Hypothesis / aims of study
Retropubic mid-urethral slings are associated with a 3% risk of chronic pelvic or groin pain. For women requesting complete excision of mesh an open or laparoscopic approach is needed to dissect out the retropubic arms. While small series have reported good short term success rates, it is unclear which patients are most likely to benefit from mesh removal, and the risks of laparoscopic total mesh excision surgery, including recurrent stress incontinence, are not well described. From our case series of 56 women, we aimed to report technical feasibility and operative outcomes for complete and partial laparoscopic excision of mid-urethral slings, and evaluate patient reported functional outcomes from surgery.
Study design, materials and methods
We identified all patients who underwent laparoscopic removal of mid-urethral sling mesh at two local hospitals between 2011 and 2016. Erosions of the lower urinary tract were excluded. We extracted data from patient records and theatre logs, and contacted all patients with a postal questionnaire, incorporating rating scales for pain, symptom severity and satisfaction. We used multivariate logistic regression to test for predictors of pain resolution and recurrence of stress incontinence.
Over the study period, 56 patients had a laparoscopic removal of a retropubic sling for chronic pain. The mean age was 48.5 years (range 30 - 71 years). The mean BMI was 28.4 (range 18 – 40). Nine women had undergone a prior attempt to remove the suburethral mesh using a vaginal approach. The median time from insertion to laparoscopic excision was 44 months (range 3 – 192 months). Most frequently reported site of pain was vaginal (n = 30), abdominal (n=28) and groin pain (n=22), but the majority of women reported pain at multiple sites (n=42). 22 women had additional non-pain symptoms they attributed to the mesh. Forty six patients (82%) had the retropubic mesh excised completely via a combined laparoscopic and vaginal procedure, and 10 (18%) underwent laparoscopic removal of the arms of the mesh with suburethral preservation for continence purposes. All were completed laparoscopically as intended. The cases took a median of 85 minutes (range 49-150 minutes). There was one return to theatre in the first 24 hours to evacuate a retropubic haematoma, but no bowel, bladder or ureteric injury occurred. The median inpatient stay was 2 days (range 1-7). Of the 46% (n=26) of patients who returned the questionnaire, 88% said they would recommend the procedure. There was a median 6 point decrease in pain scores (10 point numerical rating scale, p<0.0001). 45% experienced worse subjective stress urinary incontinence. In logistic regression looking for improvement or resolution of pain at follow up, there was no impact of age (OR 0.99/year 95%CI 0.91-1.09), length of time sling in situ (OR 1.02/month 95%CI 0.99-1.04), or prior attempt at surgical removal (OR 2.70 95%CI 0.29-25.8). In a separate logistic regression model, excision of the suburethral portion of the mesh was strongly associated with de novo or worsening of stress incontinence (OR 10.72 95%CI 1.10-104.00).
Interpretation of results
Although laparoscopic removal is feasible, there were significant early and late complications. These risks must be balanced against unpredictable efficacy; we identified no factors associated with resolution of pain, although most patients did report cure or improvements in pain. De novo or worsening stress incontinence was also common, and was strongly associated with removal of the suburethral portion of mesh.