Hypothesis / aims of study
Botulinum toxin A (BTX-A) injections have been shown to be effective at treating patients with idiopathic refractory overactive bladder (OAB) who have otherwise failed lifestyle modifications and oral pharmacotherapy. However, some patients have little or no response to BTX-A. Furthermore adverse events such as voiding dysfunction necessitating clean intermittent self-catherisation (CISC) and urinary tract infections (UTIs) are bothersome and not uncommon. The aims of this study were to ascertain whether a poor response, voiding dysfunction and UTI were predictable to first-time BTX-A injections in a patient cohort of idiopathic refractory OAB.
Study design, materials and methods
This was a single centre, retrospective analysis of a dedicated database of patients who received their first BTX-A injections (100-200 U) to treat symptoms of refractory overactive bladder and idiopathic detrusor overactivity. Patients were recorded as either having a good response or bad response based on the minimally important difference for the UDI-6 questionnaire (1), which patients filled in at baseline and at 4-6 weeks post-injection. Data on patient demographics, urodynamic parameters as well as past medical and surgical history of each patient were also collected. Preliminary independent samples T-tests or Pearson’s Chi-Square tests were performed where appropriate. Subsequent univariate and multivariate regression analysis was performed to identify risk factors. Results were considered significant if P <0.05 for a variable with a two-tailed test.
Only patients with completed UDI-6 questionnaires pre and post BTX-A injection were recruited from the database. Ultimately 74 non-consecutive patients (50 females and 24 males) were analysed. Univariate logistic regression analysis suggested the only predictors of poor response were male gender (odds ratio (OR) 4.49, p = 0.006) and age (OR 1.05, p = 0.027). Interestingly, a lower dosage of 100 U was associated with a poorer response compared to 200 U (47.6% vs 25.5% respectively) but this did not reach statistical significance. In terms of predicting UTIs, upon multivariate logistic regression lower PIP1 in females was shown to be significant (odds ratio 0.93, p-value= 0.050). Predictability of CISC, on multivariate logistic regression, suggested lower maximum urinary flow rates (OR = 0.91, p = 0.023) and male gender (OR = 5.14, p = 0.013) were found to be significant variables. In addition, on univariate logistic regression in females, hysterectomy was associated with an increased risk of CISC (odds ratio= 4.55, p-value= 0.038). A dosage of 100U vs 200U was not shown to statistically effect CISC rates in this study (38.1% vs 46.2% respectively). Finally, on multivariate logistic regression (accounting for age, gender, dosage and the urodynamic parameters), there was an association of CISC and having a UTI after injection (OR = 11.63, p-value= 0.037).
Interpretation of results
Males were 5-6 times more likely to achieve poorer response compared to females in this study. Our data would suggest poor contractility in women would put you at risk of developing UTI. This could reflect poor emptying and higher post void residuals and therefore increased risk of UTI. Males were approximately 5 times more likely to need CISC compared to females on multivariate analysis which could reflect that benign prostatic enlargement could play a role in contributing to more voiding dysfunction compared to females. A lower maximum urinary flow rate value was associated with a higher risk of needing CISC, which again could in part reflect poor detrusor contractility. Females who had undergone hysterectomy were 4.5 times more likely to need CISC compared to females without. This could reflect a predisposition to detrusor underactivity following pelvic surgery. Interestingly, a dosage of 100 vs 200 U was not shown to statistically effect CISC rates in this study, however there was a trend observed (38.1% vs 46.2%). Patients who required CISC were more likely to have UTIs which may be associated with the catheter instrumentation. Limitations of the study include the small sample size, non-consecutive selection and retrospective nature of the study design.