Hypothesis / aims of study
While suburethral sling placement with synthetic mesh has been a preferred surgical treatment for stress urinary incontinence for the past two decades, recent concerns regarding implantable mesh have led to increased adoption of non-mesh anti-incontinence procedures. Studies on perioperative outcomes following suburethral sling placement have been largely limited to non-comparative studies for individual sling types or small comparative studies between different sling types, mostly at single institutions [1, 2].
Our primary objective was to identify the incidence of 30-day perioperative adverse events following sling placement with synthetic mesh, autologous rectus fascia, and autologous fascia lata using a national surgical quality improvement database and to compare independent risk of adverse events by sling type. Secondary objectives were to describe the most common adverse events and independent factors associated with adverse events.
Study design, materials and methods
We conducted a retrospective cohort study on patients who underwent fascia or synthetic sling placement for stress urinary incontinence between 2008 and 2021 using the American College of Surgeons’ National Surgical Quality Improvement Program (ASC-NSQIP) database. The ACS-NSQIP database is a national, validated, prospective quality improvement registry that contains patient demographic and clinical characteristics, intraoperative events, and 30-day surgical outcomes data from more than 700 hospitals across the United States.
We identified patients who underwent suburethral sling placement using Current Procedural Terminology (CPT) code 57288, which includes all sling operations for stress incontinence using either fascia or synthetic material. To differentiate between patients who received autologous fascia from those who received synthetic mesh, we used CPT codes for rectus fascia harvest (CPT 15769 and 20926) and fascia lata harvest (CPT 20920 and 20922), in line with the American Urogynecologic Society coding guidelines. Patients who underwent sling placement without a concomitant fascial harvest procedure were classified as having received synthetic mesh slings. To eliminate any confounding effects of additional procedures performed at the time of sling placement, we excluded patients who underwent any concomitant procedure aside from fascia harvest or cystoscopy (CPT 52000). Patient demographic and clinical characteristics were collected.
The primary outcome was the composite rate of any adverse event within 30 days of the surgery. Adverse events included superficial surgical site infection (SSI), deep incisional SSI, organ/space SSI, urinary tract infection, blood transfusion, sepsis or septic shock, pneumonia, renal failure, myocardial infarction, pulmonary embolism, deep vein thrombosis, cerebral vascular accident or stroke, or unplanned reoperation.
We conducted two analyses: a comparison of patients who received synthetic versus autologous fascial slings, and a comparison of patients who received autologous fascial sling harvested from rectus versus fascia lata sites. Demographic and clinical characteristics between patient cohorts were compared using the Wilcoxon rank-sum, Chi-squared, and Fisher’s exact tests, as appropriate. Multivariable regression analysis with stepwise regression was used to assess for odds of composite adverse events between cohorts while controlling for possible confounders. All covariates with a p<0.1 were retained in the final model. Given that urinary tract infection is a common adverse event following sling placement, we also conducted a subanalysis of composite complications without the inclusion of urinary tract infections.
Results
Of the 41,533 patients who underwent suburethral sling placement, 41,292 (99.4%) received a synthetic mesh sling and 241 (0.6%) received an autologous facial sling. In the autologous fascial sling cohort, 160 (66%) underwent rectus fascia harvest and 81 (34%) underwent fascia lata harvest. Compared to patients undergoing synthetic sling placement, those undergoing placement with autologous fascia were more likely to have an ASA class 2 or greater to undergo surgery by a urologist compared to an obstetrician/gynecologist (p<0.01 for both).
Operative characteristics, admission characteristics, and individual and composite complications for each suburethral sling type are presented in Table 1. Compared to synthetic mesh sling placement, autologous fascial slings were associated with longer operative times (median time 102 vs. 37 minutes, p<0.01) and longer length of hospital stay (LOS) (median LOS 0 vs. 1 day, p<0.01).
The composite complication rate was lowest for synthetic mesh sling (6%); higher rates were observed for autologous slings from rectus fascia (12%) and fascia lata (23%). The most common peri-operative complications were urinary tract infections (4.4%) followed by superficial SSIs (0.5%) and blood transfusions (0.3%). There were no differences in the rate of UTIs between synthetic mesh and autologous fascial slings (p=0.07); however, the rate of UTI was higher in autologous slings from fascia lata compared to rectus fascia harvest site (27% vs. 13%, p<0.01). After excluding UTIs, the composite complication rates for synthetic mesh, autologous rectus fascia, and autologous fascia lata slings were 3%, 8%, and 12% respectively.
Multivariable logistic regression models for risk of 30-day perioperative adverse events are presented in Table 2. When comparing composite complication rates between synthetic mesh and autologous fascial slings, autologous fascial slings were associated with increased odds of complications, even after adjusting for ASA class and surgeon subspecialty (adjusted odds ratio [aOR] 3.63, 95% confidence interval [CI] 2.56-5.15); this finding persisted with the exclusion of UTIs (aOR 4.63, 95% CI 3.02-7.09).
Compared to fascial slings from rectus fascia, fascial slings from fascia lata harvest site were associated with increased odds of composite adverse events (aOR 2.11, 95% CI 1.03-4.04). However, when considering composite adverse events with the exclusion of UTIs, the adverse event rate was similar between slings using the two fascial harvest techniques (aOR 1.93, 95% CI 0.81-4.63).
Both ASA classification and surgeon subspecialty were significantly associated with adverse events. Patients with an ASA class of 2 and 3+ were associated with 1.3 and 1.8 times the odds of complication, respectively, compared to those with an ASA class of 1. Sling surgeries performed by obstetrician/gynecologists were also associated with a greater likelihood of adverse events compared to those performed by urologists (aOR 1.54, 95% CI 1.40-1.68).
Interpretation of results
This is the first study to directly compare 30-day perioperative outcomes between synthetic and autologous suburethral slings using a large, validated surgical registry. Consistent with prior studies, we found higher adverse event rates for autologous fascial slings compared to mesh slings, likely due to the increased morbidity associated with fascial harvest [1]. Furthermore, our study adds to existing literature by comparing the unadjusted and adjusted odds of adverse events between sling types.
Additionally, among autologous fascial slings, those using fascia lata were approximately twice as likely to be associated adverse events. However, when urinary tract infections were excluded, the rates of composite adverse events were similar. The etiology of this finding is unclear; further studies are warranted. The retrospective design of this study should be taken into consideration for interpretation. Nevertheless, these data suggest that rectus fascia and fascia lata are both reasonable sites for fascial harvest when comparing major perioperative adverse events.