Salvage lymph node dissection (sLND) has been proposed as a treatment option for management of prostate cancer nodal recurrence after primary treatment but a surgical template has not been identified yet. An extended template may be necessary for curative intents but at the cost of increased morbidity of the procedure. Aim of our multicentric study is to evaluate if positivity to presacral and common iliac nodes is a predictor of retroperitoneal involvement, finding an appropriate surgical template for sLND.

METHODS

We retrospectively analyzed the data of 26 men who underwent sLND after diagnosis of pelvic nodal recurrence at PET/CT scan and were previously treated with radical prostatectomy (n=24), radiotherapy (n=1) or brachitherapy (n=1). All sLND included pelvic and retroperitoneal fields, irrespectively of PET results. Dissected nodes were stratified into 3 anatomical regions: A (internal iliac, obturatory, external iliac), B (presacral, common iliac), C (retroperitoneal). Chi square test was used on crosstabs; regression analysis was run.

RESULTS

Twenty men (76.9%) had positive pelvic nodes at sLND: 15 (57.6%) in region A and 15 (57.6%) in region B. Retroperitoneal involvement was found in 10 (38.4%). Mean number of dissected and positive nodes was 19.1 and 3.1 in the pelvic region (10.1 and 1.9 in region A, 8.9 and 1.1 in region B) and 7.8 and 1.0 in the retroperitoneal region. Positivity in region A+B was associated with retroperitoneal involvement in 45% of cases, while the negative predictive value (NPV) was 83.3%. Table 1 shows the crosstabs of nodal positivity stratified by anatomical regions. According to multivariate analysis, no pelvic region significantly predicts retroperitoneal involvement. The predictive ability does not improve with the number of dissected and positive nodes.

CONCLUSIONS

We were not able to find anatomical regions predictive of retroperitoneal involvement during sLND, probably due to the low number of patients enrolled. The high NPV of pelvic regions suggest that a threshold can be identified to select patients who need retroperitoneal LND. At the moment, an extended template including pelvic and retroperitoneal regions must be adopted to follow curative intents.