The Association of Length of the Resected Membranous Urethra With Urinary Incontinence After Radical Prostatectomy Yasuo Kohjimoto, Shimpei Yamashita, Kazuro Kikkawa, Akinori Iba, Nagahide Matsumura, Isao Hara* Purpose: To retrospectively determine whether recovery of urinary continence after radical prostatectomy is as- sociated with the preoperative length of membranous urethra (MU), the amount of rhabdosphincter and the length of MU removed with the prostate. Materials and Methods: The study cohort comprised 179 consecutive patients who underwent laparoscopic rad- ical prostatectomy (LRP: n = 98) and robot-assisted radical prostatectomy (RARP: n = 81) at Wakayama Medical University between July 2010 and May 2014. The length of MU was measured by preoperative MRI. The amount of resected rhabdosphincter and the length of resected MU were assessed in hematoxylin and eosin sections at the apical margin of prostate specimens. Patient-reported urinary continence status was determined at 3, 6, 12 and 24 months postoperatively, with urinary continence considered as 0-1 pads/day. Kaplan-Meier analysis and the log- rank test were used to compare time to urinary continence recovery. Multivariate Cox regression analyses were performed to determine the predictors of urinary continence. Results: RARP vs LRP (p = 0.02) and shorter length of resected MU (p = 0.01) showed significantly better post- operative continence recovery by log-rank test. Nerve-sparing, preoperative length of MU, and amount of resected rhabdosphincter did not significantly correlate with continence recovery. Only the length of resected MU was the independent factor for predicting postoperative urinary continence by multivariate Cox regression analysis (hazard ratio 0.84, p = 0.01). Conclusion: These results demonstrated that the length of resected MU measured by specimen was an independent predictor of urinary incontinence after radical prostatectomy. Care should be taken to preserve maximal length of MU for optimal continence outcomes. Keywords: Membranous urethral length; Radical Prostatectomy; Urinary Incontinence INTRODUCTION Radical prostatectomy is a standard treatment option for localized prostate cancer. Although oncological outcome of radical prostatectomy is fairly satisfactory, postoperative complications including urinary incon- tinence and erectile dysfunction still remain problems to be solved(1). Among them, urinary incontinence is a nuisance which affects patient’s daily life and quality of life (QOL) directly. Numerous factors are involved in postoperative uri- nary continence recovery. Firstly, patient’s background including age, comorbidity, body mass index, history of prostate surgery and prostate size should be consid- ered(2-4). Second, surgical techniques of radical pros- tatectomy are other considerably important factors. Techniques for improving postoperative urinary incon- tinence are classified into 2 categories(5,6). One is a pres- ervation of the anatomical structure related to urinary continence including bladder neck, neurovascular bun- dle, puboprostatic ligament, rhabdosphincter and mem- branous urethra, and the other is a reconstrcution of the destroyed structure such as posterior reconstruction of Department of Urology, Wakayama Medical University, Wakayama, Japan. *Correspondence: Department of Urology, Wakayama Medical University 811-1 Kimiidera, Wakayama 641-0012, Japan. Tel: +81-73-441-0637; Fax: +81-73-444-8085. E-mail: hara@wakayama-med.ac.jp. Received August 2018 & Accepted January 2019 rhabdosphincter (Rocco’s stitch) and periurethral sus- pension stitch. Types of surgery are another concern, i.e. open radical prostatectomy (ORP), laparoscopic radical prostatectomy (LRP) or robot-assisted radical prostatectomy (RARP). Although recent systematic review and meta-analysis showed significantly better urinary continence rate at 12 month in RARP compared to RRP and LRP(7), clear conclusion has not been drawn yet. In this study, we investigate whether factors concerning membranous urethra (MU) and rhabdosphincter might influence postoperative urinary incontinence using pre- operative magnetic resonance imaging (MRI) and re- sected prostate specimens. PATIENTS AND METHODS Patients The cohort of this retrospective study comprised 179 consecutive patients who underwent LRP (n = 98) and RARP (n = 81) at Wakayama Medical University be- tween July 2010 and May 2014. Indications for radi- cal prostatectomy were as follows; (1) age under 75, UROLOGICAL ONCOLOGY Urology Journal/Vol 17 No. 2/ March-April 2020/ pp. 146-151. [DOI: 10.22037/uj.v0i0.4753] (2) PSA less than 20ng/ml and (3) cT1-2 or cT3a only when the extracapsular invasion was minimum. Demo- graphic data were collected from medical chart. This study was approved by the institutional review board of Wakayama Medical University (No. 1670). Surgical technique LRP and RARP were performed using a transperitoneal approach according to the standard techniques, which were previously described (8). Briefly after cutting en- dopelvic fascia, we transected bladder neck followed by the dissection of seminal vesicle and prostate in an antegrade fashion. Cavernous nerve-sparing was cho- sen according to the extent of cancer judged by MRI and digital rectal examination. Posterior rhabdosphinc- ter reconstruction and periurethral suspension stich was performed in RARP but not in LRP. Vesicourethral anastomosis was performed with a running suture. While limited lypmphadenectomy was performed in all patients in LRP, extended lymphadenectomy was per- formed only in high-risk patients in RARP. Patient reported outcome To evaluate the status of urinary continence, we used Expanded Prostate Cancer Index Composite (EPIC) questionnaire as patient reported outcome. The Jap- anese version of EPIC was purchased from iHope International (Kyoto, Japan) and validation study of Japanese version was reported previously(9). EPIC ques- tionnaire was mailed to each patient periodically (3, 6, 12 and 24 months after operation). Response rates of the questionnaire at 3, 6, 12 and 24 months were 95%, 93%, 87% and 82%, respectively. Urinary continence status was judged from the response to one of the ques- tions in urinary domain of EPIC and patients who wear 0 - 1 pad/day were considered continent. Measurement of the length of membranous urethra (MU) All patients had MRI examination before operation. Length of MU was measured on coronal T2-weighted image of preoperative MRI and determined as the dis- tance from prostatic apex to the entry of urethra into penile bulb (Figure 1, red arrow). Measurement of the amount of resected rhabdosphinc- ter and the length of resected MU A 3-4mm cone of tissue around the urethra at the apex of resected prostate was amputated and divided into left and right halves. Each half was serially sectioned per- pendicularly, typically resulting in 3 wedges of tissue from each half. These wedges were embedded in paraf- fin blocks and standard hematoxylin and eosin sections were made. The length of resected MU was measured in the cross section including urethra (Figure 2, red arrow). The amount of resected rhabdosphincter was expressed as the percentage of total apical margin sur- face area occupied by rhabdosphincter (Figure 2, blue arrow), and the overall percentage of resected rhabdos- phincter was calculated as average of typically 6 wedg- es. These measurements were assessed by two independ- ent uropathologists. When the inter-rater reliability was examined by Pearson correlation coefficient, it showed high agreement (r = 0.867, p < 0.0001). Statistical analyses Clinical data, including age at surgery, preoperative PSA, biopsy Gleason score, cT stage, nerve preserva- tion, length of MU, amount of resected rhabdosphinc- ter, and length of resected MU were compared between LRP and RARP groups by Wilcoxon signed-rank test for continuous variables and chi square test for categor- ical variables. Kaplan-Meier analysis and the log-rank test were used to compare time to urinary continence recovery and the cumulative incidence of urinary con- tinence at follow-up. The first response to EPIC report- Urinary incontinence after radical prostatectomy-Kohjimoto et al. Table 1. Patient characteristics stratified by surgery (LRP vs RARP) Total LRP RARP p = (LRP vs. RARP) No. patients (%) 179 98 (54.7) 81 (45.3) Age, year 67(63 – 71) 68 (63 – 72) 67 (64 – 71) 0.94 PSA, ng/mL 8.0 (5.9 – 11.0) 8.1 (5.9 – 12.3) 7.7 (5.9 – 10.1) 0.42 Biopsy Gleason score, n (%) 0.13 ≤ 6 50 (27.9) 28 (28.6) 22 (27.2) 7 74 (41.3) 41 (41.8) 33 (40.7) ≥ 8 55 (30.7) 29 (29.6) 26 (32.1) cT stage, n (%) 0.22 T1c 82 (45.8) 49 (50.0) 33 (40.7) T2 90 (50.3) 47 (48.0) 43 (53.1) T3a 7 (3.9) 2 (2.0) 5 (6.2) Nerve preservation, n (%) 80 (44.7) 18 (18.4) 62 (76.5) < 0.01 Length of MU 17.3 (14.6 – 19.7) 17.5 (14.9 – 19.3) 17.1 (14.5 – 19.8) 0.63 – preoperative MRI, mm Amount of resected rhabdosphincter 31.7 (25.0 – 40.0) 30.0 (25.0 – 40.0) 34.2 (25.8 – 42.1) 0.13 – pathology specimen, % Length of resected MU 1.2 (0.5 – 2.0) 1.1 (0 – 2.1) 1.2 (0.6 – 2.0) 0.42 – pathology specimen, mm Table2. Cox regression analysis of factors predictive of urinary continence recovery HR 95% CI p Age (year) 1.00 0.97 – 1.03 0.85 Type of surgery, RARP/LRP 1.37 0.92 – 2.03 0.11 Nerve preservation , Yes/No 1.01 0.67 – 1.51 0.95 Length of MU – preoperative MRI (mm) 1.04 0.99 – 1.09 0.06 Amount of resected rhabdosphincter – pathology (%) 0.99 0.98 – 1.00 0.57 Length of resected MU - pathology (mm) 0.84 0.73 – 0.97 0.01 Vol 17 No 02 March-April 2020 147 ing urinary continence recovery (0 – 1 pad/day) was considered the event at that time. Patients incontinent at their last response were censored at that time. Multi- variate Cox regression analyses were performed to de- termine the predictors of urinary continence recovery time. Clinically important or previously reported varia- bles (age, type of surgery, nerve preservation, length of MU, amount of resected rhabdosphincter and length of resected MU) were included in the model. Proportional hazards assumption was confirmed by plotting Schoen- feld residuals and the fitness of the model was estimat- ed by likelihood ratio test. In addition, multicollinearity of the model was assessed using variance inflation fac- tor. Data analyses were conducted by using the statis- tical software JMP Pro 12 (SAS Institute, Cary, USA). All p-values were 2 tailed, and p < 0.05 was defined as statistically significant. RESULTS Patients’ characteristics Baseline characteristics for all 179 patients are shown in Table 1. Although nerve preservation rate in RARP group (76.5%) was significantly higher than that in LRP group (18.4%), no significant differences were observed between LRP and RARP groups in other var- iables. Length of MU, amount of resected rhabdosphint- er and length of resected MU Table 1 shows length of MU, amount of resected rhab- dosphinter and length of resected MU. Median length of MU measured by preoperative MRI was 17.3mm (IQR, 14.6 – 19.7). Median amount of resected rhabdosphint- er and length of resected MU measured by pathology specimen was 31.7% and 1.2mm, respectively. No significant differences were found between LRP and RARP group in these variables. Status of urinary continence according to various var- iables Figure 3 shows Kaplan-Meier curve representing the percentage of patients achieving urinary continence recovery. The percentage of patients achieving urinary continence recovery after 3, 6, 12 and 24 months was 43.2%, 75.4%, 84.4% and 92.2%, respectively. Figure 4 (a) shows the same Kaplan-Meier curve ac- cording to the type of surgery. Continence rates after 12 months from surgery were 87.8% (RARP) and 81.6% (LRP). RARP showed significantly better continence recovery comparing to LRP (log-rank test, p = 0.02). Urological Oncology 148 Figure 1. Measurement of the length of membranous urethra (MU). Length of MU was measured on coronal T2-weighted image of preoperative MRI and determined as the distance from prostatic apex to the entry of urethra into penile bulb (red arrow). Figure 2. Measurement of the amount of resected rhabdosphincter and the length of resected membranous urethra (MU). The amount of resected rhabdosphincter was expressed as the percentage of to- tal apical margin surface area occupied by rhabdosphincter (blue arrow, 50% in this section) and the overall percentage of resected rhabdosphincter was calculated as average of typically 6 wedges. The length of resected MU was measured in the cross section in- cluding urethra (red arrow). Figure 3. Overall cumulative incidence of urinary continence, de- fined as 0-1 pad/day, following radical prostatectomy. Urinary incontinence after radical prostatectomy-Kohjimoto et al. Figure 4 (b) shows the same Kaplan-Meier curve ac- cording to nerve-sparing. Continence rates after 12 months from surgery were 83.7% (nerve-sparing (+)) and 84.9% (nerve-sparing (-)). No significant differ- ence was observed between nerve-sparing (+) and nerve-sparing (-) groups (log-rank test, p = 0.15). Figure5 (a) shows the same Kaplan-Meier curve ac- cording to the length of MU measured by preoperative MRI. Continence rates after 12 months from surgery were 86.5% (length of MU ≥ 17.3mm) and 81.7% (length of MU < 17.3mm). No significant difference was observed between longer (≥17.3mm) and shorter (< 17.3mm) MU groups (log-rank test, p = 0.20). Figure 5 (b) shows the same Kaplan-Meier curve ac- cording to the amount of resected rhabdosphincter. Continence rates after 12 months from surgery were 90.5% (resected rhabdosphincter < 31.7%) and 78.3% (resected rhabdosphincter ≥31.7%). No significant dif- ference was observed according to the amount of re- sected rhabdosphincter by pathology (log-rank test, p = 0.14). We also compared the apical positive surgical margin (PSM) rates according to the amount of resect- ed rhabdosphincter. Apical PSM rates were 10.3% and 16.3% in the resected rhabdosphincter < 31.7% group and ≥31.7% group, respectively. No statistical signifi- cance was observed (chi square test, p = 0.24). Fig. 5 (c) shows the same Kaplan-Meier curve accord- ing to the length of resected MU measured by resected specimen. Continence rates after 12 months from sur- gery were 88.5% (resected MU < 1.2mm) and 80.2% (resected MU ≥1.2mm). Longer resected MU group (≥ 1.2 mm) showed significantly worse continence re- covery comparing to shorter resected MU group (< 1.2 mm) (log-rank test, p = 0.01). We also compared apical PSM rates according to the length of resected MU. Api- cal PSM rates was 16.9% and 10.0% in the resected MU < 1.2 mm group and ≥1.2mm group, respectively. No statistical significance was observed (chi square test, p = 0.18). Multivariate analysis determining the factors in- fluencing urinary continence recovery Table 2 showed Cox regression analysis of factors pre- dicting urinary continence recovery. Only the length of resected MU measured by specimen was independent predictive factor for urinary continence recovery (haz- ard ratio 0.84, p = 0.01). The length of MU measured by preoperative MU seems to be marginal predictive factor for urinary continence recovery, however the difference did not reach statistically significant (p = 0.06). DISCUSSION Postoperative urinary incontinence is most annoying Figure 4. Cumulative incidence of urinary continence following radical prostatectomy stratified by (a) type of surgery (RARP vs LRP) and (b) nerve-sparing status. Figure 5. Cumulative incidence of urinary continence following radical prostatectomy stratified by (a) length of membranous urethra (MU) measured by preoperative MRI, (b) amount of resected rhabdoshpincter measured by pathology specimen, and (c) length of resected MU measured by pathology specimen. Urinary incontinence after radical prostatectomy-Kohjimoto et al. Vol 17 No 02 March-April 2020 149 complication after radical prostatectomy. Recently, the anatomical mechanism regarding urinary continence in male has been gradually revealed by many vigorous studies(10-13). Basically, the anatomical mechanism for urinary continence consists of 2 systems; a sphincter- ic system and a supportive system(5). The sphincter- ic system is composed of inner smooth muscle layer (longitudinal and circular smooth muscle) and striated urogenital sphincter muscle (rhabdosphincter). Rhab- dosphincter extends from the prostatic apex to the prox- imal bulbar urethra and considered the most important structure for urinary continence after radical prostatec- tomy. Strasser et al. showed that the contractility of the remaining rhabdosphincter after transurethral resection of prostate or radical prostatectomy was associated with postoperative urinary incontinence by 3-dimensional ultrasound(14). The other system is supportive system of pelvic floor surrounding MU and bladder neck. In males, Denonvilliers’ fascia, puboprostatic ligament, endopelvic fascia, levator ani muscle and arcus tendi- nous fascia pelvis correspond to this system. Radical prostatectomy impairs this supportive system inevita- bly. In this study, we focused on the sphincteric system in urinary incontinence after LRP and RARP since the sphincteric system is more directly controlling urinary continence than the supportive system is. We selected several measurements concerning the sphincteric sys- tem. First one was preoperative length of MU measured by MRI. Paparel et al. demonstrated that both preop- erative and postoperative membranous urethral length measured by preoperative and postoperative MRI were correlated with urinary continence status after radical prostatectomy(15). Unfortunately, we evaluated only preoperative membranous urethral length since we did not examine postoperative MRI. Second, we measured the amount of resected rhabdosphincter histologically using resected prostate specimens. Such evaluation was conducted by Skeldon et al. first(16). They measured the amount of striated muscle observed in apex of resected prostate specimen and expressed semiquantitatively as percentage of total apex tissue. They showed that the odds of a patient whose resected striated muscle occu- pied more than 11% of total apex tissue being incon- tinent was 11.7 times that of a patient whose resected striated muscle occupied less than 10%. We followed their methods exactly in this study. Third, we measured the length of resected MU histologically using resected prostate specimen. To our knowledge, such measure- ment is firstly tried in this study. Other variables we analyzed for postoperative urinary incontinence were age(17), type of surgery (RARP/LRP) (7), nerve preservation(18). All these variables were re- ported to be significant factors for postoperative incon- tinence, although negative reports were also observed. We performed Cox regression analysis to identify the most relevant factor for postoperative urinary conti- nence using the above-mentioned variables (age, type of surgery, nerve preservation, length of MU by pre- operative MRI, amount of resected rhabdosphincter by pathology, and length of resected MU by pathology). As a result, length of resected MU by pathology was the only independent factor for predicting postoperative urinary continence (Table 2). Although it did not reach statistically significant (p = 0.06), length of MU by pre- operative MRI also showed tendency to predict post- operative urinary continence. Our results are consistent with Paparel’s repot(15). They showed that both preop- erative and postoperative membranous urethral length measured by preoperative and postoperative MRI were correlated with urinary continence status after radical prostatectomy. The only difference is that our resected urethral length is an absolute length, whereas Paparel et al. showed the resected urethral length in proportion to the original length of MU. Another concern is the PSM at the apex of prostate. We wondered if the PSM rate would increase in cases where rhabdosphincter or MU at the apical region of prostate was preserved as much as possible to obtain good uri- nary continence. However, no significant differences were found in PSM rates according to the amount of resected rhabodsphincter or the length of resected MU. Limitations of this study were retrospective fashion and the relatively small number of patients. Another lim- itation is that we could only evaluated the amount of resected rhabdosphincter semi-quantitatively. In conclusion, length of resected MU was the most im- portant factor to predict urinary incontinence after LRP or RARP. In LRP or RARP, sophisticated manipulation is possible in a magnified field of view with less bleed- ing. To the extent that we do not sacrifice complete resection of the tumor, we should preserve maximal length of MU to prevent postoperative urinary incon- tinence. REFERENCES 1. Ficarra V, Novara G, Artibani W, et al. Retropubic, laparoscopic, and robot-assisted radical prostatectomy: a systematic review and cumulative analysis of comparative studies. Eur Urol. 2009;55:1037-63. 2. Hamidi N, Atmaca AF, Canda AE, et al. Does Presence of a Median Lobe Affect Perioperative Complications, Oncological Outcomes and Urinary Continence Following Robotic-assisted Radical Prostatectomy? Urol J. 2018;15:248-55. 3. Jiang DG, Xiao CT, Mao YH, et al. Impact and Predictive Value of Prostate Weight on the Outcomes of Nerve Sparing Laparoscopic Radical Prostatectomy in Patients with Low Risk Prostate Cancer. Urol J. 2018. 4. Mustafa M, Davis JW, Gorgel SN, Pisters L. Robotic or Open Radical Prostatectomy in Men with Previous Transurethral Resection of Prostate. Urol J. 2017;14:2955-60. 5. Kojima Y, Takahashi N, Haga N, et al. Urinary incontinence after robot-assisted radical prostatectomy: pathophysiology and intraoperative techniques to improve surgical outcome. Int J Urol. 2013;20:1052-63. 6. Vora AA, Dajani D, Lynch JH, Kowalczyk KJ. Anatomic and technical considerations for optimizing recovery of urinary function during robotic-assisted radical prostatectomy. Curr Opin Urol. 2013;23:78-87. 7. Ficarra V, Novara G, Rosen RC, et al. Systematic review and meta-analysis of studies reporting urinary continence recovery after robot-assisted radical prostatectomy. Eur Urol. 2012;62:405-17. Urological Oncology 150 Urinary incontinence after radical prostatectomy-Kohjimoto et al. 8. Koike H, Kohjimoto Y, Iba A, et al. Health- related quality of life after robot-assisted radical prostatectomy compared with laparoscopic radical prostatectomy. J Robot Surg. 2017. 9. Takegami M, Suzukamo Y, Sanda MG, et al. [The Japanese translation and cultural adaptation of Expanded Prostate Cancer Index Composite (EPIC)]. Nihon Hinyokika Gakkai Zasshi. 2005;96:657-69. 10. Burnett AL, Mostwin JL. In situ anatomical study of the male urethral sphincteric complex: relevance to continence preservation following major pelvic surgery. J Urol. 1998;160:1301- 6. 11. Ganzer R, Blana A, Gaumann A, et al. Topographical anatomy of periprostatic and capsular nerves: quantification and computerised planimetry. Eur Urol. 2008;54:353-60. 12. Rocco F, Carmignani L, Acquati P, et al. Early continence recovery after open radical prostatectomy with restoration of the posterior aspect of the rhabdosphincter. Eur Urol. 2007;52:376-83. 13. Walz J, Burnett AL, Costello AJ, et al. A critical analysis of the current knowledge of surgical anatomy related to optimization of cancer control and preservation of continence and erection in candidates for radical prostatectomy. Eur Urol. 2010;57:179-92. 14. Strasser H, Pinggera GM, Gozzi C, et al. Three-dimensional transrectal ultrasound of the male urethral rhabdosphincter. World J Urol. 2004;22:335-8. 15. Paparel P, Akin O, Sandhu JS, et al. Recovery of urinary continence after radical prostatectomy: association with urethral length and urethral fibrosis measured by preoperative and postoperative endorectal magnetic resonance imaging. Eur Urol. 2009;55:629- 37. 16. Skeldon SC, Gani J, Evans A, Van Der Kwast T, Radomski SB. Striated muscle in the prostatic apex: does the amount in radical prostatectomy specimens predict postprostatectomy urinary incontinence? Urology. 2014;83:888-92. 17. Hou GL, Luo Y, Di JM, et al. Predictors of urinary continence recovery after modified radical prostatectomy for clinically high-risk prostate cancer. Urol J. 2015;12:2021-7. 18. Choi WW, Freire MP, Soukup JR, et al. Nerve- sparing technique and urinary control after robot-assisted laparoscopic prostatectomy. World J Urol. 2011;29:21-7. 19. Lee SE, Byun SS, Lee HJ, et al. Impact of variations in prostatic apex shape on early recovery of urinary continence after radical retropubic prostatectomy. Urology. 2006;68:137-41. Urinary incontinence after radical prostatectomy-Kohjimoto et al. Vol 17 No 02 March-April 2020 151