Background
Urethral diverticulum is a rare entity and requires a high suspicion for diagnosis based on symptoms and physical exam with confirmation by imaging. A common presenting symptom is stress urinary incontinence (SUI). The recommended treatment is surgical excision with urethral diverticulectomy. Postoperatively, approximately 37% of patients may have persistent and 16% may have de novo SUI. An autologous fascial pubovaginal sling (PVS) placed at the time of urethral diverticulectomy (UD) has the potential to prevent and treat postoperative SUI. However, little has been published about the safety and efficacy of a concomitant pubovaginal sling.
Objective
The objective of this study was to compare the clinical presentation, outcomes, complications, and diverticulum recurrence rates in women who underwent a urethral diverticulectomy with vs without a concurrent pubovaginal sling.
Study Design
This multicenter, retrospective cohort study included women who underwent a urethral diverticulectomy between January 1, 2000, and December 31, 2016. Study participants were identified by Current Procedure Terminology codes, and their records were reviewed for demographics, medical or surgical history, symptoms, preoperative testing, concomitant surgeries, and postoperative outcomes. Symptoms, recurrence rates, and complications were compared between women with and without a concomitant pubovaginal sling. The primary outcome was the presence of postoperative stress urinary incontinence symptoms. Based on a stress urinary incontinence rate of 50% with no pubovaginal sling and 10% with a pubovaginal sling, we needed a sample size of 141 participants who underwent diverticulectomy without a pubovaginal sling and 8 participants with a pubovaginal sling to achieve 83% power with P <.05.
Results
We identified 485 diverticulectomy cases from 11 institutions who met the inclusion criteria; of these, 96 (19.7%) cases had a concomitant pubovaginal sling. Women with a pubovaginal sling were older than those without a pubovaginal sling (53 years vs 46 years; P <.001), and a greater number of women with pubovaginal sling had undergone diverticulectomy previously (31% vs 8%; P <.001). Postoperative follow-up period (14.6±26.9 months) was similar between the groups. The pubovaginal sling group had greater preoperative stress urinary incontinence (71% vs 33%; P <.0001), dysuria (47% vs 30%; P =.002), and recurrent urinary tract infection (49% vs 33%; P =.004). The addition of a pubovaginal sling at the time of diverticulectomy significantly improved the odds of stress urinary incontinence resolution after adjusting for prior diverticulectomy, prior incontinence surgery, age, race, and parity (adjusted odds ratio, 2.27; 95% confidence interval, 1.02–5.03; P =.043). It was not significantly protective against de novo stress urinary incontinence (adjusted odds ratio, 0.86; 95% confidence interval, 0.25–2.92; P =.807). Concomitant pubovaginal sling increased the odds of postoperative short-term (<6 weeks) urinary retention (adjusted odds ratio, 2.5; 95% confidence interval, 1.04–6.22; P =.039) and long-term urinary retention (>6 weeks) (adjusted odds ratio, 6.98; 95% confidence interval, 2.20–22.11; P =.001), as well as recurrent urinary tract infections (adjusted odds ratio, 3.27; 95% confidence interval, 1.26–7.76; P =.013). There was no significant risk to develop a de novo overactive bladder (adjusted odds ratio, 1.48; 95% confidence interval, 0.56–3.91; P =.423) or urgency urinary incontinence (adjusted odds ratio, 1.47; 95% confidence interval, 0.71–3.06; P =.30). A concomitant pubovaginal sling was not protective against a recurrent diverticulum (adjusted odds ratio, 1.38; 95% confidence interval, 0.67–2.82; P =.374). Overall, the diverticulum recurrence rate was 10.1% and did not differ between the groups.
Conclusion
This large retrospective cohort study demonstrated a greater resolution of stress urinary incontinence with the addition of a pubovaginal sling at the time of a urethral diverticulectomy. There was a considerable risk of postoperative urinary retention and recurrent urinary tract infections in the pubovaginal sling group.
A urethral diverticulum is a rare condition with an annual incidence of 0.02% per year. Diagnosis of urethral diverticulum is usually based on the evaluation of symptoms and a physical examination, but it is confirmed with imaging. The preoperative symptoms vary but may include dysuria, urethral pain, recurrent urinary tract infections (UTIs), a vaginal bulge, and stress urinary incontinence (SUI). Pelvic magnetic resonance imaging (MRI) is recommended before surgery to rule out other causes of an anterior vaginal bulge and to provide information on the location, size, and shape of the diverticulum to allow for more accurate surgical planning. , The recommended treatment is surgical excision during a urethral diverticulectomy (UD). Postoperatively, approximately 37% of patients may have persistent SUI compared with 16% who may have de novo SUI. ,
Why was this study conducted?
This study aimed to compare the postoperative symptoms, specifically stress urinary incontinence, and postoperative complications in women who underwent a urethral diverticulectomy with a concomitant pubovaginal sling vs those who underwent a urethral diverticulectomy without a concomitant pubovaginal sling.
Key findings
Adding a pubovaginal sling at the time of urethral diverticulectomy markedly increases the resolution of stress urinary incontinence with an expected increase in the rates of urinary retention and urinary tract infections.
What does this add to what is known?
To our knowledge, this is the largest multicenter study focusing on postoperative symptoms after urethral diverticulectomy. It is also the first multicenter study comparing the outcomes of women who underwent urethral diverticulectomy with vs without a pubovaginal sling.
An autologous fascial pubovaginal sling (PVS) placed at the time of UD has the potential to prevent and treat postoperative SUI. However, little has been published about the safety and efficacy of a concomitant PVS. The largest published case series (n=38) followed women with SUI and urethral diverticulum who underwent concomitant PVS at the time of their UD. Of the patients in that study, 19% (n=34) reported a complete resolution of SUI symptoms with a mean follow-up of 12.7 months. The morbidities reported when PVS was performed with a UD include urinary retention, harvest site complications, and an increased operating time of approximately 54 min (25–140 minute range). The published cases of urinary retention rates that required catheterization after a PVS were approximately 10% by 6 weeks postoperatively, and the majority of these were resolved by 6 months. In addition, SUI may resolve with diverticulectomy alone without further need for another surgical procedure. A 2019 study followed 100 women with urodynamic SUI (USUI) who underwent UD. Of the 27 patients (29%) who had USUI preoperatively, 8 (30%) subjects’ SUI resolved with UD alone. Thus, many surgeons do not offer or perform concomitant PVS at the time of UD. A study comparing the outcomes of patients undergoing UD with and without PVS could better guide the decision to offer PVS in patients planning to undergo a diverticulectomy.
We are not aware of any adequately powered published studies evaluating postoperative SUI in patients undergoing diverticulectomy with and without PVS. The primary aim of this study was to compare postoperative SUI (de novo and persistent) after diverticulectomy with and without a concomitant PVS. We hypothesized that the addition of a PVS during diverticulectomy would decrease postoperative SUI more than the diverticulectomy alone. We also compared the complications, recurrence rates, and symptom resolution or persistence after UD between the 2 groups.
Materials and Methods
We conducted a multicenter, retrospective cohort study comparing the outcomes and patient characteristics of women who underwent UD with and without a concomitant PVS. This study was sponsored by the American Urogynecologic Society and the Society of Gynecologic Surgeons Fellows Pelvic Research Network, and the data were collected at 11 Female Pelvic Medicine and Reconstructive Surgery institutions in the United States. This study was approved by the institutional review boards at each site. This study adheres to the Strengthening the Reporting of Observational Studies in Epidemiology guidelines.
We included all UDs that were performed by Female Pelvic Medicine and Reconstructive Surgeons over a 16-year period from January 1, 2000 to December 31, 2016. Women who underwent UD were identified electronically using the Current Procedural Terminology codes (53230—female urethral diverticulectomy and 53400—female urethroplasty). We reviewed operative reports to verify that a UD was performed and excluded Skene’s gland or other periurethral lesion excisions. We only included the index UD at an institution and excluded subsequent diverticulectomy procedures performed on the same patient during the study period. We excluded subjects who had concomitant urinary incontinence procedures other than autologous PVS (such as retropubic colposuspension, needle suspension, injection of bulking agent, and synthetic mesh slings). We also excluded subjects who had no postoperative follow-up visits.
We extracted data on patient demographics, clinical presentation, diagnostic testing, and management from inpatient and outpatient records including preoperative and postoperative notes, operative reports, and discharge summaries. The operative reports were reviewed for concomitant procedures, surgical techniques, and intraoperative complications. Cases with concomitant PVS were identified via operative reports. PVS cases were included if autologous tissue was used, harvested either from the rectus fascia or fascia lata. The length of the graft used for sling material was not collected. Postoperative outcomes included symptoms, complications, treatment for SUI or an overactive bladder (OAB), and diverticulum recurrences with or without reoperation. Study data were recorded and managed using REDCap electronic data capture tools at each site. A uniform collection tool on REDCap was used at all sites.
The primary outcome was postoperative SUI (de novo and persistent) between the 2 groups, which was diagnosed by the presence of SUI symptoms. All follow-up visits after the index procedure were reviewed for SUI symptoms, and the onset of symptoms and treatments for SUI were recorded. If any SUI symptom was documented, this was considered symptomatic SUI. No severity index for SUI was used because the documentation of SUI varied between the institutions.
Secondary outcomes included symptoms on presentation of diverticulum, surgical history, medical history, concomitant surgeries (Martius flap, PVS, or other surgeries for SUI), peri- and postoperative complications, diverticulum recurrence, resolution of the presenting symptoms after diverticulectomy, and the need for further treatment of urinary incontinence after diverticulectomy. These outcomes were then compared between the cohorts.
Symptomatic outcomes assessed included OAB, defined as urinary urgency, frequency, or nocturia, and urgency urinary incontinence (UUI), defined as urinary incontinence associated with urinary urgency. Subjects were identified as having OAB and/or UUI if documented in their history. No severity index was used. Other outcomes included recurrent UTIs (identified by >2 positive cultures in 6 months or >3 in 12 months as per records), a vaginal bulge, dysuria, vaginal pain, dyspareunia, postvoid dribbling, and voiding dysfunction (including hesitancy, sensation of incomplete emptying, and abnormal urinary stream) documented in pre- and/or postoperative follow-up documentation. Postoperative complication outcomes included UTI (positive culture >100,000 colony-forming units bacteria), urinary retention (postvoid residual >150 cc or requiring catheterization or indwelling Foley catheter), wound infection (identified as cellulitis or abscess), readmission related to the index surgery, and reoperation (not including for recurrent diverticulum or SUI) found in postoperative documentation <6 weeks postoperatively. Long-term complications identified in postoperative documentation >6 weeks postoperatively included urethral stricture (identified on exam or cystoscopy), urethrovaginal or vesicovaginal fistula (identified on examination or cystoscopy), recurrent UTIs (>2 positive cultures in 6 months or >3 in 12 months after surgery), and urinary retention (>150 cc requiring reoperation, catheterization, or indwelling Foley catheter). Diverticulum recurrence was identified in postoperative follow-up documentation by examination findings, cystoscopy, MRI, or operative reports for reoperation.
We calculated our sample size for each cohort (UD with or without PVS) based on previous reported rates of postoperative SUI after diverticulectomy. Using an estimated overall 50% rate of postoperative SUI (14% de novo and 36% persistent) after diverticulectomy alone based on previous literature, we estimated postoperative SUI rates to be 10% after diverticulectomy with PVS based on prior PVS outcome studies. We used a confidence interval (CI) of 95% and a P value of .05 to achieve 83% power to detect a difference between the groups. Group proportions of 0.4 and an estimated 1:20 ratio between the groups were selected given fewer anticipated cases of UD with concomitant PVS. We then calculated sample sizes of 141 in the diverticulectomy without PVS group and 8 in diverticulectomy with PVS group. Although a priori power calculation was performed, all cases performed at these institutions meeting inclusion and exclusion criteria in the study time frame were included in the analysis.
The collected data were exported to R 3.4 for analysis, using the epitools package. Categorical variables were analyzed using χ 2 tests. Continuous variables were analyzed using Student t-test. The McNemar test was used to determine if symptoms were reduced postoperatively within both treatment groups. Odds ratios (ORs) were calculated for postoperative symptoms. These were then adjusted (aOR, adjusted OR) by controlling for substantial differences between the cohorts in baseline characteristics, presence of symptoms before surgery, previous incontinence surgery, and diverticulectomy history using logistic regression. Similarly, aORs were calculated for perioperative and postoperative complications as well as diverticulum recurrence.
Results
From January 1, 2000 to December 31, 2016, a total of 485 subjects met inclusion criteria with 389 (80%) who underwent diverticulectomy alone and 96 (19.7%) who had concomitant PVS. We excluded 4 cases who underwent different SUI procedures (1 mesh sling, 1 Burch colposuspension, and 2 urethral bulkings) and 15 cases of repeat diverticulectomies that were performed during the follow-up period ( Figure 1 ). The majority (83%) of the PVS cases were performed at 2 institutions, Vanderbilt University and the Mayo Clinic. At baseline, the women in the PVS group were older (53.3 years vs 46.1 years; P <.001), more likely to be postmenopausal (54% vs 32%; P <.001), white (77% vs 58%; P =.001), and have a history of previous urethral surgery (37.5% vs 7.7%; P <.001) ( Table 1 ). For previous urethral surgery, women in the PVS group had significantly higher rates of injury with repair (2.1% vs 0%; P =.050), urethral fistula repair (4.2% vs 0%; P =.001), and a previous UD (31% vs 7.7%; P <.001). There were no differences in number of vaginal deliveries, tobacco use, previous sexually transmitted infections, and medical comorbidities such as diabetes and autoimmune disorders as well as chronic steroid use. There were also no differences in any prior SUI surgery including synthetic midurethral slings, PVS, Burch colposuspension, and urethral bulking. There were no differences in concomitant Martius flap procedure performed at the time of index surgery (1.8% vs 3.1%; P =.67).
| Variable | Total | No PVS (n=389) | UD+PVS (n=96) | P value |
|---|---|---|---|---|
| Institution hospital, n (%) | <.001 | |||
| Cooper Health | 9 (1.9) | 9 (2.3) | 0 (0) | |
| Emory | 38 (7.8) | 36 (9.3) | 2 (2.1) | |
| Massachusetts General Hospital | 27 (5.6) | 27 (6.9) | 0 (0) | |
| Mayo | 126 (26) | 84 (21.6) | 42 (43.8) | |
| MedStar Washington Hospital Center | 39 (8) | 34 (8.7) | 5 (5.2) | |
| New York University | 49 (10.1) | 42 (10.8) | 7 (7.3) | |
| Ohio State | 14 (2.9) | 13 (3.3) | 1 (1) | |
| University of South Florida | 34 (7) | 34 (8.7) | 0 (0) | |
| University of Pennsylvania Medical Center | 55 (11.3) | 55 (14.1) | 0 (0) | |
| University of Texas Southwestern | 16 (3.3) | 16 (4.1) | 0 (0) | |
| Vanderbilt University Medical Center | 78 (16.1) | 39 (10) | 39 (40.6) | |
| Age, mean (SD) | 47.5 (14.26) | 46.08 (14.55) | 53.34 (11.36) | <.001 |
| BMI, mean (SD) | 29.32 (7.19) | 28.69 (7.11) | 28.73 (7.24) | .384 |
| Race, n (%) | ||||
| White | 298 (61.4) | 224 (57.6) | 74 (77.1) | .001 |
| Black | 136 (28) | 122 (31.4) | 14 (14.6) | .002 |
| Asian | 8 (1.6) | 6 (1.5) | 2 (2.1) | 1 |
| Hispanic | 2 (0.4) | 2 (0.5) | 0 (0) | 1 |
| Asian | 2 (0.4) | 0 (0) | 2 (2.1) | .05 |
| Unknown | 23 (4.7) | 22 (5.7) | 1 (1) | .102 |
| Parity, n (%) | .045 | |||
| 0 | 60 (13.5) | 52 (14.8) | 8 (8.8) | |
| 1 | 82 (18.5) | 71 (20.2) | 11 (12.1) | |
| 2 | 164 (37) | 129 (36.6) | 35 (38.5) | |
| 3+ | 137 (30.9) | 100 (28.4) | 37 (40.7) | |
| Vaginal delivery, n (%) (missing=50) | .195 | |||
| 0 | 111 (25.5) | 91 (26.5) | 20 (22) | |
| 1 | 81 (18.6) | 69 (20.1) | 12 (13.2) | |
| 2 | 130 (29.9) | 101 (29.4) | 29 (31.9) | |
| 3+ | 113 (26) | 83 (24.1) | 30 (33) | |
| Tobacco use, n (%) (missing=33) | .444 | |||
| N | 321 (66.2) | 258 (66.3) | 63 (65.6) | |
| Y | 125 (25.8) | 97 (24.9) | 28 (29.2) | |
| Postmenopausal | <.001 | |||
| N | 268 (55.5) | 234 (60.5) | 34 (35.4) | |
| Y | 175 (36.2) | 123 (31.8) | 52 (54.2) | |
| Prior sexually transmitted infection | .073 | |||
| Y | 65 (13.4) | 58 (14.9) | 7 (7.3) | |
| N | 420 (86.6) | 331 (85.1) | 89 (92.7) | |
| Past medical history, n (%) | ||||
| Diabetes mellitus | 46 (9.5) | 36 (9.3) | 10 (10.4) | .645 |
| HIV | 5 (1) | 5 (1.3) | 0 (0) | .581 |
| Chronic steroid use | 8 (1.6) | 6 (1.5) | 2 (2.1) | 1 |
| Autoimmune disease | 10 (2.1) | 9 (2.3) | 1 (1) | .701 |
| Prior SUI surgery, n (%) | ||||
| Midurethral sling | 20 (4.1) | 14 (3.6) | 6 (6.2) | .377 |
| Burch | 13 (2.6) | 10 (2.6) | 3 (3.1) | 1 |
| PVS | 11 (2.2) | 8 (2.1) | 3 (3.1) | .805 |
| Urethral bulking | 7 (1.4) | 6 (1.5) | 1 (1.0) | 1 |
| Any prior SUI surgery | 49 (10.1) | 37 (9.5) | 12 (12.5) | .495 |
| Prior urethral surgery, n (%) | ||||
| Urethral injury and repair | 2 (0.4) | 0 (0) | 2 (2.1) | .050 |
| Diverticulectomy | 60 (12.3) | 30 (7.7) | 30 (31.2) | <.001 |
| Urethral fistula repair | 4 (0.8) | 0 (0) | 4 (4.2) | .001 |
| Any prior urethral surgery | 66 (13.6) | 30 (7.7) | 36 (37.5) | <.001 |
| Concomitant procedures, n (%) | ||||
| Martius Flap | 10 (2.1) | 7 (1.8) | 3 (3.1) | .676 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
