- Email: [email protected]
Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy
EUF-662; No. of Pages 8 E U R O P E A N U R O L O GY F O C U S X X X ( 2 0 18 ) X X X– X X X
available at www.sciencedirect.com journal homepage: www.europeanurology.com/eufocus
Incontinence
Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy Lawrence H.C. Kim *, Amit Patel, Netty Kinsella, Mansour T.A. Sharabiani, Derfel Ap Dafydd, Declan Cahill Royal Marsden Hospital, London, UK
Article info
Abstract
Article history: Accepted January 16, 2019
Background: Studies demonstrated the significance of membranous urethral length (MUL) as a predictor of continence following robot-assisted radical prostatectomy (RARP). There are other magnetic resonance imaging (MRI) parameters that might be linked to continence outcome. Objective: To evaluate the association between preoperative urethral parameters on MRI and continence outcome, to estimate the risk of incontinence using different cut-off values, and to assess interobserver variability in measuring urethral parameters. Design, setting, and participants: Patients with localised prostate cancer who underwent RARP were retrospectively reviewed. Baseline patient characteristics, perioperative, and pathological outcomes were assessed. Continence was defined as no pad or a safety pad with <2 g/24 h pad weight. Outcome measurements and statistical analysis: Several MRI variables were measured by a uroradiologist, a uro-oncology fellow, and a urologist. Binary logistic regression analyses were performed to identify predictors of incontinence. Interclass correlation was used to evaluate interobserver variability. Results and limitations: A total of 190 patients met the study inclusion criteria. The mean MUL was 14.6 mm. Age and MUL were significantly associated with incontinence outcome. The area under the receiver operating characteristic curve for continence based on MUL was 0.78 at 12 mo. The risk of incontinence in patients with MUL of <10 mm was 27.8% (13.8% and 39.1% for patients aged <65 and >65 yr respectively). Conversely, the risk of incontinence with MUL >15 mm was 2.7% (1.5% and 4.5% for patients aged <65 and >65 yr, respectively). The concordance rate between different observers was 89% for coronal MUL, but 77%, 74%, and 62% for sagittal MUL, membranous urethral thickness, and intraprostatic urethral length, respectively. Conclusions: This study confirmed the significance of MUL for the continence outcome following RARP. There was also excellent consistency in measuring MUL values between different observers. Patient summary: Although further studies would be required to verify our findings, we support the significance of membranous urethral length in predicting the risk of incontinence and the need to incorporate it as part of preoperative assessment and counselling. This can reliably be measured by urologists and can further facilitate a patient-tailored approach to radical treatment of prostate cancer. Crown Copyright © 2019 Published by Elsevier B.V. on behalf of European Association of Urology. All rights reserved.
Associate Editor: Derya Tilki Keywords: Membranous urethral length Radical prostatectomy Continence Interobserver variability
* Corresponding author. Department of Urology, Royal Marsden Hospital, Fulham Road, Outside U.S./ Canada, London, UK. Tel.: +614 3111 6566; Fax: +612 9633 3672. E-mail address: [email protected] (Lawrence H.C. Kim).
https://doi.org/10.1016/j.euf.2019.01.011 2405-4569/Crown Copyright © 2019 Published by Elsevier B.V. on behalf of European Association of Urology. All rights reserved.
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 2
E U R O P E A N U R O L O GY F O C U S X X X ( 2 0 18 ) X X X– X X X
1.
Introduction
therapy (n = 8), and missing follow-up data (n = 9). Baseline patient characteristics were obtained, including age, BMI, ASA score, prostate
Robot-assisted radical prostatectomy (RARP) is increasingly becoming the standard surgical therapy of choice for localised prostate cancer [1–3]. However, despite a better understanding of the anatomy of the prostate and its surrounding structures, as well as ongoing advances in technology and technical modifications, incontinence continues to affect quality of life and psychological well-being following RARP [4]. The aetiology of postradical prostatectomy incontinence is complex and still not thoroughly defined. Several preoperative factors have been demonstrated to be associated with continence outcome. Advanced age is one of the most well-known risk factors, and an anatomical study demonstrated progressive loss of the striated sphincter invaded by connective tissue, leading to weakening of its strength and tensile properties with age [5]. Other studies reported a significant correlation with higher body mass index (BMI), American Society of Anesthesiologists (ASA) score, international prostate symptom score (IPSS), international index of erectile function (IIEF), and prostate volume. Surgeon’s experience and techniques such as robotassisted procedure, nerve sparing, bladder neck preservation, preservation of anterior urethral ligaments, and urethrovesical reconstruction also seem to play a role [6,7]. Multiparametric magnetic resonance imaging (MRI) scan is increasingly being utilised for its significant role in diagnosis and staging of the disease. One of the radiological factors that has been shown to influence continence is the membranous urethral length (MUL). A review article concluded its major influence on urinary continence at every time point at 1, 3, 6, and 12 mo after radical prostatectomy [8]. Although the results are heterogeneous, there are other MRI parameters such as membranous urethral thickness (MUT) that might be linked to continence outcome [9]. We postulate that the intraprostatic urethral length (IPL) might also play a part based on the fact that the rhabdosphincter extends from the membranous urethra (MU) up to the level of the verumontanum, as demonstrated in anatomical and functional studies [10]. The majority of the studies were conducted in Asia or North America, and this is the first study that was evaluated in the UK. Furthermore, the measurements have been based on the interpretation by radiologists, and there is no documented evidence whether urologists can reliably measure these parameters preoperatively. The primary objective of this study was to evaluate the association between preoperative urethral parameters on MRI and continence outcome. The secondary objectives were to estimate the risk of incontinence using different cut-off values and to assess interobserver variability in measuring urethral parameters.
volume, and IPSS. Perioperative and pathological outcomes such as prostate-specific antigen (PSA), estimated blood loss, complications, Gleason score, pelvic lymph node dissection, staging, nerve-sparing technique, and surgical margin status were also collected. Patients were regularly followed up at 8 wk and at 3, 6, 9, and 12 mo after RARP. Clinical examination, and examination of PSA and continence status were carefully performed at each follow-up. Continence was assessed by the patient-reported pad usage per 24 h. In terms of patients who reported the use of a security pad of 24-h pad weight was employed, and continence was defined as leak <2 g per 24-h pad weight for more objective assessment. All patients were taught and advised to perform pelvic floor muscle exercises in the postoperative period until they regained continence.
2.1.
MRI measurement
Imaging was performed using a 3 Tesla Siemens (Erlangen, Germany) Skyra MRI scanner using the following parameters: transverse large field of view (FOV) T1-weighted imaging (repetition time/echo time [TR/TE], 678/20 ms; section thickness, 5 mm; intersection gap, 1 mm; FOV, 38 cm; matrix, 320 320), transverse large FOV T2-weighted imaging (TR/TE, 4370/108 ms; section thickness, 5 mm; intersection gap, 1 mm; FOV, 38 cm; matrix, 320 320), large FOV diffusion-weighted imaging (TR/TE, 10 900/70 ms; section thickness, 5 mm; intersection gap, 0 mm; FOV, 38 cm; matrix, 320 320; B50, B600, B1050), high-resolution transverse T2-weighted imaging (TR/TE, 5500/77 ms; section thickness, 5 mm; intersection gap, 0 mm; FOV, 17 cm; matrix, 256 256), and high-resolution coronal T2-weighted imaging (TR/TE, 6120/122 ms; section thickness, 3 mm; intersection gap, 0.3 mm; FOV, 20 cm; matrix, 320 320). MRI variables were measured by a uroradiologist, a urooncology fellow, and a urologist. MUL was measured from the inferior end of the prostatic apex to the level of the penile bulb. Fig. 1 depicts how every MRI urethral parameter was measured. IPL was measured from the level of the verumontanum to the most inferior end of the prostatic apex. MUT was measured in transverse and anterioposterior dimension at the level of the penile bulb. Surface area of the MU was calculated based on the following equation: (transverse MUT 1/2) (anteroposterior MUT 1/2) p; volume of MU was calculated by the following equation: (transverse MUT 1/2) (anteroposterior MUT 1/2) p cor-
onal MUL.
2.2.
Statistical analyses
Demographic and clinicopathological data were presented as frequency distribution and percentages, respectively. Non-normally distributed variables were expressed as median and interquartile range, and normally distributed variables as mean standard deviation. Known confounding variables were included in the univariate and multivariate logistic regression analyses to identify predictors of incontinence. Multicollinearity was assessed for the presence of a high correlation between variables using the variance inflation factor. Interclass correlation was used to evaluate interobserver variability. The levels of classification and prediction were also measured by the area under the receiver operating characteristic curve. A confidence interval of 95% was assumed, and p < 0.05 was considered significant. All statistical analyses were carried
2.
Patients and methods
Following institutional review board approval, prospectively collected data on patients with localised prostate cancer who underwent RARP
out using SPSS version 22.0 (IBM Corp., Somers, NY, USA).
3.
Results
from 2015 to 2016 by a single high-volume surgeon (D.C.) were retrospectively reviewed. We excluded patients with a history of prostate cancer treatment (n = 3), preoperative incontinence (n = 2), adjuvant
A total of 190 consecutive patients met the study criteria. Continence was achieved in 116 patients at 3 mo and in
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 E U RO P E A N U RO L O GY F O C U S X X X ( 2 018 ) X X X– X X X
3
Fig. 1 – (A) MUL on coronal view. (B) MUL on sagittal view. (C) IPL. (D) MUT and surface area MU. Blue line represents anteroposterior MUT, green line transverse MUT, and white line surface area MU. IPL = intraprostatic urethral length; MU = membranous urethra; MUL = membranous urethral length; MUT = membranous urethral thickness.
174 patients at 12 mo. The mean age and BMI were 62 and 27.0 kg/m2, respectively (Table 1). PSA was undetectable in 175 patients at 12 mo. The median follow-up period was 21 mo. In terms of MRI measurements, the mean MUL values were 14.6 and 14.2 mm on coronal and sagittal views, respectively. Other parameters are also shown in Table 2. Fig. 2 demonstrates the example of short and long MUL on coronal view. After several known continence predictors were incorporated, age, BMI, and MUL remained to be significant predictors on multivariate analysis at 3 mo. MU volume was also shown to be a predictor on univariate analysis at 3 mo, but no longer significant on multivariate
analysis. Univariate analysis demonstrated that age and coronal MUL were significant for continence at 12 mo (Tables 3 and 4). Other MRI parameters such as MUT, IPL, and surface area of the MU did not show any correlation with continence outcome. The area under the receiver operating characteristic curve for continence based on coronal MUL was 0.78 at 12 mo (Fig. 3). The risk of incontinence in patients with coronal MUL of <10 mm was 26.7% (Fig. 4). It was three-fold more likely in patients over 65 yr of age (39.1% vs 13.8%). Conversely, the risk of incontinence with coronal MUL >15 mm was 2.7%. Again, the risk of incontinence was three times higher in patients over 65 yr of age (4.5% vs 1.5%).
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 4
E U R O P E A N U R O L O GY F O C U S X X X ( 2 0 18 ) X X X– X X X
Table 1 – Patient clinicopathological characteristics. ASA = American Society of Anesthesiologists; BMI = body mass index; IPSS = international prostate symptom score; IQR = interquartile range; PSA = prostate-specific antigen; SD = standard deviation. Baseline characteristics Mean (SD) Age (yr) BMI (kg/m2) PSA (ng/ml) Prostate volume (cm3) IPSS ASA score 1 2 3 Estimated blood loss (ml) Operating time (min) Pathologic stage T2 T3a T3b Prostatectomy Gleason score GG1 GG2 GG3 GG4 Positive surgical margin Nerve sparing Not done Unilateral Bilateral Pelvic lymph node dissection Lymph node positivity
62.4 (6.5) 27.0 (3.1) 7.4 (5.5–10.0) 36.4 (29.5–45) 15 (9–20) 68 (35.8) 119 (62.6) 3 (1.6) 150 (100, 200) 120 (97.5, 146.3) 82 (43.2) 94 (49.4) 14 (7.4) 4 (2.3) 116 (61.1) 58 (30.7) 12 (6.3) 27 (14.2) 40 (21) 95 (50) 55 (29) 85 (46.0) 10 (12)
Data are presented as median (IQR), mean (SD), or n (%). ASA = American Society of Anesthesiologists; BMI = body mass index; IPSS = international prostate symptom score; IQR = interquartile range; PSA = prostate-specific antigen; SD = standard deviation.
Table 2 – Preoperative MRI parameters MRI variables Coronal MUL (mm) Sagittal MUL (mm) IPL (mm) Transverse MUT (mm) Anteroposterior MU (mm) Axial surface area MUa (mm2) Volume MUb (mm3)
14.6 (3.0) 14.2 (2.7) 16.1 (4.0) 12.7 (1.9) 12.8 (2.0) 131.6 (37.1) 1947 (633.7)
Data are presented as mean (SD). IPL = intraprostatic urethral length; MRI = magnetic resonance imaging; MU = membranous urethra; MUL = membranous urethral length; MUT = membranous urethral thickness; SD = standard deviation. a (Transverse MUT 1/2) (anteroposterior MUT 1/2) p. b (Transverse MUT 1/2) (anteroposterior MUT 1/2) p coronal MUL.
The concordance rate between different observers was 89% for MUL on coronal view. However, the rates were demonstrated to be lower at 77%, 74%, 72%, and 62% for sagittal MUL, transverse MUT, anteroposterior MUT, and IPL, respectively. 4.
Discussion
RARP has increasingly been adopted to be the standard surgical treatment of choice for localised prostate cancer.
Fig. 2 – MUL measurement on MRI. (A) Short MUL on coronal view. (B) Long MUL on coronal view. MRI = magnetic resonance imaging; MUL = membranous urethral length.
However, incontinence remains to be one of the complications that are most difficult to manage. Continence relies on a complex mechanism and is multifactorial in origin. RARP leads to changes in the structure and function of the components inherently related to the urinary sphincter complex. It removes internal urethral sphincter located in the proximal portion of the prostate. It is composed of smooth muscle under the regulation of autonomic innervation and is continuous with the detrusor muscle. Several different techniques such as bladder neck sparing have been attempted to preserve the internal sphincter with little significant benefit. It has been shown that postoperative continence depends largely on the external urethral sphincter—rhabdosphincter [11]. It is innervated by a branch of the pudendal nerve. This is most prominent over the MU and forms a muscular coat over it in an omega-shaped loop. Its key role in continence is thought to be due to its ability to
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 5
E U RO P E A N U RO L O GY F O C U S X X X ( 2 018 ) X X X– X X X
Table 3 – Univariate and multivariate regression analysis of continence at 3 mo Predictive factors
Univariate
Age BMI ASA Prostate volume Preop PSA IPSS EBL Gleason score GG3 vs GG4 pT stage T2 vs T3 Nerve sparing (NS) Non-NS vs Uni-NS Non-NS vs Bil-NS MRI Coronal MUL Sagittal MUL IPL Transverse MUT Anteroposterior MUT Axial surface area MU Total volume MU
Multivariate
HR
95% CI
p value
HR
95% CI
p value
0.921 0.921 0.643 0.983 0.959 0.996 1.000 0.606
0.876–0.969 0.830–1.022 0.337–1.227 0.957–1.009 0.918–1.002 0.976–1.017 0.998–1.002 0.276–1.329
0.001 0.121 0.180 0.190 0.063 0.714 0.792 0.211
0.932 0.884
0.882–0.985 0.790–0.990
0.013 0.033
0.610
0.324–1.148
0.126
2.014 3.174
0.801–4.108 1.304–6.758
0.152 0.016
2.177 2.268
0.799–5.931 0.932–5.517
0.128 0.071
1.152 1.154 1.131 1.054 1.120 1.007 1.001
1.035–1.281 0.998–1.385 0.981–1.307 0.901–1.264 0.930–1.325 0.998–1.021 1.000–1.002
0.009 0.078 0.104 0.423 0.218 0.126 0.020
1.199
1.063–1.354
0.003
1.001
0.999–1.002
0.295
ASA = American Society of Anesthesiologists; Bil = bilateral; BMI = body mass index; CI = confidence interval; EBL = estimated blood loss; HR = hazard ratio; IPL = intraprostatic urethral length; IPSS = international prostate symptom score; MRI = magnetic resonance imaging; MU = membranous urethra; MUL = membranous urethral length; MUT = membranous urethral thickness; PSA = prostate-specific antigen; Uni = unilateral.
Table 4 – Univariate and multivariate regression analysis of continence at 12 mo Predictive factors
Age BMI ASA Prostate volume Preop PSA IPSS EBL Gleason score GG3 vs GG4 pT stage T2 vs T3 Nerve sparing (NS) Non-NS vs Uni-NS Non-NS vs Bil-NS MRI Coronal MUL Sagittal MUL IPL Transverse MUT Anteroposterior MUT Axial surface area MU Total volume MU
Univariate HR
95% CI
p value
0.830 0.959 0.674 0.993 0.969 0.997 0.998 0.621
0.718–0.972 0.820–1.046 0.335–1.230 0.960–1.027 0.910–1.002 0.981–1.020 0.994–1.002 0.256–1.337
0.007 0.128 0.185 0.667 0.068 0.701 0.278 0.231
0.587
0.132–2.418
0.446
1.230 1.848
0.240–5.214 0.306–7.821
0.809 0.567
1.425 1.847 1.115 1.057 1.132 1.008 1.001
1.187–1.869 0.789–3.692 0.939–1.324 0.653–1.483 0.721–1.686 0.998–1.013 1.000–1.003
0.001 0.095 0.213 0.645 0.531 0.118 0.108
ASA = American Society of Anesthesiologists; Bil = bilateral; BMI = body mass index; CI = confidence interval; EBL = estimated blood loss; HR = hazard ratio; IPL = intraprostatic urethral length; IPSS = international prostate symptom score; MRI = magnetic resonance imaging; MU = membranous urethra; MUL = membranous urethral length; MUT = membranous urethral thickness; PSA = prostate-specific antigen; Uni = unilateral.
increase urethral closure pressures voluntarily [12–14]. Maximal MUL sparing has been recommended to achieve maximal functional urethral length and shown to improve continence [10,15,16]. This would further facilitate more effective training of pelvic floor muscle. Therefore, MUL could serve as a surrogate marker for the rhabdosphincter responsible for stress urinary incontinence after radical prostatectomy. The MU also contains smooth muscle fibres that are supplied by autonomic innervation from the prostatic plexus and cavernous nerve. It is believed to influence a resting tone contributing to maintaining urethral closure pressures. Studies demonstrated improved continence in patients who underwent nerve-sparing prostatectomy. This is postulated to be due to more preserved membranous urethral vascular integrity and membranous urethral sensitivity from less autonomic afferent denervation [17–20]. Several other technical refinements have been proposed with heterogeneous outcomes. They include puboprostatic sparing, bladder neck preservation, pubovesical complex or dorsal vein complex preservation, and posterior musculofascial reconstruction, as well as anterior pelvic restoration and anterior reconstruction using the periurethral suspension stitch [21–23]. More recently, Retzius-sparing radical prostatectomy has been demonstrated to provide superior early continence outcome by preserving a number of the supportive and integrated structures mentioned above [24,25]. Strasser et al [12] demonstrated transrectal ultrasound as a useful tool to assess the quality of the rhabdosphincter postoperatively. They detected thinning, wasting, defects, or scarring in the majority of patients with
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 6
E U R O P E A N U R O L O GY F O C U S X X X ( 2 0 18 ) X X X– X X X
Fig. 3 – Area under the ROC curve for continence at 12 mo based on coronal MUL. MUL = membranous urethral length; ROC = receiver operating characteristic.
1.00 0.80
Probability
0.60 0.40 0.20 0.00
3.0
6.0
9.0
12.0
15.0
18.0
21.0
24.0
Fig. 4 – Probability of continence following RARP based on membranous urethral length (MUL). RARP = robot-assisted radical prostatectomy.
incontinence. The rhabdosphincter-urethra distance was significantly lower in the incontinent group than in the continent group. This would in part explain a much lower rate of continence in patients following salvage radical prostatectomy, which is likely a result of an extensive fibrosis from ischaemia and loss of dissection planes. Postoperative cystographic images were also used to predict continence outcome and showed that vesicourethral anastomosis location and bladder neck angle were significantly associated with continence [26,27]. Unlike the preoperative factors, anatomical variables can be improved by surgeon’s experience and technical modifications. The aforementioned technical modifications could theoretically improve these cystographic profiles and explain improved continence.
In terms of the significance of MUL for continence, Coakley et al [28] reported a continence rate of 89% if MUL was >12 mm and was decreased down to 77% if 12 mm.Our study showed that the risk of incontinence was 27.8% for MUL <10 mm and 2.7% for MUL >15 mm. The risk of incontinence was almost three times more likely if patients are aged >65 yr, again confirming its significant influence on continence. There are studies that developed nomograms incorporating MUL into preoperative factors to predict continence [7,29]. Matsushita et al [7] showed that the addition of MUL increased the area under the curve by 0.085 over the base model. However, there is a lack of information as to the significance of other parameters such as MUT, IPL, surface area, or volume of MU. Anatomical and functional studies showed the extension of
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 E U RO P E A N U RO L O GY F O C U S X X X ( 2 018 ) X X X– X X X
the rhabdosphincter from the MU up to the level of the verumontanum [10]. Although the amount of rhabdosphincter over the intraprostatic urethra would vary, this might theoretically be associated with continence. However, there was no significant difference in continence outcome based on IPL in our study. This could partly be explained by difficulty ascertaining an accurate measurement of its length due to variable thickness of slices performed on an MRI scan. We also postulate that the proportion of IPL in relation to MUL might also be associated with continence. Tienza and colleagues [9] showed that transverse MUT played a part in continence outcome. We measured MUT in its transverse and anteroposterior diameters, and there was no significant association with continence. When the surface area of the MU at the level of the penile bulb was measured, no association was again demonstrated. However, MU volume was significant on univariate analysis. This is most likely due to MUL being a strong influential factor. This study has several limitations. It was a small singlesurgeon series. Accuracy of the surface volume or volume measurement is questionable, as it was based at the level of the penile bulb and there is an anatomical variation along MUL in each individual. Difficulty was encountered in measuring MUL on a sagittal view and IPL partly because the slices were not calibrated thin enough. We also made a presumption that the urethra is of similar calibre and the MUT would correlate with the amount of rhabdosphincter axially. There was no postoperative MRI to identify accurately and objectively the residual functional MUL and extent of scarring. However, preoperative MUL was as significant as postoperative MUL [16]. Furthermore, the percent change of MUL from pre- to postoperative measurement did not have a significant impact on continence recovery at 12 mo [30]. Multivariate logistic regression analysis for continence at 12 mo was not performed due to a small number of events leading to overfitting if all the known predictors were included. Nevertheless, age and MUL were again significant on univariate analysis. Incontinence due to bladder dysfunction was not taken into account, although it would account for only a minority. We assessed various urethral parameters only, and did not take other parameters that were shown to be associated with incontinence into account such as bladder neck angle or descent in relation to pubic symphysis. There are several strengths worthy of mention. As the data were prospectively collected, a potential confounding bias was avoided. As this was a single-surgeon series, the technical factors that arise from the heterogeneity of different surgeon experience and skill set did not need to be taken into account. The analysis of the MRI parameters was performed by an independent uroradiologist with high-volume experience interpreting a prostate MRI scan. To our knowledge, this is the first study that is reported in the UK and evaluated interobserver variability between a radiologist, a urologist, and a trainee. This demonstrates the reliability of predicting continence based on the measurements performed by a urologist or a trainee.
5.
7
Conclusions
This study confirmed the significance of MUL for the continence outcome following RARP. There was also excellent consistency in measuring MUL values between different observers. While further larger prospective studies would be required to validate our findings, we suggest that in case of a high likelihood of incontinence, alternative treatment such as radiotherapy can be considered or early measures for definitive incontinence management in the form of urethral sling or artificial urinary sphincter can be planned. Author contributions: Lawrence H.C. Kim had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Kim, Patel, Cahill. Acquisition of data: Kim, Patel, Ap Dafydd, Cahill. Analysis and interpretation of data: Kim, Patel, Sharabiani, Cahill. Drafting of the manuscript: Kim, Patel, Cahill, Kinsella. Critical revision of the manuscript for important intellectual content: Kim, Kinsella, Cahill. Statistical analysis: Kim, Sharabiani. Obtaining funding: None. Administrative, technical, or material support: None. Supervision: Cahill. Other: None. Financial disclosures: Lawrence H.C. Kim certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: None.
References [1] Hofer MD, Meeks JJ, Cashy J, Kundu S, Zhao LC. Impact of increasing prevalence of minimally invasive prostatectomy on open prostatectomy observed in the national inpatient sample and national surgical quality improvement program. J Endourol 2013;27:102–7. [2] Kumar A, Samavedi S, Bates AS, et al. Continence outcomes of robotassisted radical prostatectomy in patients with adverse urinary continence risk factors. BJU Int 2015;116:764–70. [3] The British Association of Urological Surgeons. Analyses of Radical Prostatectomies performed between January 1st and December 31st 2016. https://www.baus.org.uk/_userfiles/pages/files/ Publications/Audit/Radical%20Prostatectomy%202016%20final% 20analysis.pdf [Accessed 10 January 2019]. [4] Punnen S, Cowan JE, Chan JM, Carroll PR, Cooperberg MR. Long-term health-related qualityof life after primary treatment forlocalized prostate cancer: results from the CaPSURE registry. Eur Urol 2015;68:600–8. [5] Oelrich TM. The urethral sphincter muscle in the male. Am J Anat 1980;158:229–46. [6] Ficarra V, Novara G, Rosen RC, et al. Systematic review and metaanalysis of studies reporting urinary continence recovery after robot-assisted radical prostatectomy. Eur Urol 2012;62:405–17. [7] Matsushita K, Kent MT, Vickers AJ, et al. Preoperative predictive model of recovery of urinary continence after radical prostatectomy. BJU Int 2015;116:577–83.
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 8
E U R O P E A N U R O L O GY F O C U S X X X ( 2 0 18 ) X X X– X X X
[8] Mungovan SF, Sandhu JS, Akin O, Smart NA, Graham PL, Patel MI.
[20] Park YH, Kwon OS, Hong S, Kim SW, Hwang T, Lee JY. Effect of nerve-
Preoperative membranous urethral length measurement and con-
sparing radical prostatectomy on urinary continence in patients with
tinence recovery following radical prostatectomy: a systematic review and meta-analysis. Eur Urol 2017;71:368–78.
preoperative erectile dysfunction. Int Neurourol J 2016;20:69–74. [21] Vora AA, Dajani D, Lynch JH, Kowalczyk KJ. Anatomic and technical
[9] Tienza A, Hevia M, Benito A, Pascual JI, Zudaire JJ, Robles JE. MRI
considerations for optimizing recovery of urinary function during
factors to predict urinary incontinence after retropubic/laparo-
robotic-assisted radical prostatectomy. Curr Opin Urol 2013;23:78–87.
scopic radical prostatectomy. Int Urol Nephrol 2015;47:1343–9. [10] Schlomm T, Heinzer H, Steuber T, et al. Full functional-length urethral sphincter preservation during radical prostatectomy. Eur Urol 2011;60:320–9.
[22] Sridhar AN, Abozaid M, Rajan P, et al. Surgical techniques to optimize early urinary continence recovery post robot assisted radical prostatectomy for prostate cancer. Curr Urol Rep 2017;18:71. [23] Walz J, Epstein JI, Ganzer R, et al. A critical analysis of the current
[11] Hoyland K, Vasdev N, Abrof A, Boustead G. Post-radical prostatectomy
knowledge of surgical anatomy of the prostate related to optimisa-
incontinence: etiology and prevention. Rev Urol 2014;16:181–8.
tion of cancer control and preservation of continence and erection
[12] Strasser H, Pinggera GM, Gozzi C, et al. Three-dimensional transrectal ultrasound of the male urethral rhabdosphincter. World J Urol 2004;22:335–8.
in candidates for radical prostatectomy: an update. Eur Urol 2016;70:301–11. [24] Dalela D, Jeong W, Prasad MA, et al. A pragmatic randomized
[13] Cameron AP, Suskind AM, Neer C, et al. Functional and anatomical
controlled trial examining the impact of the Retzius-sparing
differences between continent and incontinent men post radical
approach on early urinary continence recovery after robot-assisted
prostatectomy on urodynamics and 3T MRI: a pilot study. Neurourol Urodyn 2015;34:527–32. [14] Dubbelman YD, Bosch JL. Urethral sphincter function before and after radical prostatectomy: systematic review of the prognostic value of various assessment techniques. Neurourol Urodyn 2013;32:957–63. [15] Hamada A, Razdan S, Etafy MH, Fagin R, Razdan S. Early return of
radical prostatectomy. Eur Urol 2017;72:677–85. [25] Lim SK, Kim KH, Shin TY, et al. Retzius-sparing robot-assisted laparoscopic radical prostatectomy: combining the best of retropubic and perineal approaches. BJU Int 2014;114:236–44. [26] Shao IH, Chou CY, Huang CC, et al. A specific cystography pattern can predict postprostatectomy incontinence. Ann Surg Oncol 2015;22 (Suppl 3):S1580–6.
continence in patients undergoing robot-assisted laparoscopic
[27] Jeong SJ, Yi J, Chung MS, et al. Early recovery of urinary continence
prostatectomy using modified maximal urethral length preserva-
after radical prostatectomy: correlation with vesico-urethral anas-
tion technique. J Endourol 2014;28:930–8. [16] Paparel P, Akin O, Sandhu JS, et al. Recovery of urinary continence after radical prostatectomy: association with urethral fibrosis mea-
tomosis location in the pelvic cavity measured by postoperative cystography. Int J Urol 2011;18:444–51. [28] Coakley FV, Eberhardt S, Kattan MW, Wei DC, Scardino PT, Hricak H.
sured by preoperative and postoperative magnetic resonance imag-
Urinary continence after radical retropubic prostatectomy: rela-
ing. Eur Urol 2009;55:629–37.
tionship with membranous urethral length on preoperative endor-
[17] Kadono Y, Nohara T, Kawaguchi S, et al. Investigating the mechanism underlying urinary continence recovery after radical prosta-
ectal magnetic resonance imaging. J Urol 2002;168:1032–5. [29] Jeong SJ, Yeon JS, Lee JK, et al. Development and validation of
tectomy: effectiveness of a longer urethral stump to prevent urinary
nomograms to predict the recovery of urinary continence after
incontinence. BJU Int 2018;122:456–62.
radical prostatectomy: comparisons between immediate, early,
[18] John H, Suter S, Hauri D. Effect of radical prostatectomy on urethral blood flow. Urology 2002;59:566–9. [19] Catarin MV, Manzano GM, Nobrega JA, Almeida FG, Srougi M, Bruschini
and late continence. World J Urol 2014;32:437–44. [30] Song W, Kim CK, Park BK, et al. Impact of preoperative and postoperative membranous urethral length measured by 3 Tesla magnetic
H. The role of membranous urethral afferent autonomic innervation in
resonance imaging on urinary continence recovery after robotic-
the continence mechanism after nerve sparing radical prostatectomy: a
assisted radical prostatectomy. Can Urol Assoc J 2017;11:E93–9.
clinical and prospective study. J Urol 2008;180:2527–31.
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
available at www.sciencedirect.com journal homepage: www.europeanurology.com/eufocus
Incontinence
Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy Lawrence H.C. Kim *, Amit Patel, Netty Kinsella, Mansour T.A. Sharabiani, Derfel Ap Dafydd, Declan Cahill Royal Marsden Hospital, London, UK
Article info
Abstract
Article history: Accepted January 16, 2019
Background: Studies demonstrated the significance of membranous urethral length (MUL) as a predictor of continence following robot-assisted radical prostatectomy (RARP). There are other magnetic resonance imaging (MRI) parameters that might be linked to continence outcome. Objective: To evaluate the association between preoperative urethral parameters on MRI and continence outcome, to estimate the risk of incontinence using different cut-off values, and to assess interobserver variability in measuring urethral parameters. Design, setting, and participants: Patients with localised prostate cancer who underwent RARP were retrospectively reviewed. Baseline patient characteristics, perioperative, and pathological outcomes were assessed. Continence was defined as no pad or a safety pad with <2 g/24 h pad weight. Outcome measurements and statistical analysis: Several MRI variables were measured by a uroradiologist, a uro-oncology fellow, and a urologist. Binary logistic regression analyses were performed to identify predictors of incontinence. Interclass correlation was used to evaluate interobserver variability. Results and limitations: A total of 190 patients met the study inclusion criteria. The mean MUL was 14.6 mm. Age and MUL were significantly associated with incontinence outcome. The area under the receiver operating characteristic curve for continence based on MUL was 0.78 at 12 mo. The risk of incontinence in patients with MUL of <10 mm was 27.8% (13.8% and 39.1% for patients aged <65 and >65 yr respectively). Conversely, the risk of incontinence with MUL >15 mm was 2.7% (1.5% and 4.5% for patients aged <65 and >65 yr, respectively). The concordance rate between different observers was 89% for coronal MUL, but 77%, 74%, and 62% for sagittal MUL, membranous urethral thickness, and intraprostatic urethral length, respectively. Conclusions: This study confirmed the significance of MUL for the continence outcome following RARP. There was also excellent consistency in measuring MUL values between different observers. Patient summary: Although further studies would be required to verify our findings, we support the significance of membranous urethral length in predicting the risk of incontinence and the need to incorporate it as part of preoperative assessment and counselling. This can reliably be measured by urologists and can further facilitate a patient-tailored approach to radical treatment of prostate cancer. Crown Copyright © 2019 Published by Elsevier B.V. on behalf of European Association of Urology. All rights reserved.
Associate Editor: Derya Tilki Keywords: Membranous urethral length Radical prostatectomy Continence Interobserver variability
* Corresponding author. Department of Urology, Royal Marsden Hospital, Fulham Road, Outside U.S./ Canada, London, UK. Tel.: +614 3111 6566; Fax: +612 9633 3672. E-mail address: [email protected] (Lawrence H.C. Kim).
https://doi.org/10.1016/j.euf.2019.01.011 2405-4569/Crown Copyright © 2019 Published by Elsevier B.V. on behalf of European Association of Urology. All rights reserved.
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 2
E U R O P E A N U R O L O GY F O C U S X X X ( 2 0 18 ) X X X– X X X
1.
Introduction
therapy (n = 8), and missing follow-up data (n = 9). Baseline patient characteristics were obtained, including age, BMI, ASA score, prostate
Robot-assisted radical prostatectomy (RARP) is increasingly becoming the standard surgical therapy of choice for localised prostate cancer [1–3]. However, despite a better understanding of the anatomy of the prostate and its surrounding structures, as well as ongoing advances in technology and technical modifications, incontinence continues to affect quality of life and psychological well-being following RARP [4]. The aetiology of postradical prostatectomy incontinence is complex and still not thoroughly defined. Several preoperative factors have been demonstrated to be associated with continence outcome. Advanced age is one of the most well-known risk factors, and an anatomical study demonstrated progressive loss of the striated sphincter invaded by connective tissue, leading to weakening of its strength and tensile properties with age [5]. Other studies reported a significant correlation with higher body mass index (BMI), American Society of Anesthesiologists (ASA) score, international prostate symptom score (IPSS), international index of erectile function (IIEF), and prostate volume. Surgeon’s experience and techniques such as robotassisted procedure, nerve sparing, bladder neck preservation, preservation of anterior urethral ligaments, and urethrovesical reconstruction also seem to play a role [6,7]. Multiparametric magnetic resonance imaging (MRI) scan is increasingly being utilised for its significant role in diagnosis and staging of the disease. One of the radiological factors that has been shown to influence continence is the membranous urethral length (MUL). A review article concluded its major influence on urinary continence at every time point at 1, 3, 6, and 12 mo after radical prostatectomy [8]. Although the results are heterogeneous, there are other MRI parameters such as membranous urethral thickness (MUT) that might be linked to continence outcome [9]. We postulate that the intraprostatic urethral length (IPL) might also play a part based on the fact that the rhabdosphincter extends from the membranous urethra (MU) up to the level of the verumontanum, as demonstrated in anatomical and functional studies [10]. The majority of the studies were conducted in Asia or North America, and this is the first study that was evaluated in the UK. Furthermore, the measurements have been based on the interpretation by radiologists, and there is no documented evidence whether urologists can reliably measure these parameters preoperatively. The primary objective of this study was to evaluate the association between preoperative urethral parameters on MRI and continence outcome. The secondary objectives were to estimate the risk of incontinence using different cut-off values and to assess interobserver variability in measuring urethral parameters.
volume, and IPSS. Perioperative and pathological outcomes such as prostate-specific antigen (PSA), estimated blood loss, complications, Gleason score, pelvic lymph node dissection, staging, nerve-sparing technique, and surgical margin status were also collected. Patients were regularly followed up at 8 wk and at 3, 6, 9, and 12 mo after RARP. Clinical examination, and examination of PSA and continence status were carefully performed at each follow-up. Continence was assessed by the patient-reported pad usage per 24 h. In terms of patients who reported the use of a security pad of 24-h pad weight was employed, and continence was defined as leak <2 g per 24-h pad weight for more objective assessment. All patients were taught and advised to perform pelvic floor muscle exercises in the postoperative period until they regained continence.
2.1.
MRI measurement
Imaging was performed using a 3 Tesla Siemens (Erlangen, Germany) Skyra MRI scanner using the following parameters: transverse large field of view (FOV) T1-weighted imaging (repetition time/echo time [TR/TE], 678/20 ms; section thickness, 5 mm; intersection gap, 1 mm; FOV, 38 cm; matrix, 320 320), transverse large FOV T2-weighted imaging (TR/TE, 4370/108 ms; section thickness, 5 mm; intersection gap, 1 mm; FOV, 38 cm; matrix, 320 320), large FOV diffusion-weighted imaging (TR/TE, 10 900/70 ms; section thickness, 5 mm; intersection gap, 0 mm; FOV, 38 cm; matrix, 320 320; B50, B600, B1050), high-resolution transverse T2-weighted imaging (TR/TE, 5500/77 ms; section thickness, 5 mm; intersection gap, 0 mm; FOV, 17 cm; matrix, 256 256), and high-resolution coronal T2-weighted imaging (TR/TE, 6120/122 ms; section thickness, 3 mm; intersection gap, 0.3 mm; FOV, 20 cm; matrix, 320 320). MRI variables were measured by a uroradiologist, a urooncology fellow, and a urologist. MUL was measured from the inferior end of the prostatic apex to the level of the penile bulb. Fig. 1 depicts how every MRI urethral parameter was measured. IPL was measured from the level of the verumontanum to the most inferior end of the prostatic apex. MUT was measured in transverse and anterioposterior dimension at the level of the penile bulb. Surface area of the MU was calculated based on the following equation: (transverse MUT 1/2) (anteroposterior MUT 1/2) p; volume of MU was calculated by the following equation: (transverse MUT 1/2) (anteroposterior MUT 1/2) p cor-
onal MUL.
2.2.
Statistical analyses
Demographic and clinicopathological data were presented as frequency distribution and percentages, respectively. Non-normally distributed variables were expressed as median and interquartile range, and normally distributed variables as mean standard deviation. Known confounding variables were included in the univariate and multivariate logistic regression analyses to identify predictors of incontinence. Multicollinearity was assessed for the presence of a high correlation between variables using the variance inflation factor. Interclass correlation was used to evaluate interobserver variability. The levels of classification and prediction were also measured by the area under the receiver operating characteristic curve. A confidence interval of 95% was assumed, and p < 0.05 was considered significant. All statistical analyses were carried
2.
Patients and methods
Following institutional review board approval, prospectively collected data on patients with localised prostate cancer who underwent RARP
out using SPSS version 22.0 (IBM Corp., Somers, NY, USA).
3.
Results
from 2015 to 2016 by a single high-volume surgeon (D.C.) were retrospectively reviewed. We excluded patients with a history of prostate cancer treatment (n = 3), preoperative incontinence (n = 2), adjuvant
A total of 190 consecutive patients met the study criteria. Continence was achieved in 116 patients at 3 mo and in
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 E U RO P E A N U RO L O GY F O C U S X X X ( 2 018 ) X X X– X X X
3
Fig. 1 – (A) MUL on coronal view. (B) MUL on sagittal view. (C) IPL. (D) MUT and surface area MU. Blue line represents anteroposterior MUT, green line transverse MUT, and white line surface area MU. IPL = intraprostatic urethral length; MU = membranous urethra; MUL = membranous urethral length; MUT = membranous urethral thickness.
174 patients at 12 mo. The mean age and BMI were 62 and 27.0 kg/m2, respectively (Table 1). PSA was undetectable in 175 patients at 12 mo. The median follow-up period was 21 mo. In terms of MRI measurements, the mean MUL values were 14.6 and 14.2 mm on coronal and sagittal views, respectively. Other parameters are also shown in Table 2. Fig. 2 demonstrates the example of short and long MUL on coronal view. After several known continence predictors were incorporated, age, BMI, and MUL remained to be significant predictors on multivariate analysis at 3 mo. MU volume was also shown to be a predictor on univariate analysis at 3 mo, but no longer significant on multivariate
analysis. Univariate analysis demonstrated that age and coronal MUL were significant for continence at 12 mo (Tables 3 and 4). Other MRI parameters such as MUT, IPL, and surface area of the MU did not show any correlation with continence outcome. The area under the receiver operating characteristic curve for continence based on coronal MUL was 0.78 at 12 mo (Fig. 3). The risk of incontinence in patients with coronal MUL of <10 mm was 26.7% (Fig. 4). It was three-fold more likely in patients over 65 yr of age (39.1% vs 13.8%). Conversely, the risk of incontinence with coronal MUL >15 mm was 2.7%. Again, the risk of incontinence was three times higher in patients over 65 yr of age (4.5% vs 1.5%).
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 4
E U R O P E A N U R O L O GY F O C U S X X X ( 2 0 18 ) X X X– X X X
Table 1 – Patient clinicopathological characteristics. ASA = American Society of Anesthesiologists; BMI = body mass index; IPSS = international prostate symptom score; IQR = interquartile range; PSA = prostate-specific antigen; SD = standard deviation. Baseline characteristics Mean (SD) Age (yr) BMI (kg/m2) PSA (ng/ml) Prostate volume (cm3) IPSS ASA score 1 2 3 Estimated blood loss (ml) Operating time (min) Pathologic stage T2 T3a T3b Prostatectomy Gleason score GG1 GG2 GG3 GG4 Positive surgical margin Nerve sparing Not done Unilateral Bilateral Pelvic lymph node dissection Lymph node positivity
62.4 (6.5) 27.0 (3.1) 7.4 (5.5–10.0) 36.4 (29.5–45) 15 (9–20) 68 (35.8) 119 (62.6) 3 (1.6) 150 (100, 200) 120 (97.5, 146.3) 82 (43.2) 94 (49.4) 14 (7.4) 4 (2.3) 116 (61.1) 58 (30.7) 12 (6.3) 27 (14.2) 40 (21) 95 (50) 55 (29) 85 (46.0) 10 (12)
Data are presented as median (IQR), mean (SD), or n (%). ASA = American Society of Anesthesiologists; BMI = body mass index; IPSS = international prostate symptom score; IQR = interquartile range; PSA = prostate-specific antigen; SD = standard deviation.
Table 2 – Preoperative MRI parameters MRI variables Coronal MUL (mm) Sagittal MUL (mm) IPL (mm) Transverse MUT (mm) Anteroposterior MU (mm) Axial surface area MUa (mm2) Volume MUb (mm3)
14.6 (3.0) 14.2 (2.7) 16.1 (4.0) 12.7 (1.9) 12.8 (2.0) 131.6 (37.1) 1947 (633.7)
Data are presented as mean (SD). IPL = intraprostatic urethral length; MRI = magnetic resonance imaging; MU = membranous urethra; MUL = membranous urethral length; MUT = membranous urethral thickness; SD = standard deviation. a (Transverse MUT 1/2) (anteroposterior MUT 1/2) p. b (Transverse MUT 1/2) (anteroposterior MUT 1/2) p coronal MUL.
The concordance rate between different observers was 89% for MUL on coronal view. However, the rates were demonstrated to be lower at 77%, 74%, 72%, and 62% for sagittal MUL, transverse MUT, anteroposterior MUT, and IPL, respectively. 4.
Discussion
RARP has increasingly been adopted to be the standard surgical treatment of choice for localised prostate cancer.
Fig. 2 – MUL measurement on MRI. (A) Short MUL on coronal view. (B) Long MUL on coronal view. MRI = magnetic resonance imaging; MUL = membranous urethral length.
However, incontinence remains to be one of the complications that are most difficult to manage. Continence relies on a complex mechanism and is multifactorial in origin. RARP leads to changes in the structure and function of the components inherently related to the urinary sphincter complex. It removes internal urethral sphincter located in the proximal portion of the prostate. It is composed of smooth muscle under the regulation of autonomic innervation and is continuous with the detrusor muscle. Several different techniques such as bladder neck sparing have been attempted to preserve the internal sphincter with little significant benefit. It has been shown that postoperative continence depends largely on the external urethral sphincter—rhabdosphincter [11]. It is innervated by a branch of the pudendal nerve. This is most prominent over the MU and forms a muscular coat over it in an omega-shaped loop. Its key role in continence is thought to be due to its ability to
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 5
E U RO P E A N U RO L O GY F O C U S X X X ( 2 018 ) X X X– X X X
Table 3 – Univariate and multivariate regression analysis of continence at 3 mo Predictive factors
Univariate
Age BMI ASA Prostate volume Preop PSA IPSS EBL Gleason score GG3 vs GG4 pT stage T2 vs T3 Nerve sparing (NS) Non-NS vs Uni-NS Non-NS vs Bil-NS MRI Coronal MUL Sagittal MUL IPL Transverse MUT Anteroposterior MUT Axial surface area MU Total volume MU
Multivariate
HR
95% CI
p value
HR
95% CI
p value
0.921 0.921 0.643 0.983 0.959 0.996 1.000 0.606
0.876–0.969 0.830–1.022 0.337–1.227 0.957–1.009 0.918–1.002 0.976–1.017 0.998–1.002 0.276–1.329
0.001 0.121 0.180 0.190 0.063 0.714 0.792 0.211
0.932 0.884
0.882–0.985 0.790–0.990
0.013 0.033
0.610
0.324–1.148
0.126
2.014 3.174
0.801–4.108 1.304–6.758
0.152 0.016
2.177 2.268
0.799–5.931 0.932–5.517
0.128 0.071
1.152 1.154 1.131 1.054 1.120 1.007 1.001
1.035–1.281 0.998–1.385 0.981–1.307 0.901–1.264 0.930–1.325 0.998–1.021 1.000–1.002
0.009 0.078 0.104 0.423 0.218 0.126 0.020
1.199
1.063–1.354
0.003
1.001
0.999–1.002
0.295
ASA = American Society of Anesthesiologists; Bil = bilateral; BMI = body mass index; CI = confidence interval; EBL = estimated blood loss; HR = hazard ratio; IPL = intraprostatic urethral length; IPSS = international prostate symptom score; MRI = magnetic resonance imaging; MU = membranous urethra; MUL = membranous urethral length; MUT = membranous urethral thickness; PSA = prostate-specific antigen; Uni = unilateral.
Table 4 – Univariate and multivariate regression analysis of continence at 12 mo Predictive factors
Age BMI ASA Prostate volume Preop PSA IPSS EBL Gleason score GG3 vs GG4 pT stage T2 vs T3 Nerve sparing (NS) Non-NS vs Uni-NS Non-NS vs Bil-NS MRI Coronal MUL Sagittal MUL IPL Transverse MUT Anteroposterior MUT Axial surface area MU Total volume MU
Univariate HR
95% CI
p value
0.830 0.959 0.674 0.993 0.969 0.997 0.998 0.621
0.718–0.972 0.820–1.046 0.335–1.230 0.960–1.027 0.910–1.002 0.981–1.020 0.994–1.002 0.256–1.337
0.007 0.128 0.185 0.667 0.068 0.701 0.278 0.231
0.587
0.132–2.418
0.446
1.230 1.848
0.240–5.214 0.306–7.821
0.809 0.567
1.425 1.847 1.115 1.057 1.132 1.008 1.001
1.187–1.869 0.789–3.692 0.939–1.324 0.653–1.483 0.721–1.686 0.998–1.013 1.000–1.003
0.001 0.095 0.213 0.645 0.531 0.118 0.108
ASA = American Society of Anesthesiologists; Bil = bilateral; BMI = body mass index; CI = confidence interval; EBL = estimated blood loss; HR = hazard ratio; IPL = intraprostatic urethral length; IPSS = international prostate symptom score; MRI = magnetic resonance imaging; MU = membranous urethra; MUL = membranous urethral length; MUT = membranous urethral thickness; PSA = prostate-specific antigen; Uni = unilateral.
increase urethral closure pressures voluntarily [12–14]. Maximal MUL sparing has been recommended to achieve maximal functional urethral length and shown to improve continence [10,15,16]. This would further facilitate more effective training of pelvic floor muscle. Therefore, MUL could serve as a surrogate marker for the rhabdosphincter responsible for stress urinary incontinence after radical prostatectomy. The MU also contains smooth muscle fibres that are supplied by autonomic innervation from the prostatic plexus and cavernous nerve. It is believed to influence a resting tone contributing to maintaining urethral closure pressures. Studies demonstrated improved continence in patients who underwent nerve-sparing prostatectomy. This is postulated to be due to more preserved membranous urethral vascular integrity and membranous urethral sensitivity from less autonomic afferent denervation [17–20]. Several other technical refinements have been proposed with heterogeneous outcomes. They include puboprostatic sparing, bladder neck preservation, pubovesical complex or dorsal vein complex preservation, and posterior musculofascial reconstruction, as well as anterior pelvic restoration and anterior reconstruction using the periurethral suspension stitch [21–23]. More recently, Retzius-sparing radical prostatectomy has been demonstrated to provide superior early continence outcome by preserving a number of the supportive and integrated structures mentioned above [24,25]. Strasser et al [12] demonstrated transrectal ultrasound as a useful tool to assess the quality of the rhabdosphincter postoperatively. They detected thinning, wasting, defects, or scarring in the majority of patients with
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 6
E U R O P E A N U R O L O GY F O C U S X X X ( 2 0 18 ) X X X– X X X
Fig. 3 – Area under the ROC curve for continence at 12 mo based on coronal MUL. MUL = membranous urethral length; ROC = receiver operating characteristic.
1.00 0.80
Probability
0.60 0.40 0.20 0.00
3.0
6.0
9.0
12.0
15.0
18.0
21.0
24.0
Fig. 4 – Probability of continence following RARP based on membranous urethral length (MUL). RARP = robot-assisted radical prostatectomy.
incontinence. The rhabdosphincter-urethra distance was significantly lower in the incontinent group than in the continent group. This would in part explain a much lower rate of continence in patients following salvage radical prostatectomy, which is likely a result of an extensive fibrosis from ischaemia and loss of dissection planes. Postoperative cystographic images were also used to predict continence outcome and showed that vesicourethral anastomosis location and bladder neck angle were significantly associated with continence [26,27]. Unlike the preoperative factors, anatomical variables can be improved by surgeon’s experience and technical modifications. The aforementioned technical modifications could theoretically improve these cystographic profiles and explain improved continence.
In terms of the significance of MUL for continence, Coakley et al [28] reported a continence rate of 89% if MUL was >12 mm and was decreased down to 77% if 12 mm.Our study showed that the risk of incontinence was 27.8% for MUL <10 mm and 2.7% for MUL >15 mm. The risk of incontinence was almost three times more likely if patients are aged >65 yr, again confirming its significant influence on continence. There are studies that developed nomograms incorporating MUL into preoperative factors to predict continence [7,29]. Matsushita et al [7] showed that the addition of MUL increased the area under the curve by 0.085 over the base model. However, there is a lack of information as to the significance of other parameters such as MUT, IPL, surface area, or volume of MU. Anatomical and functional studies showed the extension of
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 E U RO P E A N U RO L O GY F O C U S X X X ( 2 018 ) X X X– X X X
the rhabdosphincter from the MU up to the level of the verumontanum [10]. Although the amount of rhabdosphincter over the intraprostatic urethra would vary, this might theoretically be associated with continence. However, there was no significant difference in continence outcome based on IPL in our study. This could partly be explained by difficulty ascertaining an accurate measurement of its length due to variable thickness of slices performed on an MRI scan. We also postulate that the proportion of IPL in relation to MUL might also be associated with continence. Tienza and colleagues [9] showed that transverse MUT played a part in continence outcome. We measured MUT in its transverse and anteroposterior diameters, and there was no significant association with continence. When the surface area of the MU at the level of the penile bulb was measured, no association was again demonstrated. However, MU volume was significant on univariate analysis. This is most likely due to MUL being a strong influential factor. This study has several limitations. It was a small singlesurgeon series. Accuracy of the surface volume or volume measurement is questionable, as it was based at the level of the penile bulb and there is an anatomical variation along MUL in each individual. Difficulty was encountered in measuring MUL on a sagittal view and IPL partly because the slices were not calibrated thin enough. We also made a presumption that the urethra is of similar calibre and the MUT would correlate with the amount of rhabdosphincter axially. There was no postoperative MRI to identify accurately and objectively the residual functional MUL and extent of scarring. However, preoperative MUL was as significant as postoperative MUL [16]. Furthermore, the percent change of MUL from pre- to postoperative measurement did not have a significant impact on continence recovery at 12 mo [30]. Multivariate logistic regression analysis for continence at 12 mo was not performed due to a small number of events leading to overfitting if all the known predictors were included. Nevertheless, age and MUL were again significant on univariate analysis. Incontinence due to bladder dysfunction was not taken into account, although it would account for only a minority. We assessed various urethral parameters only, and did not take other parameters that were shown to be associated with incontinence into account such as bladder neck angle or descent in relation to pubic symphysis. There are several strengths worthy of mention. As the data were prospectively collected, a potential confounding bias was avoided. As this was a single-surgeon series, the technical factors that arise from the heterogeneity of different surgeon experience and skill set did not need to be taken into account. The analysis of the MRI parameters was performed by an independent uroradiologist with high-volume experience interpreting a prostate MRI scan. To our knowledge, this is the first study that is reported in the UK and evaluated interobserver variability between a radiologist, a urologist, and a trainee. This demonstrates the reliability of predicting continence based on the measurements performed by a urologist or a trainee.
5.
7
Conclusions
This study confirmed the significance of MUL for the continence outcome following RARP. There was also excellent consistency in measuring MUL values between different observers. While further larger prospective studies would be required to validate our findings, we suggest that in case of a high likelihood of incontinence, alternative treatment such as radiotherapy can be considered or early measures for definitive incontinence management in the form of urethral sling or artificial urinary sphincter can be planned. Author contributions: Lawrence H.C. Kim had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Kim, Patel, Cahill. Acquisition of data: Kim, Patel, Ap Dafydd, Cahill. Analysis and interpretation of data: Kim, Patel, Sharabiani, Cahill. Drafting of the manuscript: Kim, Patel, Cahill, Kinsella. Critical revision of the manuscript for important intellectual content: Kim, Kinsella, Cahill. Statistical analysis: Kim, Sharabiani. Obtaining funding: None. Administrative, technical, or material support: None. Supervision: Cahill. Other: None. Financial disclosures: Lawrence H.C. Kim certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: None.
References [1] Hofer MD, Meeks JJ, Cashy J, Kundu S, Zhao LC. Impact of increasing prevalence of minimally invasive prostatectomy on open prostatectomy observed in the national inpatient sample and national surgical quality improvement program. J Endourol 2013;27:102–7. [2] Kumar A, Samavedi S, Bates AS, et al. Continence outcomes of robotassisted radical prostatectomy in patients with adverse urinary continence risk factors. BJU Int 2015;116:764–70. [3] The British Association of Urological Surgeons. Analyses of Radical Prostatectomies performed between January 1st and December 31st 2016. https://www.baus.org.uk/_userfiles/pages/files/ Publications/Audit/Radical%20Prostatectomy%202016%20final% 20analysis.pdf [Accessed 10 January 2019]. [4] Punnen S, Cowan JE, Chan JM, Carroll PR, Cooperberg MR. Long-term health-related qualityof life after primary treatment forlocalized prostate cancer: results from the CaPSURE registry. Eur Urol 2015;68:600–8. [5] Oelrich TM. The urethral sphincter muscle in the male. Am J Anat 1980;158:229–46. [6] Ficarra V, Novara G, Rosen RC, et al. Systematic review and metaanalysis of studies reporting urinary continence recovery after robot-assisted radical prostatectomy. Eur Urol 2012;62:405–17. [7] Matsushita K, Kent MT, Vickers AJ, et al. Preoperative predictive model of recovery of urinary continence after radical prostatectomy. BJU Int 2015;116:577–83.
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011
EUF-662; No. of Pages 8 8
E U R O P E A N U R O L O GY F O C U S X X X ( 2 0 18 ) X X X– X X X
[8] Mungovan SF, Sandhu JS, Akin O, Smart NA, Graham PL, Patel MI.
[20] Park YH, Kwon OS, Hong S, Kim SW, Hwang T, Lee JY. Effect of nerve-
Preoperative membranous urethral length measurement and con-
sparing radical prostatectomy on urinary continence in patients with
tinence recovery following radical prostatectomy: a systematic review and meta-analysis. Eur Urol 2017;71:368–78.
preoperative erectile dysfunction. Int Neurourol J 2016;20:69–74. [21] Vora AA, Dajani D, Lynch JH, Kowalczyk KJ. Anatomic and technical
[9] Tienza A, Hevia M, Benito A, Pascual JI, Zudaire JJ, Robles JE. MRI
considerations for optimizing recovery of urinary function during
factors to predict urinary incontinence after retropubic/laparo-
robotic-assisted radical prostatectomy. Curr Opin Urol 2013;23:78–87.
scopic radical prostatectomy. Int Urol Nephrol 2015;47:1343–9. [10] Schlomm T, Heinzer H, Steuber T, et al. Full functional-length urethral sphincter preservation during radical prostatectomy. Eur Urol 2011;60:320–9.
[22] Sridhar AN, Abozaid M, Rajan P, et al. Surgical techniques to optimize early urinary continence recovery post robot assisted radical prostatectomy for prostate cancer. Curr Urol Rep 2017;18:71. [23] Walz J, Epstein JI, Ganzer R, et al. A critical analysis of the current
[11] Hoyland K, Vasdev N, Abrof A, Boustead G. Post-radical prostatectomy
knowledge of surgical anatomy of the prostate related to optimisa-
incontinence: etiology and prevention. Rev Urol 2014;16:181–8.
tion of cancer control and preservation of continence and erection
[12] Strasser H, Pinggera GM, Gozzi C, et al. Three-dimensional transrectal ultrasound of the male urethral rhabdosphincter. World J Urol 2004;22:335–8.
in candidates for radical prostatectomy: an update. Eur Urol 2016;70:301–11. [24] Dalela D, Jeong W, Prasad MA, et al. A pragmatic randomized
[13] Cameron AP, Suskind AM, Neer C, et al. Functional and anatomical
controlled trial examining the impact of the Retzius-sparing
differences between continent and incontinent men post radical
approach on early urinary continence recovery after robot-assisted
prostatectomy on urodynamics and 3T MRI: a pilot study. Neurourol Urodyn 2015;34:527–32. [14] Dubbelman YD, Bosch JL. Urethral sphincter function before and after radical prostatectomy: systematic review of the prognostic value of various assessment techniques. Neurourol Urodyn 2013;32:957–63. [15] Hamada A, Razdan S, Etafy MH, Fagin R, Razdan S. Early return of
radical prostatectomy. Eur Urol 2017;72:677–85. [25] Lim SK, Kim KH, Shin TY, et al. Retzius-sparing robot-assisted laparoscopic radical prostatectomy: combining the best of retropubic and perineal approaches. BJU Int 2014;114:236–44. [26] Shao IH, Chou CY, Huang CC, et al. A specific cystography pattern can predict postprostatectomy incontinence. Ann Surg Oncol 2015;22 (Suppl 3):S1580–6.
continence in patients undergoing robot-assisted laparoscopic
[27] Jeong SJ, Yi J, Chung MS, et al. Early recovery of urinary continence
prostatectomy using modified maximal urethral length preserva-
after radical prostatectomy: correlation with vesico-urethral anas-
tion technique. J Endourol 2014;28:930–8. [16] Paparel P, Akin O, Sandhu JS, et al. Recovery of urinary continence after radical prostatectomy: association with urethral fibrosis mea-
tomosis location in the pelvic cavity measured by postoperative cystography. Int J Urol 2011;18:444–51. [28] Coakley FV, Eberhardt S, Kattan MW, Wei DC, Scardino PT, Hricak H.
sured by preoperative and postoperative magnetic resonance imag-
Urinary continence after radical retropubic prostatectomy: rela-
ing. Eur Urol 2009;55:629–37.
tionship with membranous urethral length on preoperative endor-
[17] Kadono Y, Nohara T, Kawaguchi S, et al. Investigating the mechanism underlying urinary continence recovery after radical prosta-
ectal magnetic resonance imaging. J Urol 2002;168:1032–5. [29] Jeong SJ, Yeon JS, Lee JK, et al. Development and validation of
tectomy: effectiveness of a longer urethral stump to prevent urinary
nomograms to predict the recovery of urinary continence after
incontinence. BJU Int 2018;122:456–62.
radical prostatectomy: comparisons between immediate, early,
[18] John H, Suter S, Hauri D. Effect of radical prostatectomy on urethral blood flow. Urology 2002;59:566–9. [19] Catarin MV, Manzano GM, Nobrega JA, Almeida FG, Srougi M, Bruschini
and late continence. World J Urol 2014;32:437–44. [30] Song W, Kim CK, Park BK, et al. Impact of preoperative and postoperative membranous urethral length measured by 3 Tesla magnetic
H. The role of membranous urethral afferent autonomic innervation in
resonance imaging on urinary continence recovery after robotic-
the continence mechanism after nerve sparing radical prostatectomy: a
assisted radical prostatectomy. Can Urol Assoc J 2017;11:E93–9.
clinical and prospective study. J Urol 2008;180:2527–31.
Please cite this article in press as: Kim LHC, et al. Association Between Preoperative Magnetic Resonance Imaging–based Urethral Parameters and Continence Recovery Following Robot-assisted Radical Prostatectomy. Eur Urol Focus (2019), https://doi.org/ 10.1016/j.euf.2019.01.011