The Male Perineal Sling: Assessment and Prediction of Outcome Melissa C. Fischer, Chad Huckabay and Victor W. Nitti*,† From the Department of Urology, New York University School of Medicine, New York, New York

Purpose: The male perineal sling has become an option for treating male stress incontinence. We evaluated its overall efficacy and determined preoperative parameters that could predict success. Materials and Methods: A total of 62 men with stress incontinence were prospectively evaluated and underwent a male perineal sling. Preoperatively 24-hour pad weight, urodynamics and a number of validated incontinence questionnaires were completed. At a minimum of 3 months of followup 24-hour pad weight and questionnaires were repeated. Success was assessed using the 1 question Patient Global Impression of Improvement. The Patient Global Impression of Improvement was compared to a number of other subjective and objective measures of outcome. Finally, preoperative parameters were evaluated to determine predictors of outcome. Results: As determined by the Patient Global Impression of Improvement, the success rate was 58%. The only preoperative factor predictive of success was 24-hour pad weight. An individual had a 71% chance of successful surgery if preoperative pad weight was less than 423 gm. There was a statistically significant difference between successes and failures in terms of postoperative pad weight and certain questionnaires, including the UCLA/RAND Prostate Cancer Index urinary function score, International Consultation on Incontinence short form, Incontinence Impact Questionnaire, Urogenital Distress Index and International Prostate Symptom Score. The overall complication rate was 21% and the reoperation rate was 14.5%. Conclusions: The male perineal sling can be an effective surgical treatment for stress incontinence in the appropriate patient. The procedure is most successful in patients with lesser objective degrees of incontinence. The Patient Global Impression of Improvement is an effective tool for assessing outcome for this population. Key Words: prostate, urinary incontinence, prostatectomy, prostheses and implants, outcome assessment (health care)

ncontinence after treatment for prostate cancer and benign prostatic hyperplasia can be a significant problem. Most incontinence after radical prostatectomy for prostate cancer is caused by sphincteric insufficiency (SUI) due to direct or indirect compromise of the distal sphincteric mechanism.1,2 There are 2 surgical options to treat this condition, that is the gold standard AUS with patient satisfaction rates of up to 90%3,4 and the more recently described male sling procedures. The male perineal sling with bone anchors is the most widely adopted male sling procedure to date. Results of the male sling are more variable than those of the AUS with reported shortterm success rates of 39.5% to 96.2%.5–10 Some groups suggested that patients with higher pad requirements6,9 or those who received radiation therapy7 are at risk for failure. The definition of success for the male sling or male incontinence procedures in general is not standardized. Success is defined in a number of ways, including complete cure, social continence while wearing 1 small or thin pad or less daily,9,10 quality of life or incontinence impact questionnaires, such as the UCLA/RAND,8,10 patient satisfaction

I

Submitted for publication August 17, 2006. * Correspondence: Department of Urology, New York University School of Medicine, 150 East 32nd St., New York, New York 10016 (telephone: 646-825-6343; FAX: 646-825-6397; e-mail: victor.nitti@ nyumc.org). † Financial interest and/or other relationship with American Medical Systems.

0022-5347/07/1774-1414/0 THE JOURNAL OF UROLOGY® Copyright © 2007 by AMERICAN UROLOGICAL ASSOCIATION

rates4,6 and subjective measures, such as 50% or greater improvement.6 While they are legitimate ways of classifying success, they are not consistent and may not consider the patient perception of the effect of surgery on his condition. We prospectively detailed experience at a single institution with the male perineal sling and critically evaluated its efficacy and morbidity. In addition, we determined if any preoperative parameters were predictive of outcome. We used a single question assessment tool, the PGI-I, to define success. The PGI-I was validated in the female stress incontinence population.11 It was recently shown to correlate with a decrease in pad weight and the change in the validated ICIQ-SF in men undergoing a male perineal sling.12 MATERIALS AND METHODS A total of 62 men were prospectively evaluated for a primary diagnosis of SUI between April 2002 and December 2005. They subsequently underwent placement of a male perineal sling. Preoperatively a history was obtained and physical examination were performed, and patients completed the I-PSS, UDI-6,13 IIQ-713 and ICIQ-SF.14 Multichannel urodynamics were performed in all patients according to a previously described protocol15 and the standards of the International Continence Society.16 Urodynamic stress incontinence was found in all patients and ALPP, maximum cystometric capacity and the presence or absence of DO were recorded. A 24-hour pad test was done to objectively assess the degree of incontinence. Patients were

1414

Vol. 177, 1414-1418, April 2007 Printed in U.S.A. DOI:10.1016/j.juro.2006.11.061

MALE PERINEAL SLING counseled regarding treatment options and informed consent was obtained for a male perineal sling. In year 1 there were no selection criteria for the male sling and patients elected the procedure based on their desire to avoid or inability to operate a mechanical device. However, after interim analysis showed that men with higher degrees of incontinence had poorer outcomes with the male perineal sling,17 patients were informed of this and counseled appropriately, although they were not excluded from the procedure. All patients underwent a male perineal sling using the InVance™ male sling system with bone anchors according to the previously described technique.7 A 4 ⫻ 7 cm silicone coated polyester mesh sling was used in all patients. Regional anesthesia was used to allow the patient to cough during the procedure. Initially tension was adjusted by a cough test with 250 ml in the bladder.7 After it was established that there was no leakage with coughing, the sling was temporarily secured. RLPP was then determined by perfusion sphinterometry.5 If RLPP was less than 60 cm H2O, the sling was tightened until RLPP equaled or exceeded 60 cm H2O. After appropriate tension was established the sling was completely secured into position, and the cough test and RLPP were repeated. All patients had a negative cough test and an RLPP of 60 cm H2O or greater. All procedures were performed on an outpatient basis. A 14Fr Foley catheter was left in place for 1 to 3 days. Postoperatively a 24-hour pad test was repeated and questionnaires were re-administered, including the UCLA/ RAND18 and PGI-I questionnaires. Questionnaires were administered at the last followup. Additionally, noninvasive uroflowmetry and post-void residual urine measurement were performed and any postoperative complications were recorded. Success was defined by the PGI-I as very much or much improved. Failure was defined as a little better, no change, a little worse or much worse. Patients were considered padfree if they reported never wearing pads. Several parameters were evaluated by binary logistic regression analysis to determine their relationship to surgical outcome, including patient age, time from prostate surgery/radiation, followup, preoperative pad weight, ALPP, DO, and I-PSS and UDI-6 scores. In addition, successes and failures defined by the PGI-I were compared to objective (pad weight) and subjective (postoperative questionnaires) outcomes using ANOVA. Data were analyzed using SPSS®.

RESULTS Mean patient age was 67.2 years (range 45 to 84). The etiology of SUI was radical retropubic prostatectomy alone in 47 patients (76%), combined surgery and radiation therapy in 11 (18%), pelvic radiation therapy alone in 3 (5%) and transurethral prostatectomy in 1 (2%). Table 1 lists the results of preoperative pad testing, questionnaires and urodynamics. In addition, 21 patients (34%) demonstrated DO on urodynamics. A total of 14 patients (23%) had a history of bladder neck contracture or anastomotic stricture. However, urodynamics and intraoperative cystos-

1415

TABLE 1. Results of preoperative urodynamics, pad weight and questionnaires Mean ⫾ SE (range) Pad wt (gm) I-PSS IIQ-7 UDI-6 ALPP (cm H2O) Max cystometric capacity (ml)

352 ⫾ 43 (17–1479) 10.4 ⫾ 1.1 (0–31) 8.48 ⫾ 0.9 (0–28) 7.5 ⫾ 0.5 (2–15) 81 ⫾ 5 (21–204) 327 ⫾ 13 (140–595)

copy confirmed no current bladder outlet obstruction before sling placement. Mean time from treatment causing incontinence to sling surgery was 56.5 months (range 9 to 204). Mean followup was 15 months (range 3 to 37, median 12). Table 2 shows successes, which were defined using the PGI-I. A total of 36 men (58%) had a successful result (very much or much better), 21 (34%) were completely pad-free, 6 (10%) were a little better but were not considered successes and 9 (14.5%) considered themselves worse postoperatively. There was no association of a history of radiation therapy or bladder neck contracture with outcome (Fisher’s exact test p ⫽ 1.000 and 0.219, respectively). Binary logistic regression analysis was performed to determine the predictive value of age, time from treatment causing incontinence, followup, preoperative pad weight, ALPP, maximum cystometric capacity, DO, I-PSS or UDI-6 with success. Only preoperative pad weight predicted success. Risk analysis showed that an individual had a 71% chance of successful surgery if preoperative pad weight was less than 423 gm. Furthermore, the odds of successful surgery were 6-fold greater for patients with a pad weight of less than 423 gm vs those with a pad weight of more than 423 gm. The figure shows successes, pad-free patients and those who were worse postoperatively according to preoperative pad weight. Interestingly patients with the lowest preoperative pad weights (50 or less, or 50 to 100 gm) did not have the most favorable results. Table 3 shows postoperative pad weight and questionnaire results. For each postoperative parameter the cohort was divided into successes and failures based on the PGI-I. The t test was then performed to determine if there was a significant difference between the success and failure groups. There was a statistically significant difference between successes and failures in terms of pad weight, the postoperative UCLA/RAND, and postoperative scores and changes from preoperative values on the ICIQ-SF, IIQ-7, UDI-6 and I-PSS. Concern for obstructive voiding after sling placement was addressed by determining the change in Qmax. Preoperative and postoperative data were available on 45 patients. The average change in Qmax was ⫺4.3 U. Statistical analysis of the distribution revealed that with 99% certainty the mean change in Qmax was less than zero. Therefore, for the entire population the male perineal sling did not result in a decreased flow rate. However, 3 patients underwent sling adjustment for prolonged obstructive symptoms, including 2 with increased post-void residual urine on intermittent catheterization. In addition, there was a statistically higher increase in I-PSS in failures compared to successes. Overall 13 patients (21%) reported complications (table 4). De novo urge and/or urge incontinence was

1416

MALE PERINEAL SLING TABLE 2. Classification of success using PGI-I PGI-I

Circle one that best describes how your urinary tract condition is now compared with how it was before you had operation: Very much better Much better Little better No change Little worse Much worse Very much worse Total

No. Success

No. Failure

TABLE 3. Correlation of success based on PGI-I with other postoperative pad weight and questionnaires PGI-I Outcome

23 13 6 11 6 3 0 36

26

present in 1 patient. In this man postoperative urodynamics confirmed DO without outlet obstruction. Urinary retention requiring intermittent catheterization greater than 1 month after implantation occurred in 2 patients. Each man was able to void effectively after sling adjustment. One patient with poor contractility had significant obstructive voiding symptoms without retention and he underwent adjustment. Four men underwent elective revision of the sling for persistent stress incontinence, which was successful in 2. In addition, there were a total of 9 reoperations (14.5%) in 8 patients (13%) that were related specifically to the sling. Sling infections were treated in 4 patients, including 1 in whom infection was controlled with oral antibiotics and 3 in whom the sling was removed. The patient in whom the sling was not removed showed perineal discharge and discomfort 5 months after surgery. He was treated initially with oral antibiotics with resolution but continued to have periodic drainage from a sinus, which responded to oral antibiotics. He refused removal. One infection occurred with urethral erosion in a patient who previously underwent successful sling adjustment with increased tension. No osseous complications were identified. Perineal paresthesia, ie numbness or an itching sensation, and/or mild pain persisting greater than 3 months postoperatively were reported in 5

Pad wt (gm): Success Failure UCLA/RAND: Success Failure Postop ICIQ-SF: Success Failure ICIQ-SF change: Success Failure Postop IIQ-7: Success Failure IIQ-7 change: Success Failure Postop UDI-6: Success Failure UDI-6 change: Success Failure I-PSS: Success Failure I-PSS change: Success Failure

Mean ⫾ SE

p Value

22.5 ⫾ 39.2 350.3 ⫾ 48.0

⬍0.0001

322.1 ⫾ 17.0 128.5 ⫾ 21.7

⬍0.0001

5.7 ⫾ 0.7 12.8 ⫾ 1.2

⬍0.0001

⫺8.6 ⫾ 0.9 ⫺2.4 ⫾ 1.2

0.0002

2.0 ⫾ 0.8 7.6 ⫾ 1.7

⬍0.0001

⫺7.7 ⫾ 1.5 ⫺0.9 ⫾ 2.0

0.0073

4.0 ⫾ 0.6 7.8 ⫾ 1.0

0.0010

⫺3.3 ⫾ 0.8 ⫺0.3 ⫾ 5.0

0.0293

8.1 ⫾ 1.0 13.5 ⫾ 1.7

0.0058

1.7 ⫾ 1.1 6.8 ⫾ 1.7

0.0102

Change from preoperatively to postoperatively.

patients. Five patients, including 4 with failure and 1 with erosion, ultimately had an AUS placed. DISCUSSION In this series we addressed 2 critical issues in male incontinence surgery, that is how success is defined, and the outcome of and patient selection for the male perineal sling. Reported success rates for the male perineal sling using different measures and definitions of success vary widely from 39.5% to 96.2%.7–9 As defined by PGI-I, our success rate was 58%, which includes an unselected group of patients with high grade incontinence. We chose to define success as a global patient perception of the lower urinary tract condition after surgery compared to that before surgery. This makes sense for several reasons. Most men are not pad-free after surgery for stress incontinence even with an AUS and yet satisfaction rates are high. Therefore, using pad-free rates as a measure of success is too critical an assessment. Determining how much of a decrease

TABLE 4. Complications and reoperations Complication De novo urge/urinary incontinence Revision for persistent SUI Urinary retention Obstructive symptoms Paresthesia/pain greater than 3 mos Erosion/infection Infection

Results by preoperative pad weight

Totals (%)

No. Pts

No. Reoperation

1 — 2 1 5 1 3

— 3 2 1 — 1 2

13 (21)

9 (14.5)

MALE PERINEAL SLING in objective urine loss is significant to a particular patient is impossible when applied to a group as a whole. For example, some men with a 50% decrease in incontinence are pleased but many are not. A global assessment of outcome also accounts for undesirable or desirable affects on symptoms other than incontinence, such as frequency, urgency or ease of voiding. The fact that our failures had a statistically higher increase in I-PSS than our successes would support this notion. We previously noted that the PGI-I strongly correlates with the ICIQ-SF and the percent decrease in 24-hour pad weight in men undergoing a male perineal sling.12 The same result was found in the current study. Finally, we believe that the simplicity of a single, straightforward question of outcome may result in increased patient compliance and more widespread application among urologists. We were fairly critical of defining success and chose not to include cases that were a little better as successes. We believed that patients choosing surgery expect more than that. If we include that group in our successes, the rate increases to 68%. It is clear from our experience that the male perineal sling is not ideal for all men with SUI. Clearly those with high grade incontinence, as measured objectively by pad weight, do not fare as well. Preoperative pad weight was the only preoperative factor that correlated with success. Patients with a pad weight of less than 423 gm had a 71% chance of success. While this risk analysis determines the pad weight that maximizes the differences in success probability (above and below that value), it not meant to be an absolute cutoff pad weight for determining candidates for a sling. Our sample size was not large enough to determine such a cutoff. In more practical terms trends can be seen (see figure). A more surprising finding was that men with the most mild degrees of incontinence (50 or less and 50 to 100 gm) did not do as well. In fact, the lowest pad weight group had one of the highest percents of patients who considered themselves worse after surgery. We suspect that this was because of high expectations, ie total dryness, in this group. These men are probably less likely to tolerate any negative effects of surgery if they do not achieve dryness. This is clearly a difficult group to satisfy with any intervention. These findings highlight the need for an objective measurement of incontinence to properly counsel patients before surgery. Urodynamic findings other than the demonstration of urodynamic stress incontinence were not predictive of global improvement. Previous study showed that ALPP does not correlate with pad weight in men with SUI19 and, therefore, it is not surprising that ALPP did not correlate with success. Furthermore, DO on preoperative urodynamics was not associated with a worse outcome. Therefore, it is reasonable to offer patients with stress predominant mixed incontinence or DO a male sling for the stress component of symptoms. DO and low ALPP are not contraindications to a male sling. At this time the usefulness of preoperative urodynamics would seem to be to 1) diagnose urodynamic SUI (all patients had this), 2) rule out significantly impaired compliance (no patient had this) and 3) identify patients with impaired contractility and/or Valsalva voiding who may be at risk for postoperative voiding dysfunction or retention (1 patient had this and required sling revision).

1417

All slings were placed by the same technique, and a cough test and minimum retrograde urethral pressure were used to adjust tension. We cannot make any specific claims regarding the usefulness of either of these tension adjustment techniques because all cases were done in the same way. There are not enough data to draw specific conclusions about a specific retrograde urethral pressure and in most cases it was between 60 and 70 cm H2O. The overall complication rate was 21% and the reoperation rate was 14.5%. Of note, 5 patients experienced bothersome perineal paresthesia and/or pain greater than 3 months in duration. Others reported numbness and/or pain that did not persist for greater than 3 months. Longterm studies are needed to elucidate the significance of this complication and whether technical modifications are needed. To our knowledge prolonged obstruction from a male sling has not been previously reported. Ullrich and Commiter observed no urodynamic obstruction and no change in mean Qmax or mean detrusor pressure at Qmax in 22 men who underwent postoperative urodynamics.20 We also found no significant change in mean Qmax in our patients, although 2 and 1 required reoperation for retention and obstructive symptoms, respectively. Therefore, we believe that it is possible to place too much tension on a sling and some objective and/or subjective measurement of sling tension might help decrease this risk. We realize that the mean followup of only 15 months in this series is short but, nevertheless, important lessons can be learned. Even if success rates decrease with time, it is clear that men with high grade incontinence do not do as well with a male sling. Furthermore, men with low grade incontinence also do not do as well, at least when global assessment of the condition is done. While infection, erosion and obstruction are rare, they can develop. These finding have a definite impact on how we counsel patients. In addition, we observed that, although most failures are apparent within the first 3 to 6 months, some patients have failure with time. While this is not the norm, it can happen. Longer followup is needed to make significant claims regarding the longevity of the procedure.

CONCLUSIONS The male perineal sling is an attractive surgical alternative to the AUS because the results are immediate and it is nonmechanical. However, patients must have a realistic expectation of outcomes and complications. In the properly selected patient it is effective surgical treatment for SUI with modest morbidity. While our results are short to intermediate in terms of followup, we identified some important selection and expectation criteria. The most reliable predictor of success is 24-hour pad weight. Individuals must be properly counseled with regard to success and potential complications. Those with high grade and low grade objective incontinence must have appropriate expectations if they choose a male sling. PGI-I is a simple, effective tool for measuring outcomes. It correlates well with postoperative pad weight and other widely used, subjective self-assessment questionnaires.

1418

MALE PERINEAL SLING 9.

Abbreviations and Acronyms ALPP AUS DO ICIQ-SF

⫽ ⫽ ⫽ ⫽

IIQ-7 ⫽ I-PSS ⫽ MCC ⫽ PGI-I ⫽ Qmax RLPP SUI UCLA/RAND

⫽ ⫽ ⫽ ⫽

UDI-6 ⫽

abdominal leak point pressure artificial urinary sphincter detrusor overactivity International Consultation on Incontinence short form Incontinence Impact Questionnaire short form International Prostate Symptom Score maximum cystometric capacity Patient Global Impression of Improvement maximum urinary flow rate retrograde leak point pressure stress urinary incontinence UCLA/RAND Prostate Cancer Index urinary function score Urogenital Distress Index short form

10. 11.

12.

13.

14.

REFERENCES 1.

2.

3.

4.

5. 6. 7.

8.

Desautel MG, Kapoor R and Badlani GH: Sphincteric incontinence: the primary cause of post-prostatectomy incontinence in patients with prostate cancer. Neurourol Urodyn 1997; 16: 153. Ficazzola MA and Nitti VW: The etiology of post-radical prostatectomy incontinence and correlation of symptoms with urodynamic findings. J Urol 1998; 160: 1317. Litwiller SE, Kim KB, Fone PD, White RW and Stone AR: Post-prostatectomy incontinence and the artificial urinary sphincter: a long-term study of patient satisfaction and criteria for success. J Urol 1996; 156: 1975. Fleshner N and Herschorn S: The artificial urinary sphincter for post-radical prostatectomy incontinence: impact on urinary symptoms and quality of life. J Urol 1996; 155: 1260. Comiter C: The male sling for stress urinary incontinence: a prospective study. J Urol 2002; 167: 597. Onur R, Rajpurkar A and Singla A: New perineal bone-anchored male sling. Lessons learned. Urology 2004; 64: 58. Rujpurkrar AD, Onur R and Singla A: Patient satisfaction and clinical efficacy of the new perineal bone-anchored male sling. Eur Urol 2005; 47: 237. Samli M and Singla A: Absorbable versus nonabsorbable graft: outcome of bone anchored male sling for post-radical prostatectomy incontinence. J Urol 2005; 173: 499.

15.

16.

17.

18.

19.

20.

Castle E, Andrews P, Itano N, Novicki DE, Swanson SK and Ferrigni RG: The male sling for post-prostatectomy incontinence: mean followup of 18 months. J Urol 2005; 173: 1657. Comiter C: The male perineal sling: intermediate term results. Neurourol Urodyn 2005; 24: 645. Yalcin I and Bump R: Validation of two global impression questionnaires for incontinence. Am J Obstet Gynecol 2003; 189: 98. Twiss CO, Fischer MC and Nitti VW: Comparison between reduction in 24-hour pad weight, International Consultation on Incontinence-Short Form (ICIQ-SF) score, International Prostate Symptom Score (IPSS), and Post-Operative Patient Global Impression of Improvement (PGI-I) score in patient evaluation after male perineal sling. Neurourol Urodyn 2007; 26: 8. Uebersax J, Wyman J, Shumaker S, McClish DK and Fantl AJ: Short forms to assess life quality and symptom distress for urinary incontinence in women: the incontinence impact questionnaire and the urogenital distress inventory. Neurourol Urodyn 1995; 14: 131. Avery K, Donovan J and Abrams P: Validation of a new questionnaire for incontinence: the International Consultation on Incontinence Questionnaire (ICI-Q). Neurourol Urodyn 2001; 20: 510. Huckabay C, Twiss C, Berger A and Nitti VW: A urodynamics protocol to optimally assess men with post-prostatectomy incontinence. Neurourol Urodyn 2005; 24: 622. Abrams P, Cardozo L, Fall M, Griffiths D, Rosier P, Ulmsten U et al: The standardisation of terminology of lower urinary tract function: report from the Standardisation Subcommittee of the International Continence Society. Urology 2003; 61: 37. Huckabay CP, Fleischmann N, Twiss C and Nitti VW: Predicting outcomes of the male perineal sling. J Urol, suppl., 2005; 173: 150, abstract 550. Litwin M, Hays R, Fink A, Ganz PA, Leake B, Leach GE et al: The UCLA Prostate Cancer Index: development, reliability, and validity of a health-related quality of life measure. Med Care 1998; 36: 1002. Twiss C, Fleischmann N and Nitti V: Correlation of abdominal leak point pressure with objective incontinence severity in men with post-radical prostatectomy incontinence. Neurourol Urodyn 2005; 24: 207. Ullrich NFE and Commiter CV: The male sling for stress urinary incontinence: urodynamic and subjective assessment. J Urol 2004; 172: 204.