Development of a predictive tool for sacral nerve modulation implantation in the treatment of non-obstructive urinary retention and/or slow urinary stream: a study from the Neuro-Urology Committee of the French Association of Urology
Résumé
Purpose
This study aimed to seek predictive factors and develop a predictive tool for sacral nerve modulation (SNM) implantation in patients with non-obstructive urinary retention and/or slow urinary stream (NOUR/SS).
Methods
This study was designed as a retrospective study including all patients who have undergone a two-stage SNM for NOUR/SS between 2000 and 2021 in 11 academic hospitals. The primary outcome was defined as the implantation rate. Secondary outcomes included changes in bladder emptying parameters. Univariate and multivariable logistic regression analysis were performed and determined odds ratio for IPG implantation to build a predictive tool. The performance of the multivariable model discrimination was evaluated using the c-statistics and an internal validation was performed using bootstrap resampling.
Results
Of the 357 patients included, 210 (58.8%) were finally implanted. After multivariable logistic regression, 4 predictive factors were found, including age (≤ 52 yo; OR = 3.31 CI95% [1.79; 6.14]), gender (female; OR = 2.62 CI95% [1.39; 4.92]), maximal urethral closure pressure (≥ 70 cmH2O; OR: 2.36 CI95% [1.17; 4.74]), and the absence of an underlying neurological disease affecting the lower motor neuron (OR = 2.25 CI95% [1.07; 4.76]). Combining these factors, we established 16 response profiles with distinct IPG implantation rates, ranging from 8.7 to 81.5%. Internal validation found a good discrimination value (c-statistic, 0.724; 95% CI 0.660–0.789) with a low optimism bias (0.013). This allowed us to develop a predictive tool (https://predictivetool.wixsite.com/void).
Conclusion
The present study identified 4 predictive factors, allowing to develop a predictive tool for SNM implantation in NOUR/SS patients, that may help in guiding therapeutic decision-making. External validation of the tool is warranted.