Shockwave lithotripsy compared with ureteroscopic stone treatment for adults with ureteric stones: the TISU non-inferiority RCT

Article history

The research reported in this issue of the journal was funded by the HTA programme as project number 10/137/01. The contractual start date was in March 2013. The draft report began editorial review in March 2019 and was accepted for publication in January 2020. The authors have been wholly responsible for all data collection, analysis and interpretation, and for writing up their work. The HTA editors and publisher have tried to ensure the accuracy of the authors’ report and would like to thank the reviewers for their constructive comments on the draft document. However, they do not accept liability for damages or losses arising from material published in this report.

Copyright statement

© Queen’s Printer and Controller of HMSO 2022. This work was produced by Dasgupta et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This issue may be freely reproduced for the purposes of private research and study and extracts (or indeed, the full report) may be included in professional journals provided that suitable acknowledgement is made and the reproduction is not associated with any form of advertising. Applications for commercial reproduction should be addressed to: NIHR Journals Library, National Institute for Health Research, Evaluation, Trials and Studies Coordinating Centre, Alpha House, University of Southampton Science Park, Southampton SO16 7NS, UK.

2022 Queen’s Printer and Controller of HMSO

The health problem

Urinary tract stone disease, or urolithiasis, is the formation of stones or calculi in the urinary tract. Urinary tract stone disease is very common, with an estimated lifetime prevalence of 13% in the UK,4,5 and it is more common in men than in women. 6 In the UK the prevalence is rising, with the number of interventions for stone disease also increasing. 7 This increasing prevalence has also been observed in other countries, for example in the USA, where the prevalence is expected to continue to rise. 8

Ureteric stones are crystalline deposits that are originally formed in the collecting part of one or both kidneys that subsequently pass into the ureters. The ureter is the urine drainage tube connecting the kidney to the bladder (Figure 1). Ureteric stones can block the flow of urine and can cause pain (ureteric colic) in the flank, lower abdomen and groin, which is typically severe and recurrent. This sudden severe pain, which is associated with a stone migrating from the kidney into the ureter causing intermittent obstruction, can necessitate urgent attention from a general practitioner (GP) or a hospital emergency department or emergency admission to a urology department.

FIGURE 1.

Anatomy of urinary tract showing definition of ureteric segments. Reproduced with permission from Medindia. 9

Reproduced with permission from Medindia.

Urinary stones often recur and the lifetime recurrence rate is approximately 50%. 10 The interval between recurrences is variable, with approximately 10% recurring within 1 year, 35% within 5 years and 50% within 10 years. 11 The increased incidence of urinary stones in the industrialised world is associated with improved standards of living (mainly due to the high dietary intake of proteins and minerals) and there is also an association with ethnicity and region of residence. 12 Urinary stones affect younger adults (aged 20–55 years) of working age,5 and so have a personal and societal cost owing to working days missed. There is limited evidence on the impact of stone disease on patient quality of life, but patients describe the pain associated with stone disease as one of the worst imaginable. 13

There is an increasing recognition of the rising health burden due to urinary stone disease, with a 63% increase in hospital episodes from 2000 to 2010. 5 There is an increasing trend towards the use of URS as the treatment modality for stone disease. 4 URS is increasingly being used as an emergency intervention, with a 38% increase in emergency URS procedures between 2006 and 2013. 7 This change in clinical practice has occurred despite the lack of evidence of clinical or economic benefit of URS to patients or the NHS.

Treatment options

Most people with a ureteric stone can be expected to pass the stone spontaneously with conservative or supportive care, such as increased fluid intake, pain relief and metabolic expulsive therapy (MET). However, between one-fifth and one-third of cases require an active intervention (i.e. stone removal)14 because of failure to pass the stone, failure of conservative management, continuing pain, infection or obstruction to urine drainage. The two standard active intervention options are SWL and URS. In some cases, a temporising procedure (emergency procedure), such as a ureteric stent or nephrostomy, is needed to treat concurrent infection or obstruction before any active intervention can safely take place.

The size, shape and position of the stone in the ureter (upper, middle or lower; see Figure 1) influences whether or not the person is likely to pass the stone spontaneously and if the person is likely to require an intervention to facilitate stone passage. People with small stones (i.e. < 5 mm) generally do not need an active intervention. 14

The role of MET is unclear given the findings of the Spontaneous Urinary Stone Passage ENabled by Drugs (SUSPEND) trial,15 but current guidelines still advise consideration of alpha-blockers for ureteric stones < 10 mm in size in the lower ureter. The European Association of Urology (EAU’s) urolithiasis 2018 guideline16 advocates the use of either SWL or URS for stones < 10 mm in size, whereas URS is deemed marginally more suitable for stones > 10 mm in size, both in the proximal and in the distal ureters.

The treatment pathways

The treatment pathway for a patient with a stone that is judged clinically to be unlikely to pass spontaneously, will generally start with either SWL or URS.

If the treatment pathway starts with SWL, after the first SWL session the need for a potential second session is reviewed which will depend on progress with stone fragmentation and passage. If a ureteric stone has not been cleared after two sessions of SWL, urologists would generally advocate changing to ureteroscopic clearance.

If treatment starts with URS, the stone can usually successfully be cleared in a single procedure. This may be by direct removal of the entire stone, fragmentation of the stone with removal of the fragments or fragmentation of the stone with the fragments passed in the day(s) following the procedure. In 10–15% of patients,17 however, it is only possible to insert a stent because the ureter is too tight to reach the stone safely with a ureteroscope (these patients require a subsequent procedure to remove the stone and stent). After URS (with or without stone clearance) the surgeon may also insert a temporary stent to allow safe postoperative drainage of the ureter and reduce the risk of postoperative pain. Stent insertion then requires a further procedure for stent removal (usually carried out under local anaesthesia as a day case).

Some patients presenting with ureteric stones as an emergency may have continuing severe pain or evidence of infection or obstruction, and these patients may require urgent drainage of their renal collecting system either through insertion of a ureteric stent or through a nephrostomy (rather than having primary SWL or URS), with definitive treatment postponed to a later date.

The choice between a non-invasive, outpatient-based treatment (SWL) and the more invasive option of URS (requiring anaesthesia) has implications for the NHS and other health-care systems. The combination of technological advances [miniaturisation of ureteroscopes, effective fragmentation with laser (usually holmium) and improved retrieval devices] and increased availability has been reflected in the global trend of increasing URS cases, with a concomitant decline in SWL procedures. 18 Provision of primary treatment in the NHS setting can be affected by resource availability,19 although the safe clearance of a ureteric stone without the need for a stent should be an achievable target in contemporary stone management in a specialised and fully resourced unit.

Members of the British Association of Urological Surgeons (BAUS) Section of Endourology were surveyed at the association’s annual meeting in 2012. The consensus among UK endourologists was that they would accept up to a 20% inferiority level of SWL compared with URS when discussing treatment options with their patients. This level was also considered acceptable by the BAUS Section of Endourology patient group.

Current evidence base

At the time of funding, a Cochrane review2 suggested that URS was associated with better stone clearance rates but higher complication rates than SWL. A more recent systematic review3 supports these findings. A joint EAU–American Urological Association guideline for ureteric stones,20 current at the time of funding, had similar findings, but the evidence at that time was deemed insufficient to recommend either SWL or URS as the first-line treatment. One of the major conclusions from these publications was to suggest the need for large, multicentre RCTs to compare these modalities.

We describe the TISU trial, which was a large, multicentre RCT. All participants had ureteric stones diagnosed and confirmed by the contemporaneous use of computed tomography scan of the kidneys, ureters and bladder (CTKUB) for stone location and size, and were clinically judged to need active intervention (usually due to failure of the stone to progress, failure of conservative treatment, continuing pain or the size and position of the stone). Only centres with an established fixed-site lithotripter (rather than mobile machines) and with the ability to perform ureteroscopic procedures were able to recruit patients to the study. A key feature of the TISU trial design was that treatment would follow established usual NHS clinical pathways (i.e. both SWL and URS must be available as treatment options to the patient). Assessment of the outcome measure of resolution of the stone episode (stone clearance) was based on the need for any further procedures (i.e. additional to those in the treatment pathway to which the participants were initially randomised) up to 6 months after random allocation. When no further procedures took place, the participant was deemed to be stone free from their definitive initial treatment pathway.

Trial objectives

The TISU trial was a multicentre, non-inferiority RCT of SWL as the initial treatment option (vs. URS) for ureteric stones, in a UK NHS setting.

The aim of the TISU trial was to determine the clinical effectiveness and cost-effectiveness of SWL as the initial treatment option compared with primary URS.

The clinical effectiveness and cost-effectiveness was determined with respect to:

• resolution of stone episode (stone clearance), defined as no further intervention required to facilitate stone passage

• incremental cost per quality-adjusted life-year (QALY)

• participant-reported health outcomes

• treatment-related harms up to 6 months post randomisation.

Research ethics and research governance approvals

The TISU trial was given a favourable opinion prior to approaching any potential participants by the North of Scotland Research Ethics Committee 1 (reference number 13/NS/0002). It was approved by the sponsors (NHS Grampian and University of Aberdeen) and by the research and development departments of the NHS organisations at each site that took part prior to approaching any potential participant at that site. The trial was conducted in accordance with the principles of Good Clinical Practice, was registered on the UK Clinical Research Network Portfolio (study identification 13979) and was assigned an International Standard Randomised Clinical Trial Number (ISRCTN92289221). A site initiation visit took place at each site prior to starting recruitment. At the site initiation visit, the trial manager detailed and explained trial procedures to the local principal investigator (PI) and clinical research team and provided a trial-specific site file.

Participants

Potential participants were adults presenting with a diagnosis of a unilateral ureteric stone in any segment of the ureter at participating UK NHS hospitals, and were identified according to the inclusion and exclusion specified, as follows.

Identifying participants and consent

Local procedures at participating hospitals were different. The timing and mode of approach to patients and the consent process varied to accommodate both the variability at a site and the needs of the patients. Following adequate pain relief and confirmation of their ureteric stone by CTKUB, eligible patients (according to the criteria in Inclusion criteria and Exclusion criteria) were provided with a patient information leaflet (see Report Supplementary Material 1).

Each eligible patient was given the opportunity to discuss the trial with the local clinical team. Eligible patients could decide to participate during a consultation with the local clinical team, during a visit to hospital (e.g. when they attended a clinic appointment or while a patient was in hospital for their initial stone episode) or, alternatively, after consideration of the patient information leaflet at home. Some patients who agreed to be contacted at home may have been called by the local research nurse to discuss any further queries. Patients who decided to participate following telephone consultation sent their completed documents (consent form and baseline questionnaire, see Report Supplementary Material 1) through the post to the local team at their treating hospital or were told to take the documents with them if they returned to hospital for another consultation or treatment.

Signed informed consent forms were obtained from the participants in all centres. Participants who could not give informed consent (e.g. due to incapacity) were not eligible for participation. The participant’s permission was sought to inform their GP that they were taking part in the TISU trial. Patients were randomised to one of the two treatment arms following consent.

Randomisation

Participants were allocated to one of the two intervention arms: SWL or URS.

The randomisation algorithm used trial centre (site), stone size (≤ 10 mm or > 10 mm) and stone location (upper, middle or lower ureter, defined in the EAU urolithiasis guideline16) as minimisation covariates and 1 : 1 allocation was used. A web-based application or a remote telephone interactive voice-response randomisation application, both hosted by the Centre for Healthcare Randomised Trials (CHaRT), Health Services Research Unit (HSRU), at the University of Aberdeen, was used to carried out randomisation.

Trial interventions

We were investigating the care pathways that started with one of the interventions (SWL or URS).

Shockwave lithotripsy involves generation of a shockwave that is focused on the stone and causes it to fragment, with the fragments subsequently passing spontaneously. It is routinely performed in an outpatient setting, with pain relief or sedation as required. Recruitment took place only in established UK centres with fixed-site lithotripters. This allowed some standardisation of pathways on times to treatment and SWL delivery. Up to two sessions of SWL were considered as ‘one intervention’, as per standard practice (usually the second session is delivered within 2–4 weeks of the first). Details of the make and model of lithotripters used and standard site-specific lithotripsy treatment protocols can be found in Appendix 1, Table 16.

Ureteroscopic stone treatment is the use of a small semi-rigid or flexible ureteroscope, in conjunction with intracorporeal lithotripsy devices, such as the holmium laser, to directly visualise and fragment ureteric stones. Smaller stones in the lower ureter can occasionally be removed intact by using basketing devices. It is currently most often performed as a day-case procedure [but may require hospital admission depending on complexity (2014 NHS average = 1.7 days23)] and usually necessitates general anaesthesia.

Outcome measures

Resolution of the stone episode

The primary clinical outcome was defined according to the treatment care pathway (Figure 4). The outcome was derived from several fields from case report forms (CRFs) at 8 weeks and 6 months post randomisation (Table 1; see also Data collection). This was checked against treatment CRFs and any supplementary CRFs that were completed between the 8-week and the 6-month CRFs (see Data collection).

FIGURE 4.

Treatment care pathway to define the primary outcome.

TABLE 1 - Measurement of outcomes (components and timing)

NRS, Numeric Rating Scale; PQ, participant questionnaire.

See Report Supplementary Material 1.

Note

This table has been reproduced with permission from McClinton et al. 1 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The table includes minor additions and formatting changes to the original table.

Outcome measure	Source	Timing
		Baseline	Intervention		Post randomisation
			Pre	1 week post	8 weeks	6 months
Interventions received	CRF and PQa				✓	✓
Health status: EQ-5D-3L	PQ	✓	✓	✓	✓	✓
Health profile: SF-12	PQ	✓			✓	✓
Pain: NRS	PQ	✓	✓	✓	✓
Use of analgesics	PQ	✓	✓	✓	✓
Complications	CRF				✓	✓
NHS primary and secondary health-care use	CRF and PQ					✓

Data collection

Original sample size

The original sample size calculations reflect that the TISU trial was a non-inferiority design. Published literature2 suggested that the proportion of participants who were stone free without further intervention up to 6 months would be about 0.75 in the URS arm (P1) and about 0.65 in the SWL arm (P2). A survey of members of the BAUS Section of Endourology was carried out at the association’s annual meeting in 2012. The consensus among UK endourologists was that, when discussing treatment options with their patients, they would continue to recommend SWL as long as its inferiority level compared with URS was no more than 20%. This level was also agreed by the BAUS Section of Endourology patient group as being acceptable to them. The margin of inferiority deemed acceptable was, therefore, set at 0.20, so that P2 – P1 > –0.20. The sample size was estimated using simulations. The power of a non-inferiority trial can be considered as the probability that the lower bound of the estimated confidence interval (CI) around the difference between trial proportions excludes the margin of non-inferiority. Simulating thousands of trials of fixed sizes with the parameters P1 and P2, as above, indicated that a trial of 450 participants per arm was required for the lower bound of the estimated 95% CI to exclude –20%, with 90% power. Adjustment for potential of 10% dropout inflated the number of participants needed to 1000 in total. A trial of this size has 90% power to test superiority on secondary outcomes of an effect size of one-quarter of 1 SD.

Sample size reassessment

Following slower than planned recruitment, our funders requested a reassessment of the assumptions of sample calculation. We did this by looking at trial-aggregated primary outcome from 267 participants and in discussion with our independent DMC, and subsequently the funders. The sample size was amended downwards from 1000 to 750. The amendment was ratified by the trial oversight committees, the sponsor and the funder. Recruitment projections showed us that the original sample size of 1000 participants was unachievable in a realistic time frame, despite measures implemented to improve recruitment. We agreed with the funder that an extension of 18 months would be required to reach a revised sample size of 750 and that this was an achievable target. Our original sample size of 1000 included a 15% uplift from 850 to enable the primary analysis to be a suitably defined per-protocol (PP) analysis, as this approach, in the special context of a non-inferiority design, is often seen as more conservative than the more conventional intention-to-treat (ITT) approach. However, the view of the Health Technology Assessment Board, and which was confirmed by the TISU trial DMC, was that the ITT approach should be preferred over the PP approach as the former would better reflect the TISU trial’s pragmatic effectiveness focus, that is, the fact that its aim was to evaluate the policy of initiating one or other of these treatment options rather than to compare their relative performance. The PP analysis was used as a supporting analysis. Based on the 267 participants with mature outcome data (as of 16 February 2016), the DMC observed that all of the assumptions behind the power calculation remained plausible. Given that the PP analysis was deemed of secondary importance (and under the original design of assuming that the proportion of participants who were stone free would be 65% in the SWL arm and 75% in the URS arm), the achievable sample size of 750 gave 85% power.

Statistical analysis

Economic evaluation

Economic evaluation was an integral part of the TISU trial. The evaluation considered the costs of the care pathways that patients had received. Resource data collected included the costs of the interventions, SWL and URS, and simultaneous and consequent use of primary and secondary NHS services (including additional interventions received) by participants. See Chapter 4 for a detailed description of the methods used.

Management of the study

The Trial Management Team was centralised, the study office was based within CHaRT, University of Aberdeen, and it provided real-time support for the recruiting centres. Recruiting centres were led by local PIs (urologists) responsible for all aspects of the trial, including recruitment and consent of participants, delivery of interventions and notification of SAEs and breaches. PIs were supported by local research nurses.

The trial was supervised by the Project Management Group (PMG). This group consisted of grant holders and representatives from the study office.

The trial was overseen by a TSC comprising four independent members and by an independent DMC. Both committees met on a yearly basis, and the DMC always met before the TSC and would provide any recommendations to the committee. The DMC did request additional meetings to discuss slow recruitment and to discuss interim analysis conducted during the trial.

Patient and public involvement

Participant baseline characteristics

Trial recruitment

In total, 613 (out of 1291 eligible) participants were recruited from 25 centres and the median number of participants per centre was 21 (IQR 16–27). We randomised 306 participants to the SWL care pathway arm and 307 participants to the URS care pathway arm. Participants were recruited between July 2013 and June 2017, and final follow-up was to December 2017 (Figure 5).

FIGURE 5.

Recruitment graph.

Participant flow

Figure 6 shows the Consolidated Standards of Reporting Trials (CONSORT) flow diagram for the TISU trial. We screened 3209 potentially eligible patients, of whom 1918 (60%) were ineligible. The main reasons for ineligibility were that the patient was not suitable for either SWL or URS and the patient presented with abnormal urinary tract anatomy or bilateral stones. We randomised 613 out of 1291 (47.5%) eligible participants. The reasons for not being randomised were that the patient had a preference for one treatment and did not want to be randomised, clinician had a preference for a treatment and, in some cases, patients declined to give a reason. Full details of the reasons for patients being ineligible or declining are tabulated in Appendix 2, Table 17. There were four post-randomisation exclusions [three patients in the SWL arm (one patient on warfarin, one patient not fit for treatment and one patient with bilateral stones) and one patient in the URS arm (patient on cardiac medication)], leaving 303 and 306 participants, respectively.

FIGURE 6.

The CONSORT flow diagram. a, 86 had two sessions of SWL; b, 3 had two sessions of SWL; c, other = unknown as not treated within the NHS, did not attend, unable to treat. This figure has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original figure.

This figure has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original figure.

Baseline characteristics

The two arms were well balanced at baseline (Table 2). The majority of participants were men. The age distribution was similar in both study arms, with mean ages in the early fifties. Over 95% of participants had a stone size ≤ 10 mm: 45% of stones were in the upper ureter and 38% in the lower ureter. Most participants had experienced pain related to their ureteric stone in the previous 7 days and had taken pain medication.

Care pathway and treatment received

The care pathways are outlined in Chapter 2 (see Figure 4). Table 3 describes the allocated care pathway compared with actual treatment received in the TISU trial. There were 210 (69.3%) participants in the SWL pathway and 250 (81.7%) participants in the URS pathway who were treated as allocated. There were 37 (12%) participants who were allocated to the SWL care pathway but received URS as their treatment. The reasons provided were medical (n = 5), participant choice (n = 3) or stone not visible on pre-SWL imaging (n = 10) or no reason was given (n = 18). In the URS care pathway, 12 (4%) participants received SWL as their treatment. The reasons provided were medical (n = 2) or participant choice (n = 6), or no reason was given (n = 4). In 52 (17%) SWL arm participants, no stone was visible on pre-treatment imaging and a decision was made not to proceed with the SWL because of evidence that the stone had passed (no visible stone and cessation of symptoms). This was also the case for 36 (12%) participants in the URS arm. Overall, seven participants (three SWL participants and four URS participants) did not attend for treatment, and surgeons were unable to treat one participant in the SWL arm owing to the position of their stone. Four participants had their treatment outside the TISU trial and we have not been able to establish what treatments they received. Waiting times are described in Table 4. More than 90% of participants in the SWL care pathway received treatment within 8 weeks, with a slightly lower proportion (86%) of participants in the URS care pathway being treated within 8 weeks (excluding those who passed their stone before treatment in both pathways).

Prior to treatment, participants attended a pre-intervention appointment and were asked to complete questionnaires again, to assess pain and quality of life (Table 5).

Primary outcome

We analysed the primary outcome initially for all participants using an ITT approach (that is analysing participants as they were randomised, regardless of whether or not they passed their stone before intervention or of the intervention received) (Table 6). In the SWL arm, 67 out of 302 (22.2%) participants needed further treatment. In the URS arm, 31 out of 302 (10.3%) participants needed further treatment. The ARD was 11.4% (95% CI 5.0% to 17.8%); the upper bound of the 95% CI ruled out the prespecified margin of non-inferiority (which was 20%) (Figure 7).

FIGURE 7.

Plot of treatment effects for primary outcome. This figure has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original figure.

This figure has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original figure.

We explored the moderating treatment effect using subgroup analyses of stone size, stone location and gender. Within-subgroup treatment effects are summarised in Figure 8 and were fairly homogeneous across all strata; there was no evidence that subgroup moderated treatment effects. The full subgroup models are summarised (see Appendix 3, Tables 18–20, for details).

FIGURE 8.

Forest plot of subgroup treatment effects. This figure has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original figure.

This figure has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original figure.

Secondary outcomes

Patient-reported outcomes

For both the pain measures and the acceptability measure we found a similar pattern between care pathways. At 8 weeks, self-reported pain was low in both arms (Table 8). The participants in the SWL arm reported taking pain relief more frequently, but the number of days reported as requiring pain relief was low in both arms; the median number of days reported was zero. Of those who responded, > 80% in each arm stated that they would recommend their treatment to a friend with ureteric stones, and there was no evidence that this differed between arms.

TABLE 8 - Pain at 8 weeks

All treatment effect estimates adjusted for outcome at baseline (when relevant): stone size, stone location, age, gender and centre.

Mean difference between SWL and URS.

Incident rate ratio from negative binomial regression model.

Relative risk.

Note

This table has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The table includes minor additions and formatting changes to the original table.

Patient-reported measurement	SWL	URS	Effect sizea	95% CI
Pain today
n	183	184
Median (IQR)	0 (0–2)	0 (0–1)
Mean (SD)	1.3 (2.4)	0.97 (2.04)	0.3b	–0.2 to 0.9
Days with pain relief over the last 7 days
n	178	181
Median (IQR)	0 (0–2)	0 (0–1)
Mean (SD)	1.5 (2.5)	1.0 (1.9)	1.42c	0.96 to 2.11
Recommend to a friend n/N (%)	148/171 (86.6)	142/171 (83.0)	1.04d	0.97 to 1.13

Quality of life, as measured by the SF-12 PCS and MCS components, is reported in Table 9 and in Figures 9 and 10. In both arms of the trial, and on both measures, quality of life improved over the duration of the trial, from baseline to 6 months. When we used observed data only, there were small but consistent effects favouring SWL for both SF-12 PCS and MCS. However, these effects were attenuated when we used multiple imputation models.

TABLE 9 - Quality of life

All treatment effect estimates adjusted for outcome at baseline: stone size, stone location, age, gender and centre.

Multilevel model for normally distributed outcome to account for repeated time points.

Multiple imputation process included the primary outcome and was conducted on each treatment group separately.

Note

This table has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The table includes minor additions and formatting changes to the original table.

Variable	SWL (N = 303)		URS (N = 306)		Estimatea,b (95% CI); p-value	Imputed estimatea,b,c (95% CI); p-value
	n	Mean (SD)	n	Mean (SD)
SF-12 PCS
Baseline	290	43.5 (9.5)	289	44.8 (9.7)
8 weeks	150	47.0 (10.1)	156	47.9 (9.2)	–0.1 (–2.0 to 1.8); 0.95	–0.82 (–2.70 to 1.05); 0.35
6 months	137	48.0 (10.5)	146	50.9 (8.8)	–1.7 (–3.7 to 0.2); 0.080	–0.90 (–2.70 to 1.05); 0.35
SF-12 MCS
Baseline	290	48.5 (11.1)	289	50.4 (9.6)
8 weeks	150	48.9 (12.4)	156	51.4 (9.9)	–2.2 (–4.4 to –0.01); 0.056	–1.84 (–3.93 to 0.26); 0.09
6 months	137	50.3 (11.6)	146	52.0 (10.4)	–1.1 (–3.4 to 1.2); 0.33	–1.68 (–3.78 to 0.42); 0.12

FIGURE 9.

Short Form questionnaire-12 items PCS over time. This figure has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original figure.

This figure has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original figure.

FIGURE 10.

Short Form questionnaire-12 items MCS over time. This figure has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original figure.

This figure has been reproduced with permission from Dasgupta et al. 21 This is an Open Access article distributed in accordance with the terms of the Creative Commons Attribution (CC BY 4.0) license, which permits others to distribute, remix, adapt and build upon this work, for commercial use, provided the original work is properly cited. See: https://creativecommons.org/licenses/by/4.0/. The figure includes minor additions and formatting changes to the original figure.

Economics methods

We estimated resource use and costs for each participant, and our evaluation considered the costs of the care pathways that patients received. Data collected on resource use included the intervention, SWL or URS, and simultaneous and consequent use of primary and secondary NHS services (including additional interventions received) by participants. The personal costs that were collected were purchase of medications, particularly analgesics, and visits to private health-care professionals.

The economic evaluation followed the reference case of the National Institute for Health and Care Excellence’s (NICE’s)29 recommendations for economic evaluations alongside a clinical trial. 30 The study adopted the perspective of the NHS, although some personal resource data were collected from the participants. We did not use any discounting because participants were followed up for only 6 months. The cost year utilised was 2017/18 and the currency used was Great British pounds (GBP).

Quality of life

The EQ-5D-3L24 and the SF-1226 were used to measure generic health-related quality of life (HRQoL) and health status. Participants were asked to complete the EQ-5D-3L at baseline (after informed consent but before randomisation), directly prior to treatment (pre intervention), 1 week after intervention/treatment, and at 8 weeks and 6 months post randomisation. The EQ–5D-3L divides health status into five dimensions with three levels of severity. EQ-5D-3L questionnaire responses were transformed into utility values using general population time trade-off-generated preference weights. 34

Quality-adjusted life-years were calculated by multiplying quality (utility) and length of life, assuming linear extrapolation between measurement time points. For each patient, the area under the curve (AUC) was used to estimate QALYs gained (quality of life multiplied by duration of the trial). Calculation of the AUC took into account the length of time that the patient waited for treatment. Information on the time between randomisation and treatment was incorporated into the QALY calculation, when it was available. Calculation of QALYs gained in the case of those for whom a treatment date was missing was based on the post-randomisation time points.

Responses from the SF-12 questionnaire collected at baseline and at 8 weeks and 6 months post randomisation were also used to estimate QALYs. They were mapped onto the existing Short Form questionnaire-6 Dimensions (SF-6D) measure, using the algorithm by Brazier et al.,35 to allow utility values to be estimated for each time point. These utility scores were transformed into QALYs using the methods described above, to provide an alternative measure of QALYs gained for each participant.

Economics results

Quality-adjusted life-years

Table 14 shows the EQ-5D-3L, VAS and SF-6D utility scores for each care pathway at different time points. The baseline utility scores were all similar. The EQ-5D-3L utility scores pre treatment and at 8 weeks and 6 months post randomisation were higher for URS than for SWL. The mean estimated QALYs gained were 0.411 (SD 0.112) for the SWL pathway and 0.439 (SD 0.070) for the URS pathway. The adjusted mean QALY difference for the SWL care pathway was –0.029 (95% CI –0.062 to 0.005). The VAS scores were higher in the URS arm at each time point, but the differences were small. The mean estimated QALYs for SF-6D utility scores were 0.393 (SD 0.075) for the SWL arm and 0.400 (SD 0.064) for the URS arm. The adjusted mean QALY difference (SWL vs. URS) was –0.009 (95% CI –0.036 to 0.018). The QALY results should be interpreted with caution, taking into account the fact that the proportion of missing data was high.

TABLE 14 - Quality-of-life measures

EQ-5D, EuroQol-5 Dimensions.

Differences based on regression model adjusting for baseline EQ-5D-3L and minimisation variables [trial centre (site), stone size (≤ 10 mm or > 10 mm), stone location (upper, middle and lower ureter), age and gender].

The maximum the QALY value can be is 0.5, as it is measured over a 6-month period.

Measure	SWL			URS			Difference SWL vs. URSa
	n	Mean	SD	n	Mean	SD	Mean	95% CI
EQ-5D-3L
Baseline	298	0.737	0.263	297	0.729	0.303
Pre treatment	252	0.735	0.260	211	0.758	0.272	–0.041	–0.085 to 0.002
1 week post treatment	186	0.756	0.267	175	0.757	0.263	–0.007	–0.068 to 0.055
8 weeks post randomisation	149	0.797	0.293	152	0.874	0.207	–0.081	–0.152 to –0.009
6 months post randomisation	130	0.837	0.289	143	0.912	0.182	–0.081	–0.146 to –0.016
QALYb	70	0.407	0.116	74	0.436	0.070	–0.029	–0.062 to 0.005
EQ-5D VAS
Baseline	282	68	24	283	67	27
Pre treatment	235	69	25	198	74	22	–4	–7 to –1
1 week post treatment	180	74	22	172	74	20	–1	–7 to 5
8 weeks post randomisation	150	77	21	153	79	21	–4	–9 to 1
6 months post randomisation	131	78	21	143	81	18	–3	–9 to 3
SF-6D
Baseline	195	0.699	0.168	193	0.737	0.175
8 weeks post randomisation	106	0.762	0.169	107	0.782	0.151	–0.003	–0.047 to 0.040
6 months post randomisation	104	0.789	0.173	109	0.837	0.139	–0.069	–0.123 to –0.015
QALYb	45	0.393	0.075	51	0.400	0.064	–0.009	–0.036 to 0.018

There was an increase in all (EQ-5D-3L, VAS and SF-6D) scores over time, as illustrated by Figures 11–13.

FIGURE 11.

EuroQol-5 Dimensions, three-level version scores at each time point by arm. EQ-5D, EuroQol-5 Dimensions.

FIGURE 12.

EuroQol-5 Dimensions VAS scores at each time point by arm. EQ-5D, EuroQol-5 Dimensions.

FIGURE 13.

Short Form questionnaire-6 Dimensions scores at each time point by arm.

Cost-effectiveness analysis

The results of the base-case analysis are reported in Table 15. The base-case analysis (using multiple imputation) showed that the mean cost for participants on the SWL care pathway was £809 less than that for those on the URS care pathway, but resulted in a QALY gain 0.021 lower than the URS care pathway. The point estimate of the incremental cost per QALY is a cost saving of £39,118 per QALY lost and the uncertainty around this estimate is illustrated in Figures 14 and 15. This means that a decision-maker would save £39,118 for each lost QALY, with 79% probability that SWL would be considered to be cost-effective at society’s WTP £30,000 for a QALY. The CEAC is derived from the joint distribution of incremental costs and incremental effects. Most of the results fall in the south-west quadrant of the cost effectiveness plane as SWL always costs less but is less effective.

TABLE 15 - Incremental cost-effectiveness: NHS perspective

Differences based on GLM adjusting for baseline EQ-5D-3L and minimisation variables [trial centre (site), stone size (≤ 10 mm or > 10 mm), stone location (upper, middle and lower ureter), age and gender].

URS costs more and is more effective than SWL.

URS would not be considered to be cost-effective at the NICE recommended threshold of £20,000–30,000 per QALY.

Intervention	Cost (£)	Differencea (£)	QALYs	Differencea	ICER £/QALY	Probability of being cost-effective at different WTP thresholds (%)
						£0	£10,000	£20,000	£30,000	£50,000
Base-case analysis imputed data analysis
SWL	1790		0.403			1	1	0.98	0.79	0.25
URS	2599	809b	0.424	0.021b	39,118	0	0	0.02	0.21	0.75
Complete-case analysis
SWL	1584		0.407			1	1	0.96	0.80	0.46
URS	2932	1348	0.436	0.029	46,297c	0	0	0.04	0.20	0.52
Using SF-6D utility scores imputed data
SWL	1790		0.385			1	1	1	1	1
URS	2599	809	0.387	0.002	432,432c	0	0	0	0	0
Complete case using SF-6D utility scores
SWL	2102		0.388			0.81	0.76	0.70	0.64	0.53
URS	2502	500	0.398	0.010	52,313c	0.20	0.24	0.30	0.36	0.47
Assuming all patients with missing EQ-5D-3L 6-month scores are in full health at 6 months as stones have passed
SWL	1790		0.423			1	1	1	0.95	0.65
URS	2599	795	0.437	0.014	57,889	0	0	0	0.05	0.35
Higher cost of SWL assuming 25% of patients are inpatient
SWL	1952		0.403			1	1	0.90	0.58	0.13
URS	2614	663	0.424	0.021	32,034	0	0	0.10	0.42	0.87
Higher cost of SWL assuming 50% of patients are inpatient
SWL	2073		0.403			1	0.98	0.77	0.42	0.08
URS	2627	555	0.424	0.021	26,820	0	0.02	0.23	0.58	0.92
Higher cost of SWL assuming 75% of patients are inpatient
SWL	2190		0.403			1	0.94	0.60	0.23	0.04
URS	2642	453	0.424	0.021	21,888	0	0.06	0.40	0.77	0.96
Higher cost of SWL assuming 100% of patients are inpatient
SWL	2306		0.403			0.99	0.81	0.38	0.13	0.02
URS	2652	346	0.424	0.021	16,710	0.01	0.19	0.62	0.87	0.98

FIGURE 14.

Cost-effectiveness acceptability curve: SWL vs. URS – base-case analysis based on multiple imputation data.

FIGURE 15.

Scatterplot of incremental cost and incremental QALYs (imputed data): SWL vs. URS.

Figure 15 shows that a large number of bootstrapped iterations lie to the left of the vertical axis and below the horizontal axis, indicating that the SWL is less costly than URS, with marginally fewer QALYs achieved throughout the intervention.

Sensitivity analysis

Complete-case analysis

The results of the complete-case analysis are reported in Table 15 and Figures 16 and 17. The cost results were in the same direction as in the base-case analysis: SWL cost less but the cost difference was higher [£1348 (complete case) vs. £809 (imputed data)]. The QALY difference was similar (0.029). The ICER is higher (£46,297) than the base-case analysis and the probability that SWL is cost-effective is 80% at a threshold of society’s WTP for a QALY of £30,000.

FIGURE 16.

Cost-effectiveness acceptability curve for sensitivity analysis based on complete-case data: SWL vs. URS.

FIGURE 17.

Scatterplot of incremental costs and QALYs based on complete-case data: SWL vs. URS.

Imputation analysis using Short Form questionnaire-6 Dimensions data

The results of the sensitivity analysis using imputed SF-6D data are reported in Table 15 and Figures 18 and 19. The direction of the difference in cost result remained the same as in the base-case analysis: SWL cost less (£809) than URS (£1790 vs. £2599). The direction of the QALY difference was the same as in the base case (lower), but the magnitude changed. The lower difference in quality-of-life estimates (0.002) meant that the ICER increased to £432,432. The probability that society will be willing to pay for a QALY loss was 100% if the threshold of society’s WTP for a QALY is £30,000.

FIGURE 18.

Cost-effectiveness acceptability curve for sensitivity analysis based on SF-6D QALY multiple imputation data: SWL vs. URS.

FIGURE 19.

Scatterplot of incremental costs and incremental QALYs based on imputation data: SWL vs. URS.

The results for the complete-case analysis using complete-case SF-6D data are reported in Table 15 and Figures 20 and 21. These results should be interpreted cautiously, as the number of patients with complete data (costs and SF-6D QALYs) was very low [138/609 (23%)]. The cost difference was lower than in the base case (£500 vs. £809) and the QALY difference was also lower than in the base-case analysis (0.010 vs. 0.021), and this increased the ICER to £52,313. The probability that SWL would be considered cost-effective was 64% at a threshold of society’s WTP of £30,000. The scatterplot (see Figure 21) illustrates the uncertainty, as the estimates are distributed over all quadrants of the cost-effectiveness plane.

FIGURE 20.

Cost-effectiveness acceptability curve for SF-6D QALYs complete-case data: SWL vs. URS.

FIGURE 21.

Scatterplot of incremental costs and QALYs based on SF-6D QALYs complete-case data: SWL vs. URS.

Best-case scenario for missing EQ-5D-3L 6-month data (assuming that all the participants have perfect health)

These results varied from the base-case analysis, as the QALY difference between the two arms reduced from 0.021 to 0.014 (base-case analysis) (see Table 15). The cost difference remained the same as in the base-case analysis (£795). The smaller QALY difference led to an increase in the ICER to £57,899. The probability that SWL would be considered cost-effective at the £30,000 WTP threshold was 95% (Figures 22 and 23).

FIGURE 22.

Cost-effectiveness acceptability curve assuming all missing EQ-5D-3L 6-month scores are full health: SWL vs. URS.

FIGURE 23.

Scatterplot of incremental costs and QALYs assuming all missing EQ-5D-3L 6-month scores are full health: SWL vs. URS.

Four scenarios of higher cost of shockwave lithotripsy imputed data

The cost-effectiveness of SWL when different proportions of patients undergo the procedure as an elective inpatient is reported in Table 15 (see also Figures 24, 25 and 28–33). Figures 24 and 25 present the results of the worst-case scenario, that is when 100% of patients receive SWL as an elective inpatient procedure (further results are in Appendix 4). The cost difference was smaller than that in the base-case analysis (£346 vs. £809), whereas the QALY difference was the same, 0.021. The ICER was £16,710. The smaller cost difference leads to a reduction in the probability that SWL would be considered cost-effective to 13%, if society’s WTP for a QALY is £30,000. The scenario ICERs range from £17,000 to £32,000.

FIGURE 24.

Cost-effectiveness acceptability curve for higher cost of SWL using EQ-5D-3L and imputed data: SWL vs. URS.

FIGURE 25.

Scatterplot of incremental costs and QALYs for higher cost of SWL: SWL vs. URS.

Higher cost shockwave lithotripsy complete-case analysis

The results of the complete-case analysis of higher costs for SWL (see Table 15 and Figures 26 and 27) were in the same direction as in the base-case analysis: SWL cost less and resulted in a lower QALY gain; however, the magnitude of the cost difference was higher [£818 vs. £809 (base case)]. This resulted in an ICER of £29,434 and a probability of 51% that SWL would be considered cost-effective at society’s threshold of WTP for a QALY of £30,000.

FIGURE 26.

Cost-effectiveness acceptability curve for higher cost of SWL using EQ-5D-3L QALYs and complete-case data: SWL vs. URS.

FIGURE 27.

Scatterplot of incremental costs and QALYs for higher cost of SWL: SWL vs. URS.

Economics discussion

To our knowledge, this is the first economic evaluation that performs a direct comparison of clinical pathways that start with SWL and URS in a RCT setting. The results suggest that SWL, on average, costs less, but results in a lower QALY gain, than URS. The difference in cost is mainly driven by the unit intervention cost of URS, which is, on average, four times higher than the unit cost of SWL. Additionally, more participants who received URS had a stent inserted, leading to more stent removal procedures in the URS arm. The unit cost of stent insertion and removal is twice that of SWL. The initial cost of endoscopic stent insertion during the URS procedure was not costed separately, as it was included in the overall cost of URS. Although, on average, participants who followed the pathway starting with SWL had greater subsequent resource use, the cost was not high enough to offset the initial high cost of URS and the stent removal costs that URS care pathway participants incur.

The direction of difference in utility scores derived from both the EQ-5D-3L and the SF-6D was the same; however, the magnitude of the difference was higher for the EQ-5D-3L scores than for the SF-6D scores. On average, SWL resulted in fewer additional QALYs than URS; however, although the QALY difference based on the EQ-5D-3L scores was statistically significant, the QALY difference based on SF-6D scores was not. When interpreting results, it should be noted that there were more missing SF-12 data than missing EQ-5D-3L data. Overall, participants’ utility scores increased over time.

One of the limitations of the study was the low return and low completion rates of the participant questionnaires. This was addressed by imputing the missing data. The overall conclusions were sensitive to the data that were used and the assumptions that were made. On average, SWL cost less than URS and resulted in a smaller QALY gain than URS. The base-case analysis results indicated that SWL had a 79% chance of being considered cost-effective at a WTP threshold of £30,000, whereas the complete-case analysis indicated that SWL had an 80% chance of being considered cost-effective at the same threshold. Similar results were noted in the SF-6D sensitivity analysis, which explored the effect of the higher cost of SWL. However, the results of the assumption surrounding the cost of SWL suggested that the higher the cost of SWL (the more people who received SWL as an elective inpatient procedure), the lower the probability that SWL would be considered cost-effective at the different WTP thresholds. The ICERs of the different analyses ranged from £16,710 to £432,432.

The cost results of the TISU trial are similar to those which have been published in a similar setting. 40,41 In these studies40,41 SWL was the primary treatment for stones between 10 mm and 20 mm in size. SWL was undertaken in an outpatient setting and patients did not receive sedation or anaesthesia, but had analgesia according to their tolerance. Flexible ureteroscopy was performed under general anaesthesia. The authors of both studies concluded that SWL, on average, is cheaper than URS. A recent systematic review18 reported that URS was more cost-effective than SWL for stone treatment. However, the measurement of cost varied across studies and the cost of procedures also varied between health-care systems. The systematic review was based on retrospective case series and the authors reported that the evidence base was poor and suggested that there is a need for large RCTs. The Lotan et al. 42 study performed cost-effectiveness analyses using a decision tree model and reported that URS was the most cost-effective treatment strategy for ureteral stones at all locations after observation failed. The Lotan et al. 42 study cited the high cost of purchasing and maintaining a lithotriptor as the driver of the higher SWL cost in their study. Pearle et al. 43 also reported that SWL was slightly more costly than URS. Similarly, Cone et al. 44 reported that URS had superior clinical effectiveness and cost-effectiveness to SWL.

One of the strengths of this study is that it is the first study that we know of to measure and report quality of life in patients following these two care pathways using generic tools. 18 Over time, quality of life in increased in both groups, and at both the 8-week and the 6-month follow-ups the QALY gains in the URS arm, based on EQ-5D-3L scores and calculated from the AUC, were statistically significant. On average, QALY gain based on EQ-5D-3L scores was significantly lower in the SWL arm than in the URS arm, and this difference translated into approximately 10 more healthy days over the 6-month period for patients in the URS care pathway. In the case of utility scores based on the SF-12 questionnaire, the results were in the same direction in both arms. Scores in the URS arm were higher, on average than in the SWL group, and were statistically significantly higher at 6 months. However, the difference in SF-6D QALY score between the two groups over the 6-month period was not statistically significant.

Primary outcomes

Research to date about the relative merits of URS and SWL has focused on the balance between achieving a stone-free state for the patient and the need for further procedures or reintervention, with consideration also given to complication rates. In the TISU trial we found that the reintervention rate in the SWL arm was 11.4% (95% CI 5.0% to 17.8%) higher than in the URS arm. This is within the 20% limit set at the outset of the study to demonstrate non-inferiority, which would make it acceptable to patients and urologists as the initial treatment pathway. Our results corroborate previous findings,3 demonstrating better stone clearance with URS as the initial treatment pathway, but at a higher cost.

The higher reintervention rate seen in the SWL arm is almost certainly related to the fact that SWL can be limited by certain factors (e.g. skin-to-stone distance, patient tolerability of the shockwave treatment and stone density) that do not affect outcomes in a procedure under general anaesthesia using (typically) laser energy to fragment the stone, as is the case with URS. It has also been shown that effective delivery of SWL requires as much expertise and attention to detail as URS, and this is known to vary from centre to centre. 46

Secondary outcomes

As well as the need for further intervention, a key factor for patient choice and clinical advice is the complication rate associated with each of the two treatment pathways. For the purposes of this study we did not record expected, low-grade complications, such as bleeding or pain on passing urine (see Chapter 2, Safety reporting), but we did collect data on serious complications. The serious complication rate was similarly low in both treatment pathways, with SWL having a marginally higher serious complication rate of 3.6% (compared with 2.7% for URS).

There was only one death, which occurred in the URS arm, and this was unrelated to the treatment; this low number (< 1%) was as expected, as these procedures are generally associated with extremely low mortality. There were two life-threatening complications, both in the URS arm: one was cardiac (in a patient with a previous myocardial infarction) and the other pulmonary (pulmonary embolus in a patient who had also undergone recent orthopaedic surgery). Neither was assessed as being attributable to the trial intervention (or anaesthesia).

Quality of life in patients in this study was measured by the two generic HRQoL measures, as there is no condition-specific tool for stone disease. In both treatment pathways, by 8 weeks there was an overall improvement from baseline, but this was greater in the URS arm (and was maintained up to 6 months). The real extent of this difference is important to patients and the development of condition-specific HRQoL tools for use in urinary stone disease trials in the future would allow a more accurate measurement of quality of life following treatment of ureteric stones.

Pain and the need for analgesia over the treatment period was similar in both pathways, with no indication that one treatment pathway resulted in less pain or need for analgesia than the other.

However, the methods used to assess pain are more suited to chronic pain measurement and we are less certain of the impact of acute pain related to treatments delivered during their pathway. The acceptability of the treatments received by patients was assessed at 8 weeks and showed that both interventions were reassuringly acceptable to a high proportion of patients (SWL 86% and URS 83%).

Economics

On average, the URS care pathway cost more than the SWL care pathway and this cost difference was mainly driven by the unit cost of URS. The estimated unit cost of SWL was less than that of URS. Koo et al. 41 reported the cost in a UK setting of the SWL and flexible ureteroscopic laser lithotripsy procedure. The cost was calculated by the specialty costing department of the hospital and was based on the purchase, maintenance, repair and service costs of the lithotripter machine and flexible ureteroscope, as well as the cost of medical and nursing staff, the overhead cost for administration and the pharmaceutical and utility costs. This was defined as the ‘perceived cost’ and was formulated as the cost per session or procedure of SWL or flexible ureteroscopic laser lithotripsy. 1 The ‘actual cost’ was the perceived cost combined with the cost of additional procedures. The additional procedural cost was formulated according to the cost of instruments used that were specific to each patient (e.g. guidewires, stents, stone retrieval basket and stent removal) and the additional overhead cost was formulated as the cost per day’s stay in the hospital specific to each patient. The total mean costs of the interventions were similar to those used in our study. The measurement of the economic cost of a procedure and of ancillary procedures is clearly difficult to standardise within one health-care system, let alone across different health-care systems. These shortcomings are illustrated in a recent meta-analysis,18 in which the authors attempt to tabulate measurements of URS compared with SWL, in studies from several countries.

Strengths and weaknesses of the trial

Access to treatment

When assessing and designing a clinical pathway for a patient with ureteric colic, it is important to note that accessibility to the treatment is likely to be crucially important. For instance, although the planning for a session of SWL is relatively straightforward, particularly in centres with an in-house lithotripter, booking an urgent theatre slot (for URS) requires several hurdles to be overcome. For example, most emergency theatres will have resources for the insertion of a ureteric stent; however, planned laser stone fragmentation is a semi-elective procedure and requires the co-ordination of technology (URS, laser machine) and the appropriate theatre team (e.g. anaesthetist, radiographer and scrub team), not to mention the logistics of adding a case to an already full elective theatre list. Adopting the clinical pathway in which SWL is the preferred initial treatment option would potentially mean that 78% of all patients would not need further intervention with URS and, therefore, remove the need to find theatre time and space. From the NHS perspective, it is important to note that many hospitals dealing with patients presenting with ureteric stones currently do not have an on-site lithotripter and so cannot offer urgent treatment with SWL (a mobile service may be available on a monthly basis only). 47

A further, and more subtle, issue relates to the availability of flexible ureterorenoscopes, as opposed to semi-rigid ureteroscopes. The latter can treat most stones in the distal ureter and some mid-ureteric stones (55% of those in this study), but proximal ureteric stones in a male patient can sometimes be reached only with a flexible ureterorenoscope. Furthermore, a more distal stone is sometimes retropulsed proximally, again requiring clearance using flexible instrumentation. This is significant if the premise is that URS has a 90% chance of resulting in the patient being stone free, regardless of stone location within the ureter. Although all centres in this study had access to this flexible technology (as well as in-house lithotripsy), this is not true of all NHS hospitals and this means that decision-making will continue to depend on local equipment availability.

Waiting time

There was a substantial difference in waiting time to first treatment: 7 days from randomisation to treatment for the SWL arm, compared with 25 days from randomisation to treatment for the URS arm. This is likely to be representative of the challenges of booking semi-urgent cases into busy NHS theatre timetables. Although some NHS centres have trialled the use of dedicated theatre slots for urgent urology cases (as opposed to booking onto an emergency theatre list), this practice is not universally feasible. Although this route has the added attraction of ensuring that the necessary technology (e.g. laser and X-ray) is available, individual centres still face the pressure of having to allocate resources. There is recent evidence to support better outcomes (reduced need for further procedures and reduced hospitalisation) if SWL is delivered within 24 hours,48 just as the recent NICE guidelines have also suggested to aim for treatment within 48 hours. 45 The recommendations in the NICE guidelines were based on evidence of benefit to patients in terms of stone removal, repeated or ancillary procedures, and need for stent insertion.

There are other consequences for patients of waiting > 3 weeks after presentation, as this will inevitably mean that a proportion of patients will reattend the emergency department because of pain. Furthermore, among those in whom a stent has been inserted, a waiting time of several weeks for definitive surgery is known to be associated with an increasing risk of stent-related sepsis.

An exact time to achieve stone-free status from time of presentation is difficult to measure, even if all patients were to undergo post-treatment computerised tomography, and patient-reported passage of stones is unreliable. This pragmatic study based stone-free status on the need for further intervention to achieve stone clearance using standard NHS pathways and imaging techniques (mainly plain X-ray and ultrasound). The time to become stone free can, in the case of the URS arm, be the time to URS (with a proportion needing further time to be fragment free and stent free), whereas in the SWL arm it generally takes a few more days for stone fragments to be passed and, therefore, time to become stone free is much less clear in this arm. An agreed, standardised core outcome set for stone trials would be very helpful, as it could address issues such as time to stone-free measurements and ensure that all future studies use comparable data.

Other limitations

We had initially powered the study for 500 patients in each arm, but reached 303 and 306 participants in the SWL and URS arms, respectively. The CONSORT diagram shows that a large number of patients were screened for the study, but were not randomised. This may be because waiting times for treatment to start were different in the two arms of the study or patients had already received treatments before (i.e. were experiencing stone recurrence) or wanted to avoid a general anaesthetic.

The non-inferiority margin was placed at a figure of 20% based on the views of urologists and patients from the BAUS Section of Endourology; this is what they judged would be an acceptable limit, and is not a validated limit. However, the overall difference in stone-free rates of 11% was in accordance with our knowledge of contemporary URS and is still an acceptable outcome for the vast majority of patients with ureteric colic in this study.

Project Management Group

Samuel McClinton (chief investigator), James N’Dow, Graeme MacLennan, Mary Kilonzo, Frank Keeley, Ken Anson, Charles Clark, John Norrie, Rob Pickard, Sara MacLennan, Ruth Thomas, Kath Starr, Neil Burgess and Thomas Lam.

Independent members of the Trial Steering Committee

Roger Kockelbergh (TSC chairperson), John McGrath, Sarah Meredith and Stanley Coutts.

Independent members of the Data Monitoring Committee

Elaine McColl (DMC chairperson), Simon Harrison and Richard Emsley.

Principal investigators

David Thomas, Ken Anson, Joe Philip, Graham Young, Raj Gowda, Tony Browning, Ranjan Thilagarajah, Chris Betts, Ben Turney, Nitin Shrotri, Andrew Myatt, Oliver Wiseman, Benjamin Jenkins, Michael Kimuli, Simon Phipps, Ranan Dasgupta, Jeff Webster, Jake Patterson, Kay Thomas, Seshadri Sriprasad, Iqbal Shergill, James Forster, Giacomo Caddeo, John McCabe and Sachin Agrawal.

Research nurses/fellows/clinical trial assistants

Wendy Robson, Peter Murphy, Bernadette Kilbane, Nicola Brown, Leigh Morrison, Rebecca Ilyas, Susan Walker, Beverley Taylor, Jim Anderson, Fiona McNeela, Lauren Fergey, Tracy Camburn, Vicky Thomas, Emily Grout, Fiona Hammonds, Joshua O’Donnell, Tracey Cosier, Julie Rawlings, Kelly Leonard, Eleanor Dungca, Louise Fairlie, Lorraine Wiseman, Louise Williamson, Katherine Lawrence, Finny Patterson, Geraldine Cummings, Kareen Darnley, Gillian Hornzee, Sanela Andijac, Daisy Floyd, Annamaria Harmathova, Byiravey Pathmanathan, Susannah Hulton, Helen Bowyer, Joanna Peel, Rachel Muir, Golda-Grace Azanu, Rebecca Gare, Matthew Hogben, Kate Ripalda, Sherma Turner, Jnine Travis, Olivia McHugh, Stacy Ackerley, Claire Watkins, Linzi Williams, Hyley Inman, Dawn McNulty, Deborah Morgan, Charlotte Downes, Sharon Dealing, Susan Dowling, Catherine Gray, Maria Croft and Victoria Frost.

Contributions of authors

Ranan Dasgupta (https://orcid.org/0000-0001-8044-8902) (Consultant Urologist) contributed to the interpretation of the data and writing of the final report.

Sarah Cameron (https://orcid.org/0000-0002-4308-1416) (Trial Manager, Trialist) was responsible for the day-to-day management of the trial and contributed to the interpretation of the data and the writing of the report.

Lorna Aucott (https://orcid.org/0000-0001-6277-7972) (Senior Statistician) conducted the statistical analyses and contributed to the interpretation of the data and the writing of the report.

Graeme MacLennan (https://orcid.org/0000-0002-1039-5646) (Professor, CHaRT Director, Statistician and Trialist) contributed to the conception and design of the trial, the conduct of the trial and the interpretation of results and made significant contributions to the writing and editing of the report.

Mary M Kilonzo (https://orcid.org/0000-0002-3450-4536) (Health Economist) contributed to the conception and design of the study, the analysis of the health economics data and the drafting of the health economics chapters.

Thomas BL Lam (https://orcid.org/0000-0003-1582-3387) (Honorary Senior Lecturer in Urology and Consultant Urologist) contributed his clinical experience to the design of the study and writing of the final report.

Ruth Thomas (https://orcid.org/0000-0002-8316-2616) (Trialist) contributed to the conception and design of the study and the conduct of the trial and made significant contributions to the interpretation of the data and to the writing of the report.

John Norrie (https://orcid.org/0000-0001-9823-9252) (Professor of Medical Statistics and Trial Methodology, Director of Edinburgh Clinical Trials Unit) contributed to the design of the study, the delivery of the trial and the writing of the final report.

Alison McDonald (https://orcid.org/0000-0002-0256-2889) (Senior Trial Manager, Trialist) contributed to the design of the study, the delivery of the trial and the writing of the final report.

Ken Anson (https://orcid.org/0000-0001-8111-2802) (Consultant Urological Surgeon and Reader in Urology) contributed to the design of the study, the delivery of the trial and the writing of the final report.

James N’Dow (https://orcid.org/0000-0001-5340-0081) (Professor of Urology) contributed to the design of the study and the writing of the final report.

Neil Burgess (https://orcid.org/0000-0003-1986-0839) (Consultant Urologist) contributed to the design of the study and the writing of the final report.

Charles T Clark (https://orcid.org/0000-0003-1813-9767) (Member of Stone Patient Advisory Group) contributed to the design of the study, the delivery of the trial and the writing of the final report.

Francis X Keeley (https://orcid.org/0000-0001-7826-5664) (Consultant Urological Surgeon) contributed to the design of the study, the delivery of the trial and the writing of the final report.

Sara J MacLennan (https://orcid.org/0000-0003-1405-6964) (Health Psychologist) contributed to the conception and design of the study and made significant contributions to interpretation of the data and the writing of the report.

Kath Starr (https://orcid.org/0000-0003-3356-7751) (Trial Manager, Trialist) contributed to the design of the study, the delivery of the trial and the writing of the final report.

Samuel McClinton (https://orcid.org/0000-0002-0539-9570) (Chief Investigator and Professor of Urology) contributed to the conception and the design of the trial, the conduct of the trial, the interpretation of the results and the writing of the report.

Contributions of others

Rob Pickard (Professor of Urology) contributed his clinical expertise to the design of the study, recruitment, interpretation of the trial findings and the writing of final report.

Publications

Skea ZC, Treweek S, Gillies K. ‘It’s trying to manage the work’: a qualitative evaluation of recruitment processes within a UK multi-centre trial. BMJ Open 2017;7:e016475.

Dasgupta R, Cameron S, Aucott L, MacLennan G, Thomas RE, Kilonzo M, et al. Shockwave lithotripsy versus ureteroscopic treatment as therapeutic interventions for stones of the ureter (TISU): a multicentre randomised controlled non-inferiority trial. Eur Urol 2021;80:46–54.

Data-sharing statement

All data requests should be submitted to the corresponding author for consideration. Access to anonymised data may be granted following review.

Patient data

This work uses data provided by patients and collected by the NHS as part of their care and support. Using patient data is vital to improve health and care for everyone. There is huge potential to make better use of information from people’s patient records, to understand more about disease, develop new treatments, monitor safety, and plan NHS services. Patient data should be kept safe and secure, to protect everyone’s privacy, and it’s important that there are safeguards to make sure that it is stored and used responsibly. Everyone should be able to find out about how patient data are used. #datasaveslives You can find out more about the background to this citation here: https://understandingpatientdata.org.uk/data-citation.

Disclaimers

This report presents independent research funded by the National Institute for Health Research (NIHR). The views and opinions expressed by authors in this publication are those of the authors and do not necessarily reflect those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health and Social Care. If there are verbatim quotations included in this publication the views and opinions expressed by the interviewees are those of the interviewees and do not necessarily reflect those of the authors, those of the NHS, the NIHR, NETSCC, the HTA programme or the Department of Health and Social Care.