Robot-assisted radical cystectomy (RARC) is a technically challenging procedure associated with high overall postoperative morbidity [1, 2]. Relative to other forms of urinary diversion, neobladder reconstruction is associated with a higher complication rate [3]. As low morbidity represents one of the pivotal surgical outcomes, efforts must be focused on minimizing the morbidity rate. To achieve this goal, it is crucial to understand the most common complications associated with RARC to allow surgeons to take action to prevent their occurrence. To the best of our knowledge, only a few studies have evaluated the type of complications that may occur after RARC with intracorporeal neobladder, and the generalizability of their results is limited [1-5]. Previous studies have reported complications from small single-institution or single-surgeon series of patients who underwent RARC with either extracorporeal or intracorporeal neobladder, or did not use standardized criteria to report complications [6]. The present study aimed to characterize the type of complications occurring after RARC with intracorporeal neobladder in order to improve peri-operative management, highlight areas of need for future studies and, ultimately, reduce RARC morbidity.

On behalf of the European Association of Urology (EAU) Robotic Urology Section Scientific Working Group, we created a multi-institutional database of 980 patients who underwent RARC and intracorporeal neobladder at 16 high-volume European centres between 2003 and 2022. All patients with incomplete information on postoperative outcomes were excluded. Our final population included 858 individuals. The surgical technique was previously described [7]. Postoperative outcomes were collected according to the EAU recommendations (11/14 criteria; Table S1) [6]. Complications were scored using the Clavien–Dindo classification system, grouped by type and severity (severe: Clavien–Dindo score ≥3), and divided into early (<30 days) and late (31–90 days). We compared the type of complication between patients who received and those who did not receive neoadjuvant therapy (nadjT) using Fisher's exact test. Finally, by using a multivariable logistic regression model, we evaluated if the rate of complications was associated with amount of surgical experience of each centre (coded as the total number of RARCs with intracorporeal neobladder performed in that centre before the patient operation) or operating time after accounting for potential confounders (age, body mass index [BMI], American Society of Anesthesiologists [ASA] score, sex, preoperative T stage, and nadjT).

The majority of patients were male (87%), healthy (76% had an ASA score ≤2), and had muscle-invasive tumours (55%; Table S2). The median (interquartile range [IQR]) age at surgery and median (IQR) BMI were 64 years (58, 69) years and 26 (23.8, 28.4) kg/m², respectively. Almost all patients underwent pelvic lymphadenectomy (98%). The neobladder reconstruction technique was the Studer/Wiklund in 557 patients (65%), S pouch in 16 (1.9%), Gaston in 163 (19%), vescica ileale Padovana in 102 (12%) and Hautmann in 20 (2.3%). The median (IQR) operating time was 366 (300, 440) min and the urinary diversion time was 166 min. The median (IQR) length of stay was 10 (7, 15) days (Table S3). Overall, 514 (60%) and 223 patients (26%) experienced at least one complication and at least one severe complication within 3 months, respectively. One patient died within 1 month due to a complication (sepsis), and six patients died between 2 and 3 months after surgery (two from thromboembolic events, one from sepsis and three from cancer progression). Overall, 13 (1.5%) and 28 patients (3%) required intra-operative and postoperative blood transfusion, respectively. The rate of complications was highest in the first month post-surgery (early 52% vs late 20%). UTI was the most common complication at both time points (early: 20%, late: 10%; Fig. 1, Table S4). The most common early complications after UTI were paralytic ileus (10%) and urinary leakage/fistula (9%). The most frequent late complications after UTI were ureteric stricture (3%) and urinary leakage/fistula (2%). Although UTI was the most common occurrence, this was not severe in the majority of cases (Fig. 1, Table S4). Urinary leakage/fistula and urinary device complication (defined as mispositioning or obstruction of the bladder catheter and/or ureteric stents) were the most common severe early complications, whereas ureteric stricture and lymphocele/lymphorrea were the most common severe late complications.

Among female patients, the most common early complications were UTI (20%), urinary leakage/fistula (7%) and paralytic ileus (6.5%), whereas the most common late complications were UTI (9%), ureteric stricture (3%) and urinary leakage/fistula (3%).

Approximately half of the population received nadjT (48%; Table S5). There were no significant differences in the rate of overall (59% vs 60%; P = 0.8) and severe (26% vs 26%; P > 0.9) complications between patients who received and those who did not receive nadjT. However, the rates of fever of unknown origin and paralytic ileus were significantly higher in the nadjT group (fever of unknown origin: 9% vs 4%, P = 0.012; paralytic ileus: 13% vs 7%, P = 0.008). On multivariable analysis, the centre's surgical experience was inversely associated with both overall (odds ratio [OR] 0.97, 95% CI 0.94, 0.99) and severe complications (OR 0.97, 95% CI 0.93, 1.00) after accounting for potential confounders (Table S6). By contrast, operating time was associated only with overall complications (OR 1.03, 95% CI 1.02, 1.05; Table S7).

In the present study, we evaluated complications after RARC with intracorporeal neobladder. Our findings have several clinical implications. First, despite the rate of complications decreasing with increasing centre experience (multivariable analysis results), probably as a reflection of surgical learning and improvement in surgical technique, our results highlight a high overall rate of complications after RARC. This study therefore supports the need for future efforts aimed at improving surgical technique and peri-operative management. In particular, investigations are needed to improve bowel function recovery and ureteric-neobladder anastomosis, which, as emerged in our results, is frequently affected by leakage or stenosis. In the meantime, our findings could be used for modifying the peri-operative management of these patients. The enhanced recovery after surgery (ERAS) protocol was implemented in an effort to reduce some of these complications (e.g., paralytic ileus) [8] and should therefore be applied in clinical practice. Regarding the high incidence of urinary leakage/fistula, clinicians should consider performing a leakage test before removing stents and bladder catheters. Overall, UTI was the most common complication, although often not severe. Thus, performing a urine test and/or urine culture before surgery could be useful to prevent the incidence of this occurrence. Our results also underline the importance of careful management of the urinary device. Indeed, mispositioning or obstruction of the urinary device was one of the most common early severe complications. Finally, clinicians should be aware that the incidence of fever of unknown origin and paralytic ileus is higher in patients who received nadjT.

Despite several strengths, this study has some limitations. First, all patients underwent surgery at a tertiary referral centre with significant experience in robotic surgery. Second, the complication rate may be underestimated as it was based on the presence of condition codes, with the absence of such a record taken to indicate the absence of complications. However, it is important to note that our focus was on the most common type of complication following RARC and intracorporeal neobladder, and this aspect may be less affected by the aforementioned limitation compared to the overall rate of complications. Indeed, we can speculate that all types of complications were underestimated equally. Finally, because the follow-up was relatively short, the incidence of some late complications (e.g., ureteric stricture) may be underestimated.

To conclude, our results provide useful information for improving the management of patients who underwent RARC with intracorporeal neobladder. Moreover, our findings point to the need for future studies aimed at optimizing RARC and intracorporeal neobladder and reducing the incidence of complications following surgery.

机器人辅助根治性膀胱切除术 (RARC) 是一项技术上具有挑战性的手术，术后总体发病率较高[ 1, 2 ]。相对于其他形式的尿流改道，新膀胱重建的并发症发生率较高[ 3 ]。由于低发病率是关键的手术结果之一，因此必须集中精力最大限度地降低发病率。为了实现这一目标，了解与 RARC 相关的最常见并发症至关重要，以便外科医生能够采取行动防止其发生。据我们所知，只有少数研究评估了 RARC 体内新膀胱术后可能发生的并发症类型，并且其结果的普遍性有限 [ 1-5 ]。先前的研究报告了单机构或单外科医生系列小规模患者接受体外或体内新膀胱 RARC 的并发症，或未使用标准化标准报告并发症 [ 6 ]。本研究旨在描述体内新膀胱 RARC 后发生的并发症类型，以改善围手术期管理，突出未来研究需要的领域，并最终降低 RARC 发病率。

我们代表欧洲泌尿外科协会 (EAU) 机器人泌尿科科学工作组，创建了一个多机构数据库，其中包含 2003 年至 2022 年间在 16 个高容量欧洲中心接受 RARC 和体内新膀胱治疗的 980 名患者。有关术后结果的信息被排除在外。我们的最终人口包括 858 人。先前描述了手术技术[ 7 ]。根据 EAU 建议（11/14 标准；表 S1）收集术后结果[ 6 ]。使用Clavien-Dindo分类系统对并发症进行评分，按类型和严重程度分组（严重：Clavien-Dindo评分≥3），并分为早期（<30天）和晚期（31-90天）。我们使用 Fisher 精确检验比较了接受新辅助治疗 (nadjT) 的患者和未接受新辅助治疗 (nadjT) 的患者之间的并发症类型。最后，通过使用多变量逻辑回归模型，我们评估了并发症发生率是否与每个中心的手术经验量（编码为患者手术前在该中心进行的体内新膀胱的 RARC 总数）或手术时间相关。在考虑了潜在的混杂因素（年龄、体重指数 [BMI]、美国麻醉医师协会 [ASA] 评分、性别、术前 T 分期和 nadjT）后。

大多数患者为男性（87%），健康（76% ASA 评分≤2），患有肌肉浸润性肿瘤（55%；表 S2）。手术时中位年龄（四分位距[IQR]）和中位BMI（IQR）分别为64岁（58, 69）岁和26（23.8, 28.4）kg/m ² 。几乎所有患者均接受了盆腔淋巴结切除术（98%）。新膀胱重建技术为 Studer/Wiklund 557 例 (65%)、S pouch 16 例 (1.9%)、Gaston 163 例 (19%)、vescica ileale Padovana 102 例 (12%) 和 Hautmann 20 例 (2.3%) ）。中位手术时间 (IQR) 为 366 (300, 440) 分钟，尿流改道时间为 166 分钟。中位住院时间 (IQR) 为 10 (7, 15) 天（表 S3）。总体而言，3 个月内分别有 514 名 (60%) 和 223 名患者 (26%) 经历了至少一种并发症和至少一种严重并发症。一名患者在 1 个月内因并发症（败血症）死亡，六名患者在术后 2 至 3 个月内死亡（两名患者死于血栓栓塞事件，一名患者死于败血症，三名患者死于癌症进展）。总体而言，分别有 13 名患者 (1.5%) 和 28 名患者 (3%) 需要术中和术后输血。术后第一个月并发症发生率最高（早期 52% vs 晚期 20%）。 UTI 是两个时间点最常见的并发症（早期：20%，晚期：10%；图 1，表 S4）。 UTI 后最常见的早期并发症是麻痹性肠梗阻 (10%) 和尿漏/瘘管 (9%)。 UTI 后最常见的晚期并发症是输尿管狭窄 (3%) 和尿漏/瘘管 (2%)。尽管尿路感染是最常见的情况，但大多数病例并不严重（图 1，表 S4）。尿漏/瘘管和泌尿装置并发症（定义为膀胱导管和/或输尿管支架错位或阻塞）是最常见的严重早期并发症，而输尿管狭窄和淋巴囊肿/淋巴管是最常见的严重晚期并发症。

在女性患者中，最常见的早期并发症是尿路感染（20%）、尿漏/瘘管（7%）和麻痹性肠梗阻（6.5%），而最常见的晚期并发症是尿路感染（9%）、输尿管狭窄（3%） ）和尿漏/瘘管（3%）。

大约一半的人口接受了 nadjT（48%；表 S5）。 接受 nadjT 治疗的患者与未接受 nadjT 治疗的患者之间，总体并发症发生率（59% vs 60%； P  = 0.8）和严重并发症发生率（26% vs 26%；P > 0.9）没有显着差异。然而，nadjT 组的不明原因发热和麻痹性肠梗阻发生率明显较高（不明原因发热：9% vs 4%，P  = 0.012；麻痹性肠梗阻：13% vs 7%，P  = 0.008）。在多变量分析中，考虑到潜在的混杂因素后，该中心的手术经验与总体并发症（比值比 [OR] 0.97，95% CI 0.94，0.99）和严重并发症（OR 0.97，95% CI 0.93，1.00）呈负相关（表S6）。相比之下，手术时间仅与总体并发症相关（OR 1.03，95% CI 1.02，1.05；表 S7）。

在本研究中，我们评估了 RARC 体内新膀胱术后的并发症。我们的研究结果具有多种临床意义。首先，尽管并发症发生率随着中心经验的增加而降低（多变量分析结果），这可能反映了手术学习和手术技术的改进，但我们的结果强调了 RARC 后并发症的总体发生率很高。因此，这项研究支持未来需要努力改进手术技术和围手术期管理。特别是，需要进行研究以改善肠道功能恢复和输尿管-新膀胱吻合术，正如我们的结果所示，这经常受到渗漏或狭窄的影响。与此同时，我们的研究结果可用于修改这些患者的围手术期管理。实施加速康复外科（ERAS）方案是为了减少其中一些并发症（例如麻痹性肠梗阻）[ 8 ]，因此应在临床实践中应用。鉴于尿漏/瘘管的高发生率，临床医生应考虑在移除支架和膀胱导尿管之前进行漏尿测试。总体而言，尿路感染是最常见的并发症，尽管通常并不严重。因此，在手术前进行尿液测试和/或尿液培养可能有助于预防这种情况的发生。我们的结果还强调了仔细管理泌尿装置的重要性。事实上，泌尿装置错位或阻塞是最常见的早期严重并发症之一。最后，临床医生应该意识到，接受 nadjT 的患者不明原因发热和麻痹性肠梗阻的发生率较高。

尽管有很多优点，这项研究也有一些局限性。首先，所有患者均在具有丰富机器人手术经验的三级转诊中心接受手术。其次，并发症发生率可能会被低估，因为它是基于条件代码的存在，而没有这样的记录则表明不存在并发症。然而，值得注意的是，我们的重点是 RARC 和体内新膀胱后最常见的并发症类型，与并发症的总体发生率相比，这一方面可能受上述限制的影响较小。事实上，我们可以推测所有类型的并发症都被同等地低估。最后，由于随访时间相对较短，一些晚期并发症（例如输尿管狭窄）的发生率可能被低估。

总之，我们的结果为改善接受 RARC 体内新膀胱患者的管理提供了有用的信息。此外，我们的研究结果表明，未来需要进行旨在优化 RARC 和体内新膀胱并降低手术后并发症发生率的研究。