- Email: [email protected]
Device Survival after Primary Implantation of an Artificial Urinary Sphincter for Male Stress Urinary Incontinence
Author's Accepted Manuscript Device survival following primary implantation of the AMS 800 artificial urinary sphincter for male stress urinary incontinence Faysal A. Yafi , Kenneth J. DeLay , Carrie Stewart , Jason Chiang , Premsant Sangkum , Wayne J.G. Hellstrom
PII: DOI: Reference:
S0022-5347(16)31207-1 10.1016/j.juro.2016.08.107 JURO 13987
To appear in: The Journal of Urology Accepted Date: 23 August 2016 Please cite this article as: Yafi FA, DeLay KJ, Stewart C, Chiang J, Sangkum P, Hellstrom WJG, Device survival following primary implantation of the AMS 800 artificial urinary sphincter for male stress urinary incontinence, The Journal of Urology® (2016), doi: 10.1016/j.juro.2016.08.107. DISCLAIMER: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our subscribers we are providing this early version of the article. The paper will be copy edited and typeset, and proof will be reviewed before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to The Journal pertain.
Embargo Policy All article content is under embargo until uncorrected proof of the article becomes available online. We will provide journalists and editors with full-text copies of the articles in question prior to the embargo date so that stories can be adequately researched and written. The standard embargo time is 12:01 AM ET on that date. Questions regarding embargo should be directed to [email protected].
ACCEPTED MANUSCRIPT
Device survival following primary implantation of the AMS 800 artificial urinary sphincter for male stress urinary incontinence
a
RI PT
Faysal A. Yafi, MDa, Kenneth J. DeLay, MDa, Carrie Stewart, MDa, Jason Chiang, MDa, Premsant Sangkum, MDb, Wayne J.G. Hellstrom, MDa
M AN U
Corresponding Author: Wayne J.G. Hellstrom Tulane University Health Sciences Center Department of Urology 1430 Tulane Ave. 86-42 New Orleans, LA, 70112 USA Phone: +1 504 988 3361 Fax: +1 504 988 5059 Email: [email protected]
SC
Department of Urology, Tulane University School of Medicine, New Orleans, LA, USA Division of Urology, Department of Surgery, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand b
Running head: AUS device survival following primary implantation
TE D
Main text word count: 2316 Abstract word count: 237
AC C
EP
Keywords: urinary incontinence, artificial urinary sphincter, device survival, explantation, revision
ACCEPTED MANUSCRIPT
Abstract: Purpose: The AMS 800TM artificial urinary sphincter (AUS) remains the gold standard for the
RI PT
surgical management of male stress urinary incontinence. We reviewed AUS device survival following primary implantation.
Materials and Methods: Retrospective data was collected from the AMS 800TM patient
SC
information form (PIF) database. Since 1972, 77,512 PIFs for primary AUS implantation have been completed in the United States. Following exclusion of procedures performed in children
M AN U
and females, and those labeled with an unknown surgical technique, 27,096 AUS cases were included in the analysis. Collected variables included patient age, surgical approach, number of cuffs, and surgeon volume. Measured outcomes included device explantation, device revision, component revision, and time to each event.
TE D
Results: AUS insertion was performed by low-volume implanters in 22,165 (82.6%) of cases. Approach was perineal in 18,373 cases (67.8%), and a tandem cuff was used in 2,224 cases (8.2%). Overall, 5,723 cases required either revision or explantation (21.1%). Younger age and
EP
penoscrotal approach were associated with higher device explantation and revision rates, while use of a tandem cuff was associated with higher explantation rates. On multivariate analysis,
AC C
younger age, penoscrotal approach, and use of a tandem cuff, but not surgeon volume, were significant factors associated with device explantation and component revision. Conclusions: These data provide a general overview on AUS device survival, and may serve urologists when counseling patients. Younger age, penoscrotal approach, and use of tandem cuff may be associated with inferior outcomes.
2
ACCEPTED MANUSCRIPT
Introduction The artificial urinary sphincter (AUS), first reported by Dr. Brantley Scott in 1972, has become the gold standard surgical procedure for men with moderate to severe stress urinary incontinence
RI PT
(SUI) 1. Its newest iteration, the AMS 800TM (American Medical Systems, Boston Scientific, Marlborough, MA, USA) consists of a cuff placed around the bulbar urethra, a pressure-
regulating balloon (PRB), and a scrotal pump. It can be placed through either a perineal or a
SC
penoscrotal approach.
SUI in adult men most often occurs following radical prostatectomy for prostate
M AN U
adenocarcinoma, with rates of post-prostatectomy incontinence of 2.5% to 40% 2. Following implantation of an AUS, continence rates range from 73% to 88%, depending on the study design and definition of continence 3-7. Infection, erosion, and mechanical failure are the most commonly reported post-operative complications, requiring device revision, explantation, or
TE D
replacement 8-10.
A recent systematic review found that 5-year device survival rates following AUS insertion ranged from 59% to 79% 8. Consistent with the literature on the topic, this study was
EP
limited by a small series lacking long-term follow-up. Factors that have been believed to impact device survival include surgical approach, history of radiation, surgeon experience, presence of
AC C
tandem cuffs, and cuff size 11-16.
We reviewed the largest database to date to assess device survival following primary
implantation of the AMS 800TM AUS. Furthermore, we sought to examine various variables that may influence device survival.
Materials and methods
3
ACCEPTED MANUSCRIPT
Patient selection Retrospective data was collected from the AMS 800TM patient information form (PIF) database. From January 1972 to September 2015, there were 201,637 PIFs for AUS surgery completed
RI PT
worldwide, of which 77,512 were primary implantations in the United States. Following the exclusion of procedures performed in children and females, and those labeled with an unknown surgical technique, 27,096 AUS cases were included in the analysis.
SC
PIF databases do not systematically capture many salient variables such as patient
demographics (besides age), clinical characteristics (previous radiotherapy, previous urethral
M AN U
surgery, etc.), device characteristics (size of cuff, size of PRB, etc.), and certain surgical details (transcorporal cuff placement, location of PRB, etc.). However, consistently available variables were collected and used for analysis. These included patient age, surgical approach (perineal vs. penoscrotal), number of cuffs (single vs. tandem), and surgeon volume (low [<10 AUS/year] vs.
of data.
TE D
high [≥10 AUS/year]). Low vs. high volume implanter definition was determined after analysis
Complaint forms were then used to assess for post-implantation device complications.
EP
Similarly, these forms are completely voluntary and are filled in at the time of revision and/or explantation surgery. As such, they are often lacking information regarding pertinent
AC C
complications, such as mechanical failure and urinary retention, amongst others. Consistently available ones that were included in the analysis were cuff erosion, fluid loss, sub-cuff atrophy, PRB herniation, and infection. Finally, measured outcomes included device explantation and revision, and time to each of these events. Explantation was defined as any surgical intervention requiring removal of all components of the device. Revision was defined as any surgical intervention that removed one or more but not all of the device components. Component revision
4
ACCEPTED MANUSCRIPT
was defined as any surgical intervention requiring either revision or explantation of that particular component. Non-surgical clinical outcomes such as dry rates, number of pads per day,
RI PT
and patient satisfaction, amongst others, were not uniformly available for collection.
Statistical analysis
Continuous variables were summarized with mean and standard deviation; categorical variables
SC
were summarized with frequency counts and percentages. Multivariable Cox proportional
hazards models were used to estimate hazard ratios (HRs) for associations between clinical and
M AN U
surgical characteristics and post-operative complications, explantation and/or revision. Probability of device survival was estimated using the Kaplan-Meier method with time from AUS primary implantation to the subsequent re-surgery intervention (including explantation and/or revision for any reason); patients were censored at the time of dataset cutoff
TE D
date if no event occurred. All statistical tests were 2-sided, with a P value < 0.05 considered statistically significant. Statistical analysis was performed using the SAS 9.4 software (SAS
Results
EP
Institute, Inc., Cary, NC).
AC C
Overall characteristics
Mean age of the population was 68.6 years (SD 8.9). Surgery was performed with high- and lowvolume implanters in 4,666 (17.4%) and 22,165 (82.6%) of cases, respectively. Approach was perineal in 18,373 cases (67.8%) and penoscrotal in 8,723 cases (32.2%), and a tandem cuff was used in 2,224 cases (8.2%) (table 1).
5
ACCEPTED MANUSCRIPT
Post-operative complications Post-implantation complications were as follows: cuff erosion (1,085, 4.0%), loss of fluid (1,020, 3.8%), sub-cuff atrophy (661, 2.4%), device infection (518, 1.8%), and PRB herniation (49,
RI PT
0.2%). Other or unspecified complications were reported in 14.9% of patients (table 2). Younger age was associated with higher rates of sub-cuff atrophy, fluid loss, and PRB herniation, but less cuff erosion. A penoscrotal surgical approach and use of a tandem cuff were both significantly
SC
associated with higher rates of device infection, cuff erosion, and fluid loss. Finally, device
insertion by low-volume implanters was significantly associated with higher cuff erosion rates.
M AN U
Similarly, on multivariate analysis, younger patient age was again a significant independent predictor of high rates of sub-cuff atrophy, fluid loss, and PRB herniation, but less cuff erosion; penoscrotal incision was a predictor of infection and erosion; tandem cuff placement a predictor of fluid loss, erosion, and infection; and a low-volume surgeon a predictor
Explantation and revision
TE D
of erosion (table 3).
EP
Overall, 5,723 patients (21.1%) required explantation and/or revision, including 52.0% (2,978/5,723) of explantation only, 37.7% (2,160/5,723) of revision only, and 10.2% (585/5,723)
AC C
of explantation and revision. Younger age, penoscrotal approach, use of tandem cuff and lowvolume implanter were significantly correlated with higher explantation and component revision rates (table 4).
Similarly, on multivariate analysis, younger age, penoscrotal approach, use of tandem
cuffs, but not surgeon volume, were independent predictors of higher device explantation and component revision (table 5).
6
ACCEPTED MANUSCRIPT
Device survival On Kaplan-Meier survival analysis, 5- and 10-year explantation-free device survival rates were
RI PT
87.1% and 78.3%, respectively (figure 1a). The 5- and 10-year explantation and/or revision-free device survival rates were 78.1% and 68.7%, respectively (figure 1b).
SC
Discussion
Results of this large retrospective study demonstrate that the AMS 800TM AUS continues to be a
M AN U
reliable, durable option for men with stress urinary incontinence. Younger age, penoscrotal approach, and use of a tandem cuff are associated with significantly inferior overall outcomes. In our study, 21.1% of cases required re-intervention by way of device explantation and/or revision. This number is slightly lower than the 26.0% re-intervention rate reported in the
TE D
aforementioned systematic review of the literature by Van der Aa et al 8. Similarly, in the largest, single-center study to date, Linder et al demonstrated an even higher re-operation rate of 31.2% following primary perineal AUS implantation in 1,082 men 10. It is worth noting that, while not
EP
shown to be associated with worse outcomes in their study, a significant number of patients in their cohort had associated co-morbidities and urethra-compromising factors (27% prior pelvic
AC C
radiation, 32% prior vesico-urethral anastomotic stricture). Furthermore, in the aforementioned study, device survival rates, free of any revision, were 74% at 5 years and 57% at 10 years, consistent with other previously published reports 3,7,9,10,15,17-21. In the present study, however, 5and 10-year device survival rates were relatively higher (78.1% and 68.7%, respectively). One likely explanation for this discrepancy is the fact that explantation and/or revision events were collected from the manufacturer’s patient information forms. These forms were optional and are
7
ACCEPTED MANUSCRIPT
filled at the time of surgical revision. It is reasonable to assume that a fair number of these forms may have been incompletely filled out, thus leading to under-reporting of complications and resulting in better outcomes.
RI PT
Two-thirds of cases performed in this series were implanted through a perineal approach. This is consistent with the recently published 2015 International Continence Society (ICS) consensus conference on AUS which states that the perineal approach is preferred to the
SC
penoscrotal one (C grade of recommendation) 22. This stems from the perception that a
penoscrotal approach does not allow for optimal bulbar urethral exposure and proximal
M AN U
placement of the cuff. Our results seem to support this notion, with the penoscrotal approach being associated with higher erosion and infection rates, leading to higher explantation and revision rates. This is in line with a retrospective review of 126 patients that demonstrated significantly better continence and less need for tandem cuff placement with the perineal
TE D
compared to penoscrotal approach 23. On the other hand, Wilson reported excellent outcomes with the use a modified perineal approach for cuff placement, with no reported complications 24. Currently, the ICS guidelines recommend that the penoscrotal approach be reserved for redo
EP
cases, and in patients with conditions that preclude them from being placed in lithotomy position, such as morbid obesity, spine or limb deformities, or neuro-motor conditions, as well as those
AC C
undergoing dual AUS and inflatable penile prosthesis placement through a single penoscrotal incision 22.
In our series, a tandem cuff was placed at the time of initial implantation in 8.2% of
cases. While the merits of using a tandem cuff in cases of revision surgery for cuff erosion or sub-cuff atrophy have been well reported 25,26, its role in patients undergoing primary AUS implantation is uncertain. In a long-term follow-up study of 47 men with post-prostatectomy
8
ACCEPTED MANUSCRIPT
incontinence (25 single and 22 tandem cuffs), O’Connor et al. showed that the use of a tandem cuff imparted no significant differences in dry rates, overall continence, or quality of life as compared to a single cuff 15. Furthermore, these authors demonstrated that men receiving double
RI PT
cuff implants were at higher risk of complications requiring revision surgery. Similarly, in our study, placement of a tandem cuff was a significant predictor of sub-cuff atrophy, cuff erosion and device infection, and ultimately higher explantation and revision rates.
SC
In our study, a younger mean age at the time of surgery appeared to be associated with significantly higher rates of sub-cuff atrophy, PRB herniation, and fluid loss, as well as higher
M AN U
explantation and revision rates. It is unclear why younger age would lead to higher complication rates. One potential explanation is that younger patients in this series had longer documented follow-ups, and consequentially a higher chance for complications and need for revision surgery. There are limited reports that have specifically examined the effect of age on AUS outcomes . While one study did demonstrate that octogenarians were more likely to experience erosion
TE D
27,28
or infection compared with younger patients 28, the 5-year device survival rates were comparable to those reported in younger men (63% to 70%) 27,28. As such, current evidence remains
age.
EP
insufficient to recommend making decisions on AUS placement and outcomes solely based on
AC C
Finally, we defined a high volume surgeon as any surgeon who performs 10 or more
implants per year. Using this definition, we noted that high volume surgeons accounted for 17.4% of all primary AUS implantations. Furthermore, we observed a small association between greater surgeon experience and decreased cuff erosion, device explantation, and component revision. After accounting for other variables, however, surgeon volume was not a significant predictor of complications or need for revision surgery. While one would expect that surgeon
9
ACCEPTED MANUSCRIPT
experience, a likely surrogate of advanced training, would be a predictor of improved outcomes, there are a few theories as to why this benefit was not observed in our study. First and foremost, it is likely that surgeons with higher volume are more specialized urologists working in large
RI PT
academic centers who undertake more complex cases with expectedly higher complication rates. Second, the cut-off value for surgeon volume might be too high. In a case log analysis from certifying and recertifying urologists obtained from the American Board of Urology (ABU)
SC
between 2004 and 2010, among urologists who placed any sphincters, the median number of cases was only two per year with only 4% of urologists implanting ≥10 29. Similarly, in a more
M AN U
recent 6-months case log data of certifying urologists (2003-2013) obtained from the ABU, amongst the 90th percentile of certifying urologists performing AUS, four cases were logged over 6 months, whereas the top 4% (comprising 42 urologists) logged only eight cases during this time period 30. The number of surgeons who fit our definition of a “high-volume implanter”
TE D
(≥10 AUS/year) in this series is only 18 out of 4092 (0.4%). Using stricter cut-offs of 5 AUS/year or 2 AUS/year would have resulted in 54 (1.3%) and 326 (8.0%) high-volume implanters. Accordingly, it could be argued that using a lower threshold for a low vs. high
EP
volume surgeon may have yielded more definite results. Finally, this data puts in perspective the fact that most urologists implanting AUS only perform a few cases per year.
AC C
This study has limitations. First and foremost, it is based on retrospective data collected
from patient information forms that are filled in the operative room at the time of implantation by the surgeon, the circulating nurse or the product specialist. These forms mostly require “ticking a box” but also contain free text sections, which are naturally difficult to capture when reviewing large databases. Furthermore, these optional forms are often incompletely filled. As noted, outcomes were obtained from the manufacturer’s complaint forms, which are similarly filled in
10
ACCEPTED MANUSCRIPT
at the time of revision and/or explantation surgery. These are frequently incomplete, thus affecting outcomes. Finally, as previously mentioned, many important variables relating to patient, clinical, surgical, and post-surgical characteristics, such as continence rates, were either
RI PT
not captured or not deemed reliable after assessment, and were excluded from analysis. As such, the results of this study may not allow for strict clinical recommendations. This study’s—the largest collected AUS database to date for the management of SUI in adult men—main purposes,
SC
however, are to offer a general overview on device survival, and serve as a counseling tool for
Conclusions
M AN U
urologists when discussing surgical options with their patients.
These results provide a general overview on AUS device survival, and may serve urologists when counseling patients. Younger age, penoscrotal approach, and use of tandem cuff may be
References
EP
TE D
associated with inferior outcomes.
AC C
1. Scott FB, Bradley WE, Timm GW. Treatment of urinary incontinence by an implantable prosthetic urinary sphincter. J Urol. 1974; 112(1); 75-80.
2. Begg CB, Riedel ER, Bach PB, et al. Variations in morbidity after radical prostatectomy. N Engl J Med. 2002; 346; 1138-44.
3. Lai HH, Hsu EI, Teh BS, Butler EB, Boone TB. 13 years of experience with artificial urinary sphincter implantation at Baylor College of Medicine. J Urol. 2007; 177; 1021-5.
11
ACCEPTED MANUSCRIPT
4. Petero VG Jr, Diokno AC. Comparison of the long-term outcomes between incontinent men and women treated with artificial urinary sphincter. J Urol. 2006; 175; 605-9.
RI PT
5. Montague DK, Angermeier KW, Paolone DR. Long-term continence and patient satisfaction after artificial sphincter implantation for urinary incontinence after prostatectomy. J Urol. 2001; 166; 547-9.
SC
6. Gousse AE, Madjar S, Lambert MM, Fishman IJ. Artificial urinary sphincter for postradical prostatectomy urinary incontinence: long-term subjective results. J Urol. 2001;
M AN U
166: 1755-8.
7. Venn SN, Greenwell TJ, Mundy AR. The long-term outcome of artificial urinary sphincters. J Urol. 2000; 164: 702-6.
8. Van der Aa F, Drake MJ, Kasyan GR, Petrolekas A, Cornu JN. The artificial urinary sphincter after a quarter of a century: a critical systematic review of its use in male non-
TE D
neurogenic incontinence. Eur Urol. 2013; 63: 681–9. 9. Raj GV, Peterson AC, Toh KL, Webster GD. Outcomes following revisions and secondary implantation of the artificial urinary sphincter. J Urol. 2005; 173(4): 1242–5.
EP
10. Linder BJ, Rivera ME, Ziegelmann MJ, Elliott DS. Long-term Outcomes Following
AC C
Artificial Urinary Sphincter Placement: An Analysis of 1082 Cases at Mayo Clinic. Urology. 2015; 86(3): 602-7.
11. Henry GD, Graham SM, Cornell RJ, et al. A multicenter study on the perineal versus penoscrotal approach for implantation of an artificial urinary sphincter: cuff size and control of male stress urinary incontinence. J Urol.2009; 182(5): 2404–9.
12
ACCEPTED MANUSCRIPT
12. Sandhu JS, Maschino AC, Vickers AJ. The surgical learning curve for artificial urinary sphincter procedures compared to typical surgeon experience. Eur Urol. 2011; 60: 1285– 90.
RI PT
13. Lai HH, Boone TB. Complex artificial urinary sphincter revision and reimplantation cases--how do they fare compared to virgin cases? J Urol 2012; 187: 951-5.
14. Gomha MA, Boone TB. Artificial urinary sphincter for post-prostatectomy incontinence
SC
in men who had prior radiotherapy: A risk and outcome analysis. J Urol. 2002; 167: 5916.
M AN U
15. O'Connor RC, Lyon MB, Guralnick ML, Bales GT. Long-term follow-up of single versus double cuff artificial urinary sphincter insertion for the treatment of severe postprostatectomy stress urinary incontinence. Urology. 2008; 71: 90–3. 16. Yafi FA, Powers MK, Zurawin J, Hellstrom WJ. Contemporary review of artificial
TE D
urinary sphincters for male stress urinary incontinence. Sex Med Rev. 2016; 4: 157-6. 17. Leon P, Chartier-Kastler E, Roupret M,.Ambrogi V, Mozer P, Phe V. Long-term functional outcomes after artificial urinary sphincter implantation in men with stress
EP
urinary incontinence. BJU Int. 2014; 115: 951-7. 18. Linder BJ, de Cogain M, Elliott DS. Long-term device outcomes of artificial urinary
AC C
sphincter reimplantation following prior explantation for erosion or infection. J Urol. 2014; 191: 734-8.
19. Guralnick ML, Miller E, Toh KL, Webster GD. Transcorporal artificial urinary sphincter cuff placement in cases requiring revision for erosion and urethral atrophy. J Urol. 2002; 167: 2075-8.
13
ACCEPTED MANUSCRIPT
20. Magera JS, Elliott DS. Tandem transcorporal artificial urinary sphincter cuff salvage technique: surgical description and results. J Urol. 2007; 177: 1015-9. 21. Kim SP, Sarmast Z, Daignault S, Faerber GJ, McGuire EJ, Latini JM.. Long-term
RI PT
durability and functional outcomes among patients with artificial urinary sphincters: a 10year retrospective review from the University of Michigan. J Urol. 2008; 179: 1912-6. 22. Biardeau X, Aharony S, AUS Consensus Group, Campeau L, Corcos J. Artificial Urinary
SC
Sphincter: Report of the 2015 Consensus Conference. Neurourol Urodyn. 2016; 35 Suppl 2: S8-S24.
M AN U
23. Henry GD, Graham SM, Cleves MA, Simmons CJ, Flynn B. Perineal approach for artificial urinary sphincter implantation appears to control male stress incontinence better than the transscrotal approach. J Urol. 2008; 179(4): 1475-9; discussion 1479. 24. Wilson SK, Delk JR, Henry GD, Siegel AL. New surgical technique for sphincter urinary
TE D
control system using upper transverse scrotal incision. J Urol 2003; 169: 261. 25. Brito CG, Mulcahy JJ, Mitchell ME, Adams MC. Use of a double cuff AMS800 urinary sphincter for severe stress incontinence. J Urol. 1993; 149(2): 283-5.
EP
26. DiMarco DS, Elliott DS. Tandem cuff artificial urinary sphincter as a salvage procedure following failed primary sphincter placement for the treatment of post-prostatectomy
AC C
incontinence. J Urol. 2003; 170(4 Pt 1): 1252-4.
27. O'Connor RC, Nanigian DK, Patel BN, Guralnick ML, Ellision LM, Stone AR. Artificial urinary sphincter placement in elderly men. Urology. 2007; 69(1): 126-8.
28. Ziegelmann MJ, Linder BJ, Rivera ME, Viers BR, Rangel LJ, Elliott DS. Int J Urol. 2016; 23(5): 419-23. Outcomes of artificial urinary sphincter placement in octogenarians.
14
ACCEPTED MANUSCRIPT
29. Poon SA, Silberstein JL, Savage C, Maschino AC, Lowrance WT, Sandhu JS. Surgical practice patterns for male urinary incontinence: analysis of case logs from certifying American urologists. J Urol. 2012; 188(1): 205-10.
RI PT
30. Liu JS, Hofer MD, Milose J, et al. Male Sling and Artificial Urethral Sphincter for Male Stress Urinary Incontinence Among Certifying American Urologists. Urology. 2016; 87:
M AN U
SC
95-9.
Figure legends:
Figure 1a. Artificial urinary sphincter device explanation free rate estimated by Kaplan Meier
TE D
method.
Figure 1b. Artificial urinary sphincter device explanation and/or revision free rate estimated by
AC C
EP
Kaplan Meier method.
15
ACCEPTED MANUSCRIPT
Table 1. Clinical and surgical characteristics of 27,096 adult men undergoing primary implantation of AMS 800 artificial urinary sphincter Total N=27,096 68.6 ± 8.9
RI PT
Mean age at surgery (years ± SD) Surgical Approach
N=27,096
Peno-scrotal Perineal
8,723 (32.2%) 18,373 (67.8%) N=27,033
SC
Tandem Cuff
24,809 (91.8%) 2,224 (8.2%)
Physician volume
AC C
EP
TE D
Low (< 10 surgeries/year) High (≥ 10 surgeries/year)
M AN U
No Yes
N=26,831
22,165 (82.6%) 4,666 (17.4%)
ACCEPTED MANUSCRIPT
68.6 ± 8.9
66.8 ± 8.8
2.8 ± 2.4 <0.001
68.7 ± 8.9
65.8 ± 8.7
0.231
8,496 (97.4%)
227 (2.6%)
8,361 (95.9%)
362 (4.1%)
Perineal
17,939 (97.6%)
434 (2.4%)
17,715 (96.4%)
658 (3.6%)
0.628
<0.001
68.6 ± 8.9
24,199 (97.5%)
610 (2.5%)
23,937 (96.5%)
872 (3.5%)
Yes
2,173 (97.7%)
51 (2.3%)
2,077 (93.4%)
147 (6.6%)
0.328
65.0 ± 9.2
21,323 (96.2%)
Yes
4,542 (97.3%)
124 (2.7%)
4,497 (96.4%)
No, N=26,011 (96.0%)
1.4 ± 1.9
0.006
68.6 ± 8.9
67.9 ± 10.1
Yes, N=1,085 (4.0%)
0.089
68.5 ± 8.9
69.7 ± 8.4
<0.001
227 (2.6%)
8,265 (94.7%)
458 (5.3%)
18,342 (99.8%)
31 (0.2%)
18,082 (98.4%)
291 (1.6%)
17,746 (96.6%)
627 (3.4%)
2,222 (99.9%)
<0.001
<0.001
47 (0.2%)
24,373 (98.2%)
436 (1.8%)
23,892 (96.3%)
917 (3.7%)
2 (0.1%)
2,145 (96.4%)
79 (3.6%)
2,060 (92.6%)
164 (7.4%)
0.764
0.098
<0.001
<0.001
8,496 (97.4%)
0.434
P Value
2.3 ± 2.3
18 (0.2%)
24,762 (99.8%)
EP
535 (2.4%)
<0.001
842 (3.8%)
22,121 (99.8%)
44 (0.2%)
21,737 (98.1%)
428 (1.9%)
21,212 (95.7%)
953 (4.3%)
169 (3.6%)
4,662 (99.9%)
4 (0.1%)
4,579 (98.1%)
87 (1.9%)
4,537 (97.2%)
129 (2.8%)
AC C
21,630 (97.6%)
Cuff Erosion P Value
8,705 (99.8%)
0.564
No
Yes, N=518 (1.9%)
0.496
TE D
No
High Volume Physician
<0.001
0.022
Penoscrotal
Tandem Cuff
1.7 ± 2.2
SC
4.2 ± 2.7
M AN U
Mean years to event ± SD Mean age at Surgery ± SD Surgical Approach
Infection No, N=26,578 (98.1%)
RI PT
Table 2. Complications following primary implantation of AMS 800 artificial urinary sphincter in 27,096 adult men Sub-cuff atrophy Fluid loss Herniation No, Yes, P No, Yes, P No, Yes, P N=26,435 N=661 Value N=26,076 N=1,020 Value N=27,047 N=49 Value (97.6%) (2.4%) (96.2%) (3.8%) (99.8%) (0.2%)
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 3. Multivariate Cox Proportional Hazard regression models for complications following implantation of AMS 800 artificial urinary sphincter in 27,096 adult men Cuff atrophy Fluid loss Cuff Erosion Infection PRB herniation Variable HR P HR P HR P HR P HR P Age 0.980 <0.001 0.970 <0.001 1.013 <0.001 0.991 0.064 0.963 0.004 Perineal 1.116 0.189 1.015 0.823 0.785 <0.001 0.669 <0.001 1.000 0.999 approach Tandem 0.710 0.020 1.581 <0.001 1.626 <0.001 1.769 <0.001 0.421 0.232 cuff High 1.151 0.164 1.000 0.995 0.730 0.001 1.095 0.452 0.422 0.101 volume surgeon
ACCEPTED MANUSCRIPT
Mean years to event ± SD Mean age at Surgery ± SD Surgical Approach
68.8 ± 8.9
Penoscrotal l
7,242 (83.0%)
1,481 (17.0%)
7,718 (88.5%)
1,005 (11.5%)
Perineal
16,291 (88.7%)
2,082 (11.3%)
16,633 (90.5%)
1,740 (9.5%)
<0.00 1
68.7 ± 8.9
2.6 ± 2.4
67.5 ± 8.9
<0.00 1
<0.00 1
Tandem Cuff
<0.00 1
<0.00 1
0.356
21,736 (87.6%)
3,073 (12.4%)
22,280 (89.8%)
2,529 (10.2%)
Yes
1,738 (78.1%)
486 (21.9%)
2,011 (90.4%)
213 (9.6%)
19,176 (86.5%)
2,989 (13.5%)
Yes
4,119 (88.3%)
547 (11.7%)
19,920 (89.9%)
4,177 (89.5%)
<0.00 1
68.8 ± 8.9
67.2 ± 8.9
Yes, N=3,94 7 (14.6%)
<0.00 1
68.8 ± 8.9
67.3 ± 8.9
<0.00 1
7,072 (81.1%)
1,651 (18.9%)
7,061 (80.9%)
1,662 (19.1%)
15,223 (82.9%)
3,150 (17.1%)
15,941 (86.8%)
2432 (13.2%)
16,088 (87.6%)
2,285 (12.4%)
<0.00 1
<0.00 1
20,189 (81.4%)
4,620 (18.6%)
21,259 (85.7%)
3,550 (14.3%)
21,409 (86.3%)
3,400 (13.7%)
1,645 (74.0%)
579 (26.0%)
1,695 (76.2%)
529 (23.8%)
1,681 (75.6%)
543 (24.4%)
<0.00 1
<0.00 1 <0.00 1
2,056 (23.6%)
<0.00 1
P Value
2.9 ± 2.6
<0.00 1
<0.00 1
2,245 (10.1%)
17,866 (80.6%)
4,299 (19.4%)
18,772 (84.7%)
3,393 (15.3%)
18,830 (85.0%)
3,335 (15.0%)
489 (10.5%)
3793 (81.3%)
873 (18.7%)
4,007 (85.9%)
659 (14.1%)
4,083 (87.5%)
583 (12.5%)
AC C
No
No, N=23,14 9 (85.4%)
6,667 (76.4%)
0.471
EP
<0.00 1
67.3 ± 8.9
<0.00 1
TE D
No
High Volume Physician
68.9 ± 8.9
Pump revision/explantation
2.9 ± 2.6
M AN U
67.2 ± 8.9
1.8 ± 1.9
SC
3.0 ± 2.6
RI PT
Table 4. Explantation and revision rates following primary implantation of AMS 800 artificial urinary sphincter in 27,096 adult men Any explantation Any revision Cuff revision/explantation PRB revision/explantation No, Yes, P No, Yes, P No, Yes, P No, Yes, P N=23,53 N=3,56 Value N=24,35 N=2,74 Value N=23,01 N=4,08 Value N=21,89 N=5,20 Value 3 3 1 5 3 3 0 6 (86.9%) (13.1%) (89.9%) (10.1%) (84.9%) (15.1%) (80.8%) (19.2%)
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 5. Multivariate Cox Proportional Hazard regression models for device explantation and revision following AMS 800 artificial urinary sphincter implantation in 27,096 adult men Any Any revision Cuff Pump PRB explantation revision/explantation revision/explantation revision/explantation Variable HR P HR P HR P HR P HR P Age 0.984 <0.001 0.987 <0.001 0.985 <0.001 0.984 <0.001 0.984 <0.001 Perineal 0.806 <0.001 0.887 0.003 0.835 <0.001 0.780 <0.001 0.828 <0.001 approach Tandem 1.414 <0.001 0.826 0.008 1.141 0.003 1.449 <0.001 1.358 <0.001 cuff High0.968 0.493 1.077 0.146 1.053 0.170 0.923 0.079 1.016 0.718 volume
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
artificial urinary sphincter
PIF
patient information form
SUI
stress urinary incontinence
PRB
pressure-regulating balloon
HRs
hazard ratios
AC C
EP
TE D
M AN U
SC
AUS
RI PT
Keywords
PII: DOI: Reference:
S0022-5347(16)31207-1 10.1016/j.juro.2016.08.107 JURO 13987
To appear in: The Journal of Urology Accepted Date: 23 August 2016 Please cite this article as: Yafi FA, DeLay KJ, Stewart C, Chiang J, Sangkum P, Hellstrom WJG, Device survival following primary implantation of the AMS 800 artificial urinary sphincter for male stress urinary incontinence, The Journal of Urology® (2016), doi: 10.1016/j.juro.2016.08.107. DISCLAIMER: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our subscribers we are providing this early version of the article. The paper will be copy edited and typeset, and proof will be reviewed before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to The Journal pertain.
Embargo Policy All article content is under embargo until uncorrected proof of the article becomes available online. We will provide journalists and editors with full-text copies of the articles in question prior to the embargo date so that stories can be adequately researched and written. The standard embargo time is 12:01 AM ET on that date. Questions regarding embargo should be directed to [email protected].
ACCEPTED MANUSCRIPT
Device survival following primary implantation of the AMS 800 artificial urinary sphincter for male stress urinary incontinence
a
RI PT
Faysal A. Yafi, MDa, Kenneth J. DeLay, MDa, Carrie Stewart, MDa, Jason Chiang, MDa, Premsant Sangkum, MDb, Wayne J.G. Hellstrom, MDa
M AN U
Corresponding Author: Wayne J.G. Hellstrom Tulane University Health Sciences Center Department of Urology 1430 Tulane Ave. 86-42 New Orleans, LA, 70112 USA Phone: +1 504 988 3361 Fax: +1 504 988 5059 Email: [email protected]
SC
Department of Urology, Tulane University School of Medicine, New Orleans, LA, USA Division of Urology, Department of Surgery, Faculty of Medicine, Ramathibodi Hospital, Mahidol University, Bangkok, Thailand b
Running head: AUS device survival following primary implantation
TE D
Main text word count: 2316 Abstract word count: 237
AC C
EP
Keywords: urinary incontinence, artificial urinary sphincter, device survival, explantation, revision
ACCEPTED MANUSCRIPT
Abstract: Purpose: The AMS 800TM artificial urinary sphincter (AUS) remains the gold standard for the
RI PT
surgical management of male stress urinary incontinence. We reviewed AUS device survival following primary implantation.
Materials and Methods: Retrospective data was collected from the AMS 800TM patient
SC
information form (PIF) database. Since 1972, 77,512 PIFs for primary AUS implantation have been completed in the United States. Following exclusion of procedures performed in children
M AN U
and females, and those labeled with an unknown surgical technique, 27,096 AUS cases were included in the analysis. Collected variables included patient age, surgical approach, number of cuffs, and surgeon volume. Measured outcomes included device explantation, device revision, component revision, and time to each event.
TE D
Results: AUS insertion was performed by low-volume implanters in 22,165 (82.6%) of cases. Approach was perineal in 18,373 cases (67.8%), and a tandem cuff was used in 2,224 cases (8.2%). Overall, 5,723 cases required either revision or explantation (21.1%). Younger age and
EP
penoscrotal approach were associated with higher device explantation and revision rates, while use of a tandem cuff was associated with higher explantation rates. On multivariate analysis,
AC C
younger age, penoscrotal approach, and use of a tandem cuff, but not surgeon volume, were significant factors associated with device explantation and component revision. Conclusions: These data provide a general overview on AUS device survival, and may serve urologists when counseling patients. Younger age, penoscrotal approach, and use of tandem cuff may be associated with inferior outcomes.
2
ACCEPTED MANUSCRIPT
Introduction The artificial urinary sphincter (AUS), first reported by Dr. Brantley Scott in 1972, has become the gold standard surgical procedure for men with moderate to severe stress urinary incontinence
RI PT
(SUI) 1. Its newest iteration, the AMS 800TM (American Medical Systems, Boston Scientific, Marlborough, MA, USA) consists of a cuff placed around the bulbar urethra, a pressure-
regulating balloon (PRB), and a scrotal pump. It can be placed through either a perineal or a
SC
penoscrotal approach.
SUI in adult men most often occurs following radical prostatectomy for prostate
M AN U
adenocarcinoma, with rates of post-prostatectomy incontinence of 2.5% to 40% 2. Following implantation of an AUS, continence rates range from 73% to 88%, depending on the study design and definition of continence 3-7. Infection, erosion, and mechanical failure are the most commonly reported post-operative complications, requiring device revision, explantation, or
TE D
replacement 8-10.
A recent systematic review found that 5-year device survival rates following AUS insertion ranged from 59% to 79% 8. Consistent with the literature on the topic, this study was
EP
limited by a small series lacking long-term follow-up. Factors that have been believed to impact device survival include surgical approach, history of radiation, surgeon experience, presence of
AC C
tandem cuffs, and cuff size 11-16.
We reviewed the largest database to date to assess device survival following primary
implantation of the AMS 800TM AUS. Furthermore, we sought to examine various variables that may influence device survival.
Materials and methods
3
ACCEPTED MANUSCRIPT
Patient selection Retrospective data was collected from the AMS 800TM patient information form (PIF) database. From January 1972 to September 2015, there were 201,637 PIFs for AUS surgery completed
RI PT
worldwide, of which 77,512 were primary implantations in the United States. Following the exclusion of procedures performed in children and females, and those labeled with an unknown surgical technique, 27,096 AUS cases were included in the analysis.
SC
PIF databases do not systematically capture many salient variables such as patient
demographics (besides age), clinical characteristics (previous radiotherapy, previous urethral
M AN U
surgery, etc.), device characteristics (size of cuff, size of PRB, etc.), and certain surgical details (transcorporal cuff placement, location of PRB, etc.). However, consistently available variables were collected and used for analysis. These included patient age, surgical approach (perineal vs. penoscrotal), number of cuffs (single vs. tandem), and surgeon volume (low [<10 AUS/year] vs.
of data.
TE D
high [≥10 AUS/year]). Low vs. high volume implanter definition was determined after analysis
Complaint forms were then used to assess for post-implantation device complications.
EP
Similarly, these forms are completely voluntary and are filled in at the time of revision and/or explantation surgery. As such, they are often lacking information regarding pertinent
AC C
complications, such as mechanical failure and urinary retention, amongst others. Consistently available ones that were included in the analysis were cuff erosion, fluid loss, sub-cuff atrophy, PRB herniation, and infection. Finally, measured outcomes included device explantation and revision, and time to each of these events. Explantation was defined as any surgical intervention requiring removal of all components of the device. Revision was defined as any surgical intervention that removed one or more but not all of the device components. Component revision
4
ACCEPTED MANUSCRIPT
was defined as any surgical intervention requiring either revision or explantation of that particular component. Non-surgical clinical outcomes such as dry rates, number of pads per day,
RI PT
and patient satisfaction, amongst others, were not uniformly available for collection.
Statistical analysis
Continuous variables were summarized with mean and standard deviation; categorical variables
SC
were summarized with frequency counts and percentages. Multivariable Cox proportional
hazards models were used to estimate hazard ratios (HRs) for associations between clinical and
M AN U
surgical characteristics and post-operative complications, explantation and/or revision. Probability of device survival was estimated using the Kaplan-Meier method with time from AUS primary implantation to the subsequent re-surgery intervention (including explantation and/or revision for any reason); patients were censored at the time of dataset cutoff
TE D
date if no event occurred. All statistical tests were 2-sided, with a P value < 0.05 considered statistically significant. Statistical analysis was performed using the SAS 9.4 software (SAS
Results
EP
Institute, Inc., Cary, NC).
AC C
Overall characteristics
Mean age of the population was 68.6 years (SD 8.9). Surgery was performed with high- and lowvolume implanters in 4,666 (17.4%) and 22,165 (82.6%) of cases, respectively. Approach was perineal in 18,373 cases (67.8%) and penoscrotal in 8,723 cases (32.2%), and a tandem cuff was used in 2,224 cases (8.2%) (table 1).
5
ACCEPTED MANUSCRIPT
Post-operative complications Post-implantation complications were as follows: cuff erosion (1,085, 4.0%), loss of fluid (1,020, 3.8%), sub-cuff atrophy (661, 2.4%), device infection (518, 1.8%), and PRB herniation (49,
RI PT
0.2%). Other or unspecified complications were reported in 14.9% of patients (table 2). Younger age was associated with higher rates of sub-cuff atrophy, fluid loss, and PRB herniation, but less cuff erosion. A penoscrotal surgical approach and use of a tandem cuff were both significantly
SC
associated with higher rates of device infection, cuff erosion, and fluid loss. Finally, device
insertion by low-volume implanters was significantly associated with higher cuff erosion rates.
M AN U
Similarly, on multivariate analysis, younger patient age was again a significant independent predictor of high rates of sub-cuff atrophy, fluid loss, and PRB herniation, but less cuff erosion; penoscrotal incision was a predictor of infection and erosion; tandem cuff placement a predictor of fluid loss, erosion, and infection; and a low-volume surgeon a predictor
Explantation and revision
TE D
of erosion (table 3).
EP
Overall, 5,723 patients (21.1%) required explantation and/or revision, including 52.0% (2,978/5,723) of explantation only, 37.7% (2,160/5,723) of revision only, and 10.2% (585/5,723)
AC C
of explantation and revision. Younger age, penoscrotal approach, use of tandem cuff and lowvolume implanter were significantly correlated with higher explantation and component revision rates (table 4).
Similarly, on multivariate analysis, younger age, penoscrotal approach, use of tandem
cuffs, but not surgeon volume, were independent predictors of higher device explantation and component revision (table 5).
6
ACCEPTED MANUSCRIPT
Device survival On Kaplan-Meier survival analysis, 5- and 10-year explantation-free device survival rates were
RI PT
87.1% and 78.3%, respectively (figure 1a). The 5- and 10-year explantation and/or revision-free device survival rates were 78.1% and 68.7%, respectively (figure 1b).
SC
Discussion
Results of this large retrospective study demonstrate that the AMS 800TM AUS continues to be a
M AN U
reliable, durable option for men with stress urinary incontinence. Younger age, penoscrotal approach, and use of a tandem cuff are associated with significantly inferior overall outcomes. In our study, 21.1% of cases required re-intervention by way of device explantation and/or revision. This number is slightly lower than the 26.0% re-intervention rate reported in the
TE D
aforementioned systematic review of the literature by Van der Aa et al 8. Similarly, in the largest, single-center study to date, Linder et al demonstrated an even higher re-operation rate of 31.2% following primary perineal AUS implantation in 1,082 men 10. It is worth noting that, while not
EP
shown to be associated with worse outcomes in their study, a significant number of patients in their cohort had associated co-morbidities and urethra-compromising factors (27% prior pelvic
AC C
radiation, 32% prior vesico-urethral anastomotic stricture). Furthermore, in the aforementioned study, device survival rates, free of any revision, were 74% at 5 years and 57% at 10 years, consistent with other previously published reports 3,7,9,10,15,17-21. In the present study, however, 5and 10-year device survival rates were relatively higher (78.1% and 68.7%, respectively). One likely explanation for this discrepancy is the fact that explantation and/or revision events were collected from the manufacturer’s patient information forms. These forms were optional and are
7
ACCEPTED MANUSCRIPT
filled at the time of surgical revision. It is reasonable to assume that a fair number of these forms may have been incompletely filled out, thus leading to under-reporting of complications and resulting in better outcomes.
RI PT
Two-thirds of cases performed in this series were implanted through a perineal approach. This is consistent with the recently published 2015 International Continence Society (ICS) consensus conference on AUS which states that the perineal approach is preferred to the
SC
penoscrotal one (C grade of recommendation) 22. This stems from the perception that a
penoscrotal approach does not allow for optimal bulbar urethral exposure and proximal
M AN U
placement of the cuff. Our results seem to support this notion, with the penoscrotal approach being associated with higher erosion and infection rates, leading to higher explantation and revision rates. This is in line with a retrospective review of 126 patients that demonstrated significantly better continence and less need for tandem cuff placement with the perineal
TE D
compared to penoscrotal approach 23. On the other hand, Wilson reported excellent outcomes with the use a modified perineal approach for cuff placement, with no reported complications 24. Currently, the ICS guidelines recommend that the penoscrotal approach be reserved for redo
EP
cases, and in patients with conditions that preclude them from being placed in lithotomy position, such as morbid obesity, spine or limb deformities, or neuro-motor conditions, as well as those
AC C
undergoing dual AUS and inflatable penile prosthesis placement through a single penoscrotal incision 22.
In our series, a tandem cuff was placed at the time of initial implantation in 8.2% of
cases. While the merits of using a tandem cuff in cases of revision surgery for cuff erosion or sub-cuff atrophy have been well reported 25,26, its role in patients undergoing primary AUS implantation is uncertain. In a long-term follow-up study of 47 men with post-prostatectomy
8
ACCEPTED MANUSCRIPT
incontinence (25 single and 22 tandem cuffs), O’Connor et al. showed that the use of a tandem cuff imparted no significant differences in dry rates, overall continence, or quality of life as compared to a single cuff 15. Furthermore, these authors demonstrated that men receiving double
RI PT
cuff implants were at higher risk of complications requiring revision surgery. Similarly, in our study, placement of a tandem cuff was a significant predictor of sub-cuff atrophy, cuff erosion and device infection, and ultimately higher explantation and revision rates.
SC
In our study, a younger mean age at the time of surgery appeared to be associated with significantly higher rates of sub-cuff atrophy, PRB herniation, and fluid loss, as well as higher
M AN U
explantation and revision rates. It is unclear why younger age would lead to higher complication rates. One potential explanation is that younger patients in this series had longer documented follow-ups, and consequentially a higher chance for complications and need for revision surgery. There are limited reports that have specifically examined the effect of age on AUS outcomes . While one study did demonstrate that octogenarians were more likely to experience erosion
TE D
27,28
or infection compared with younger patients 28, the 5-year device survival rates were comparable to those reported in younger men (63% to 70%) 27,28. As such, current evidence remains
age.
EP
insufficient to recommend making decisions on AUS placement and outcomes solely based on
AC C
Finally, we defined a high volume surgeon as any surgeon who performs 10 or more
implants per year. Using this definition, we noted that high volume surgeons accounted for 17.4% of all primary AUS implantations. Furthermore, we observed a small association between greater surgeon experience and decreased cuff erosion, device explantation, and component revision. After accounting for other variables, however, surgeon volume was not a significant predictor of complications or need for revision surgery. While one would expect that surgeon
9
ACCEPTED MANUSCRIPT
experience, a likely surrogate of advanced training, would be a predictor of improved outcomes, there are a few theories as to why this benefit was not observed in our study. First and foremost, it is likely that surgeons with higher volume are more specialized urologists working in large
RI PT
academic centers who undertake more complex cases with expectedly higher complication rates. Second, the cut-off value for surgeon volume might be too high. In a case log analysis from certifying and recertifying urologists obtained from the American Board of Urology (ABU)
SC
between 2004 and 2010, among urologists who placed any sphincters, the median number of cases was only two per year with only 4% of urologists implanting ≥10 29. Similarly, in a more
M AN U
recent 6-months case log data of certifying urologists (2003-2013) obtained from the ABU, amongst the 90th percentile of certifying urologists performing AUS, four cases were logged over 6 months, whereas the top 4% (comprising 42 urologists) logged only eight cases during this time period 30. The number of surgeons who fit our definition of a “high-volume implanter”
TE D
(≥10 AUS/year) in this series is only 18 out of 4092 (0.4%). Using stricter cut-offs of 5 AUS/year or 2 AUS/year would have resulted in 54 (1.3%) and 326 (8.0%) high-volume implanters. Accordingly, it could be argued that using a lower threshold for a low vs. high
EP
volume surgeon may have yielded more definite results. Finally, this data puts in perspective the fact that most urologists implanting AUS only perform a few cases per year.
AC C
This study has limitations. First and foremost, it is based on retrospective data collected
from patient information forms that are filled in the operative room at the time of implantation by the surgeon, the circulating nurse or the product specialist. These forms mostly require “ticking a box” but also contain free text sections, which are naturally difficult to capture when reviewing large databases. Furthermore, these optional forms are often incompletely filled. As noted, outcomes were obtained from the manufacturer’s complaint forms, which are similarly filled in
10
ACCEPTED MANUSCRIPT
at the time of revision and/or explantation surgery. These are frequently incomplete, thus affecting outcomes. Finally, as previously mentioned, many important variables relating to patient, clinical, surgical, and post-surgical characteristics, such as continence rates, were either
RI PT
not captured or not deemed reliable after assessment, and were excluded from analysis. As such, the results of this study may not allow for strict clinical recommendations. This study’s—the largest collected AUS database to date for the management of SUI in adult men—main purposes,
SC
however, are to offer a general overview on device survival, and serve as a counseling tool for
Conclusions
M AN U
urologists when discussing surgical options with their patients.
These results provide a general overview on AUS device survival, and may serve urologists when counseling patients. Younger age, penoscrotal approach, and use of tandem cuff may be
References
EP
TE D
associated with inferior outcomes.
AC C
1. Scott FB, Bradley WE, Timm GW. Treatment of urinary incontinence by an implantable prosthetic urinary sphincter. J Urol. 1974; 112(1); 75-80.
2. Begg CB, Riedel ER, Bach PB, et al. Variations in morbidity after radical prostatectomy. N Engl J Med. 2002; 346; 1138-44.
3. Lai HH, Hsu EI, Teh BS, Butler EB, Boone TB. 13 years of experience with artificial urinary sphincter implantation at Baylor College of Medicine. J Urol. 2007; 177; 1021-5.
11
ACCEPTED MANUSCRIPT
4. Petero VG Jr, Diokno AC. Comparison of the long-term outcomes between incontinent men and women treated with artificial urinary sphincter. J Urol. 2006; 175; 605-9.
RI PT
5. Montague DK, Angermeier KW, Paolone DR. Long-term continence and patient satisfaction after artificial sphincter implantation for urinary incontinence after prostatectomy. J Urol. 2001; 166; 547-9.
SC
6. Gousse AE, Madjar S, Lambert MM, Fishman IJ. Artificial urinary sphincter for postradical prostatectomy urinary incontinence: long-term subjective results. J Urol. 2001;
M AN U
166: 1755-8.
7. Venn SN, Greenwell TJ, Mundy AR. The long-term outcome of artificial urinary sphincters. J Urol. 2000; 164: 702-6.
8. Van der Aa F, Drake MJ, Kasyan GR, Petrolekas A, Cornu JN. The artificial urinary sphincter after a quarter of a century: a critical systematic review of its use in male non-
TE D
neurogenic incontinence. Eur Urol. 2013; 63: 681–9. 9. Raj GV, Peterson AC, Toh KL, Webster GD. Outcomes following revisions and secondary implantation of the artificial urinary sphincter. J Urol. 2005; 173(4): 1242–5.
EP
10. Linder BJ, Rivera ME, Ziegelmann MJ, Elliott DS. Long-term Outcomes Following
AC C
Artificial Urinary Sphincter Placement: An Analysis of 1082 Cases at Mayo Clinic. Urology. 2015; 86(3): 602-7.
11. Henry GD, Graham SM, Cornell RJ, et al. A multicenter study on the perineal versus penoscrotal approach for implantation of an artificial urinary sphincter: cuff size and control of male stress urinary incontinence. J Urol.2009; 182(5): 2404–9.
12
ACCEPTED MANUSCRIPT
12. Sandhu JS, Maschino AC, Vickers AJ. The surgical learning curve for artificial urinary sphincter procedures compared to typical surgeon experience. Eur Urol. 2011; 60: 1285– 90.
RI PT
13. Lai HH, Boone TB. Complex artificial urinary sphincter revision and reimplantation cases--how do they fare compared to virgin cases? J Urol 2012; 187: 951-5.
14. Gomha MA, Boone TB. Artificial urinary sphincter for post-prostatectomy incontinence
SC
in men who had prior radiotherapy: A risk and outcome analysis. J Urol. 2002; 167: 5916.
M AN U
15. O'Connor RC, Lyon MB, Guralnick ML, Bales GT. Long-term follow-up of single versus double cuff artificial urinary sphincter insertion for the treatment of severe postprostatectomy stress urinary incontinence. Urology. 2008; 71: 90–3. 16. Yafi FA, Powers MK, Zurawin J, Hellstrom WJ. Contemporary review of artificial
TE D
urinary sphincters for male stress urinary incontinence. Sex Med Rev. 2016; 4: 157-6. 17. Leon P, Chartier-Kastler E, Roupret M,.Ambrogi V, Mozer P, Phe V. Long-term functional outcomes after artificial urinary sphincter implantation in men with stress
EP
urinary incontinence. BJU Int. 2014; 115: 951-7. 18. Linder BJ, de Cogain M, Elliott DS. Long-term device outcomes of artificial urinary
AC C
sphincter reimplantation following prior explantation for erosion or infection. J Urol. 2014; 191: 734-8.
19. Guralnick ML, Miller E, Toh KL, Webster GD. Transcorporal artificial urinary sphincter cuff placement in cases requiring revision for erosion and urethral atrophy. J Urol. 2002; 167: 2075-8.
13
ACCEPTED MANUSCRIPT
20. Magera JS, Elliott DS. Tandem transcorporal artificial urinary sphincter cuff salvage technique: surgical description and results. J Urol. 2007; 177: 1015-9. 21. Kim SP, Sarmast Z, Daignault S, Faerber GJ, McGuire EJ, Latini JM.. Long-term
RI PT
durability and functional outcomes among patients with artificial urinary sphincters: a 10year retrospective review from the University of Michigan. J Urol. 2008; 179: 1912-6. 22. Biardeau X, Aharony S, AUS Consensus Group, Campeau L, Corcos J. Artificial Urinary
SC
Sphincter: Report of the 2015 Consensus Conference. Neurourol Urodyn. 2016; 35 Suppl 2: S8-S24.
M AN U
23. Henry GD, Graham SM, Cleves MA, Simmons CJ, Flynn B. Perineal approach for artificial urinary sphincter implantation appears to control male stress incontinence better than the transscrotal approach. J Urol. 2008; 179(4): 1475-9; discussion 1479. 24. Wilson SK, Delk JR, Henry GD, Siegel AL. New surgical technique for sphincter urinary
TE D
control system using upper transverse scrotal incision. J Urol 2003; 169: 261. 25. Brito CG, Mulcahy JJ, Mitchell ME, Adams MC. Use of a double cuff AMS800 urinary sphincter for severe stress incontinence. J Urol. 1993; 149(2): 283-5.
EP
26. DiMarco DS, Elliott DS. Tandem cuff artificial urinary sphincter as a salvage procedure following failed primary sphincter placement for the treatment of post-prostatectomy
AC C
incontinence. J Urol. 2003; 170(4 Pt 1): 1252-4.
27. O'Connor RC, Nanigian DK, Patel BN, Guralnick ML, Ellision LM, Stone AR. Artificial urinary sphincter placement in elderly men. Urology. 2007; 69(1): 126-8.
28. Ziegelmann MJ, Linder BJ, Rivera ME, Viers BR, Rangel LJ, Elliott DS. Int J Urol. 2016; 23(5): 419-23. Outcomes of artificial urinary sphincter placement in octogenarians.
14
ACCEPTED MANUSCRIPT
29. Poon SA, Silberstein JL, Savage C, Maschino AC, Lowrance WT, Sandhu JS. Surgical practice patterns for male urinary incontinence: analysis of case logs from certifying American urologists. J Urol. 2012; 188(1): 205-10.
RI PT
30. Liu JS, Hofer MD, Milose J, et al. Male Sling and Artificial Urethral Sphincter for Male Stress Urinary Incontinence Among Certifying American Urologists. Urology. 2016; 87:
M AN U
SC
95-9.
Figure legends:
Figure 1a. Artificial urinary sphincter device explanation free rate estimated by Kaplan Meier
TE D
method.
Figure 1b. Artificial urinary sphincter device explanation and/or revision free rate estimated by
AC C
EP
Kaplan Meier method.
15
ACCEPTED MANUSCRIPT
Table 1. Clinical and surgical characteristics of 27,096 adult men undergoing primary implantation of AMS 800 artificial urinary sphincter Total N=27,096 68.6 ± 8.9
RI PT
Mean age at surgery (years ± SD) Surgical Approach
N=27,096
Peno-scrotal Perineal
8,723 (32.2%) 18,373 (67.8%) N=27,033
SC
Tandem Cuff
24,809 (91.8%) 2,224 (8.2%)
Physician volume
AC C
EP
TE D
Low (< 10 surgeries/year) High (≥ 10 surgeries/year)
M AN U
No Yes
N=26,831
22,165 (82.6%) 4,666 (17.4%)
ACCEPTED MANUSCRIPT
68.6 ± 8.9
66.8 ± 8.8
2.8 ± 2.4 <0.001
68.7 ± 8.9
65.8 ± 8.7
0.231
8,496 (97.4%)
227 (2.6%)
8,361 (95.9%)
362 (4.1%)
Perineal
17,939 (97.6%)
434 (2.4%)
17,715 (96.4%)
658 (3.6%)
0.628
<0.001
68.6 ± 8.9
24,199 (97.5%)
610 (2.5%)
23,937 (96.5%)
872 (3.5%)
Yes
2,173 (97.7%)
51 (2.3%)
2,077 (93.4%)
147 (6.6%)
0.328
65.0 ± 9.2
21,323 (96.2%)
Yes
4,542 (97.3%)
124 (2.7%)
4,497 (96.4%)
No, N=26,011 (96.0%)
1.4 ± 1.9
0.006
68.6 ± 8.9
67.9 ± 10.1
Yes, N=1,085 (4.0%)
0.089
68.5 ± 8.9
69.7 ± 8.4
<0.001
227 (2.6%)
8,265 (94.7%)
458 (5.3%)
18,342 (99.8%)
31 (0.2%)
18,082 (98.4%)
291 (1.6%)
17,746 (96.6%)
627 (3.4%)
2,222 (99.9%)
<0.001
<0.001
47 (0.2%)
24,373 (98.2%)
436 (1.8%)
23,892 (96.3%)
917 (3.7%)
2 (0.1%)
2,145 (96.4%)
79 (3.6%)
2,060 (92.6%)
164 (7.4%)
0.764
0.098
<0.001
<0.001
8,496 (97.4%)
0.434
P Value
2.3 ± 2.3
18 (0.2%)
24,762 (99.8%)
EP
535 (2.4%)
<0.001
842 (3.8%)
22,121 (99.8%)
44 (0.2%)
21,737 (98.1%)
428 (1.9%)
21,212 (95.7%)
953 (4.3%)
169 (3.6%)
4,662 (99.9%)
4 (0.1%)
4,579 (98.1%)
87 (1.9%)
4,537 (97.2%)
129 (2.8%)
AC C
21,630 (97.6%)
Cuff Erosion P Value
8,705 (99.8%)
0.564
No
Yes, N=518 (1.9%)
0.496
TE D
No
High Volume Physician
<0.001
0.022
Penoscrotal
Tandem Cuff
1.7 ± 2.2
SC
4.2 ± 2.7
M AN U
Mean years to event ± SD Mean age at Surgery ± SD Surgical Approach
Infection No, N=26,578 (98.1%)
RI PT
Table 2. Complications following primary implantation of AMS 800 artificial urinary sphincter in 27,096 adult men Sub-cuff atrophy Fluid loss Herniation No, Yes, P No, Yes, P No, Yes, P N=26,435 N=661 Value N=26,076 N=1,020 Value N=27,047 N=49 Value (97.6%) (2.4%) (96.2%) (3.8%) (99.8%) (0.2%)
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 3. Multivariate Cox Proportional Hazard regression models for complications following implantation of AMS 800 artificial urinary sphincter in 27,096 adult men Cuff atrophy Fluid loss Cuff Erosion Infection PRB herniation Variable HR P HR P HR P HR P HR P Age 0.980 <0.001 0.970 <0.001 1.013 <0.001 0.991 0.064 0.963 0.004 Perineal 1.116 0.189 1.015 0.823 0.785 <0.001 0.669 <0.001 1.000 0.999 approach Tandem 0.710 0.020 1.581 <0.001 1.626 <0.001 1.769 <0.001 0.421 0.232 cuff High 1.151 0.164 1.000 0.995 0.730 0.001 1.095 0.452 0.422 0.101 volume surgeon
ACCEPTED MANUSCRIPT
Mean years to event ± SD Mean age at Surgery ± SD Surgical Approach
68.8 ± 8.9
Penoscrotal l
7,242 (83.0%)
1,481 (17.0%)
7,718 (88.5%)
1,005 (11.5%)
Perineal
16,291 (88.7%)
2,082 (11.3%)
16,633 (90.5%)
1,740 (9.5%)
<0.00 1
68.7 ± 8.9
2.6 ± 2.4
67.5 ± 8.9
<0.00 1
<0.00 1
Tandem Cuff
<0.00 1
<0.00 1
0.356
21,736 (87.6%)
3,073 (12.4%)
22,280 (89.8%)
2,529 (10.2%)
Yes
1,738 (78.1%)
486 (21.9%)
2,011 (90.4%)
213 (9.6%)
19,176 (86.5%)
2,989 (13.5%)
Yes
4,119 (88.3%)
547 (11.7%)
19,920 (89.9%)
4,177 (89.5%)
<0.00 1
68.8 ± 8.9
67.2 ± 8.9
Yes, N=3,94 7 (14.6%)
<0.00 1
68.8 ± 8.9
67.3 ± 8.9
<0.00 1
7,072 (81.1%)
1,651 (18.9%)
7,061 (80.9%)
1,662 (19.1%)
15,223 (82.9%)
3,150 (17.1%)
15,941 (86.8%)
2432 (13.2%)
16,088 (87.6%)
2,285 (12.4%)
<0.00 1
<0.00 1
20,189 (81.4%)
4,620 (18.6%)
21,259 (85.7%)
3,550 (14.3%)
21,409 (86.3%)
3,400 (13.7%)
1,645 (74.0%)
579 (26.0%)
1,695 (76.2%)
529 (23.8%)
1,681 (75.6%)
543 (24.4%)
<0.00 1
<0.00 1 <0.00 1
2,056 (23.6%)
<0.00 1
P Value
2.9 ± 2.6
<0.00 1
<0.00 1
2,245 (10.1%)
17,866 (80.6%)
4,299 (19.4%)
18,772 (84.7%)
3,393 (15.3%)
18,830 (85.0%)
3,335 (15.0%)
489 (10.5%)
3793 (81.3%)
873 (18.7%)
4,007 (85.9%)
659 (14.1%)
4,083 (87.5%)
583 (12.5%)
AC C
No
No, N=23,14 9 (85.4%)
6,667 (76.4%)
0.471
EP
<0.00 1
67.3 ± 8.9
<0.00 1
TE D
No
High Volume Physician
68.9 ± 8.9
Pump revision/explantation
2.9 ± 2.6
M AN U
67.2 ± 8.9
1.8 ± 1.9
SC
3.0 ± 2.6
RI PT
Table 4. Explantation and revision rates following primary implantation of AMS 800 artificial urinary sphincter in 27,096 adult men Any explantation Any revision Cuff revision/explantation PRB revision/explantation No, Yes, P No, Yes, P No, Yes, P No, Yes, P N=23,53 N=3,56 Value N=24,35 N=2,74 Value N=23,01 N=4,08 Value N=21,89 N=5,20 Value 3 3 1 5 3 3 0 6 (86.9%) (13.1%) (89.9%) (10.1%) (84.9%) (15.1%) (80.8%) (19.2%)
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
Table 5. Multivariate Cox Proportional Hazard regression models for device explantation and revision following AMS 800 artificial urinary sphincter implantation in 27,096 adult men Any Any revision Cuff Pump PRB explantation revision/explantation revision/explantation revision/explantation Variable HR P HR P HR P HR P HR P Age 0.984 <0.001 0.987 <0.001 0.985 <0.001 0.984 <0.001 0.984 <0.001 Perineal 0.806 <0.001 0.887 0.003 0.835 <0.001 0.780 <0.001 0.828 <0.001 approach Tandem 1.414 <0.001 0.826 0.008 1.141 0.003 1.449 <0.001 1.358 <0.001 cuff High0.968 0.493 1.077 0.146 1.053 0.170 0.923 0.079 1.016 0.718 volume
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
AC C
EP
TE D
M AN U
SC
RI PT
ACCEPTED MANUSCRIPT
ACCEPTED MANUSCRIPT
artificial urinary sphincter
PIF
patient information form
SUI
stress urinary incontinence
PRB
pressure-regulating balloon
HRs
hazard ratios
AC C
EP
TE D
M AN U
SC
AUS
RI PT
Keywords