Summary of results
The FUTURE study is the largest RCT to date evaluating the clinical and cost-effectiveness of urodynamics investigation in the treatment pathway of women with refractory OAB or urgency-predominant MUI. FUTURE compared treatment outcomes in women following urodynamics plus CCA (referred to as the urodynamics arm) versus CCA only (referred to as the CCA only arm). The results confirm that the participant-reported success rates following treatments in women who underwent urodynamics and CCA were not superior to those who underwent CCA only [adjusted OR 1.12 (95% CI 0.73 to 1.74); p = 0.601]. We undertook sensitivity analyses and further per protocol analyses and the effect sizes were consistent with the main ITT estimates, providing confirmation and confidence in the results.
The results are consistent with Rovner et al., where 331 patients (male and female) were randomised to receive various doses of BoNT-A treatment versus placebo.50 All participants underwent baseline urodynamics and 75% had proven DO. They reported that patients with DO at baseline experienced similar reductions in the primary efficacy endpoint (mean change in weekly frequency of UUI episodes from baseline to week 12) compared with those without DO at baseline. However, their RCT was an efficacy study of BoNT-A versus placebo and the above comparison was a subgroup analysis and hence did not have sufficient power to detect significant differences between these relatively small subgroups. Nevertheless, they concluded that the similarity in symptom improvement between patients with refractory OAB with/without evidence of DO on urodynamics suggests it is not necessary to confirm DO using urodynamic testing prior to initiating BoNT-A treatment. In their study, patients with refractory OAB benefited from BoNT-A treatment regardless of baseline urodynamics diagnosis of DO. Similarly, Groenendijk et al. analysed the outcomes for all 111 women who underwent SNM and showed a statistically significant improvement in first sensation and maximum filling volume in women with UUI with or without baseline DO following SNM.56 They concluded that women with UUI but no DO are at least as successful as women with UUI and DO and therefore baseline DO should not be a prerequisite selection criterion for using SNM.
Participant-reported success rates
The primary outcome in FUTURE was the proportion of participants who reported success at their final follow-up time point (15 or 24 months post randomisation). Success was a participant response of either ‘very much improved’ or ‘much improved’ to the question ‘How would you describe your urinary/bladder problems (urgency and/or incontinence) now compared to when you joined the study?’ on the PGI-I assessment tool (i.e. comparing their symptoms in the last 2 weeks prior to completing the questionnaire to their symptoms on the day they were randomised). We considered all other responses (‘improved’, ‘same’, ‘worse’, ‘much worse’ and ‘very much worse’) as unsuccessful.
Our participant-reported success rates in both trial arms were noted to be lower than those reported in the literature. At the final follow-up time point, 117 women (23.6%) in the urodynamics arm and 114 women (22.7%) in the CCA only arm reported success as per our definition; that is, ‘very much improved’ or ‘much improved’. The majority of participants in the FUTURE trial received BoNT-A treatment (n = 620).
Brubaker et al. conducted a RCT evaluating BoNT-A treatment against placebo, in women with refractory OAB and associated DO on urodynamics, and showed that approximately 60% of women who received BoNT-A had a positive clinical response based on the PGI-I.49 However, they defined participant-reported success as a PGI-I score of 4 or greater at least 2 months after injection; that is, they defined success as ‘very much improved’, ‘much improved’ or ‘improved’ 2 months post treatment. Similarly, Chapple et al. (2013), in a double-blind, placebo-controlled RCT, showed that BoNT-A was associated with participant-reported success rates of 62%, where they defined success as ‘improved and greatly improved’ on the Treatment Benefit Scale – a four-point scale (‘greatly improved’, ‘improved’, ‘no change’ and ‘worsened’) at 3 months post treatment.51 In the Chapple et al. RCT 86% of the participants were women.
The reasons behind the apparent lower success rates in FUTURE are therefore trifold: the other studies in the literature (1) used a less strict definition of success, that is, ‘improved’ was classed as success, (2) had significantly shorter follow-up duration and (3) assessed the outcomes at set time points triggered by receiving the BoNT-A treatment, as compared to set time points post randomisation (as in FUTURE). The latter is especially important when considering BoNT-A treatment due to the transient nature of its success (i.e. improvement tends to wane 4–6 months after the injection). Unlike the above studies which were designed to assess the efficacy of BoNT-A treatment, FUTURE was as a pragmatic RCT evaluating the effectiveness of urodynamics in women with refractory OAB/urgency-predominant MUI. The treatment pathways therefore included other options of management, such as surgery for SUI, SNM, PTNS and others, including no treatment (see www.fundingawards.nihr.ac.uk/award/15/150/05). Hence it was necessary to set up the follow-up points post randomisation.
The above challenges were, however, identified early in the trial by the PMG and the independent DMC and to address these:
- A secondary analysis utilising a similarly less strict definition of success on PGI-I was included; that is, ‘very much improved’, ‘much improved’, and ‘improved’. Inclusion of ‘improved’ as a successful outcome had the effect of increasing the number of participants reporting success at the last follow-up time point in both the urodynamic arm (43.8%) and the CCA only arm (41.6%). However, the effect size was similar to our primary analysis [OR 1.14 (95% CI 0.79 to 1.65); p = 0.469].
- An additional PGI-I question specific to women who underwent BoNT-A treatment was introduced, asking them to rate their symptoms 2 months after receiving treatment, that is, short term post-treatment assessment as used by Brubaker et al. and Chapple et al.49,51 The participant responses to this question would therefore best represent the participant-reported success rates for the subgroup who received BoNT-A treatment. Using the primary definition of success on PGI-I, the participant-reported success rates were 63.8% versus 60.0% [OR 1.17 (99% CI 0.73 to 1.89); p = 0.518] in the urodynamics arm and the CCA only arm, respectively. Using the less strict definition of success (inclusion of ‘improved’) the participant-reported success rates were 83.3% versus 76.4% [OR 1.47 (99% CI 0.82 to 2.63); p = 0.195] in the urodynamics arm and the CCA only arm, respectively. These results are comparable to those reported by Brubaker et al.49 and Chapple et al.51 The effect sizes for both analyses were similar to our primary analysis.
The above extra analyses provide further reassurances of the robustness of the results of the primary outcome analysis.
Participant-reported success rates at earlier time points (3 and 6 months) showed significant differences between groups favouring CCA only, which was consistent using the original and less strict definition of success. However, by the last follow-up time point, the difference had disappeared and there was no significant difference between groups. The main explanation is that women receiving CCA only were more likely to receive their treatment without the delay of waiting for the urodynamics test, and therefore experienced earlier improvement in their symptoms. In women receiving BoNT-A treatment, the mean time to receiving the first dose of BoNT-A was 234.3 days in the urodynamics arm versus 188.4 days in the CCA only arm.
Recruitment
One thousand one hundred and three women were randomised, slightly over the recruitment target of 1096, making FUTURE the largest trial to date worldwide to recruit women with refractory OAB and the only one to date to evaluate the clinical and cost effectiveness of urodynamics in women with refractory OAB. The successful recruitment to target has been described by both the independent TSC and the DMC as a big achievement given the halt to all non-COVID-19 research in the UK from March 2020 till late 2020. A more detailed description of the impact of COVID-19 on the FUTURE trial in terms of recruitment, data collection and analyses of outcomes is outlined later in this chapter.
Response rates
Questionnaire response rates were over 90% for the 6-, 15- and 24-month post-randomisation questionnaire, an excellent achievement in these types of trials and comparable with the three largest studies assessing the effectiveness of BoNT-A as a treatment for refractory OAB. Chapple et al. randomised 548 patients; 89% completed the 6-month follow-up timepoint.51 Tincello et al. recruited 240 patients; 83% completed the 6-month follow-up time point (n = 199).102 In Rovner et al., 87% (272/313) completed the 36-week follow-up time point.50 In Brubaker et al., a total of 43 subjects were randomised, including 28 to BoNT-A and 15 to placebo; the study was stopped prematurely due to the higher-than-expected rates of increased PVR and associated UTIs.49
Choice of the primary outcome
The primary outcome measure in FUTURE was participant-reported success at 15 months post randomisation as measured by PGI-I.
We surveyed surgeon opinions (urologists/urogynaecologists) in 45 units at the time of trial design and the vast majority recommended follow-up time points at 6 and 15 months post randomisation. These two timings are appropriate for measuring the outcomes of treatments for refractory OAB: 6 months is adequate to capture participant-reported and objective outcomes and early AEs for each treatment (such BoNT-A or SNM), while 15 months post randomisation is appropriate to compare the overall outcomes (participant-reported/objective success rates, participant satisfaction, AEs, further treatment, cost–utility and cost-effectiveness) between trial arms.
The PGI-I is a global index that is widely used to rate the response of a condition to a therapy (transition scale). It is a simple, direct and easy to use scale that is intuitively understandable to clinicians and patients.103 The PGI-I has excellent construct validity compared to various assessment variables: incontinence episode frequency, the incontinence HRQoL questionnaire, and fixed volume (400 ml) stress pad test.104
In a benchmark study, Yalcin and Bump reported a secondary analysis of data from two double-blind, placebo-controlled clinical trials (n = 1133) that evaluated treatment of women with predominant SUI.105 The authors showed that significant correlations (p < 0.0001) were observed between the PGI-I response categories and three independent measures of improvement in SUI (0.49, 0.33 and −0.43 with incontinence episode frequency, stress test and Incontinence HRQoL Questionnaire results, respectively). This important study established the construct validity of the PGI-I for the evaluation of the baseline severity and treatment response in women with UI.
The PGI-I has been widely used in clinical trials assessing surgical and conservative interventions for UI in women:
- The RELAX trial evaluated the PGI-I as an outcome assessment tool in women with refractory OAB symptoms requiring BoNT-A treatment (i.e. a similar cohort to the FUTURE study).102 The results showed that ‘the PGI-I scales are robust and valid instruments to assess disease severity, bother and improvement after treatment in women with detrusor overactivity’.
- Brubaker et al. used PGI-I as the primary outcome in their RCT evaluating BoNT-A treatment versus placebo in women with refractory OAB.49 The results showed a successful outcome in approximately 60% of the women who received BoNT-A treatment. The authors showed that PGI-I was able to detect differences in responses between groups.
- The PGI-I was utilised in the SIMS RCT (n = 600), evaluating surgical interventions for SUI in women.106 The results showed excellent response rate at 1- and 3-year follow-up.
- Two clinical trials on 10-year outcomes following surgical treatment of SUI in women utilised PGI-I as their primary participant-reported outcomes.107,108
One study compared the PGI-I versus the change in ICIQ-SF score for women undergoing UI or pelvic organ prolapse surgery; and reported that PGI-I may overestimate participant-reported success.109 The accompanying editorial questioned their methodology of the arbitrary conversion of individual scores for both questionnaires to the same scale as the numerical values assigned to each category in the underlying items.110
Our team has considered the use of disease-specific questionnaires (such as ICIQ-OAB or ICIQ-LUTSQoL) and the change in the post-treatment scores as the primary outcome. However, such an approach would have a number of potential drawbacks: (1) OAB-specific questionnaires/scores would inevitably miss the impact of the interventions on other urinary symptoms, such as new-onset or worsening of pre-existing stress incontinence and/or voiding dysfunction; (2) the post-intervention score-change that represents minimal clinical important difference was found to differ between studies, hence is not considered reliable.111,112 We have therefore included these validated questionnaires as secondary outcomes.
In summary, PGI-I provides a robust validated and more global review of the treatment outcome and is more encompassing of the range of benefits and potential harms.110 We used appropriate disease-specific symptom severity and HRQoL questionnaires as secondary outcomes.
Subgroups and participant-reported success rates
We have further analysed the participant-reported success rates in both groups according to the baseline clinical diagnosis of OAB and urgency-predominant MUI. In line with the main results, at the 3-month follow-up time point, women in the CCA only arm had significantly higher participant-reported success rates in both subgroups. However, at the final time point, the difference in both subgroups was not significant. The subgroup effects of 1.14 (99% CI 0.33 to 3.90), p = 0.788, and 1.07 (99% CI 0.39 to 2.95), p = 0.861, for the original and less strict definitions of success respectively show that there is no evidence of a significant difference in the effect of urodynamics between participants with a baseline clinical diagnosis of OAB compared to urgency-predominant MUI.
As the majority of women in FUTURE received BoNT-A treatment, we undertook a further post hoc analysis to evaluate the effect of urodynamics in the cohort of participants who underwent BoNT-A treatment according to their baseline clinical diagnosis (OAB vs. urgency-predominant MUI). The participant-reported success rates 2 months following BoNT-A treatment showed urodynamics to be more effective in the urgency-predominant MUI group compared to OAB, but the subgroup interaction was not significant [OR 0.74 (99% CI 0.19 to 2.96); p = 0.581].
Other studies in the literature did not specifically make a similar comparison. Tincello et al. tested whether any baseline covariates had an impact on outcomes following BoNT-A treatment.102 They found the only associated factor was severity of frequency at baseline: for each additional voiding episode at baseline, they showed an additional 0.37 (95% CI 0.17 to 0.58; p < 0.001) reduction in voiding frequency.
The small number of women in the subgroups receiving other treatments [PTNS (48 women), SUI (21 women) or SNM (19 women)] would make comparison of outcomes by group clinically and statistically non-meaningful.
Urinary symptoms
We analysed urinary symptoms at baseline and at the follow-up time points using the ICIQ-FLUTS and bladder diaries.
International Consultation on Incontinence Questionnaire female lower urinary tract symptoms
ICIQ-FLUTS is a validated symptom severity questionnaire for three domains: filling, voiding and incontinence. As expected, there was no improvement in the voiding domain score but slight deterioration. This is best explained by the fact that treatments for refractory OAB can relax the detrusor muscle in parallel to reduction in the sensitivity of the detrusor receptors. BoNT-A treatment is known to lead to an increase in PVR volumes and may require women to perform CISC to achieve complete bladder emptying. Rovner et al. reported a dose-dependent increase in PVR with BoNT-A treatment which declined steadily to week 36.50 Tincello et al. showed that voiding difficulty requiring CISC was reported by 16% of women in the BoNT-A group and by 4% of the placebo group [OR at 6 months 4.87 (95% CI 1.52 to 20.33); p = 0.003].102
In FUTURE it was reassuring to see improvement in both the filling and incontinence domain scores in both groups at the final time points compared to baseline. The between-group difference was significant only in the filling domain, favouring urodynamics. The filling domain best represents improvement in overall OAB symptoms (daytime/night frequency and urgency) while the incontinence domain reflects improvement in the frequency of the UI episodes.
We utilised two disease-specific assessment tools for OAB symptoms: the ICIQ-OAB scores and the UPS. At the final follow-up time point, ICIQ-OAB scores showed improvement in both groups compared to baseline, with no significant differences between the study groups. Similarly, the percentages of women reporting cure/improvement in urgency on the UPS were similar: urodynamics 45% versus CCA only 42%.
Bladder diaries
One-third of women in FUTURE completed the 3-day bladder diary at 6- and 15-month follow-up. Diaries represent excellent semi-objective assessments of various urinary symptoms, such as daytime/night frequency, urgency and its severity, UI episodes and their severity, number of pads and others. The baseline findings were similar to other RCTs in the literature which included women with refractory OAB: urgency and UI episodes were 7 and 4/day, respectively. Chapple et al. and Tincello et al. reported urgency episodes of ~9/day and 8/day and UI episodes of ~5/day and 6/day at baseline, respectively.51,102 Rovner et al. reported a mean baseline daytime frequency of ~7 to 8/day.50
We analysed results for daytime frequency, nocturia, urgency episodes (mild/moderate and severe) and UI episodes (SUI/UUI). Apart from significant reduction in daytime frequency favouring CCA only at 6 months, all other parameters were similar between groups at 6- and 15-month follow-up.
There are no other RCTs in the literature that compared the impact of urodynamics versus CCA only on post-treatment urinary symptoms and parameters of bladder diaries. However, several studies assessed the impact of BoNT-A treatment in women with refractory OAB:
- UI episodes: Chapple et al. reported that participants with refractory OAB who received BoNT-A treatment perceived a significant overall improvement in their UI episodes: a mean decrease of 2.95 episodes.51 The EMBARK study group assessed the efficacy of 100 units of BoNT-A in women with refractory OAB: they reported a mean 47.9% reduction of UI episodes at 3-month follow-up compared to baseline;113 60% of the participants who received BoNT-A treatment experienced ≥ 50% improvement in their UUI episodes.
- Urgency episodes: the EMBARK study also showed a mean reduction of 2.93 (95% CI −3.43 to −2.44) in urgency episodes at 3 months compared to baseline in participants who received BoNT-A treatment; this equated to a mean change of 30% reduction in urgency episodes.113
Health-related quality of life
We assessed the HRQoL in both groups using both general and disease-specific validated tools; hence the results are robust. There were no significant differences in any of the HRQoL scores between the groups. The ICIQ-LUTSQoL showed improvement in HRQoL compared to baseline in both groups. However, this was not reflected in improvement in the general HRQoL scores on the EQ-5D-5L.
The EMBARK study group utilised disease-specific questionnaires (incontinence QoL and the KHQ) and showed significant improvement in HRQoL compared to baseline in women with refractory OAB who received BoNT-A treatment (compared to placebo).113 Similarly, Chapple et al. utilised both questionnaires and showed > 10 points improvement in all three domains of incontinence QoL questionnaire compared to baseline and > 5 points improvement in six out of seven domains in the KHQ in participants with refractory OAB who received BoNT-A treatment.51 The single KHQ domain that did not show a clinically significant improvement was the ‘General Health Domain’, which is in agreement with the results where the general EQ-5D-5L scores did not show an improvement.
Adverse events
The current literature shows that from the patients’ perspective, many describe urodynamics as invasive, intimate, embarrassing and a painful investigation associated with emotional distress.62,63 Urodynamics is also associated with a risk of discomfort and UTIs.52
Reassuringly, in the FUTURE trial the AE rates were low and similar in the two groups. UTI was the most common AE (7%); as these were participant-reported it was difficult to establish which were related to urodynamics and which were related to BoNT-A. Reassuringly, other AEs reported following urodynamics were low, including pain/burning sensation (0.7%) and fainting (0.4%). The most commonly reported AEs were mainly those expected after BoNT-A treatment, as it was the most commonly utilised treatment; the use of long-term prophylactic antibiotics was reported in 7% and CISC in 5% of participants. Buttock numbness after SNM occurred in one participant. The Chapple et al. results showed ‘uncomplicated UTI’ as the most frequently reported AE, with only one case of complicated UTI, which occurred approximately 4 months after BoNT-A treatment.51
This study results showed nine SAEs during the follow-up period, including five unrelated deaths and two CISC. CISC is not normally classed as a SAE in clinical practice, hence in our opinion there were only two SAEs in FUTURE that were serious, related and expected; a case of upper UTI following BoNT-A that required hospitalisation and a general-anaesthetic-related complication following SUI surgery. In addition, limb weakness was reported (as an AE) by 2.2% of participants, which is relatively higher than other studies in the literature; these could be classified as AE related to BoNT-A treatment. In the RELAX study, there were three BoNT-A-related SAEs: two women reported generalised muscle weakness severe enough to interfere with daily activities, and one woman suffered a bronchopneumonia within 3 weeks of the BoNT-A injection.102
Impact of urodynamics on clinical decision-making
In order to answer this question, we analysed the results for the 499 participants who underwent urodynamics within FUTURE to answer two important questions:
1. Did urodynamics change the baseline clinical diagnosis?
Of the 499 participants with refractory OAB symptoms who underwent urodynamics, a diagnosis following urodynamics was available for 494 participants: 58% of the participants were diagnosed with DO and/or DOI, 13% had USI, 8% had urodynamics MUI and 21% had no evidence of USI or DO.
When adjusted for the baseline clinical diagnosis:
- in the group of participants with a baseline diagnosis of OAB, their urodynamics diagnoses were DO/DOI in 62.3% of cases, MUI in 6.2%, no evidence of USI or DO in 21.9% and USI in 8.3% of cases
- in the group of participants with a baseline diagnosis of urgency-predominant MUI, their urodynamics diagnoses were DO/DOI in 48.6% of cases, MUI in 10.9%, no evidence of USI or DO in 18.3% and USI in 21.7% of cases.
The results therefore showed that DO/DOI was not detected on urodynamic assessment in 34% of women with a clinical diagnosis of refractory OAB/urgency-predominant MUI. These findings are consistent with previous literature showing urodynamics fails to show evidence of DO in up to 45% of women with OAB.60 In the latter, they assessed urodynamics diagnoses in the general population of women with OAB as compared to FUTURE where the cohort of women had refractory OAB.
One in five women in both groups had no evidence of either USI or DO on urodynamics; that is, the urodynamic test did not make any advances in diagnosis of their condition. In clinical practice, these women would receive symptomatic treatment according to their predominant symptoms and CCA diagnosis.
The results also showed that urodynamics clearly changed the diagnosis in 13% of women with refractory OAB/urgency-predominant MUI to USI. More women in the urgency-predominant MUI group were likely to be diagnosed with USI compared with women with a baseline clinical diagnosis of OAB (22% vs. 8%). This would have the potential to change the management plan to USI surgery according to clinical practice and the study treatment pathways, especially knowing that participants in FUTURE would have failed conservative treatment in the form of supervised PFMT prior to randomisation.
2. Did the clinicians make the treatment decisions based on the urodynamics diagnosis where it was different to the baseline clinical diagnosis?
In total, 65 women were diagnosed with USI in the urodynamics arm. Interestingly, only 20% (n = 13) had a treatment decision to undergo SUI surgery, while 37% (n = 24) had a treatment decision to undergo BoNT-A treatment. The remaining participants had other treatment decisions that were predominantly conservative.
By the final follow-up time point, 54/65 participants had received treatment and only 24% (n = 13) had received SUI surgery, while 39% (n = 21) received BoNT-A and a further 37% received treatment directed to their OAB symptoms, such as medical treatment, PTNS and cysto-distention.
The above clearly shows that among women with refractory OAB/urgency-predominant MUI at baseline who were subsequently diagnosed with USI on urodynamics, only one in five had a treatment decision according to their urodynamics diagnosis. In addition, by the final follow-up time point, only one in four of the participants received surgery for USI. Obviously, there may be more participants awaiting surgery for USI due to the COVID-19-related long waiting lists and/or participants who will undergo such treatments in the future.
Nevertheless, these results clearly indicate that clinicians are heavily influenced by their baseline clinical assessment, and a clear urodynamics diagnosis that contradicts this will not necessarily change their management decision. Similarly, in an observational study within the BUS RCT, 666 women with non-refractory OAB underwent urodynamics. Their results showed that clinicians and patients appeared to be guided in part by the urodynamics diagnosis in selecting treatment options.52
One explanation is the clinician’s awareness of the limitation of urodynamics in detecting DO in women with OAB. They are also fully aware that the accuracy of urodynamics relies on well-calibrated equipment, the experience of investigators and their objective interpretation of subjective parameters. Hence the standardisation of the test is difficult and is affected by the wide variation in staff practice and type of equipment used.61
We assessed the clinical outcomes in the group of participants with baseline clinical diagnosis of MUI which was later changed to USI following the urodynamic assessment (n = 65). The participant-reported success rate at the final follow-up time point was 17% compared to 24% in the full FUTURE cohort (i.e. only one in six participants reported being ‘very much improved or ‘much improved’), rising to 37% (compared to 44% in the full FUTURE cohort) when using the less strict definition of success on PGI-I (inclusion of the response ‘improved’).
Of those who underwent SUI surgery, the participant-reported success rate was available for 11/13 participants, with only one participant meeting the definition of success, that is, ‘very much improved’ or ‘much improved’ (9%). These results show a poor success rate for participants with refractory OAB/urgency-predominant MUI who are diagnosed with USI on urodynamics and subsequently undergo surgical treatment for SUI. In the recent SIMS trial, participant-reported success rates following various mid-urethral slings were 77% at a similar 15-month follow-up time point and a similar definition of success (‘very much improved’/‘much improved’) on PGI-I.106
The main explanation is that participants with a USI diagnosis in FUTURE are a different cohort of women, with baseline refractory OAB symptoms. One possible explanation for poor outcomes following SUI surgery in the participants with a USI diagnosis is that they were primarily due to the failure of urodynamics to show concomitant DO/DOI. If DO/DOI was shown, these participants would have undergone treatment directed to their predominant clinical symptoms (i.e. BoNT-A, SNM or PTNS). However, this explanation is challenged by the fact that in the CCA only arm more participants received BoNT-A treatment [72% (n = 343) vs. 59% (n = 277)]; however, this was not associated with significantly better participant-reported success rates in the CCA only arm.
It is important to highlight the small number of participants who received surgery for SUI in FUTURE making it difficult to draw a robust conclusion.
In FUTURE, participants in the urodynamics arm underwent fewer invasive procedures. However, participant-reported success rates were not superior to CCA only. The BUS study economic evaluation modelling suggested that urodynamics can be a cost-effective diagnostic strategy in women with predominant symptoms of refractory OAB, based on fewer women undergoing invasive treatment in the urodynamics group and a possible small reduction in clinical effectiveness.52 In contrast with this, the FUTURE trial confirmed no reduction in clinical effectiveness in the urodynamics group.
Quality assurance
Concerns had been raised previously in the literature about the impact of the quality of urodynamic studies on the reliability of diagnostic results.99 A ‘think tank’ had considered it was clear that technique affects the quality of a urodynamic test, and with other factors it will affect the utility and perceived value of that test.99
A robust methodology for obtaining a measure of QA was therefore developed for FUTURE to reduce the potential variability of urodynamic assessments. The key elements included developing a robust protocol for urodynamic testing based on the ICS Good Urodynamics Practices (2002) and sharing this protocol with all sites from time of invitation to participate. We evaluated urodynamics traces from all sites prior to participation and shared the assessment and action plan (if required) with them. Throughout the trial, all sites submitted the urodynamics traces/reports that they performed, and the trial office arranged anonymous independent structured central review of randomly collected urodynamic data from each site. Online training and one-to-one training with experts were available and utilised as required. Full details are described in Chapter 7.
However, as a pragmatic effectiveness study, the role of the QA process was not to force departments into compliance with a rigid protocol, but to ensure that the data used for analysis were based on a reasonable level of assurance for its quality yet reflected the standard current clinical practice in the UK. For example, a uniform brand of urodynamics equipment was not required, but the capacity of machines was specified according to the ICS Guidelines on Urodynamic Equipment.
The presence of a well-defined and agreed QA system within FUTURE was key for keeping all collaborating sites aware that urodynamics was regularly monitored in a supportive and constructive way. The response and engagement from our collaborating sites were variable, but as a whole were reasonable, especially given we had over 60 participating sites representing all types of practice in the UK. It was reassuring to see that 95% of pre-trial submitted traces were of acceptable quality to our expert panel. During the study the urodynamics diagnosis was changed for five participants following the QA process and discussion with the local team.
Over 60 sites from all four nations in the UK participated in FUTURE, representing urology, gynaecology, tertiary centres, district general hospitals and teaching hospitals. Our QA system represented a key strength in ensuring the generalisability of the results and enabled the FUTURE trial to ensure a reasonably high-quality urodynamics practice while keeping the ethos of an effectiveness pragmatic study that represents the standard clinical practice in the UK.
Qualitative study
One strength of the FUTURE trial was the embedded qualitative study. The study findings confirm previous findings in this area related to women undergoing urodynamics investigation, not only for investigation of refractory OAB.96–98 Some studies found that women report urodynamic investigations to be uncomfortable or even painful, particularly during the insertion of the catheter into the bladder. However, other research has suggested that the level of discomfort varies widely among women, and some may not experience any discomfort at all.
Additionally, some women have reported feeling embarrassed or anxious during the test, particularly if it was their first time undergoing a urodynamic investigation. External influences include the type of catheter used, positioning during the test, skill of the clinicians, age, medical history and previous experiences of medical investigations. As with the study results, urodynamics is generally considered acceptable and it is important that healthcare staff support women to minimise associated discomfort or anxiety. Crucially, counselling is vital and the voices of women with lived experience included in this report will be extremely valuable in fully informing women about their assessment/investigation to enable them to prepare appropriately.
Cost-effectiveness
Urodynamics is shown to be more costly, principally due to the testing itself and more clinic visits. There is evidence of greater numbers of interventions for SUI in patients undergoing urodynamics, but all other effects are highly uncertain, and not statistically significant. There is no clear evidence of differences in HRQoL (as measured by the EQ-5D-5L) at any time point, nor in total QALYs. The higher mean costs and QALYs lead to urodynamics not being cost-effective at a funding threshold of £20,000 per QALY gained, with only a 34% chance of it being cost-effective. However, this is sensitive to imputation, with the complete case analysis showing a 67% chance of urodynamics being cost-effective. The subgroup analysis suggests larger health benefits for patients with an initial diagnosis of MUI, which is associated with a 72% chance of cost-effectiveness. This is principally due to the higher QALY gains in that participant group compared to participants with an initial diagnosis of OAB.
Further analysis of the effects that the 24-month data have on the results is warranted, as this leads to two methodological uncertainties. First, the pattern of missing data and its impact on costs and outcomes is more complex than is generally the case in RCTs. Second, the incorporation of both 15-month and 24-month data in the estimation of total costs and QALYs is based on a simple, additive specification. Alternative approaches to these issues are possible and need consideration.
Extrapolation of the estimated 24-month results using final treatment designations and long-term success rates taken from the literature reduce the chance of urodynamics being cost-effective further. This finding is driven by the higher rates of ongoing treatment with BoNT-A in the CCA group, to which the model applies favourable EQ-5D-5L values and long-term success rates (as sourced from FUTURE and our literature review, respectively). However, if the long-term success rates for BoNT-A are later found to be reduced, that will impact the long-term cost-effectiveness analysis.
Impact of the coronavirus disease-19 pandemic
The COVID-19 pandemic had an impact on the FUTURE trial in terms of recruitment, data collection and analysis of the trial results.
Recruitment
Recruitment to the FUTURE trial was paused in March 2020 due to the COVID-19 pandemic, at which point 1022 participants had been recruited: 93% of the overall target. Approvals to restart recruitment were obtained in August 2020 and the recruitment end date was extended to 31 January 2021. A further 81 participants (7.3% of the total randomised) were recruited after the pause; 43 to receive urodynamics and 38 to receive CCA only.
Data collection
The primary aim of the FUTURE trial was to assess participant-reported improvement in symptoms following treatment for refractory OAB. However, during the COVID-19 pandemic, routine NHS treatments were suspended, leading to treatment delays. As a result, some FUTURE participants received the 6- and 15-month post-randomisation questionnaires before treatment delivery.
Therefore, for participants whose treatment had been delayed because of the pandemic, an additional 24-month post-randomisation questionnaire was issued. This questionnaire contained the same suite of questions as the 15-month post-randomisation questionnaire and followed the same reminder system. A case-note review was also conducted for these participants at 24 months post randomisation.
A participant was deemed eligible for the additional 24-month questionnaire (i.e. their treatment was delayed) if they were randomised 6 months prior to the recruitment pause (September 2019 to March 2020; n = 207), no intervention (urodynamics or CCA-only) CRF had been completed prior to March 2020 (n = 13) and the site confirmed the participant was on the waiting list for treatment in March 2020 (n = 85). Therefore, 27.8% of the FUTURE cohort received the additional 24-month post-randomisation questionnaire.
Analysis of the trial results
The COVID-19 pandemic impacted on the analysis by delaying assessment and treatment. Therefore, the addition of the 24-month questionnaire benefited the analysis as participants’ final outcomes were at the correct point in their treatment pathway and the number of participants who reported outcome data without having received treatment was reduced.
Another impact of COVID-19 was a higher rate of missing data. The analysis model used was planned in the protocol and pre-COVID-19 drafts of the statistical analysis plan with two additions. A dummy variable indicating a participant received a 24-month follow-up questionnaire and the time from randomisation to all observations used in the repeated-measures mixed model were also included. As recommended in Cro et al. (2020), multiple imputation with chained equations was used to explore the effect of missing data.114 A variable indicating a participant’s assessment or treatment was affected by COVID-19 was included in the imputation model. Sensitivity analyses for the effect of missing data using pattern-mixture modelling were also performed.
The adoption of two different lengths of follow-up poses additional problems for the cost-effectiveness analysis as both costs and QALYs are calculated as cumulative values over the follow-up period. As such, participants who are followed-up for longer necessarily generate higher costs and QALYs. Estimation of incremental costs and QALYs, therefore, needs to account for this through the inclusion of additional terms in the relevant regression analyses. In tandem with the associated patterns of missing data, there is some uncertainty relating to the best specifications for the regression and imputation. As such, further exploration of this is warranted.
COVID-19 also had a specific direct impact on the FUTURE trial in that the subgroup analysis comparing participants who started on BoNT-A to those who started on SNM could not be conducted due to the low number of participants who received SNM, nor was it possible to make the comparison of SNM and BoNT-A on participants whose treatment assessment was made using urodynamics.
Patient and public involvement
Pre-funding application and design of the research
Prior to the initial funding application, we established a collaboration with the largest relevant patient-support group in the UK to provide insights from the patient’s perspective. The Bowel and Bladder Foundation advised on the treatment pathways, proposed assessment tools and outcome measures. However, they ceased operating in July 2016 prior to the funding start date.
Bladder Health UK were therefore approached and agreed to support the research by becoming a co-applicant and a member of the PMG, providing clear leadership on the patient perspective, and were integral to the development of the trial protocol and all the trial documents, including the patient information sheet, letters of invitation/reminders, participant questionnaires and the bladder diary.
Oversight of the study
One of the independent members of the TSC was a patient representative. The TSC met throughout the study and reviewed all study documentation, including patient-facing documents, newsletters and questionnaires that were sent to potential and recruited participants in FUTURE. In addition to being an integral part of the study oversight, they provided the following feedback when reflecting on their experience:
I enjoyed working with clinicians and others to simplify documentation to make it easier to understand, and I was gratified when my suggestions were embraced. I hope my input helped those patients taking part in the trial give proper, informed consent.
Report writing, academic paper preparation and dissemination
The Bladder Health UK representative and the PPI partner on the TSC have contributed towards the preparation of the plain language summary and have been actively involved in discussions of the trial results with the TSC. They will continue to be involved in dissemination activities, including the dissemination of results to the FUTURE participants and academic papers.
Equality, diversity and inclusion
The participants in FUTURE were similar in age to the UK female population with OAB. They were recruited from over 60 sites from all four nations in the UK, representing urology, gynaecology, tertiary sites, district general hospitals and teaching hospitals.
The research team included representatives from Bladder Health UK, the largest bladder-patient-support charity in the UK. They, along with the PPI representative on the TSC, were actively involved throughout the trial contributing to the trial design, development of trial materials and contributing to discussions of the trial results, as well as preparation of the plain language summary.
The trial team represented a broad range of expertise in quantitative and qualitative methodologies. Less experienced members of the team were encouraged to lead discrete components of the study (under senior supervision) during all aspects of the trial, including start-up, recruitment, data collection and data cleaning.
Key strengths
Superiority design
A major issue in designing the trial was that urodynamics has become embedded in standard clinical care without a solid evidence base to support its routine or selective use, as highlighted by the NICE guideline.3 Usually, when designing a study to test a standard of care, a non-inferiority design is used. In such a design it could be argued that if the removal of urodynamics preserved at least x% of the proven benefit of urodynamics, and by removing it the patient journey is made better, and/or safer, and/or cheaper, then the acceptable reduction (the so-called non-inferiority margin, 100 − x%) in the clinical effectiveness can be traded-off against these other benefits. However, if urodynamics has been accepted into practice without demonstrating its actual benefit over clinical assessment only, then it is not possible to set a credible non-inferiority margin. Instead, it seems imperative to go back one step in the evidence base and conduct the superiority-type design that has been missed.
The trial design also included a well-designed and adequately executed QA system, an embedded qualitative study to explore participants’ experiences, and a parallel cost-effectiveness analysis, which are all key strengths of the FUTURE trial.
Large sample size and recruitment to target within reasonable time frame
To our knowledge, FUTURE is the largest RCT worldwide for evaluating the clinical and cost-effectiveness of urodynamics in women.
Independent data monitoring and Trial Steering Committees
The independent DMC and TSC met regularly throughout the study and were instrumental in successfully delivering the trial. The decision to introduce a secondary analysis considering ‘improved’ as success was a key decision supported by both committees.
Key limitations
The majority of participants in FUTURE underwent BoNT-A treatment, which can be seen as a limitation. While it represents the standard clinical practice in the UK, the small number of participants receiving other treatments made pre-planned secondary analyses for outcomes of SNM and PTNS not possible. In addition, we were not able to test the best sequence of treatments.
Follow-up was limited to 15 to 24 months; hence, we lack information on whether women in the CCA only arm would end up having urodynamics at a later stage, which will impact the clinical and cost-effectiveness results. Longer-term follow-up to 5 years is under way. Lastly, the impact of the COVID-19 pandemic led to a number of participants not receiving the intervention and or treatments during the original 15-month follow-up period. However, we were able to mitigate this to a great extent by excellent communication with our collaborating sites and introducing an additional follow-up time point at 24 months for participants who suffered these delays.
Key take-home messages
- In women with refractory OAB/urgency-predominant MUI, the participant-reported success rates following treatment in women who underwent urodynamics and CCA were not superior to those who underwent CCA only [adjusted OR 1.12 (95% CI 0.73 to 1.74); p = 0.601]. Sensitivity analyses and further per protocol analyses showed similar effect sizes and provided confirmation and confidence in the trial results.
- A secondary analysis with the inclusion of ‘improved’ as a successful outcome on the PGI-I had the effect of increasing the number of participants reporting success at the last follow-up time point. However, the effect size was similar to the primary analysis [OR 1.14 (95% CI 0.79 to 1.65); p = 0.469].
- A secondary analysis asking women who underwent BoNT-A treatment to rate their symptoms 2 months after receiving treatment had the effect of increasing the number of participants reporting success; however, the effect sizes remained insignificant [OR 1.17 (95% CI 0.73 to 1.89); p = 0.518].
- A secondary analysis of participant-reported success rates according to the baseline clinical diagnosis of OAB and urgency-predominant MUI, at the final time point, showed no evidence of a significant difference in the effect of urodynamics when using both the original and less strict definition of success [1.14 (99% CI 0.33 to 3.90); p = 0.788 and 1.07 (99% CI 0.39 to 2.95); p = 0.861], respectively.
- Women in the urodynamics arm underwent fewer invasive procedures; however, participant-reported success rates were not superior to those in the CCA only arm.
- Over 60 sites from all four nations in the UK participated in FUTURE, representing urology, gynaecology, tertiary centres, district general hospitals and teaching hospitals. Our QA system represented a key strength in ensuring the generalisability of the results and enabled the FUTURE trial to ensure a reasonably high-quality urodynamics practice while keeping the ethos of an effectiveness pragmatic study that represents the standard clinical practice in the UK.
- The cost-effectiveness data showed that urodynamics had a 34% probability of being cost-effective at £20,000 per QALY gained. This probability reduced further when the results are extrapolated over the patients’ lifetimes. In a subgroup analysis for participants with baseline clinical diagnoses of OAB and urgency-predominant MUI separately, the two groups of participants were found to have similar costs, but women with a baseline diagnosis of MUI had a notably greater gain in QALYs associated with urodynamics (0.053 vs. −0.010). This led to a lower ICER (£8357 per QALY gained) and a commensurately higher probability of being cost-effective at £20,000 per QALY (72%).
- DO/DOI was ‘not’ detected on urodynamic assessment in 34% of women with a clinical diagnosis of refractory OAB at baseline. One in five participants had no evidence of USI or DO on urodynamics. In clinical practice, these women would receive symptomatic treatment according to their CCA diagnosis. Finally, participants with a USI diagnosis had poor participant-reported success rates compared to the rest of the FUTURE cohort. Those who underwent SUI surgery had poor success rates, but the numbers were very low to make robust conclusions in this subgroup.
Conclusion
In women with refractory OAB/urgency-predominant MUI, the participant-reported success rates following treatment in women who undergo urodynamics and CCA are not superior to those who undergo CCA only, up to 15 months post randomisation. Significantly more women who undergo CCA only report earlier improvement in their symptoms. Urodynamics plus CCA is not cost-effective at a threshold of £20,000 per QALY gained.
Future research
- Longer-term outcomes are being assessed up to 5 years to explore further treatments received in both arms and whether women in the CCA only arm undergo urodynamics later and to compare the participant-reported outcomes after longer-term follow-up in both groups.
- It is clear from the results that women with urgency-predominant MUI have different outcomes. Similar findings are reported in the literature with regard to women with a baseline diagnosis of stress-predominant MUI having different outcomes following SUI surgery. Hence future research should assess the clinical effectiveness of urodynamics in women with MUI (both stress- and urgency-predominant MUI).
- Once the results of the extended follow-up are available, a value-of-information analysis should be undertaken by updating the parameters within the cost-effectiveness model in order to help identify those research questions that are most valuable to commissioners.