Introduction

Pelvic lymph node dissection (PLND) is the gold standard for LN staging in localized prostate cancer. The prevalence of LN involvement (LNI) is directly related to the number of dissected LNs or extent of PLND. Most published outcomes of PLND are based on single-center series. Unselected data on current PLND practice and influence of different surgical approaches on the performance of PLND are rare. The present study aims to assess the differential LN counts and rates of LNI detection depending on different PLND techniques by analysis of data from a multicenter German database.

Methods

We identified men in a German internet-based multicentre database (www.prostata-ca.net) who had open PLND, open radioisotope guided sentinel PLND (sPLND), robotic-assisted PLND (rPLND) or laparoscopic PLND (lPLND) and radical prostatectomy from 2005 to 2015. Differences in demographic characteristics, clinicopathological features, N-stage, LN counts and rates of LNI detection were examined using Mann-Whitney U-test or Pearson Chi-square test.

Results

A total of 6892 men (median age 67 years, median PSA 8ng/ml) from 17 hospitals met inclusion criteria. The patients were subjected to open PLND (59.4%), sPLND (28.2%), lPLND (11.2%) and rPLND (1.3%). sPLNDs and rPLNDs were carried out only by one center each. Except for a trend (p<0.0001) towards better differentiated Gleason-Scores in men with sPLND, there were no differences in patient characteristics comparing the groups. The median number of LNs removed was 10 (IQR 7-14) with no significant differences between PLND techniques: open PLND 11 (IQR 7-15), sPLND 10 (8-14), lPLND 10 (6-15) and rPLND 11 (8-13.5). Overall, 12.33% of patients had LNI. In sPLND, the LNI rate (18.9%) was significantly (p<0.001) higher than in open (10.26%), lPLND (7.27%) and rPLND (7.95%). In open PLND techniques, the rate of LNI was significantly higher (p<0.001) than in the two minimal invasive approaches. Overall, the median number of positive LNs was 2 (IQR 1-4) without significant difference between the techniques.

Conclusions

In current practice in Germany, using open PLND techniques results in a higher LNI rate than minimal invasive approaches. The use of targeted sPLND results in by far the highest rate of LNI despite better differentiated tumors. More LN+ patients were detected by sPLND than expected according to the data of single-center extended PLND series, too. These results should stimulate a reevaluation of patient selection or type of PLND.

Funding

none