Objective
To describe surgical complications in 597 women over a 24-month period after randomization to retropubic or transobturator midurethral slings.
Study Design
During the Trial of Midurethral Slings study, the Data Safety Monitoring Board regularly reviewed summary reports of all adverse events using the Dindo Surgical Complication Scale. Logistic regression models were created to explore associations between clinicodemographic factors and surgical complications.
Results
A total of 383 adverse events were observed among 253 of the 597 women (42%). Seventy-five adverse events (20%) were classified as serious (serious adverse events); occurring in 70 women. Intraoperative bladder perforation (15 events) occurred exclusively in the retropubic group. Neurologic adverse events were more common in the transobturator group than in retropubic (32 events vs 20 events, respectively). Twenty-three (4%) women experienced mesh complications, including delayed presentations, in both groups.
Conclusion
Adverse events vary by procedure, but are common after midurethral sling. Most events resolve without significant sequelae.
Retropubic and transobturator midurethral synthetic slings are now considered gold standard procedures for primary surgical treatment of stress urinary incontinence in women. The less invasive nature of the midurethral slings (MUS) has significantly reduced many forms of surgical morbidity; however, long-term follow-up had not been available to address clinical concerns about delayed complications including mesh-related complications over time. Although mesh-associated complications are common to all mesh midurethral slings, adverse events (AEs) specific to the retropubic midurethral sling include bladder perforation, postoperative voiding dysfunction, de novo urgency and urge incontinence, and rare complications such as bowel perforation, vascular injury, and neurologic injuries. Although the transobturator midurethral sling was designed to minimize bladder and bowel perforation, complications such as thigh pain and neurologic pain are known to occur.
The investigators of the Urinary Incontinence Treatment Network (UITN) recently reported 1 year outcomes of a randomized equivalence design trial (the Trial of Mid-Urethral Slings [TOMUS] study) comparing these 2 approaches. The 12-month objective success rates for the retropubic and transobturator procedures were relatively high at 80.8% and 77.7%, and met the prespecified criteria for equivalence. In that report, we also briefly summarized adverse events during the first postoperative year. These included bladder perforation and voiding dysfunction, which were uncommon and occurred only in the retropubic group; women in this group were also more likely than women in the transobturator group to have postoperative urinary tract infections. In contrast, the frequency of neurologic symptoms was higher in the transobturator-sling group.
This report details the 2-year AE experience of women enrolled in the TOMUS Study. The relationship between clinical and demographic factors measured at baseline with the occurrence was also examined.
Materials and Methods
To ensure standardization across the participating sites, uniform definitions of AEs and serious AEs (SAEs) were established before trial initiation. AEs were collected at each study visit or between visits when known to the research team. AEs were classified with a modified version of the Dindo classification system as described previously. A complications work group, comprised of 5 principal investigators, reviewed all AE reports for quality control purposes and accuracy. The Complications Work Group members were masked to site, surgeon, and randomization assignment, although for some of the complications, narrative descriptions may have revealed treatment assignment. The group met regularly to review each AE report. Before each meeting, 2 work group members independently reviewed each AE report, graded the event using the modified Dindo classification and assigned the body system that was affected. During the group meetings, if there was consensus of the 2 primary reviewers and no concerns expressed by other members, the AE coding was entered into the data system. In case of disagreement, the work group members discussed the case until consensus was reached. For internal consistency, an on-going record of prior consensus decisions was maintained to facilitate recollection of previous decision. Participants were asked to identify specific locations of pain, using anatomic pictures, and rate the intensity of the pain associated with the study surgery. Patient self-reports were collected daily for the first 2 weeks after surgery, then at each subsequent study visit. Patients who reported surgical pain at the 2-week follow-up visit were asked to complete a daily pain diary for an additional 2 weeks.
Neurologic symptoms were defined as new paresthesias or alterations in motor function that developed within the first 6 weeks after surgery. Participants were considered to have a neurologic complication related to TOMUS surgery if the patient responded affirmatively to either of the following questions; “do you have any numbness in your legs or pelvic area that has developed since surgery?” and “do you have any weakness in your legs or pelvic area that has developed since surgery?” Patients who responded affirmatively were asked to specify the location and magnitude of the symptoms. They were also asked about bother with response categories of “not at all, slightly, moderately or greatly bothersome.”
Reporting of urinary tract infections (UTIs) was based on the time from study surgery. Within the first 6 weeks postoperatively, bother from presumed (not culture-proven) and/or culture-proven UTIs was reported. In the interval between 6 weeks and 1 year postoperative, only recurrent UTIs were considered as AEs and were defined as ≥3 episodes of symptoms characteristic of UTI symptoms that resulted in antibiotic treatment, regardless of urine culture results.
Mesh-related AEs included erosion (defined as occurring after primary healing, into an organ or surrounding tissue) or exposure (defined as mesh visualized through a prior incision area with or without an inflammatory reaction). There was no time limit for reporting mesh-related AEs.
Descriptive statistics were generated for the following clinical and demographic factors: age, body mass index (BMI), self-reported pelvic surgical history, concomitant surgery, operative time, blood loss, prolapse stage, self-reported history of UTI, smoking, menopausal status/hormone replacement therapy use, and diabetes. Bivariate associations between these factors and presence/absence of any AE were explored; odds ratios were calculated from logistic regression models. We compared intraoperative vs postoperative complications by event type between treatment groups. P values were calculated using Fisher exact test. Logistic regression models were created to explore associations between clinical-demographic factors and surgical complications.
Results
Baseline clinical and demographic characteristics and primary results of the TOMUS Study have been previously described. Five hundred twenty-eight of the original 597 randomized participants completed the 24-month assessments or failed treatment at or before that visit (258 [86.6%] retropubic and 270 [90.3%] transobturator). Over a period of 24 months, 42% (253/597) of all study participants experienced at least 1 AE including 12% (70/597) that experienced at least 1 SAE. These 253 patients experienced a total of 383 AEs; 75 AEs (20%) were classified as SAEs ( Table 1 ). Most (77%) of the AEs had an onset date on or before 6-week postoperative visit. Participants were more likely to experience at least one AE ( Table 2 ) if they reported a prior UTI (odds ratio, 2.37; 95% confidence interval, 1.24–4.52; P = .01, prior continence surgery (odds ratio, 1.99; 95% confidence interval, 1.23–3.22; P = .01), experienced longer surgical times or increased blood loss. AEs had no effect on subjective or objective surgical success.
| No. of events/No. patients (% complication rate) | |||
|---|---|---|---|
| Variable | Retropubic | Transobturator | P value a |
| Total no. of patients | 298 | 299 | |
| SAEs | |||
| Bladder perforation | 15/15 (5.0%) | 0/0 | < .0001 |
| Urethral perforation | 1/1 (0.3%) | 0/0 | .50 |
| Pulmonary embolus | 0/0 | 1/1 (0.3%) | > .99 |
| Postoperative bleeding | 1/1 (0.3%) | 0/0 | .50 |
| Mesh complication: erosion | 1/1 (0.3%) | 1/1 (0.3%) | > .99 |
| Mesh complication: exposure | 10/9 (3.0%) | 6/6 (2.0%) | .45 |
| Surg site inf: deep incisional | 0/0 | 1/1 (0.3%) | > .99 |
| Surg site inf: organ/space | 0/0 | 2/2 (0.7%) | .50 |
| Recurrent UTI | 3/3 (1.0%) | 0/0 | .12 |
| Neurologic symptoms | 1/1 (0.3%) | 0/0 | .50 |
| Granulation tissue | 0/0 | 1/1 (0.3%) | > .99 |
| Vaginal epithelium perforation | 6/6 (2.0%) | 13/13 (4.4%) | .16 |
| Voiding dysfunction requiring surgery (and/or catheter use) | 9/9 (3.0%) | 0/0 | .002 |
| Other | 0/0 | 3/3 (1.0%) | .25 |
| Total SAEs | 47/45 (15.1%) | 28/25 (8.4%) | .01 |
| AEs | |||
| Intraoperative bleeding | 14/14 (4.7%) | 7/7 (2.3%) | .13 |
| Postoperative bleeding | 6/6 (2.0%) | 0/0 | .02 |
| Mesh complication: exposure | 4/4 (1.3%) | 2/2 (0.7%) | .45 |
| Surg site inf: superficial incisional | 2/2 (0.7%) | 0/0 | .25 |
| UTI culture proven | 27/25 (8.4%) | 16/14 (4.7%) | .07 |
| Empiric | 16/15 (5.0%) | 9/9 (3.0%) | .22 |
| Recurrent | 18/16 (5.4%) | 10/10 (3.4%) | .24 |
| Neurologic symptoms | 20/15 (5.0%) | 32/29 (9.7%) | .04 |
| Voiding dysfunction | 10/10 (3.4%) | 6/6 (2.0%) | .33 |
| Pain per patient self-report ≥6 wk | 7/7 (2.3%) | 7/6 (2.0%) | .79 |
| De novo urge incontinence | 0/0 | 1/1 (0.3%) | > .99 |
| Persistent urge incontinence | 42/42 (14.1%) | 38/38 (12.8%) | .63 |
| Other | 8/7 (2.3%) | 6/6 (2.0%) | .79 |
| Total AEs | 174/121 (40.6%) | 134/98 (32.8%) | .051 |
| Factors | OR (95% CI) a | P value a | P value b |
|---|---|---|---|
| Treatment group | .003 c | ||
| Retropubic vs transobturator | 1.63 (1.18–2.26) | ||
| Age (per 10 y interval) | 1.01 (0.87–1.17) | .88 | .83 |
| BMI (per 5 units) | 1.12 (0.99–1.26) | .08 | .10 |
| Diabetes | .87 | .73 | |
| Yes vs no (ref) | 1.05 (0.55–2.03) | ||
| Prior UI surgery | .005 c | .004 c | |
| Yes vs no (ref) | 1.99 (1.23–3.22) | ||
| Prior POP surgery | .051 | .04 c | |
| Yes vs no (ref) | 0.37 (0.13–1.01) | ||
| Previous hysterectomy surgery | .07 | .052 | |
| Yes vs no (ref) | 0.71 (0.49–1.03) | ||
| Concomitant surgery | .25 | .23 | |
| Yes vs no (ref) | 1.24 (0.86–1.80) | ||
| Operative time (per 30 min interval) | 1.23 (1.09–1.37) | < .001 c | < .001 c |
| Blood loss -Entire case (per 50 mL) | 1.24 (1.11–1.39) | < .001 c | < .001 c |
| -Mid-urethral sling (per 50 mL) | 1.82 (1.40–2.37) | < .0001 c | < .0001 c |
| POP-Q stage | .87 | .87 | |
| Stage 0/1 vs Stage 3/4 | 0.87 (0.47–1.61) | ||
| Stage 2 vs Stage 3/4 | 0.85 (0.46–1.57) | ||
| History of UTI | .009 c | .005 c | |
| Yes vs no (ref) | 2.37 (1.24–4.52) | ||
| Smoking | .97 | .93 | |
| Never smoker vs current smoker (ref) | 1.05 (0.64–1.72) | ||
| Former smoker vs current smoker (ref) | 1.07 (0.63–1.81) | ||
| Menopausal status/HRT | .56 | .57 | |
| No vs premenopausal (ref) | 1.21 (0.82–1.79) | ||
| Yes vs premenopausal (ref) | 1.23 (0.80–1.88) |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
