Vein bypass first vs. best endovascular treatment first revascularisation strategy for chronic limb-threatening ischaemia due to infra-popliteal disease: the BASIL-2 RCT

Article history

The research reported in this issue of the journal was funded by the HTA programme as award number 12/35/45. The contractual start date was in February 2014. The draft manuscript began editorial review in June 2023 and was accepted for publication in February 2024. 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’ manuscript 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 article.

Copyright statement

Copyright © 2024 Moakes et al. This work was produced by Moakes et al. under the terms of a commissioning contract issued by the Secretary of State for Health and Social Care. This is an Open Access publication distributed under the terms of the Creative Commons Attribution CC BY 4.0 licence, which permits unrestricted use, distribution, reproduction and adaptation in any medium and for any purpose provided that it is properly attributed. See: https://creativecommons.org/licenses/by/4.0/. For attribution the title, original author(s), the publication source – NIHR Journals Library, and the DOI of the publication must be cited.

2024 Moakes et al.

Trial design

The BASIL-2 trial was a pragmatic, open-label, parallel, multicentre, randomised, superiority trial. The trial had a favourable ethical opinion from National Research Ethics Service Committee 14/WM/0057 obtained in March 2014. The final protocol version was v7.0 dated 22 September 2022. The protocol for the trial is published22 (based on an earlier version of the protocol), and the full, most recent protocol is available at the BASIL-2 website (www.birmingham.ac.uk/BASIL2).

Participants

The participants in the BASIL-2 trial were recruited from vascular surgery units across the UK as well as in Stockholm, Sweden and Kolding, Denmark. Eligible participants were patients who presented to hospital-based vascular surgery units with CLTI due to atherosclerotic PAD and who required an IP revascularisation with or without an additional more proximal infra-inguinal revascularisation procedure to restore limb perfusion. Participants had to have an anticipated life expectancy of more than 6 months, judged by a minimum of two consultants (at least one of whom was performing IP VB and one of whom was performing IP BET in their clinical practice) to require and be suitable for both IP VB and IP BET. Eligible patients were not permitted to have had previous vascular intervention to the target IP artery within the previous 12 months, but did need to have adequate aortoiliac inflow to support both revascularisation strategies. Patients had to be able and willing to complete HRQoL and health economic questionnaires (with help if required) and be able to understand sufficient English, Swedish, or Danish (depending on country of recruitment) to ensure informed consent.

Randomisation

Once eligibility was confirmed and consent obtained, participants were randomised to the BASIL-2 trial by the research staff at participating centres using a secure online, central randomisation service provided by the Birmingham Clinical Trials Unit. A central telephone back-up service was also available. Participants were randomised on a 1 : 1 basis to either a VB-first or a BET-first revascularisation strategy. A minimisation procedure, incorporating a random element, using a computer-based algorithm, was used to reduce the chance of imbalances in important prognostic variables. Strata used in the minimisation were:

• Age (≤ 60, 61–70, 71–80, > 80 years).

• Gender (male, female).

• DM and chronic kidney disease (CKD) {DM, CKD [CKD defined as stage 3 or worse based on estimated glomerular filtration rate (eGFR) of < 60 (ml/minute/1.73 m²)]23, DM and CKD or neither}.

• Severity of clinical disease (ischaemic rest/night pain only, tissue loss only, or both).

• Previous (permissible) intervention to the trial leg (yes, no).

• Intention for hybrid procedure (yes, no).

Procedures

The pragmatic trial design encouraged vascular and endovascular surgeons and interventional radiologists to perform VB and BET using their preferred equipment, devices, and surgical and anaesthetic techniques. For VB, any vein deemed suitable by the responsible vascular surgeons could be used. If at operation it was discovered that the vein could not be used, then composite or prosthetic grafts could be inserted at the surgeon’s discretion in the patient’s best interests. Regarding BET, any device being used as part of standard care in that country was permissible. Drug-coated balloons (DCBs), bare metal stents (BMSs) and drug-eluting stents (DESs) could be used at the operator’s discretion. Atherectomy devices were permitted but not used. In this pragmatic trial, all additional management strategies, including additional procedures, were at the responsible clinicians’ discretion in the patient’s best interests.

Blinding

BASIL-2 was an open-label study; participants, study staff, and investigators were not masked to treatment allocation.

Adherence monitoring

Adherence to the allocated intervention was confirmed on the intervention form where the first revascularisation procedure details were collected. The researchers defined adherence as those participants in the VB-first group who received a VB as their first revascularisation intervention and those participants in the BET-first group who received a BET as their first revascularisation intervention. Further revascularisations, which have been captured as secondary outcomes, were permitted as deemed clinically indicated by the responsible clinicians and were not considered part of the adherence definition.

Participant withdrawal

Participants could voluntarily withdraw from the trial (or any component of it) at any time. Reasons for withdrawal were captured where possible. If a participant explicitly withdrew consent to have any further data recorded (including NHS data), their decision was respected and recorded on the electronic data capture system. All communication surrounding the withdrawal was noted in the patient’s medical notes, and no further data were collected for that participant. In rare instances where a participant was randomised and was later found to have violated one or more inclusion criteria they remained in the trial.

Outcomes and assessments

Serious adverse events

Serious adverse events (SAEs) were recorded on a purpose-designed SAE form and notified by local investigators to the clinical trials unit within 24 hours of the local investigators becoming aware of these events. Local investigators were responsible for additionally reporting SAEs to their host institutions, according to local regulations. SAEs that were categorised by the Chief Investigator as both suspected to be related to the trial intervention and unexpected were classified as related unexpected serious adverse events (RUSAEs) and were subject to expedited reporting. SAEs were collected up to 30 days following the first revascularisation.

Statistical considerations

Trial oversight

Study oversight was provided by a Trial Steering Committee (TSC) (Chaired by Professor Jonathan Michaels, University of Sheffield) and a DMC (Chaired by Professor Charles McCollum, University of Manchester).

The TSC provided independent supervision for the trial, providing advice to the Chief and Co- Investigators and the Sponsor on all aspects of the trial throughout the study. The DMC adopted the DAMOCLES charter to define its terms of reference and operation in relation to oversight of the BASIL-2 trial. 30

Data sharing

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

Recruitment

The trial recruited its first participant on 22 July 2014 and closed to recruitment on 30 November 2020 after 345 had participants had been randomised. The trial completed follow-up on 30 November 2022, after all surviving participants were followed up for a minimum of 2 years from randomisation. The database was locked for statistical analysis on 5 January 2023. Recruitment occurred across 39 vascular surgery units in the UK, and one each in Sweden (Stockholm) and Denmark (Kolding) (Table 3).

Participant flow (up to 24 months) is illustrated in Figure 1. Three hundred and forty-five participants were randomised to BASIL-2, with 172 participants allocated to a VB-first and 173 participants allocated to a BET-first revascularisation strategy. At the end of follow-up, 74 participants were withdrawn from the study and a further 10 participants were lost to follow-up. Type of withdrawal and median time to withdrawal are presented in Table 4. At the end of follow-up {median 40.0 months [interquartile range (IQR) 20.9–60.6]}, 200 primary outcome events had been observed. All patients were included in the primary ITT analysis of AFS.

FIGURE 1.

Participant flow. LFU, lost to follow-up.

Baseline data

The baseline demographic characteristics of participants in the VB-first group and the BET-first group were comparable, with the minimisation algorithm achieving balance across age, gender, DM and CKD, severity of clinical disease, previous (permissible) intervention to the trial leg and intention for a hybrid procedure.

The participants had a mean age of 70.7 years [standard deviation (SD) 11.0 years] and 81% were male. Sixty-nine percentage of participants had predominantly type 2, DM [207/237 (87%)] and a high proportion were insulin-dependent [123/236 (52%)]. In regard to their severity of clinical disease at randomisation, 88% presented with tissue loss (with or without rest pain), which was mainly at or below the ankle. Twenty-one percent of participants declared themselves as current smokers, with an additional 49% declaring to be ex-smokers (Table 5).

Adherence data

Adherence to allocated intervention was high: (n = 145, 84%) in the VB-first group and (n = 165, 95%) in the BET-first group. In total, 17 participants received no revascularisation intervention (n = 15 VB-first, n = 2 BET-first), and reported reasons included deterioration or improvement in condition and patient declining. Eighteen participants (n = 12 VB-first, n = 6 BET-first) received alternative revascularisation interventions to the one to which they were allocated at randomisation. Ninety-four percentage of participants who received a first revascularisation intervention did so within 4 weeks of randomisation (Table 6). Further details of the first revascularisation procedures are provided in Tables 7 and 8.

Primary outcome

One hundred and eight (63%) of 172 patients in the VB-first group and 92 (53%) of 173 patients in the BET-first group had a major amputation or died [adjusted HR 1.35 (95% CI 1.02 to 1.80); p = 0.037]. Thirty-five participants (n = 18 VB-first, n = 17 BET-first) had a major amputation and then died (major amputation was considered their first event for the time-to-event analysis) (see Table 9 and Figure 2).

FIGURE 2.

Kaplan–Meier plot of AFS (ITT analysis).

Sensitivity analyses which included a per-protocol and an as-treated analysis supported a trend towards an increased risk of major amputation or death in the VB-first group (Table 10). No evidence of non-proportional hazards was observed.

Subgroup analysis was performed for the pre-specified variables used in the minimisation algorithm and baseline ABPI and TBPI. There was no evidence for varying treatment effects in each subgroup analysis performed. The proportion of participants who had a major amputation or death in each subgroup is shown in Figure 3.

FIGURE 3.

Forest plot for subgroup analyses for AFS (ITT analysis).

Secondary outcomes

Secondary outcomes are reported in Tables 11 and 12.

Ninety-one (53%) of 172 patients in the VB-first group and 77 (45%) of 173 in the BET-first group died from any cause [adjusted HR 1.37 (95% CI 1.00 to 1.87); Figure 4]. Cardiovascular and respiratory diseases were the most common causes of death in both groups as expected. There were no specific causes of death identified in either group which would explain the differences in number of deaths observed between the two groups (Tables 13 and 14).

FIGURE 4.

Kaplan–Meier plot for OS (ITT analysis).

Thirty-five (20%) of 172 in the VB-first group and 32 (18%) of 173 patients in the BET-first group had a major amputation [adjusted HR 1·23 (95% CI 0.75 to 2.01); Figure 5]. There was no difference in the number of participants who had at least one further revascularisation procedure in the trial leg between the VB-first group [50 (29%) of 172 patients] and the BET-first group [56 (32%) of 173; adjusted RR 0.94 (95% CI 0.68 to 1.28)]. However, the number of patients who had a re-intervention was higher in the BET-first group [33 (19%) patients] than in the VB-first group [9 (5%) patients; RR 0.27 (95% CI 0.13 to 0.55)]. Conversely, cross-over interventions were more common in the VB-first group [46 (27%) patients] than in the BET-first group [33 (19%) patients; RR 1.43 (95% CI 0.94 to 2.18)]. There were no differences between the two treatment groups in 30-day morbidity and death, MALE, MACE, relief of ischaemic pain, or HRQoL (see Tables 11 and 12). The results from the post hoc sensitivity analyses for time-to-event secondary outcomes are reported in Table 15.

FIGURE 5.

Kaplan–Meier plot for major amputation (ITT analysis).

Serious adverse events

Twenty-nine (17%) patients in the VB-first group and 23 (13%) patients in the BET-first group reported 33 and 26 SAEs, respectively. Only one SAE (biliary sepsis due to gallstones complicated by pancreatitis and organ failure) in the BET-first group was considered related to the trial intervention and a RUSAE.

Introduction

This chapter contains material that has been reproduced with permission from Bradbury et al. 1 This is an Open Access article distributed under the terms of the Creative Commons Attribution Licence, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited. The text below includes minor additions and formatting changes to the original text.

This chapter describes the health economic analysis conducted for the BASIL-2 trial. The objective of the analysis was to assess the cost-effectiveness of a VB-first versus a BET-first revascularisation strategy in CLTI patients who required an IP, with or without a more proximal infra-inguinal revascularisation procedure, to restore limb perfusion. A comprehensive health economic analysis plan was specified before analysis.

Methods

Results

Sensitivity analysis

The joint distributions of incremental costs and effects (AFLY), derived from the non-parametric adjusted bootstrap, provide an estimate of the distribution of the population costs and effects presented on the cost-effectiveness plane in Figure 6. Most of the point estimates support the base-case finding and were located in the southeast and northeast quadrants where BET-first is more effective and either less costly or more costly than VB-first. In the absence of an official WTP value for the primary outcome, the probability of BET-first being cost saving was 80% and the probability of the intervention being cost-effective can increase to 98% if an arbitrary WTP value per AFLY of £15,000 is used (Figure 7).

FIGURE 6.

Cost-effectiveness plane (BET-first vs. VB-first): years alive without amputation including hospital costs only, 7 years after randomisation (base-case analysis – discounted).

FIGURE 7.

Cost-effectiveness acceptability curve: years alive without amputation including hospital costs only, 7 years after randomisation (base-case analysis – discounted).

Considering the cost per additional QALY gained, the CE planes of the CUA for 2 and 3 years post randomisation are displayed in Figures 8 and 9, respectively. The figures revealed that BET-first is likely to be less costly and less or more effective when compared with VB-first.

FIGURE 8.

Cost-effectiveness plane (BET-first vs. VB-first): QALY including hospital costs only, 2 years after randomisation (base-case analysis – discounted).

FIGURE 9.

Cost-effectiveness plane (BET-first vs. VB-first): QALY including hospital costs only, 3 years after randomisation (base-case analysis – discounted).

The probability that BET-first is cost-effective at different WTP thresholds over 2 and 3 years is plotted on the CEAC in Figures 10 and 11, respectively. Compared to VB-first, BET-first is likely to be 93% and 89% cost saving at 2 years and 3 years, respectively. When NICE’s WTP threshold of 20,000 is applied, BET-first is 91% cost-effective at 2 years and 94% cost-effective at 3 years.

FIGURE 10.

Cost-effectiveness acceptability curve: QALY including hospital costs only, 2 years post randomisation (base-case analysis – discounted).

FIGURE 11.

Cost-effectiveness acceptability curve: QALY including hospital costs only, 3 years after randomisation (base-case analysis – discounted).

Scenario analyses

Results of different scenario analyses are presented in Table 24. The inclusion of primary health care and other hospital-based services costs, taking the wider societal perspective and considering those who were adherent to the trial intervention, did not change the direction of the findings from the base-case analyses. Overall, BET-first remained a cost-saving and more effective strategy compared with VB-first. When including only patients with complete data, BET-first was still cost saving by £4366 less hospital costs and resulted in additional 0.211 QALY gains. The complete case analysis suggests that the differences in cost and effect (QALY) between BET-first compared to VB-first were statistically significant. The CE plane in Figure 12 shows that a high proportion of the generated point estimates of incremental costs and QALYs using complete data of hospital costs and QALYs imply that BET-first is less costly and more effective than VB-first.

TABLE 24 - Scenario analysis discounted results – 2 years after randomisation

ICER, incremental cost-effectiveness ratio.

Adjusted for age group (≤ 60, 61–70, 71–80, > 80 years), gender, DM and CKD, severity of clinical disease, previous (permissible) intervention to the trial leg. Values > 0 favour BET in comparison to VB.

Adjusted for baseline EQ-5D-5L, age group (≤ 60, 61–70, 71–80, > 80 years), gender, DM and CKD, severity of clinical disease, previous (permissible) intervention to the trial leg. Values < 0 favour BET in comparison to VB.

Dominant = intervention (BET) less costly and more effective.

Difference is statistically significant.

	Mean total (SD) £cost	Incremental £cost (95% CI)a	Mean QALY (SD)	Incremental QALY (95% CI)b	ICER (£cost per QALY)
UK NHS perspective [secondary (hospital based), other secondary and primary NHS healthcare costs]
VB n = 172	17,927.84 (16,366.52)		0.832 (0.515)
BET n = 173	15,912.35 (15,748.35)	−2078.36 (−5409.00 to 1858.61)	0.884 (0.515)	0.016 (−0.083 to 0.116)	Dominantc
Societal perspective (all NHS, patient incurred and productivity costs)
VB n = 172	18,425.47 (16,410.69)		0.832 (0.515)
BET n = 173	16,036.25 (15,781.83)	−2420.54 (−5791.18 to 1290.36)	0.884 (0.515)	0.016 (−0.083 to 0.116)	Dominant
Per protocol (adherent)
VB n = 145	16,338.23 (17,208.12)		0.863 (0.497)
BET n = 165	13,372.85 (15,712.11)	−3283.94 (−7181.42 to 239.27)	0.887 (0.523)	0.038 (−0.082 to 0.146)	Dominant
Complete case
VB n = 66	14,975.07 (13,060.71)		0.743 (0.523)
BET n = 66	10,412.35 (9986.97)	−4366.45 (−9117.07 to −600.90)d	0.933 (0.516)	0.211 (0.043 to 0.382)d	Dominant

FIGURE 12.

Cost-effectiveness plane: QALY including hospital costs only, 2 years after randomisation (complete case analysis – discounted).

Subgroup analysis

The cost-effectiveness results, associated with the prespecified subgroups, are illustrated in Figure 13. Overall, BET-first is dominant (less costly and more effective) over VB-first in most of the subgroups. However, the results showed that differences in costs and QALYs varied from positive to negative, and this highlights the level of uncertainty over the direction of cost and QALY changes. The differences were also associated with relatively wide CIs, for selected groups,indicating the uncertainty in the size of the estimated QALY gain or loss associated with BET-first relative to VB-first strategy.. In some subgroups (female, those who had unknown intervention to the trial leg, patients with tissue loss only and patients below the age of 60 years), BET-first was dominated by VB-first where it resulted in higher costs and lower QALYs. However, the number of patients in these subgroups was relatively small. In other subgroups, BET-first was more costly and more effective than VB-first and the estimated ICER was above NICE threshold of 20,000 per QALY gained; therefore BET-first was not deemed to be cost-effective. These findings should be interpreted with caution given the level of uncertainty around QALY changes.

FIGURE 13.

Forest plot of BET-first vs. VB-first estimated difference in hospital costs and difference in QALY, 2 years after randomisation for specific subgroups. ICER, incremental cost-effectiveness ratio; dominant, intervention (BET-first) less costly and more effective; dominated, intervention (BET-first) more costly and less effective.

Clinical findings

The BASIL-2 trial shows that, in patients with CLTI who require an IP, with or without an additional more proximal infra-inguinal revascularisation procedure, to restore limb perfusion, a VB-first revascularisation strategy was associated with an approximately one-third increased risk of major amputation or death from any cause compared with a BET-first revascularisation strategy. The difference in AFS was mainly driven by fewer deaths in the BET-first group as limb-based outcomes were similar between the groups. The reported 30-day post-procedural morbidity and mortality rates were not significantly different between the two groups, and the causes of death in the two groups were not unexpected. Further analyses of the BASIL-2 data set and other similar cohorts of patients with CLTI will be required to understand the reasons for the differences observed. However, differences in time to first revascularisation, the timing and nature of additional procedures, and in BMT might be important. 13,54,55 BASIL-2 recruitment was stopped early due to the COVID-19 pandemic. COVID-19 also had a major adverse effect on follow-up, particularly for end points that required face-to-face assessment. However, the study showed no evidence that the observed difference in outcomes between the groups is COVID-19 related.

In 2012, when the BASIL-2 trial was being developed, an IP VB-first revascularisation strategy was hypothesised to be superior to an IP BET-first strategy. This was mainly based on the results of the BASIL-1 trial. 13 However, only around 25% of participants in the BASIL-1 trial had an IP revascularisation. This reflected the fact that at the time (1999–2004) most patients with CLTI were having only more proximal (mainly FP) infra-inguinal revascularisation procedures.

Despite advances in vascular and endovascular techniques and devices, in patients with often heavily calcified IP disease both IP VB and BET are technically challenging. As a result, controversy remains as to whether patients with CLTI, requiring an IP revascularisation and who are suitable for both procedures, should be offered VB- or BET-first. 56 For all these reasons, the results from the BASIL-2 trial have been keenly awaited.

At first glance, this study’s results appear to conflict with the BEST-CLI trial17 which showed that VB in patients with an optimal, single segment GSV (cohort 1), VB was associated with a better outcome, in terms of a composite primary outcome of any MALE or death, than BET. Of note, there were no differences in outcome in participants who did not have an optimal vein for bypass (cohort 2). 17 However, there are many differences between the two trials, importantly including the primary end point. The clinical experience of the BASIL-2 investigators suggests that relatively few patients with CLTI are deemed suitable and have an optimal vein for bypass. Further work is required to determine if BASIL-2 participants are more like patients randomised in cohort 2 of the BEST-CLI trial. Although BASIL-2 and the BEST-CLI trial19 were developed, run and analysed entirely independently, the researchers of this study have collaborated closely with the BEST-CLI trial investigators, and entered into a data-sharing agreement with the BEST-CLI trial investigators before either trial was analysed. A data-sharing agreement will allow an in-depth comparison of the two trials that will hopefully explain why some of the outcomes observed appear to be different. One of the outputs will be an individual-patient-level data meta-analysis. Until this is completed, this study can only speculate as to why the two trials appear to have reached different conclusions in certain respects.

Outcomes for the patients in the BASIL-2 trial were poor [median AFS of 3.8 years (IQR 3.1–4.4)] and not materially different from those reported in the BASIL-1 trial. 57 In both trials, around half of the patients had died by 5 years post randomisation. Although limb salvage rates of around 80% at 5 years seem encouraging, many patients had died and were no longer at risk of limb loss. 58

At randomisation, most BASIL-2 participants were on what has been termed BMT. The severe multilevel atherosclerotic disease that causes CLTI develops over many years. The researchers do not have information on the quality of previous BMT and lifestyle (e.g. stop smoking) interventions. Randomisation, as researchers have described, helps ensure that variations in the quality of previous BMT and lifestyle interventions are equally represented in both groups. Around 20% of patients admitted that they were still smoking and around 70% of patients had (mostly type 2) diabetes of whom around 50% required insulin. At presentation around 90% of the participants had, often quite extensive tissue loss. These baseline data suggest that there may still be missed opportunities in public health and primary care to prevent CLTI through medical therapy and lifestyle interventions, as well as missed opportunities to refer patients to secondary care earlier once CLTI begins to develop. 1,59 Better prevention and timely referral are important as the BASIL-2 trial shows once again that by the time patients present to vascular and endovascular surgeons and interventional radiologists with established CLTI, their prognosis is often poor in terms of major amputation and death regardless of what form of revascularisation they are offered. 60

The BASIL-2 trial has some statistical and clinical limitations. The total number of participants aimed to enrol was not met due to challenges in recruitment, including COVID-19. Since the planned sample size was based on the number of events required, the reduction in participants recruited was mitigated by the increased duration of follow-up. However, as the study did not reach its planned target of event numbers, it is important that the uncertainty in the findings be considered. The uncertainty in the estimate is best judged by the point estimate and the CI of the primary end point [AFS; HR 1.35 (95% CI 1.02 to 1.80)]. This shows that an increase in risk of 35% (HR 1.35) is the most likely value, with increases in risk of 2% (HR 1.02) and 80% (HR 1.80) the least probable points in this range. Although most of this range covers point estimates likely to be considered clinically important differences, it also includes smaller differences. However, it is very unlikely (around a 2% chance) that a VB-first could be more clinical effective than BET-first revascularisation strategy in this patient cohort.

Clinical colleagues will need to consider some important issues when deciding to what extent this study’s findings can be applied to their patients, practice, and healthcare system. Vascular specialists in some countries might be unable to offer BET due to cost. 61 Patients with CLTI, presenting in other countries, might be different in several ways, for example, inter alia age, sex, prevalence of risk factors, and racial background. 3,17 It is also important to note that many patients with CLTI are offered primary amputation or conservative (end of life) care rather than revascularisation; many patients do not have a suitable vein for VB, so they are deemed only suitable for BET; and even patients who are deemed suitable for VB and have a good vein might choose the less invasive endovascular option. 3

Collecting reliable screening data, other than in Birmingham, proved impossible due to changes in the funding model and the need to increase in the number of centres. However, in Birmingham, the researchers established the BASIL prospective cohort study, which includes all (nearly 500) patients with CLTI admitted to Heartlands Hospital during the recruitment period. These data will be the subject of a further report. The researchers hope the prospective cohort study will allow the BASIL-2 findings to be viewed in the context of the CLTI patient population as a whole, as well as help establish the generalisability of the BASIL-2 trial to similar patients who were not recruited to the trial.

Recruiting patients to the BASIL-2 trial was much more difficult than anticipated. The BEST-CLI trial17 also faced similar difficulties despite much greater funding and a larger potential patient population.

As well as many patients being deemed unsuitable for both procedures (especially for VB) for a variety of reasons, an absence of equipoise on the part of clinicians and patients was an important issue. Colleagues explained it was often easier to offer early BET than it was to offer early VB, and easier to obtain imaging confirming suitability for BET. These logistical issues may cause delay to revascularisation, and this can be associated with a worse outcome. 62 In BASIL-2, most patients received their allocated revascularisation procedures in a timely manner that was clinically appropriate for that individual. Nevertheless, the BASIL-2 trial is a pragmatic randomised trial that compares two different revascularisation strategies, rather than two different sets of procedures, in a real-world context of what can be realistically achieved within the national, publicly funded healthcare systems of the UK, Sweden and Denmark. It is important that the BASIL-2 primary comparative analysis be performed, and the results interpreted, on an ITT basis as results in alternative analysis populations may provide biased estimates (since randomisation can no longer preserve balance for known risk factors).

In conclusion, BASIL-2 has shown that, in patients with CLTI who require an IP revascularisation, with or without an additional more proximal infra-inguinal revascularisation procedure, and who are deemed suitable for both VB and BET, randomisation to a VB-first revascularisation strategy led to a 35% increased risk of major amputation or death when compared to a BET-first revascularisation strategy. This difference was mainly driven by fewer deaths in the BET-first revascularisation group as limb-based outcomes were similar between the two groups.

Health economics findings

Patient and public involvement

The researchers have been supported throughout the trial by two patient and public involvement (PPI) members who comprised part of their TSC. The researchers engaged with the PPI members throughout, improving their understanding of the needs of patients with CLTI. PPI members commented on all patient-facing materials to ensure that they were clear and comprehensive. The researchers organised collaboration days during the course of the trial and the PPI representatives attended these meetings. The researchers will engage with their PPI members regarding the dissemination of their results. Any future research groups, taking forward the research recommendations from this project, would benefit from engaging with PPI representatives.

Equality, diversity and inclusion

As this trial had restrictive inclusion criteria, the researchers were unable to target specific under-represented groups. However, they were open to recruitment across 41 centres in Scotland, England, Wales, Sweden and Denmark, some rural and some inner city, and the ethnic diversity of the principal investigators (PIs) and research staff at these sites was broad. Of the 345 participants recruited to this trial, 280 were male (81%), 315 were white (92%); 10 were Asian (3%); 17 were black (4%) and 3 (1%) were from other ethnic groups or declined to provide information. Data from the BASIL prospective cohort study, which includes all (nearly 500) patients with CLTI admitted to Heartlands Hospital during the BASIL-2 recruited period, showed a similar demographic composition of ethnicity and gender which suggests that the BASIL-2 population was generalisable to the CLTI population as a whole (BASIL prospective cohort study awaiting publication).

CRediT contribution statement

Catherine A Moakes (https://orcid.org/0000-0002-0473-0532): Data curation (equal), Formal analysis (lead), Methodology (supporting), Writing – original draft (lead), Writing – editing and reviewing (lead), Visualisation (equal), Writing – original draft (supporting).

Andrew W Bradbury (https://orcid.org/0000-0003-0530-9667): Conceptualisation (lead), Funding acquisition (lead); Methodology (lead), Writing – editing and reviewing (supporting), Visualisation (equal), Writing – original draft (supporting).

Zainab Abdali (https://orcid.org/0000-0002-2736-5427): Formal analysis (lead), Writing – editing and reviewing (supporting), Visualisation (equal), Writing – original draft (supporting).

Gareth R Bate (https://orcid.org/0009-0005-3084-070X): Conceptualisation (supporting), Data curation (equal), Funding acquisition (supporting), Methodology (supporting), Visualisation (equal), Writing – original draft (supporting).

Jack Hall (https://orcid.org/0000-0002-3016-7624): Data curation (equal), Validation (lead), Visualisation (equal), Writing – original draft (supporting).

Hugh Jarrett (https://orcid.org/0000-0001-7374-3700): Data curation (equal), Project administration (equal), Resources (equal), Supervision (equal), Visualisation (equal), Writing – original draft (supporting).

Lisa Kelly (https://orcid.org/0009-0000-0105-1967): Data curation (equal), Visualisation (equal), Writing – original draft (supporting).

Jesse Kigozi (https://orcid.org/0000-0001-7608-4923): Validation (lead), Visualisation (equal), Writing – original draft (supporting).

Suzanne Lockyer (https://orcid.org/0009-0000-9188-2261): Data curation (equal), Project administration (equal), Resources (equal), Visualisation (equal), Writing – original draft (supporting).

Lewis Meecham (https://orcid.org/0000-0003-4698-7333): Methodology (supporting), Visualisation (equal), Writing – original draft (supporting).

Smitaa Patel (https://orcid.org/0000-0002-4997-3829): Conceptualisation (supporting), Data curation (equal), Funding acquisition (supporting), Methodology (supporting), Visualisation (equal), Writing – original draft (supporting).

Matthew Popplewell (https://orcid.org/0000-0002-6239-4039): Methodology (supporting), Visualisation (equal), Writing – original draft (supporting).

Gemma Slinn (https://orcid.org/0000-0003-1096-4148): Project administration (equal), Resources (equal), Supervision (equal), Visualisation (equal), Writing – original draft (supporting).

Jonathan J Deeks (https://orcid.org/0000-0002-8850-1971): Conceptualisation (supporting), Funding acquisition (supporting), Methodology (supporting), Visualisation (equal), Writing – original draft (supporting).

Acknowledgements

The researchers would like to thank all the participants who agreed to take part in this study, without whom this would not have been possible.

Patient data statement

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 they are 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.

Data-sharing statement

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

Ethics statement

The BASIL-2 trial received ethical approval from the West Midlands Research Ethics Committee (reference 14/WM/0057) on 4 March 2014. Study registration: ISRCTN 27728689.

Information governance statement

The University of Birmingham is committed to handling all personal information in line with the UK Data Protection Act (2018) and the General Data Protection Regulation (EU GDPR) 2016/679. Under the Data Protection legislation, the University of Birmingham is the Data Controller, and you can find out more about how the researchers handle personal data, including how to exercise your individual rights and the contact details for our Data Protection Officer here: https://www.birmingham.ac.uk/privacy.

Disclosure of interests

Full disclosure of interests: Completed ICMJE forms for all authors, including all related interests, are available in the toolkit on the NIHR Journals Library report publication page at https://doi.org/10.3310/YTFV4524.

Primary conflicts of interest: Andrew W Bradbury has the following memberships: HTA Efficient Study Designs – 2, HTA Efficient Study Designs Board, HTA IP Methods Group, HTA IP Panel, HTA Surgery Themed Call Board, HTA Prioritisation Committee B (In hospital), HTA Remit and Competitiveness Group, HTA Prioritisation Committee B Methods Group, HTA Post-Funding Committee teleconference (POC members to attend) and HTA Programme Oversight Committee.

Jonathan J Deeks has the following memberships: DTSP Methods Group Teleconference, HTA Additional Capacity Funding Board, HTA Efficient Study Designs – 2, HTA End of Life Care and Add-on Studies, HTA Medical Tests Methods Group, HTA Primary Care Themed Call board, HTA Primary Care Themed Call Remit meeting and PGfAR EOIs – HTA projects Remit meeting.

Publication

Bradbury AW, Moakes CA, Popplewell M, Meecham L, Bate GR, Kelly L, et al. A vein bypass first versus a best endovascular treatment first revascularisation strategy for patients with chronic limb threatening ischaemia who required an infra-popliteal, with or without an additional more proximal infra-inguinal revascularisation procedure to restore limb perfusion (BASIL-2): an open-label, randomised, multicentre, phase 3 trial. Lancet 2023;401:1798–809.

Disclaimers

This article presents independent research funded by the National Institute for Health and Care 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, 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, the HTA programme or the Department of Health and Social Care.