Article Text

Original research
Impact on stable chest pain pathways of CT fractional flow reserve
  1. Rachel A O'Leary1,
  2. Julie Burn1,
  3. Samuel G Urwin1,
  4. Andrew J Sims1,2,
  5. Anna Beattie3,
  6. Alan Bagnall2,4
  1. 1 Northern Medical Physics and Clinical Engineering, Newcastle Upon Tyne Hospitals NHS Foundation Trust, Newcastle Upon Tyne, UK
  2. 2 Faculty of Medical Sciences, Newcastle University, Newcastle upon Tyne, UK
  3. 3 Radiology, Royal Victoria Infirmary, Newcastle upon Tyne, UK
  4. 4 Cardiology, Newcastle Upon Tyne Hospitals NHS Foundation Trust, Newcastle Upon Tyne, UK
  1. Correspondence to Dr Andrew J Sims, Northern Medical Physics and Clinical Engineering, Newcastle Upon Tyne Hospitals NHS Foundation Trust, Newcastle Upon Tyne NE7 7DN, UK; andrew.sims5{at}


Objectives To evaluate the impact of introducing CT fractional flow reserve (FFRCT) on stable chest pain pathways, concordance with National Institute for Health and Care Excellence (NICE) chest pain guidelines, resource usage and revascularisation of patients from a tertiary UK cardiac centre rapid access chest pain clinic (RACPC).

Methods Single-centre before and after study comparing data from electronic records and Strategic Tracing Service of all RACPC patients attending between 1 July 2017 and 31 December 2017, and 1 August 2018 and 31 January 2019.

Results Two hundred and sixty-eight and 287 patients (overall mean age 62 years, range 26–89 years, 48.3% male), were eligible for first-line CT coronary angiography (CTCA) pre-FFRCT and post-FFRCT, respectively. First-line CTCA use per NICE Guideline CG95 increased (50.6% pre-FFRCT vs 75.7% post-FFRCT, p<0.001). More patients reached pathway endpoint (revascularisation or assumed medical management) after one investigation (74.9% pre-FFRCT vs 84.9% post-FFRCT, p=0.005). There were fewer stress (22.8% pre-FFRCT vs 7.7% post-FFRCT, p<0.001) and rest (10.4% pre-FFRCT vs 4.2% post-FFRCT, p=0.007) myocardial perfusion scans and diagnostic-only angiograms (25.5% vs 13.7%, p<0.001). Despite fewer invasive procedures (29.3% pre-FFRCT vs 17.6% post-FFRCT, p=0.002), revascularisation rates remained similar (10.4% pre-FFRCT vs 8.8% post-FFRCT, p=0.561). Avoiding invasive investigations reduced inpatient admissions (39.0% pre-FFRCT vs 24.3% post-FFRCT, p<0.001). Time to revascularisation was unchanged (153.5 days pre-FFRCT vs 142.0 post-FFRCT, p=0.925). Unplanned hospital attendances, emergency admissions and adverse events were similar.

Conclusions FFRCT adoption was associated with greater compliance with NICE guidelines, reduced invasive diagnostic angiography, planned admissions and needing more than one test to reach a pathway endpoint.

  • computed tomography angiography
  • coronary artery disease
  • chest pain
  • coronary stenosis
  • percutaneous coronary intervention

Data availability statement

Data may be obtained from a third party and are not publicly available. This study used identifiable data sets collected for the purpose of service evaluation within the same care organisation.

This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See:

Statistics from

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.


  • Fractional flow reserve CT (FFRCT) facilitates the anatomical and functional assessment of coronary artery disease (CAD) using a single, non-invasive radiation exposure, which can inform clinical management decisions.


  • Implementation of FFRCT in diagnostic pathways for stable chest pain reduced the number of diagnostic-only invasive angiograms, planned hospital admissions and need for myocardial perfusion scans. Patients were more likely to need only one diagnostic test to reach a pathway endpoint, which was achieved with no increase in time to revascularisation or adverse events.


  • This study confirms the feasibility of National Institute for Health and Care Excellence CG95 in routine clinical practice and supports evidence that CT coronary angiography (CTCA) and FFRCT is a safe alternative for guiding revascularisation decisions in patients presenting with new-onset chest pain without known CAD. Guideline adoption will likely require additional CTCA capacity but may release capacity for revascularisation by reducing the number of diagnostic-only invasive procedures.


Patients presenting with chest pain commonly require investigation to determine if they have coronary artery disease (CAD) and to plan revascularisation. Simulation software is now available to non-invasively estimate coronary fractional flow reserve from CT angiography (FFRCT). FFR is the reference standard measure of the functional impact of a coronary stenosis on blood flow.1 FFRCT uses standard CT coronary angiography (CTCA) image data to create a three-dimensional model of the coronary arteries and associated stenoses to predict this impact. FFRCT analysis therefore facilitates the anatomical and functional assessment of CAD within a single, non-invasive radiation exposure and reduces the requirement for invasive angiography when employed in diagnostic pathways for stable chest pain.2–7

Prior to the 2016 update, the UK National Institute for Health and Care Excellence (NICE) Clinical guideline CG95 (‘Recent-Onset Chest Pain of Suspected Cardiac Origin: Assessment and Diagnosis’) recommended CTCA only for patients with low pretest probability of significant CAD.8 The updated guidance recommends CTCA first line for all patients at any risk level to detect the presence of CAD if symptoms are suggestive of typical or atypical angina, or in patients with non-anginal chest pain if ECG abnormalities are present. If CTCA shows CAD of uncertain significance or is non-diagnostic, a second-line functional investigation is recommended, such as nuclear myocardial perfusion scan (NMPS), stress echocardiography or stress MRI perfusion imaging. If results remain inconclusive, invasive coronary angiography is then offered.

In 2017, NICE published MTG32,9 which appraised the utility of FFRCT for anatomical and functional assessment of CAD. A guidance update in 2021 featured economic analysis estimating a cost saving of £391 per patient for FFRCT-based assessment pathways, potentially saving the National Health Service (NHS) >£9 million annually. In August 2018, the Newcastle on Tyne Hospitals NHS Foundation Trust (NuTH) was designated a pilot site for FFRCT adoption under the Innovation and Technology Payment (ITP) programme funded by NHS England. The service was made available to patients presenting to the rapid access chest pain clinic (RACPC).

This retrospective service evaluation aimed to assess the effect of introducing FFRCT on (1) uptake of CTCA for first-line investigation of stable chest pain, (2) requirement for other non-invasive functional tests and (3) efficiency of use of the cardiac ‘cath lab’ (defined by the proportion of patients undergoing invasive investigation having immediate follow-on revascularisation). The study also assessed the impact on pathway times, hospital attendances and adverse events.



The study was registered as a service evaluation on the NuTH Clinical Effectiveness Register and jointly undertaken by the Directorates of Northern Medical Physics and Clinical Engineering, Radiology and Cardiothoracic Services.


NuTH is an NHS Trust with >2250 beds providing tertiary cardiac services to the region, performing >1600 CTCA scans and around 2800 percutaneous coronary interventions (PCIs) per year.

The RACPC assesses patients with suspected angina within 14 days of referral with capacity for approximately 1000 patients annually. Nine consultant cardiologists and one chest pain specialist nurse managed patients attending the RACPC during the study.

Inclusion and exclusion criteria

Eligible patients were ≥18 years with stable chest pain with an appointment at the RACPC in one of two recruitment periods: before the introduction of FFRCT, ‘pre-FFRCT’ between 1 July 2017 and 31 December 2017, or after FFRCT adoption under the ITP programme, ‘post-FFRCT’ between 1 August 2018 and 31 January 2019, thus avoiding a cross-over period when FFRCT was used infrequently. The index appointment for each patient was the earliest appointment or earliest appointment within any 3-month period where multiple referrals were made.

Patients with previously confirmed CAD (eg, previous myocardial infarction (MI), revascularisation or angiography) were excluded from analysis as CG95 recommends first-line non-invasive functional testing in this group. Because our study was designed to examine the influence of FFRCT on radiological investigation pathways, patients discharged from the RACPC without radiological tests were excluded from analysis. It was assumed such patients had non-cardiac chest pain diagnosed clinically by the attending cardiologist.

Data sources

Data were obtained retrospectively from the RACPC database for the pre-FFRCT and post-FFRCT time periods, plus 3 months prior to each, to exclude patients with index appointments before the recruitment periods. Patient-level identifiers (NHS number and local medical record number) were used to quantify healthcare encounters from local administrative data, including accident and emergency (A&E), outpatient attendances and inpatient admissions. Clinical events identified from inpatient data included acute MI, revascularisation with coronary artery bypass grafting (CABG) and invasive cath lab procedures including diagnostic angiography with or without pressure wire functional testing, and revascularisation with PCI. Patient-level identifiers were used to extract records from the radiology information system to ensure all relevant investigations were identified, including CTCA, MRI and NMPS (stress and rest) scans, intravascular ultrasound and optical coherence tomography. Investigations and events were identified using International Classification of Diseases, 10th Revision, and Office of Population Censuses and Surveys (OPCS) Classification of Interventions and Procedures codes, and standardised exam descriptions used locally (see online supplemental file). An in-house database was checked for stress echocardiography investigations. Additional data sources included the Strategic Tracing Service to identify patient deaths and the software provider’s web application for FFRCT.

Supplemental material

All healthcare encounters and clinical event data between 1 July 2017 and 30 September 2019 were extracted. This allowed 8 months of follow-up for all patients, which was deemed appropriate for completion of related investigations.

Data cleaning

Records were excluded if duplicated, unrelated to an index appointment, or if patient identifiers were missing or invalid. Patients with known CAD were excluded if a previous inpatient episode contained relevant OPCS or ICD codes, or existing CAD was recorded at the RACPC (see online supplemental file). As the latter used a text-matching algorithm, the authors manually verified exclusions.

Definitions and outcomes

Pathway start point was defined as the date of RACPC attendance when a decision was made to refer for additional radiological investigation. Pathway endpoint was defined as either revascularisation (PCI or CABG), death or MI without revascularisation, or an assumed decision to manage medically (ie, no further investigations, procedures or adverse events recorded during follow-up).

Investigations were ordered chronologically and only the earliest instance of each considered. Times between RACPC attendance and investigations, and the pathway endpoint were calculated for each patient. Overall numbers and types (emergency/non-emergency, cardiac/non-cardiac) of inpatient and outpatient attendances, number of cardiac radiological investigations and invasive procedures were determined. Safety outcomes were frequencies of MI, death and unplanned hospital attendance.

For analysis, related investigations and procedures were grouped into ‘other’ non-invasive functional tests (NMPS, stress echo and stress MRI); diagnostic-only cath lab procedures (angiography or angiography plus pressure wire); and invasive revascularisation (PCI or CABG).

Statistical analysis

Analysis used ‘R’ statistical programming language10 and a significance level of 0.05 using Bonferroni correction for related groups of observations. Comparisons of pre-FFRCT and post-FFRCT cohorts used Student’s t-test, χ2 test, Mann-Whitney U test or Fisher’s exact test, as appropriate. Tests were two-sided. Kaplan-Meier analysis and Cox proportional hazards test were used to compare time to revascularisation events.

Patient and public involvement

This was a retrospective evaluation of standard care pathways and routinely collected data. Therefore, no patients or members of the public were directly involved.


Study population

In total, 995 patients (483 pre-FFRCT and 512 post-FFRCT) attended the RACPC. After data cleaning and exclusion of patients with known CAD, or discharge without further investigation, 268 pre-FFRCT and 287 post-FFRCT patients were eligible for CTCA first line (figure 1). The proportions of patients discharged without additional investigation were similar (pre-FFRCT 37.2% (n=159) vs 33.3% (n=143) post-FFRCT, p=0.225). Of those discharged, two died and one suffered an MI in the pre-FFRCT cohort, and one died in the post-FFRCT cohort.

Figure 1

Recruitment flowchart for pre-FFRCT (left) and post-FFRCT (right) time periods. CTCA, CT coronary angiography; FFRCT, CT fractional flow reserve; MI, myocardial infarction; RACPC, rapid access chest pain clinic.

Patient characteristics and extent of missing data are summarised in table 1. Smoking was more prevalent in the post-FFRCT cohort (p=0.003). Hyperlipidaemia and hypertension were also more prevalent (p=0.0063 and 0.035, respectively), although not significantly following Bonferroni correction (Bonferroni threshold 0.0063).

Table 1

Clinical characteristics of patients included in the analysis.

After referral for investigation, 12 eligible patients (9 pre-FFRCT and 3 post-FFRCT) experienced an MI before planned first-line investigations and were excluded from the main analysis. The risk of event prior to first-line investigation did not differ (9/268, 3.4% pre-FFRCT vs 3/287, 1.0% post-FFRCT; p=0.08).

Investigations and pathways

The sequence and number of tests between RACPC attendance and pathway endpoint are shown in figures 2 and 3. Flow diagrams detailing a chronological breakdown of investigations and times between them are shown in online supplemental file 1.

Figure 2

Pre-FFRCT: sequence of testing between RACPC attendance and endpoint. CABG, coronary artery bypass grafting; CTCA, CT coronary angiography; FFRCT, CT fractional flow reserve; NM, nuclear medicine; PCI, percutaneous coronary intervention; RACPC, rapid access chest pain clinic.

Figure 3

Post-FFRCT; sequence of testing between RACPC attendance and endpoint. CTCA, CT coronary angiography; FFRCT, CT fractional flow reserve; NM, nuclear medicine; PCI, percutaneous coronary intervention; RACPC, rapid access chest pain clinic.

Time from RACPC attendance to first-line investigation did not differ between cohorts (p=0.98). Median time was 52 days (IQR 35–74) pre-FFRCT vs 53 days (IQR 46–59) post-FFRCT.

The proportion of patients offered CTCA as a first-line investigation increased (131/259, 50.6% pre-FFRCT vs 215/284, 75.7% post-FFRCT; p<0.001).

The impact of FFRCT on cardiac test referral within 8 months of attendance at the RACPC is shown in table 2. After introduction of FFRCT, more patients reached their pathway endpoint (revascularisation or assumed decision to manage medically) following one investigation (74.9% pre-FFRCT vs 84.9% post-FFRCT, p=0.005; table 3).

Table 2

Cardiac test usage within 8 months of attendance at the rapid access chest pain clinic, pre-FFRCT/post-FFRCT adoption

Table 3

Number of investigations after rapid access chest pain clinic visit required to reach pathway endpoint

Non-invasive cardiac investigations

After introducing FFRCT, use of stress NMPS decreased from 22.8% to 7.7% of patients (p<0.001). Use of rest NMPS also decreased (10.4% pre-FFRCT vs 4.2% post-FFRCT), although with Bonferroni correction, this was not statistically significant (p=0.007). The use of stress cardiac MRI (cMRI) and stress echocardiography was unchanged.

The proportion of patients with any functional investigation (including FFRCT) before cath lab attendance increased from 5/86 (5.8%) pre-FFRCT, to 17/59 (28.8%) post-FFRCT (p<0.001).

Invasive coronary angiography and revascularisation

The overall proportion of patients undergoing diagnostic-only invasive angiography decreased significantly after introducing FFRCT (25.5% pre-FFRCT vs 13.7% post-FFRCT, p<0.001). This reduced the overall proportion of patients requiring cath lab procedures from 86/259 (33.2%) to 59/284 (20.8%) (p=0.001).

The proportion of patients undergoing invasive coronary angiography as a first-line investigation was reduced (65/259, 25.1% pre-FFRCT vs 41/284, 14.4% post-FFRCT; p=0.002), as was, non-significantly, the use of invasive coronary angiography as a second-line investigation following initial CTCA (16/131, 12.2% pre-FFRCT vs 15/215, 7.0% post-FFRCT; p=0.12).

The proportion of patients revascularised remained similar between cohorts (27/259, 10.4% pre-FFRCT vs 25/284, 8.8% post-FFRCT; p=0.561).

Pre-FFRCT, the revascularisation rate was 8/16 (50.0%) among patients with first-line CTCA only before cath lab attendance. After first-line CTCA plus FFRCT prior to cath lab attendance, the revascularisation rate was 12/14 (85.7%), although this was not statistically significant (p=0.058).

Time to revascularisation

The time taken to reach an endpoint of revascularisation was unchanged between cohorts (figure 4), with a median time of 153.5 days (maximum of 231 days) pre-FFRCT, vs 142.0 days (maximum of 236 days) post-FFRCT (p=0.925).

Figure 4

Kaplan-Meier plot comparing time to revascularisation for pre-FFRCT and post-FFRCT cohorts. The Cox proportional hazards model found no significant difference in the risk of revascularisation between cohorts. The coloured bands are 95% CIs. FFRCT, CT fractional flow reserve.

Hospitalisation and adverse events

The proportion of patients with ≥1 planned inpatient admissions reduced from 39.0% pre-FFRCT to 24.3% post-FFRCT (p<0.001) due to fewer elective invasive angiography procedures. Unplanned hospital attendances did not differ with 5.8% of pre-FFRCT patients attending A&E at least once for cardiovascular reasons, compared with 4.9% of post-FFRCT patients (p=0.705). Both cohorts had similar rates of ≥1 emergency admissions (6.9% pre-FFRCT vs 7.7% post-FFRCT, p=0.745), table 4.

Table 4

Hospital attendance for cardiovascular emergency or elective treatment within 8 months of rapid access chest pain clinic attendance

Of patients completing first-line investigations, there were no deaths and one MI in the pre-FFRCT cohort, and one death and two MIs post-FFRCT. All MIs were treated with revascularisation, and the death in the post-FFRCT cohort occurred following revascularisation.


Summary of findings

We assessed RACPC diagnostic pathways and concordance with NICE Guideline CG958 before and after introducing FFRCT at a large NHS tertiary cardiac centre. FFRCT adoption was associated with improved compliance through increased CTCA as a first-line investigation and reduced use of some alternative functional tests. Elective inpatient admissions were reduced, as were the number of diagnostic-only invasive angiography procedures. The proportion of patients having percutaneous revascularisation after an invasive angiogram increased. The overall proportion of patients with stable chest pain requiring revascularisation was unchanged. Unplanned hospital admissions were not increased.

Within a single non-invasive test, FFRCT provides anatomical and functional information based on the modelled impact of epicardial stenoses. It does not identify patients with chest pain and evidence of ischaemia in the absence of fixed epicardial stenosis, such as those with microvascular dysfunction or vasospastic angina who may require alternative invasive or non-invasive assessment.

NICE economic modelling estimates potential cost savings of £391 per patient compared with alternative stable chest pain pathways,9 though this saving may only be achieved in those otherwise undergoing invasive angiography. Despite reduced use of invasive coronary angiography and other diagnostic modalities, similar to that observed in our own study, cost savings were not substantiated in the large UK randomised Fractional Flow Reserve Derived From Computed Tomography Coronary Angiography in the Assessment and Management of Stable Chest Pain (FORECAST) trial.7 Of note, first-line CTCA use was higher (63%) in the standard care arm of FORECAST compared with our pre-FFRCT cohort (51%), which may reflect local differences in guideline adoption, service provision or the influence of trial participation.

In common with previous studies,2–7 we observed fewer patients undergoing invasive, diagnostic-only angiography. Despite performing fewer invasive angiograms, revascularisation rates were similar between cohorts, with no increase in adverse events. This suggests that patients received the interventions they needed when triaged using FFRCT, while avoiding the associated risks of invasive angiography. Our data and those of other studies4 11 support the concept that patients with prior positive FFRCT results are more likely to undergo immediate follow-on PCI if referred for invasive assessment. Prioritising cath lab use for revascularisation (rather than diagnosis) is important to minimise revascularisation waiting times for those with proven stable CAD with symptoms inadequately controlled with medical therapy. This may release capacity for urgent or emergency angiography in patients with unstable coronary syndromes. Shifting towards cath lab usage for revascularisation will be particularly important to manage the backlog of patients awaiting PCI following the COVID-19 pandemic.

We observed reduced use of nuclear stress tests, previously our most used functional test. With resource constraints, this has important implications for service planning and may alleviate waiting times for nuclear investigations for other conditions. Reduced use of nuclear stress tests may particularly benefit patients as an abnormal scan requires a second resting test on a separate visit. Therefore, FFRCT reduces not only total patient visits but also patient exposure to ionising radiation. The potential reduction in radiation exposure associated with FFRCT is quantified elsewhere.5 12 Despite differences in study design, we found our decreased reliance on diagnostic cath lab procedures and non-FFRCT functional tests to be concordant with other studies7 and NHS hospitals.13 However, it is challenging to disaggregate the influence of increased CTCA use alone, compared with CTCA plus selective FFRCT.

FFRCT analysis is available within 12 hours, potentially enabling communication of diagnostic and treatment decisions to patients within a day of their CTCA. Conversely, pathways that ‘layer’ diagnostic tests (eg, CTCA followed by functional testing with stress cMRI, stress echocardiography or NMPS) typically have additive waiting times, potentially taking longer to reach diagnosis. Although we saw no difference in adverse events while patients awaited investigation, longer waiting times may prolong anxiety for patients and may delay return to work, travel or investigation of alternative causes of symptoms.

There was no significant difference in either the overall time patients spent on pathways or the median number of hospital attendances and procedures. However, this disguised changes in how attendances and investigation types were distributed. There was greater reliance on invasive testing earlier in the pathway before adopting FFRCT, with more patients undergoing first-line invasive coronary angiography. Changes in pathway time may also have been masked because waiting times for clinic review or elective revascularisation procedures did not change. We found a greater proportion of patients reached pathway endpoint after one test, which may be important for patient satisfaction and planning new ‘one-stop-shop’ services for diagnosis and management of new-onset stable chest pain.

Strengths and limitations

Our study is from a large UK NHS Trust and provides a comprehensive picture of all cardiac investigation modalities and hospital settings used to assess and diagnose stable chest pain. The work was retrospective and therefore reflects a ‘real-world’ change in pathways, without the influence of trial participation, although as an uncontrolled study, there were potential confounders.

Using NHS numbers and national datasets permitted identification of all patient deaths, although cause of death information was not available. RACPC patients undergo subsequent investigations at our hospital, and our records, combined with manual review of coded events, enabled identification of all encounters. We could not, however, identify additional investigations or emergency admissions at other hospitals. Exclusion of patients ineligible for CTCA under CG95 was achieved using local data and clinical history, although CAD events recorded outside of NuTH would rely on patient recall. However, it is likely that missed exclusions would occur proportionally across both cohorts and would not, therefore, alter our results.

The two cohorts differed significantly in certain clinical characteristics, with higher proportions of hyperlipidaemia, hypertension and smoking in the post-FFRCT cohort. There were no changes in RACPC referral criteria during the study, and it is unlikely that these factors would have affected investigation choices. However, we could not determine the factors influencing first-line or second-line investigation choices. Patients with persistent symptoms and non-obstructive CAD on CTCA or FFRCT may have undergone functional assessment for identification of ischaemia in the context of normal coronary arteries (eg, due to microvascular dysfunction) or because the physician suspected a false-negative result. Increased awareness of non-obstructive causes of ischaemic chest pain may have influenced second-line referral patterns.

Data relating to symptoms in the RACPC dataset were recorded as a free-text summary rather than a standardised scale and so were unsuitable for analysis, and no information on quality of life was available.

Similar results may not be realised in smaller hospitals or those without the necessary high-quality cardiac CT service capacity or compliance with information governance standards. The implications of the latter were presented in NICE shared learning from Liverpool Heart & Chest Hospital NHS Foundation Trust14 and from Bath Hospital.13


Introduction of FFRCT was associated with greater compliance with NICE CG95, reduction in invasive diagnostic coronary angiography, planned hospital admissions and need for more than one diagnostic test to reach a stable CAD pathway endpoint.

Data availability statement

Data may be obtained from a third party and are not publicly available. This study used identifiable data sets collected for the purpose of service evaluation within the same care organisation.

Ethics statements

Patient consent for publication

Ethics approval

This study involves human participants. The study was a retrospective service evaluation of standard care pathways and routinely collected data. It took place in our own organisation that delivered the patient care, was carried out in accordance with clinical governance guidelines and was approved by the NHS Trust’s Caldicott Guardian. It was not intended to generate new, generalisable knowledge, was therefore not research and did not require ethical review.The study was a retrospective service evaluation of standard care pathways and routinely collected data within our own organisation. Approval was granted by the NHS Trust’s Caldicott Guardian for the use of patient data in this way.


The authors thank David Astrop for providing radiology information system data, Susan Hollands for providing commissioning data and Tim Irvine for providing stress echo data.


Supplementary materials

  • Supplementary Data

    This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.


  • Twitter @doctorradiology

  • Contributors ABe, JB, RO and SU were responsible for the design of the study. JB, RO and SU were responsible for the protocol and data management. JB, RO, SU and AJS were responsible for the analysis of the data. ABe, ABa, JB, RO and AJS were responsible for data review and interpretation of the analysis. ABe, ABa, JB, RO, SU and AJS drafted, revised and approved the final version of the manuscript. AJS is the guarantor.

  • Funding The study was facilitated by the Academic Health Science Network for the North East and North Cumbria. JB, RO and AJS are employed by the Northern Medical Physics and Clinical Engineering directorate of The Newcastle upon Tyne Hospitals NHS Foundation Trust, which hosts an External Assessment Group funded by the National Institute for Health and Care Excellence.

  • Competing interests ABe has received honoraria for travel and speaker engagements from HeartFlow. ABa is UK chief investigator of the P4 trial that is part-funded by HeartFlow.

  • Patient and public involvement Patients and/or the public were not involved in the design, conduct, reporting or dissemination plans of this research.

  • Provenance and peer review Not commissioned; externally peer reviewed.

  • Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.