**2. Results**

A total of 161 patients with voiding dysfunction refractory to medical therapy with a mean age of 58.8 ± 20.2 years who underwent urethral BoNT-A injection were enrolled. The patients were divided into three subgroups as follows: 60 DU patients (19 males and 41 females), 77 DV patients (10 males and 67 females), and 24 PRES patients (14 males and 10 females). The probable underlying comorbidities and bladder conditions that could be related to voiding dysfunction are listed in Table 1.

**Table 1.** Baseline characteristics and demographics of non-spinal cord injured patients with voiding dysfunction.


DU: detrusor underactivity, DV: dysfunctional voiding, PRES: poor relaxation of the external sphincter, DM: diabetes, CVA: Cerebrovascular accident, UTI: urinary tract infection.

The VUDS characteristics before and after treatment in the three study groups were compared, and the results are listed in Table 2. In female patients with DV, Pdet.Qmax significantly decreased after urethral BoNT-A injection (from 60.1 ± 36.0 to 47.6 ± 32.6, *p* = 0.004), and bladder outlet obstruction indext (BOOI) also significantly decreased in female DV patients (from 48.2 ± 35.8 to 30.0 ± 33.5, *p* = 0.000). The changes in Pdet.Qmax and BOOI in female DV patients were also statistically significant compared with the DU and PRES groups (*p* = 0.000 and *p* = 0.002). Other videourodymamic parameters did not differ significantly after urethral BoNT-A injections.

**Table 2.** Comparison of videourodynamic parameters before and after the urethral Botox injection in non-spinal cord injured patients with voiding dysfunction.


FSF: first sensation of filling; FS: full sensation; US: urge sensation; Pdet: detrusor pressure; Qmax: maximum flow rate; Vol: voided volume; PVR: post-void residual; CBC: cystometric bladder capacity; VE: voiding efficiency; BOOI: bladder outlet obstruction index. P, comparison of the changes in variables from baseline and after treatment among each group. \* *p* value < 0.05 comparison between baseline and after treatment.

Treatment outcomes, per the scaled GRA as described in the methodology, are listed in Table 3. The GRA was recorded a month after treatment. Per the postoperative GRA, we divided patients into three groups (0–1, 2, and 3). GRA ≥ 2 was considered a successful outcome. Overall, 100 of 161 (62.1%) non-SCI patients with voiding dysfunction were suc-

cessfully treated using urethral BoNT-A injections. As shown in Table 3, younger patients responded better to treatment (*p* = 0.016). On the other hand, sex was not significantly associated with treatment outcomes (*p* = 0.127). Finally, among patients with different voiding dysfunctions, we found that DV patients had better treatment outcomes than those with DU and PRES (*p* = 0.002). Approximately 76.6% of DV patients reported GRA ≥ 2, while 50% of DU patients and 45.8% of PRES patients reported GRA ≥ 2. 64 patients were under CIC and baseline and 30 patients voided well without CIC.

**Table 3.** Treatment outcome per the scaled Global Response Assessment (GRA) after urethral Botulinum toxin A injections.


DU: detrusor underactivity, DV: dysfunctional voiding, PRES: poor relaxation of the external sphincter.

We searched the predictive factors related to the treatment outcome of the baseline characteristics, including the underlying disease, lower urinary tract condition, and VUDS parameters. During multivariate analyses of factors associated with GRA ≥ 2 in the treatment outcome of patients with non-SCI voiding dysfunction, a diagnosis of DV (OR = 3.630, *p* = 0.002), more voided volume (OR = 1.004, *p* = 0.014) at baseline, and a history of recurrent urinary tract infection (UTI) (OR = 3.949, *p* = 0.007) were predictors of good treatment response. On the other hand, cervical cancer was a predictor of a poor treatment outcome (OR = 0.214, *p* = 0.008) (Table 4). Because only approximately 50% of DU patients had satisfactory outcomes, we further analyzed which factors could be better indicators of a good response. We found that a large PVR was a negative predictive factor for DU patients (OR = 0.995, *p* = 0.011). Because PVR is a predictor of a poor outcome in DU patients, a ROC curve analysis was performed. Figure 1 shows that PVR > 250 mL is a negative predictive factor for urethral BoNT-A injection in DU patients (Sensitivity = 0.567, specificity = 0.767, *p* = 0.008).

**Table 4.** Multivariate analysis of factors associated with a global response assessment ≥2 in nonspinal cord injured voiding dysfunction.


DV: dysfunctional voiding, s/p: status post operation, UTI: urinary tract infection.

**Figure 1.** Receiver operating characteristic analysis of the baseline post-void residual (PVR) volume in patients with detrusor underactivity.
