Authors: Anna Beattie, Alan Bagnall, Julie Burn, Rachel O'Leary
Patients commonly present with chest pain and many require investigation to determine whether they have coronary artery disease (CAD). NICE clinical guideline CG95 recommended that CT coronary angiography (CTCA) be used first line to detect the presence of CAD if symptoms were suggestive of typical or atypical angina, or non-anginal chest pain if ECG abnormalities were present. In 2017 NICE also published MTG 32 which appraised the utility of HeartFlow FFRCT to provide both anatomical and functional assessment of CAD within a single non-invasive test.
Our project was to determine the uptake of CG95 recommendations for CTCA and whether the subsequent addition of HeartFlow FFRCT impacted on guideline concordance, on utilisation of other non-invasive functional tests, and on cardiac catheterisation lab utilisation.
Aims and objectives
CG95 recommends that patients with suspected CAD undergo CTCA to detect the anatomical presence and severity of CAD. Thereafter, patients in whom CTCA is inconclusive (<70% stenosis in a branch coronary artery or <50% stenosis of the left main stem) are recommended to undergo a 2nd line functional investigation (such as stress cMRI, stress echocardiography, or nuclear myocardial perfusion scanning), or, as a 3rd line, invasive coronary angiography. HeartFlow CT FFR is a coronary physiology simulation software.
The software provides a non-invasive method of estimating fractional flow reserve (FFR; the gold standard measure of the functional impact of a coronary stenosis on blood flow) using standard CCTA image data. Using HeartFlow would thus enable the anatomical and functional significance of any CAD to be assessed with a single test and single radiation exposure.
Our principle aim was to assess whether the addition of HeartFlow CT FFR testing in patients presenting to a Rapid Access Chest Pain Clinic (RACPC) would a) increase physician uptake of CTCA as a first line investigation of chest pain (CG95) b) reduce the requirement (and therefore cost) of provision of other non-invasive functional tests as a 2nd line (NICE 126.96.36.199.2) c) Increase the efficiency of cardiac cath lab utilisation by ensuring that patients ultimately undergoing invasive investigation were those most likely to require revascularisation rather than a diagnostic procedure only and d) evaluate the patient experience.
We envisaged that the patient experience might be improved by minimising the number of diagnostic tests required to reach definitive diagnosis, avoidance of unnecessary invasive diagnostic procedures and providing opportunities to discuss primary prevention in patients with non-functionally significant CAD which may in turn reduce morbidity and mortality. CG95 acknowledges the SCOTHEART trial (188.8.131.52.4) which demonstrated a 38% reduction in fatal and non-fatal MI when patients were randomised to cardiac CT rather than standard care.
Reasons for implementing your project
Prior to the 2016 up-date of the NICE guidelines, cardiac CT was only recommended for patients with a low probability of significant CAD. The update suggested that patients at any risk level should undergo cardiac CT as an initial investigation. The recommendation was based on the cost effectiveness model, demonstrating that CTCA had the lowest cost per correct diagnosis for all levels of pre-test likelihood.
A functional test was recommended as the 2nd line investigation if CT was non diagnostic or inconclusive. Conventional functional tests are not convenient for patients and increase the overall number of investigations and clinic appointments for those patients who ultimately get referred for invasive angiogram.
Prior to HeartFlow, such patients were required to attend for separate functional investigations with attendant additional risk, inconvenience and delays to final diagnosis. Clinicians were therefore reluctant to use cardiac CT and instead referred directly for either functional or invasive testing in higher risk groups. HeartFlow is coronary physiology simulation software used for the analysis of computerised tomography DICOM data to simulate coronary physiology. The software provides a non-invasive method of estimating fractional flow reserve (FFR) using standard coronary CT angiography (CCTA) image data.
Using HeartFlow would enable the anatomical and functional significance to be assessed with no additional tests and should improve the patient pathway and efficiency. A baseline assessment was made of the patient pathway and investigations utilised in the 6 months prior to HeartFlow adoption The Newcastle upon Tyne Hospitals NHS Foundation Trust has over 2250 beds, performs over 1600 cardiac CTs per year and around 3000 percutaneous coronary interventions per year.
A patient experience survey was initiated for all investigations used to investigate for coronary disease to assess patient acceptability. Stakeholders were engaged through a series of educational meetings, quality improvement project presentations and case review sessions.
How did you implement the project
Newcastle Hospitals was already equipped with the infrastructure to acquire and report cardiac CT and had previously participated with Heartflow in their pivotal efficacy and cost effectiveness trials. The company therefore approached Newcastle as a pilot site for the Innovation Technology Payment scheme where funding is available directly from NHS England.
No additional capital investment or significant additional training was required, although increased CT appointment capacity was required. Training and educational events for the clinical adoption of HeartFlow were organised with the support of the company to ensure the technology was used effectively and its limitations recognised. However there was still a considerable investment of time and resources from both NHS staff and the company to ensure IT governance was secure in transferring anonymised patient data for off-site analysis.
Overall it took approximately 4 months to embed Heartflow FFRCT into practice and the first clinical scans were sent in August 2018. To reduce costs, only those cases identified as having indeterminate CAD by the health profession reading the CT were sent for HeartFlow analysis. Results were returned within 48 hours and incorporated into the radiology report. Snapshot images were also made available on PACS and an interactive 3-D model provided on a secure website. Patients were then seen in clinic again to discuss their results and next steps which might include medical management or referral for revascularisation. This project was designed to assess the impact of a novel technology on a real-life clinical pathway which will permit more accurate financial models. The data analysis was conducted by staff from the Northern Medical Physics and Clinical Engineering Department (NMPCE) and supported by the NENC AHSN.
Data was collected from local hospital systems and used to assess the patient pathways and investigations utilised before and after the adoption of HeartFlow. The baseline assessment covered a 6 month period approximately one year prior to HeartFlow adoption, with a second 6 month period following the introduction of HeartFlow.
Aim 1 – increase proportion of patients getting cardiac CT as first line investigation and reduce functional and invasive investigations as initial test:
Prior to HeartFlow, first line investigations were: Cardiac CT 131 of 470 patients (27.9%) Functional test 81/470 (17.2%) Invasive angiogram 85/470 (18.1%)
Post HeartFlow this aim was achieved, first line investigations were: Cardiac CT 225/496 (45.4%) Functional test 48/496 (9.7%) Invasive angiogram 60/496 (12.1%).
Aim 2 – Reduce the requirement for and provision of non-HeartFlow non-invasive functional tests (DSE, stress cMRI, NMPS)
Prior to HeartFlow 6/131 patients had non-HeartFLow functional testing as the 2nd line investigation following CTCA 16/131 underwent invasive angiography as a 2nd line test 50% of patients who ultimately underwent invasive angiography underwent revascularisation.
Following HeartFlow 89/225 had HeartFlow CT FFR testing following CTCA 1/225 patients had invasive testing without a prior functional test (patient choice).
Aim 3 – reduce the rates of diagnostic only invasive testing.
13/15 (86.7%) patients who had invasive angiography following HeartFlow assessment were re-vascularised compared with 8 of 16 patients (50%) following cardiac CT alone prior to HeartFlow
Aim 4 – shorten the pathway. In our experience the time on the pathway was similar (54 days pre HeartFlow and 55 days post HeartFlow.
There was no reduction in clinic appointment with an average of 1.9 per patient. This is likely to be due to the higher reliance of invasive testing earlier in the pathway prior to HeartFlow.
Therefore we have demonstrated better adherence to guidelines and cath lab utilisation without delays in care or increase in clinic appointments. Our patient experience survey is on-going but initial data demonstrates: Time from clinical appointment to CT scan was rated as: 17/25 excellent, 7/25 satisfactory, 1/25 poor Maintaining dignity: 22/25 excellent, 1/25 poor, 2/25 satisfactory Discomfort during the test: 17/25 acceptable, 3/25 satisfactory, 1/25 poor Willingness to have the test again: 22/25 yes, 2/25 no and 1/25 not sure.
Key learning points
- Additional of functional testing to CT coronary angiogram improves adherence to NICE guidelines in CG95 and increases proportion of patients undergoing revascularisation at invasive angiography
- Organisations are required to have a high-quality cardiac CT service and will need to collaborate closely with the company to ensure IT governance standards are met which takes time and resources. There may be a need to increase capacity for performing cardiac CTs.
- Patients find cardiac CT an acceptable test and we intend to assess whether other tests to investigate coronary disease are also acceptable.
- Engagement with the stakeholders in necessary to design a pathway and facilitate the most effective use of resources. On-going educational events and case reviews are important.
- As a novel technology, access through the innovation technology payment scheme facilitated earlier adoption and an opportunity to assess the impact on a clinical service. Through the medtech mandate, reimbursement through commissioners should be approved for centres looking to implement HeartFlow.
- In the future we intend to assess the cost implications of change in pathways to better predict the financial impact.