Renal cell carcinoma (RCC) is a commonly diagnosed malignancy in Europe, with approximately 100,000 new cases reported annually, affecting both men and women across the region [1, 2]. Surgical excision remains the primary curative treatment for localized RCC, with partial nephrectomy (PN) preferred as a nephron-sparing option whenever suitable. PN may be performed using a traditional open technique (OPN) or through minimally invasive methods, such as robotic-assisted partial nephrectomy (RAPN). Current international guidelines advise that the choice of surgical approach should be based on the surgeon’s proficiency and experience [3, 4], as there is a lack of high-quality randomized trials definitively favoring one technique over the other. Nevertheless, the widespread adoption of RAPN has accelerated in recent years [5], underscoring the pressing need for randomized studies to validate this trend.

Although OPN is a well-established and standardized operation, it is associated with considerable morbidity, including postoperative complications like hernias and flank bulges, which can affect up to 50% of patients and may have a lasting impact on quality of life [6, 7]. The shift toward minimally invasive PN is largely driven by its potential benefits, including reduced postoperative pain, lower incidence of wound-related complications, and shorter hospitalization, all contributing to a faster recovery process [8]. While some randomized clinical trials (RCTs) have demonstrated that robotic-assisted surgery offers advantages over open surgery in terms of estimated blood loss (EBL), length of hospital stay (LOS), and preservation of renal function [9–12], direct comparative data between RAPN and OPN remain limited. Therefore, the objective of this study is to evaluate and compare intraoperative, perioperative, and oncologic outcomes of open versus robotic-assisted partial nephrectomy using a propensity score-matched analysis.

A total of 449 patients underwent partial nephrectomy during the study period. Of these, 63 were excluded due to missing preoperative data or inadequate follow-up. The final analysis included 386 patients, comprising 172 who underwent OPN and 214 who underwent RAPN. Following propensity score matching, 152 patients in the open group were matched to 125 patients in the robotic group (Table 1). Matching was performed using a logistic regression model based on age at surgery, Charlson Comorbidity Index, PADUA and RENAL scores, clinical T stage, and tumor location (Supplementary materials). Post-matching comparison demonstrated a balanced distribution of baseline characteristics across groups, with no statistically significant differences in age, BMI, CCI, or nephrometrics scores, confirming an adequate balance between cohorts. Intraoperative complications were significantly more frequent in the open group than in the robotic group (Table 2). Before matching, complications occurred in 9% of open procedures compared to only 0.9% in robotic surgeries (p = 0.001). This difference remained significant even after matching, with 10% of patients in the open group experiencing complications versus 2% in the robotic group (p = 0.003). Length of hospital stay (LOS) was also longer among patients undergoing open surgery. After matching, the median LOS was 6 days (IQR 6–7) in the open group and 6 days (IQR 5–7) in the robotic group, with the difference reaching statistical significance (p = 0.001). Among secondary outcomes, estimated blood loss was substantially higher in the open group compared to the robotic group. After matching, the median blood loss was 300 mL (IQR 200–600) in the open group versus 200 mL (IQR 100–300) in the robotic group, which was statistically significant (p = 0.001). Operative time was also longer in open procedures (170 vs. 147 minutes, p = 0.001). In contrast, warm ischemia time showed no significant difference between the two approaches following matching (p = 0.12). Postoperative complications were more frequent in the open group, with overall complications occurring in 25% of patients compared to 14% in the robotic group (p = 0.01). However, no significant differences were observed in the rate of severe complications, defined as Clavien-Dindo grade ≥ III (2% vs. 1%, p = 0.25), nor in the incidence of positive surgical margins (PSM) (3% vs. 6%, p = 0.23).

Oncological outcomes at 5 years, including OS, CSS, and RFS, were similar between the two surgical approaches. The Kaplan-Meier analysis showed no statistically significant difference in OS (log-rank p = 0.68), CSS (log-rank p = 0.68), or RFS (log-rank p = 0.67) between patients undergoing open or robotic partial nephrectomy. The number of patients at risk at each time point is displayed in the corresponding Kaplan-Meier curves (Fig. 1).

In this single‑center, propensity‑matched cohort, RAPN was associated with fewer intraoperative and overall postoperative complications, lower estimated blood loss, shorter operative time, and shorter length of stay compared with OPN. Oncologic outcomes were comparable between approaches at the longest available follow‑up, with no differences in overall, cancer‑specific, or recurrence‑free survival. Together, these data suggest that the perioperative efficiency of RAPN does not compromise intermediate‑term oncologic control.

The lower complication rates observed in the RAPN group likely reflect the enhanced visualization, dexterity, and precision afforded by robotic platforms. These technical advantages translate clinically into less tissue trauma and more accurate vascular control, which in turn reduce bleeding and facilitate faster recovery [14]. Analyzing data of other surgical specialties, robotic surgical approach is strongly related to lower complication rates, so our results can be defined in line with the literature [15, 16]. The reduction in overall postoperative complication rates could lead to the spread of robot-assisted surgical approach for more challenging, bigger, and more complex renal masses, setting the line to higher postoperative standards.

The shorter LOS and the reduced EBL not only benefit patients but may also lessen the burden on healthcare systems by decreasing postoperative resource utilization. Indeed, a population-based analysis demonstrated that although RAPN incurs higher initial costs, these may be offset by fewer complications and readmissions [17]. Conversely, other studies have reported greater total expenditures for robotic procedures, indicating that the overall cost-effectiveness of RAPN remains to be fully elucidated [18, 19]. Notably, the reduced rate of intra- and postoperative complications of the robotic approach has led to an increase of its use in frail patients, who particularly benefit from the use of RAPN instead of OPN, as supported by data across various surgical specialties [20, 21].

Operative time is another relevant point. Surprisingly, in our study robotic surgery showed faster operative time than OPN. Probably proficiency acquired by both our surgeons and scrub nurses plays a pivotal role. Fast docking time and technical surgical skills can lead to a relevant reduction in work-time [22]. In addition, practice can improve operative time after each procedure [23]. Moreover, differences in terms of operative time could lay on risk factors: prior abdominal surgery and tumor size [24]. However, data in the literature about this topic reveal a longer operative time in robotic-assisted surgeries, it could be related to different docking techniques or the use of robot-assisted surgery for more complex procedures [25, 26]. More studies are needed, having consequences in hospitals’ economies.

Oncological safety was confirmed by equivalent rates of positive surgical margins (PSM) and comparable survival outcomes. This finding aligns with emerging evidence endorsing RAPN even for anatomically complex renal tumors, including those with elevated RENAL or PADUA scores [27–30]. Accurate tumor classification is therefore critical for anticipating surgical challenges and selecting the optimal approach [31]. To this end, the introduction of three-dimensional virtual modeling enhances preoperative planning by offering superior spatial resolution compared with conventional two-dimensional imaging, thereby improving risk stratification and facilitating nephron-sparing decisions [32]. Future trials incorporating refined complexity metrics may further delineate the contexts in which RAPN offers maximal advantage over OPN.

Our results have both clinical and systemic implications. Clinically, RAPN appears to deliver a safer perioperative course and equivalent long-term cancer control, which may translate into expedited convalescence and reduced morbidity. From a health-system perspective, these advantages could yield cost savings and greater efficiency as robotic technology becomes increasingly prevalent.

This study has limitations. Key strengths include granular perioperative data, use of nephrometry scores, and efforts to mitigate selection bias via propensity methods. Limitations include the retrospective single‑center design, residual post‑matching imbalances, absence of detailed renal functional endpoints, and follow‑up immaturity for definitive 5‑year conclusions. These limitations notwithstanding, the consistent perioperative advantages without apparent oncologic compromise support the use of RAPN in appropriately selected patients.

RAPN is associated with lower perioperative morbidity and resource utilization compared with OPN, with comparable oncologic outcomes at the longest available follow‑up. Enhanced reporting of renal functional preservation, rigorous balance diagnostics, and sensitivity analyses will further substantiate these findings and inform patient‑centered surgical decision‑making.