Robotic, three-dimensional laparoscopic and two-dimensional laparoscopic pyeloplasty achieved success rates of 96.4%, 88.9% and 85.2% respectively in a retrospective cohort of 109 patients treated at a tertiary referral centre in India, a difference that was not statistically significant. The study, conducted at the Department of Urology of the All India Institute of Medical Sciences in Jodhpur and published in the African Journal of Urology, is one of the relatively few three-arm comparisons to treat 3D laparoscopy as a distinct intermediate technology between conventional laparoscopy and robotic surgery, and it comes from a resource-constrained, publicly funded setting where the cost gap between these platforms matters most.
All patients underwent a primary Anderson–Hynes dismembered pyeloplasty for ureteropelvic junction obstruction, a condition that blocks urine flow from the renal pelvis into the ureter and, if untreated, may cause progressive loss of renal function. Dismembered pyeloplasty is the reference standard repair, and minimally invasive access has largely replaced open surgery because it offers magnified visualisation, reduced analgesic requirement, faster recovery and shorter hospital stay. Between January 2018 and June 2022, a single team of three consultant urological surgeons operated on 27 patients with 2D laparoscopy, 27 with 3D laparoscopy and 55 with the da Vinci Xi robotic system, while the 3D procedures used an Olympus three-dimensional camera platform. All three surgeons had accumulated 40 to 85 laparoscopic and 25 to 45 robot-assisted pyeloplasties each before the study began, placing them beyond the reported learning curves for both techniques.
A crucial caveat shapes every conclusion of the study: the patients were not randomised. Group assignment reflected the modality actually used, determined in daily practice by the availability of the shared robotic platform on designated operating days, by patient preference after counselling, and by affordability, since robotic surgery carried an additional out-of-pocket cost at the institution. None of these factors was recorded as a structured field, so the authors describe the grouping as post hoc and expect it to have introduced selection bias. Because affordability gated access to the robot, socioeconomic status may be entangled with unmeasured prognostic factors, and robotic cases accrued disproportionately late in the study period, confounding secular improvements in peri-operative care with the robotic approach. The authors explicitly present all comparisons as associations rather than causal effects.
The six-month outcome was a composite requiring all three of the following: resolution of obstructive symptoms, an unobstructed drainage pattern on diuretic renography graded by the O’Reilly system, and preserved or improved differential renal function, defined as a fall of no more than 5 percentage points from baseline. Meeting any single failure criterion was enough to count as a failure. All 109 patients completed this protocol-defined assessment, and median follow-up reached 26 months. One hundred patients met all three criteria; there were nine failures, four in the 2D group, three in the 3D group and two in the robotic group. Notably, the three outcome domains agreed in every patient, and all nine failures had both an obstructed renographic curve and recurrent symptoms.
The statistical power behind that null result deserves close attention. With only nine failures in total, the study could have detected a difference of roughly 22 percentage points in success against the robotic group, and about 34 points between the two laparoscopic groups, with 80% power. The confidence intervals around the between-group differences are wide: the absolute difference between 2D and robotic success was −11.2 percentage points (95% CI −29.0 to +1.3), and between 3D and robotic −7.5 points (95% CI −24.6 to +3.8). Both intervals include differences that would be clinically important. The authors are unambiguous that this is a failure to show a difference, not a demonstration that the approaches are equivalent, and no equivalence margin was set in advance.
Where the approaches clearly diverged was in operative logistics. Nursing set-up time, which includes table set-up, patient positioning and robot docking, was 27.7 minutes longer in the robotic group than in the 2D group and 31.0 minutes longer than in the 3D group, both differences highly significant. Surgeon operative time was also longer with the robot, by 20.5 minutes against 2D and 27.4 minutes against 3D laparoscopy, and this association persisted after multivariable adjustment, with an adjusted mean difference of 21.3 minutes against 2D laparoscopy. Total operating-theatre time was 55.5 minutes longer in the robotic group than with 2D optics and 61.2 minutes longer than with 3D optics. There was no significant difference between the 2D and 3D groups in any of these timing measures.
Safety and recovery measures showed no significant between-group differences. Estimated blood loss was modest in all groups, one patient in the 2D group required a single unit of packed red cells, and hospital stay did not differ significantly. Complications within 30 days occurred in 20 of 109 patients, 18.3%, predominantly low-grade Clavien–Dindo grade I or II events such as self-limiting fever, transient haematuria and urinary tract infection. Four patients, 3.7%, sustained major complications of grade IIIa or above, and one 2D case was converted to open surgery because of dense adhesions. No grade IV or V complications occurred. The authors note that complications were graded retrospectively by investigators who were not blinded to modality, so minor events may have been under-ascertained, possibly unevenly across groups.
An exploratory multivariable logistic regression, restricted to three covariates because only nine outcome events were available, found no independent association between surgical approach and success after adjustment for pelvic diameter and the presence of a crossing vessel. The adjusted odds ratio for robotic versus 2D laparoscopy was 4.18 with a 95% confidence interval of 0.68 to 25.7, an interval so wide it is compatible with substantial benefit or harm. The authors label these estimates exploratory rather than confirmatory. A surgeon-rated ease-of-procedure score, recorded contemporaneously on a one-to-ten scale, was lowest for 2D laparoscopy, but the authors stress that this measure is subjective, unvalidated and single-rater, provided by a surgeon necessarily aware of the modality used, and should not on its own guide the choice of approach.
The fate of the nine failures offers a practical footnote: eight underwent formal redo pyeloplasty and one was managed conservatively, resolving clinically and radiologically by twelve months. Seven of the eight redo procedures succeeded on repeat renography, and the remaining patient underwent ureterocalicostomy with a satisfactory result, so all nine patients eventually achieved resolution, at the cost of nine additional procedures. The authors suggest that in settings like theirs, the choice of approach may reasonably be guided by equipment availability, cost and surgeon expertise, and that the role of 3D laparoscopy in resource-constrained environments merits formal evaluation.
What would settle the question is adequately powered, prospective randomised trials comparing all three modalities, with follow-up extending well beyond one year and pre-specified equivalence or non-inferiority margins, ideally multicentre and incorporating blinded renography adjudication, validated difficulty measures, patient-reported outcomes and formal cost-effectiveness analysis. Until such trials exist, this single-team cohort stands as a hypothesis-generating real-world snapshot: three minimally invasive routes to the same operation, broadly similar functional results that cannot be declared equivalent, and a robotic platform whose extra theatre time, in a shared public hospital, may be its most measurable signature.
Subject of Research: Retrospective comparison of robotic, three-dimensional and two-dimensional laparoscopic pyeloplasty outcomes
Article Title: Minimally invasive pyeloplasty at a tertiary care centre: a retrospective cohort study of robotic, three-dimensional and two-dimensional laparoscopic approaches
Article References: Soni, J. S., Tripathi, S., Singh, M., Bhirud, D. P., Navriya, S. C., Choudhary, G. R., & Sandhu, A. S. (2026). Minimally invasive pyeloplasty at a tertiary care centre: a retrospective cohort study of robotic, three-dimensional and two-dimensional laparoscopic approaches. African Journal of Urology, 32(1), Article 67. https://doi.org/10.1186/s12301-026-00615-0
Image Credits: AI Generated
DOI: 10.1186/s12301-026-00615-0
Keywords: pyeloplasty, ureteropelvic junction obstruction, robotic surgery, 3D laparoscopy, 2D laparoscopy, retrospective cohort, tertiary care, surgical outcomes, Clavien-Dindo complications, operative time, non-randomised study, India
News Source: Ophelia Keating. (October 29, 2026). Robotic, 3D and 2D pyeloplasty show no clear success difference in Indian cohort. Scienmag.
