Surgical Removal of Midurethral Sling in Women Undergoing Surgery for Presumed Mesh-Related Complications

A Systematic Review

Abstract

OBJECTIVE: 

To assess whether some, or all, of the mesh needs to be removed when a midurethral sling is removed for complications.

DATA SOURCES: 

A systematic review and meta-analysis was conducted. MEDLINE, Cochrane, and ClinicalTrials.gov databases from January 1, 1996, through May 1, 2021, were searched for articles that met the eligibility criteria with total, partial, or a combination of anti-incontinence mesh removal.

METHODS OF STUDY SELECTION: 

All study designs were included (N≥10), and a priori criteria were used for acceptance standards. Studies were extracted for demographics, operative outcomes, and adverse events. Meta-analysis was performed when possible.

TABULATION, INTEGRATION, AND RESULTS: 

We double-screened 11,887 abstracts; 45 eligible and unique studies were identified. Thirty-five were single-group studies that evaluated partial mesh removal, five were single-group studies that evaluated total mesh removal, and five were studies that compared partial mesh removal with total mesh removal. All of the studies were retrospective in nature; there were no randomized controlled studies. Comparative studies demonstrated that partial mesh removal had lower rates of postoperative stress urinary incontinence (SUI) than total mesh removal (odds ratio 0.46, 95% CI 0.22–0.96). Single-group studies supported lower rates of postoperative SUI with partial mesh removal compared with total mesh removal (19.2% [95% CI 13.5–25.7] vs 48.7% [95% CI 31.2–66.4]). Both methods were similar with respect to associated pain, bladder outlet obstruction, mesh erosion or exposure, and lower urinary tract symptoms. Adverse events were infrequent.

CONCLUSION: 

Postoperative SUI may be lower with partial mesh removal compared with total mesh removal. Other outcomes were similar regardless of the amount of mesh removed.

SYSTEMATIC REVIEW REGISTRATION: 

PROSPERO, CRD 42018093099.

In Brief

Postoperative stress urinary incontinence may be lower with partial mesh removal compared with total mesh removal; other outcomes are similar regardless of the amount of mesh removed.

Since the U.S. Food and Drug Administration’s approval of the use of mesh midurethral slings for the surgical treatment of female stress urinary incontinence (SUI) in 1996, these devices have gained worldwide popularity owing to the simplicity of use, equal or superior continence outcomes to more invasive anti-incontinence procedures,¹ and decreased recovery time. Synthetic midurethral sling procedures are the most commonly performed surgery for SUI, with more than 200,000 slings placed every year in the United States. Reported success for relief of incontinence ranges between 70% and 91%. Synthetic midurethral slings are considered to be the procedure of choice for treatment of SUI.²

Midurethral slings for SUI have an overall complication rate of less than 5%,³ with readmission and reoperation rates of 0.6–0.8%.⁴ Complications can present in the short- or long-term and include mesh exposure, urinary retention, persistent postoperative pain, and persistent or worsening SUI. The management of mesh complications depends on patient symptomatology and ranges from expectant management to surgical interventions. Surgical management differs based on the indication, approach (abdominal vs vaginal), timing, and amount of mesh removed (partial mesh removal vs total mesh removal).

The Society of Gynecologic Surgeons (SGS) Systematic Review Group performed a systematic review and formulated evidence-based clinical practice guidelines that address the risks and benefits of partial mesh removal compared with total mesh removal to address complications of previously placed anti-incontinence midurethral slings. These guidelines considered the quality and scope of evidence in the current literature, focusing on preservation of stress-related continence and minimizing adverse events.

SOURCES

The SGS Systematic Review Group includes practicing gynecologists and surgeons who specialize in female pelvic medicine and reconstructive surgery, in addition to a methodology expert in systematic review. MEDLINE, Cochrane, and clinicaltrials.gov databases were searched through May 1, 2021, using terms for anti-incontinence mesh surgery and revisions. The heterogeneity of mesh removal terms in addition to American Urogynecologic Society's (AUGS) recommended language⁵ to describe midurethral mesh removal were searched.

STUDY SELECTION

Based on an a priori criteria, we included studies of women with previous mesh sling for SUI who were undergoing surgical mesh removal for any reason. We excluded studies that had concomitant vaginal mesh removal for pelvic organ prolapse. We did not exclude studies that preformed a colporrhaphy or apical suspension in fewer than 25% of the patients at the time of mesh removal. Our intervention of interest was total mesh sling removal, which was defined as the removal of the entire vaginal portion of the sling, with or without the removal of the retropubic or obturator arms, through vaginal, abdominal, or both surgical approaches. Comparators included partial midurethral mesh excision, transection of the sling, or early mobilization through vaginal or abdominal approach. No studies included urethral dilations. We included studies of surgeries conducted for any presumed mesh-related complications. Specific outcomes were identified during data extraction and were not predefined or limited. To be included as relevant, outcomes had to relate to mesh-related symptoms, lower urinary tract symptoms, or surgical adverse events.

We included randomized controlled trials (RCTs) and nonrandomized studies that compared surgeries and single-group studies. Studies had to report both preoperative and postoperative outcomes (except operative outcomes and adverse events) for each surgical technique and with a minimum of 10 patients in each group. Studies could be published in any language and any format.

Abstracts were screened independently in duplicate by 12 researchers. Discrepancies were resolved by the lead author. Abstract screening was conducted in Abstrackr software (http://abstrackr.cebm.brown.edu), which incorporates machine learning to facilitate the abstract review process. The full texts of all accepted abstracts were screened in duplicate for eligibility, with discrepancies resolved by a third team member. From each study, a single researcher extracted data on study and participant characteristics, surgery and outcome descriptions, results, and risk of bias. At least one other researcher reviewed extracted data for accuracy.

Each study was assessed based on the Cochrane Risk of Bias tool (designed for RCTs)⁶; selected questions from the ROBINS-I tool (designed for nonrandomized comparative studies) addressing patient selection bias, confounding bias, and use of intention-to-treat principals⁷; and questions about reporting adequacy. Each study was then classified as being of overall methodologic quality good (A), fair (B), or poor (C). As appropriate, specific outcomes within studies were further downgraded.

Outcomes were categorized as: 1) lower urinary tract symptoms that include urinary urgency, frequency, urgency urinary incontinence, and subjective irritative voiding symptoms; 2) mesh erosion or exposure, such as symptomatic and asymptomatic mesh in the vagina, bladder, or urethra; 3) pain, defined as dyspareunia, penile pain during vaginal intercourse, groin pain, suprapubic pain, or pelvic pain; 4) bladder outlet obstruction, defined as an elevated postvoid residual measurement, urodynamics, or intermittent self-catheterization; 5) SUI, including objective and subjective reporting of urine leakage with activity, urodynamics, and questionnaires; and 6) adverse events.

For each outcome, a summary table was assembled to summarize and evaluate the strength of the evidence comparing total mesh removal and partial mesh removal. Concordant with GRADE (Grading of Recommendations Assessment, Development and Evaluation) methodology,⁸ we evaluated study quality (risk of bias), consistency across studies, directness of the evidence, sufficiency of evidence, and related features. Each outcome was graded to be of high, moderate, low, or very low strength of evidence.

Guideline statements were developed and given a level of strength based on the strength of the supporting evidence. Strong recommendations were made if the benefits (primarily clinical outcomes) outweighed the harms (adverse events, costs), with at least a moderate strength of evidence level. Weak recommendations were made based on strength of evidence and an uncertainty that benefits outweigh harms.

This review and associated clinical recommendations were presented at the SGS annual scientific meeting on July 9, 2020. On completion of the manuscript, comments were petitioned for 2 weeks, and the manuscript was approved by the membership.

RESULTS

Our database searches yielded 11,887 citations, of which 205 articles were retrieved in full text. Forty-eight eligible studies were found; however, four originated from the same database, hence the largest and most current article was used. Of the 45 eligible studies, 35 were single-group studies that evaluated partial mesh removal; five single-group studies that evaluated total mesh removal, and five studies that compared total mesh removal compared with partial mesh removal (Fig. 1). We found no eligible RCTs. None of the five nonrandomized comparative studies attempted to control for preoperative difference between groups or potential confounders. All studies were of retrospective design, with two retrospectively reviewing prospectively collected data. Time from initial surgery to mesh removal surgery was a median of 9 (range 2–153) months across all studies.

Fig. 1.: Results of study selection and characteristics. RCT, randomized controlled trial.Doyle. Removing Midurethral Sling for Mesh-Related Complications. Obstet Gynecol 2022.

Among the five^9–13 nonrandomized comparative studies that evaluated partial and total mesh removal, three^11–13 evaluated SUI rates before and after surgery, two^9,12 evaluated the effect of surgery on pain, and two^10,11 evaluated multiple symptoms before and after surgery, including vaginal and urethral mesh erosion, urinary tract infections, pain, and lower urinary tract symptoms. Two studies compared different types or timing of partial sling removal: One article¹⁴ evaluated the difference between sling mobilization and transection, and another¹⁵ looked at early mesh transection (less than 1 year since the index surgery) compared with late transection (more than 1 year since the index surgery).

Of the 35 single-group studies that report partial mesh removal surgeries,^14–41 nine assessed the outcomes of sling transection^{15–17,19,20,24,28,31,32} and four studied mesh mobilization.^27,29,39,42 The five total mesh removal studies^3,43–46 looked at a combination of retropubic and transobturator slings, with the exception of Coskun,⁴² who report only on mini-slings.

Across studies, the reasons for mesh removal included lower urinary tract symptoms, mesh erosion or extrusion, pain, and bladder outlet obstruction. The overall quality of the studies was fair to poor. No studies were classified as good quality, because they were lacking in a prospective study design or control for potential confounders. A summary of the main study outcome measures, levels of evidence, number of patients evaluated for each study, and whether partial or total mesh removal was performed is presented in Appendix 1, available online at https://links.lww.com/AOG/C531.

Four^7–10 fair-to-poor–quality studies with a total of 486 women compared total mesh removal and partial mesh removal and report on lower urinary tract symptom-related outcomes. Reported outcome measures for lower urinary tract symptoms in these four studies included an incontinence severity index questionnaire, urodynamic or uroflowmetry results, voiding diary, or subjective documentation on chart review. One study saw no association when comparing preoperative and postoperative lower urinary tract symptoms, two demonstrated an association in symptom improvement postoperatively but no difference between groups, and one demonstrated an association favoring partial mesh removal for improvement in lower urinary tract symptoms.

More specifically, Misrai et al¹¹ report no symptom change in either group, with 13.3% (10/75) experiencing lower urinary tract symptoms postoperatively compared with 12.0% (9/75) preoperatively. Although they report an equal number of lower urinary tract symptoms preoperatively and postoperatively, they do not identify whether this is de novo or persistent urgency. In contrast, Drain¹² et al report that new onset lower urinary tract symptoms developed more frequently in total mesh removal patients (21.4%) when compared with partial mesh removal patients (7.3%, P=.039). Jambusaria et al¹³ report an associated improvement in lower urinary tract symptoms in both groups, with 54.3% (82/151) having lower urinary tract symptoms preoperatively and 24.7% (37/150) having lower urinary tract symptoms postoperatively, but no difference between groups, with 29% (7/24) in the partial mesh removal group and 38% (12/32) in the total mesh removal group (P=.52). Even though their article reports a relationship between preoperative lower urinary tract symptom improvement with either mesh removal techniques, there was an increase in urgency after total mesh removal in 80% (44/55) of patients compared with 20% (11/55) in the partial mesh removal group. They hypothesized that this was the result of a more extensive periurethral dissection. This outcome was consistent at a mean short-term follow-up of 5.9 weeks and a mean long-term follow-up of 29.1 weeks.

Goodall et al⁹ report associated symptom improvement in both groups, with 55% of patients (31/56) reporting preoperative urgency or subjective voiding difficulties, compared with 21% of patients (12/56) postoperatively.

For mesh transection for lower urinary tract symptoms, two^15,21 poor-quality studies looked at the timing of transection. South et al¹⁵ associate an overall improvement in lower urinary tract symptoms in patients who had mesh transected less than 1 year (early lysis group) and more than 1 year (late lysis group) after the index surgery. They report a greater associated improvement in the early lysis group compared with the late lysis group (91% vs 71%; P=.01). Agnew et al²¹ also show an overall associated improvement in lower urinary tract symptoms with mesh transection but did not demonstrate an association between early or late mesh transection, with 6 of 45 (13%) of participants in the less-than-1-year revision group and 2 of 18 (11%) of those in the more-than-1-year revision group experiencing persistent voiding dysfunction.

Three of the comparative studies associated symptom improvement in preoperative lower urinary tract symptoms after mesh removal in both the total and partial mesh removal subgroups. A meta-analysis was not performed owing to the heterogeneity of outcome measures.

Overall, 23^{14–23,25–27,30–32,36,38–41,47,48} single-group studies that evaluated partial mesh removal, four^3,44–46 single-group studies that looked at total mesh removal, and four comparative studies support an association between improvement in lower urinary tract symptoms after surgery, regardless of mesh removal amount.

A total of 15 single-cohort studies of fair and poor quality evaluated mesh erosion preoperatively and postoperatively; 11^{17–19,22,33–38,49} report on partial mesh removal, and four^3,43,45,50 report on total mesh removal. The single-group studies all implied resolution of mesh erosion postoperatively, but very few considered this a primary outcome, and postoperative statistics were lacking. Two studies,^7,9 with a total of 320 patients, compared total and partial mesh removal on patients with mesh erosion or extrusion. However, neither study reports postoperative outcomes related to mesh erosion. Forde and Creagh³³ suggest that clinicians are poor at diagnosing erosive complications after midurethral sling placement. With a lack of comparative studies or homogenous outcomes, no meta-analysis was performed.

Overall, 11 single-group studies that evaluated partial mesh removal, four single-group studies that looked at total mesh removal, and two comparative studies suggest an association with improvement in mesh erosion or exposure after surgery regardless of mesh removal amount.

Four^9,11–13 comparative studies of poor to fair quality evaluated a total of 486 patients. Outcome measures for pain included visual analog scale, chart review, and Douler Neuropathique 4 questionnaire. The studies all found no difference in pain resolution between total and partial mesh removal. Jambusaria et al¹³ described resolution of pain in 72% of patients undergoing partial mesh removal and 76% of complete sling removal for the indication of pain (P=.67). Goodall et al⁹ report a cohort of 56 women undergoing midurethral mesh removal for pain presumed to be caused by the sling. Of women who completed the questionnaire (visual analog scale), there was an associated median 6-point decrease in pain scores postoperatively compared with preoperative findings (P<.001) independent of partial or total mesh removal techniques. Drain et al¹² associate a 72.7% improvement in postoperative pain with total mesh removal and an 89.7% associated improvement with partial mesh excision (P=.15); however, women with pain and mesh erosion were categorized together, possibly distorting evaluation of pain. Similarly, Misrai et al¹¹ report an associated total decrease in pain in all patients. Although four comparative studies looked at mesh removal for pelvic pain, we were unable to perform meta-analysis of these data because different specific outcomes are reported across studies and outcomes are either reported as a combination of symptoms, such as pain and erosion together,⁸ or the outcomes for pain and mesh erosion do not differentiate between partial removal and total mesh removal. Nevertheless, overall, there was no difference found between partial or total mesh removal in regard to improvement in pain outcomes in all four studies. Nine^{16–18,38,40,47,51–53} single-group studies that evaluated partial removal all demonstrated an associated improvement in pain outcomes; these outcome measures were varied and included the FSFI (Female Sexual Function Index), the UDI-6 (Urogenital Distress Index), subjective reports, and a numeric pain scale. Two^45,46 single-group studies looking at total mesh removal also report an associated improvement in pain outcome measures.

Overall, nine single-group studies that evaluated partial mesh removal, two single-group studies that looked at total mesh removal, and four comparative studies suggest an associated improvement in pain after surgery regardless of mesh removal amount.

Two^11,12 studies that evaluated 185 patients compared partial mesh removal and total mesh removal and report on bladder outlet obstruction outcomes from presumed midurethral mesh obstruction. Outcome measures include urodynamics and elevated postvoid residual measurements. Drain et al¹² report that 76.5% of patients who underwent total mesh removal and 85.7% who underwent partial mesh removal as improved or cured of obstruction (P=.51). Misrai et al¹¹ did not report a difference in obstructive voiding outcomes between surgeries but noted that bladder outlet obstruction was associated with improved outcomes in 82% of women overall (with either surgery). One²⁹ single-group study of fair quality that evaluated mesh mobilization within the first 2 weeks after midurethral sling placement for outlet obstruction found 87.9% (29/33) resolution. Of the four patients with persistent bladder outlet obstruction, mesh sling transection was performed 4–8 months later with 100% resolution. Comparative and single-group studies^{3,14,17,19,20,23,26–32,39,41,43–45,52,54} demonstrated no difference in bladder outlet obstruction outcomes between total and partial mesh excision; both techniques showed an associated improvement in bladder outlet obstruction symptoms.

Overall, 16 single-group studies that evaluated partial mesh removal, four single-group studies looking at total mesh removal, and two comparative studies suggest an associated improvement in bladder outlet obstruction after surgery regardless of mesh amount removed.

Three^11–13 studies with a total of 430 women compared total and partial mesh removal and report on postoperative SUI. Outcome measures included urodynamic testing, voiding diaries, chart review for subsequent interventions for SUI, or documentation of patient-reported SUI. All three studies report a higher rate of postoperative SUI with total mesh removal when compared with partial mesh removal. For example, Jambusaria et al¹³ report a statistically significant difference between the two groups with 21% of patients in the partial sling removal group developing postoperative SUI compared with 69% in the total mesh removal group (P<.001). Drain et al¹² report postoperative SUI in 21.6% after partial mesh removal and 25.0% after total mesh removal. Misrai et al¹¹ report postoperative SUI in 45% (18/40) undergoing partial mesh removal and in 60% (21/35) undergoing total mesh removal. The three studies^11–13 were sufficiently similar to allow meta-analysis. The risk of postoperative SUI was lower after partial mesh removal than total mesh removal (odds ratio 0.46, 95% CI 0.22–0.96, P=.037, I²=0.23; Fig. 2).

Fig. 2.: Meta-analysis of postoperative stress urinary incontinence (SUI) in three studies that compare total mesh removal with partial mesh removal. The risk of postoperative SUI was lower after partial mesh removal than total mesh removal.Doyle. Removing Midurethral Sling for Mesh-Related Complications. Obstet Gynecol 2022.

When we further evaluated postoperative SUI in the partial mesh removal studies, we found four^27,29,39,42 studies that specifically looked at sling mobilization. Three of the studies^27,29,39 described surgical techniques as being a vaginal “pull down” method to loosen the sling. One study,⁴² described it as “tape loosening”. Postoperative SUI was 3.7% with this technique, and the time from initial sling placement to intervention was 9.3 days. The remaining study²⁷ described a vaginal and abdominal sling urethrolysis technique that mobilized the urethra off of the sling and created mobility at the proximal urethra. In this study, postoperative SUI was 30%, and the time from sling placement to intervention was 38 months.

Regarding mesh transection, one study¹⁵ looked at postoperative SUI after early transection (less than 1 year) and late transection (more than 1 year) and found that early transection of the sling resulted in 9.5% of patients having SUI and late transection resulted in 7.9% of patients having SUI, although the difference was not statistically significant. Conversely, another²¹ did not find a difference between early (13%) and late (11%) transection in their study.

Overall, 26 single-group studies that evaluated partial mesh removal, four single-group studies that looked at total mesh removal (Table 1), and three comparative studies suggest low rates of SUI after surgery (19.2% partial vs 48.7% total) with partial mesh having significantly less SUI (odds ratio 0.46, 95% CI 0.22–0.96). It should be noted that a potential difference in participant baseline characteristics and follow-up times were not accounted.

Table 1.: Postoperative Stress Urinary Incontinence Rates After Partial and Total Anti-incontinence Mesh Excision

Of the 45 studies included in this systematic review, 33 (25 single-group, partial removal; five single-group, total removal; three comparative) commented on adverse events over 2,721 surgeries. Among these 33 studies, 18 noted no adverse events in 1,805 surgeries. None of the comparative studies report a statistically significant difference in complications based on surgical method.

Among eight studies (n=688) that report adverse events with total mesh removal, there were reports of one case of urinary retention, one urethral obstruction, one urethral stricture, one infected retropubic tape, one vaginal hematoma, one cystotomy, four retropubic hematomas, one groin hematoma, five patients who needed additional surgery, one gluteal abscess, five port site hernias, and one port site fibroma. This was a total of 23 adverse events out of 688 surgeries, or 3.34%. It should be noted that one study⁹ reports 26% of the complications related to abdominal retropubic mesh removal, and this could be skewing the total mesh removal safety profile.

Among 28 studies (n=2,033) that report adverse events with partial mesh removal, there were 32 adverse events out of 2,033 surgeries, or 1.57%. Adverse events after surgery for partial removal of midurethral mesh included 12 urinary tract infections, one seroma, three wound infections, two vaginal infections, one extended catheter placement, three fistulas, five hematomas, one blood transfusion, one urethrotomy, and three cystotomies.

DISCUSSION

The results of this systematic review confirm that both partial and total mesh removal are associated with improved symptoms for the majority of patients. We specifically evaluated lower urinary tract symptoms, mesh erosion or exposure, pain, bladder outlet obstruction, postoperative SUI and adverse events. Postoperative SUI was present with partial and total mesh removal; however, postoperative SUI is lower in partial mesh removal surgeries compared with total mesh removal. It is a reasonable plausibility that leaving part of the midurethral sling intact, a portion of the sling's functionality would also remain intact. However, it should be noted that of the three comparative studies that evaluated postoperative SUI, follow-up time varied and the study with the shortest follow-up time demonstrated the most significant difference.

Data for postoperative mesh erosion or extrusion outcomes were of very low strength evidence. The majority of studies inferred a resolution in mesh erosion or extrusion, but this was not statistically reported.

All studies demonstrated an associated improvement in pain outcomes independent of the amount of mesh removed. However, we were unable to break down the data into type of midurethral sling being removed (tension-free vaginal tape vs transobturator tape). Nevertheless, we believe it is reasonable to individualize surgical approach when removing mesh for pain related complications as there may be a difference related to the type of sling being removed. Fuentes et al⁵¹ suggest performing partial sling excision for pain followed by total mesh removal in the subset of patients whose pain does not improve with partial mesh removal.

Bladder outlet obstruction outcomes were associated with improvement irrespective of amount of mesh removal or even by mesh transection. Of note, one study¹⁵ looked at early transection (less than 1 year) and late transection (more than 1 year). It showed more postoperative SUI in the early lysis group compared with delayed transection. The presumed mechanism is more tissue ingrowth and scarring from a longer healing period, resulting in more midurethral support after sling transection.

Although adverse events were uncommon in both partial and total mesh removal, the types of complications were different. For example, with total mesh removal procedures, there was commonly a retropubic or transobturator dissection. This resulted in hematomas, abscesses, and port site complications. It should be noted that one study⁹ reports 26% of the complications related to abdominal retropubic mesh removal, and this could be skewing the total mesh removal safety profile. It is unclear whether this surgical approach has an increase in complications or if the researchers used a lower reporting threshold. Additional complications included fistulas and a blood transfusion. It would be anatomically logical for fistula formation to be a risk regardless of surgical approach, but none were reported with total mesh removal. Incidental cystotomies, which were repaired intraoperatively, were reported with both types of mesh removal surgery. Overall, the adverse event rate shows an acceptable safety profile for both types of sling excision.

The strengths of this article are in the inclusive study design and a research team composed of female pelvic medicine and reconstructive surgery specialists and a systematic review content expert. Rigorous systematic review processes were used. The review was vetted by a professional society of gynecologists, and the clinical practice guidelines were agreed on.

The limitations of the article include the overall quality of the studies—the majority were considered to be poor- to fair-quality. Given the overall design of retrospective single-cohort studies, there were data gaps on relevant outcomes. There were no prospective or randomized studies available. Despite fundamental differences between the women for whom total or partial mesh removal is performed, none of the studies accounted for these or adjusted for potential confounders. Furthermore, data are limited by heterogeneity of measuring outcomes, differing definitions of lower urinary tract symptoms, bladder outlet obstruction, or postoperative pain between studies.

The results of this study allow preoperative counseling in women who are contemplating partial or total mesh removal for complaints presumed to be related to a previously placed midurethral sling. This study found postoperative SUI is higher with total mesh removal than with partial mesh removal, and early mobilization for bladder outlet obstruction resulted in the lowest amount of postoperative SUI (3.7%). With the exception of postoperative SUI, all outcomes studied in this review were associated with improved outcomes after partial and total mesh removal of a midurethral sling. Only postoperative SUI is lower with partial mesh removal, compared with total mesh removal, after a midurethral sling. This supports a more conservative approach to mesh sling removal that use approaches such as early mobilization, transection, and targeted excision. The evidence did not support better outcomes with total mesh removal on pain, bladder outlet obstruction, lower urinary tract symptoms, or mesh erosion. These outcomes resolved equally well with total or partial mesh excision, and adverse outcomes were low for both mesh excision strategies.

In addition, the heterogeneity of mesh removal terms used in research publications made it difficult to analyze techniques and outcomes despite AUGS recommended language⁵ to describe midurethral mesh removal. As the evidence is of moderate to low quality, we hope for more rigorous and prospective studies to this clinical question in the coming years.

Based on our systematic review and meta-analyses, the SGS Systematic Review Group makes the following clinical practice guideline recommendations:

1. We suggest counseling patients preoperatively that the rate of postoperative SUI is lower with partial mesh removal compared with total mesh removal, with no difference in other outcomes or adverse events (weak recommendation).
2. We recommend that future research incorporate AUGS/International Urogynecologic Association⁵ terminology to describe midurethral mesh removal (strong recommendation based on preferred reporting standards).