Clinical and Urodynamic Predictors for Sacral Neuromodulation Outcomes in Overactive Bladder

1.0 Introduction: Overactive bladder syndrome (OAB) is a prevalent disorder that affects about 10% of the adult population and > 40% of elderly. It is defined by the presence of urgency, with or without urge incontinence, in the absence of infection or other pathology. In 1997 through 1999, sacral nerve stimulation SNS (InterStim, Medtronic Inc., Minneapolis, Minnesota) was approved by the U.S. Food and Drug Administration (FDA) for treating urge urinary incontinence, urinary urgency, and frequency. Despite the large numbers of SNS performed, the only objective clinical evaluation of OAB is urodynamic detrusor instability (UDI) with some evidence suggesting a correlation with outcomes after sacral neuromodulation.

Interestingly, the mechanism of action of SNS is not fully understood. Theories include direct activation of efferent fibers to the striated urethral sphincter causing reflex relaxation of the detrusor or potential activation of afferent fibers selectively which can lead to inhibition at spinal and supraspinal levels. Somatic sacral afferent inflow activation at sacral level affects the storage and emptying reflexes in the bladder and central nervous system, explaining the beneficial effects of neuromodulation on both storage and emptying functions of the bladder. Malaguti and his colleagues detected somatosensory evoked potentials during sacral neuromodulation, revealing that sacral neuromodulation works by both sacral afferent activity and somatosensory cortex activation. As sacral neuromodulation is clinically proven for both storage and emptying bladder dysfunctions, it is difficult to isolate its action to either sacral afferent or efferent circuits in the micturition reflex pathway. In our study, we are going to study sacral neuromodulation outcome predictors from the clinical and urodynamic perspectives in order to help identifying the right candidates for sacral neuromodulation procedure.

2.0 Specific Aims:

• Specific aim#1: Using the medical reporting systems, we will identify all patients who, underwent sacral neuromodulation at University Hospitals of Cleveland from January 1, 2012 through February, 2017.
• Specific aim#2: Reporting on the efficacy outcomes including: rate of successful placement, rate of reported patient improvement after peripheral nerve (PNE) stage 1 and stage 2 procedure. The proportion of patients who undergo stage 2 compared to patients who had PNE. Rate of removal after full implantation and rate of battery change. Finally, will report on patient satisfaction, change in disease-specific severity scores and change in health-related quality of life
• Specific aim #3: Will identify preoperative clinical and urodynamic factors that can predict successful outcome and or/ failure.
• Specific aim #4: to identify predictors in subgroups of patients with OAB including:

1. Wet versus dry subgroups.
2. With urodynamic detrusor instability (UDI) versus no UDI subgroups.
3. Neurogenic versus non-neurogenic.
4. Staged versus non-staged procedure.

• Specific aim #5: Validating the identified predictors in a different established set of patients from Cleveland Clinic.
• Specific aim#6: Reporting on safety outcomes including intraoperative complications, early postoperative complications (within 30 days) and later postoperative complications. A modified Dindo classification will be utilized.

3.0 Data management and analysis: Data collection: Data will be collected from the medical reporting systems (Ambulatory medical reporting system and physician portal), University Hospitals Cleveland Medical Center and affiliated Hospitals, Cleveland, OH.

Data storage: REDCap software will be used for data storage. Computer software: SPSS 24 will be used for statistical analysis. Statistical tests: Continuous characteristics will be summarized with mean, median, and range; categorical characteristics will be summarized with number and percentage. For specific aim#2 and #3, #4, #5: associations of baseline characteristics with outcomes will be further evaluated using univariate, and multivariate logistic regression models and summarized with odds ratios (ORs) and 95% confidence intervals (CIs). Model calibration will be evaluated using the Hosmer and Lemeshow goodness-of-fit test to identify the best model. The comparison will be performed using independent sample T -test for normally disturbed continuous variables and Wilcoxon test for skewed data. Chi-square tests will be used for comparing categorical variables. For specific aim#6, Cox proportional hazard models will be used to evaluate the predictors of failure after device implantation, we will be using Hosmer and Lemeshow goodness-of-fit test. All tests will be 2 sided; P values less than .05 were considered statistically significant. Statistical analyses will be performed using SAS software (SAS Institute, Inc, Cary, North Carolina).