The Use of Artificial Intelligence-Enhanced Electrocardiograms in the Chest Pain Clinic to Risk Stratify Patients, Provide Rapid Reassurance and Enable a Low-Cost Clinical Pathway
试验速览
- 阶段
- 不适用
- 状态
- 尚未招募
- 发起方
- 入组人数
- 4,000
- 试验地点
- 2
- 主要终点
- The healthcare utilisation costs of each pathway over a one-year period.
研究概览
简要总结
Our current pathway for investigating patients with chest pain differs depending on if the pain is cardiac sounding or not. National guidelines advise us that patients with non-cardiac chest pain do not need further tests beyond seeing a clinician and having a test called an electrocardiogram (ECG), but often we do unnecessary additional investigations for these patients. Some of the tests we do involve invasive procedures or radiation, which have associated risks. We have recently developed an artificial intelligence (AI) ECG technology, which has been shown in various studies to reliably predict risk of heart disease, including heart attacks and death, from just one AI-ECG reading, which is a test that is painless with no radiation. We have shown that this AI-ECG is more accurate at predicting outcomes than the standard risk prediction models we use now.
We propose investigating whether this new technology helps to nudge our clinicians to avoid risk averse behaviour so that they undertake fewer unnecessary investigations, by comparing its use to our current treatment pathway.
The main questions our study aims to answer are:
- Will an AI-ECG assisted chest pain clinic pathway result in lower healthcare resource costs than the standard pathway?
- Will an AI-ECG assisted chest pain clinic pathway reduce the time from referral to diagnosis and treatment?
- Will an AI-ECG assisted chest pain clinic pathway perform equally as well as our current pathway in resolving symptoms and preventing future heart disease?
We will randomly allocate half of the patients with non-cardiac pain in our chest pain clinics to have an AI-ECG, using it to determine which patients are low risk and which are higher risk. Feedback from the analysis will be given to the assessing clinician, with our hypothesis being that patients triaged as low risk by the AI-ECG will be reassured and discharged from clinic, with patients identified as higher risk undergoing further investigation.
The other half of patients not allocated to receive an additional AI-ECG test will be managed as usual. All patients' clinical assessment and management plans will be assessed by a Consultant Cardiologist, who will not have access to the AI-ECG data so that there is assurance that all assigned management pathways are clinically safe and appropriate. We will compare the cost spent for each group at one year, as well as how quickly we can provide a diagnosis/management plan to patients, the number of cardiac events and the number of patients prescribed cholesterol and blood pressure lowering medications. We propose that this study will allow us to safely reassure more patients with chest pain more quickly.
详细描述
Our group has been at the forefront of developing AI-ECG models that are actionable, biologically plausible and explainable, with the aim of fostering clinical trust and maximising their potential for integration into routine clinical care. Recently, we developed the artificial-intelligence risk-estimation (AIRE) platform, comprising a series of AI-ECG models capable of predicting time-to-mortality, and further disease-specific AIRE models encompassing a broad range of future cardiovascular conditions from a single 12-lead ECG.
The AIRE platform was developed using the Beth Israel Deaconess Medical Centre (BIDMC) dataset, comprising more than 1.1 million ECGs from a secondary care population in Boston, USA. The tool has been externally validated across five large international cohorts from the USA, UK, and Brazil, encompassing diverse demographics, a range of baseline cardiovascular risks, including both primary and secondary care, as well as volunteer populations.
AIRE generates patient-specific survival curves using data from a single 12-lead ECG and can predict time-to-death. In our validation studies, AIRE demonstrated strong predictive performance for cardiovascular (CV) mortality, achieving a concordance-index (c-index) of 0.844 (95% confidence interval, 0.839-0.849). This outperformed current predictive models based on demographic data and traditional risk factors alone, which had a c-index of 0.733 (0.726-0.740).
We believe the rapid access chest pain clinic provides an ideal setting in which to evaluate the clinical translation of this validated AI-ECG technology. Risk prediction is central to clinical-decision making in this population, making it well-suited for assessing the added value of AI-enhanced tools. In low-risk cohorts such as our proposed study population, the high negative predictive value of AIRE offers the potential to confidently reassure patients at the lowest risk, in line with national guidelines. Importantly, we have demonstrated that AIRE can stratify risk even among patients whose ECGs are labelled as normal by clinicians. In this subgroup, the model effectively distinguishes between low- and higher-risk individuals, with a significant difference in mortality observed between these groups.
ORIGINAL HYPOTHESIS
研究设计
- 研究类型
- Interventional
- 分配方式
- Randomized
- 干预模型
- Single Group
- 主要目的
- Diagnostic
- 盲法
- None
盲法说明
The reviewing Cardiologist will be masked as to the outcomes of any artificial-intelligence enhanced ECG analysis. The reviewing cardiologist will intervene if the managing clinician investigation plan is felt to be clinically inappropriate.
入排标准
- 年龄范围
- 18 Years 至 —（Adult, Older Adult）
- 性别
- All
- 接受健康志愿者
- 否
入选标准
- •Our study population will include all patients aged 18 years and older presenting with non-anginal chest pain (chest pain that is not typical to the heart) to 6 rapid access chest pain clinic sites across North West London. Chest pain typicality will be defined using the standardised Rose Angina questionnaire, based on clinical history.
排除标准
- •Patients with typical cardiac (heart-related) chest pain, as defined by the Rose Angina classification
- •Patients with known moderate or severe stenosis (narrowing) in an epicardial coronary artery (the blood vessels supplying the heart)
- •Known left ventricular impairment (left ventricular ejection fraction <50%), otherwise known as heart failure
- •Left bundle branch block (a significant electrical abnormality of the heart on ECG)
- •End-stage kidney failure requiring renal replacement therapy such as dialysis or a kidney transplant
- •Moderate or severe valvular heart disease (serious narrowing or leaking of the heart valves)
- •Paced rhythm at the time of ECG acquisition (lots of patients with pacemakers will fall into this category, but only if their pacemaker is firing at the time the ECG was taken).
研究组 & 干预措施
AI-ECG arm
Patients in this arm will have AI-ECG undertaken alongside their standard care. For patients with non-cardiac chest pain, the clinician will have access to the AI-ECG prediction result denoting the risk level from the ECG. This risk prediction will contribute to their clinical assessment of the patient in addition to history and examination.
干预措施: AI-ECG (Diagnostic Test)
Standard care
Patients will be reviewed in clinic as per best current practice. Decisions on their further care will be undertaken by clinicians as usual.
结局指标
主要结局
The healthcare utilisation costs of each pathway over a one-year period.
时间窗: 12 months following enrolment.
The cost of each pathway will be assessed over a one-year period, taking an English NHS perspective. Healthcare resource use will include subsequent consultations with the general practitioner, outpatient cardiology appointments, accident and emergency attendances, inpatient admissions and additional investigations/interventions. This data will be extracted from the Whole Systems Integrated Care (WSIC) dashboard, which captures data on healthcare contacts across the North West London region. Data will be collected at 12 months following enrolment. Resource use will be valued using unit costs of health and social care from the Care and Outcomes Research Centre and national cost collection for the NHS. Differences in healthcare resource use and costs (both planned and unplanned) between the between the AI-ECG and the standard care pathways will be reported at 12 months.
次要结局
- Time from referral to completion of the clinical pathway(From enrolment to date of established diagnosis and management plan.)
- Composite outcome of hospital admission with acute coronary syndrome and cardiovascular mortality.(Measured up to one year from enrolment.)
- Statin and antihypertensive medication use at one year.(One year from enrolment.)
- Symptom burden at one year, assessed using the Rose Angina questionnaire-based angina quantification app.(One year following enrolment.)
研究者
Jamil Mayet
Professor of Cardiology
Imperial College London