Randomised trials in vascular surgery: strengths, weaknesses, opportunities and threats
Professor Alison Halliday, Nuffield Department of Surgical Sciences
Clear outcomes from large well-conducted randomised surgical trials change clinical practice. Vascular Surgery trials and meta-analyses of carotid artery interventions provide the highest level of procedural evidence for stroke prevention. Use of international clinical networks can optimise recruitment, minimise bias and random error and improve generalisability, whilst simple trial design and follow up procedures should reduce busy clinicians’ workload. Over 6000 patients have now been recruited in the ACST-1 and ongoing ACST-2, two large trials which have changed practice [Strengths]. These trials are twice as big and cost X10 less than equivalent North American studies, and their hitherto understated study design deserves to be disseminated and adopted widely.
RESEARCH EXPERIENCE, RESEARCH METHODS AND TRAINING
The DPhil student will review the streamlined and evolutionary design of these two landmark trials and describe how this has enabled large-scale international recruitment, with good compliance and follow-up. A critical appraisal of ACST methodology will focus on the following unique streamlined features: Integration of recruitment activities with routine clinical care; In-person follow-up restricted to 1-month post-procedure review, with any subsequent strokes detected by annual questionnaires mailed directly to participants and detailed follow-up limited to the relatively small numbers of patients with incident strokes. The DPhil student may identify methodological improvements for subsequent ACST studies.
They will also be involved with the close-out (2020) and reporting (2021) of ACST-2, thereby gaining direct clinical trial experience. Specific ACST-2 activities will include: Writing a baseline / study design paper (2020); Phenotyping several hundred post-procedural strokes reported via annual questionnaires (eg, laterality, relationship to randomised carotid artery, severity), which will entail close liaison with ACST-2 collaborators and requires specialist clinical input.
Whilst randomised trials can compare long-term efficacy of procedures reliably, the procedural hazards seen in trials may not be generalisable to contemporary surgical practice, which are better seen in large registries [Weaknesses]. The DPhil student will review examples of randomised trials comparing novel vascular interventions with more established treatments (eg. early trials of CABG v medical therapy, peripheral angioplasty / stenting v open bypass surgery and carotid stenting v carotid surgery) and compare the results of such early trials with subsequent studies conducted once these innovations were established as part of routine clinical care.
Registry-based RCTs allow recruitment, randomisation and follow-up via national registries. Several have been completed (predominantly in Sweden) and others are proposed in the UK and elsewhere. This approach may enable efficient randomisation of large numbers of patients to different procedures. Generalisability of results from such registry-enrolled trials may be better than a traditional RCT if a substantial proportion of registry participants are included [Opportunities]. The DPhil student will perform a systematic review of registry-based RCTs with a focus on recruitment and retention rates and cost.
RCTs have become increasingly expensive. This is a particular challenge for trials of one strategy versus another (eg, open v endovascular surgery), which depend on non-commercial funding. Industry, healthcare payers and research funders are keen to explore alternative ways to assess effectiveness of different treatments and it is suggested that this may include non-randomised comparisons of outcomes in large registries [Threats]. The DPhil student will critically appraise the use of such approaches and identify both where this could be appropriate and where it is likely to produce misleading results.
FIELD WORK, SECONDMENTS, INDUSTRY PLACEMENTS AND TRAINING
Mr Bulbulia and Professor Halliday are members of the Carotid Stenosis Trialists’ Collaboration (CSTC) which includes all recent and ongoing large carotid trials and also work closely with several other groups conducting RCTs in other aspects of vascular surgery (eg, BEST-CLI). Opportunities exist for collaborative work with these groups, which may include working abroad with other key principal investigators. The supervisory team have extensive experience in the design, conduct and reporting of clinical trials and will provide the successful applicant with all necessary support. Dr Pan has extensive experience in statistics pertaining to clinical trials and IPD meta-analysis and will provide statistical support to the successful applicant. Formal statistical training may also be available.
The project will be suitable for vascular surgeons/stroke physicians/neurologists with a particular interest in evidence-based medical care, randomised trials and stroke prevention.