Robotic-assisted retroperitoneal lymph node dissection: defining selection parameters-a B-SAFE approach.

The objective of the study is to define selection parameters for RRPLND and evaluate the outcomes from cases selected via this method. Patients undergoing RRPLND from 2017 to 2023 (n = 57) were included. Initial criteria for robotic case selection were defined via 'B-SAFE' parameters. Safety was assessed via complication rate and oncological outcome. Analysis of both robotic and open RPLND outcomes including data from across the Anglican Germ Cell Cancer Collaborative Group was done. Mean lesion size was 30 mm (9-72). No cases required open conversion. Positive margin rate was 5.2%. Median length of stay (LOS) was 2 days (1-5). Overall complication rate was 15.7%. One patient required radiological intervention via embolization for a post-operative bleed. No in-field recurrences was observed at a median follow-up of 25 months (1-81). Analysis of parallel open RPLND cohort (n = 57) showed some differences in LOS (2 vs 6 [p =  < 0.05]) and bloods loss (130 vs 865 [p =  < 0.05]) likely explained by case complexity. Nodal yield higher in RRPLND (23 vs 10 [p =  < 0.05]). No significant difference in operation time (4.5 vs 4.6 [p = 0.5]), positive margins (5.2 vs 15.8% [p = 0.06]) or complication rates (15.7 vs 17% [p = 0.85]). This series proposed six parameters that can be used to appropriately select cases for RRPLND which we have defined using the 'B-SAFE' system. Our results using this framework are encouraging, with no instances of open conversion, excellent short-term oncological outcomes and no compromise of peri-operative morbidity with a short LOS. We also demonstrate an evolution in our practice towards more complex cases suggesting that as unit experience grows, initial selection criteria can be expanded to tackle more complex lesions.