Changing clinical trends in 10 000 robot-assisted laparoscopic prostatectomy patients and impact of the 2012 US Preventive Services Task Force's statement against PSA screening.

ABSTRACT

OBJECTIVES: To evaluate the clinical trend changes in our robot-assisted laparoscopic prostatectomy (RALP) practice and to investigate the effect of 2012 US Preventive Services Task Force (USPSTF) statement against PSA screening on these trends. PATIENTS AND METHODS: Data of 10 000 RALPs performed by a single surgeon between 2002 and 2017 were retrospectively analysed. Time trends in successive 1000 cases for clinical, surgical and pathological characteristics were analysed with linear and logistic regression. Time-trend changes before and after the USPSTF's statement were compared using a logistic regression model and likelihood-ratio test. RESULTS: Unfavourable cancer characteristics rate, including D'Amico high risk, pathological non-organ-confined disease and Gleason score ≥4+4 increased from 11.5% to 23.3%, 14% to 42.5%, and 7.7% to 20.9%, respectively, over time (all P < 0.001). Significant time-trend changes were detected after the USPSTF's statement with an increase in the positive trend of Gleason ≥4+4 and increase in the negative trends of Gleason ≤3+4 tumours. There was a significant negative trend in the rate of full nerve-sparing (NS) with a decrease from 59.3% to 35.7%, and a significant positive trend in partial NS with an increase from 15.8% to 62.5% over time (both P < 0.001). The time-trend slope in 'high-grade' partial NS significantly decreased and 'low-grade' partial NS significantly increased after the USPSTF's statement. The overall positive surgical margin rate increased from 14.6% to 20.3% in the first vs last 1000 cases (P < 0.001), with a significant positive slope after the USPSTF's statement. CONCLUSIONS: The proportion of high-risk patients increased in our series over time with a significant impact of the USPSTF's statement on pathological time trends. This stage migration resulted in decreased utilisation of high-quality NS and increased performance of poor-quality NS.