Plain versus drug balloon and stenting in severe ischaemia of the leg (BASIL-3): open label, three arm, randomised, multicentre, phase 3 trial.
Bradbury AW, Hall JA, Popplewell MA, Meecham L, Bate GR, Kelly L, Deeks JJ, Moakes CA, BASIL-3 Investigators
- DOI
- 10.1136/bmj-2024-080881
- Record issued
- 2026-08-15
- Engine
- 7.39.0
- Exported
- 2026-09-19
Prepared by Alpha1. This document is confidential: it is intended for the recipient it was shared with and must not be redistributed. The live record at alpha1science.com/verify/325b7c42-9ef5-4ba0-a4ca-ffb1094e4b4e is authoritative.
How this rating was calculated
- IntegrityIntegrity concern ×2−1★
- ReportingData & code availability partially met−0.25★
- No data or code availability links were detected to verify.
This Kaimen Rigor review uses Kaimen Rigor reviewers trained on a curated corpus of high-fidelity and retracted papers, with expert supervision and curation. It can still make mistakes; verify each finding against the source before relying on it.
This is a well-conducted and well-reported phase 3 randomised controlled trial with a clear scientific premise, rigorous design, and comprehensive reporting of ethics, biological variables, and statistical methods. The main weakness is the vague data availability statement, which lacks a concrete access mechanism or platform. Minor copyedit issues and a few internal consistency points (e.g., 481 vs 480 participants) are noted but are explained in the text.
This is a post-publication audit of a published trial. Both reviewers independently scored all eight dimensions and agreed on all statuses, so no divergence needed reconciliation. The statistics verification component checked only 2 reported tests (both consistent); the rest of the statistical analyses were not machine-verified and should not be assumed correct. The citation check found no retracted or unresolved references.
Numerical inconsistencies
1 finding · worst lowValues that contradict each other or are impossible for the stated sample: recomputed p-values and test statistics, GRIM/GRIMMER checks on summary numbers, percentages against their own counts, totals against their parts, and estimates against their own confidence intervals.
- Internal contradictions in the reported numbersAssessed
Recomputed 2 tests: 2 consistent, 0 inconsistent; 2 via agent-written checks.
- CONSISTENTreported p = .220 · recomputed p = .288Reviewers 1, 2Recompute p-value for primary comparison PBA±BMS vs DCBA±BMS from reported HR and 97.5% CI.
“adjusted hazard ratios: PBA±BMS v DCBA±BMS: 0.84, 97.5% confidence interval 0.61 to 1.16, P=0.22”
Taken as given: The reported HR is 0.84 and the 97.5% CI is 0.61 to 1.16.; The CI is two-sided at 97.5% (as stated).; The p-value is two-sided and derived from the same model.Method: Used pCI function to derive p-value from HR and CI, assuming a log-normal distribution for the HR.How we recomputed it: pCI(0.84, 0.61, 1.16, 1) - CONSISTENTreported p = .200 · recomputed p = .262Reviewers 1, 2Recompute p-value for primary comparison PBA±BMS vs DES from reported HR and 97.5% CI.
“PBA±BMS v DES: 0.83, 0.60 to 1.15, P=0.20”
Taken as given: The reported HR is 0.83 and the 97.5% CI is 0.60 to 1.15.; The CI is two-sided at 97.5%.; The p-value is two-sided and derived from the same model.Method: Used pCI function to derive p-value from HR and CI.How we recomputed it: pCI(0.83, 0.60, 1.15, 1)
- lowinternal contradictionThe abstract reports 481 participants randomised, but the results section states 480 participants were included in analyses after one participant was removed for lack of consent. This is a minor discrepancy but is explained in the text.
481 participants were randomised (167 (35%) women, mean age 71.8 years (standard deviation 10.8)). ... Of the 480 participants included in the analyses, 167 (35%) were women
Abstractreviewer’s wording - lowinternal contradictionTable 1 shows DES group n=159, but the text states 160 were randomised to DES. The footnote explains this is excluding the participant who did not provide consent, so it is consistent.
Table 1 Baseline characteristics by randomised group ... DES (n=159)* ... * Excluding participant who did not provide consent.
Table 1reviewer’s wording
Overstated conclusions
1 finding · worst lowConclusions that reach past what the paper's own results support — including a significance claim that no longer holds when the statistic is recomputed, and efficacy resting on an unvalidated surrogate endpoint.
- Conclusions only partially backed by the presented evidenceAssessed
5 major claims checked against the paper's own evidence: 1 only partially supported (evidence backs part of the claim; gaps or caveats remain); the rest adequately supported.
- partialReviewer 2The results are likely to be applicable to other countries with similar chronic limb threatening ischaemia populations and healthcare systems.The trial was conducted in UK NHS settings, which may limit generalisability to other healthcare systems, but the pragmatic design supports some generalisability.Evidence: The paper states the trial involved 35 UK NHS vascular units and reflects standard of care across the NHS.
“The results are likely to be applicable to other countries with similar chronic limb threatening ischaemia populations and healthcare systems.”
LimitationsFind in source - supportedReviewers 1, 2Neither DCBA±BMS nor DES conferred significant clinical benefit over PBA±BMS in the femoro-popliteal segment in patients with chronic limb threatening ischaemia.The primary outcome analysis shows no statistically significant difference between groups, with hazard ratios close to 1 and confidence intervals crossing 1.Evidence: Primary outcome: adjusted hazard ratios 0.84 (0.61 to 1.16) for DCBA vs PBA and 0.83 (0.60 to 1.15) for DES vs PBA, with p-values 0.22 and 0.20.
“Neither DCBA±BMS nor DES conferred significant clinical benefit over PBA±BMS in the femoro-popliteal segment in patients with chronic limb threatening ischaemia undergoing endovascular femoro-popliteal, with or without infra-popliteal, revascularisation.”
ConclusionFind in source - supportedReviewer 1The trial was powered to detect a 40% relative reduction in amputation free survival, and the confidence intervals excluded this target.The paper states the target difference was a 40% relative reduction, and the 97.5% confidence intervals for the hazard ratios (0.61 to 1.16 and 0.60 to 1.15) exclude 0.60, which corresponds to a 40% reduction.Evidence: The confidence intervals for the primary outcome are reported as 0.61 to 1.16 and 0.60 to 1.15, which do not include 0.60.
“The 97.5% confidence intervals for the hazard ratios ranged from 0.61 to 1.16 for DCBA±BMS and 0.60 to 1.15 for DES, which include values representing potential benefit and harm. The 97.5% confidence intervals narrowly excluded (or bordered) the 40% relative reduction (13% absolute difference at two years) in the rate of major amputation or death, which was set as the target difference a priori.”
DiscussionFind in source - supportedReviewer 1No differences in serious adverse events were reported between the groups.The paper reports the numbers of serious adverse events and the risk ratios with confidence intervals that include 1, indicating no statistically significant difference.Evidence: Serious adverse events: 16 (10%) in PBA, 9 (6%) in DCBA, 17 (11%) in DES; RR 0.53 (0.22 to 1.30) and 0.98 (0.48 to 2.00).
“No differences in serious adverse events were reported between the groups.”
AbstractFind in source - supportedReviewer 2The trial was powered to detect a 40% relative reduction in amputation free survival.The sample size calculation is described, and the paper notes that the number of primary endpoints required for 90% power was exceeded.Evidence: Sample size calculation states 861 participants would provide 90% power to detect a 40% reduction (HR 0.60). The trial observed 296 primary outcomes, exceeding the required 291.
“861 participants (having 342 primary outcome events) would provide 90% power to detect a reduction in amputation free survival of 40% (hazard ratio 0.60, equivalent to an absolute difference in amputation free survival of 13% at year 2”
Statistical analysisFind in source
Efficacy claim is anchored to an adequate endpoint and a meaningful effect.
- ADEQUATESurrogate endpointThe primary outcome is amputation-free survival, a hard clinical outcome (time to major amputation or death). No surrogate endpoint is used as the primary basis for the efficacy claim.
“The primary outcome was amputation free survival defined as time to first major amputation or death from any cause.”
- ADEQUATEEffect sizeThe trial was powered to detect a 40% relative reduction (hazard ratio 0.60) in amputation-free survival, corresponding to an absolute difference of 13% at two years. The observed hazard ratios were 0.84 and 0.83 with confidence intervals including 1, and the authors explicitly state that the results do not support a clinically important benefit. The effect size is anchored to a prespecified target difference.
“The 97.5% confidence intervals for the hazard ratios ranged from 0.61 to 1.16 for DCBA±BMS and 0.60 to 1.15 for DES, which include values representing potential benefit and harm. The 97.5% confidence intervals narrowly excluded (or bordered) the 40% relative reduction (13% absolute difference at two years) in the rate of major amputation or death, which was set as the target difference a priori.”
Data authenticity concerns
None foundAn adversarial read for patterns associated with data that may not be genuine: results that look too clean, implausibly large effects, duplicated data or images, and methods that do not match the results reported.
Checked — nothing surfaced.
Reporting gaps
1 finding · worst mediumRequired detail the manuscript never states — study design, biological variables, ethics approval and consent, key resources, statistical reporting, data and code availability, and overall transparency.
- Data/code availability incompleteAssessed
The introduction cites three prior publicly funded RCTs (BASIL, BEST-CLI, BASIL-2) and notes the uncertainty about the relative effectiveness of PBA, DCBA, and DES, supported by systematic reviews and guidelines. The rationale is logical: NICE recommended a trial, leading to NIHR HTA funding. The paper acknowledges limitations of prior work (e.g., many studies analysed DCBA and DES together) and addresses them by comparing the three strategies separately.
“To date, three published, publicly funded randomised controlled trials have compared the clinical and cost effectiveness of different revascularisation strategies in chronic limb threatening ischaemia.”
“In 2012, the UK National Institute for Health and Care Excellence (NICE) recommended a randomised controlled trial to compare the clinical and cost effectiveness of PBA, BMS, DCBA, and DES in patients with chronic limb threatening ischaemia.”
“Unlike BASIL-3, many studies analysed DCBA and DES together, even though they are very different technologies with their own advantages and disadvantages in different clinical and anatomic scenarios.”
“To date, three published, publicly funded randomised controlled trials have compared the clinical and cost effectiveness of different revascularisation strategies in chronic limb threatening ischaemia.”
“In 2012, the UK National Institute for Health and Care Excellence (NICE) recommended a randomised controlled trial to compare the clinical and cost effectiveness of PBA, BMS, DCBA, and DES in patients with chronic limb threatening ischaemia.”
“Unlike BASIL-3, many studies analysed DCBA and DES together, even though they are very different technologies with their own advantages and disadvantages in different clinical and anatomic scenarios.”
Randomisation was performed using a secure online system with minimisation to balance key prognostic variables. The sample size calculation is detailed, including a correction for an error in the original macro. Inclusion/exclusion criteria are pre-specified. Blinding was not used, but the trial is open-label by design, which is appropriate for a device trial. The analysis populations (intention-to-treat and per-protocol) are defined. Outlier handling is addressed through the pre-specified analysis population and missing data approach.
“Allowing for 5% attrition and the BASIL-1 survival estimates, 861 participants (having 342 primary outcome events) would provide 90% power to detect a reduction in amputation free survival of 40% (hazard ratio 0.60, equivalent to an absolute difference in amputation free survival of 13% at year 2, corresponding to one out of seven participants needing to benefit from DES or DCBA over PBA (number needed to treat) at the 2.5% significance level (to account for increased type I error risk associated with making two key comparisons) using the artsurv (version 1.0.7) programme in Stata (version 17.0). However, an error was identified in the implementation of the macro used to compute the original sample size, whereby a smaller effect size (than the targeted 40% relative reduction) had been applied for one of the two comparisons. This resulted in an overestimation of the sample size and number of primary outcome events needed. Using the same parameters as in the original calculation (listed above), the corrected sample size was 749 and the number of primary outcome events required was 291.”
“Patients were excluded if they were expected to live for less than six months (pragmatic decision by randomising team) or had undergone an intervention to the target femoro-popliteal vessel within the past 12 months.”
“Participants were randomly assigned (1:1:1) using a secure online system to femoro-popliteal PBA±BMS, DCBA±BMS, or DES as their first revascularisation procedure.”
“Participants, study staff, and investigators were not masked to treatment allocation.”
“However, an error was identified in the implementation of the macro used to compute the original sample size, whereby a smaller effect size (than the targeted 40% relative reduction) had been applied for one of the two comparisons.”
Sex is reported for all participants (167 women, 35%). Age is reported as mean and SD, and also categorised. Demographics include ethnicity, smoking status, comorbidities, and mobility. Since both sexes are included, sex_justified is not applicable. Species/strain and housing conditions are not applicable for a human trial.
“481 participants were randomised (167 (35%) women, mean age 71.8 years (standard deviation 10.8)).”
“Table 1 Baseline characteristics by randomised group”
“Of the 480 participants included in the analyses, 167 (35%) were women, and the mean age was 71.8 years (standard deviation 10.8; , ).”
The paper states ethical approval was obtained from the National Research Ethics Committee, North of Scotland, on 26 August 2015 (15/NS/0070). Informed consent is described (written informed consent was required; one participant was removed for lack of consent). Compliance with the Declaration of Helsinki and Good Clinical Practice guidelines is stated.
“The National Research Ethics Committee, North of Scotland, provided ethical approval on 26 August 2015 (15/NS/0070).”
“Additionally, they needed to speak sufficient English (where translation facilities were insufficient) and to have capacity to provide written informed consent.”
“Declaration of Helsinki and Good Clinical Practice guidelines were followed.”
“The National Research Ethics Committee, North of Scotland, provided ethical approval on 26 August 2015 (15/NS/0070).”
“Additionally, they needed to speak sufficient English (where translation facilities were insufficient) and to have capacity to provide written informed consent.”
“Declaration of Helsinki and Good Clinical Practice guidelines were followed.”
The trial uses drug-coated balloons and drug-eluting stents as investigational products. The paper names the device classes and states that any UK licensed PBA, DCBA, BMS, or DES was permitted, but does not specify manufacturers or lot numbers, which is typical for a pragmatic trial. Statistical software (SAS, Stata) is identified with versions. Antibodies, cell lines, mycoplasma, and organisms are not applicable.
“Any UK licensed PBA, DCBA, BMS, or DES was permitted.”
“All analyses were done in SAS (version 9.4) or Stata (version 18.0).”
“Any UK licensed PBA, DCBA, BMS, or DES was permitted.”
“All analyses were done in SAS (version 9.4) or Stata (version 18.0).”
The primary analysis uses Cox proportional hazards models with adjusted hazard ratios and 97.5% confidence intervals. Secondary outcomes use appropriate methods (competing risks, mixed models, GEE). Exact p-values are reported for the primary outcome. Effect sizes are reported with confidence intervals. Software is identified. Data presentation follows clinical trial conventions (Kaplan-Meier curves, CONSORT diagram). Mathematical plausibility checks are not applicable due to large sample size and continuous outcomes.
“Amputation free survival was analysed using a Cox proportional hazards model to generate an adjusted hazard ratio.”
“HR 0.84 (0.61 to 1.16),† P=0.22”
“HR 0.84 (0.61 to 1.16),† P=0.22”
“Amputation free survival was analysed using a Cox proportional hazards model to generate an adjusted hazard ratio.”
“adjusted hazard ratio from the Cox proportional hazards model 0.84; 97.5% confidence interval 0.61 to 1.16; P=0.22”
“adjusted hazard ratio 0.86, 97.5% confidence interval 0.62 to 1.20”
The data availability statement says 'Requests for data should be directed to the corresponding author. Participant level data will be made available within six months of publication. Requests will be assessed for scientific rigour before being granted. Data will be anonymised and securely transferred. A data sharing agreement might be required.' This is more concrete than a bare 'on request' but lacks a named platform or data access committee. Repository deposit and accession numbers are not applicable for patient-level data. Code sharing is not applicable as no bespoke code is mentioned.
“Requests for data should be directed to the corresponding author. Participant level data will be made available within six months of publication. Requests will be assessed for scientific rigour before being granted. Data will be anonymised and securely transferred. A data sharing agreement might be required.”
“Requests for data should be directed to the corresponding author. Participant level data will be made available within six months of publication. Requests will be assessed for scientific rigour before being granted. Data will be anonymised and securely transferred. A data sharing agreement might be required.”
The trial is registered (ISRCTN14469736). Methods are detailed enough for replication. A CONSORT diagram is included, indicating adherence to CONSORT. All pre-specified outcomes are reported, including negative results. Limitations are discussed extensively. Conclusions are proportional to the evidence. Funding and competing interests are disclosed.
“ISRCTN registry ISRCTN14469736”
“Fig 1 Consolidated standards of reporting trials (CONSORT) diagram.”
“Because of concerns about the safety of paclitaxel, BASIL-3 recruitment was paused between December 2018 and September 2019. Although we have no evidence to suggest that this was the case, we cannot exclude the possibility that patients recruited before and after the pause were different.”
“ISRCTN registry ISRCTN14469736”
“Fig 1 Consolidated standards of reporting trials (CONSORT) diagram.”
“Because of concerns about the safety of paclitaxel, BASIL-3 recruitment was paused between December 2018 and September 2019.”
Registered (1 ID: ISRCTN). Reporting guideline cited: CONSORT.
Broken references and links
None found · partly checkedReferences checked against Crossref, OpenAlex and Retraction Watch for retractions and resolvability, plus declared data and code links probed for whether they resolve to content matching the paper.
Nothing surfaced — but not everything feeding this category ran (missing: data/code link verification), so read this as a partial clean bill.
Checked 24 references by DOI: 23 verified — 1 no DOI (shown, not verified).
- NO DOILower Limb Peripheral Arterial Disease: Diagnosis and Management (Clinical Guideline 147)No DOI in the reference — shown for manual review; not independently verifiable (not a fabrication signal).
Copyediting
4 minorWording, consistency and formatting errors that need correcting before submission.
No major wording or formatting errors. 4 minor suggestions below.
4 copyedit issues flagged: mostly consistency, typo, clarity.
- MINORtypoTable 2, row 'Nether antiplatelet nor anticoagulant'“Nether antiplatelet nor anticoagulant”→ Neither antiplatelet nor anticoagulantTypographical error in the word 'Neither'.
- MINORconsistencyTable 3, footnote for PEDIS“PEDIS=perfusion, extent, depth, infection and sensation”→ Ensure abbreviation expansion is consistent with text (PEDIS is defined in Methods as perfusion, extent, depth, infection, and sensation).Minor inconsistency in comma usage.
- MINORclarityResults, paragraph 1“On 12 December 2018, recruitment was stopped by the trial management group, supported by the trial steering committee, because a meta-analysis had shown excess mortality in patients treated with a paclitaxel DCBA or DES.”→ Consider clarifying that the meta-analysis was external and prompted a safety pause.Clarity improvement.
- MINORconsistencyAbstract, Results“106 of 160 (66%) participants in the PBA±BMS group”→ Ensure consistent use of 'participants' vs 'patients' throughout.Minor inconsistency in terminology.
The published work is robust and well-reported. An informed reader should weigh the vague data availability statement and the minor internal consistency points (481 vs 480 participants) as minor limitations. No erratum or correction appears necessary, but the authors could improve transparency by providing a more concrete data access route and clarifying the participant count discrepancy.
- 1.HIGHdata codeIn the Data availability statement, specify a concrete access mechanism such as a named data access committee or a repository (e.g., YODA, Vivli) and detail the request process and timeline.The current statement is vague and only directs requests to the corresponding author, which is a reporting gap that could hinder reproducibility.
- 2.HIGHdata codeConsider depositing anonymised participant-level data in a recognised repository (e.g., a clinical trial data repository) and provide an accession number or DOI.Depositing data in a repository would enhance transparency and allow independent verification, while respecting patient privacy.
- 3.MEDIUMdata codeIf bespoke analysis code was used, share it in a public repository (e.g., GitHub) with a DOI to allow independent verification of the statistical analyses.Code sharing is currently not reported, and providing it would improve reproducibility.
- 4.MEDIUMreportingClarify the discrepancy between 481 participants randomised (abstract) and 480 participants included in analyses (results) by explicitly stating that one participant was removed for lack of consent in the abstract or results section.The internal contradiction is explained in the text but could confuse readers; making it explicit improves clarity.
- 5.MEDIUMreportingIn Table 1, ensure the DES group n=159 is clearly footnoted as excluding the participant who did not provide consent, to avoid apparent inconsistency with the text stating 160 were randomised.The footnote exists but could be more prominent to prevent misinterpretation.
- 6.MEDIUMreportingAdd a statement about the availability of the statistical analysis plan (SAP) and any protocol amendments in the supplementary materials.Providing the SAP would enhance transparency and allow readers to verify that analyses were pre-specified.
- 7.MEDIUMreportingClarify the specific devices used in each arm (e.g., brand names) in the supplementary materials, as the paper only states that any UK licensed device was permitted.Specifying devices would improve reproducibility and allow readers to assess generalisability.
- 8.MEDIUMstatisticsReport the results of the flexible parametric model for non-proportional hazards in more detail, including the number of knots and model selection process.This would enhance transparency of the sensitivity analysis and allow readers to assess its validity.
- 9.LOWcopyeditIn Table 2, correct the typo 'Nether antiplatelet nor anticoagulant' to 'Neither antiplatelet nor anticoagulant'.This is a minor typographical error that should be fixed for professionalism.
- 10.LOWcopyeditIn Table 3, ensure the abbreviation expansion for PEDIS is consistent with the text (e.g., use 'perfusion, extent, depth, infection, and sensation' consistently).Minor inconsistency in comma usage could be confusing.
- 11.LOWcopyeditIn the Results section, clarify that the meta-analysis that prompted the recruitment pause was external and not part of this trial.This improves clarity about the reason for the pause.
- 12.LOWcopyeditEnsure consistent use of 'participants' vs 'patients' throughout the manuscript.Minor terminology inconsistency could be polished.
The star rating is the report’s one-glance summary. Every paper starts at 5★ and loses stars for the concrete problems the review finds — so a rating is never a vague average, it’s a running total you can read line by line under “How this rating was calculated.”
- Reporting — 8 dimensionseach dimension that fully fails−½★
- each dimension partially met−¼★
- Statistics · Integrity · Claimseach serious problem−1★
- each medium problem−½★
- Citationseach retracted or unverifiable reference−¼★
- Copyeditonly when the manuscript needs a full edit−½★
The rating never drops below 1★, and a demonstrable critical failure (an impossible statistic, a proven ethics violation) caps it at 1★ on its own — so the stars can never look healthy when the verdict is CRITICAL.
The rating draws on a panel of agents. Three independent Kaimen Rigor reviewers grade the eight dimensions below across several independent passes (the shown verdict is their majority vote — steadier than any single run), isolate the paper’s major claims and check its own evidence backs them, and flag integrity concerns. Alongside them, a citation agent resolves every reference against Crossref, OpenAlex, and Retraction Watch; a statistics agent recomputes reported tests; and rule-based checks verify that declared data/code links actually resolve. Full text is required — an abstract-only submission is not analyzed.
Graded against NIH, MDAR, ARRIVE 2.0, CONSORT, EQUATOR, and RRID guidelines. A dimension that doesn’t apply to the study type is skipped, never penalized.
This Kaimen Rigor review is model-assisted and is not a substitute for formal expert review. It complements human evaluation by surfacing potential methodological concerns — verify each finding against the source.