Original Research
Repeat Midurethral Sling Compared With Urethral Bulking for Recurrent Stress Urinary Incontinence Anthony Gaddi, MD, Noelani Guaderrama, and Emily L. Whitcomb, MD, MAS
MD,
Nora Bassiouni,
OBJECTIVE: To compare the effectiveness and safety of repeat midurethral sling with urethral bulking after failed midurethral sling. METHODS: This is a retrospective cohort study of patients within Kaiser Permanente Southern California Medical Group who underwent a midurethral sling for stress urinary incontinence (SUI) from 2008 to 2011 and subsequently had either a midurethral sling or urethral bulking for recurrent SUI. Current Procedural Terminology codes were used to identify patients and electronic medical records were queried for individual patient information. Our primary outcome was either subjective failure defined by SUI or objective failure defined as a positive cough stress test, urodynamic stress incontinence, or retreatment for SUI. Secondary outcomes included perioperative complications and adverse events. RESULTS: Of 6,914 midurethral slings performed, 165 patients underwent a repeat procedure for recurrent SUI, including 98 midurethral slings and 67 urethral bulking. Of the 165 patients who underwent repeat procedures, there were 11 failures (11.2%) in the midurethral sling group and 26 failures (38.8%) in the urethral bulking group (P5.004). There were no differences in perioperative complications or adverse events between the groups. In multivariable logistic regression, From the Departments of Obstetrics and Gynecology and Research and Evaluation, Southern California Permanente Medical Group, Irvine Medical Center, Irvine, California. Supported by a grant from the Regional Research Committee, Southern California Permanente Medical Group. Corresponding author: Emily L. Whitcomb, MD, MAS, Kaiser Permanente Irvine Medical Center, 6650 Alton Parkway, Irvine, CA 92618; e-mail: Emily. [email protected]. Financial Disclosure The authors did not report any potential conflicts of interest. © 2014 by The American College of Obstetricians and Gynecologists. Published by Lippincott Williams & Wilkins. ISSN: 0029-7844/14
VOL. 123, NO. 6, JUNE 2014
MD,
Judith Bebchuk,
ScD,
risk of failure was significantly higher in those undergoing urethral bulking compared with those undergoing midurethral sling (odds ratio 3.49, 95% confidence interval 1.34–9.09, P5.01). CONCLUSIONS: In a managed care population, urethral bulking was associated with higher risk of failure than repeat midurethral sling after primary midurethral sling failure with no differences in perioperative complications or adverse events. (Obstet Gynecol 2014;123:1207–12) DOI: 10.1097/AOG.0000000000000282
LEVEL OF EVIDENCE: II
M
idurethral slings are widely used for the surgical treatment of stress urinary incontinence (SUI). Although midurethral slings are associated with high success rates ranging from 77% to 90% depending on the definition of success,1 approximately 20,000 patients will have recurrent SUI in the United States annually.2 Review of the literature reveals little comparative data on the effectiveness and safety of midurethral slings and urethral bulking for the treatment of recurrent SUI after a midurethral sling. In a 2010 consensus statement drawn from the International Continence Society Research Society, the need for a structured approach to the assessment and management of midurethral sling failures was highlighted.3 Therefore, we aimed to evaluate the effectiveness and safety of repeat midurethral slings and urethral bulking in a large managed care population with recurrent SUI.
MATERIALS AND METHODS We conducted a Kaiser Permanente Southern California institutional review board-approved retrospective cohort study of patients who underwent a midurethral sling for SUI from January 2008 to
OBSTETRICS & GYNECOLOGY
1207
December 2011 within the Southern California Permanente Medical Group (encompassing eight major metropolitan areas) and subsequently had a repeat midurethral sling or urethral bulking for treatment failure. Repeat procedures performed through January 2013 were included in the review. The midurethral slings performed were a variety of retropubic, transobturator, or single-incision approaches. Urethral bulking was performed using either transurethral or periurethral approaches and included Contigen, Coaptite, or Macroplastique. A medical billing query of the electronic medical record was performed using the Current Procedural Terminology code 57288 (sling procedure for female stress incontinence) to identify all patients who received midurethral slings from 2008 to 2011. This group was then queried using Current Procedural Terminology codes 51715 (endoscopic injection of implant material into the submucosal tissues of the urethra, bladder neck, or both) and 57288 (sling procedure for female stress incontinence) to identify patients who underwent a subsequent antiincontinence procedure. An electronic medical record review of patients undergoing repeat procedures after midurethral sling was conducted. We included patients with a documented history of a prior midurethral sling and at least 12-month follow-up after the repeat procedure. Postoperative follow-up intervals were widely variable among surgeons. Although a patient may not have had a 1-year postoperative visit with the surgeon, the medical record was searched through 12 months with any physician to ensure that patients with complaints of SUI were captured. Patient demographic and clinical characteristics including age, race or ethnicity, body mass index (BMI, calculated as weight (kg)/[height (m)]2), menopausal status, and past medical, surgical, and obstetric history were abstracted. Patient characteristics were collected, including preoperative physical examination findings (including cough stress test and Q-tip evaluation), urodynamic observations, preoperative diagnoses, type of initial midurethral sling performed, and repeat procedure performed. Perioperative complications and adverse events (intraoperative complications, readmission within 2 weeks of a repeat procedure, reoperation, infection, or urinary retention) were also abstracted. Although patients with a preoperative diagnosis of intrinsic sphincter deficiency before the repeat procedure were characterized by urethral mobility, urodynamic variables, or both, including Valsalva leak point pressure and maximal urethral closure pressure, this diagnosis
1208
Gaddi et al
was made at the discretion of the surgeon. There was no standardized definition of intrinsic sphincter deficiency. Our primary outcome was subjective or objective failure (subsequently referred to as failure) of the repeat procedure. Subjective failure was defined as symptoms of SUI (involuntary leakage on effort or exertion or on sneezing or coughing), whereas objective failure was defined as a positive stress test (leakage of urine during routine office evaluation with cough or Valsalva in either supine or standing position), urodynamic stress incontinence (leakage of urine with either cough or Valsalva during urodynamic testing), or retreatment for SUI. In addition, patients who experienced failure but reported subjective improvement, defined as less leakage than before the repeat procedure, were captured as “improved” in our chart abstraction. Improvement was based solely on subjective report because no objective measures were used in its determination. Our secondary outcomes were perioperative complications and adverse events. In univariate analysis, categorical variables were assessed using the x2 test and continuous variables with the Kruskal-Wallis test. To address the nonrandom allocation to the type of repeat procedure, we created a propensity score to determine the probability of undergoing a midurethral sling (rather than urethral bulking) for each patient. We then created a logistic model with failure as the outcome, and type of repeat procedure and propensity score as the covariates, to determine the significance of the propensity score. Multivariate logistic regression analysis, including variables significant in univariate analysis or that were deemed to be clinically significant (including vaginal parity, ethnicity, and mixed urinary incontinence) were used to assess independent risk factors for failure while controlling for potential confounders. P values ,.05 were considered to statistically significant. SAS 9.2 software for Windows was used for the statistical analyses.
RESULTS A total of 6,914 midurethral slings were performed within the Southern California Permanente Medical Group between January 2008 and December 2011. Of those patients, 165 (2.4% reoperation rate) underwent repeat anti-incontinence procedures within our system, 98 (59.4%) of whom received midurethral slings and 67 (40.6%) urethral bulking. Patients in the urethral bulking group were older, had lower BMI, and were less likely to have undergone menopause than those undergoing repeat midurethral sling
Surgical Treatment for Recurrent Stress Incontinence
OBSTETRICS & GYNECOLOGY
(Table 1). There were 11 of 98 failures (11.2%) in the midurethral sling group and 26 of 65 failures (38.8%) in the urethral bulking group (P5.004). In univariate analysis, a preoperative diagnosis of intrinsic sphincter deficiency and type of repeat procedure were associated with failure (Table 2). In the logistic model with failure as the outcome and type of repeat procedure and propensity score as the covariates, type of repeat procedure was strongly associated with failure (odds ratio [OR] 4.31, 95% confidence interval [CI] 1.84–10.07, P,.001), but the propensity score was nonsignificant (OR 0.337, 95% CI 0.04–2.82, P5.32). As such, in the full multivariable model, we included age, ethnicity, BMI, and menopausal status and did not include the propensity score. In this model, urethral bulking was associated with higher risk of failure compared with the midurethral sling (OR 3.49, 95% CI 1.34–9.09, P5.01) (Table 3). Failure of the repeat procedure was not associated with age, BMI, vaginal parity, or menopausal status. In addition, a preoperative diagnosis of intrinsic sphincter deficiency was associated with a higher risk of failure compared with absence of intrinsic sphincter deficiency, regardless of procedure performed (OR 4.76, 95% CI 1.32–17.22, P5.02). When excluding patients with a preoperative
diagnosis of intrinsic sphincter deficiency (n520) from the multivariate regression analysis, the OR for treatment failure increased in the urethral bulking group (OR 5.15, 95% CI 2.1–12.6, P,.01). Additional analysis of those who failed the repeat procedure was conducted. Improvement in symptoms was seen in 5 of 11 (four retropubic, one transobturator) patients who underwent the midurethral sling and in 12 of 26 (six Contigen, five Coaptite, one Macroplastique) who underwent urethral bulking. There were no differences in perioperative complications and adverse events between patients undergoing the midurethral sling and urethral bulking. Specifically, there were two perioperative complications in both groups. Of those undergoing the midurethral sling, one patient developed recurrent urinary tract infection and one developed urinary retention requiring sling revision. In the urethral bulking group, two patients had transient urinary retention requiring intermittent self-catheterization.
DISCUSSION Midurethral slings are widely used for the surgical treatment of SUI; however, there are limited data to inform the management of recurrent or persistent SUI after midurethral sling failure. In the present study,
Table 1. Demographic Characteristics of Patients Undergoing Urethral Bulking Compared With Midurethral Sling Repeat Procedure Demographic Characteristic Age (y) Younger than 40 40–49 50–59 60 or older BMI (kg/m2) 25 or less 26–30 31–35 36 or higher Ethnicity Asian Black Hispanic White Menopausal Vaginal parity No vaginal births 1 or 2 vaginal births 3 or more vaginal births
Urethral Bulking (n567)
Midurethral Sling (n598)
P
62.3613.85 4 (6) 9 (13.4) 18 (26.9) 36 (53.7) 28.165.73 22 (32.8) 22 (32.8) 15 (22.4) 8 (11.9)
55.0610.81 8 (8.2) 28 (28.6) 27 (27.6) 35 (35.7) 30.165.45 15 (15.3) 36 (36.7) 32 (32.7) 15 (15.3)
,.001 .06
1 (1.5) 2 (3) 18 (26.9) 46 (68.7) 16 (23.9)
0 (0) 2 (2) 45 (45.9) 51 (52) 42 (42.9)
.06
4 (6.3) 32 (50.8) 27 (42.9)
7 (7.3) 50 (52.1) 39 (40.6)
.95
.02 .06
.01
BMI, body mass index. Data are mean6standard deviation or n (%) unless otherwise specified. Significant P values ,.05 indicated in bold.
VOL. 123, NO. 6, JUNE 2014
Gaddi et al
Surgical Treatment for Recurrent Stress Incontinence
1209
Table 2. Demographic Characteristics of Patients Experiencing No Failure Compared With Failure Stress Urinary Incontinence Demographic Characteristic
No Failure (n5128)
Failure (n537)
P
Age (y) Younger than 40 40–49 50–59 60 or older BMI (kg/m2) 25 or less 26–30 31–35 36 or higher Ethnicity Asian Black Hispanic White Menopausal Vaginal parity No vaginal births 1 or 2 vaginal births 3 or more vaginal births Intrinsic sphincter deficiency Mixed urinary incontinence Repeat procedure Coaptite Macroplastique Contigen Retropubic midurethral sling Transobturator tape Single incision sling Sling as repeat procedure
57.1612.15 9 (7) 31 (24.2) 36 (28.1) 52 (40.6) 29.565.5 26 (20.3) 46 (35.9) 38 (29.7) 18 (14.1)
61.1613.82 3 (8.1) 6 (16.2) 9 (24.3) 19 (51.4) 28.866.1 11 (29.7) 12 (32.4) 9 (24.3) 5 (13.5)
.12 .62
.49 .68
1 4 52 71 79
(0.8) (3.1) (40.6) (55.5) (61.7)
0 (0) 0 (0) 11 (29.7) 26 (70.3) 28 (75.7)
.33
7 70 46 8 20
(5.7) (56.9) (37.4) (6.3) (15.7)
4 (11.1) 12 (33.3) 20 (55.6) 12 (32.4) 5 (13.5)
.05
14 (10.9) 5 (3.9) 22 (17.2) 48 (37.5) 21 (16.4) 18 (14.1) 87 (68)
13 (35.1) 2 (5.4) 11 (29.7) 10 (27) 1 (2.7) 0 (0) 11 (29.7)
.12
,.001 .74 .004
,.001
BMI, body mass index. Data are mean6standard deviation or n (%) unless otherwise specified. Significant P values ,.05 indicated in bold.
patients who underwent urethral bulking experienced a greater than threefold increased risk of failure compared with a repeat midurethral sling. Nevertheless, there was improvement in symptoms in an additional 46% of patients in the urethral bulking group, which compares favorably with the expected benefit of urethral bulking in the literature; that is, of those undergoing urethral bulking (n567), 26 of 67 (38.8%) experienced failure, but 12 of 26 (46%) had subjective improvement of their symptoms. There were baseline differences in age, BMI, and menopausal status between those undergoing the midurethral sling and urethral bulking; however, in the propensity score analysis, these differences were not associated with failure of the repeat procedure. Regardless of the repeat procedure, patients with a preoperative diagnosis of intrinsic sphincter deficiency had a higher risk of failure compared with those without intrinsic sphincter deficiency. When
1210
Gaddi et al
excluding patients with the diagnosis of intrinsic sphincter deficiency from the multivariate analysis, the OR for failure in the urethral bulking group increased. This suggests that including this oftenrefractory population, typically managed with urethral bulking, may not have uniquely influenced overall risk of failure in that group. Comparison of our findings with the literature is made difficult by the fact that we found no studies that included both midurethral slings and urethral bulking for midurethral sling failure. There are two retrospective studies that have evaluated repeat procedures after failed midurethral sling; however, they are limited by small sample sizes, and neither includes both the midurethral sling and urethral bulking. Parden et al2 and Stav et al4 found overall subjective cure rates among repeat midurethral slings to be 54% and 62%, respectively. In a prospective study evaluating the efficacy of the midurethral sling after a failed
Surgical Treatment for Recurrent Stress Incontinence
OBSTETRICS & GYNECOLOGY
Table 3. Multivariable Logistic Regression for Failure of the Repeat Procedure Demographic Characteristic Age (y) Younger than 40 vs 60 or older 40–49 vs 60 or older 50–59 vs 60 or older BMI (kg/m2) Less than 25 vs 36 or higher 26–30 vs 36 or higher 31–35 vs 36 or higher Vaginal parity 3 or more vaginal births vs none 1–2 vaginal births vs none White vs other Menopausal (yes vs no) Intrinsic sphincter deficiency (yes vs no) Mixed urinary incontinence (yes vs no) Urethral bulking vs midurethral sling
Multivariable Model†
P
0.91 (0.22–3.73) 0.53 (0.19–1.47) 0.68 (0.28–1.68)
3.81 (0.29–49.17) 3.07 (0.41–22.82) 1.14 (0.37–3.48)
.31 .27 .82
1.52 (0.45–5.14) 0.94 (0.29–3.05) 0.85 (0.25–2.91)
1.35 (0.31–5.98) 0.93 (0.23–3.78) 0.93 (0.22–3.93)
.69 .92 .93
0.76 0.30 1.90 1.93 7.14 0.84 4.31
0.70 0.22 1.31 3.24 4.76 1.12 3.49
.67 .08 .59 .22 .02 .86 .01
Univariate Models*
(0.20–2.89) (0.07–1.18) (0.86–4.17) (0.84–4.43) (2.64–19.27) (0.29–2.40) (1.84–10.07)
‡
(0.14–3.60) (0.04–1.18) (0.49–3.47) (0.50–21.19) (1.32–17.22) (0.33–3.81) (1.34–9.09)
BMI, body mass index. Data are odds ratio (95% confidence interval) unless otherwise specified. Significant P values ,.05 indicated in bold. * Logistic models with failure as the outcome and the variable in the far left column as a predictor. † The multivariable model includes failure as the outcome and all the variables in the far left column as predictors. The ORs presented in this column are adjusted for all other variables in the model. ‡ Type of repeat procedure was significant at a P value of ,.01. This model includes type of repeat procedure as well as propensity score. The P value for the propensity score was .32.
midurethral sling, Van Baelan et al5 found physicianand patient-based subjective cure rates of 55% and 53%, respectively. This study was limited by the inclusion of only 26 patients at a single institution. In our study with a primary outcome of failure (rather than cure), we found a lower rate of failure (11.2%) in the midurethral sling group, a finding that may be attributed to our definition of failure and the nature of our retrospective review. Lee et al6 retrospectively evaluated a cohort of 23 patients who failed midurethral slings and subsequently underwent transurethral bulking with either Macroplastique or Durasphere. Their rates of cure, defined as a negative cough stress test, were 34.8%, lower than that found in the present study, which may again be attributed to our definition of failure and the nature of our retrospective review. A recent large meta-analysis of Macroplastique conducted by Ghoniem et al7 including 958 patients from 23 cohorts showed short-term (less than 6 months) cure and improvement rates of 43% and 74%, midterm (6– 18 months) of 37% and 73%, and long-term (greater than 18 months) of 36% and 64%, respectively. There were no serious adverse events. In our study, we collected patient data to at least 12 months postoperatively. When extrapolating for cure (nonfailures) in the bulking group, we found a slightly higher rate of 61.2%, but similarly had no serious adverse events.
VOL. 123, NO. 6, JUNE 2014
Gaddi et al
The strengths of our study include analysis of a large cohort of patients who sought repeat procedures for a failed midurethral sling. Our study population consisted of patients from multiple clinical sites within a large and captive health maintenance organization that improve the generalizability of our findings and limit the bias introduced by loss to follow-up. These findings provide preliminary robust ORs that may inform future prospective study. The limitations of this study include those inherent to a retrospective chart review, including the selection bias introduced by lack of standardized assignment to the repeat procedure. That is, patients who underwent urethral bulking were older, had lower BMI, and were less likely to be menopausal than those undergoing the midurethral sling. However, when controlling for treatment allocation in the propensity analysis, it appears these differences did not appear to influence overall failure of the repeat procedure. Given the retrospective nature of this review, standardized definitions or outcomes were not always available in the medical record. The definition of intrinsic sphincter deficiency was not standardized, but rather left to the discretion of the surgeon, and patients may have been underidentified if the surgeon did not diagnose them specifically as such. In addition, our definition of failure makes comparisons with
Surgical Treatment for Recurrent Stress Incontinence
1211
the existing literature difficult. It does not include validated measures of patient satisfaction and quality of life nor does it include validated measures of subjective or objective failure or standardized follow-up intervals. The low reoperation rate after a failed midurethral sling in our population precludes meaningful comparisons between specific types of slings and urethral bulking agents. We also had a low number of complications among repeat procedures, which also limits the conclusions that can be made about the safety of these procedures. Nevertheless, this study provides preliminary baseline data for surgeons when faced with midurethral sling failure. In summary, we found a higher risk of failure in those undergoing urethral bulking compared with the midurethral sling for recurrent SUI in this managed care population. We will use this information to inform preoperative counseling of our patients undergoing repeat procedures for recurrent SUI. We are planning a prospective randomized surgical trial using validated, condition-specific objective and subjective outcome measures.
REFERENCES 1. Oliphant SS, Wang L, Bunker CH, Lowder JL. Trends in stress urinary incontinence inpatient procedures in the United States, 1979-2004. Am J Obstet Gynecol 2009;200: 521.e1–6. 2. Parden AM, Gleason JL, Jauk V, Garner R, Ballard A, Richter HE. Incontinence outcomes in women undergoing primary and repeat midurethral sling procedures. Obstet Gynecol 2013;121:273–8. 3. Smith AR, Artibani W, Drake MJ. Managing unsatisfactory outcome after mid-urethral tape insertion. Neurourol Urodyn 2011; 30:771–4. 4. Stav K, Dwyer P, Rosamilia A, Schierlitz L, Lim YN, Chao F, et al. Repeat mid urethral sling procedure for women with recurrent stress urinary incontinence. J Urol 2010;183:241–6. 5. Van Baelen AA, Delaere KP. Repeat transobturator tape after failed mid-urethral sling procedure: follow-up with questionnaire-based assessment. Urol Int 2009;83:399– 403. 6. Lee HN, Lee YS, Han JY, Jeong JY, Choo MS, Lee KS. Transurethral injection of bulking agent for treatment of failed mid-urethral sling procedures. Int Urogynecol J 2010;21: 1479–83. 7. Ghoniem GM, Miller CJ. A systematic review and meta-analysis of Macroplastique for treating female stress urinary incontinence. Int Urogynecol J 2013;24:27–36.
Artículos de las Series de Especialidad Clínica— Ahora en Español! La traducciones de los artículos de las Series de Especialidad Clínica publicados a partir de abril de 2010 están disponibles en línea solamente en http://www.greenjournal.org. Para ver la colección entera de artículos traducidos, haga click en “Collections” y luego seleccione “Translations (Español).” rev 7/2013
1212
Gaddi et al
Surgical Treatment for Recurrent Stress Incontinence
OBSTETRICS & GYNECOLOGY
Repeat Midurethral Sling Compared With Urethral Bulking for Recurrent Stress Urinary Incontinence Anthony Gaddi, MD, Noelani Guaderrama, and Emily L. Whitcomb, MD, MAS
MD,
Nora Bassiouni,
OBJECTIVE: To compare the effectiveness and safety of repeat midurethral sling with urethral bulking after failed midurethral sling. METHODS: This is a retrospective cohort study of patients within Kaiser Permanente Southern California Medical Group who underwent a midurethral sling for stress urinary incontinence (SUI) from 2008 to 2011 and subsequently had either a midurethral sling or urethral bulking for recurrent SUI. Current Procedural Terminology codes were used to identify patients and electronic medical records were queried for individual patient information. Our primary outcome was either subjective failure defined by SUI or objective failure defined as a positive cough stress test, urodynamic stress incontinence, or retreatment for SUI. Secondary outcomes included perioperative complications and adverse events. RESULTS: Of 6,914 midurethral slings performed, 165 patients underwent a repeat procedure for recurrent SUI, including 98 midurethral slings and 67 urethral bulking. Of the 165 patients who underwent repeat procedures, there were 11 failures (11.2%) in the midurethral sling group and 26 failures (38.8%) in the urethral bulking group (P5.004). There were no differences in perioperative complications or adverse events between the groups. In multivariable logistic regression, From the Departments of Obstetrics and Gynecology and Research and Evaluation, Southern California Permanente Medical Group, Irvine Medical Center, Irvine, California. Supported by a grant from the Regional Research Committee, Southern California Permanente Medical Group. Corresponding author: Emily L. Whitcomb, MD, MAS, Kaiser Permanente Irvine Medical Center, 6650 Alton Parkway, Irvine, CA 92618; e-mail: Emily. [email protected]. Financial Disclosure The authors did not report any potential conflicts of interest. © 2014 by The American College of Obstetricians and Gynecologists. Published by Lippincott Williams & Wilkins. ISSN: 0029-7844/14
VOL. 123, NO. 6, JUNE 2014
MD,
Judith Bebchuk,
ScD,
risk of failure was significantly higher in those undergoing urethral bulking compared with those undergoing midurethral sling (odds ratio 3.49, 95% confidence interval 1.34–9.09, P5.01). CONCLUSIONS: In a managed care population, urethral bulking was associated with higher risk of failure than repeat midurethral sling after primary midurethral sling failure with no differences in perioperative complications or adverse events. (Obstet Gynecol 2014;123:1207–12) DOI: 10.1097/AOG.0000000000000282
LEVEL OF EVIDENCE: II
M
idurethral slings are widely used for the surgical treatment of stress urinary incontinence (SUI). Although midurethral slings are associated with high success rates ranging from 77% to 90% depending on the definition of success,1 approximately 20,000 patients will have recurrent SUI in the United States annually.2 Review of the literature reveals little comparative data on the effectiveness and safety of midurethral slings and urethral bulking for the treatment of recurrent SUI after a midurethral sling. In a 2010 consensus statement drawn from the International Continence Society Research Society, the need for a structured approach to the assessment and management of midurethral sling failures was highlighted.3 Therefore, we aimed to evaluate the effectiveness and safety of repeat midurethral slings and urethral bulking in a large managed care population with recurrent SUI.
MATERIALS AND METHODS We conducted a Kaiser Permanente Southern California institutional review board-approved retrospective cohort study of patients who underwent a midurethral sling for SUI from January 2008 to
OBSTETRICS & GYNECOLOGY
1207
December 2011 within the Southern California Permanente Medical Group (encompassing eight major metropolitan areas) and subsequently had a repeat midurethral sling or urethral bulking for treatment failure. Repeat procedures performed through January 2013 were included in the review. The midurethral slings performed were a variety of retropubic, transobturator, or single-incision approaches. Urethral bulking was performed using either transurethral or periurethral approaches and included Contigen, Coaptite, or Macroplastique. A medical billing query of the electronic medical record was performed using the Current Procedural Terminology code 57288 (sling procedure for female stress incontinence) to identify all patients who received midurethral slings from 2008 to 2011. This group was then queried using Current Procedural Terminology codes 51715 (endoscopic injection of implant material into the submucosal tissues of the urethra, bladder neck, or both) and 57288 (sling procedure for female stress incontinence) to identify patients who underwent a subsequent antiincontinence procedure. An electronic medical record review of patients undergoing repeat procedures after midurethral sling was conducted. We included patients with a documented history of a prior midurethral sling and at least 12-month follow-up after the repeat procedure. Postoperative follow-up intervals were widely variable among surgeons. Although a patient may not have had a 1-year postoperative visit with the surgeon, the medical record was searched through 12 months with any physician to ensure that patients with complaints of SUI were captured. Patient demographic and clinical characteristics including age, race or ethnicity, body mass index (BMI, calculated as weight (kg)/[height (m)]2), menopausal status, and past medical, surgical, and obstetric history were abstracted. Patient characteristics were collected, including preoperative physical examination findings (including cough stress test and Q-tip evaluation), urodynamic observations, preoperative diagnoses, type of initial midurethral sling performed, and repeat procedure performed. Perioperative complications and adverse events (intraoperative complications, readmission within 2 weeks of a repeat procedure, reoperation, infection, or urinary retention) were also abstracted. Although patients with a preoperative diagnosis of intrinsic sphincter deficiency before the repeat procedure were characterized by urethral mobility, urodynamic variables, or both, including Valsalva leak point pressure and maximal urethral closure pressure, this diagnosis
1208
Gaddi et al
was made at the discretion of the surgeon. There was no standardized definition of intrinsic sphincter deficiency. Our primary outcome was subjective or objective failure (subsequently referred to as failure) of the repeat procedure. Subjective failure was defined as symptoms of SUI (involuntary leakage on effort or exertion or on sneezing or coughing), whereas objective failure was defined as a positive stress test (leakage of urine during routine office evaluation with cough or Valsalva in either supine or standing position), urodynamic stress incontinence (leakage of urine with either cough or Valsalva during urodynamic testing), or retreatment for SUI. In addition, patients who experienced failure but reported subjective improvement, defined as less leakage than before the repeat procedure, were captured as “improved” in our chart abstraction. Improvement was based solely on subjective report because no objective measures were used in its determination. Our secondary outcomes were perioperative complications and adverse events. In univariate analysis, categorical variables were assessed using the x2 test and continuous variables with the Kruskal-Wallis test. To address the nonrandom allocation to the type of repeat procedure, we created a propensity score to determine the probability of undergoing a midurethral sling (rather than urethral bulking) for each patient. We then created a logistic model with failure as the outcome, and type of repeat procedure and propensity score as the covariates, to determine the significance of the propensity score. Multivariate logistic regression analysis, including variables significant in univariate analysis or that were deemed to be clinically significant (including vaginal parity, ethnicity, and mixed urinary incontinence) were used to assess independent risk factors for failure while controlling for potential confounders. P values ,.05 were considered to statistically significant. SAS 9.2 software for Windows was used for the statistical analyses.
RESULTS A total of 6,914 midurethral slings were performed within the Southern California Permanente Medical Group between January 2008 and December 2011. Of those patients, 165 (2.4% reoperation rate) underwent repeat anti-incontinence procedures within our system, 98 (59.4%) of whom received midurethral slings and 67 (40.6%) urethral bulking. Patients in the urethral bulking group were older, had lower BMI, and were less likely to have undergone menopause than those undergoing repeat midurethral sling
Surgical Treatment for Recurrent Stress Incontinence
OBSTETRICS & GYNECOLOGY
(Table 1). There were 11 of 98 failures (11.2%) in the midurethral sling group and 26 of 65 failures (38.8%) in the urethral bulking group (P5.004). In univariate analysis, a preoperative diagnosis of intrinsic sphincter deficiency and type of repeat procedure were associated with failure (Table 2). In the logistic model with failure as the outcome and type of repeat procedure and propensity score as the covariates, type of repeat procedure was strongly associated with failure (odds ratio [OR] 4.31, 95% confidence interval [CI] 1.84–10.07, P,.001), but the propensity score was nonsignificant (OR 0.337, 95% CI 0.04–2.82, P5.32). As such, in the full multivariable model, we included age, ethnicity, BMI, and menopausal status and did not include the propensity score. In this model, urethral bulking was associated with higher risk of failure compared with the midurethral sling (OR 3.49, 95% CI 1.34–9.09, P5.01) (Table 3). Failure of the repeat procedure was not associated with age, BMI, vaginal parity, or menopausal status. In addition, a preoperative diagnosis of intrinsic sphincter deficiency was associated with a higher risk of failure compared with absence of intrinsic sphincter deficiency, regardless of procedure performed (OR 4.76, 95% CI 1.32–17.22, P5.02). When excluding patients with a preoperative
diagnosis of intrinsic sphincter deficiency (n520) from the multivariate regression analysis, the OR for treatment failure increased in the urethral bulking group (OR 5.15, 95% CI 2.1–12.6, P,.01). Additional analysis of those who failed the repeat procedure was conducted. Improvement in symptoms was seen in 5 of 11 (four retropubic, one transobturator) patients who underwent the midurethral sling and in 12 of 26 (six Contigen, five Coaptite, one Macroplastique) who underwent urethral bulking. There were no differences in perioperative complications and adverse events between patients undergoing the midurethral sling and urethral bulking. Specifically, there were two perioperative complications in both groups. Of those undergoing the midurethral sling, one patient developed recurrent urinary tract infection and one developed urinary retention requiring sling revision. In the urethral bulking group, two patients had transient urinary retention requiring intermittent self-catheterization.
DISCUSSION Midurethral slings are widely used for the surgical treatment of SUI; however, there are limited data to inform the management of recurrent or persistent SUI after midurethral sling failure. In the present study,
Table 1. Demographic Characteristics of Patients Undergoing Urethral Bulking Compared With Midurethral Sling Repeat Procedure Demographic Characteristic Age (y) Younger than 40 40–49 50–59 60 or older BMI (kg/m2) 25 or less 26–30 31–35 36 or higher Ethnicity Asian Black Hispanic White Menopausal Vaginal parity No vaginal births 1 or 2 vaginal births 3 or more vaginal births
Urethral Bulking (n567)
Midurethral Sling (n598)
P
62.3613.85 4 (6) 9 (13.4) 18 (26.9) 36 (53.7) 28.165.73 22 (32.8) 22 (32.8) 15 (22.4) 8 (11.9)
55.0610.81 8 (8.2) 28 (28.6) 27 (27.6) 35 (35.7) 30.165.45 15 (15.3) 36 (36.7) 32 (32.7) 15 (15.3)
,.001 .06
1 (1.5) 2 (3) 18 (26.9) 46 (68.7) 16 (23.9)
0 (0) 2 (2) 45 (45.9) 51 (52) 42 (42.9)
.06
4 (6.3) 32 (50.8) 27 (42.9)
7 (7.3) 50 (52.1) 39 (40.6)
.95
.02 .06
.01
BMI, body mass index. Data are mean6standard deviation or n (%) unless otherwise specified. Significant P values ,.05 indicated in bold.
VOL. 123, NO. 6, JUNE 2014
Gaddi et al
Surgical Treatment for Recurrent Stress Incontinence
1209
Table 2. Demographic Characteristics of Patients Experiencing No Failure Compared With Failure Stress Urinary Incontinence Demographic Characteristic
No Failure (n5128)
Failure (n537)
P
Age (y) Younger than 40 40–49 50–59 60 or older BMI (kg/m2) 25 or less 26–30 31–35 36 or higher Ethnicity Asian Black Hispanic White Menopausal Vaginal parity No vaginal births 1 or 2 vaginal births 3 or more vaginal births Intrinsic sphincter deficiency Mixed urinary incontinence Repeat procedure Coaptite Macroplastique Contigen Retropubic midurethral sling Transobturator tape Single incision sling Sling as repeat procedure
57.1612.15 9 (7) 31 (24.2) 36 (28.1) 52 (40.6) 29.565.5 26 (20.3) 46 (35.9) 38 (29.7) 18 (14.1)
61.1613.82 3 (8.1) 6 (16.2) 9 (24.3) 19 (51.4) 28.866.1 11 (29.7) 12 (32.4) 9 (24.3) 5 (13.5)
.12 .62
.49 .68
1 4 52 71 79
(0.8) (3.1) (40.6) (55.5) (61.7)
0 (0) 0 (0) 11 (29.7) 26 (70.3) 28 (75.7)
.33
7 70 46 8 20
(5.7) (56.9) (37.4) (6.3) (15.7)
4 (11.1) 12 (33.3) 20 (55.6) 12 (32.4) 5 (13.5)
.05
14 (10.9) 5 (3.9) 22 (17.2) 48 (37.5) 21 (16.4) 18 (14.1) 87 (68)
13 (35.1) 2 (5.4) 11 (29.7) 10 (27) 1 (2.7) 0 (0) 11 (29.7)
.12
,.001 .74 .004
,.001
BMI, body mass index. Data are mean6standard deviation or n (%) unless otherwise specified. Significant P values ,.05 indicated in bold.
patients who underwent urethral bulking experienced a greater than threefold increased risk of failure compared with a repeat midurethral sling. Nevertheless, there was improvement in symptoms in an additional 46% of patients in the urethral bulking group, which compares favorably with the expected benefit of urethral bulking in the literature; that is, of those undergoing urethral bulking (n567), 26 of 67 (38.8%) experienced failure, but 12 of 26 (46%) had subjective improvement of their symptoms. There were baseline differences in age, BMI, and menopausal status between those undergoing the midurethral sling and urethral bulking; however, in the propensity score analysis, these differences were not associated with failure of the repeat procedure. Regardless of the repeat procedure, patients with a preoperative diagnosis of intrinsic sphincter deficiency had a higher risk of failure compared with those without intrinsic sphincter deficiency. When
1210
Gaddi et al
excluding patients with the diagnosis of intrinsic sphincter deficiency from the multivariate analysis, the OR for failure in the urethral bulking group increased. This suggests that including this oftenrefractory population, typically managed with urethral bulking, may not have uniquely influenced overall risk of failure in that group. Comparison of our findings with the literature is made difficult by the fact that we found no studies that included both midurethral slings and urethral bulking for midurethral sling failure. There are two retrospective studies that have evaluated repeat procedures after failed midurethral sling; however, they are limited by small sample sizes, and neither includes both the midurethral sling and urethral bulking. Parden et al2 and Stav et al4 found overall subjective cure rates among repeat midurethral slings to be 54% and 62%, respectively. In a prospective study evaluating the efficacy of the midurethral sling after a failed
Surgical Treatment for Recurrent Stress Incontinence
OBSTETRICS & GYNECOLOGY
Table 3. Multivariable Logistic Regression for Failure of the Repeat Procedure Demographic Characteristic Age (y) Younger than 40 vs 60 or older 40–49 vs 60 or older 50–59 vs 60 or older BMI (kg/m2) Less than 25 vs 36 or higher 26–30 vs 36 or higher 31–35 vs 36 or higher Vaginal parity 3 or more vaginal births vs none 1–2 vaginal births vs none White vs other Menopausal (yes vs no) Intrinsic sphincter deficiency (yes vs no) Mixed urinary incontinence (yes vs no) Urethral bulking vs midurethral sling
Multivariable Model†
P
0.91 (0.22–3.73) 0.53 (0.19–1.47) 0.68 (0.28–1.68)
3.81 (0.29–49.17) 3.07 (0.41–22.82) 1.14 (0.37–3.48)
.31 .27 .82
1.52 (0.45–5.14) 0.94 (0.29–3.05) 0.85 (0.25–2.91)
1.35 (0.31–5.98) 0.93 (0.23–3.78) 0.93 (0.22–3.93)
.69 .92 .93
0.76 0.30 1.90 1.93 7.14 0.84 4.31
0.70 0.22 1.31 3.24 4.76 1.12 3.49
.67 .08 .59 .22 .02 .86 .01
Univariate Models*
(0.20–2.89) (0.07–1.18) (0.86–4.17) (0.84–4.43) (2.64–19.27) (0.29–2.40) (1.84–10.07)
‡
(0.14–3.60) (0.04–1.18) (0.49–3.47) (0.50–21.19) (1.32–17.22) (0.33–3.81) (1.34–9.09)
BMI, body mass index. Data are odds ratio (95% confidence interval) unless otherwise specified. Significant P values ,.05 indicated in bold. * Logistic models with failure as the outcome and the variable in the far left column as a predictor. † The multivariable model includes failure as the outcome and all the variables in the far left column as predictors. The ORs presented in this column are adjusted for all other variables in the model. ‡ Type of repeat procedure was significant at a P value of ,.01. This model includes type of repeat procedure as well as propensity score. The P value for the propensity score was .32.
midurethral sling, Van Baelan et al5 found physicianand patient-based subjective cure rates of 55% and 53%, respectively. This study was limited by the inclusion of only 26 patients at a single institution. In our study with a primary outcome of failure (rather than cure), we found a lower rate of failure (11.2%) in the midurethral sling group, a finding that may be attributed to our definition of failure and the nature of our retrospective review. Lee et al6 retrospectively evaluated a cohort of 23 patients who failed midurethral slings and subsequently underwent transurethral bulking with either Macroplastique or Durasphere. Their rates of cure, defined as a negative cough stress test, were 34.8%, lower than that found in the present study, which may again be attributed to our definition of failure and the nature of our retrospective review. A recent large meta-analysis of Macroplastique conducted by Ghoniem et al7 including 958 patients from 23 cohorts showed short-term (less than 6 months) cure and improvement rates of 43% and 74%, midterm (6– 18 months) of 37% and 73%, and long-term (greater than 18 months) of 36% and 64%, respectively. There were no serious adverse events. In our study, we collected patient data to at least 12 months postoperatively. When extrapolating for cure (nonfailures) in the bulking group, we found a slightly higher rate of 61.2%, but similarly had no serious adverse events.
VOL. 123, NO. 6, JUNE 2014
Gaddi et al
The strengths of our study include analysis of a large cohort of patients who sought repeat procedures for a failed midurethral sling. Our study population consisted of patients from multiple clinical sites within a large and captive health maintenance organization that improve the generalizability of our findings and limit the bias introduced by loss to follow-up. These findings provide preliminary robust ORs that may inform future prospective study. The limitations of this study include those inherent to a retrospective chart review, including the selection bias introduced by lack of standardized assignment to the repeat procedure. That is, patients who underwent urethral bulking were older, had lower BMI, and were less likely to be menopausal than those undergoing the midurethral sling. However, when controlling for treatment allocation in the propensity analysis, it appears these differences did not appear to influence overall failure of the repeat procedure. Given the retrospective nature of this review, standardized definitions or outcomes were not always available in the medical record. The definition of intrinsic sphincter deficiency was not standardized, but rather left to the discretion of the surgeon, and patients may have been underidentified if the surgeon did not diagnose them specifically as such. In addition, our definition of failure makes comparisons with
Surgical Treatment for Recurrent Stress Incontinence
1211
the existing literature difficult. It does not include validated measures of patient satisfaction and quality of life nor does it include validated measures of subjective or objective failure or standardized follow-up intervals. The low reoperation rate after a failed midurethral sling in our population precludes meaningful comparisons between specific types of slings and urethral bulking agents. We also had a low number of complications among repeat procedures, which also limits the conclusions that can be made about the safety of these procedures. Nevertheless, this study provides preliminary baseline data for surgeons when faced with midurethral sling failure. In summary, we found a higher risk of failure in those undergoing urethral bulking compared with the midurethral sling for recurrent SUI in this managed care population. We will use this information to inform preoperative counseling of our patients undergoing repeat procedures for recurrent SUI. We are planning a prospective randomized surgical trial using validated, condition-specific objective and subjective outcome measures.
REFERENCES 1. Oliphant SS, Wang L, Bunker CH, Lowder JL. Trends in stress urinary incontinence inpatient procedures in the United States, 1979-2004. Am J Obstet Gynecol 2009;200: 521.e1–6. 2. Parden AM, Gleason JL, Jauk V, Garner R, Ballard A, Richter HE. Incontinence outcomes in women undergoing primary and repeat midurethral sling procedures. Obstet Gynecol 2013;121:273–8. 3. Smith AR, Artibani W, Drake MJ. Managing unsatisfactory outcome after mid-urethral tape insertion. Neurourol Urodyn 2011; 30:771–4. 4. Stav K, Dwyer P, Rosamilia A, Schierlitz L, Lim YN, Chao F, et al. Repeat mid urethral sling procedure for women with recurrent stress urinary incontinence. J Urol 2010;183:241–6. 5. Van Baelen AA, Delaere KP. Repeat transobturator tape after failed mid-urethral sling procedure: follow-up with questionnaire-based assessment. Urol Int 2009;83:399– 403. 6. Lee HN, Lee YS, Han JY, Jeong JY, Choo MS, Lee KS. Transurethral injection of bulking agent for treatment of failed mid-urethral sling procedures. Int Urogynecol J 2010;21: 1479–83. 7. Ghoniem GM, Miller CJ. A systematic review and meta-analysis of Macroplastique for treating female stress urinary incontinence. Int Urogynecol J 2013;24:27–36.
Artículos de las Series de Especialidad Clínica— Ahora en Español! La traducciones de los artículos de las Series de Especialidad Clínica publicados a partir de abril de 2010 están disponibles en línea solamente en http://www.greenjournal.org. Para ver la colección entera de artículos traducidos, haga click en “Collections” y luego seleccione “Translations (Español).” rev 7/2013
1212
Gaddi et al
Surgical Treatment for Recurrent Stress Incontinence
OBSTETRICS & GYNECOLOGY
