da Vinci^® single-port versus da Vinci^® Xi multiport robot-assisted radical prostatectomy for localized prostate cancer: a GRADE-assessed systematic review and meta-analysis

Abstract

Background

The adoption of the da Vinci^® single-port (SP) platform for robot-assisted radical prostatectomy (RARP) has expanded; however, comparative evidence versus the da Vinci^® Xi multiport system remains heterogeneous across access corridors and perioperative pathways. We performed a GRADE-assessed systematic review and meta-analysis to quantify platform-level differences in oncologic surrogates, perioperative recovery, and early functional outcomes.

Methods

This systematic review and meta-analysis was prospectively registered in PROSPERO (CRD420261299189). We systematically searched MEDLINE (via PubMed), Embase, Scopus, Web of Science, the Cochrane Central Register of Controlled Trials (CENTRAL), and ClinicalTrials.gov from inception to the final date of February 1, 2026. Eligible studies were comparative cohorts (prospective or retrospective, including propensity-matched designs) of adult patients with localized prostate cancer undergoing da Vinci^® SP RARP compared directly with da Vinci^® Xi multiport RARP. Studies were required to report at least one extractable perioperative, oncologic, or functional outcome measure. Outcomes were pooled using random effect models. The risk of bias was assessed using ROBINS-I, and the certainty of evidence was appraised using GRADE. No external funding was received for this study.

Results

Five comparative cohorts were included, comprising 968 patients who underwent SP-RARP and 668 patients who underwent Xi or multiport RARP. The positive surgical margin rates were not significantly different between the approaches (RR 0.90, 95% CI 0.75–1.09; I²=0%). Operative time was longer with SP (MD 10.39 min, 95% CI 0.72–20.07; I²=88%), whereas estimated blood loss (MD − 68.95 mL, 95% CI − 128.27 to − 9.63; I²=98%) and length of hospital stay (MD − 17.01 h, 95% CI − 24.12 to − 9.89 h; I²=91%) favored SP. Lymph node dissection rates (RR 0.49, 95% CI 0.20–1.23; I²=99%), postoperative complications (RR 1.06, 95% CI 0.71–1.60; I²=0%), and urinary continence at ~ 3 months (RR 1.22, 95% CI 0.87–1.71; I²=92%) were not significantly different between the two groups. Corridor-based subgrouping did not modify the PSM effects (p for subgroup differences = 0.51). The certainty of evidence ranged from low to very low, limited by non-randomized designs and inconsistencies for several recovery endpoints.

Conclusions

Based on low-to very-low-certainty evidence from five nonrandomized cohorts, SP-RARP showed similar margin control and complication risk to Xi/multiport RARP, with a directional recovery signal favoring SP for blood loss and hospital stay; however, these estimates carried extreme heterogeneity (I²=98% and 91%) and should be regarded as exploratory. No strong platform-level recommendations can be made without prospective confirmatory evidence.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12893-026-03950-9.

Introduction

Localized prostate cancer remains a principal driver of urologic surgical volume, and robot-assisted radical prostatectomy (RARP) is a cornerstone of definitive management owing to its capacity to achieve oncologic clearance while simultaneously preserving the urinary continence and erectile function. The evolution of robotic urologic surgery has progressed from early robotic systems and conventional multiport platforms to reduced-port, single-site, and single-port approaches, reflecting a broader historical trend toward access minimization while maintaining the precision of robotic surgery [1, 2]. The purpose-built da Vinci single-port (SP) platform has advanced this trajectory toward access minimization in RARP; nonetheless, whether reduced-port SP strategies confer clinically meaningful advantages over the well-established multiport (MP) transperitoneal technique remains to be investigated. Contemporary comparative cohorts have directly contrasted SP approaches with MP techniques through propensity-matched analyses, reporting divergent findings with respect to operative efficiency, perioperative recovery, and early functional outcomes [1, 2]. Platform-level comparisons between da Vinci SP and the da Vinci Xi have raised the question of whether the observed benefits reflect approach selection, system design, or the cumulative effect of perioperative pathway optimization [3]. More recent propensity-matched evaluations of transvesical SP versus standard MP RARP have extended this discourse to alternative access corridors that may further modify recovery [4]. Earlier comparative studies have similarly identified differences in postoperative pain control and length of hospital stay between SP and MP prostatectomy, underscoring the enduring clinical relevance of access strategy selection [5].

Several biological and technical mechanisms have been proposed as plausible mediators of these differences. Extraperitoneal and transvesical corridors may reduce the extent of peritoneal insufflation and bowel manipulation, thereby attenuating the postoperative ileus cascade, systemic inflammatory stress responses, and opioid analgesic requirements, each of which is intimately linked to early mobilization capacity and discharge readiness after pelvic surgery [1, 2, 4]. Conversely, the constrained instrument geometry inherent to single-site access limits surgical triangulation and operative workspace, which may increase technical complexity and adversely influence operative time, extent of lymphadenectomy, and margin control in anatomically demanding configurations [1, 2, 5]. The critical importance of technique maturation is further illuminated by outcome and learning curve analyses of extraperitoneal single-port RARP performed with the da Vinci Xi platform, demonstrating that proficiency effects meaningfully shape both perioperative metrics and functional recovery trajectories [6].

Despite the expansion of contemporary experience, the existing evidence base remains heterogeneous and methodologically constrained. Most comparative studies are retrospective in design, frequently limited to single institutional settings, and differ with respect to access routes, perioperative management protocols, and outcome endpoint definitions. These inconsistencies complicate cross-study inferences and restrict the degree of certainty that can be imparted to shared clinical decision-making [1–6]. Even within propensity-matched frameworks, residual confounding may persist, and clinically consequential endpoints, such as biochemical disease control and durable functional recovery, are frequently reported with a limited follow-up duration or non-uniform assessment methodology [1, 3–5]. Critically, existing higher-level evidence syntheses have predominantly addressed the broader question of novel robotic platforms versus conventional da Vinci multiport systems, rather than isolating the specific SP versus Xi comparison that clinicians confront when selecting between the two widely deployed da Vinci configurations [7].

Accordingly, a focused quantitative synthesis is required to guide platform selection, perioperative pathway design, and patient counseling in the context of localized disease. Contemporary appraisals have framed SP RARP as a potential paradigm shift and have summarized evolving technical refinements alongside early outcome signals; however, none have provided a dedicated meta-analytic estimate directly contrasting da Vinci SP with da Vinci Xi multiport prostatectomy [8, 9]. Related reduced-port innovations, including single-site multiport techniques, further underscore the diversity of “less invasive” robotic strategies and reinforce the necessity of distinguishing device-specific effects from those attributable to access route and perioperative pathway design [10]. Therefore, we conducted the first systematic review and meta-analysis specifically comparing da Vinci SP with da Vinci Xi multiport RARP for the treatment of localized prostate cancer, uniquely incorporating prespecified sensitivity analyses, subgroup analyses, and a formal GRADE certainty of evidence assessment to strengthen clinical interpretability and facilitate informed decision-making.

Methods

Study design

This systematic review and meta-analysis aimed to compare the perioperative, oncologic, and functional outcomes of da Vinci^® single-port (SP) and da Vinci^® Xi multiport (MP) robot-assisted radical prostatectomy (RARP) in patients with localized prostate cancer. The methodology followed standard systematic review principles with pre-specified eligibility criteria, outcomes, and analytic plans to limit selection and reporting biases. This systematic review and meta-analysis was prospectively registered in PROSPERO (CRD420261299189).

Literature search strategy

A comprehensive search of major biomedical databases, including MEDLINE (via PubMed), Embase, Scopus, Web of Science, the Cochrane Central Register of Controlled Trials (CENTRAL), and ClinicalTrials.gov, was performed from inception to the 1st February, 2026. The search was supplemented by manual screening of the reference lists of eligible studies and relevant review articles. The search terms combined controlled vocabulary (where applicable) and free-text keywords related to radical prostatectomy, robotic surgery, single-port approaches/platforms, and multiport/Xi comparisons. Grey literature sources and preprint servers were not formally searched; however, trial registries were searched via ClinicalTrials.gov (n = 6 records identified). The decision not to search grey literature or preprint servers was prespecified, reflecting the low likelihood of unpublished platform-specific comparative cohorts meeting our inclusion criteria, and is acknowledged as such. The complete electronic search strategies and database-specific yields are summarized in Supplementary Table 1, which was used to document the reproducibility of the search and the flow of records into the screening (Supplementary Table 1).

Inclusion criteria

We included comparative studies enrolling adult patients with localized prostate cancer undergoing RARP, in which the da Vinci^® SP approach (including extraperitoneal or transvesical SP strategies when performed using the SP platform) was directly compared with the da Vinci^® Xi multiport prostatectomy. Eligible designs included comparative cohorts (prospective or retrospective), including propensity score–matched analyses. Studies were required to report extractable group-level data for at least one prespecified perioperative, oncologic, or functional endpoints.

Exclusion criteria

We excluded non-comparative case series, single-arm feasibility reports, conference abstracts without sufficient extractable data, editorials/letters, narrative reviews, and studies that did not report separable SP and Xi multiport outcomes. Studies focusing on non-prostate cancer indications, irretrievable duplicate cohorts (where the most complete dataset was retained), or comparisons not isolating SP from Xi multiport prostatectomy were excluded from the study.

To assess the risk of overlapping patient cohorts, we reviewed the institutional affiliations, recruitment periods, surgical approaches, and patient numbers of all included studies. Chavali 2024 and Soputro 2025 both originate from the same institution (Cleveland Clinic, Glickman Urological Institute), the same single experienced robotic surgeon, and overlapping recruitment periods (2015–2023 and 2020–2024, respectively). Chavali 2024 enrolled a mixed SP-EP/SP-TV cohort (SP-TV, n = 226), whereas Soputro 2025 enrolled SP-TV patients exclusively (n = 285 after matching). Given the overlapping time periods and shared institutional databases, the SP-TV patients in Chavali 2024 likely represent a subset of those in Soputro 2025. Similarly, the MP transperitoneal control arms in both studies were drawn from the same institutional database and surgeon. Both studies were retained in the primary analysis because they partially addressed distinct research questions: Chavali 2024 compared a mixed EP/TV approach versus MP, while Soputro 2025 isolated the TV approach versus MP, and propensity matching in each study produced non-identical matched cohorts.

Data extraction and outcome measures

Two reviewers (W.M. and M.B.A.) independently screened the titles/abstracts, evaluated the full texts for eligibility in an unblinded manner, and extracted the data using a standardized data collection form. Software tools were not used in the screening process. Disagreements were resolved through discussion and arbitration by a third author (M.A.). Where primary studies reported outcomes in a format incompatible with meta-analytic pooling (e.g., medians without ranges or graphical data), no attempts were made to contact study authors for missing or unclear data. Instead, validated conversion methods were applied where feasible, or the specific outcome was excluded from the pooled analysis where conversion was not possible. The extracted variables included the study design, setting, eligibility criteria, surgical approach descriptors (extraperitoneal, transperitoneal, and transvesical), baseline characteristics, and outcomes. The prespecified outcomes included operative time (total/console where available), estimated blood loss, length of stay, catheter duration (when reported), postoperative pain/opioid outcomes (when reported), complications, positive surgical margins, biochemical recurrence measures (timepoint-specific or time-to-event where available), and functional recovery (urinary continence and erectile function measures, as defined by each study). All prespecified outcomes were extracted at all time points reported in primary studies. Where studies used divergent definitions for the same outcome (e.g., urinary continence defined as 0 pads/day versus ≤ 1 pad/day), the most commonly used definition across studies was selected as the primary pooled estimate, and sensitivity was explored by restricting analyses to studies using equivalent definitions, where feasible. No post-hoc selective extraction of time points or subgroups was performed.

Quality assessment

The risk of bias for non-randomized comparative studies was assessed using ROBINS-I, which evaluates confounding, participant selection, intervention classification, deviations from intended interventions, missing outcome data, outcome measurement and selective reporting. Assessments were performed independently with consensus resolution, and risk of bias judgments were incorporated into the interpretation and certainty of the evidence grading.

Statistical analysis

Meta-analyses were performed using RevMan 5.4 review manager random-effects models to account for the anticipated clinical and methodological heterogeneity across cohorts (e.g., port strategy, access route, selection criteria, and perioperative pathways). For dichotomous outcomes, effect sizes were summarized as risk ratios (RRs) with 95% confidence intervals (CIs), and for continuous outcomes, mean differences (MDs) with 95% CIs were used. When studies reported medians with interquartile ranges (IQR), data were converted to approximate means and standard deviations using the method described by Wan et al. for samples reporting median, minimum, and maximum, or the method by Luo et al. for samples reporting median and IQR only, depending on the available summary statistics in each study [11, 12]. Statistical heterogeneity was quantified using I² and τ², and the influence of heterogeneity was explored through pre-specified sensitivity and subgroup analyses. Outcomes with very high or extreme statistical heterogeneity (I²≥75%) were interpreted with caution. Random-effects pooling was retained for all outcomes, including those with I²>90%, as this model is more conservative and produces wider confidence intervals that better reflect between-study uncertainty than fixed-effects models would. For outcomes with I²>90%, such as estimated blood loss (I²=98%), lymph node dissection rate (I²=99%), and urinary continence at approximately 3 months (I²=92%), pooled estimates were presented as exploratory and hypothesis-generating only, and were supplemented by narrative descriptions of individual study findings to allow independent assessment of directionality and magnitude. Meta-regression was not performed because the number of included studies was limited (n = 5), which would make covariate-based exploration of heterogeneity statistically unstable and prone to over-interpretation. Formal assessment of small-study effects and reporting bias using funnel plot asymmetry was not performed for any outcome, as the minimum threshold of ten studies per outcome required for reliable detection of asymmetry was not met in this review. This limitation is acknowledged, and the possibility of publication bias cannot be excluded from consideration.

Sensitivity analysis, subgroup analysis, and GRADE certainty of evidence

Sensitivity analyses will be performed using a leave-one-out approach to evaluate the robustness of the pooled estimates and identify influential studies. A prespecified clinical subgroup analysis will be conducted according to the surgical access corridor of the single-port arm, separating studies that used an off-peritoneal SP technique (SP-EP or SP-TV) from those that used an intraperitoneal/transperitoneal SP technique (SP-TP), because peritoneal violation and bowel handling can modify postoperative pain, ileus physiology, recovery of diet/bowel function, opioid requirements, and length of stay. This yields a 3-study subgroup (off-peritoneal SP: Chavali 2024 [SP-EP], Ko 2024 [SP-EP], Soputro 2025 [SP-TV]) and a 2-study subgroup (transperitoneal SP: Moschovas 2023 [SP-TP], Vigneswaran 2020 [SP-TP]). Subgroup analyses will be performed only for outcomes with sufficient reporting across studies to avoid unstable results. The certainty of evidence for each outcome will be assessed using GRADE, considering the risk of bias, inconsistency, indirectness, imprecision, and publication bias, and rated as high, moderate, low, or very low.

Results

As shown in Figs. 1 and 782 records were identified through database searching and 12 additional records through citation searching, with no grey literature records included. After the removal of 264 duplicates ( 0 duplicates identified manually, 0 records marked ineligible by automation tools, and 0 removed for other reasons), 530 records underwent title/abstract screening. Thirty full texts were assessed for eligibility, with no reports unretrieved, and 25 were excluded for incorrect comparators, incorrect outcomes, or incorrect study design. Ultimately, five studies were included in the review (Fig. 1).

Fig. 1.

PRISMA 2020 flow diagram showing identification, screening, eligibility assessment, and final inclusion of studies comparing da Vinci^® single-port with da Vinci^® Xi multiport robot-assisted radical prostatectomy

Characteristics of included studies

As shown in Table 1, five comparative cohorts were included, with four propensity-matched analyses to improve the baseline balance. Most of the evidence is from the USA, with one matched cohort study from South Korea. The SP approach varied across studies (extraperitoneal and/or transvesical). Simultaneously, the controls were consistently multiport Xi transperitoneal RARP/RALP, with outcomes spanning perioperative recovery, pathology/oncology, and functional recovery (Table 1).

Table 1.

Study characteristics of the included comparative studies (da Vinci^® SP vs. da Vinci^® Xi multiport RARP)

Study (ID)	Country	Experimental group (n)	Control group (n)	Inclusion criteria	Exclusion criteria	Primary endpoints	Secondary endpoints
Chavali 2024	USA	SP extraperitoneal RARP (EP ± TV), 485 (propensity-matched)	MP transperitoneal RARP (Xi), 121 (propensity-matched)	Primary RARP cases; SP vs. MP; database cohort (2015–2023)	Salvage prostatectomy	Not prespecified; perioperative recovery comparisons	Pathology/oncology (PSM, BCR); function (continence, SHIM)
Ko 2024	South Korea	SP extraperitoneal RARP, 50	MP transperitoneal RARP (Xi), 50 (1:1 matched)	Localized PCa; staging workup completed	Prostate > 50 g; suspected rectal invasion; radiologic N+	Operative time metrics; early recovery	Continence outcomes: bowel recovery (diet/gas)
Moschovas 2023	USA	SP RARP, 100 (1:1 PS matched)	Xi RARP, 100 (1:1 PS matched)	Consecutive SP RARP; matched Xi cohort (2019–2021)	Not specified	Postop pain (6 h, 12 h, 18 h)	EBL, operative time, complications, and pathology (including PSM)
Soputro 2025	USA	SP transvesical RARP, 285 (1:1 PS matched)	MP transperitoneal RARP (Xi), 285 (1:1 PS matched)	Consecutive SP-TV and MP-TP RARP; PS-matched cohort	Not specified	Not prespecified; perioperative outcome comparisons	Pathology/oncology (PSM, 12-mo BCR); function (continence; erectile function at 12 mo)
Vigneswaran 2020	USA	SP RARP, 50	MP RARP (Xi), 113	Localized PCa; consecutive SP vs. contemporary MP cohort	Combined surgeries; unreliable pain reporting	Pain outcomes (POD1 pain-free/opioid use); LOS	Time metrics; PSM; complications; 90-d stress incontinence

Abbreviations: BCR biochemical recurrence, EBL estimated blood loss, EP extraperitoneal, LOS length of stay, MP multiport, N+ node-positive, PCa prostate cancer POD postoperative day, PS propensity score, PSM positive surgical margin, RARP robot-assisted radical prostatectomy, SHIM Sexual Health Inventory for Men, SP single-port, TP transperitoneal, TV transvesical, Xi da Vinci Xi

Across the included cohorts, the baseline age was generally in the early to-mid 60s, except for Ko 2024, which enrolled an older population. Where clinical staging was available, most patients had ≤T2 disease; however, a substantial fraction had ≥pT3a disease on pathology in both arms (Table 2). Soputro 2025 and Vigneswaran 2020 provided the most comprehensive baseline reporting (including ASA and clinical stage breakdowns), whereas ASA and staging fields were not consistently available across all studies. Overall, the experimental and control arms appear to be broadly comparable within studies, particularly in propensity-matched cohorts.

Table 2.

Baseline demographics and clinical characteristics of the included studies (da Vinci^® SP vs. da Vinci^® Xi multiport RARP)

Study (ID)	Country	Group	Robotic system	n	Age (years)	BMI (kg/m²)	PSA (ng/mL)	Prostate volume (mL)	Clinical stage ≤T2 (n)	Clinical stage > T2 (n)	Pathologic stage ≥pT3a (n)	Prior abdominal surgery (n)	ASA 1–2 (n)	ASA 3–4 (n)	ISUP ≤ 3 (n)	ISUP > 3 (n)
Chavali 2024	USA	Experimental	SP (SP-EP)	485	63.6 (6.74)	29.5 (5.1)	8.3 (8.4)	NR	484	1	246	201	NR	NR	423	62
Chavali 2024	USA	Control	MP-Xi	121	62.2 (7.3)	29.1 (4.5)	8.2 (6.2)	NR	120	1	66	38	NR	NR	99	22
Ko 2024	Korea	Experimental	SP (SP-EP)	50	69.8 (6.0)	NR	11.3 (10.3)	27.1 (8.5)	16	NR	16	NR	NR	NR	41	9
Ko 2024	Korea	Control	MP-Xi	50	68.7 (5.4)	NR	60.8 (160.1)	29.0 (13.8)	16	NR	16	NR	NR	NR	41	9
Moschovas 2023	USA	Experimental	SP (SP-TP)	100	62 (3.47)	25.4 (1.17)	5.75 (1.01)	48.25 (4.91)	NR	NR	NR	NR	NR	NR	96	4
Moschovas 2023	USA	Control	MP-Xi	100	63 (2.89)	25.3 (1.05)	5.45 (0.9)	51.75 (5.47)	NR	NR	NR	NR	NR	NR	96	4
Soputro 2025	USA	Experimental	SP (SP-EP)	285	63.5 (7.2)	28.4 (4.7)	6.4 (3.2)	37.1 (15.6)	283	2	129	142	84	202	272	13
Soputro 2025	USA	Control	MP-Xi	285	62.1 (7.4)	28.8 (4.5)	6.2 (3.1)	39.5 (15.6)	280	5	136	92	72	210	267	18
Vigneswaran 2020	USA	Experimental	SP (SP-TP)	50	63 (7.41)	30 (5.19)	7.2 (5.63)	53 (14.82)	50	0	20	11	28	22	40	10
Vigneswaran 2020	USA	Control	MP-Xi	113	62 (6.67)	28 (5.19)	8.0 (6.07)	44 (11.11)	113	0	50	30	73	40	85	28

Abbreviations: ASA American Society of Anesthesiologists, BMI body mass index, ISUP International Society of Urological Pathology grade group, MP multiport, NR not reported in the highlighted baseline table, PSA prostate-specific antigen, RARP robot-assisted radical prostatectomy, SP single-port, TP transperitoneal, EP extraperitoneal, Xi da Vinci Xi

Methodological quality and study integrity

As shown in Supplementary Figs. 1–2 and Supplementary Table 2, the overall risk of bias was judged to be moderate in four studies and serious in one study. Chavali 2024, Ko 2024, Moschovas 2023, and Soputro 2025 were judged to have a moderate overall risk of bias, primarily because all were non-randomized comparative cohorts with residual concerns related to confounding, participant selection, institutional case-selection practices, and variable completeness of functional follow-up. Vigneswaran 2020 was judged to have serious overall risk of bias because of greater concerns in the confounding and participant-selection domains, reflecting its nonrandomized design, lack of comparable adjustment to the propensity-matched cohorts, and potential influence of temporal adoption, learning-curve effects, and perioperative pathway differences.

Across all included studies, the intervention classification was judged to be at low risk because the SP and Xi/multiport robotic approaches were clearly defined. The domains of deviations from the intended interventions, missing outcome data, outcome measurement, and selection of the reported result were generally judged to be low to moderate, reflecting variations in follow-up duration, completeness of functional outcomes, and inconsistent reporting of longer-term oncologic and patient-centered endpoints. Detailed domain-level ROBINS-I judgments and supporting rationales for each included study are provided in Supplementary Table 2, allowing verification of the study-level and overall risk of bias assessments.

Primary outcome

Positive surgical margin (PSM)

Across five comparative cohorts, PSM rates were similar between da Vinci® SP and da Vinci® Xi multiport RARP, with a pooled RR of 0.90 (95% CI: 0.75–1.09), indicating no statistically significant difference (P = 0.29) (Fig. 2). Heterogeneity was negligible (I² = 0%), supporting consistency across the included studies.

Fig. 2.

Forest plot comparing the risk of positive surgical margins after da Vinci^® single-port versus da Vinci^® Xi multiport RARP (random-effects model)

Secondary outcomes

Operative time (minutes)

Operative time was significantly longer with da Vinci® SP, with a pooled MD of 10.39 min (95% CI 0.72–20.07) (P = 0.04) (Fig. 3). Therefore, the direction of the effect favors the da Vinci® Xi robot/multiport for a shorter operative duration. However, heterogeneity was substantial (I² = 88%), suggesting meaningful across-study variation in the operative workflow and case mix.

Fig. 3.

Forest plot comparing operative time between da Vinci® single-port and da Vinci® Xi multiport RARP (random-effects model; mean difference)

Estimated blood loss (EBL; mL)

Estimated blood loss was significantly lower with the da Vinci® SP (pooled MD, − 68.95 mL (95% CI, − 128.27 to − 9.63) (P = 0.02) (Fig. 4). This indicates a statistically significant reduction in blood loss, favoring the single-port approach. Heterogeneity was very high (I² = 98%), implying strong between-study differences in measurement/reporting, and perioperative practices.

Fig. 4.

Forest plot comparing estimated blood loss after da Vinci^® single-port versus da Vinci^® Xi multiport RARP (random-effects model; mean difference)

Length of stay (LOS)

Length of stay was significantly shorter with da Vinci® SP, with a pooled MD of − 17.01 h (95% CI − 24.12 to − 9.89 h) (P < 0.00001) (Fig. 5). The effect direction consistently favored single-port surgery for a faster discharge. Heterogeneity remained substantial (I² = 91%), consistent with the cross-study differences in discharge pathways and unit conventions.

Fig. 5.

Forest plot comparing postoperative length of stay between da Vinci^® single-port and da Vinci^® Xi multiport RARP (random-effects model; mean difference)

Lymph node dissection performed (LND rate)

The likelihood of LND being performed was lower with da Vinci® SP (pooled RR 0.49), but the confidence interval crossed unity (95% CI 0.20–1.23), and this finding was not statistically significant (P = 0.13) (Fig. 6). Heterogeneity was extreme (I² = 99%), indicating major differences in LND indications/extent and institutional practices among the studies.

Fig. 6.

Forest plot comparing the likelihood of lymph node dissection performed in da Vinci^® single-port versus da Vinci^® Xi multiport RARP (random-effects model; risk ratio)

Postoperative complications

Overall, postoperative complications were comparable between the approaches, with a pooled RR of 1.06 (95% CI 0.71–1.60) and no statistically significant difference (P = 0.77) (Fig. 7). Heterogeneity was low (I² = 0%), suggesting broadly consistent complication findings across the included studies, despite differences in definitions.

Fig. 7.

Forest plot comparing postoperative complication rates between da Vinci^® single-port and da Vinci^® Xi multiport RARP (random-effects model; risk ratio)

Urinary continence at ~ 3 months

Across three studies, the urinary continence at ~ 3 months was not significantly different between the da Vinci® SP and da Vinci® Xi/multiport RARP groups, with a pooled RR of 1.22 (95% CI 0.87–1.71) (P = 0.26) (Fig. 8). Although one study showed higher continence with SP, overall heterogeneity was considerable (I² = 92%), indicating substantial inconsistency across cohorts and follow-up definitions.

Fig. 8.

Forest plot comparing urinary continence rates at approximately 3 months after da Vinci^® single-port versus da Vinci^® Xi multiport RARP (random-effects model)

Across the included studies, the primary outcome (PSM) showed no statistically significant difference between the da Vinci® SP and da Vinci® Xi/multiport RARP groups (Fig. 2). Among the secondary outcomes, operative time was significantly longer with SP (favoring Xi/multiport) (Fig. 3), whereas EBL and LOS were significantly lower/shorter with SP (Figs. 4 and 5). The LND rate and postoperative complications were not significantly different (Figs. 6 and 7, respectively). Urinary continence at ~ 3 months was also not significantly different overall but showed high heterogeneity, which limited its interpretability (Fig. 8). Overall, continuous outcomes demonstrated substantial heterogeneity, whereas PSM and complications showed minimal heterogeneity, supporting greater consistency for these safety and oncologic endpoints

Sensitivity analysis

The leave-one-out sensitivity analyses (Supplementary Table 3) identified single influential studies that materially contributed to between-study heterogeneity for several endpoints. Removing Ko 2024 reduced heterogeneity for operative time (I² 88% to 58%) while preserving statistical significance and strengthening the pooled estimate. For LOS, excluding Vigneswaran 2020 markedly reduced heterogeneity (I²: 91% to 37%), with a larger magnitude of reduction favoring the single-port group. For the LND rate, heterogeneity was eliminated (I ² 99% to 0%), and the pooled effect shifted to a statistically significant reduction in LND use with single-port, excluding Soputro 2025. In contrast, for EBL and ~ 3-month continence, excluding Soputro 2025 only partially reduced heterogeneity and did not yield a more consistent direction of effect for continence (Supplementary Table 3),

Subgroup analysis

In the transperitoneal SP access (SP-TP) subgroup (Moschovas 2023; Vigneswaran 2020), pooled effects showed no difference in PSM between platforms (RR 0.99, 95% CI 0.71–1.40; I²=0%), indicating that when SP is performed via a transperitoneal corridor, margin positivity appears comparable to Xi in the available studies. In the off-peritoneal SP access (SP-EP/SP-TV) subgroup (Chavali 2024; Ko 2024; Soputro 2025), the pooled estimate again demonstrated no statistically significant difference (RR 0.87, 95% CI 0.69–1.09; I²=0%), although the point estimate tended toward fewer PSMs with SP. Importantly, the test for subgroup differences was not significant (Chi² 0.44, df = 1; p = 0.51), supporting the lack of evidence of effect modification by the SP access corridor on PSM outcomes. Overall, combining all studies, the pooled effect remained non-significant (RR 0.90, 95% CI 0.75–1.09; I²=0%), suggesting that across surgical corridors, PSM rates were broadly similar between da Vinci SP and da Vinci Xi in the included comparative cohorts (Supplementary Fig. 3).

GRADE certainty of evidence

Across outcomes, certainty was very low (Supplementary Table 4) owing to the nonrandomized evidence base (risk of bias downgraded) and frequent between-study inconsistency for several continuous endpoints, with additional downgrading for imprecision, where confidence intervals crossed no effect (e.g., PSM, LND, complications, and continence). Nevertheless, the pre-sensitivity pooled effects suggested reductions favoring SP for operative time, EBL, and LOS (MDs <0 for EBL/LOS, MD >0 for operative time depending on direction conventions), while dichotomous outcomes showed no definitive differences for PSM, complications, LND rate, or ~3-month continence (CIs crossing unity) (Supplementary Table 4).

Discussion

Across five comparative cohorts, this meta-analysis found that da Vinci single-port radical prostatectomy yielded postoperative recovery advantages over da Vinci Xi multiport surgery without a clear oncologic trade-off in early margin control [1]. The pooled estimate for positive surgical margins did not demonstrate a statistically significant difference between platforms, and the absolute impact was modest when anchored to the control event rate (approximately 253 per 1,000 in the Xi arm), corresponding to approximately 25 fewer margins per 1,000 patients with single-port surgery under the pooled relative effect [2]. In contrast, several secondary recovery and resource utilization outcomes favored the single-port approach, aligning with the concept that minimizing peritoneal violation and abdominal wall trauma can translate into earlier functional milestones [3]. These signals refine and contextualize the rapidly expanding platform literature by separating outcomes that are plausibly technique-sensitive (pain, bowel recovery, discharge readiness) from those more likely to be dominated by tumor biology and surgical judgment (margin status and nodal staging strategy) [4].

The principal clinical interpretation is that single-port technology appears to function as a recovery-enabling strategy rather than an oncologic intensification strategy, and its value proposition is strongest when paired with extraperitoneal access corridors [5]. In our dataset, early continence recovery at approximately three months was consistently directionally favorable for single-port surgery, where reported, and using the pooled control baseline (approximately 496 continent patients per 1,000 in the Xi arm), the relative effect translated to approximately 109 additional continent patients per 1,000 at this early time point [6]. Confounding by indication represents the most important threat to the internal validity of this body of evidence. Lymph node dissection was performed less frequently in SP arms across the included cohorts; however, this differential is most plausibly explained by systematic differences in patient selection, disease-risk stratification, and institutional staging thresholds rather than an inherent technical limitation of the SP platform. In centers early in SP adoption, surgeons may preferentially select lower-risk patients (lower PSA, lower ISUP grade, smaller prostate volume, absence of suspicious nodal disease on imaging) for the SP approach, particularly when the extraperitoneal or transvesical corridor limits pelvic access compared with a transperitoneal multiport dissection [7]. This selection effect, which operates at the level of the referring clinician and the operating surgeon, is only partially mitigated by propensity score matching on observed covariates, as unobserved determinants of staging intent, such as surgeon comfort with extended pelvic lymph node dissection via a single-port corridor, remain as residual confounders. Moreover, the surgical corridor itself introduces confounding factors: extraperitoneal and transvesical SP approaches provide a different anatomical workspace for lymphadenectomy than transperitoneal multiport surgery, and any effect on LND yield or extent may reflect access geometry rather than surgeon intent or platform capability. For decision-making, this means that any apparent efficiency in recovery should be interpreted alongside staging intent, as lower dissection rates can reflect case mix and practice patterns rather than platform-specific limitations [8].

Mechanistically, the clinical advantages most plausibly arise from access corridor-dependent physiology rather than the number of skin incisions alone, particularly when extraperitoneal or transvesical routes reduce bowel manipulation and peritoneal inflammation [9]. Extraperitoneal single-port approaches are designed to preserve peritoneal integrity and limit ileus-prone stimuli, which can reasonably explain the observed direction of effect on the length of stay and early postoperative recovery endpoints [10]. The broader urologic literature has repeatedly emphasized that the single-port ecosystem does not represent a single technique but rather constitutes a family of approaches, and outcomes are therefore a composite of platform characteristics, corridor selection, patient selection, and surgeon learning effects [13]. Patient-centered endpoints, such as satisfaction and convalescence, may therefore diverge from traditional perioperative metrics, reinforcing the importance of standardized reporting frameworks that capture both functional recovery and the experience of care [14].

Our findings also clarify why prior pooled analyses often reported heterogeneous conclusions, as earlier syntheses frequently combined platforms, corridors, and study designs, thereby diluting corridor-specific signals [15]. Updated meta-analyses that aggregate single-port and multiport radical prostatectomy across heterogeneous comparators have typically suggested comparable safety with potential advantages in selected recovery endpoints, which is directionally concordant with our estimates while underscoring that effect sizes depend heavily on the corridors and eras included [16]. Technical evolution studies have highlighted that as surgeons transition across robotic generations, workflow and efficiency changes can occur independent of platform characteristics, creating confounding factors that can masquerade as a device effect unless they are rigorously controlled [17]. This is particularly relevant for operative time and blood loss, where learning curve dynamics, team familiarity, docking workflow, and perioperative pathways can shift substantially during the adoption phase [18].

The expanding ecosystem of non-Da Vinci single-port systems and single-site multiport variants further complicates inference, as device-specific ergonomics, instrument articulation, and assistant roles can influence both outcomes and the surgeon’s workload [19]. Systematic comparisons of single-site and multiport constructs suggest that a reduced port burden may not uniformly translate into superior outcomes, and the net clinical effect depends on how the approach reshapes exposure, traction, and reconstructive steps [20]. From a technology translation standpoint, this is important because early feasibility series of novel single-port systems demonstrate procedural viability but offer limited generalizability and sparse long-term oncologic follow-up [21]. Even within the da Vinci ecosystem, the literature emphasizes the rapid diffusion of single-port applications across urologic practice; however, consistent reporting of functional and oncologic endpoints remains uneven, limiting the certainty with which platform-wide claims can be made [22].

The substantial-to-extreme heterogeneity observed for several secondary outcomes means that these pooled estimates should be interpreted as exploratory and hypothesis-generating, rather than definitive. Heterogeneity in our pooled estimates should be interpreted by distinguishing the direction from the magnitude, as the direction of the effect for recovery endpoints was generally consistent, whereas the magnitude varied across cohorts [23]. Leave-one-out sensitivity analyses reduced statistical heterogeneity for the selected outcomes, indicating that one or two influential cohorts can meaningfully shift the pooled dispersion without necessarily reversing the clinical signal [24]. A plausible clinical explanation is that corridor choice (extraperitoneal or transvesical versus transperitoneal), discharge pathways, and institutional enhanced recovery protocols are effect modifiers that alter the magnitude of benefit in terms of length of stay, diet advancement, and opioid exposure [25]. Additionally, differences in surgeon and team workload between single-port and multiport platforms may influence efficiency and error risk in ways not captured by standard complication reporting, contributing to unexplained heterogeneity in operative metrics [26].

The certainty of evidence is constrained primarily by the study design, as comparative platform data are predominantly observational and often rely on propensity matching, which cannot eliminate unmeasured confounding [27]. From a GRADE perspective, we downgraded the risk of bias due to residual confounding and selection mechanisms, and we downgraded the inconsistency for outcomes with high I², where corridor and pathway variation likely introduced genuine between-study differences [28]. We also downgraded for imprecision, where confidence intervals encompassed clinically meaningful benefits and harms, a common limitation when only a small number of cohorts report a given endpoint in a compatible format [29]. Functional outcomes should be interpreted with particular caution because continence was primarily assessed at early follow-up (approximately 3 months), and longer-term functional recovery data were limited and inconsistently reported. Therefore, while the directionality of recovery advantages appears robust, the overall certainty is best characterized as low to moderate depending on the outcome, and conclusions should be framed as probabilistic rather than definitive [30].

In practice, these data support a nuanced corridor-informed approach to platform selection rather than a universal conclusion regarding single-port superiority [31]. Patients who are most likely to benefit clinically are those in whom minimizing bowel manipulation and facilitating early discharge are the most important, such as those prioritizing outpatient or short-stay pathways, those with prior abdominal surgery where extraperitoneal access may be advantageous, and those for whom early functional recovery is a central preference [32]. Conversely, patients in whom extensive nodal staging is anticipated, or where complex anatomy mandates a familiar multiport workflow, may derive less incremental benefit from a single-port strategy and may be better served by individualized platform selection based on the surgeon’s expertise [33, 34]. Future trials and prospective registries should be stratified by corridor, use standardized functional and patient-reported outcome sets, prespecify discharge and analgesia pathways, and ensure comparable staging. Only then can the field determine whether recovery gains translate into durable quality of life or oncologic equivalence across the risk strata.

Strengths and limitations

Strengths

Strengths include a tightly defined platform comparison, inclusion of directly comparative cohorts (with propensity adjustment where available), and clinically anchored analytic safeguards: random-effects pooling, prespecified leave-one-out sensitivity analyses to address influential studies driving heterogeneity, corridor-based subgroup analysis grounded in recovery physiology, and GRADE-informed certainty appraisal to enhance clinical interpretability.

Limitations

The two most fundamental constraints on inference in this review are the small number of included studies (n=5 comparative cohorts) and the complete absence of randomized controlled trial evidence. With only five cohorts contributing to each pooled estimate, and in some cases fewer, individual studies exert a disproportionate influence on the pooled results. The absence of randomized evidence means that even the most rigorous propensity score matching cannot fully eliminate confounding by indication. Unmeasured factors such as surgeon preference, patient risk tolerance, and institutional case-selection thresholds remain residual confounders that matching on observed variables cannot address. Additional limitations include the predominance of single-surgeon, single-institution designs across most of the included cohorts (Chavali 2024, Soputro 2025, Ko 2024, Moschovas 2023), which limits the generalizability of the results to broader surgical practice. Potential cohort overlap between Chavali 2024 and Soputro 2025 (both from the Cleveland Clinic database) was identified. One included study (Ko 2024) used the da Vinci X rather than the da Vinci Xi as the multiport comparator, introducing a degree of device-level heterogeneity within the control arm. Length of stay was reported in hours by some studies (Chavali 2024, Soputro 2025) and in days by others (Ko 2024, Vigneswaran 2020), with Moschovas 2023 not reporting LOS at all, values were converted to a common unit prior to pooling, but this cross-study inconsistency in reporting contributed to the extreme heterogeneity observed (I²=91%). Variable definitions for functional outcomes further limited cross-study comparability; continence was defined as zero pads (Moschovas 2023), single pad (Ko 2024, Vigneswaran 2020), or zero-to-one pad (Chavali 2024, Soputro 2025), and follow-up timing ranged from 90 days to 6 months. Substantial to extreme heterogeneity for several secondary outcomes, such as estimated blood loss, lymph node dissection rate, length of stay, and urinary continence, further limited the robustness of pooled estimates. Therefore, these findings should be considered exploratory and hypothesis-generating rather than definitive.

Conclusions

In this meta-analysis of low-to very-low-certainty evidence, da Vinci® SP RARP demonstrated comparable PSM rates and complication risks relative to da Vinci® Xi. Directional recovery advantages were observed for EBL and LOS, but extreme heterogeneity for both outcomes (I²=98% and 91%) limited their reliability. Lower LND rates in the SP cohorts reflect case-selection differences rather than platform limitations. Early continence was directionally favorable for SP in two of the three studies, but was not significant overall and was highly heterogeneous. Given the absence of randomized evidence and low to very low GRADE certainty across all outcomes, SP-RARP should be considered a potentially safe alternative to Xi in experienced programs with appropriate patient selection, pending confirmation from prospective studies with harmonized outcome definitions and longer follow-up.

Abbreviations

ASA

American Society of Anesthesiologists

BMI

Body mass index

ISUP

International Society of Urological Pathology grade group

MP

Multiport

NR

Not reported in the highlighted baseline table

PSA

Prostate-specific antigen

RARP

Robot-assisted radical prostatectomy

SP

Single-port

TP

Transperitoneal

EP

Extraperitoneal

Xi

da Vinci Xi

Authors' contributions

W.M. conceived the study, designed the protocol, performed the literature search, and drafted the manuscript. M.B.A. and M.B.M.D. independently screened the records and extracted the data; disagreements were resolved by R.N.A. and W.M. conducted the statistical analyses and GRADE assessment with input from A.M.H. and M.A. All authors critically revised the manuscript for important intellectual content and approved the final version of the manuscript.

Funding

No external funding was received for this study.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.