Prostatic and pelvic imaging parameters to predict post radical prostatectomy erectile function recovery: a systematic review.

With radical prostatectomy for prostate cancer, there may be associated long-term postoperative sequalae: urinary incontinence and erectile dysfunction (ED). It is important to predict the functional recovery of erections for better patient counselling and timely treatment of ED. This systematic review looks at imaging parameters to predict the recovery of erectile function (EF) after laparoscopic or robotic prostatectomy. A systematic search was performed to capture publications from January 2000 up to December 2023 (PROSPERO; Registrations ID CRD 42022359557). The considered studies applied an imaging parameter obtained by any form of imaging modality and in any operative phase (pre- or intraoperative) to assess the potential impact on EF after surgery. An essential criterion was a formal EF assessment at both baseline and postoperatively, by means of a validated questionnaire. A total of 8 studies met our inclusion criteria. We categorised the studies based on the imaging modality into three groups: MRI (n = 4), diffusion tensor imaging (DTI) (n = 2), and intraoperative ultrasonography (n = 2). Preoperative MRI parameters were, firstly, dynamic contrast enhancement of prostatic tissue measured as ratio of change of contrast from baseline to 120 s (Ratio120) showing weak correlation to postoperative IIEF5 scores (r = 0.31; p = 0.044), and secondly, area of neurovascular bundle (NVB) was a predictor of EF recovery on univariate analysis (odds ratio = 1.30; P = 0.001). Bony pelvic dimensions, prostate surface area, and fascial thickness measured on MRI did not show correlation with EF scores. Two studies using DTI were included, with one showing the change in the number of periprostatic nerve fibres before and after surgery, which also demonstrated a correlation with the change in EF scores (r = 0.35; P < 0.05). While the other study using DTI showed the change direction of the periprostatic nerve fibres before and after surgery measured as a value of functional anistropy, it showed a weak negative correlation to postoperative EF scores on the left (r = -0.66120; p = 0.0006) and right (r = -0.420068; p = 0.0456). Additionally, intraoperative ultrasound assessment of the NVB, such as the number of visible vessels within the NVB, has also been shown to correlate (r = 0.34, p = 0.0001) with postoperative EF. Our systematic review could not identify an imaging parameter strongly correlated with EF recovery. Enhancement of the prostate on mpMRI and intraoperative ultrasound might be worth investigating through better-designed studies. More research is needed to establish which parameters can reliably predict EF post-prostatectomy to best inform the patient and mitigate the risk.