EARLY EXPERIENCE WITH SMALL VOLUME PERIURETHRAL POLYTETRAFLUOROETHYLENE FOR FEMALE STRESS URINARY INCONTINENCE

0022-5347/00/1636-1838/0 THE JOURNAL OF UROLOGY® Copyright © 2000 by AMERICAN UROLOGICAL ASSOCIATION, INC.®

Vol. 163, 1838 –1842, June 2000 Printed in U.S.A.

EARLY EXPERIENCE WITH SMALL VOLUME PERIURETHRAL POLYTETRAFLUOROETHYLENE FOR FEMALE STRESS URINARY INCONTINENCE SENDER HERSCHORN*

AND

ABRAHAM A. GLAZER

From the Division of Urology, Department of Surgery, University of Toronto, Sunnybrook and Women’s Health Sciences Centre, Toronto, Ontario, Canada

ABSTRACT

Purpose: Injectable polytetrafluoroethylene paste has been used to treat female stress urinary incontinence. Since the substance is antigenically nonreactive we postulated that the reported morbidity may be due in part to the relatively large volumes injected. Therefore, we assessed the outcome of injecting small volumes of polytetrafluoroethylene with attention to the durability of success and morbidity. Materials and Methods: A total of 46 women with a median age of 73.8 years (range 26 to 88) with stress urinary incontinence were treated. The procedure was done on an outpatient basis using topical urethral lidocaine and periurethral lidocaine as anesthesia. A 20Fr urethroscope was used to inject polytetrafluoroethylene through an 18 gauge needle periurethrally. Patients were reinjected 1 to 3 months after initial injection if not cured (no incontinence or no pads). Continence was assessed by questionnaire from an interviewer not known to the patient. Subsequent top-up injections were given if required. Morbidity was monitored with clinical examination and imaging as required. Results: Of the patients 14 (30.4%) were dry, 19 (41.3%) were improved and 13 (28.3%) had treatment failure at a mean of 17.9, 15.9 and 9.1 months, respectively, after last injection. For cured and improved patients (71.7%) an average of 2 treatments were needed, for a total mean volume of 5.4 cc polytetrafluoroethylene. Using a Kaplan-Meier survival curve the probability of the 14 patients remaining dry without additional polytetrafluoroethylene was 90% at 1 and 60% at 2 years. Outcome was not affected by radiological type of incontinence (that is with or without hypermobility), age, pretreatment bladder instability or previous interventions. Conclusions: Periurethral injection of small volumes of polytetrafluoroethylene is effective and safe for treatment of female stress urinary incontinence. Early analysis shows that success is durable with minimal morbidity. KEY WORDS: polytetrafluoroethylene; injections; urinary incontinence, stress; urethra

Various materials have been injected into and around the urethra to treat urinary incontinence since 1938 when Murless first reported the injection of sodium morrhuate around the urethra.1 Polytetrafluoroethylene was used by Arnold2 for vocal cord augmentation and first introduced by Berg3 for the treatment of urinary incontinence before being popularized by Politano et al4 in the 1970s. Although a 73% improvement rate was demonstrated in patients with stress urinary incontinence treated with polytetrafluoroethylene transurethrally,5 these results have been overshadowed by reports of particle migration and granuloma formation.6, 7 Politano4 and Lopez5 et al described injecting 10 to 20 cc polytetrafluoroethylene per treatment to obtain an obstructed appearing urethra. More recent reports using collagen describe using 4 to 6 cc per treatment as a bulking agent.8, 9 We hypothesized that using volumes of polytetrafluoroethylene similar to those being used today for collagen might achieve a similar success rates with less risk of particle migration and granuloma formation. We report our early experience injecting small volumes of polytetrafluoroethylene periurethrally, with attention to the durability of success and morbidity. Accepted for publication December 17, 1999. * Financial interest and/or other relationship with Mentor.

MATERIALS AND METHODS

The study design was a prospective, open label, single arm trial and the primary outcome measure was dryness on the questionnaire. A total of 46 women 26 to 88 years old (mean age 68.9, median 73.8) with stress urinary incontinence were selected to receive polytetrafluoroethylene injections between May 1996 and June 1998. This injectable agent was previously approved by the Health Protections Branch of the Government of Canada for stress urinary incontinence. Evaluation of all cases consisted of video urodynamics and cystoscopy in addition to clinical examination. A history of multiple sclerosis was noted in 2 patients who were ambulatory, and 1 had bladder exstrophy closed at birth. No patient had cognitive impairment. A total of 40 anti-incontinence procedures had been performed in 27 patients (59%), including 12 needle suspensions, 10 retropubic suspensions, 8 slings, 8 anterior repairs and 2 laparoscopic Burch procedures. Of the patients 21 (46%) had also received previous urethral injections of collagen (10), autologous fat (10) or both (1). Injections had been received 2 years or longer before the study. All patients underwent video urodynamic studies before injection, consisting of cystometrography and pressure flow studies when possible. Stress incontinence was classified radiologically as types 1 to 3 according to the method of

1838

1839

PERIURETHRAL POLYTETRAFLUOROETHYLENE FOR STRESS URINARY INCONTINENCE 10

Blaivas. Abdominal leak point pressure was measured simultaneously. All procedures were performed on an outpatient basis with local anesthesia. After topical urethral lidocaine was instilled a 20Fr urethroscope with a 30-degree telescope was inserted into the urethra. The 1% lidocaine was infiltrated at the 3 and 9 o’clock positions, and an 18 gauge 1.5-inch needle was inserted near the urethral meatus and advanced parallel to the urethra up to the bladder neck while its position was confirmed cystoscopically. Additional lidocaine was injected through the needle to confirm submucosal placement as well as for patient comfort. Polytetrafluoroethylene paste was softened by heating in microwaved sterile water and then injected periurethrally to obtain a submucosal bleb that occluded the urethra. Additional injections were performed if required at other sites to obtain mucosal apposition. When urethral occlusion was observed patients were asked to cough and strain in the supine position. If leakage occurred, more polytetrafluoroethylene was given. A maximum of 2 periurethral locations were injected with a maximum of 3 cc at each location for a total maximum of 6 cc polytetrafluoroethylene per treatment. If no incontinence was demonstrated, the patient voided and was discharged home. If the patient could not void a 12Fr Foley catheter was inserted for 1 to 3 days. Patients were reevaluated approximately 6 weeks after initial treatment and polytetrafluoroethylene was reinjected 2 to 3 months after treatment when necessary. Additional injections during followup were performed as needed. The primary outcome measure was achievement of continence as determined by direct patient questioning about symptom grade and pad requirement by a physician interviewer not known to the patient (A. A. G.). Cure was defined as no incontinence, improvement as a decrease in the number of pads as well as subjective improvement described by the patient and success as cure or improvement. Morbidity was monitored with history, physical examination and imaging as required. Data were analyzed to assess duration of success after the last polytetrafluoroethylene injection. Variables included impact of age, radiological type of incontinence, pretreatment bladder instability and previous surgery or injectable agent. Statistical analysis was performed using the chi-square test except for 1-way analysis of variance (ANOVA) to compare the amount of polytetrafluoroethylene required to achieve success for the different radiological types of incontinence and the relationship of outcomes to abdominal leak point pressure.11 A Kaplan-Meier survival curve was used to determine the natural history of treatment for patients who were cured for at least 2 months. Commercial computer software used for analysis. RESULTS

A total of 14 patients (30.4%) were cured at the most recent followup, 19 (41.3%) were improved and 13 (28.3%) had treatment failure (table 1). Mean followup for cured, improved and treatment failure groups was 17.9, 15.9 and 9.1 months, respectively, after last injection. All cured cases were followed at least 12 months after last injection. For successful cases mean amount of polytetrafluoroethylene was 2.7 cc per

treatment, mean number of treatments was 2.0 and mean amount of total polytetrafluoroethylene was 5.4 cc per patient. There were no statistically significant differences in outcomes among women with types 1, 2 or 3 incontinence (p ⫽ 0.1765, table 2). Mean amount of polytetrafluoroethylene to achieve a successful outcome (cured or improved) was significantly lower in women with types 1 and 2 compared to those with type 3 incontinence (1-way ANOVA p ⫽ 0.0085). The relationship of outcomes to pretreatment abdominal leak point pressure is shown in table 3. There were no statistically significant differences in pretreatment leak point pressure for any outcome group. Other factors were also evaluated (table 4). Patient age had no effect on outcome and results were similar for different age groups of patients younger than 60, 60 to 75 and older than 75 years (p ⫽ 0.2851). Success was noted in 7 of 10 patients (70%) with pretreatment urodynamics demonstrating instability and 26 of 36 (72%) with stable bladders (p ⫽ 0.7996). Since surgery and/or injectable agents had failed in 36 patients (78%), the relationship between these interventions and outcomes was analyzed. Table 5 illustrates no significant difference in results with or without previous injectable agent (p ⫽ 0.4404), surgery (p ⫽ 0.4483) or both (p ⫽ 0.67263). Of the patients 15 were dry for at least 2 months after initial injection or series of injections. Followup results showing the time to deterioration from cure or the limit of followup while still dry (censored) were plotted on a Kaplan-Meier survival curve (fig. 1). The probability of remaining dry without additional polytetrafluoroethylene was 90% at 1 year and 60% at 2 years (14 patients were at risk at 12 months and 3 at 24). Morbidity was monitored during followup, which ranged from 11 to 38 months (mean 28.4) after first injection. Of the 7 patients who required a Foley catheter after the procedure 5 had urinary retention following catheter removal, which was treated with clean intermittent catheterization for approximately 1 week. Nonfebrile urinary tract infections in 2 patients resolved with oral antibiotics. Another patient who was dry for 1 year after a 2.5 cc polytetrafluoroethylene injection complained of recent onset slow stream and incomplete bladder emptying. Vaginal examination was negative but transvaginal ultrasound showed a 4 ⫻ 2 ⫻ 2 cm. echo poor mass in the bladder neck region consistent with a reaction to polytetrafluoroethylene. Post-void residual was 100 cc. Obstructive symptoms completely resolved within 4 months. Another 3 cc polytetrafluoroethylene injection was performed 6 months later and the patient has remained dry after 13 months of followup. Abdominal computerized tomography, performed for other reasons 1 year after the obstructive symptoms had resolved showed no abnormality in the bladder neck region. Another patient was treated with nonsteroidal anti-inflammatory medications for prolonged pain that resolved after 3 months. No other problems have been noted. DISCUSSION

Polytetrafluoroethylene has been used to treat urinary incontinence in children,12 men with post-prostatectomy in-

TABLE 1. Status after last polytetrafluoroethylene injection No. pts. (%) Total cc polytetrafluoroethylene (range) Mean treatments (range) Mean cc polytetrafluoroethylene/treatment (range) Mean total cc polytetrafluoroethylene (range) Mean mos. after treatment (range)

Cure

Improvement

Failure

14 (30.4) 5.2 (1.5–10) 1.9 (1–3) 2.7 (1.5–5.0) 5.1 (1.5–10) 12.8 (3–23)

19 (41.3) 5.5 (1.0–18) 2.1 (1–6) 2.6 (1.0–5) 5.5 (1–11.5) 11.6 (2–30)

13 (28.3) 7.8 (3.0–11.5) 2.6 (1–4) 3 (1.5–5) 7.8 (3–11.5) 8.8 (1–24)

1840

PERIURETHRAL POLYTETRAFLUOROETHYLENE FOR STRESS URINARY INCONTINENCE TABLE 2. Radiological type of incontinence and outcome

Incontinence Type

No. Cases*

No. Cured (%)

No. Improved (%)

No. Failure (%)

Mean Cc Polytetrafluoroethylene for Cured ⫹ Improved†

1 7 3 (43) 3 (43) 1 (14) 3.65 2 20 7 (35) 7 (35) 6 (30) 4.64 3 19 4 (21) 9 (47) 6 (32) 7.00 * No significant differences (p ⫽ 0.1765) in comparisons of cured, improved and failure. † One-way ANOVA difference among the group means p ⫽ 0.0085, type 1 versus 2 p ⫽ 0.3909, type 2 versus 3 p ⫽ 0.0136 and type 1 versus 3 p ⫽ 0.0091.

TABLE 3. Outcome and pretreatment abdominal leak point pressure Cured

Improved

Failure

No. pts. 14 19 13 Mean cm. water abdominal leak 90.14 (38) 79.76 (38.37) 99.83 (22.33) point pressure (SD) No significant differences (p ⫽ 0.2766) in comparisons of cure, improved and failure (1-way ANOVA). Cured versus improved p ⫽ 0.3999, cured versus failed p ⫽ 0.4730 and improved versus failed p ⫽ 0.1166.

TABLE 4. Relationship of patient age, bladder instability and outcome

Pt. age (p ⫽ 0.2851): Younger than 60 60–75 Older than 75 Bladder instability (p ⫽ 0.7996): Yes No

No. Pts.

No. Cured (%)

No. Improved (%)

No. Failure (%)

13 14 19

2 (15.4) 4 (28.6) 8 (42.1)

5 (38.4) 6 (42.8) 8 (42.1)

6 (46.2) 4 (28.6) 3 (15.8)

10 36

2 (20) 12 (33.3)

5 (50) 14 (38.9)

3 (30) 10 (27.8)

TABLE 5. Relationship of outcome to previous intervention

Previous surgical repair (p ⫽ 0.4483) No previous repair Previous injectable agent (p ⫽ 0.4404) No injectable agent Previous injectable or surgery (p ⫽ 0.67263) No injectable or surgery

No. Pts.

No. Cured (%)

No. Improved (%)

No. Failure (%)

29

8 (27.6)

14 (48.3)

7 (24.1)

17 21

6 (35.3) 7 (33.3)

5 (29.4) 10 (47.6)

6 (35.3) 4 (19.1)

25 36

7 10

(28) (28)

9 16

(36) (44)

9 10

(36) (28)

10

4

(40)

3

(30)

3

(30)

continence13 and women with stress urinary incontinence.4, 5, 14 –17 In these studies between 7 and 20 cc polytetrafluoroethylene per treatment were used to achieve urethral coaptation and continence. Lopez et al treated 128 women with urinary incontinence, of whom 54.7% became completely continent after treatment and 18% improved for a success rate of 72.7%.5 They used 18.9 cc polytetrafluoroethylene per treatment and averaged 1.5 treatments, for a total of 28.4 cc for each patient. We report similar results (71.7% success rate) using an average of 2.7 cc polytetrafluoroethylene per treatment and an average of 2 treatments, for a total of 5.4 cc for cured and improved cases. Achieving success with smaller volumes of polytetrafluoroethylene depends on placement at the level of the bladder neck and within the walls of the urethra, similar to that reported with other injectables.18 Once the 18 gauge spinal needle is placed periurethrally lidocaine is injected to confirm submucosal location. Only enough polytetrafluoroethylene is given to achieve urethral coaptation and, due to its high molecular weight and high viscosity, it coalesces into a homogeneous mass. At each treatment setting we injected up to 2 locations using a maximum of 6 cc polytetrafluoroethylene. According to Lopez et al the probability of remaining dry at

FIG. 1. Kaplan-Meier curve shows durability of cure of incontinence after last polytetrafluoroethylene injection in 14 cured patients.

1 and 5 years was 82% and 61%, respectively, but whether additional injections were given during the followup period was not specified.5 Others have reported poorer success with long-term followup studies. Beckingham et al reported that only 7% of 26 patients were dry and 20% were improved after 3 years,14 and Harrison et al reported that 11% of 36 patients were dry and 22% were improved after 5 years.15 Deterioration of benefit is seen with other injectables as well.8, 9, 18 However, except for the first report5 larger injected volumes did not ensure good long-term results. Maintenance of good results during reported followup may be due to persistence of the effect or reinjections. All of our results are given after the last injection. Because entry of patients into the study and decline of the effect of polytetrafluoroethylene or limit of followup were staggered, the Kaplan-Meier curve was useful to display the survival rate of cured patients (fig. 1). The probability of remaining dry was 90% at 1 and 60% at 2 years. Median survival has not yet been reached. An important question is efficacy compared to other agents. Glutaraldehyde cross-linked bovine collagen in phosphate buffered saline begins to degrade 12 weeks after injection and is completely degraded in 10 to 19 months.19 However, there is neovascularization, fibroblast ingrowth and host collagen formation in the implant.20 On the other hand, polytetrafluoroethylene is composed of strong carbon-carbon bonds surrounded by a protective screen of fluorine atoms and, thus, has a high molecular weight and viscosity, and is chemically inert.21 Due to its chemical composition and local inflammatory response6 polytetrafluoroethylene might prove to be more durable than collagen. Additionally, we compared the Kaplan-Meier curve to previously reported results with collagen injections and found no significant difference (log rank test p ⫽ 0.3345) (fig. 2).9 Longer followup is needed to demonstrate a difference. The use of injectable agents in patients with intrinsic sphincter deficiency is accepted22 but in those with hypermobility it is controversial. Normal continence results from the musculofascial components with an intrinsically intact urethra. Failure of a component will not invariably produce

PERIURETHRAL POLYTETRAFLUOROETHYLENE FOR STRESS URINARY INCONTINENCE

FIG. 2. Kaplan-Meier curves illustrate durability of cure of 14 patients treated with polytetrafluoroethylene and 78 treated with collagen.9 Log rank test indicates no significant difference.

stress incontinence because of the compensatory effect of the other component, which may explain why many patients with bladder and urethral prolapse can be dry. If women with hypermobility have stress incontinence, the intrinsic sphincter mechanism may not be totally compensating.23 The agent may augment urethral mucosa, and improve coaptation and intrinsic sphincter function as evidenced by an increase in posttreatment abdominal leak pressure.8, 9, 22 It may compensate for hypermobility and render the patient continent. We found no significant difference in outcome for patients with or without hypermobility (table 2). This finding is similar to other reports of collagen injections in patients with hypermobility and genuine stress incontinence.9, 24 However, our patients with type 3 incontinence required the largest amount of polytetrafluoroethylene to achieve the same result compared to those with type 1 or 2. This finding may indicate that the patients with the most severe intrinsic sphincteric deficiency require the largest amount of injected substance. We also found no differences in outcomes and pretreatment leak point pressure. Patient age did not affect the outcome as older women had benefit equal to younger women, which has also been described by others.16, 17 The presence of bladder instability did not impact outcome, whereas it has been noted previously by others to be an adverse risk factor.5, 17 Previous surgery did not affect outcome. Although Vesey et al found an improved success rate in the previous surgery group it did not reach statistical significance.16 Lopez et al reported a worse outcome after multiple previous surgeries due to periurethral fibrosis.5 Prior injections also did not affect outcome. To our knowledge there are no previous reports of using a secondary type of injection following failure of an agent. Minor complications, such as temporary urinary retention and urinary tract infections, are common to all urethral bulking agents18 but the main criticisms of polytetrafluoroethylene for treatment of stress urinary incontinence are particle migration and distant granuloma formation. Although Malizia et al showed distant migration of polytetrafluoroethylene particles to pelvic nodes, lung, brain and kidneys in experimental animals treated with periurethral polytetrafluoroethylene paste,6 to our knowledge only 1 case of clinical significance in humans has been reported in the literature. Claes et al described a woman previously treated with periurethral polytetrafluoroethylene for urinary incontinence who later presented with lymphocytic alveolitis and fever.7 Light microscopy showed polytetrafluoroethylene particles in the lungs. A subsequent letter to the editor questioned the amount of periurethral polytetrafluoroethylene used in this patient, and the authors replied, “The amounts [of polytetrafluoroethylene] used to stop urinary incontinence in the young woman presented, indeed, were large.”25 Mittleman and Marraccini reported an incidental finding of

1841

postmortem interstitial pulmonary granulomas in a previously asymptomatic man who had received 20 to 30 cc polytetrafluoroethylene.26 In a series of 22 women Kiilholma et al reported 3 complications which required surgical intervention, including sterile periurethral abscess, urethral diverticulum and urethral granuloma.27 Their mean polytetrafluoroethylene volume was 7.3 cc per treatment but 2 patients underwent 2 treatments. Although the risk of migration is related to particle size of less than 100 ␮m.6 and the reported clinically significant complication rate is low, it may also be related in part to the volumes injected. We postulated that using small volumes of polytetrafluoroethylene would decrease the risk of future serious complications and after a mean of 28.4 months none had appeared, although longer followup is needed. Our study was designed to test the feasibility of using smaller volumes of polytetrafluoroethylene than previously reported. Major drawbacks are the lack of objective outcome data, such as pad test and validated quality of life questionnaires, which are being addressed in another prospective trial using the same injection technique. CONCLUSIONS

Periurethral injection of small volumes of polytetrafluoroethylene appears to be effective and safe for treatment of female stress urinary incontinence. Success was not affected by patient age, pretreatment bladder instability, radiological type of incontinence or previous intervention. Early analysis shows that the durability of success for small volumes of polytetrafluoroethylene is comparable to that reported with larger volumes, although longer followup is required. Morbidity has been acceptable and we are continuing to monitor patients for long-term complications, which we anticipate may be less than previously reported. Mentor Medical Systems, Canada, provided polytetrafluoroethylene free of charge. REFERENCES

1. Murless, B. C.: The injection treatment of stress incontinence. J Obstet Gynaecol Br Emp, 45: 67, 1938 2. Arnold, G. E.: Vocal rehabilitation of paralytic dysphonia IX. Technique of intracordal injection. Arch Otolaryngol, 76: 358, 1962 3. Berg, S.: Polytef augmentation urethroplasty. Correction of surgically incurable urinary incontinence by injection technique. Arch Surg, 107: 379, 1973 4. Politano, V. A., Small, M., P., Harper, J. M. et al: Periurethral Teflon injection for urinary incontinence. J Urol, 111: 180, 1974 5. Lopez, A. E., Padron, O. F., Patsias, G. et al: Transurethral polytetrafluoroethylene injection in female patients with urinary continence. J Urol, 150: 856, 1993 6. Malizia, A. A., Jr., Reiman, H. M., Myers, R. P. et al: Migration and granulomatous reaction after periurethral injection of polytef (Teflon). JAMA, 251: 3277, 1984 7. Claes, H., Stroobants, D., Van Meerbeek, J. et al: Pulmonary migration following periurethral polytetrafluoroethylene injection for urinary incontinence. J Urol, 142: 821, 1989 8. Winters, J. C. and Appell, R.: Periurethral injection of collagen in the treatment of intrinsic sphincteric deficiency in the female patient. Urol Clin North Am, 22: 673, 1995 9. Herschorn, S., Steele, D. J. and Radomski, S. B.: Follow-up of intraurethral collagen for female stress urinary incontinence. J Urol, 156: 1305, 1996 10. Blaivas, J. G.: Sphincteric incontinence in the female: pathophysiology, classification and choice of corrective surgical procedure. AUA Update Series, vol. VI, lesson 25, 1987 11. Matthews, D. E. and Farewell, V. T.: Using and Understanding Medical Statistics, 3rd ed. Basel: S. Karger, 1996 12. Vorstman, B., Lockhart, J. L., Kaufman, M. et al: Polytetrafluoroethylene injection for urinary incontinence in children. J Urol, 133: 248, 1985

1842

PERIURETHRAL POLYTETRAFLUOROETHYLENE FOR STRESS URINARY INCONTINENCE

13. Politano, V. A.: Transurethral Polytef injection for postprostatectomy urinary incontinence. Br J Urol, 69: 26, 1992 14. Beckingham, I. J., Wemyss-Holden, G. and Lawrence, W. T.: Long-term follow-up of women treated with periurethral Teflon injections for stress incontinence. Br J Urol, 69: 580, 1992 15. Harrison, S. C. W., Brown, C. and O’Boyle, P. J.: Periurethral Teflon for stress urinary incontinence: medium-term results. Br J Urol, 71: 25, 1993 16. Vesey, S. G., Rivett, A. and O’Boyle, P. J.: Teflon injection in female stress incontinence. Effect on urethral pressure profile and flow rate. Br J Urol, 62: 39, 1988 17. Lotenfoe, R., O’Kelly, J. K., Helal, M. et al: Periurethral polytetrafluoroethylene paste injection in incontinent female subjects: surgical indications and improved surgical technique. J Urol, 149: 279, 1993 18. Appell, R. A.: Periurethral injection therapy. In: Campbell’s Urology, 7th ed. Edited by P. C. Walsh, A. B. Retik, E. D. Vaughan, Jr. et al. Philadelphia: W. B. Saunders Co., vol. 1, chapt. 35, pp. 1109 –1120, 1998 19. Canning, D. A., Peters, C. A., Gearhart, J. P. et al: Local tissue reaction to glutaraldehyde crosslinked bovine collagen in the rabbit bladder. J Urol, suppl., 139: 258, abstract 381, 1988

20. Ford, C. N.: Histologic studies on the fate of soluble collagen injected into canine vocal folds. Laryngoscope, 96: 1248, 1986 21. Gangal, S. V.: Polytetrafluoroethylene. In: Encyclopedia of Chemical Technology, 3rd ed. New York: John Wiley & Sons Inc., vol. 2, pp. 1–24, 1980 22. McGuire, E. J. and Appell, R. A.: Transurethral collagen injection for urinary incontinence. Urology, 43: 413, 1994 23. Raz, S., Little, N. and Juma, S.: Female urology. In: Campbell’s Urology, 6th ed. Edited by P. C. Walsh, A. B. Retik, T. A. Stamey et al. Philadelphia: W. B. Saunders Co., vol. 3, chapt. 75, pp. 2782–2828, 1992 24. 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 25. Yachia, D.: Re: Pulmonary migration following periurethral polytetrafluoroethylene injection for urinary incontinence. J Urol, 145: 839, 1991 26. Mittleman, R. E. and Marraccini, J. V.: Pulmonary Teflon granulomas following periurethral Teflon injection for urinary incontinence. Letter to the editor. Arch Pathol Lab Med, 107: 611, 1983 27. Kiilholma, P. J., Chancellor, M. B., Makinen, J. et al: Complications of Teflon injection for stress urinary incontinence. Neurourol Urodyn, 12: 131, 1993