Urinary incontinence (UI) and pelvic organ prolapse (POP) are common conditions that may impair women's quality of life. 
Complications that can occur after pelvic floor surgery may be serious for a functional pathology. To describe postoperative complications, the most widely used classification is the Clavien-Dindo classification [1] because of its simplicity of use, its adaptability to different specialties and its reduction in subjectivity of reporting postoperative complications. A recent survey that administered online urologic scenarios to the members of the EAU (European Association of Urology) found moderate to high interrater agreement for the majority of the cases and validated the Clavien Dindo classification to grade and report postoperative complications in urology [2]. However, this classification does not apply for intraoperative complications, and does not take into account the perception of the women. 
Our main objective was to assess women’s perception of postoperative UI or POP surgery complications and compare it to the classification of Clavien-Dindo.

Women's participating to the VIGI-MESH registry about pelvic floor repair surgery were invited to quote their perception of complications severity through a clinical vignette-based survey. The questionnaire consisted of 30 clinical vignettes based on the complications described in the registry and presented in random order, the same for all participants.
For each vignette, 4 grades of severity were proposed: “not serious”, “a little serious”, “serious”, “very serious”. Women could explain their answers with a short commentary. The same questionnaire was completed by surgeons participating in registry. 
We performed a pretest in a sample of 60 patients to estimate the understanding of the clinical vignettes and to make modifications accordingly to the women comments.
We classified the vignettes according to Clavien-Dindo classification: some vignettes were easily classifiable, others could be classified by analogy to situations described by Clavien-Dindo. When we had no resources, we considered the clinical vignettes to be unclassifiable.
The responses to the questionnaire were described as relative frequencies and plotted in a histogram. We considered that the response of the women or the surgeons was concordant with the Clavien classification when the majority of them answered: "not serious" for a grade I Clavien-Dindo complication, "a little serious" for a grade II Clavien-Dindo complication, "serious" for a grade III Clavien-Dindo complication and "very serious" for a grade IV Clavien-Dindo complication. We considered perceptions of women and surgeons as concordant if the most frequent item of response was the same between the two types of participants.

Among all registered women, 1146 received the survey invitation by e-mail. We received 529 responses (46.2%). Mean age of the respondents was 57.5 (±13.8), 225 (42,5%) had midurethral sling (MUS) placement alone, 249 (47,1%) had POP repair alone, and 52 (9,8%) had POP repair and MUS. There was no significant difference between respondents and non-respondents regarding age, body mass index, menopausal status, type of surgery. Respondents had not experienced more complications than non-responders (10.6% had a grade 3 or higher complications vs 8.4%, p = 0.22). 
One hundred and forty-one surgeons were contacted and 70 of them returned the completed questionnaire (49.6%).
Comparing Clavien-Dindo and women’s classification of the clinical vignettes, 12/25 (48.0%) were concordant (# 1,5,11,12,16,19,22,24,26-29); 8 (61.5%) were estimate less serious by women (# 3,8,10,13-15,17,18), and 5 (38.5%) more serious (# 4,6,8,9,18,20,21).
Among the 5 unclassifiable clinical vignettes, 2 were classified as not or a little serious (# 2,23), 1 situation as not very serious (# 25) and 2 situations as serious (# 7,30).
Comparing Clavien-Dindo and classification of surgeons of the clinical vignettes, 13/25 (52.0%) were concordant (#1,11,16-19,21,22,24,26-29), 8 (61.5%) were estimate less serious by surgeons (#3,5,8,10,12-15) and 4 more serious (#4,6,9,20).
Women and surgeons responses were concordant in 20 out of 30 clinical vignettes (66.7%).

Our study shows that women’s perception of pelvic floor repair surgery complications is not concordant with the Clavien-Dindo classification for about half of the presented clinical vignettes. 
This classification, defined from analyses in a population of 650 patients undergoing cholecystectomy, is based on the invasiveness of the therapy engaged to treat these complications and it has been validated over the duration of hospitalization. The results of our survey suggested that duration and burden of the complication management before recovery is in favor of a more serious perception: a situation classified as grade II according to Clavien-Dindo such as anticoagulation for pulmonary embolism is perceived as serious by women; surgeons also classified this complication as more serious than the Clavien-Dindo classification. Long-term use of antibiotics due to recurrent urinary tract infections was also perceived as serious by women (and surgeons); probably because of the treatment duration and the risk of sequelae.
Women’s responses to clinical vignettes about urinary catheterization showed that a urinary indwelling catheter placed for a few days was better accepted by women than self-catheterization for chronic retention, whereas these situations are classified in the same grade according to Clavien-Dindo. This can be explained by the absence of recovery in case of self-catheterization. Self-catheterization can be perceived as a disability, altering self-image and relationship then provoking social anxiety. Compared to self-catheterization, a JJ urethral probe placement, considered as more invasive according to Clavien’s classification, appears to be a less serious complication for women; perhaps because there is no impairment of women’s autonomy.
Concerning intraoperative complications, our results show that if the planned surgery has been carried out even if it is not according to the planned technique, the women consider that the complication is not serious. If the planned surgery has not been performed, whether with or without complications, women feel it is serious. 
A qualitative analysis based on the comments of the women for each clinical vignette 
will be carried out to specify these results.

Women’s perception of pelvic floor repair surgery seems different from the Clavien-Dindo classification. Among women, lack of repair and long-term disability seem to be two major factors in favor of perception of the surgical complication as serious. We need to develop a classification of complication severity that better reflects patients’ perceptions.

Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004;240(2):205–213.
Mitropoulos D, Artibani W, Biyani CS, Bjerggaard Jensen J, Rouprêt M, Truss M. Validation of the Clavien-Dindo Grading System in Urology by the European Association of Urology Guidelines Ad Hoc Panel. Eur Urol Focus. 2018 Jul;4(4):608-613.