Timo Soeterik

88 CHAPTER 5 METHODS Patients and data collection Patients diagnosed with prostate cancer undergoing robot-assisted radical prostatectomy (RARP) as primary treatment at 4 Dutch teaching hospitals (Martini Hospital Groningen, Hospital Group Twente, St. Antonius Hospital Nieuwegein/Utrecht and the Canisius Wilhelmina Hospital Nijmegen) from 2013 to 2018 were included in the study (IRB No. Z18.023). Data were captured in a prospective manner. Patients were excluded from analysis if they underwent salvage RARP or were treated with up-front androgen deprivation therapy. Baseline characteristics (age, clinical T-stage based on DRE, radiological T-stage based on multiparametric magnetic resonance imaging (mpMRI), preoperative serum PSA, total biopsy cores taken and number of positive cores at diagnosis, biopsy International Society of Urological Pathology (ISUP) grade, prostate volume measured using transrectal ultrasonography (TRUS) or mpMRI, treatment information (date of surgery, surgeon, nerve sparing as mentioned in the surgical report) and definitive pathology data (pathological T-stage, Gleason score, margin status) were documented. In addition, prostate side-specific radiological, surgical and pathological data were retrospectively collected. Predictor and outcome definitions The most recent preoperative PSA and prostate volume measured by TRUS or MRI were used to calculate PSAD (serum PSA [ng/ml] divided by prostate volume [ml]). DRE was subdivided into the 3 stages of T1 (benign), T2 (nodule) or T3 (EPE). All radical prostatectomies performed in the study period were robot-assisted. Interfascial nerve sparing was performed using an antegrade approach. After the upward traction of the vas deferens and seminal vesicles, the prostatic pedicle was observed and controlled athermally at the base of the prostate. Then the prostate was pulled to the opposite side and the lateral pelvic fascia was exposed. The triangular space between the lateral pelvic fascia, Denonvilliers’ fascia and the prostate was observed and the neurovascular bundle was defined. Subsequently, the lateral pelvic fascia was exposed and the interfascial dissection was performed. The non-nerve sparing technique included dissection posterior to Denonvilliers’ fascia and incision on to the perirectal fat lateral to the neurovascular bundles. RARP was performed by 14 surgeons. Surgical experience per surgeon varied from 0 procedures (least experienced) to 500 procedures (most experienced) at the beginning of the study period. For the analysis the most experienced surgeon was used as the reference category. Prostatic carcinoma was graded using the 2014 ISUP grading system. 11 A positive surgical margin, assessed by dedicated uropathologists, was defined as tumour cells present at the inked margin. 12