COLLAGEN IMPLANT FOR TREATING STRESS URINARY INCONTINENCE IN WOMEN WITH URETHRAL HYPERMOBILITY

0022-5347/01/1664-1354/0 THE JOURNAL OF UROLOGY® Copyright © 2001 by AMERICAN UROLOGICAL ASSOCIATION, INC.®

Vol. 166, 1354 –1357, October 2001 Printed in U.S.A.

COLLAGEN IMPLANT FOR TREATING STRESS URINARY INCONTINENCE IN WOMEN WITH URETHRAL HYPERMOBILITY A. E. BENT,* J. FOOTE,† S. SIEGEL, G. FAERBER,‡ R. CHAO§

AND

E. A. GORMLEY㛳

From the Greater Baltimore Medical Center, Baltimore, Maryland, Shepherds Center, Atlanta, Georgia, University of Michigan Medical Center, Ann Arbor, Michigan, Metropolitan Urologic Specialists, St. Paul, Minnesota, Cleveland Clinic Florida, Ft. Lauderdale, Florida, and Dartmouth-Hitchcock Medical Center, Lebanon, New Hampshire

ABSTRACT

Purpose: We evaluated the use of collagen in women with stress urinary incontinence and urethral hypermobility. Materials and Methods: We treated 90 patients with stress urinary incontinence and urethral hypermobility with 1 to 3 injections of Contigen implant (C. R. Bard, Inc., Covington, Georgia) between June 1996 and October 1998. Observations at 3, 6 and 12 months included continence grade, 7-day voiding diaries and quality of life questionnaires. Abdominal leak point pressure was determined at baseline and 12 months. Results: At the 6-month followup 68 patients remained in the study, of whom 30 (44%) were dry and 24 (35%) were improved. Of the 58 patients who reached the 12-month followup 19 (33%) were dry and 19 (33%) were improved. Considering all patients entered into the study at 6 months 30 (33%) were dry and 24 (27%) were improved. Of the original 90 patients 19 (21%) were dry and 19 (21%) were improved at 12 months. The probability of maintaining initial improvement for 12 months was 44%. The success of bulking agent therapy was not predicted by the initial incontinence grade. Improved continence grade correlated with improved leak point pressure. Conclusions: This therapy is appropriate in women with urethral hypermobility who wish to avoid surgical risks and in those in whom surgery is ill advised. KEY WORDS: urethra; urinary incontinence, stress; collagen

Stress urinary incontinence is the demonstrable involuntary loss of urine with an increase in intra-abdominal pressure and without detrusor activity. Its 2 key components are intrinsic sphincter deficiency, a term developed in 1992 to describe a damaged urethral sphincter mechanism (the most severe subtype of stress urinary incontinence), and urethral hypermobility, that is abnormal urethral displacement as a result of underlying structural damage or poor support at the bladder neck.1 While the majority of women with stress incontinence have urethral hypermobility, the combination of urethral hypermobility and intrinsic sphincter deficiency is not uncommon.2 Treatment for stress urinary incontinence includes conservative pelvic muscle reeducation with biofeedback, avoidance techniques, mechanical barriers, medication, surgery and urethral bulking agents. A number of bulking agents have been used, including polytetrafluoroethylene, autologous fat, blood, ear cartilage, silicone polymers, carbon coated beads and glutaraldehyde cross-linked collagen (Contigen collagen implant).3 Collagen has been used for intrinsic sphincter deficiency in patients who have minimal urethral mobility with reasonable success.4 – 6 As outlined by Medicare policy and followed by private insurers, the diagnostic criteria for the collagen implant in-

clude abdominal leak point pressure 100 cm. water or less and absent urethral hypermobility.7, 8 However, Contigen implant has been efficacious in patients with urethral hypermobility.9, 10 Clinical experience reveals that in many women of all ages who do not fit the current diagnostic criteria for collagen therapy, that is they have urethral hypermobility, conservative therapy fails and many do not wish to undergo major surgery.11, 12 We assessed the effect of collagen implantation in women with stress incontinence and urethral hypermobility. MATERIALS AND METHODS

Contigen implant is a sterile nonpyrogenic device composed of highly purified bovine dermal collagen lightly crosslinked with glutaraldehyde and dispersed in phosphate buffered physiological saline. It is composed of 95% type I and 5% type III collagen. Injection of collagen materials into the submucosal urethral tissues just distal to the bladder neck creates increased tissue bulk and coaptation of the urethral lumen. Six centers participated in our nonrandomized prospective study of patients receiving collagen implantation for stress incontinence with urethral hypermobility. Incontinence was 12 months in duration and resistant to a 3-month trial of conservative therapy. Previous application of a periurethral bulking agent excluded a patient from study. Stress incontinence was defined according to the classification of Blaivas and Olsson as type I, IIA, IIB or III,13 as determined by videourodynamics or stress cystometrography. Type I stress urinary incontinence is defined as a closed bladder neck at rest situated at or above the inferior margin of the symphysis pubis. During stress the bladder neck and proximal urethra open and descend 2 cm. or less, causing

Accepted for publication May 18, 2001. Supported by a research grant from Bard Urological Division * Financial interest and/or other relationship with Bard, Curis, Pharmacia and Upjohn, Neotonus, Curatek, Genyx, Cook and IMET. † Financial interest and/or other relationship with Alza, Medtronic, Pfizer, Pharmacia and Upjohn, Bayer, Astra-Zeneca and Bard. ‡ Financial interest and/or other relationship with Bard and Pfizer. § Financial interest and/or other relationship with Bard. 㛳 Financial interest and/or other relationship with C. R. Bard, Pharmacia, Alza Pharmaceuticals and Reprogenesis. 1354

1355

COLLAGEN IMPLANT FOR URINARY INCONTINENCE

leakage. Type IIA is defined as a closed bladder neck at rest situated above the inferior margin of the symphysis pubis. During stress the bladder neck and proximal urethra open and descend more than 2 cm., causing leakage. Type IIB is defined as a closed bladder neck at rest situated below the inferior margin of the symphysis pubis. During stress the bladder neck and proximal urethra open, causing leakage. Type III, commonly referred to as intrinsic sphincter deficiency, is defined as an open bladder neck at rest with associated urine leakage during absent detrusor activity. Patients with type III incontinence were excluded from study, as were those with predominant urge incontinence, bladder capacity less than 250 ml., post-void residual urine greater than 50 ml., grade 3 or 4 uterine prolapse or cystocele, neurogenic bladder, fistula, bladder tumor, ␣-adrenergic drug therapy and skin test results positive to collagen. Participants had no active urinary tract infection, autoimmune disease or interstitial cystitis and none was pregnant. Patients were given a total of 3 collagen injections in 6 months with injections at least 1 month apart. Injections were performed using the periurethral or transurethral technique according to investigator preference. All patients also received 3 days of antibiotic prophylaxis. Cases were followed 12 months after the last injection. Incontinence status was determined at the time of the initial injection and at each followup based on the Stamey classification scale, including grade 0 — continence, grade 1— urine loss with sudden increases in abdominal pressure due to coughing, sneezing or laughing, grade 2—leakage with lesser degrees of physical stress, such as walking, standing erect from a sitting position or sitting up in bed and grade 3— complete incontinence, that is urine loss without any relation to physical activity or position.14 Data collection involved 7-day voiding diaries, quality of life questionnaires and incontinence grade assignment completed at baseline, and 1, 3, 6 and 12 months after the last injection. The study included cystometry and abdominal leak point pressure determination at baseline. The protocol required repeat cystometry and leak point pressure measurement at 12 months in women who were not dry. Patients were questioned regarding adverse events at each follow-up and urinalysis was performed at each visit. The primary outcome measure was the number of patients dry at 12 months, as determined by an incontinence grade of zero. Secondary analysis involved the evaluation of patients improved at 12 months, defined as a decrease of at least 1 grade on the incontinence status scale. The duration of improvement, including dryness, was assessed by comparing results at the followup visits. Quality of life assessments were evaluated. Data analysis was performed using Statgraphics Plus for Windows, Version 3.1 and MicroSoft Excel 97SR-2 (Microsoft, Redmond, Washington). The research protocol was approved by the institutional review board at each participating institution. Each patient provided written informed consent. RESULTS

We treated 90 women 35 to 86 years old (mean age 60.9) with 1 to 3 injections of Contigen implant between June 1996 and October 1998. Baseline incontinence classification

graded 16 (18%), 71 (79%) and 3 (3%) cases as 1 to 3, respectively. Of the 32 withdrawals before study completion 14 were due to patient choice, 14 were due to a lack of improvement and 4 women were lost to followup. One to 3 collagen injections were given in 23, 46 and 21 patients, respectively. The average treatment volume was 6.8 ⫾ 3.0 ml. collagen (range 1.5 to 15). The average cumulative volume injected per patient was 13.4 ⫾ 7.6 ml. (range 2.5 to 37.5). Improvement was achieved after an average of 1.9 injections (11.5 ml.). Six months after treatment was completed 30 patients (33%) were dry and 24 (27%) were improved for a total of 60%. Of the 58 women with urethral hypermobility with 12 months of followup 19 (33%) were dry, 19 (33%) were improved and 20 (34%) did not improve by at least 1 continence grade. At the last followup of all 90 patients 19 (21%) were dry and 19 (21%) were improved, while the remainder were not improved and/or had withdrawn from study (table 1). We applied intent-to-treat analysis that considered all patient study withdrawals not improved regardless of status at withdrawal. Lifetable survival analysis to estimate initial improvement duration was done. The probability of maintaining initial improvement for 6 and 12 months was projected to be 52% and 44%, respectively (see figure). The success of bulking agent therapy was not predicted by the initial grade of incontinence. Baseline incontinence grade 1 or 2 was associated with comparable success (table 2). The data show that at 12 months 37% of patients with an initial incontinence grade of 1 and 41% with an initial grade of 2 were dry or improved. Improvement after collagen injection was also evident in grade 3 urethral hypermobility cases. These patients also improved by at least 1 incontinence grade. Of the 49 women with improved leak point pressure at 12 months 21 also had an improved grade of incontinence. Conversely 11 of 49 patients with no improvement in leak point pressure also had improved incontinence. In this study improved leak point pressure correlated with improved incontinence grade. Leak point pressure improved in 21 patients with and 4 without improved grade but it did not improve in 11 with and 13 without improved grade (Fisher’s exact test p ⬍0.007). However, baseline leak point pressure was not predictive of the outcome of urethral hypermobility cases in our study and leak point pressure did not correlate with the amount of collagen injected. A relatively consistent proportion of patients achieved improvement or dryness regardless of the baseline leak point pressure category of less than 60, 60 to 100 or greater than 100 cm. water (table 3). Side effects were those expected with this treatment. Ten patients had 1 or more episodes of urinary retention related to the implant or injection therapy. Most episodes were transient and treated with self-catheterization. Urinary tract infection in 11 cases was related to the implant procedure in 3. An abscess that formed at the injection site under the urethra caused obstructed voiding but resolved spontaneously just before surgical incision. The patient remained dry. Data provided by the 58 women with urethral hypermobility who completed the quality of life questionnaire at 12 months revealed considerable difference in subjective perception and objective clinical measure of therapy results (see

TABLE 1. Urethral hypermobility treatment success at 6 and 12 months, including intent-to-treat analysis Mos. Followup

No. Pts.

No. Dry (%)

No. Improved 1 Continence Grade (%)

No. Dry Improved (%)

No. Not Improved (%)

6 6 Intent-to-Treat 12 12 Intent-to-Treat

68 90 58 90

30 (44) 30 (33) 19 (33) 19 (21)

24 (35) 24 (27) 19 (33) 19 (21)

54 (79) 54 (60) 38 (66) 38 (42)

14 (21) 36 (40) 20 (34) 52 (58)

1356

COLLAGEN IMPLANT FOR URINARY INCONTINENCE

Estimated lifetime survival analysis according to interval approach for duration of initial improvement after 1 to 3 collagen implant injections.

TABLE 2. Treatment success at 12 months by baseline incontinence grade Grade

No. Dry (%)

No. Improved 1 or More Continence Grades (%)

No. Dry Improved (%)

No. Not Improved (%)

16 (18) 71 (79) 3 (3)

6 (37) 13 (18) 0

0 16 (23) 3 (100)

6 (37) 29 (41) 3 (100)

10 (63) 42 (59) 0

90 (100)

19 (21)

19 (21)

38 (42)

52 (58)

No. Pts. (%)

1 2 3 Totals

TABLE 3. Distribution by dry or improved status with baseline leak point pressure No. Cm. Water Leak Point Pressure/Total No. (%) Followup

Less Than 60

60–100

Greater Than 100

Totals

6 Mos. 12 Mos. Last (intent-to-treat)

9/11 (82) 6/9 (67) 6/13 (46)

31/40 (77) 23/31 (74) 23/54 (43)

9/10 (90) 7/11 (64) 7/16 (44)

49/61 (80) 36/51 (71) 36/83 (43)

Appendix). This difference was apparent in patients who were improved but not dry. Clinically 19 of 58 patients were dry and according to subjective quality of life data 20 of 58 considered themselves dry, which was a strong similarity. However, of those who did not achieve dryness objective clinical data indicated that 19 of 58 were improved, while on the subjective quality of life questionnaire 36 of 58 considered themselves improved (wet but sociably acceptable) after therapy. In addition, 96% of patients who increased activity after bulking agent therapy attributed the increase to successful collagen treatment (table 4). DISCUSSION

The data show that Contigen implant was effective for treating patients with urethral hypermobility associated TABLE 4. Select quality of life results at 12 months (see Appendix) Question No.

Question

No. 0 (%)

No. 1 (%)

No. 2 (%)

Totals

1 2 3 4

Incontinence status Incontinence feelings Activity level Higher activity cause

20 (34) 48 (83) 26 (45) 25 (96)

36 (62) 8 (14) 28 (48) 1 (4)

2 (3) 2 (3) 4 (7) Not applicable

58 58 58 26

with stress urinary incontinence. Six months after the completion of 1 to 3 injections of collagen 60% of the patients were dry or improved. At 12 months 42% of women with urethral hypermobility benefited from collagen injection with 21% each dry and improved. These results were slightly lower than those in studies of intrinsic sphincter deficiency. Kim et al noted initial improvement in 86% of intrinsic sphincter deficiency cases and long-term efficacy in 43% after 1 to 4 collagen injections.15 Benshushan et al reported 69% to 77% of cases subjectively cured or improved after 24 months.16 Data on the 94 women with intrinsic sphincter deficiency treated by Smith et al indicated that 67% were dry or much improved (socially continent) at a median followup of 14 months.6 In their patients an average of 2.1 injections (11.9 ml. collagen) were required to achieve continence. Others have indicated that repeat injections are useful and necessary for maintaining satisfactory continence.3 Most retreatment patterns vary. Some groups perform unlimited periodic re-treatment up to the last planned followup, others perform a specified number of injections during 12 months after initial treatment, while yet others limit re-treatment to patients with an unsatisfactory response within an established period. The patients in our urethral hypermobility study were given up to 3 treatments during the initial 6 months of study enrollment, were not re-treated after that initial 6 months and were then followed for 12 months beyond the final injection. In our study the probability of maintaining initial improvement for 6 and 12 months was projected to be 52% and 44%, respectively. A number of study patients received a maximum benefit from only 1 injection and were dry at the 1-year point. The large number of study withdrawals was discouraging but most involved no improvement or the desire for

COLLAGEN IMPLANT FOR URINARY INCONTINENCE

other therapy despite some improvement. All patients, including those who withdrew, were included in the final statistical analysis. Smith et al indicated that more than 80% of intrinsic sphincter deficiency cases with significant improvement after treatment remained successful at 1 year.6 Herschorn and Radomski reported that the probability of remaining dry without additional collagen was 72% at 1 year and 57% at 2.9 Their study also indicated that there was no difference in outcome in patients with intrinsic sphincter deficiency or urethral hypermobility. Likewise Steele et al noted that a small number of patients with hypermobility did at least as well as those without hypermobility at 6 months of followup.10 In contrast, the results of our series are promising for patients with urethral hypermobility but the success rate was lower than for combined intrinsic sphincter deficiency and urethral hypermobility. This finding may reflect the importance of urethral hypermobility in the etiology of many cases of stress incontinence. Improving sphincter bulk may improve a number of cases but it is not sufficient to overcome the forces that promote incontinence in hypermobility of the bladder neck. Most studies mentioned indicate that incontinence grade does not predict the response to collagen treatment, as in our series. Others have also reported improved leak point pressure after collagen therapy,17, 18 as in our study. Overall the literature is inconclusive on the association of improved incontinence grade and increased leak point pressure after treatment as well as the predictive nature of baseline leak point pressure. An expectation of our study was that we would observe a rate of improvement in a significant number of patients with urethral hypermobility that would be maintained at 1 year. If proved, women with urethral hypermobility who would otherwise have only a surgical choice after failed conservative therapy, would have the option of bulking agent therapy. In conclusion, there has been enough success in patients with stress incontinence and urethral hypermobility to recommend collagen injection in those who wish to avoid the risks or recovery time associated with surgery and in those in whom surgery is ill advised due to advanced age, severe medical illness or possible additional pregnancies. APPENDIX: QUALITY OF LIFE QUESTIONNAIRE

Question 1. Do you consider yourself to be (0) Dry; (1) Wet, but socially acceptable; (2) Wet, but not socially acceptable? Question 2. Since receiving your treatment for incontinence, do you feel your incontinence is (0) Better since receiving treatment; (1) The same since receiving treatment; (2) Worse since receiving treatment? Question 3. Since receiving your treatment for incontinence, do you feel your level of activity is (0) Higher since receiving treatment; (1) The same since receiving treatment; (2) Less since receiving treatment? Question 4. If you answered question 3 above with answer “0”, please answer: If your level of activity is

1357

higher since receiving the treatment, do you believe it is due to (0) Treatment with Contigen implant; (1) Other factors? REFERENCES

1. Urinary Incontinence in Adults. Clinical Practice Guideline. Urinary Incontinence Guideline Panel. Agency for Health Care Policy and Research Publication 92-0038. Rockville, Maryland: Public Health Service, United States Department of Health and Human Services, March 1992 2. Urinary Incontinence in Adults: Acute and Chronic Management. Clinical Practice Guideline Update. Urinary Incontinence Guideline Panel. Agency for Health Care Policy and Research Publication 96-0686. Rockville, Maryland: Public Health Service, United States Department of Health and Human Services, March 1996 3. Kershen, R. T. and Atala, A.: New advances in injectable therapies for the treatment of incontinence and vesicoureteral reflux. Urol Clin North Am, 26: 81, 1999 4. McGuire, E. J. and Appell, R. A.: Transurethral collagen injection for urinary incontinence. Urology, 43: 413, 1994 5. Winters, J. C. and Appell, R.: Periurethral injection of collagen in the treatment of intrinsic sphincter deficiency in the female patient. Urol Clin North Am, 22: 673, 1995 6. Smith, D. N., Appell, R. A., Winters, J. C. et al: Collagen injection therapy for female intrinsic sphincter deficiency. J Urol, 157: 1275, 1997 7. Medicare Coverage Issues Manual: Incontinence Control Devices. Washington, D. C.: Department of Health and Human Services, Health Care Financing Administration, Pub. 6 transmittal 70, sect. 65-9, June 1994 8. Medicare Coverage Issues Manual: Incontinence Control Devices. Washington, D. C.: Department of Health and Human Services, Health Care Financing Administration, Pub. 6 transmittal 89, sect. 65-9, September 1996 9. Herschorn, S. and Radomski, S. B.: Collagen injections for genuine stress urinary incontinence: patient selection and durability. Int Urogynecol J Pelvic Floor Dysfunct, 8: 18, 1997 10. Steele, A. C., Kohli, N. and Karram, M. M.: Periurethral collagen injection for stress incontinence with and without urethral hypermobility. Obstet Gynecol, 95: 327, 2000 11. Monga, A. K., Robinson, D. and Stanton, S. L.: Periurethral collagen injections for genuine stress incontinence: a 2-year follow-up. Br J Urol, 76: 156, 1995 12. Gorton, E., Stanton, S., Monga, A. et al: Periurethral collagen injection: a long-term follow-up study. BJU Int, 84: 966, 1999 13. Blaivas, J. G. and Olsson, C. A.: Stress incontinence: classification and surgical approach. J Urol, 139: 727, 1988 14. Dairiki Shortliffe, L. M. and Stamey, T. A.: Stress urinary incontinence. In: Campbell’s Urology, 5th ed. Edited by M. F. Campbell and P. C. Walsh. vol. 3, chapt. 73, p. 2680, 1986 15. Kim, Y. H., Kattan, M. W. and Boone, T. B.: Correlation of urodynamic results and urethral coaptation with success after transurethral collagen injection. Urology, 50: 941, 1997 16. Benshushan, A., Brzezinski, A., Shoshani, O. et al: Periurethral injection for the treatment of urinary incontinence. Obstet Gynecol Surv, 53: 383, 1998 17. Herschorn, S., Radomski, S. B. and Steele, D. J.: Early experience with intraurethral collagen injections for urinary incontinence. J Urol, 148: 1797, 1992 18. Elsergany, R., Elgamasy, A. N. and Ghoniem, G. M.: Transurethral collagen for female stress incontinence. Int Urogynecol J Pelvic Floor Dysfunct, 9: 13, 1998