LAUSR

BACKGROUND Postchemotherapy retroperitoneal lymph node dissection (pcRPLND) is mandated in patients with nonseminomatous germ cell tumor found to have residual masses after chemotherapy. Performed via the open approach, pcRPLND can incur significant perioperative morbidity. OBJECTIVE To demonstrate the feasibility of robotic pcRPLND (r-pcRPLND) and provide evidence for its selection criteria. DESIGN, SETTING, AND PARTICIPANTS A retrospective search identified 93 patients undergoing pcRPLND between April 2007 and March 2018, comprising 30 r-pcRPLND and 63 open pcRPLND (o-pcRPLND) procedures performed by a single surgeon. INTERVENTION r-pcRPLND and o-pcRPLND. OUTCOME MEASUREMENTS AND STATISTICAL ANALYSIS Baseline clinicopathologic characteristics and intraoperative variables including operating room (OR) time, estimated blood loss (EBL), resection of adjacent organs, and intraoperative consultation with other surgical services were recorded. Hospital length of stay (LOS) and perioperative complications were assessed as per the Clavien-Dindo classification, and oncologic outcomes such as nodal yield, histologic distribution, pathologic staging, time to recurrence, and cancer-specific survival were compared. RESULTS AND LIMITATIONS r-pcRPLND was performed in a well-selected cohort with lower clinical stage (p=0.006), favorable International Germ Cell Cancer Collaborative Group classification (p=0.01), and smaller retroperitoneal mass (p=0.001). o-pcRPLND required more frequent bilateral template dissection (88.9% vs 43.3%; p<0.001), resection of adjacent organs (36.5% vs 10%; p=0.007), consultation with other surgical services (46% vs 2%; p<0.001), and auxiliary procedures (54.0% vs 20%; p=0.003) to achieve complete oncologic control. OR time was similar between the two groups (o-pcRPLND 375min vs r-pcRPLND 388min; p=0.16) and EBL was significantly lower in r-pcRPLND (234 vs 825ml; p<0.001). Median LOS was significantly shorter after r-pcRPLND (2 vs 7d; p<0.001). A total of 31 patients (33%) suffered postoperative complications, of whom 18 (19.4%) had major complications. Nodal yield was similar (o-pcRPLND 23 vs r-pcRPLND 24; p=0.8). The distribution of lesion histology (necrosis/teratoma/GCT) was also similar pcRPLND (o-pcRPLND 25.4%/57.1%/17.4% vs r-pcPLND 33.3%/50%/16.7%; p=0.51). Overall, tumor recurred in 15 patients (16.1%), including three following r-pcRPLND (10%), all outside the operative field. On univariate analysis, surgical approach was not a significant predictor of time to recurrence (p=0.34). One limitation was that antegrade ejaculation was not assessed. CONCLUSIONS With rigorous patient selection, r-pcRPLND can be safely performed and may reduce perioperative morbidity while maintaining oncologic proficiency. PATIENT SUMMARY Resection of residual retroperitoneal mass after chemotherapy in patients with metastatic testicular cancer can be performed safely via a robotic approach. Robotic surgery can reduce the morbidity of the procedure.