Other than mesh-specific complications, there is little understanding of the incidence and type of adverse events that may follow prolapse surgery with vaginal mesh. The objective of this study is to determine whether 30-day postoperative morbidity is higher in women undergoing vaginal mesh placement compared to those who undergo native tissue repair using propensity score matching.

This was a retrospective cohort study using data from the American College of Surgeons National Surgical Quality Improvement Program database from 2010 to 2017.  Common Procedural Terminology (CPT) codes were used to identify women who underwent vaginal procedures for prolapse repair. Propensity scores for the likelihood of undergoing mesh related repair were calculated using the following variables: age, race, body mass index, smoking status, medical comorbidities, functional status, American Society of Anesthesiology (ASA) class, length of stay, operating time, hysterectomy, extra- or intraperitoneal apical suspension, and sling placement.  The propensity score was then used to match women with and without mesh implantation at a ratio of 1:1. 

The primary outcome was the composite complication rate. Descriptive statistics were reported as means with standard deviations. Pairwise analysis using Student’s t-test and Fisher’s exact test was performed where appropriate. Stepwise backward multivariable logistic regression was used to adjust for confounders to identify statistically significant factors associated with the likelihood of undergoing apical suspension.  Univariate logistic regression was performed and any variable that was associated with treatment failure and a p-value ≤ 0.2 was included in the multivariable regression analysis.

There were 1,962 vaginal mesh repairs matched to 1,962 native tissue repairs. There were no differences in preoperative characteristics or procedure type (Table 1).   Procedures using mesh were longer, 121.3 +/-60.1 minutes vs 115.5+/-55.8 minutes, p=0.001, and were less likely to involve adhesiolysis, 0.2% vs 0.8%, p=0.01. Postoperative transfusions were more common in mesh repairs 2.4% vs 1.0%, p=0.002; while readmissions were more common for the native tissue group, 3.1% vs 1.8%, p=.007. There were no other differences in specific complications (Table 2). There was no difference in the composite complication rates between mesh and native tissue repairs 10.9% vs 11.7%, respectively, p=0.76.   After multivariable logistic regression, mesh was not an independent predictor of complications. Dependent functional status (aOR=4.31, 95%CI: 1.96-13.58) was most predictive of any complication occurring. Additional statistically significant risk factors for postoperative complications included: chronic obstructive pulmonary disease (aOR=1.73, 95%CI 1.17-2.98), length of stay (aOR=1.21 per day, 95%CI: 1.13-1.29), Operating time (aOR=1.04 per minute, 95%CI: 1.03-1.05), and American Society of Anesthesiology class > 2 (aOR=1.32, 95%CI: 1.08-1.62).

Mesh-related complications associated with vaginal prolapse repair, and erosion in particular, have been well described in the literature [1]. However, the data on incidence and risk factors of non-mesh specific complications following vaginal repair is relatively sparse. Two previous Theofanides et al and Caveney et al previously evaluated perioperative morbidity between mesh and native tissue vaginal surgery; both finding that perioperative morbidity increased with mesh related surgery [2,3].  However, both studies failed to account for significant differences in procedure type and patient comorbidities both of which are significant confounders. 

In our study, we utilized data from a prospectively maintained national surgical database to perform a propensity scored matched analysis of mesh and native tissue pelvic reconstructive surgery complications. The patients were matched by clinical, demographic data and procedure type. This matching ameliorates some of the confounding found in the previous studies. 

In this matched cohort, there was no difference in the composite complication rate between groups, though patients undergoing mesh-related surgery were more likely to require a transfusion and patients undergoing native tissue repair were more likely to be readmitted. The former may be due to the need for more extensive dissection required for mesh placement. The results of the multivariable regression suggest, that a patient's state of health is a more significant determinant of perioperative complications than mesh utilization at the time of vaginal repair of pelvic organ prolapse.

This study suggests that in carefully selected patients, vaginal repair involving mesh is not associated with increased perioperative non-mesh specific morbidity compared to native tissue repair. Additional randomized prospective studies are needed to confirm this finding.

Keys T, Campeau L, Badlani G. Synthetic mesh in the surgical repair of pelvic organ prolapse: current status and future directions. Urology. 2012 Aug 1;80(2):237-43.
Caveney M, Haddad D, Matthews C, Badlani G, Mirzazadeh M. Short-term complications associated with the use of transvaginal mesh in pelvic floor reconstructive surgery: Results from a multi-institutional prospectively maintained dataset. Neurourology and urodynamics. 2017 Nov;36(8):2044-8.
Theofanides MC, Onyeji I, Matulay J, Sui W, James M, Chung DE. Safety of mesh for vaginal cystocele repair: analysis of national patient characteristics and complications. The Journal of urology. 2017 Sep;198(3):632-7.