Current literature on incidence and risk factors for acute retention of urine following urogynaecologic surgery is heterogeneous and inconclusive. Moreover, despite it being a well-recognized postoperative complication, there is no standardized protocol for postoperative trial of void to identify acute retention of urine. The goals of this study are thus threefold: 1- report the incidence of acute retention of urine after midurethral sling procedures, 2 - identify risk factors for postoperative acute retention of urine, and 3 - validate a recently proposed voiding trial algorithm (1), whereby post-operative trial of void was successful if the voided volume was ≥200mL or 100-199 with post-void residual <50%.

A retrospective cohort study was conducted on charts of patients undergoing outpatient mid-urethral sling procedure by a Urogynaecologist between January 2010 and June 2017. Patients were excluded if the they were male or if the chart lacked a 1st void voided volume or post-void residual. If the patient underwent two midurethral sling procedures in the study period, only the first one was included. Baseline data, urodynamic reports, surgical records, and outpatient records were reviewed to collect demographics, urodynamic, perioperative variables, trial of void data, and diagnosis of acute retention of urine. Acute retention of urine was defined as being discharged from day surgery needing catheterization if after retrograde filling with 300ml sterile water, the patient could not void at least 150ml on up to two separate attempts. Logistic regression was utilized to determine factors influencing acute retention of urine.  Sensitivity, specificity, positive predictive value and negative predictive value were calculated to determine screening value of a 1st voided volume of ≥200mL or 100-199 with post-void residual <50% as ‘no retention’. A p value <0.05 was used to denote statistical significance and analyses were performed using SPSS v24 and GraphPad Prism v6.07.

During the study period 316 mid-urethral sling procedures were performed. 60 were excluded because treatment was done by another specialist, 39 because of missing charts, 1 was male, and 15 were repeat procedures during the study period. Of the 201 included patients, the mean age was 54 years (SD=10), median BMI was 29 (IQR=26-34), and 93% of patients for whom the information was available self-identified as Caucasian (113/122). Chief presenting complaint was known for 175 patients. The most prevalent was stress urinary incontinence (54%), followed by mixed urinary incontinence (38%).  

Incidence of acute retention of urine after surgery was 17% (35/201). Patients presenting with acute retention of urine were significantly older, median (IQR) 55 years old (50-62) vs. 51 years old (46-61), and more likely to have had a retropubic mid-urethral sling (51% vs 16%, adjusted OR 5.4 [2.4-11.9]) than a transobturator mid-urethral sling. Spinal anesthesia, BMI, weight, race, operative time, and urodynamic factors were not associated with acute retention of urine. 

Complete trial of void data was available for 166 women (Table 1).

In our study, the algorithm therefore has sensitivity of 96% (95% CI 79-100), specificity of 91% (95% CI 85-95), positive predictive value of 64% (95% CI 46-79), and negative predictive value of 99% (95% CI 96-100).

Overall, 17% of our patients had acute retention of urine postoperatively. The proposed algorithm appeared to be reliable if ‘negative’ for retention, incorrectly identifying only one woman out of 130. However, the positive predictive value was poor as the proposed algorithm would have incorrectly identified 36% (13/36) of women as having acute retention of urine. Advancing age and retropubic route were associated with increased acute retention of urine, but not spinal anesthetic, BMI or operating time.

Acute retention of urine occurs in a significant proportion of patients undergoing mid-urethral sling surgery, necessitating better understanding of risk factors and standardized diagnostic criteria. Previously described risk factors including BMI and operative time were not significant in our population, though we present additional evidence that increased age and retropubic transvaginal tape procedures may be associated with higher risk of acute retention of urine. We also confirm that the proposed algorithm has a reliable negative predictive value and agree that a postoperative voided volume of ≥200mL is appropriate for use in a protocol to identify patients who are not in acute urinary retention. Ultimately, however, we recommend further studies to better identify a voided volume with greater positive predictive value.

FPMRS 2015;21(5 suppl), S49