Complications of stress urinary incontinence surgery in men: a comparative analysis of urethral sling versus artificial urinary sphincter from a large national database

Abstract

Background:

Stress urinary incontinence (SUI) in men, often due to radical prostatectomy or sphincter deficiency, impacts quality of life. Surgical options include artificial urinary sphincter (AUS) and male urethral slings, valued for lower risks and cost. Understanding their outcomes aids in patient care.

Objectives:

To compare 30-day postoperative outcomes in AUS versus sling implantation in males.

Design:

Retrospective cohort study using a multicenter database.

Methods:

Male patients who underwent sling or AUS implantation between 2008 and 2022 were identified in the National Surgical Quality Improvement Program (NSQIP) database using current procedural terminology (CPT) codes. Patient characteristics, intraoperative factors, and 30-day outcomes were extracted and compared. Multivariate logistic regression adjusted for age, body mass index (BMI), race, ASA classification, anesthesia technique, smoking status, history of chronic obstructive pulmonary disease (COPD), congestive heart failure (CHF), diabetes, hypertension requiring medication, bleeding disorders, and chronic steroid use. A 1:1 propensity score–matched analysis was also conducted.

Results:

Among 4,498 patients, 39.6% received slings and 60.4% AUS. After adjustment, AUS was associated with higher odds of 30-day complications (OR 1.48 (1.09–2.02), p = 0.012), including surgical site infections (OR 2.19), overall infections (OR 1.84), implant complications (OR 4.08), genitourinary complications (OR 2.31), unplanned reoperation (OR 2.04), Clavien-Dindo Grade I–II (OR 1.58) and Grade III complications (OR 2.10), and prolonged hospital stay (OR 4.66–5.71; all p < 0.001). The 1:1 matched analysis largely supported these findings.

Conclusion:

AUS implantation is associated with higher 30-day postoperative complication rates compared to male urethral sling placement. These results may guide surgeons in their perioperative counseling regarding the short-term complication rates of both procedures, but further studies are needed to assess the long-term outcomes.

Plain language summary

Comparing complications after two surgeries for male stress urinary incontinence

Stress urinary incontinence (SUI) in men is a condition where men experience involuntary urine leakage, often caused by previous surgeries like prostate removal or a weakened urinary sphincter. Two common surgeries to treat SUI are the artificial urinary sphincter (AUS) and the male urethral sling. While AUS is typically used for severe cases, the sling is often preferred for its lower risks and cost. This study analyzed outcomes from 4,498 men who underwent these surgeries between 2008 and 2022, using a large national database. It compared short-term complications, focusing on issues occurring within 30 days after surgery. The findings revealed that men who had AUS surgery were more likely to experience complications, including infections, surgical site issues, problems with the implant, and the need for unplanned follow-up surgeries. They also had longer hospital stays compared to those who had sling procedures. These results can help doctors discuss the risks and benefits of each surgery with their patients and better prepare them for what to expect after surgery. However, more research is needed to understand the long-term outcomes of these procedures.

Keywords

artificial men prostatectomy suburethral sling urinary incontinence urinary sphincter

Introduction

Urinary incontinence is a common problem defined as the involuntary leakage of urine and has been found to increase in prevalence with age.¹ Stress urinary incontinence (SUI) is one type of leakage attributed to insufficient bladder outlet resistance. Reports suggest that, besides radical prostatectomy, the second most common cause of stress urinary incontinence in men is intrinsic sphincter deficiency.² According to a large cohort study, 25% of patients undergoing radical prostatectomy develop a bothersome leakage that remains at 10 years after treatment.³ If left untreated, SUI can affect the quality of life in men, leading to personal distress that might eventually cause social withdrawal.⁴

The initial treatment of male stress urinary incontinence takes on a conservative approach with nonpharmacologic therapies like pelvic floor muscle exercises, lifestyle modifications, bladder training, in addition to other adjunctive measures. It is recommended that this conservative approach be given a period of at least 6–12 months in males with incontinence following radical prostatectomy before offering the patients any surgical options.⁵

The two most common surgical interventions for the treatment of male incontinence are artificial urinary sphincter (AUS) implantation and perineal sling implantation. While the AUS is widely regarded as the “gold standard” for managing moderate to severe SUI in men due to its safety profile and predictable success rates, the male urethral sling has gained popularity, particularly with the introduction of the transobturator sling.^6,7 This is attributed to its lower complication rates, reduced financial cost, and favorable postoperative functional outcomes. Both procedures are effective for addressing SUI, with overlapping indications in cases of moderate incontinence, where patient characteristics, severity, and preference often guide the choice of intervention.

Achieving successful outcomes in either procedure is dependent largely on multiple factors, including patient selection, degree and character of the urinary leakage, patient’s cognitive abilities and manual dexterity, and prior surgeries for incontinence.⁸ Nonetheless, it is important to understand the relative postoperative complication rates of both procedures to be able to better counsel patients about available surgical options.

A couple of studies reported data about complications of AUS implantation and male urethral slings from retrospectively reviewed large single-institution case series^7,9 and a National Surgical Quality Improvement Program (NSQIP) dataset-based study, including patients from 2008 to 2011.¹⁰

Our study serves to provide updated data on the NSQIP database from 2008 to 2022. This is relevant to the recent increased adoption of male sling implantation procedures, while the AUS has not changed substantially in the past 30 years.¹¹

Our objective was to compare the 30-day postoperative complications for patients undergoing male sling versus AUS placement using the American College of Surgeons (ACS) NSQIP registry between 2008 and 2022.

Materials and methods

For our retrospective cohort study, the ACS NSQIP database was used, which is a variable-based data registry built to improve quality across surgical departments.¹² NSQIP data are collected directly from patients’ medical charts and track patients for 30 days after their operation. More than 150 variables are included in the database, and they assess factors such as patients’ baseline characteristics, perioperative conditions, and postoperative outcomes. More than 700 hospitals are participants and contribute data to NSQIP, most of which are in the United States (US).¹³

The ACS NSQIP and the hospitals participating in the ACS NSQIP are the source of the data used herein; they are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.

Data retrieval

We identified patients who underwent a sling or AUS implantation using Current Procedural Terminology (CPT) codes, a U.S.-based standardized system used in NSQIP to designate the primary surgical procedure performed. We retrieved all the cases from 2008 to 2022 using CPT codes 53440 and 53445. Sling operations for correction of male urinary incontinence with fascia or synthetic material are represented by 53440, and insertion of inflatable urethral/bladder neck sphincter procedures, including placement of pump, reservoir, and cuff, are represented by 53445. While the NSQIP database is subject to coding limitations, it has been validated and is maintained by trained surgical clinical reviewers to reduce such errors. Nonetheless, misclassification bias cannot be entirely excluded.^12,13 We excluded all patients who were not identified as male, including female and non-binary individuals, and patients with CPT surgical codes other than 53440 or 53445. Additional variables were created from existing ones, such as body mass index (BMI), occurrence of implant complications, incontinence, or retention (Supplemental Table 3), length of stay (LOS) with two or more days grouped (greater than 75th percentile (1 day)), and others (Supplemental Table 4).

The modified Charlson Comorbidity Index (mCCI)¹⁴ was created from existing variables in NSQIP. Postoperative complication variables were included in the analysis individually, as well as in stratified form according to the Clavien-Dindo classification system (CD).^15,16

Intraoperative factors included operation time and the principal anesthesia technique used. Preoperative risks chosen were history of chronic obstructive pulmonary disease (COPD), hypertension requiring medication, diabetes mellitus, congestive heart failure (CHF) in the past 30 days, bleeding disorders, and steroid use for a chronic condition, since these are the only preoperative risks available on NSQIP from 2008 to 2022.

Missing data were handled by creating a separate “unknown” category, where applicable, such as for the race variable. No formal sample size calculation was performed as this was a retrospective analysis of all eligible cases in the NSQIP database between 2008 and 2022. The large sample size and sufficient number of outcome events allowed for adequately powered comparative analyses.

For visualizing the trends of male sling and sphincter surgeries over the years within the NSQIP dataset, we used IBM SPSS Statistics version 29.0.0 (IBM Corp., 2022, Armonk, NY, USA). It is important to note that the manuscript was prepared in accordance with the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement.¹⁷

Statistical analysis

Patients’ baseline characteristics, intraoperative factors, and preoperative risks were compared based on the type of surgery as the exposure, using Pearson’s chi-squared test or independent samples t-test, where appropriate. Patients’ postoperative morbidity and mortality variables were compared using Pearson’s chi-squared test or Fisher’s exact test, where appropriate. Multivariate logistic regression was employed for analysis to create a model for each outcome of interest, except for length of stay, which required multinomial logistic regression due to its three categories.

Multivariate regression was done by including each postoperative outcome of interest as the dependent variable, producing a total of 17 different models. For each model, the type of surgery was included as a covariate using the “Enter” method. Perioperative variables were controlled for differently in each model, as deemed appropriate. These included age as a continuous variable, BMI as a continuous variable, American Society of Anesthesiologists (ASA) class, anesthesia type, smoking status, history of COPD, hypertension requiring medication, diabetes, history of CHF, bleeding disorder, and chronic steroid use as covariates. They were entered using the “Forward: Conditional” method, and the 95% confidence interval of the odds ratios was calculated in each model. The same was applied for the only multinomial logistic regression for length of stay, creating the final 17th model. This was done using IBM SPSS Statistics version 29.0.0 (IBM Corp., 2022, Armonk, NY, USA).

In addition to multivariate logistic regression, to account for potentially confounding baseline differences between groups, we conducted 1:1 propensity score matching using nearest neighbor matching with a caliper of 0.2 and no replacement. Matching was performed on the same covariates included in the multivariate analysis to ensure balanced baseline characteristics between the AUS and sling groups and to minimize selection bias.

Subgroup analysis

To assess whether patients with a higher comorbidity burden experienced increased complication rates, we stratified both the sling and sphincter groups into high and low comorbidity subgroups using the modified mCCI, with values >2 classified as high comorbidity burden. Within each procedure group, we compared the incidence of postoperative complications across comorbidity strata using Pearson’s chi-squared test or Fisher’s exact test, where appropriate. This was conducted using RStudio (Version 2024.04.2, Build 764; RStudio, PBC, 2024, Boston, MA, USA)

Results

A total of 4498 male patients were included from 2008 to 2022 (Figure 1), 39.6% were in the sling group, and 60.4% were in the sphincter group. Figure 2 shows a greater increase in sphincter over the years, compared with the sling group.

Figure 1.

Patient selection flowchart. Of 4552 patients identified by CPT codes, 54 non-males were excluded, leaving 4498 male patients: 1779 underwent sling placement and 2719 received a sphincter.

Figure 2.

Trends of male sling and sphincter surgeries over the years using the National Surgical Quality Improvement Program database. The solid line represents sling surgeries, while the dashed line represents sphincter surgeries.

In Table 1, it can be observed that most perioperative factors were significantly different between the sling and sphincter groups, patients in the sling group were older (p < 0.001) and had a lower BMI (p < 0.001). The sling group was more heterogeneous with 78.9%, 18.1%, and 3.0% white, black, and Asian people, respectively, compared to 87.3%, 10.8%, and 1.8% in the sphincter group (p < 0.001). People undergoing the sphincter procedure were higher risk preoperatively, with 55.7% having ASA class III–IV versus 41.6% in the sling group (p < 0.001). The same trend was seen with the mCCI score of 4+ being 26.1% versus 15.0% in sphincter and sling, respectively (p < 0.001). Operation time was higher in the sphincter, with a median of 89 min versus 70 min in the sling (p < 0.001), with general anesthesia being used more often in the sphincter group (97.9% vs 95.0%, p < 0.001). Other preoperative risk factors were all higher in sphincter versus sling, such as hypertension requiring medication (67.3% vs 59.1%, p < 0.001), diabetes mellitus (23.9% vs 18.1%, p < 0.001), CHF in 30 days before surgery (1.3% vs 0.6%, 0.022), and steroid use for chronic condition (3.3% vs 1.6%, p < 0.001).

Table 1.

Patients’ baseline, intraoperative, and preoperative factors stratified by surgery type.

	Type of surgery (n = 4498)
	Sling (1779)	Sphincter (2719)
	Count (%)	Count (%)	p-Value
Age, mean (SD), year	67.7 (8.3)	69.8 (8.6)	<0.001
Age as categories, year
18 to <60	247 (13.9%)	287 (10.6%)	<0.001
60 to <75	1187 (66.7%)	1629 (59.9%)
75 to <85	314 (17.7%)	712 (26.2%)
85+	31 (1.7%)	91 (3.3%)
BMI, mean (SD), kg/m²	29.1 (4.9)	29.8 (5.0)	<0.001
BMI as categories, kg/m²
Normal weight	311 (17.7%)	377 (13.9%)	<0.001
Overweight	791 (44.9%)	1147 (42.5%)
Obesity Class I	468 (26.6%)	776 (28.8%)
Obesity Class II	133 (7.6%)	299 (11.1%)
Obesity Class III	51 (2.9%)	90 (3.3%)
Race
White	1071 (78.9%)	2029 (87.3%)	<0.001
Black	246 (18.1%)	252 (10.8%)
Asian	41 (3.0%)	42 (1.8%)
Unknown	421 (23.7%)	396 (14.6%)
ASA classification
I	52 (2.9%)	31 (1.1%)	<0.001
II	986 (55.5%)	1172 (43.2%)
III	722 (40.6%)	1454 (53.6%)
IV	17 (1.0%)	58 (2.1%)
mCCI categories
0	26 (1.5%)	40 (1.5%)	<0.001
1	183 (10.3%)	183 (6.7%)
2	638 (35.9%)	717 (26.4%)
3	665 (37.4%)	1069 (39.3%)
4	227 (12.8%)	564 (20.7%)
5+	40 (2.2%)	146 (5.4%)
Operation time, median (IQR), min	70 (53 - 92)	89 (69–114)	<0.001
Principal anesthesia technique
General	1675 (95.0%)	2644 (97.9%)	<0.001
Regional	89 (5.0%)	56 (2.1%)
Smoker within 1 year	145 (8.2%)	209 (7.7%)	0.572
History of severe COPD	54 (3.0%)	94 (3.5%)	0.439
Hypertension requiring medication	1052 (59.1%)	1830 (67.3%)	<0.001
Diabetes mellitus	322 (18.1%)	649 (23.9%)	<0.001
CHF in 30 days before surgery	10 (0.6%)	34 (1.3%)	0.022
Bleeding disorders	35 (2.0%)	74 (2.7%)	0.108
Steroid use for a chronic condition	29 (1.6%)	89 (3.3%)	<0.001

Values are presented as counts with column percentages in parentheses, except if indicated otherwise. Age is presented as mean with standard deviation in parentheses, while BMI and operation time are presented as median with interquartile range in parentheses. Comparison across types of surgery was done using Pearson’s chi-squared test or independent samples t-test, where appropriate.

ASA, American Society of Anesthesiologists; BMI, body mass index; CHF, congestive heart failure; COPD, chronic obstructive pulmonary disease; IQR, interquartile range; MCCI, modified Charlson Comorbidity Index; SD, standard deviation.

In the sphincter group, patients with high comorbidity burden (mCCI > 2) had a significantly higher overall complication rate than those with lower burden (6.9% vs 3.4%, p < 0.001), including increased rates of urinary tract infection (p = 0.029) and progressive renal insufficiency (p = 0.042). By contrast, the sling group showed no significant difference in overall complications (3.4% vs 3.7%, p = 0.796), and no individual complication varied by comorbidity status (Supplemental Table 2). The 30-days postoperative outcomes between the two groups revealed the sphincter group to have higher overall complications (5.7% vs 3.5%, p < 0.001), including surgical site infections (SSI) (1.7% vs 0.7%, p = 0.004), any infections (2.9% vs 1.6% p = 0.005), implant complications (1.1% vs 0.2%, p = 0.002), any GU complications (2.2% vs 0.8%, p < 0.001), and unplanned reoperations (2.0% vs 1.1%, p = 0.022). CD Grade I–II (3.4% vs 2.0%, p = 0.008) and Grade III complications (3.3% vs 1.3%, p < 0.001) were also found to be significantly higher in the sphincter group compared to the sling group. Patients who underwent the sphincter procedure had a higher chance of a prolonged length of stay of 1 day and 2 or more days (67.0% vs 34.0% and 6.5% vs 3.1%, respectively, p < 0.001) (Table 2).

Table 2.

A descriptive table of 30-day postoperative complications stratified by surgery type.

	Type of surgery (n = 4498)
	Sling (1779)	Sphincter (2719)
	Count	Count	p-Value
Length of stay, days
0	1119 (62.9%)	719 (26.4%)	<0.001
1	604 (34.0%)	1823 (67.0%)
2+	56 (3.1%)	177 (6.5%)
Overall complications	63 (3.5%)	155 (5.7%)	<0.001
Any infection (other than C. difficile)	29 (1.6%)	80 (2.9%)	0.005
UTI	14 (0.8%)	34 (1.3%)	0.139
SSI	13 (0.7%)	47 (1.7%)	0.004
Sepsis	3 (0.2%)	10 (0.4%)	0.224
Any GU complications (excluding UTI)	15 (0.8%)	61 (2.2%)	<0.001
Implant complications	4 (0.2%)	28 (1.1%)	0.002
Incontinence	2 (0.1%)	8 (0.3%)	0.333
Retention	8 (0.4%)	17 (0.6%)	0.439
Other GU complications	9 (0.5%)	31 (1.1%)	0.027
Unplanned reoperation	15 (1.1%)	50 (2.0%)	0.022
CD 1–2	36 (2.0%)	92 (3.4%)	0.008
CD 3	23 (1.3%)	89 (3.3%)	<0.001
CD 4	10 (0.6%)	22 (0.8%)	0.370
CD 5 (death)	2 (0.1%)	3 (0.1%)	1.000

All variables were presented as counts with column percentages in parentheses. Comparison across types of surgery was done using Pearson’s chi-squared test or Fisher’s exact test, where appropriate.

CD, Clavien-Dindo; GU, Genitourinary; SSI, Surgical Site Infection; UTI, Urinary Tract Infection.

Results from the multivariate regression analysis revealed SSI (odds ratio (OR) 2.19 (1.18–4.07), p = 0.021), any infection (OR 1.84 (1.19–2.84), p = 0.006), implant complications (OR 4.08 (1.42–11.71), p = 0.009), any GU complications (OR 2.31 (1.30–4.09), p = 0.004), length of stay of 1 day (OR 4.66 (4.06–5.34), p < 0.001) and 2+ days (OR 5.71 (4.09–7.98), p < 0.001), unplanned reoperation (OR 2.04 (1.12–3.70), p = 0.020), CD Grade I–II (OR 1.58 (1.06–2.37), p = 0.026) and Grade III (OR 2.10 (1.33–3.32), p = 0.001), and overall complications (OR 1.48 (1.09–2.02), p = 0.012) to be significantly higher in the sphincter group compared to the sling group. Table 3 shows the odds ratio of the type of surgery within each model, with each model having one outcome of interest as the dependent variable.

Table 3.

Binomial and multinomial logistic regression results with odds ratios of sphincter to sling.

Outcome	Odds ratio (95% CI)	p-Value
Length of stay, 1 day	4.66 (4.06–5.34)	<0.001
Length of stay, 2+ days	5.71 (4.09–7.98)	<0.001
Overall complications	1.48 (1.09–2.02)	0.012
Any infection (other than C. difficile)	1.84 (1.19–2.84)	0.006
UTI	1.56 (0.82–2.98)	0.180
SSI	2.19 (1.18–4.07)	0.021
Sepsis	2.18 (0.60–7.91)	0.238
Any GU complications (excluding UTI)	2.31 (1.30–4.09)	0.004
Implant complications	4.08 (1.42–11.71)	0.009
Incontinence	2.50 (0.53–11.83)	0.247
Retention	1.02 (0.43–2.41)	0.972
Other GU complications	3.83 (0.86–17.00)	0.078
Unplanned reoperation	2.04 (1.12–3.70)	0.020
CD 1–2	1.58 (1.06–2.37)	0.026
CD 3	2.10 (1.33–3.32)	0.001
CD 4	1.44 (0.68–3.04)	0.344
CD 5 (death)	0.98 (0.16–5.85)	0.979

Bivariate regression was employed for analysis to create a model for each outcome of interest, except for length of stay, which required multinomial logistic regression due to its three categories. This resulted in 17 models, and the odds ratios with 95% confidence intervals are shown in parentheses. The odds ratio should be interpreted as sphincter to sling, with values >1 having a higher rate in sphincters and values <1 having a higher rate in slings.

CD, Clavien-Dindo; GU, genitourinary; SSI, surgical site infection; UTI, urinary tract infection.

The findings from the propensity score–matched analysis were largely consistent with those of the multivariate logistic regression (Table 4). The primary difference between the two analyses was overall infections (p = 0.087) and CD I–II complications (p = 0.094), which were not statistically significant in the matched cohort.

Table 4.

Propensity score–matched analysis of 30-day postoperative outcomes.

Outcome	Sling	Sphincter	p-Value
Length of stay, days
0	1053 (63.0%)	438 (26.2%)	<0.001
1	568 (34.0%)	1120 (67.0%)
2+	50 (3.0%)	113 (6.8%)
Overall complications	59 (3.5%)	90 (5.4%)	0.013
Any infection (except C. difficile)	26 (1.6%)	41 (2.5%)	0.087
UTI	13 (0.8%)	17 (1.0%)	0.584
SSI	11 (0.7%)	25 (1.5%)	0.030
Sepsis	3 (0.2%)	3 (0.2%)	1.000
Any GU complication	15 (0.9%)	40 (2.4%)	0.001
Implant complication	4 (0.2%)	18 (1.1%)	0.006
Incontinence	2 (0.1%)	6 (0.4%)	0.289
Retention	8 (0.5%)	11 (0.7%)	0.646
Other GU complications	2 (0.1%)	8 (0.5%)	0.114
Progressive renal insufficiency	1 (0.1%)	4 (0.3%)	0.371
Pulmonary embolism	4 (0.2%)	0 (0.0%)	0.125
Unplanned reoperation	13 (1.0%)	31 (2.1%)	0.030
CD I–II	32 (1.9%)	48 (2.9%)	0.094
CD III	23 (1.7%)	59 (3.9%)	<0.001
CD IV	10 (0.6%)	10 (0.6%)	1.000
CD V (death)	2 (0.1%)	1 (0.1%)	1.000

Propensity score–matched comparison of 30-day postoperative outcomes between sling and sphincter patients. Outcomes include length of stay, complications, infections, genitourinary events, reoperations, and Clavien-Dindo classifications. Significant differences were observed in overall complications, surgical site infections, genitourinary and implant complications, and CD III events.

CD, Clavien-Dindo; GU, genitourinary; SSI, surgical site infection; UTI, urinary tract infection.

Discussion

This study evaluated the 30-day outcomes of surgery for male urinary incontinence, comparing the urethral sling versus AUS over an 11-year analysis of the ACS NSQIP.¹⁷ Examining the early complications along with the patient’s characteristics would provide insight into device choice and utilization, as well as improving practice patterns to prevent adverse outcomes, which would directly affect device survival and effectiveness. Several studies showed that the choice of surgery for male incontinence is based solely on the severity of incontinence, which lacks a precise definition in the guidelines, without thoroughly evaluating patient’s characteristics.^18,19

Baseline differences were observed between the sling and AUS groups across several demographic and clinical characteristics, including age, BMI, comorbidity burden, and ASA classification. These differences likely reflect underlying variations in patient selection and surgical decision-making. To minimize confounding and ensure comparability between the groups, we performed 1:1 propensity score matching using key preoperative variables. After matching, the cohorts were well-balanced on most characteristics, allowing for a more accurate comparison of postoperative outcomes. Understanding the surgical complication profile of AUS and male urethral slings through accurate comparison allows for optimized counseling about the best choice for the patient rather than the ultimate better technique. Both slings and AUS are associated with long-term functional complications. Urinary retention, urinary tract infections, pain, and erosion are the most common complications for both procedures.²⁰ AUSs may carry worse complication rates due to mechanical failure and higher reoperation rates.²¹

Regarding short-term complications, Constable et al.²² found in their non-inferiority randomized controlled trial, a small rate of serious postoperative adverse events between the AUS and male sling groups. These results fail to highlight major differences in the postoperative complication rates, mainly due to a relatively small sample size.

An analysis of the NSQIP data from 2006 to 2013¹⁰ focused on the comparison of complication rates between male urethral slings and AUS.¹⁰ Their results also showed an overall low 30-day postoperative complication rate. However, urethral slings were associated with a significantly lower rate of overall adverse outcomes (2.8% vs 5.1%, p = 0.046). In our cohort, the overall rate of complications had demonstrated a slight uptrend in both groups in comparison with the prior analysis, where the rate of complications in the sling cohort had risen to 3.5% and that of the AUS had risen to 5.7%. Nevertheless, the AUS group still carries a significantly higher rate of complications than the sling group in both the regression analysis and the propensity score matching (p = 0.012, p = 0.013). Moreover, the prior analysis of NSQIP showed that the AUS was associated with a higher incidence of urinary tract infections (0.3 vs 2.0%, p = 0.020).¹⁰ Interestingly, our results did not show any significant difference between rates of postoperative urinary tract infections (p = 0.180). In the regression analysis, surgical site infections were higher in the AUS group (p = 0.021), and this was confirmed in the matched cohort results (p = 0.030) (Table 4). This is important because postoperative infections are an independent predictor of decreased device survival.²³ This was also solidified in another study that demonstrated that surgical site infections were associated with device reimplantation.²⁴

Given the relative implant complexity, the AUS group also had a higher rate of 30-day implant complications. Implant-related complications or dysfunction are expected to be higher given the mechanism of AUS, but it is interesting to note how this higher incidence is not only a long-term issue but also relevant to the early postoperative period. This is in concordance with results from a paper by Chughtai et al.,²⁵ comparing complications of different treatment modalities for SUI from Medicare beneficiaries. Their results showed similarly a higher day-of-procedure (4.4% vs 0.9%) and 90-day (3.2% vs 0.7%) device-related complications. The AUS is potentially prone to mechanical issues, including pump and cuff failure, as well as urethral atrophy and erosion requiring evaluation, which a study found to have led to a decline in implantation from 2000 to 2012, with the male slings becoming more common.²⁶ However, our graphical representation of more recent data on sling and sphincter frequencies from 2008 to 2022 shows an increase in sphincter procedures compared with sling procedures frequency’s relative stability over the years. This might represent the effect that established guidelines had on the frequency of both procedures by declaring AUS to be the gold standard.²⁷

In the regression analysis, the length of hospital stays (p < 0.001), CD III complications (p = 0.001), and unplanned reoperation (p = 0.020) rates were significantly higher in the AUS group. These findings were supported by similar trends observed in the propensity score–matched analysis. The higher incidence of reoperation rates aligns with results from older data by Alwaal et al.¹⁰ from the NSQIP database. We are limited in knowing the specific indications of reoperation; however, it is probably explained by the higher rate of device malfunction and higher rate of surgical site infection that were previously described in our results. Longer hospital stays are also concordant with the overall higher short-term complication rate of AUS.

Regression analysis of medical complications classified by severity as per the CD classification found a significantly higher incidence of Grade I–II complications (p = 0.026), defined as deviations from normal postoperative course with or without pharmacological intervention. However, this difference did not persist in the propensity score–matched analysis, suggesting that the initial finding may have been influenced by baseline differences between the groups. Despite an overall low incidence of medical complications in both groups, this discrepancy highlights the need for further investigation into the minor adverse events associated with AUS implantation and their potential clinical implications.

In our exploratory subgroup analysis of comorbidity burden using the mCCI, we observed that high comorbidity (mCCI > 2) was significantly associated with higher overall complication rates following AUS placement (6.9% vs 3.4%, p < 0.001), but not following sling procedures (3.4% vs 3.7%, p = 0.796). This finding partly contrasts with Inouye et al., who reported no association between comorbidity or frailty and short-term complications in a similarly large AUS cohort.²⁸ However, our results align more closely with a study that found frailty was associated with major complications and device removal procedures.²⁹ Notably, in our dataset, specific complications such as urinary tract infection and progressive renal insufficiency occurred more frequently in high comorbidity patients within the AUS group, suggesting that while mCCI may not be predictive of all complication types, but may be a useful marker of risk in select postoperative outcomes.

There are, in fact, several limitations to our study. Although we adjusted for a range of clinical and demographic covariates, residual confounding and selection bias remain possible due to the observational nature of the dataset and lack of randomization. Moreover, the NSQIP database lacks certain variables that might affect outcomes, including the etiology and severity of incontinence, history of incontinence procedures, history of anastomotic or bladder neck contracture, history of urethral strictures, and history of pelvic radiation beyond the 3-month preoperative period. In addition, the NSQIP database does not distinguish between autologous and synthetic male urethral slings, preventing us from assessing the impact of this variability on the complication rates, namely erosion and infection rates. This represents a potential source of misclassification bias due to unmeasured variation in sling type. Another major limitation is the short follow-up period. The NSQIP dataset only includes 30-day postoperative outcomes, which limits the ability to assess mid- and long-term complications, functional outcomes, and device durability—key endpoints for evaluating the long-term effectiveness of surgical interventions for male stress urinary incontinence. Future studies are warranted using prospective, longitudinal data sources or registry-based cohorts to evaluate the comparative effectiveness, durability, and patient-reported outcomes of AUS and sling procedures over longer follow-up periods.

Conclusion

Despite the overall low incidence of 30-day postoperative complications in both male sling and AUS implantation, AUS may carry worse complication rates compared to the sling, ranging from longer hospital stays to overall infections and unplanned reoperations. The findings of our study reinforce the results of previous research, highlighting the higher complication rates associated with AUS. Although both procedures offer a safe treatment option for stress urinary incontinence in the short term, further studies are required to study the long-term complications and functional outcomes of those procedures. Nonetheless, our study will enhance our ability to counsel patients about the 30-day postoperative outcomes of SUI surgery.

Supplemental Material

sj-docx-1-tau-10.1177_17562872251375528 – Supplemental material for Complications of stress urinary incontinence surgery in men: a comparative analysis of urethral sling versus artificial urinary sphincter from a large national database

Supplemental material, sj-docx-1-tau-10.1177_17562872251375528 for Complications of stress urinary incontinence surgery in men: a comparative analysis of urethral sling versus artificial urinary sphincter from a large national database by Towfik Sebai, Marwan Zein, Yara Ghandour, Baraa AlJardali, Hani Tamim and Albert El Hajj in Therapeutic Advances in Urology

Footnotes

Acknowledgements

Towfik Sebai and Yara Ghandour would like to acknowledge the training received under the Scholars in HeAlth Research Program (SHARP) that was in part supported by the Fogarty International Center and Office of Dietary Supplements of the National Institutes of Health (Award Number D43 TW009118). The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health). The American College of Surgeons National Surgical Quality Improvement Program and the hospitals participating in the ACS NSQIP are the source of the data used herein; they are not responsible for the statistical validity of the data analysis or the conclusions derived by the authors.

Declarations

ORCID iDs

Towfik Sebai

Marwan Zein

Supplemental material

Supplemental material for this article is available online.

References

Tennstedt

Chiu

Link

, et al. The effects of severity of urine leakage on quality of life in Hispanic, white, and black men and women: the Boston community health survey. Urology 2010; 75: 27–33.

Yafi

Powers

Zurawin

, et al. Contemporary review of artificial urinary sphincters for male stress urinary incontinence. Sex Med Rev 2016; 4: 157–166.

Al Hussein

Awamlh

Wallis

CJD

, et al. Functional outcomes after localized prostate cancer treatment. JAMA 2024; 331(4): 302–317.

Bernardes

Chagas

Izidoro

LCR

, et al. Impact of urinary incontinence on the quality of life of individuals undergoing radical prostatectomy. Rev Lat Am Enfermagem 2019; 27: e3131.

Bauer

Bastian

Gozzi

, et al. Postprostatectomy incontinence: all about diagnosis and management. Eur Urol 2009; 55: 322–333.

Lucas

Bosch

Burkhard

, et al. EAU guidelines on surgical treatment of urinary incontinence. Eur Urol 2012; 62: 1118–1129.

Bauer

Mayer

May

, et al. Complications of the advance transobturator male sling in the treatment of male stress urinary incontinence. Urology 2010; 75: 1494–1498.

Boone

TB.

How to decide whether an artificial urinary sphincter or a male sling is best for male stress urinary incontinence. J Urol 2016; 196: 641–642.

Linder

Piotrowski

Ziegelmann

, et al. Perioperative complications following artificial urinary sphincter placement. J Urol 2015; 194: 716–720.

10.

Alwaal

Harris

Awad

, et al. Comparison of complication rates related to male urethral slings and artificial urinary sphincters for urinary incontinence: national multi-institutional analysis of ACS-NSQIP database. Int Urol Nephrol 2016; 48: 1571–1576.

11.

Rehder

Haab

Cornu

, et al. Treatment of postprostatectomy male urinary incontinence with the transobturator retroluminal repositioning sling suspension: 3-year follow-up. Eur Urol 2012; 62: 140–145.

12.

ACS National Surgical Quality Improvement Program. ACS, 2024.

13.

User Guide for the 2022 ACS NSQIP Participant Use Data File (PUF). ACS, 2023.

14.

Ondeck

Nwachukwu

, et al. What associations exist between comorbidity indices and postoperative adverse events after total shoulder arthroplasty? Clin Orthop Relat Res 2019; 477: 881–890.

15.

Clavien

Barkun

de Oliveira

, et al. The Clavien-Dindo classification of surgical complications: five-year experience. Ann Surg 2009; 250: 187–196.

16.

Akhavan

Liu

Alerte

, et al. Predicting postsurgical complications of peripheral nerve decompression: NSQIP analysis of frailty measures versus historic proxies. Eplasty 2024; 24: e12.

17.

Von Elm

Altman

Egger

, et al. The Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) statement: guidelines for reporting observational studies. Lancet 2007; 370(9596): 1453–1457.

18.

Bhatt

Pavithran

Ilie

, et al. Post-prostatectomy incontinence: a guideline of guidelines. BJU Int 2024; 133: 513–523.

19.

Kumar

Litt

Ballert

, et al. Artificial urinary sphincter versus male sling for post-prostatectomy incontinence–what do patients choose? J Urol 2009; 181: 1231–1235.

20.

Kovacevic

Padmanabhan

Surgical management of post prostatectomy incontinence. Prostate Int 2024; 12: 65–69.

21.

Van Bruwaene

De Ridder

Van der Aa

. The use of sling vs sphincter in post-prostatectomy urinary incontinence. BJU Int 2015; 116: 330–342.

22.

Constable

Abrams

Cooper

, et al. Synthetic sling or artificial urinary sphincter for men with urodynamic stress incontinence after prostate surgery: the MASTER non-inferiority RCT. Health Technol Assess 2022; 26: 1–152.

23.

Kretschmer

Hüsch

Thomsen

, et al. Targeting moderate and severe male stress urinary incontinence with adjustable male slings and the perineal artificial urinary sphincter: focus on perioperative complications and device explantations. Int Neurourol J 2017; 21: 109–115.

24.

Jella

Fernstrum

Callegari

, et al. Peri-operative outcomes between primary and replacement artificial urinary sphincter surgery: an ACS-NSQIP analysis. Turk J Urol 2021; 47: 427–435.

25.

Chughtai

Sedrakyan

Isaacs

, et al. National study of utilization of male incontinence procedures. Neurourol Urodyn 2016; 35: 74–80.

26.

MacDonald

Colaco

Terlecki

Waves of change: national trends in surgical management of male stress incontinence. Urology 2017; 108: 175–179.

27.

Biardeau

Aharony

Campeau

, et al. Artificial urinary sphincter: executive summary of the 2015 Consensus Conference. Neurourol Urodyn 2016; 35(Suppl. 2): S5–S7.

28.

Inouye

Sexton

Boysen

, et al. Increasing comorbidity and frailty is not associated with early postoperative complications among men undergoing artificial urinary sphincter or sling implantation: a real-world application of NSQIP® data to reconstructive urology. Urology Practice 2021; 8(2): 264–269.

29.

Medendorp

Anger

Jin

, et al. The impact of frailty on artificial urinary sphincter placement and removal procedures. Urology 2019; 129: 210–216.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.03 MB