Female Pelvic Medicine & Reconstructive Surgery:
1Obstetrics and Gynecology, Vanderbilt Medical Center, Nashville, TN; 2Urology, Vanderbilt Medical Center, Nashville, TN; 3Biostatistics, Vanderbilt Medical Center, Nashville, TN
DISCLOSURE OF RELEVANT FINANCIAL RELATIONSHIPS: None.
To assess the degree to which smoking, age, menopausal status, hormone replacement therapy, diabetes, body-mass index (BMI) and perioperative complications are associated with the risk of mesh erosion following transvaginal placement of polypropylene mesh for pelvic organ prolapse.
MATERIALS AND METHODS:
A matched case-control study was performed. Women with transvaginally placed polypropylene mesh for the repair of pelvic organ prolapse between January 1, 2004 and May 31, 2009 were included. Cases were all women treated at our tertiary care center with mesh erosion requiring surgical revision. Controls did not develop mesh erosion and were matched for similar date and type of surgery (location and brand of mesh). Cases were identified by CPT codes 57295 and 57296, indicating a surgical mesh revision from a vaginal or abdominal approach. Controls were identified by CPT code 57267, indicating the vaginal placement of mesh. Charts were manually reviewed to assess for all inclusion criteria. Patient demographics, characteristics, comorbidities, perioperative variables, complications and erosion status were ascertained from the medical records. Complications were those recorded by the surgeon, including: bleeding, transfusion, cystotomy, rectal injury, hematoma, sciatic pain, and urinary retention. A conditional logistic regression was used to control for matching; adjusted odds ratios (OR) with 95% confidence intervals (CI) were obtained from this model. The model included smoking status, menopausal status, hormone replacement therapy, diabetes status, BMI and age. A restricted cubic spline was used for BMI and age to relax assumptions of linearity in these variables. Analysis was performed with STATA version 9.
Forty-eight cases were matched with 48 controls. The mean age for cases was 54.5 years and controls was 59.9 years. Parity, body mass index and menopausal status were similar between both groups (Mean parity 3; BMI 28.1 and 27.2, 38% and 40% menopausal, respectively). Results were inconclusive regarding the factors of interest and mesh erosion: The adjusted odds of developing mesh erosion were 3.4 fold higher among smokers than non-smokers (95% CI, 0.34 to 33.66), 2.3 fold higher among patients taking hormonal replacement therapy (95% CI, 0.57 to 9.19), and lower among those who were younger, (OR 0.96; 95% CI, 0.92 to 0.99), and those with diabetes (OR 0.46; 95% CI, 0.07 to 2.86). Mesh erosions tended to be more common among patients with perioperative complications, (OR, 6.04; 95% CI, 0.66 to 55.53). Complications among the cases included bleeding requiring transfusion (n = 1), cystotomy (n = 2), hematoma (n = 2), rectal perforation (n = 1), sciatic pain (n = 1) and urinary retention (n = 1). Only one control had a complication: postoperative urinary retention which resolved.
Our results were inconclusive regarding whether smoking, age, menopausal status, hormone replacement therapy, diabetes and body-mass index are risk factors for the development of mesh erosion. There was a trend toward mesh erosion being more common after surgeries with perioperative complications.