Since 1973, the artificial urinary sphincter (AUS) has emerged as the gold standard treatment in moderate to severe male stress urinary incontinence (SUI). 
Although urinary incontinence is known to worsen the condition of elderly patients, many surgeons are still hesitant about placing an AUS in these patients.
At this time, literature is scarce, making therapeutic decision difficult and patient information poor. 
The main objective was to assess the efficacy of male AUS in elderly patients, defined as 75 years old (yo) or more. 
Secondary we sought to determine the role of age over 75yo in safety and reoperation-free survival after AUS placement. We evaluated risk factors that may influence efficacy and reoperation-free survival specifically in the population over 75yo. Finally, we assessed the factors of erosion.

The charts of all male patients who underwent AUS implantation between 1991 and 2020 at 13 centers were retrospectively reviewed. The AMS-800 AUS (Boston Scientific™) was used in all cases. Patients with neurogenic SUI as well as with bladder neck implantation were excluded for the sake of population homogeneity. We compared patients over 75yo at AUS placement and patients under 75.
Our primary endpoint was AUS efficacy assessed by the postoperative social continence, defined as the use of 0 to 1 pad per day.
We secondary reported postoperative complications and reoperations (revisions, replacements, explantations and overall). A revision was defined as a reoperation consisting in replacement or repositioning of one or several components of the device.
We performed a survival analysis using Kaplan-Meier curves and Cox proportional hazard models to assess predictive factors of revisions, replacements or explantations. Predictive factors of continence were assessed using a logistic regression.

A total of 1233 patients met the inclusion criteria with 330 patients over 75yo (26.8%).
Regarding the characteristics of population, groups differed on comorbidities (BMI, Charlson index, ASA score, antiplatelet/anticoagulant drugs intake), but also on radiation exposure (40.5% among more than 75yo vs 31.5%), preoperative 24h pad-test (400.0mL vs 300.0mL for the youngest group), and etiologies of incontinence. Indeed although radical prostatectomy remained the first etiology in both groups (77.5% of the eldest vs 88.6%), focal therapies and endourology affected more patients in the older group (15.4% vs 6.6%). 
Concerning operative outcomes, we did not find any difference on surgical approach, pressure regulating balloon size, cuff size or intraoperative complications. Only cuff position showed a higher rate of transcorporeal position among the elderlies (13.1% vs 5.5%, p<0.001).
Social continence was achieved for 74.4% of the patients over 75yo, with no significant difference compared to the other group (80.1%, p=0.114).
We observed significantly earlier complications among elderly patients (18.8% vs 12.6%, p=0.014) but their Clavien-Dindo grade was significantly lower (p=0.025).They presented only 11.8% of Grade ≥ 3b complications (vs 28.4%). There was no difference on median hospital stay duration (3 days for both groups, p=0.901).
We observed similar overall reoperation rates in both groups (38.2% and 34.8%, p=0.299) but explantation rate was higher among the patients over 75yo (31.7% vs 22.6%, p=0.002). We also found a larger proportion of infections and erosions in the elderly patients (57.8% vs 41.9%, p=0.003). 
After a median followup of 24 and 19 months according to the group, the estimated median overall reoperation-free survival was 5 years for the patients over 75 years, and 6 years for the patients under 75 years, with no significant difference (p=0.076). 
The over 75yo group survival curve showed a tendency for early (<1year) reoperations. Also, we found a significantly worse explantation-free survival curve for the elderly group (p<0.0001). Otherwise we did not find any difference between revision-free and replacement-free survivals.
Among the predictive factors of  social continence (age over 75, center annual caseload, Charlson index, history of pelvic radiation, history of incontinence surgery, perineal surgical approach, cuff size, transcorporeal cuff position and post focal therapies/endourology incontinence), only the incontinence etiology was significant. Therefore, post endourology or focal therapies incontinence is associated with worse functional results (OR=0.36, [0.15-0.92], p=0.03).
In the Cox model risk factors of early reoperations were annual caseload and transcorporeal cuff position (HR= 1.06 [1.02-1.1], p=0.009 and HR=2.66 [1.49-4.7], p<0.001, respectively).

Our results appear to be in accordance with literature. 
Indeed we did not find any impact of age on functional results after an AUS implantation. [1]
Also we found a larger number of postoperative complications in elderly patients but most of them were low grade and didn’t affect hospital stay duration. [2]
There was similar reoperation-free survival in both groups [3] with nonetheless a tendency for early explantations in the group over 75yo, probably due to more frequent infections and erosions. 
We identified high annual caseload as a risk factor of early reoperations. This finding might be explained by a selection bias of expert centers - and a tendency to choose early reoperation when it’s needed.
Transcorporeal cuff position was a risk factor of early reoperations. There was not enough data to know if it was a matter of erosion or urethral atrophy leading to persistent SUI. It could be a confusion bias as patients are offered transcorporeal cuff position because of their history (multioperation, radiation exposure, fragile urethra).
We identified post-endourology or post-focal therapies incontinence as a risk factor of poor post-operative continence. This finding is difficult to discuss given the group heterogeneity.
Our study is not without limitations. Indeed we have few long-term postoperative urodynamic data and questionnaires scores. We also lack information about population characteristics -like cognitive status, dexterity or immune impairment.

AUS implantation in male patients over 75 years old appears to be an effective option to treat SUI with few early postoperative risks.
Yet we observed higher rates of early (2 years) explantations among elderly patients, probably due to the increased rate of device erosions and infections.
These findings would improve patients information and therapeutic decision-making.

Tutolo M, Cornu JN, Bauer RM, et al. Efficacy and safety of artificial urinary sphincter (AUS): Results of a large multi-institutional cohort of patients with mid-term follow-up. Neurourol Urodyn. 2019;38(2):710-718. doi:10.1002/nau.23901
Medendorp AR, Anger JT, Jin C, et al. The Impact of Frailty on Artificial Urinary Sphincter Placement and Removal Procedures. Urology. 2019;129:210-216. doi:10.1016/j.urology.2019.04.015
Raup VT, Eswara JR, Marshall SD, Vetter J, Brandes SB. Artificial Urinary Sphincters for Treatment of Urinary Incontinence in Elderly Males. Urol Int. 2016;97(2):200-204. doi:10.1159/000445254

Continence 2S2 (2022) 100228
DOI: 10.1016/j.cont.2022.100228