Stress urinary incontinence (SUI) is a common diagnosis amongst women. A popular and effective surgical treatment is the minimal invasive midurethral slings (MUS). Up to 85% of women treated with these operations are cured or experience significant improvements after surgery. Overall, the complications are few. A common long-term complication to MUS is exposure of the mesh to the vagina, urethra, or bladder. This complication is difficult to diagnose as symptoms vary considerably. Symptomatic treatment may include local estrogen, covering the mesh with vaginal tissue, or ultimately removal of the mesh, resulting in recurrence of urinary incontinence. Therefore, prevention of mesh exposure is paramount. It has been suggested that mesh complications after insertion of MUS is related to presence of inflammation and infection. There is no consensus about prophylactic use of perioperative antibiotics. Current use of prophylactic antibiotics for MUS is depending on local tradition, surgeon’s experience, and individual preoperative evaluation of patient risk. The Retropubic MUS (R-MUS) and the Trans-obturator MUS (O-MUS) are placed anatomically differently; the R-MUS has a close relation to the bladder, which might increase the number of bladder exposures, while the O-MUS has a more horizontal approach in the vagina, which might increase the risk of vaginal exposures. 
The aim of the study was to estimate the effect of single dose perioperative antibiotics on mesh exposure in patients undergoing MUS surgeries in a Danish national cohort study. Secondary, we wanted to compare the exposure rate after R-MUS compared to O-MUS.

We included women undergoing MUS during 2010-2018 in this nationwide register-based cohort study and followed them until December 31, 2018.  
Data was collected from the Danish Urogynecological Database (DugaBase). The DugaBase contains clinical data from all urogynecological surgical interventions performed in Denmark since 2006. 
The Dugabase was supplemented with data from the Danish National Patient Registry (DNPR). The DNPR includes diagnoses and surgical interventions at all Danish hospitals since 1977. In Denmark, it is mandatory to register surgical procedures in both databases, which ensures high data validity and completeness of data. The outcome, mesh exposure, was retrieved through diagnosis or procedure codes from the DNPR. 

Patients who previously underwent pelvic organ prolapse (POP) surgery with mesh and 
SUI surgeries were identified via the DNPR and excluded. Patients undergoing POP surgery with mesh after initial MUS were censored. 
We conducted Cox regression analyses to determine if a single dose perioperative antibiotic affected the risk of mesh exposure. We adjusted for patient age, body mass index (BMI), income, education, smoking habits, alcohol consumption, American Society of Anesthesiologist’s (ASA) score, surgeon’s experience and type of MUS. 
To investigate the risk of mesh exposure after R-MUS versus O-MUS, we conducted Cox regression analyses adjusted like mentioned above plus for antibiotics.
Due to the observational and register-based nature of this non-interventional study, a sample size calculation was not computed.

A total of 6,706 women were included, out of whom 5,178 (77.2%) were given antibiotics perioperatively, while 1528 (22.8%) women were not. Mean age was 52.1 in both groups. Furthermore, the groups were comparable according to BMI, ASA score, surgeon’s experience, smoking habits, level of income and education.
Totally 3,991 women underwent R-MUS surgery and 2,715 women underwent O-MUS surgery. Likewise, these groups of women were comparable. However, a higher proportion of women undergoing R-MUS received perioperative antibiotics (81.9%) compared with women undergoing O-MUS (70.4%).  

A total of 1.3% (87/6706) mesh exposures were diagnosed within the 9 year study period, 1.2% (62/5178) in the group that received perioperative antibiotics and 1.6% (25/1528) in the group that did not. 0.9% (34/3991) of the women were diagnosed with mesh exposures after R-MUS and 2.0% (53/2715) after O-MUS. 

We found an insignificant tendency toward reduced risk of mesh exposure when perioperative antibiotics were administrated in the total population (HR=0.7 (CI 0.4-1.1)). However, it was very different for the two types of MUS. After O-MUS significantly fewer women were diagnosed with mesh exposure if perioperative antibiotics were administrated HR=0.4 (CI 0.3-0.8). The same tendency was not found for women undergoing R-MUS  (HR=3.1(CI 0.7-13.1)). 
We found that the type of MUS highly affected the risk of mesh exposure. Undergoing R-MUS significantly reduced the risk compared with undergoing O-MUS: HR=0.4 (CI 0.3-0.6).

Cumulative incidence plots for mesh exposure comparing R-MUS and O-MUS (Figure 1) and comparing use of perioperative antibiotics or not in the women undergoing O-MUS surgery (Figure 2) are presented.

This nationwide cohort study found a significantly reduced risk of mesh exposures after O-MUS if perioperative antibiotics are administered. We did not find same significant association in the R-MUS group and in the total group.

This could reflect a type II error as few women undergoing R-MUS did not receive perioperative antibiotics and the number of mesh exposures was low. The findings indicate that perioperative antibiotics should be administered to women undergoing O-MUS to reduce the risk of mesh exposure. Whether perioperative antibiotics should be standard care for women undergoing R-MUS operation may be debated.

Furthermore we found that women undergoing O-MUS surgery have significantly more mesh exposures compared with women undergoing R-MUS surgery.

The risk of mesh exposure after O-MUS surgery is significantly reduced if perioperative antibiotics are administered.
Significantly fewer mesh exposures are seen after R-MUS, which might be preferred.

Continence 2S2 (2022) 100344
DOI: 10.1016/j.cont.2022.100344