Introduction

Overactive bladder (OAB) syndrome is defined as the symptom of urgency, with or without urgency urinary incontinence, usually alongside symptoms of frequency and nocturia, in the absence of urinary tract infection (UTI) or other pathology [1]. It is estimated that 10 % of adult women in the UK suffer with OAB, of whom 33 % suffer from urgency urinary incontinence [2]. The pathophysiology of OAB is complex and incompletely understood. Sacral nerve roots S2, S3 and S4 supply the detrusor muscle and elicit contraction via the parasympathetic nervous system [3]. Acetylcholine (ACh) mediates this response and acts on muscarinic receptors (M2 and M3 subtypes), with the M3 subtype being primarily responsible for bladder contraction. More recently the urothelium and suburothelium have been identified as important factors in the sensory function of the bladder, and several abnormalities in the expression of receptors have been identified in patients with OAB [4].

Botulinum neurotoxin (BoNT) is derived from the gram-positive bacterium Clostridium botulinum and is a potent neurotoxin that inhibits ACh release from the presynaptic terminal at the neuromuscular junction, resulting in muscle relaxation [5]. The mechanism of action is currently not fully understood, but several mechanisms of action have been postulated. BoNT is thought to inhibit release of ACh, ATP and substance P from the urothelium, and these are implicated in spinal reflexes contributing to OAB [4]. The processes involved in neurotransmitter exocytosis are complex and involve three proteins; vesicle-associated membrane protein (VAMP), synaptosome-associated protein 25 kDa (SNAP-25) and syntaxin. Together these form soluble N-ethylmaleimide-sensitive factor attachment protein receptor (SNARE). Several serotypes of BoNT have been implicated in the disruption of SNARE formation and hence inhibition of synaptic exocytosis [6]. There is also a further action on afferent C-fibres, which results in a reduction in the sensation of urgency [4].

There are currently two forms of BoNT available for clinical use, A and B. BoNT-A is currently recommended for injection into the detrusor as it has been shown to be effective and its effects last longer than its counterpart BoNT-B [4]. There are also two available forms of BoNT-A, onabotulinumA, marketed as Botox (Actavis, previously Allergan, Parsippany, NJ), and abobotulinumA, marketed as Dysport (Ipsen Biopharmaceuticals Inc., Basking Ridge, NJ). The National Institute for Health and Care Excellence (NICE) in the UK, the American Urological Association (AUA) and the European Association of Urology (EAU) recommend bladder wall injection with onabotulinumA in women with idiopathic detrusor overactivity (DO) who have not responded to conservative treatment and who are willing and able to self-catheterize [7–9]. OnabotulinumA has been used for several years in the treatment of lower urinary tract disorders, although it has only recently received approval by the Food and Drug Administration (FDA) in the USA (January 2013), and the Medicines and Healthcare Products Regulatory Agency (MHRA) in the UK (September 2013).

In the UK, NICE, currently recommends 200 units onabotulinumA, with consideration of a lower dose of 100 units in women who prefer a dose with which they would be less likely to need catheterization but who accept a reduced chance of success [9]. However, the AUA and, EAU currently recommend 100 units onabotulinumA as a starting dose [7, 8]. OnabotulinumA may be injected into the bladder wall via a rigid or flexible cystoscope under either local or general anaesthesia. It is injected systematically at 10 – 30 sites around the bladder wall, depending on the dose employed (10 units per site), avoiding the trigone [4].

Intradetrusor injection of onabotulinumA is a relatively minor procedure when compared to alternative surgical treatments and is usually undertaken as a day-case procedure. It therefore provides a viable minimally invasive alternative to major surgical intervention for OAB. However, the effects of an individual onabotulinumA treatment are of limited duration, and patients usually require repeated treatments. Although national guidance in the UK [9] suggests that patients should receive no more than three treatments per year, there is little evidence available regarding the duration of response to inform clinical practice.

The objective of this study was to investigate the efficacy and duration of response to intradetrusor onabotulinumA injections for the treatment of DO in women in a single-centre tertiary care setting.

Materials and methods

A retrospective review of patients receiving intradetrusor onabotulinumA injections (200 units Botox®) for DO in a single tertiary care urogynaecology centre between 2005 and 2012 was conducted. Data were collected manually from both operation records and follow-up clinic notes, using a data collection proforma. The local Clinical Governance and Risk Department (CGARD, equivalent to an institutional review board) declared that ethical approval was not required, and registered the study as a service review project (Newcastle upon Tyne Hospitals NHS Foundation Trust CGARD reference no. 5030).

The primary outcome measure was the duration of symptomatic relief received, defined as the difference in months between the date of the procedure and the date the patient requested to be added to the waiting list for repeat treatment. The discontinuation rate and reasons for discontinuation of treatment were also recorded. Primary treatment failure was also recorded; this was defined as less than 3 months of symptom relief where other reasons for discontinuation could not be identified. Secondary outcome measures were recorded to determine whether current unit practice was in line with national guidance [9]. These included indication for treatment with onabotulinumA, conservative treatments prior to onabotulinumA treatment, preoperative urodynamic investigation, and rates of postoperative complications.

Data were analysed using Microsoft Excel Mac:2011 and IBM SPSS 16. One-way analysis of variance was performed to determine the significance of differences in the mean durations of symptomatic relief for treatment cycles one to six. Kaplan-Meier survival analysis was used to determine differences in ‘survival’ of patients within the treatment programme.

Results

A total of 136 patients underwent intradetrusor onabotulinumA injections during the 7-year study period (Table 1). All patients (100 %) who received intradetrusor onabotulinumA injections had undergone urodynamic investigation prior to surgery, 92.6 % had idiopathic DO and 4.4 % had a diagnosis of neurogenic DO. All patients had received at least one conservative treatment including anticholinergic medication, bladder retraining or transcutaneous electrical nerve stimulation (TENS) of sacral dermatomes, with 96 % of patients having received at least two modalities of conservative treatment before proceeding to onabotulinumA treatment.

Table 1 Mean age of patients, indication for treatment, prior treatment modalities and total number of injections patients have received
Full size table

A large proportion of patients (80.1 %, 109 of the 136 patients) discontinued treatment over the study period, with only four patients (3 %) receiving a seventh treatment, and most discontinuing treatment following treatments 1 and 2 (24 % after treatment 1, 27 % after treatment 2). Of those who discontinued treatment, 52.3 % (41.9 % of patients) had no contact with the unit for more than 12 months after their last treatment. Primary treatment failure was observed in 5 patients (3.7 %). The most common reasons for discontinuing treatment were an alternative surgical treatment being performed in the Urogynaecology service (8.8 %) or referral to the Urology team (10.3 %). The Urogynaecology and Urology services have a well-established multidisciplinary team with a locally agreed division of specialist clinical services, so relatively small numbers of patients are referred between specialties. Other reasons for discontinuing the programme are given in Table 2.

Table 2 Patients continuing and discontinuing in the programme, including reasons for discontinuance where known
Full size table

The mean number of treatments received during the study period was 2.6, and the mode was 2 treatments. Only 9.9 % of patients received five or more treatments during the study period (Fig. 1).The time between patients receiving onabotulinumA treatment and contacting the unit to request that they be added to the surgical waiting list for retreatment ranged from 8.5 to 10.4 months for the first five cycles of treatment with a peak between the third and fourth cycles. This decreased to 5.5 and 5.3 months (p = 0.015) for the fifth and sixth cycles of treatment, respectively (Fig. 2). The Kaplan-Meier analysis of ‘survival’ within the programme showing time between treatments showed significantly lower ‘survival’ between the fifth and sixth treatments and between the sixth and seventh treatments compared to that between earlier treatments (χ 2 trend = 14.573; p = 0.012; Fig. 3).

Fig. 1
figure 1

Proportions of patients in relation to the number of onabotulinumA treatments received during the 7-year study period

Full size image
Fig. 2
figure 2

Mean times between onabotulinumA treatments and patient request to be added to the waiting list for repeat treatment

Full size image
Fig. 3
figure 3

Kaplan-Meier plot, showing intervals between treatments (log rank test for trend: χ 2 trend = 14.573; p = 0.012)

Full size image

Only eight women went on to have more than five treatments. Analysis of this group showed a similar trend with a mean of 8.4 months between their third and fourth treatments, that reduced to 5.3 months between the sixth and seventh treatments. During the study period, only one patient received more than seven treatments (she actually received a total of 11 treatments during the study period, and at the time of this report remained in the programme beyond 15 treatments). The duration of symptomatic relief following onabotulinumA treatment in this patient remained consistent over time with durations of 6, 7, 7 and 7 months between injections 7 and 8, 8 and 9, 9 and 10, and 10 and 11, respectively.

The main postoperative complications recorded were UTI and voiding difficulties. Following onabotulinumA treatment, 27 % of patients had a UTI and 22 % required intermittent self-catheterization (ISC). Although 11 women (8.1 %) were known to have a voiding problem and were already using ISC on a regular basis prior to the procedure, six of these (4.4 % of the total group) had a urodynamic diagnosis of neurogenic DO, and four (2.9 %) required ISC having had an earlier augmentation ileocystoplasty. Of those patients who initiated ISC postoperatively, 42 % also experienced a postoperative UTI.

Discussion

All patients treated with intradetrusor injection of onabotulinumA in our unit had urodynamic investigation and all underwent at least one modality of conservative treatment such as anticholinergic medication, bladder retraining or TENS of sacral dermatomes, prior to treatment. Hence, our unit practices were in keeping with the then relevant national guidance [10].

The mean interval between patients receiving onabotulinumA treatment and being added to the waiting list for repeat injections was used as an estimate of the duration of response to treatment. The intervals between injections showed a peak between the third and fourth treatments, with a significant reduction in the duration of response after the fifth treatment. Current guidance in the UK recommends no more than three treatments per year, i.e. a minimum interval of 4 months between treatments [9]. Extrapolating the line of best fit, our data indicate that at the ninth or tenth treatment the mean duration of response to treatment would be less than the recommended 4 months. The local unit definition of response is that treatments with symptomatic benefit of less than 3 months should be regarded as ‘treatment failure’, i.e. patients who undergo further onabotulinumA treatments may not get a sufficient therapeutic benefit to justify repeating the procedure.

Few studies have investigated the duration of response to onabotulinumA after repeat injections; those that have often included relatively small numbers of participants, included both men and women, and investigated the effects of onabotulinumA on neurogenic DO only. It is therefore difficult to directly compare the results of these studies with the data from this study. Although, Rechberger et al. reported data for a relatively large female population with idiopathic DO, they report only inter-treatment intervals for five treatments [11]. Previous studies are summarized in Table 3 [11–17]. Most of these studies evaluated up to five or six treatments, and suggest an inter-treatment interval between 6.0 and 11.1 months. The data from the current study suggest a range of 8.5 to 10.4 months between the first and fifth treatments, but, uniquely, a decrease to 5.4 months between the fifth and sixth treatments and between the sixth and seventh treatments. However, Tincello et al. investigated recurrence of symptoms after onabotulinumA treatment in cycles one to three and found an increased rate of symptom recurrence following treatments two and three when compared to symptom recurrence after the first treatment [18].

Table 3 Results from papers reporting time intervals between onabotulinumA injections
Full size table

The necessity for repeat treatment and treatment failure are attributed to axonal regeneration and possible activation of antitoxin antibodies [12]. The toxin effect on the detrusor muscle fades after about 1 year [12]. Therefore, patients need to be informed about the potentially temporary nature of the treatment and the possibility of requiring repeated treatments.

There are important limitations to this study. It was a retrospective analysis from a single centre, and inevitably included a small number of patients, particularly in the later injection cycles. However, analysis of the small group of women (eight) who went on to have more than five treatments showed a similar trend to the analysis of the total group, and therefore the size of this group did not affect our conclusions. No objective data in the form of urodynamic investigations and symptom/quality of life questionnaires were collected following injections with onabotulinumA. Therefore, the data are limited to subjective patient-reported outcomes. At the 3-month follow-up appointments, symptomatic relief was assessed and patients were requested to call the unit when symptoms recurred with a minimum interval of 3 to 4 months since the previous injection and were placed on the surgical waiting list for treatment usually within 4 to 6 weeks. Patients were not routinely seen back in the clinic prior to receiving further treatment and did not routinely undergo further urodynamic investigations. An assumption was made that patients requested repeat onabotulinumA treatment when symptoms recurred. The most bothersome symptom in virtually all women undergoing injection treatment in our service is urgency incontinence, and it was recurrence of incontinence that led to most patients contacting the unit. The data on duration of symptomatic relief were based on patients’ subjective assessment of recurrence of symptoms, and it is possible that some patients may have requested repeat treatment at the earliest indication of recurrence of symptoms in anticipation of time on the waiting list.

A large proportion of patients (42 % overall, or 52 % of those discontinuing treatment) had no contact with the unit for more than 12 months after their last injection; these were considered 'lost to follow-up', and are likely to have accepted the persistence of their symptoms, or to have moved on to alternative management plans elsewhere. Whilst it is possible that some might have remained symptom-free during this time, we do not think this very likely: the average time in these patients between prior treatments (for those receiving more than one) was 9 months, so relief of symptoms for over 12 months following later injections would not seem likely. Contacting patients to determine more accurately their level of symptoms at the point of withdrawal from the programme, and their reasons for not contacting the unit for in the preceding 12 months, would have been ideal. However, as a service review, without formal ethical approval, this approach was not open to us in this study.

A smaller number of patients (19.1 %) underwent an alternative surgical treatment or were referred to the Urology Service for further management due to treatment failure with onabotulinumA. However, a small number of patients discontinued due to primary treatment failure (3.6 %) defined as a ‘symptomatic response of less than 3 months’ and postoperative voiding difficulty (2.2 %). Ruffion et al. found a markedly higher rate of primary treatment failure (defined as ‘no response to treatment at 1 month follow up’) of 24 % (11 of 45 patients) [14]. However, Grosse et al. found a treatment failure rate of 6.4 % (12 of 187 patients). This disparity in treatment failure rates may be attributed to differences in when postoperative follow-up was undertaken, although initial symptomatic response to onabotulinumA injections may range from 4 days to 3 weeks [12].

The main postoperative complications recorded were UTI, voiding difficulties and the need for ISC. Interestingly, of patients in whom ISC was instituted postoperatively, 42 % also had a postoperative UTI. The cause/effect relationship here was not established. Patients who have high postvoid residuals are prone to UTIs, and yet UTIs are known to precipitate high residual urine volumes in patients [19]. Adverse effects following onabotulinumA treatment may be attributed to the drug itself or to the associated procedure [4]. Rates of UTI found in studies of intradetrusor onabotulinumA are comparable to the rates of UTI following cystoscopy alone [4]. When ISC is instituted, the rates of UTI increase to levels comparable to those in patients undertaking ISC for other indications [20]. It is not our current practice (nor was it our practice during the study period) to prescribe antibiotic prophylaxis routinely either in women undergoing onabotulinumA injections or in those undertaking (or initiating) ISC; we continue to reserve antibiotic therapy for patients who show symptoms of UTI.

Current guidance suggests that postoperative complications of haematuria, pelvic pain, transient dysuria, transient retention, difficulty urinating, feeling of incomplete emptying and UTI are prevalent in 10 % of patients undergoing onabotulinumA therapy [7, 9]. Postoperative voiding difficulty is a recognized adverse effect of onabotulinumA injection into the detrusor muscle and it is recommended that onabotulinumA therapy should only be offered to women who are able and willing to self-catheterize [4, 9]. Although the use of ISC is seen as an adverse effect, Brubaker et al. found a significant improvement in quality of life in patients after onabotulinumA treatment, whether or not ISC was instituted postoperatively [20]. Many patients are incontinent prior to treatment and most look on ISC as an acceptable price to pay for continence [4].

Notwithstanding the high rate of ‘lost to follow-up’, and discontinuance of treatment in this study, a small number of patients do continue to achieve an acceptable level of symptom benefit in the very long term (beyond 10 years and 15 treatments). Hence, with realistic counselling, we continue to see an important place for onabotulinumA treatment in our management of DO.

Conclusions

Although many reports indicate the efficacy and safety of onabotulinumA for the treatment of intractable DO, there are limited data on the duration of symptomatic relief following treatment. The results of this study suggest that in patients who show a response to onabotulinumA treatment, the duration of response declines after the fifth treatment, suggesting a possible tolerance effect to onabotulinumA and a subsequent decline in efficacy. There were, however, limitations with this study, and further research with additional objective outcomes and more complete follow-up would provide more conclusive evidence to support these findings.