Accuracy of Surgeons' Prediction of Lead Efficacy During Stage I Sacral Neuromodulation

Objectives: A variety of pre-implantation testing paradigms are available to the surgeon offering sacral neuromodulation (SNM) for the treatment of urinary urgency/frequency, non-obstructive urinary retention, or fecal incontinence. We wondered, is bilateral advanced evaluation (Stage I) necessary, if the surgeon can accurately predict the lead which will be more effective for his or her patient? This study aims to describe surgeons’ accuracy in prediction of the effective lead during Stage I testing for implantation of SNM devices. Secondarily, we seek to evaluate patient-specific, condition-specific, and intraoperative factors to elucidate any predictors of accuracy.

Methods: Patients who underwent SNM-related procedures between 2016-2021 at a single institution with 4 FPMRS-trained surgeons were included in this retrospective record review. Patients who underwent successful staged SNM implantation with bilateral stage I testing were included in outcome analysis. Demographic, clinical, and intraoperative characteristics were collected. The first lead tested (predicted intraoperatively during Stage I) was compared to the lead ultimately implanted, yielding 2 cohorts: accurate prediction and inaccurate prediction. To analyze differences between the cohorts, independent T-test was used for continuous variables, Mann-Whitney U for non-normally distributed outcomes, and Chi-square/Fisher's exact test for proportions as appropriate.

Results: A total of 225 charts were reviewed; 144 had a staged procedure, and 126 progressed to Stage II (87.5% success). 22 with unilateral Stage I testing were excluded, and in 12, the order of testing was unknown. 97 were therefore analyzed: surgeons accurately predicted the more effective lead 58.8% (57/97; 95% CI, 48.3 - 68.7%) of the time (accurate prediction cohort), and were inaccurate in 41.3% (inaccurate prediction cohort). The two cohorts did not vary in subjective short-term or long-term improvement of symptoms (median duration of treatment 32.5 months); device-related complications, or revision/removal rates. No demographic (age, race, BMI, comorbidities) predicted accuracy, nor were there differences between the two cohorts in indication for SNM, rates of incontinence, rates of pelvic organ prolapse, symptom duration, or previous treatments. Fluoro time did differ; on average almost 10 seconds longer in the accurate prediction cohort. Interestingly, one surgeon's accuracy was higher than the others' (73.7% vs 57.1%, 52.2%, and 58.3%).

Conclusions: Our study demonstrates that, while utilizing the bilateral Stage I testing paradigm, surgeons' prediction of the more effective lead is no better than chance. Moreover, there are no clinical predictors (such as age, BMI, or comorbidities) that influence surgeons' accuracy. Our results thus support continued use of a 2-stage, bilateral testing paradigm to maximize efficacy of SNM. Weaknesses of the study include retrospective design, use of first lead tested as proxy for surgeon's prediction, and lack of availability of intraoperative motor response parameters or postoperative objective outcome measures. Strengths include relatively high-volume implantation, and success and complication/explantation rates similar to those already published.