen studies referred to partial-nephrectomy, seven over the previous 5 years. Finally, 16 studies referred to radical cystectomy, 7 over the previous five years. For radical prostatectomy, LC was based on operative time (OT), estimated blood loss (EBL), length of hospital stays, complication rate, positive surgical margin (PSM), biochemical recurrence (BCR), continence, and potency with ranges of 100–400, 90–290, 200, 15–250, 50–300, 30–250, 200–500 and 200–300 cases, respectively. For partial nephrectomy, the LC was based on OT, EBL, length of hospital stay, complication rate, warm ischemia time (WIT), and trifecta, with unclear ranges for the first three categories and 20–50, 26–140, and 50–77 cases, respectively, for the rest. Finally, for radical cystectomy, the LC was based on OT, EBL, length of hospital stay, complication rate, PSM, and lymph node yield, with ranges 20–75, 88, 40–198, 16–100, no difference, and 30–50 cases, respectively. We could not identify any study assessing the LCs in RARN and RANU. Conclusions: Robot-assisted surgery does not have a standard definition of LC regardless of the type of operation, which causes heterogeneity between the studies. Nevertheless, LCs appear to be steep and continuous. Training curriculums are essential to optimize outcomes and prepare new surgeons.
背景：尽管已有大量研究评估泌尿外科肿瘤手术中的学习曲线，但仍缺乏高质量证据和完全结构化的培训课程。本研究旨在系统检索并综述泌尿系统肿瘤机器人辅助手术学习曲线的现有文献。方法：系统检索截至2024年12月的Medline数据库，筛选符合PRISMA报告规范的机器人辅助根治性前列腺切除术、机器人辅助根治性膀胱切除术、机器人辅助根治性及部分肾切除术、机器人辅助根治性肾输尿管切除术学习曲线研究。将最近五年的研究结果与早期研究进行对比分析。结果：共纳入82项研究，其中前列腺切除术相关47项（近五年9项），部分肾切除术相关18项（近五年7项），根治性膀胱切除术相关16项（近五年7项）。前列腺切除术学习曲线评估指标包括手术时间（100-400例）、估计失血量（90-290例）、住院时长（200例）、并发症发生率（15-250例）、切缘阳性率（50-300例）、生化复发率（30-250例）、控尿功能（200-500例）和性功能恢复（200-300例）。部分肾切除术学习曲线涵盖手术时间、失血量、住院时长（范围未明确），以及并发症发生率（20-50例）、热缺血时间（26-140例）和三联指标达标率（50-77例）。根治性膀胱切除术学习曲线涉及手术时间（20-75例）、失血量（88例）、住院时长（40-198例）、并发症发生率（16-100例）、切缘阳性率（无差异）及淋巴结检出数（30-50例）。未发现评估机器人辅助根治性肾切除术与肾输尿管切除术学习曲线的研究。结论：机器人辅助手术缺乏统一的学习曲线定义标准，导致研究间存在异质性。但学习曲线普遍呈现快速提升且持续改善的特征。建立规范化培训课程对优化手术效果及培养新术者至关重要。
肿瘤(癌症)患者之家