Whether there are between-group differences (solifenacin versus mirabegron) about the non-persistence and retreatment probablilities is unclear. The aim of this study is to identify the predictors of nonpersistence and retreatment in women with overactive bladder syndrome (OAB) after first-line medical treatment, especially comparing solifenacin with mirabegron.

Overactive bladder women, who underwent urodynamic studies and received first-line medical treatment (5 mg solifenacin or 25 mg mirabegron per day) in a tertiary referral center, were reviewed from the outpatient medical records.

Between July 2010 and December 2020, a total of 608 women were reviewed in this study. Except age and pad weight,there were no differences between solifenacin and mirabegron (Table 1). 
      There was a statistical difference in the curves between the subgroups (log-rank test, p<0.0001). The median time to event was 0.9 months (95% confidence interval [CI] = 0.9 to 0.9 months) for lost to follow-up, 1.6 months (95% CI = 0.9 to 2.5 months) for improvement without medication, 2.8 months (95% CI = 1.6 to 3.5 months) for switched medication, and 4.2 months (95% CI = 0.9 to 6.1 months) for added OAB medication (Figure 1).  
      There was a statistical difference in nonpersistence, lost to follow-up, switched medication, added OAB medication and retreatment probability between solifenacin and mirabegron (Table 1, Figures 2A-E). 
      The following independent predictors were identified: (1) mirabegron (hazard ratio [HR] = 0.670) for predicting non-persistence; (2) mirabegron (HR=0.713) for predicting lost to follow-up; (3) low parity (HR=0.654) and Incontinence Impact Questionnaire (IIQ-7) score (HR=0.828) for predicting improvement without medication; (4)mirabegron (HR=0.257) and high Urogenital Distress Inventory (UDI-6) score (HR=1.089) for predicting switched medication; (5) mirabegron (HR=4.568), old age (HR=1.041), and high voided volume (dL, HR=1.520) for predicting added OAB medication; (6) mirabegron (HR=5.805) for predicting retreatment. 
      The 4-year retreatment rates were 25.9% in mirabegron and 7.6% in solifenacin. 
      The following cut-off values for predicting events were identified: (1) parity ≤3 and IIQ-7 score ≤7 for predicting improvement without medication; (2) age ≥74 years-old and voided volume ≥241 mL for predicting added OAB medication.

In this study, women who received solifenacin treatment tended to have higher nonpersistence, lost to follow-up and switched medication probabilities, compared with mirabegron. In addition, parity, Incontinence Impact Questionnaire score,  Urogenital Distress Inventory, age and voided volume  could be used as factors for predicting causes of nonpersistence.

Women who received solifenacin tend to switch medication and lost to follow-up. However, women who received 25 mg mirabegron tend to add medication and be retreated.