- Email: [email protected]
Comparison of anterior colporrhaphy and retropubic urethropexy for patients with genuine stress urinary incontinence
Comparison of anterior colporrhaphy and retropubic urethropexy for patients with genuine stress urinary incontinence Robert L. Harris, MD, Christopher A. Yancey, MD, Winfred L. Wiser, MD, John C. Morrison, MD, and G. Rodney Meeks, MD Jackson, Mississippi OBJECTIVE: Our purpose was to compare the efficacy of anterior colporrhaphy and retropubic ureth(opexy performed for genuine stress urinary incontinence. STUDY DESIGN: A retrospective analysis was performed on women who underwent either anterior colporrhaphy or retropubic urethropexy for genuine stress urinary incontinence. Patients were identified by a computer-assisted search, and these women were contacted by telephone. The interview was used to assess current continence status. Variables reviewed included demographic data, medications, hormonal status, current smoking history, significant medical and surgical history, and time to recurrence of incontinence. Operative procedure, prior or concomitant hysterectomy, history of previous incontinence procedures, concomitant surgery for repair of other pelvic floor defects, experience level of the primary surgeon, and duration of postoperative catheterization were also documented. RESULTS: Seventy-six women who had undergone surgery for genuine stress incontinence during a 4-year period were identified and evaluated by telephone interview. Fifty-six had undergone anterior colporrhaphy and 20 retropubic urethropexy. Both groups of patients were comparable in age, social status, race, parity, and weight. The duration of follow-up (mean + SD) was 66.6 + 14.2 months (range 48 to 96 months). Concurrent surgery to repair other pelvic floor defects was more common in patients undergoing anterior colporrhaphy than in patients undergoing retropubic urethropexy (p < 0.05). Of the 56 patients treated with anterior colporrhaphy, 26 (46%) were continent at the time of interview versus 15 of 20 (75%) treated with retropubic urethropexy (p < 0.05). Times to recurrence for anterior colporrhaphy and retropubic urethropexy were not significantly different. History of previous incontinence procedures, concomitant hysterectomy, previous hysterectomy, duration of postoperative catheterization, obesity, chronic lung disease, and smoking were not correlated with success for either procedure. Experience of the primary surgeon did have a significant effect on success, with attending staff having a better cure rate than resident surgeons (p < 0.05). CONCLUSION: Retropubic urethropexy was significantly more effective than anterior colporrhaphy for long-term cure of genuine stress urinary incontinence. We believe these conclusions should be tempered because of the complex nature of genuine stress incontinence. Patients having anterior colporrhaphy may represent a high-risk group because nearly all of them had associated pelvic floor defects. Experience of the surgeon seems to enhance the liklihood of success and may reflect subtle modifications of technique. (AM J OBSTETGYNECOL1995;173:1671-5.)
Key words: Cure, anterior colporrhaphy, retropubic urethropexy, urinary i n c o n t i n e n c e
Few disorders p r o d u c e m o r e patient distress and disc o m f o r t than urinary i n c o n t i n e n c e does. Approximately 5% of w o m e n r e p o r t daily i n c o n t i n e n c e , and one third of w o m e n r e p o r t at least one episode o f i n c o n t i n e n c e per
From the Department of Obstetrics and Gynecology, University of Mississippi Medical Center. Supported in part by the VicksburgHospital Medical Foundation. Presented at the Twenty-firstAnnual Meeting" of the Society of Gynecologic Surgeons, Orlando, Florida, March 6-8, 1995. Reprint requests: G. Rodney Meeks, MD, Department of Obstetrics and Gynecology, University of Mississippi Medical Center, 2500 N. State St., Jackson, MS 39216-4505. Copyright © 1995 by Mosby-Year Book, Inc. 0002-9378/95 $5.00 + 0 6/6/68501
year. 1A c c o r d i n g to the National Institutes of Health, this disorder affects approximately 10 million Americans at a cost >10 billion dollars per annum. 2 T h e most c o m m o n type of urinary i n c o n t i n e n c e in w o m e n is g e n u i n e stress urinary incontinence, defined as socially or hygienically unacceptable involuntary loss of urine when intravesical pressure exceeds m a x i m u m urethral pressure in the absence of a detrusor contraction. "~ Childbirth, especially w h e n associated with trauma to the vagina and p e r i n e u m , seems to be one of the most significant p r e d i s p o s i n g factors for g e n u i n e stress incontinence. 4 A n a t o m i c changes involving the pelvic floor musculature and fascia and loss of urethrovesical angle are 1671
1672
Harriset al.
common in these patients and, although it should not be considered a normal consequence of aging, urinary stress incontinence is often accentuated with loss of estrogenic stimulation and advancing age. Surgery is the therapeutic mainstay for genuine stress incontinence. Both the anterior colporrhaphy with KellyKennedy plication and the retropubic urethropexy (Marshall-Marchetti-Krantz or Burch) are commonly performed procedures. Most studies comparing the efficacy of these procedures report relatively short follow-up, which may be misleading because the incidence of recurrent incontinence after surgery increases over time. 5' 6 The purpose of this study was to compare the long-term success of patients with genuine stress urinary incontinence who are treated with either anterior colporrhaphy or retropubic urethropexy.
Material and methods The hospital records of women at the University of Mississippi Medical Center, Jackson, Mississippi, with a diagnosis of genuine stress urinary incontinence who underwent either anterior colporrhaphy or retropubic urethropexy were identified b y a computer-assisted search. The study interval was 4 years, January 1986 through December 1989. Because the Marshall-Marchetti-Krantz and Burch procedures appear to have similar success for treatment of genuine stress incontinence, they were categorized as retropubic urethropexy without further stratification.7Variables reviewed included demographic data, medications, hormonal status, current smoking history, and significant medical and surgical history. Operative procedure, prior or concomitant hysterectomy, history of previous incontinence procedures, concomitant surgery for repair of other pelvic floor defects, experience level of the primary surgeon, and duration of postoperative catheterization were also documented. Each patient was interviewed by telephone and questioned regarding current status of urinary continence. The interval between surgery and recurrence of incontinence was documented. Treatment success was defined as complete resolution of urinary incontinence, whereas treatment failure was reported if the patient gave any history of recurrent urinary incontinence, without regard to severity. All patients had initial complaints of involuntary loss of urine during physical exertion. Preoperative evaluation included a urine culture and a urinary history in an attempt to exclude other disorders that might contribute to the incontinence. Physical examination included a screening neurologie examination with specific attention directed to the bulbocavernosus and anal reflexes. Pelvic examination was performed, and pelvic supports were noted with the urethrovesical j u n c t i o n assessed by the cotton swab test. A deflection of _>20 degrees from hori-
December 1995 AmJ ObstetGynecol
zontal indicated hypermobility.8 Simple office cystometry was performed in a m a n n e r similar to that described by Ouslander. 9After the patient voided, residual volume was measured and a value of <100 ml was considered normal. The sign of stress incontinence3 was demonstrated by direct visualization of leakage during Valsalva maneuver or cough with the bladder full. Multichannel urodynamic testing was reserved for patients with a history not typical for stress incontinence or with any high-risk history such as urge incontinence, nocturia, nocturnal enuresis, urinary retention, or previous continence surgery. Also, on office cystometry, if any evidence of bladder instability was noted or if the stress test had negative results, multichannel urodynamic studies were performed. Twentyone patients underwent multichannel urodynamic testing to confirm the diagnosis of genuine stress incontinence. Anterior colporrhaphy was performed by extensive mobilization of the vaginal epithelium from the urethral meatus to the vaginal cuff, exposing the pubocervical fascia. The fascia beneath the bladder neck was buttressed with multiple Kelly-Kennedyplication stitches and the fascia over the cystocele was plicated in the midline. Both were performed with absorbable sutures. Rectoceles and enteroceles, if present, were also repaired after appropriate dissection. Retropubic urethropexy was performed by first exposing the space of Retzius and identifying the bladder neck. For the Butch procedure, two nonabsorbable sutures were placed through the paravaginal fascia on each side of the urethrovesical angle and then through the ipsilateral Cooper's ligament. For the Marshall-MarchettiKrantz procedure, a single nonabsorbable suture was placed in the pubocervical fascia on each side of the urethrovesical angle and then through the periosteum of the symphysis pubis. Correction of urethrovesical angle was assessed by vaginal examination. Postoperatively all patients had continuous transurethral bladder drainage for 72 hours, at which time voiding trims were initiated. Consecutive postvoid residual volumes <100 ml were documented before discontinuing use of the catheter. Statistical analysis was performed using Instat (Graph PAD, San Diego), a personal computer statistical program. Student t test or Welch's alternate t test and Z2 or Fisher's exact analyses were applied where appropriate. A p value <0.05 was considered significant.
Results Of the 56 patients treated with anterior colporrhaphy, 26 (46.4%) remained continent. Fifteen of 20 patients (75%) treated with retropubic urethropexy were cured (p = 0.037) (Table I). The mean recurrence intervals for failures were 12.4 _+11.5 months (range 1 to 24 months) for anterior colporrhaphy and 13.6 +_13.6 months ( r a n g e
Harris at al. 1673
Volume 173, Number 6 Am J Obstet Gynecol
Table I. Comparison of incontinence procedures Procedure outcome
Anterior colporrhaphy (n = 56)
Retropubic urethropexy (n = 20)
Cure Failure
26 (46.4%) 30 (53.6%)
15 (75%) 5 (25%)
p = 0.037. 1 to 42 months) for retropubic urethropexy, a nonsignificant difference. The two groups were comparable in age, race, social status, parity, and weight. Medical and surgical comorbid factors and medications did not have a significant impact on success in either group. Diuretic use was noted in 30% of study patients in each group but did not affect outcome. Hormonal status could not be ascertained for every patient during review, so an analysis of effect was not performed. The mean duration of postoperative catheterization after anterior colporrhaphy (5.5 _+7.9 days) and retropubic urethropexy (7.52 _+4.3 days) were similar and did not affect outcome. Obesity (defined as weight >20% above mean weight for height and age), chronic lung disease, current smoking status, concomitant hysterectomy, previous hysterectomy, and history of a previous procedure for incontinence did not have a significant impact on success for either group. The primary surgeon (i.e., performed >50% of the operation) was listed as such in the operative notes. Attending staff as primary surgeons showed a significantly better cure rate than did resident surgeons (p < 0.05) (Table II). Significantly more patients who underwent anterior colporrhaphy also underwent concurrent repair of other pelvic floor defects (93%) than did those undergoing retropubic urethropexy (15%) (p < 0.05) (Table III).
Comment Reported cure rates after surgery for genuine stress urinary incontinence are variable, which may be due to varying definitions of success and to inadequate followup. After retropubic urethropexy Parnell et al. TM reported a cure rate of 90% at 3.75 years, Stanton and Cardoza ~1 87.5% at 0.5 to 2.5 years, and Bergman et al. 5 91% at 1 year. After anterior colporrhaphy Beck and McCormick 12 and Bergman et al. ~ found cure rates of 87% at 2 years and 65% at 1 year, respectively. These are favorable results but with relatively short follow-up. Studies with long-term follow-up are clinically more applicable. In our study the 75% cure rate for the retropubic urethropexy was similar to that reported by Van Geelen et al., 13who also followed up patients for up to 7 years. Although the cure rate for anterior colporrhaphy reported by Van Geelen et al. was 31%, our results were somewhat better, with 46.4% of patients remaining continent. Cure rates for both anterior colporrhaphy and retropubic urethropexy in our study were based on an ex-
Table II. Experience level of primary surgeon
Attending staff Resident
Cures
Failures
22 19
10 25
p = o.o37. tended follow-up interval of 48 to 96 months (66.6 + 14.2 months). The criterion for success included only a report of complete continence. Any degree of urinary incontinence was considered a failure. A "partial resolution" group was not included because we felt it would have interjected subjective bias. Although the effects of prior or concomitant hysterectomy and the effect of prior incontinence surgery remain controversial, these variables had no significant impact on success or failure in this study and these findings are consistent with those of other investigators.12' 14, 15Although other authors have reported obesity, chronic lung disease, and smoking as risk factors for surgical failure, 16'17this was not confirmed by our data. However, our numbers are relatively small to show any difference. Interestingly, diuretic use was noted in 30% of patients in each group and, although known to affect lower urinary tract function, no difference in outcome was noted. Reproducibility of results among surgeons has been suggested as a significant variable when success of individual procedures is assessed. Experience alone is thought to improve outcome. This may explain the fact that attending gynecologists, as primary surgeons, had a better success rate than did resident gynecologists (Table II). The increased cure rate for attending staff surgeons was related primarily to enhanced success with anterior colporrhaphy rather than with retropubic urethropexy. The anterior colporrhaphy is a less standardized procedure than either of the retropubic procedures. This finding suggests that greater surgical experience is advantageous in successfully performing surgery for stress incontinence and pelvic prolapse. Experience may allow subtle modifications of surgical procedure, which may enhance success. Although we emphasize the importance of experience in gynecologic surgery, we also recognize that, although adequate supervision is important, resident surgeons must gain this experience on a somewhat independent level. The data suggest that retropubic urethropexy has a better long-term success than does anterior colporrhaphy
1674
Harris et al.
December 1995 AmJ Obstet Gynecol
Table III. C o n c u r r e n t surgery for o t h e r pelvic floor defects
Posterior colporrhaphy
Anterior colporrhaphy
Retropubic urethropexy
52 (93%)*
3 (15%)
p < 0.05. *Four patients also had iliococcygeus vaginal vault suspension.
for surgical t r e a t m e n t of g e n u i n e stress i n c o n t i n e n c e . Because of the complexity of surgical studies, we believe caution should be applied to the in{erpretation of o u r results. We addressed only i n c o n t i n e n c e and did n o t address r e c u r r e n c e of o t h e r pelvic floor defects. I n d e e d , the decision to choose the vaginal a p p r o a c h and p e r f o r m anterior c o l p o r r h a p h y rather than retropubic urethropexy may have b e e n significantly biased toward patients with o t h e r existing pelvic floor defects, who also may be at h i g h e r risk for r e c u r r e n t i n c o n t i n e n c e (Table III). We are n o t advocating a b a n d o n m e n t of the vaginal a p p r o a c h if p e M c relaxation is coexisting. However, in instances where m o r e severe anatomic changes are present, p e r h a p s a c o m b i n a t i o n of vaginal and abdominal approaches should be used. No single o p e r a t i o n can be used to treat all patients with g e n u i n e stress urinary i n c o n t i n e n c e . However, w h e n the long-term cure of g e n u i n e stress urinary i n c o n t i n e n c e alone is assessed, the r e t r o p u b i c u r e t h r o p e x y was significantly m o r e effective than the anterior colporrhaphy. REFERENCES
1. Wiskind AK, Stanton SL. The Burch colposuspension for genuine stress urinary incontinence. TeLinde's Operative Gynecol Updates 1993;1:1-13. 2. Urinary Incontinence Guideline Panel. Urinary incontinence in adults: clinical practice guidelineS. Rockville, Maryland: Agency for Health Care Policy and Research, 1992:1-12; Public Health Service, United StatesDepartment of Health and Human Services; AHCPR publication no 920038. 3. Abrams P, Blavias JG, Stanton SL, et al. Standardization of terminology of lower urinary tract function. Neurourol Urodynam 1988;7:403-11. 4. Thomas TM. Epidemiology of micturition disorders. In: Stanton SL, ed. Clinical gynecologic urology. St. Louis: Mosby-Year Book, 1984:35-42. 5. Bergman A, Ballard CA, Koonings PP. Comparison of three different surgical procedures for genuine stress incontinence: prospective randomized study. AMJ OBSTETG~ECOL 1989;160:1102-6. 6. Stanton SL, Williams JE, Ritchie D. The colposuspension operation for urinary incontinence. Br J Obstet Gynaecol 1976;83:890-5. 7. Milani R, Scalambrino S, Qadri G, et al. Marshall-MarchettiKrantz procedure and Burch colposuspension in the surgical treatment of female urinary incontinence. Br J Obstet Gynaecol 1985;92:1050-3. 8. Crystle CD, Charme LS, Copeland WE. Q-tip test in stress urinary incontinence. Obstet Gynecol 1971;38:313. 9. OuslanderJG. Diagnostic evaluation of geriatric urinary incontinence. Clin Geriatr Med 1986;2:715-30.
10. Parnell JP, Marshall VM, Vaughu ED. Primary management of urinary stress incontinence by the Marshall-MarchettiKrantz vesicourethropexy. J Urol 1982;127:679-82. 11. Stanton SL, Cardoza LD. A comparison ofvaginal and suprapubic surgery in the correction of incontinence due to urethral correction of incompetence. BrJ Urol 1979;51:497-9. 12. Beck RP, McCormick S. Treatment of stress urinary incontinence with anterior colporrhaphy. Obstet Gynecol 1982;59: 269-74. 13. Van GeelenJM, Theeuwes AGM, Eskes TKAB, Martin CBJr. The clinical and urodynamic effects of anterior vaginal repair and Burch colposuspension. AMJ OBSTETGYNECOL1988; 159:137-44. 14. Langer R, Ron-E1 R, Neuman N, et al. The value of simultaneous hysterectomy during Burch colposuspension for urinary stress incontinence. Obstet Gynecol 1988;72:866-9. 15. Sand PK, Bowen LW, Ostergard DR, Nakanishi AM. Hysterectomy and prior incontinence surgery as risk factors for failed reproductive cystourethropexy. J Reprod Med 1988; 33:171-4. 16. Dwyer PL, Lee ETC, Hay DM. Obesity and urinary incontinence in women. BrJ Obstet Gynaecol 1988;95:91-6. 17. Bump RC, McClish DK. Cigarette smoking and urinary incontinence in women. A~,~ J OBSTET G'e~ECOL 1992;167: 1213-8. Discussion
DR. MARKWAUTFa~S,Cleveland, Ohio. T h e authors present a retrospective review of their surgical e x p e r i e n c e in 76 w o m e n o p e r a t e d on f r o m 1986 to 1989 with a clinical diagnosis of stress urinary incontinence. T h e diagnostic evaluation was apparently standardized and was c o m p l e t e e x c e p t for diagnostic urodynamic testing on all patients. This is a critical deficiency in a study such as this in that, to ensure diagnostic accuracy in all patients so that m e a n ing can be m a d e of the results, u r o d y n a m i c testing is r e q u i r e d on all patients. T h e follow-up interval is >4 years, which is acceptable and meaningful. However, follow-up analysis for cure is by t e l e p h o n e interview only, which is generally conside r e d to be inadequate for a scientific study. A l t h o u g h the a r g u m e n t can be m a d e that if patients were n o t dry they would say so, studies have shown that there is a difference between objective criteria for a cure (physical examination, full b l a d d e r stress testing, and u r o d y n a m i c testing) and subjective definitions of cure (patient reporting). It can only be assumed f r o m prior studies that the cure rates for these surgeries would be somewhat lower if they were objectively studied. I have several questions and comments. (1) T h e authors use the t e r m genuine stress urinary incontinence t h r o u g h o u t this study, but this strictly d e f i n e d diagnostic t e r m implies that every patient had some f o r m of urodynamic testing. Because this was n o t the case, the authors should either say stress urinary incontinence or presumed genuine stress urinary incontinence. (2) Only those w o m e n who could be contacted by t e l e p h o n e were i n c l u d e d in the analysis. H o w m a n y patients were identified in the c o m p u t e r search for the study interval that could n o t be contacted by telephone? (3) Because this was a retrospective chart review, it can be assumed that diagnostic data were occasionally missing or uninterpretable. How was this p r o b l e m h a n d l e d in the analysis? (4) At your university d u r i n g this time how do you think the
Voiume 173, Nmnber 6 AmJ Obstet Gynecol
surgeons chose anterior repair versus retropubic urethropexy for incontinence procedures? (5) You refer several times to "correction of urethrovesical angle" as the goal of surgery. This is outdated terminology and should probably be rephrased. (6) In the Results section, the interval of time that the patient reported recurrent incontinence was noted. Was this determined by patient recall, or were any objective means used? (7) You note that attending physicians had better cure rates with anterior colporrhaphy than resident physicians did. However, with adequate supervision I see no reason why this should be the case. Rather than making the point that attending surgeons are better surgeons than residents are, which is obvious, perhaps you could use this as a quality control measure for your gynecologic surgeons to be better teachers. These data do, however, lend support to the fact that retropubic urethropexies are user friendly with satisfactory results being attainable by even young surgeons. This has also been my experience, and it is one of the reasons I try to make sure that all residents know how to do a Butch procedure before they leave residency. My final question is rhetorical and has to do with whether a study like this still needs to be done. Does anyone still believe that anterior colporrhaphy compares with retropubic urethropexy for the treatment of genuine stress urinary incontinence? Over the past 5 to 10 years nearly all studies, both good and bad, have been relatively consistent. Burch procedures cure incontinence 75% to 90% of the time, and the cure rates drift down only slightly over the 5- to 7-year follow-up time. Anterior colporrhaphy, on the other hand, has 1-year cure rates that range from 40% to 90% but deteriorate from 1 to 5 years in some studies to well below 50%. It has gotten so that, as a urogynecologist, I hardly even consider incontinent patients that have had an anterior colporrhaphy to have recurrent incontinence. For the future, I think that studies like these are valuable only if they are done per fectlywith acceptable urodynamic diagnostic criteria, prospective randomization of surgical techniques, and careful long-term fob low-up with objective outcome criteria. DR. HARRIS (Closing). Because multichannel urodynamic testing was not performed on each patient, the term "genuine stress urinary incontinence," which has been strictly defined by the International Continence Society, should be rephrased to "stress urinary incontinence." We are guilty of using the term to define incon-
Harris et al. 1675
tinence associated with an anatomic defect. Eighty percent of patients who were identified could be contacted for evaluation. All reported data were complete. The hormonal status could not be ascertained for all patients and was thus not reported. In general, patients who underwent colporrhaphy had much more severe pelvic relaxation than those who underwent retropubic urethropexy, and thus patients selected for anterior repair may have been at high risk for failure. Our data and that of many other authors support the fact that when only urinary incontinence is considered the retropubic approach is superior. The term "correction of urethrovesical angle" is one well understood by clinicians who perform this surgery, but support of the bladder neck is a better term. All patients were evaluated by recall and some also by objective data. Because we wished to be consistent, only the recall data were used for this report. Senior resident physicians are given some degree of autonomy in performance of these procedures. Before being allowed to perform these surgeries independently, residents in our program have been involved with >50 procedures and have actually performed 15 to 20 procedures u n d e r supervision. Our point in separating attending and resident physicians is that the retropubic urethropexy is a more standardized operation and therefore easier to learn. Anterior colporrhaphy is less standardized and subject to a great deal of variation from one physician to another. Controversy surrounds whether anterior colporrhaphy or retropubic urethropexies are actually done on comparable patients. Certainly, when only genuine stress urinary incontinence is examined, retropubic urethropexy appears to be superior. The impact that other pelvic floor defects and other associated surgical procedures have on incontinence remains controversial. Although prospective randomized trials of different surgical techniques will ultimately provide objective data, a great deal of information can be gathered when surgeons review their own surgical outcomes. Many surgeons have reservations with strictly randomized protocols because we examine a patient and determine a surgical technique that we think will be most appropriate for that patient. Careful long term follow-up with multichannel urodynamic testing will ultimately answer the question about which techniques have the best success rates.
Key words: Cure, anterior colporrhaphy, retropubic urethropexy, urinary i n c o n t i n e n c e
Few disorders p r o d u c e m o r e patient distress and disc o m f o r t than urinary i n c o n t i n e n c e does. Approximately 5% of w o m e n r e p o r t daily i n c o n t i n e n c e , and one third of w o m e n r e p o r t at least one episode o f i n c o n t i n e n c e per
From the Department of Obstetrics and Gynecology, University of Mississippi Medical Center. Supported in part by the VicksburgHospital Medical Foundation. Presented at the Twenty-firstAnnual Meeting" of the Society of Gynecologic Surgeons, Orlando, Florida, March 6-8, 1995. Reprint requests: G. Rodney Meeks, MD, Department of Obstetrics and Gynecology, University of Mississippi Medical Center, 2500 N. State St., Jackson, MS 39216-4505. Copyright © 1995 by Mosby-Year Book, Inc. 0002-9378/95 $5.00 + 0 6/6/68501
year. 1A c c o r d i n g to the National Institutes of Health, this disorder affects approximately 10 million Americans at a cost >10 billion dollars per annum. 2 T h e most c o m m o n type of urinary i n c o n t i n e n c e in w o m e n is g e n u i n e stress urinary incontinence, defined as socially or hygienically unacceptable involuntary loss of urine when intravesical pressure exceeds m a x i m u m urethral pressure in the absence of a detrusor contraction. "~ Childbirth, especially w h e n associated with trauma to the vagina and p e r i n e u m , seems to be one of the most significant p r e d i s p o s i n g factors for g e n u i n e stress incontinence. 4 A n a t o m i c changes involving the pelvic floor musculature and fascia and loss of urethrovesical angle are 1671
1672
Harriset al.
common in these patients and, although it should not be considered a normal consequence of aging, urinary stress incontinence is often accentuated with loss of estrogenic stimulation and advancing age. Surgery is the therapeutic mainstay for genuine stress incontinence. Both the anterior colporrhaphy with KellyKennedy plication and the retropubic urethropexy (Marshall-Marchetti-Krantz or Burch) are commonly performed procedures. Most studies comparing the efficacy of these procedures report relatively short follow-up, which may be misleading because the incidence of recurrent incontinence after surgery increases over time. 5' 6 The purpose of this study was to compare the long-term success of patients with genuine stress urinary incontinence who are treated with either anterior colporrhaphy or retropubic urethropexy.
Material and methods The hospital records of women at the University of Mississippi Medical Center, Jackson, Mississippi, with a diagnosis of genuine stress urinary incontinence who underwent either anterior colporrhaphy or retropubic urethropexy were identified b y a computer-assisted search. The study interval was 4 years, January 1986 through December 1989. Because the Marshall-Marchetti-Krantz and Burch procedures appear to have similar success for treatment of genuine stress incontinence, they were categorized as retropubic urethropexy without further stratification.7Variables reviewed included demographic data, medications, hormonal status, current smoking history, and significant medical and surgical history. Operative procedure, prior or concomitant hysterectomy, history of previous incontinence procedures, concomitant surgery for repair of other pelvic floor defects, experience level of the primary surgeon, and duration of postoperative catheterization were also documented. Each patient was interviewed by telephone and questioned regarding current status of urinary continence. The interval between surgery and recurrence of incontinence was documented. Treatment success was defined as complete resolution of urinary incontinence, whereas treatment failure was reported if the patient gave any history of recurrent urinary incontinence, without regard to severity. All patients had initial complaints of involuntary loss of urine during physical exertion. Preoperative evaluation included a urine culture and a urinary history in an attempt to exclude other disorders that might contribute to the incontinence. Physical examination included a screening neurologie examination with specific attention directed to the bulbocavernosus and anal reflexes. Pelvic examination was performed, and pelvic supports were noted with the urethrovesical j u n c t i o n assessed by the cotton swab test. A deflection of _>20 degrees from hori-
December 1995 AmJ ObstetGynecol
zontal indicated hypermobility.8 Simple office cystometry was performed in a m a n n e r similar to that described by Ouslander. 9After the patient voided, residual volume was measured and a value of <100 ml was considered normal. The sign of stress incontinence3 was demonstrated by direct visualization of leakage during Valsalva maneuver or cough with the bladder full. Multichannel urodynamic testing was reserved for patients with a history not typical for stress incontinence or with any high-risk history such as urge incontinence, nocturia, nocturnal enuresis, urinary retention, or previous continence surgery. Also, on office cystometry, if any evidence of bladder instability was noted or if the stress test had negative results, multichannel urodynamic studies were performed. Twentyone patients underwent multichannel urodynamic testing to confirm the diagnosis of genuine stress incontinence. Anterior colporrhaphy was performed by extensive mobilization of the vaginal epithelium from the urethral meatus to the vaginal cuff, exposing the pubocervical fascia. The fascia beneath the bladder neck was buttressed with multiple Kelly-Kennedyplication stitches and the fascia over the cystocele was plicated in the midline. Both were performed with absorbable sutures. Rectoceles and enteroceles, if present, were also repaired after appropriate dissection. Retropubic urethropexy was performed by first exposing the space of Retzius and identifying the bladder neck. For the Butch procedure, two nonabsorbable sutures were placed through the paravaginal fascia on each side of the urethrovesical angle and then through the ipsilateral Cooper's ligament. For the Marshall-MarchettiKrantz procedure, a single nonabsorbable suture was placed in the pubocervical fascia on each side of the urethrovesical angle and then through the periosteum of the symphysis pubis. Correction of urethrovesical angle was assessed by vaginal examination. Postoperatively all patients had continuous transurethral bladder drainage for 72 hours, at which time voiding trims were initiated. Consecutive postvoid residual volumes <100 ml were documented before discontinuing use of the catheter. Statistical analysis was performed using Instat (Graph PAD, San Diego), a personal computer statistical program. Student t test or Welch's alternate t test and Z2 or Fisher's exact analyses were applied where appropriate. A p value <0.05 was considered significant.
Results Of the 56 patients treated with anterior colporrhaphy, 26 (46.4%) remained continent. Fifteen of 20 patients (75%) treated with retropubic urethropexy were cured (p = 0.037) (Table I). The mean recurrence intervals for failures were 12.4 _+11.5 months (range 1 to 24 months) for anterior colporrhaphy and 13.6 +_13.6 months ( r a n g e
Harris at al. 1673
Volume 173, Number 6 Am J Obstet Gynecol
Table I. Comparison of incontinence procedures Procedure outcome
Anterior colporrhaphy (n = 56)
Retropubic urethropexy (n = 20)
Cure Failure
26 (46.4%) 30 (53.6%)
15 (75%) 5 (25%)
p = 0.037. 1 to 42 months) for retropubic urethropexy, a nonsignificant difference. The two groups were comparable in age, race, social status, parity, and weight. Medical and surgical comorbid factors and medications did not have a significant impact on success in either group. Diuretic use was noted in 30% of study patients in each group but did not affect outcome. Hormonal status could not be ascertained for every patient during review, so an analysis of effect was not performed. The mean duration of postoperative catheterization after anterior colporrhaphy (5.5 _+7.9 days) and retropubic urethropexy (7.52 _+4.3 days) were similar and did not affect outcome. Obesity (defined as weight >20% above mean weight for height and age), chronic lung disease, current smoking status, concomitant hysterectomy, previous hysterectomy, and history of a previous procedure for incontinence did not have a significant impact on success for either group. The primary surgeon (i.e., performed >50% of the operation) was listed as such in the operative notes. Attending staff as primary surgeons showed a significantly better cure rate than did resident surgeons (p < 0.05) (Table II). Significantly more patients who underwent anterior colporrhaphy also underwent concurrent repair of other pelvic floor defects (93%) than did those undergoing retropubic urethropexy (15%) (p < 0.05) (Table III).
Comment Reported cure rates after surgery for genuine stress urinary incontinence are variable, which may be due to varying definitions of success and to inadequate followup. After retropubic urethropexy Parnell et al. TM reported a cure rate of 90% at 3.75 years, Stanton and Cardoza ~1 87.5% at 0.5 to 2.5 years, and Bergman et al. 5 91% at 1 year. After anterior colporrhaphy Beck and McCormick 12 and Bergman et al. ~ found cure rates of 87% at 2 years and 65% at 1 year, respectively. These are favorable results but with relatively short follow-up. Studies with long-term follow-up are clinically more applicable. In our study the 75% cure rate for the retropubic urethropexy was similar to that reported by Van Geelen et al., 13who also followed up patients for up to 7 years. Although the cure rate for anterior colporrhaphy reported by Van Geelen et al. was 31%, our results were somewhat better, with 46.4% of patients remaining continent. Cure rates for both anterior colporrhaphy and retropubic urethropexy in our study were based on an ex-
Table II. Experience level of primary surgeon
Attending staff Resident
Cures
Failures
22 19
10 25
p = o.o37. tended follow-up interval of 48 to 96 months (66.6 + 14.2 months). The criterion for success included only a report of complete continence. Any degree of urinary incontinence was considered a failure. A "partial resolution" group was not included because we felt it would have interjected subjective bias. Although the effects of prior or concomitant hysterectomy and the effect of prior incontinence surgery remain controversial, these variables had no significant impact on success or failure in this study and these findings are consistent with those of other investigators.12' 14, 15Although other authors have reported obesity, chronic lung disease, and smoking as risk factors for surgical failure, 16'17this was not confirmed by our data. However, our numbers are relatively small to show any difference. Interestingly, diuretic use was noted in 30% of patients in each group and, although known to affect lower urinary tract function, no difference in outcome was noted. Reproducibility of results among surgeons has been suggested as a significant variable when success of individual procedures is assessed. Experience alone is thought to improve outcome. This may explain the fact that attending gynecologists, as primary surgeons, had a better success rate than did resident gynecologists (Table II). The increased cure rate for attending staff surgeons was related primarily to enhanced success with anterior colporrhaphy rather than with retropubic urethropexy. The anterior colporrhaphy is a less standardized procedure than either of the retropubic procedures. This finding suggests that greater surgical experience is advantageous in successfully performing surgery for stress incontinence and pelvic prolapse. Experience may allow subtle modifications of surgical procedure, which may enhance success. Although we emphasize the importance of experience in gynecologic surgery, we also recognize that, although adequate supervision is important, resident surgeons must gain this experience on a somewhat independent level. The data suggest that retropubic urethropexy has a better long-term success than does anterior colporrhaphy
1674
Harris et al.
December 1995 AmJ Obstet Gynecol
Table III. C o n c u r r e n t surgery for o t h e r pelvic floor defects
Posterior colporrhaphy
Anterior colporrhaphy
Retropubic urethropexy
52 (93%)*
3 (15%)
p < 0.05. *Four patients also had iliococcygeus vaginal vault suspension.
for surgical t r e a t m e n t of g e n u i n e stress i n c o n t i n e n c e . Because of the complexity of surgical studies, we believe caution should be applied to the in{erpretation of o u r results. We addressed only i n c o n t i n e n c e and did n o t address r e c u r r e n c e of o t h e r pelvic floor defects. I n d e e d , the decision to choose the vaginal a p p r o a c h and p e r f o r m anterior c o l p o r r h a p h y rather than retropubic urethropexy may have b e e n significantly biased toward patients with o t h e r existing pelvic floor defects, who also may be at h i g h e r risk for r e c u r r e n t i n c o n t i n e n c e (Table III). We are n o t advocating a b a n d o n m e n t of the vaginal a p p r o a c h if p e M c relaxation is coexisting. However, in instances where m o r e severe anatomic changes are present, p e r h a p s a c o m b i n a t i o n of vaginal and abdominal approaches should be used. No single o p e r a t i o n can be used to treat all patients with g e n u i n e stress urinary i n c o n t i n e n c e . However, w h e n the long-term cure of g e n u i n e stress urinary i n c o n t i n e n c e alone is assessed, the r e t r o p u b i c u r e t h r o p e x y was significantly m o r e effective than the anterior colporrhaphy. REFERENCES
1. Wiskind AK, Stanton SL. The Burch colposuspension for genuine stress urinary incontinence. TeLinde's Operative Gynecol Updates 1993;1:1-13. 2. Urinary Incontinence Guideline Panel. Urinary incontinence in adults: clinical practice guidelineS. Rockville, Maryland: Agency for Health Care Policy and Research, 1992:1-12; Public Health Service, United StatesDepartment of Health and Human Services; AHCPR publication no 920038. 3. Abrams P, Blavias JG, Stanton SL, et al. Standardization of terminology of lower urinary tract function. Neurourol Urodynam 1988;7:403-11. 4. Thomas TM. Epidemiology of micturition disorders. In: Stanton SL, ed. Clinical gynecologic urology. St. Louis: Mosby-Year Book, 1984:35-42. 5. Bergman A, Ballard CA, Koonings PP. Comparison of three different surgical procedures for genuine stress incontinence: prospective randomized study. AMJ OBSTETG~ECOL 1989;160:1102-6. 6. Stanton SL, Williams JE, Ritchie D. The colposuspension operation for urinary incontinence. Br J Obstet Gynaecol 1976;83:890-5. 7. Milani R, Scalambrino S, Qadri G, et al. Marshall-MarchettiKrantz procedure and Burch colposuspension in the surgical treatment of female urinary incontinence. Br J Obstet Gynaecol 1985;92:1050-3. 8. Crystle CD, Charme LS, Copeland WE. Q-tip test in stress urinary incontinence. Obstet Gynecol 1971;38:313. 9. OuslanderJG. Diagnostic evaluation of geriatric urinary incontinence. Clin Geriatr Med 1986;2:715-30.
10. Parnell JP, Marshall VM, Vaughu ED. Primary management of urinary stress incontinence by the Marshall-MarchettiKrantz vesicourethropexy. J Urol 1982;127:679-82. 11. Stanton SL, Cardoza LD. A comparison ofvaginal and suprapubic surgery in the correction of incontinence due to urethral correction of incompetence. BrJ Urol 1979;51:497-9. 12. Beck RP, McCormick S. Treatment of stress urinary incontinence with anterior colporrhaphy. Obstet Gynecol 1982;59: 269-74. 13. Van GeelenJM, Theeuwes AGM, Eskes TKAB, Martin CBJr. The clinical and urodynamic effects of anterior vaginal repair and Burch colposuspension. AMJ OBSTETGYNECOL1988; 159:137-44. 14. Langer R, Ron-E1 R, Neuman N, et al. The value of simultaneous hysterectomy during Burch colposuspension for urinary stress incontinence. Obstet Gynecol 1988;72:866-9. 15. Sand PK, Bowen LW, Ostergard DR, Nakanishi AM. Hysterectomy and prior incontinence surgery as risk factors for failed reproductive cystourethropexy. J Reprod Med 1988; 33:171-4. 16. Dwyer PL, Lee ETC, Hay DM. Obesity and urinary incontinence in women. BrJ Obstet Gynaecol 1988;95:91-6. 17. Bump RC, McClish DK. Cigarette smoking and urinary incontinence in women. A~,~ J OBSTET G'e~ECOL 1992;167: 1213-8. Discussion
DR. MARKWAUTFa~S,Cleveland, Ohio. T h e authors present a retrospective review of their surgical e x p e r i e n c e in 76 w o m e n o p e r a t e d on f r o m 1986 to 1989 with a clinical diagnosis of stress urinary incontinence. T h e diagnostic evaluation was apparently standardized and was c o m p l e t e e x c e p t for diagnostic urodynamic testing on all patients. This is a critical deficiency in a study such as this in that, to ensure diagnostic accuracy in all patients so that m e a n ing can be m a d e of the results, u r o d y n a m i c testing is r e q u i r e d on all patients. T h e follow-up interval is >4 years, which is acceptable and meaningful. However, follow-up analysis for cure is by t e l e p h o n e interview only, which is generally conside r e d to be inadequate for a scientific study. A l t h o u g h the a r g u m e n t can be m a d e that if patients were n o t dry they would say so, studies have shown that there is a difference between objective criteria for a cure (physical examination, full b l a d d e r stress testing, and u r o d y n a m i c testing) and subjective definitions of cure (patient reporting). It can only be assumed f r o m prior studies that the cure rates for these surgeries would be somewhat lower if they were objectively studied. I have several questions and comments. (1) T h e authors use the t e r m genuine stress urinary incontinence t h r o u g h o u t this study, but this strictly d e f i n e d diagnostic t e r m implies that every patient had some f o r m of urodynamic testing. Because this was n o t the case, the authors should either say stress urinary incontinence or presumed genuine stress urinary incontinence. (2) Only those w o m e n who could be contacted by t e l e p h o n e were i n c l u d e d in the analysis. H o w m a n y patients were identified in the c o m p u t e r search for the study interval that could n o t be contacted by telephone? (3) Because this was a retrospective chart review, it can be assumed that diagnostic data were occasionally missing or uninterpretable. How was this p r o b l e m h a n d l e d in the analysis? (4) At your university d u r i n g this time how do you think the
Voiume 173, Nmnber 6 AmJ Obstet Gynecol
surgeons chose anterior repair versus retropubic urethropexy for incontinence procedures? (5) You refer several times to "correction of urethrovesical angle" as the goal of surgery. This is outdated terminology and should probably be rephrased. (6) In the Results section, the interval of time that the patient reported recurrent incontinence was noted. Was this determined by patient recall, or were any objective means used? (7) You note that attending physicians had better cure rates with anterior colporrhaphy than resident physicians did. However, with adequate supervision I see no reason why this should be the case. Rather than making the point that attending surgeons are better surgeons than residents are, which is obvious, perhaps you could use this as a quality control measure for your gynecologic surgeons to be better teachers. These data do, however, lend support to the fact that retropubic urethropexies are user friendly with satisfactory results being attainable by even young surgeons. This has also been my experience, and it is one of the reasons I try to make sure that all residents know how to do a Butch procedure before they leave residency. My final question is rhetorical and has to do with whether a study like this still needs to be done. Does anyone still believe that anterior colporrhaphy compares with retropubic urethropexy for the treatment of genuine stress urinary incontinence? Over the past 5 to 10 years nearly all studies, both good and bad, have been relatively consistent. Burch procedures cure incontinence 75% to 90% of the time, and the cure rates drift down only slightly over the 5- to 7-year follow-up time. Anterior colporrhaphy, on the other hand, has 1-year cure rates that range from 40% to 90% but deteriorate from 1 to 5 years in some studies to well below 50%. It has gotten so that, as a urogynecologist, I hardly even consider incontinent patients that have had an anterior colporrhaphy to have recurrent incontinence. For the future, I think that studies like these are valuable only if they are done per fectlywith acceptable urodynamic diagnostic criteria, prospective randomization of surgical techniques, and careful long-term fob low-up with objective outcome criteria. DR. HARRIS (Closing). Because multichannel urodynamic testing was not performed on each patient, the term "genuine stress urinary incontinence," which has been strictly defined by the International Continence Society, should be rephrased to "stress urinary incontinence." We are guilty of using the term to define incon-
Harris et al. 1675
tinence associated with an anatomic defect. Eighty percent of patients who were identified could be contacted for evaluation. All reported data were complete. The hormonal status could not be ascertained for all patients and was thus not reported. In general, patients who underwent colporrhaphy had much more severe pelvic relaxation than those who underwent retropubic urethropexy, and thus patients selected for anterior repair may have been at high risk for failure. Our data and that of many other authors support the fact that when only urinary incontinence is considered the retropubic approach is superior. The term "correction of urethrovesical angle" is one well understood by clinicians who perform this surgery, but support of the bladder neck is a better term. All patients were evaluated by recall and some also by objective data. Because we wished to be consistent, only the recall data were used for this report. Senior resident physicians are given some degree of autonomy in performance of these procedures. Before being allowed to perform these surgeries independently, residents in our program have been involved with >50 procedures and have actually performed 15 to 20 procedures u n d e r supervision. Our point in separating attending and resident physicians is that the retropubic urethropexy is a more standardized operation and therefore easier to learn. Anterior colporrhaphy is less standardized and subject to a great deal of variation from one physician to another. Controversy surrounds whether anterior colporrhaphy or retropubic urethropexies are actually done on comparable patients. Certainly, when only genuine stress urinary incontinence is examined, retropubic urethropexy appears to be superior. The impact that other pelvic floor defects and other associated surgical procedures have on incontinence remains controversial. Although prospective randomized trials of different surgical techniques will ultimately provide objective data, a great deal of information can be gathered when surgeons review their own surgical outcomes. Many surgeons have reservations with strictly randomized protocols because we examine a patient and determine a surgical technique that we think will be most appropriate for that patient. Careful long term follow-up with multichannel urodynamic testing will ultimately answer the question about which techniques have the best success rates.