Introduction
The European Society of Cardiology (ESC) updated guidelines for diagnosis and management of chronic coronary syndromes (ESC-CCS) in coronary artery disease (CAD) [
1]. Computed tomography coronary angiography (CTCA) and non-invasive functional imaging for myocardial ischaemia are recommended (Class 1) as initial test for diagnosing CAD instead of exercise electrocardiography [
1,
2]. The choice of initial non-invasive imaging test primarily depends on the patient’s pre-test probability of obstructive CAD: CTCA is recommended for those with a lower range pre-test probability, while functional imaging is recommended for those with a higher range pre-test probability. Considering that the majority of patients have a pre-test probability in the lower range, these new guideline recommendations may pose challenges for the availability of CTCA-services [
3,
4]. For example, in the Netherlands, the prevalence of CAD is around 800,000 [
5]. The prevalence of patients with chest pain and suspected CAD is higher, but the exact number remains unknown. However, we know that 252,449 patients visited a cardiologist for chest pain in 2012. Of the new patients with suspected CAD 61% underwent an exercise ECG [
6]. Even if a small proportion of exercise ECGs and invasive coronary angiography (ICA) were to be substituted by CTCA, the demand for CTCA-services is expected to increase substantially. Such a shift would require sufficient numbers of CTCA-capable scanners and competent cardiovascular imaging experts to guarantee national coverage. Moreover, variations in clinical practice need to be addressed to ensure high image quality as well as standardised interpretation and reporting of CTCA findings.
It is currently unknown, what percentage of hospital organisations in Europe provide CTCA-services and how many CTCA-capable scanners are available. There is no overview about indications for which CTCA is deployed, and variations in clinical practice across a country are unknown. The goal of our study is to track these issues in the Netherlands, a country with an advanced healthcare system, as an example. We performed a national survey among members of the Dutch Societies of Radiology (NVvR) and Cardiology (NVVC) in every hospital organisation in the Netherlands to study the current utilisation of CTCA-services and status of CTCA-protocols, and modeled the expected effect of these guidelines on CTCA capacity in the Netherlands. Accordingly, the survey is endorsed by the NVvR and NVVC.
Discussion
This national CTCA-survey in the Netherlands provides an overview of the current and modeled CTCA provision and utilisation. Sixty-three Dutch hospital organisations (92.6% of all hospital organisations) provide CTCA-services on a total of 99 CTCA-capable CT-scanners with a total of 37,283 CTCA-examinations per year. There is substantial variation between hospitals considering CTCA indications, available CTCA equipment and applied CTCA-protocols. Implementation of the 2019 ESC-CCS guideline will substantially increase the demand for high quality CTCA-examinations with a high impact on healthcare systems.
However, it should be noted that according to 2019 ESC-CSS guidelines, no type of cardiac imaging should take place before appropriate cardiological assessment, as chest complaints may develop in various clinical contexts. However, if diagnostic work-up is indicated, CTCA or non-invasive functional imaging for myocardial ischaemia is recommended as the initial tests [
1]. Clinical implementation of these recommendations will affect a substantial proportion of cardiac healthcare and is likely to lead to a shift away from exercise ECG and the other traditional tests for CAD assessment. This may pose challenges in organisation of CAD-related healthcare and impose an extra burden on CT capacity and cardiac imagers. However, the initial deployment of CTCA or non-invasive functional imaging for myocardial ischaemia may improve the prognosis of patients. The use of CTCA as initial test in patients with suspected CAD has been shown to better guide preventive medical therapy, which subsequently led to a reduction in cardiovascular death or non-fatal myocardial infarction in the SCOT-HEART trial [
1,
12,
13]. Additionally, the improved diagnostic accuracy may result in early diagnosis and selective treatment of CAD that may prove more cost-effective compared to traditional diagnostic algorithms, thereby considering the direct costs of the procedure and potential downstream cost reduction through preventive medical therapy. However, additional randomised and sufficiently powered studies are needed to make solid statements about cost-effectiveness and improved prognosis.
Challenges for CTCA implementation that need to be taken into account comprise general factors such as organisational culture, networks, communication, leadership, evaluation and feedback, as well as scarce resources such as time, financial resources and education and training of staff [
14,
15]. Moreover, there are specific barriers for the implementation of CTCA, consisting of availability of CTCA-services to provide national coverage with sufficient CTCA-capable CT-scanners and competent cardiovascular imaging experts to facilitate the number of required CTCA-examinations. Almost all hospital organisations in the Netherlands provide CTCA-services and have performed 37,283 examinations last year. However, the projected increase in CTCA-services will call for additional investments in CTCA-capable CT-scanners, cardiovascular imaging experts and CT technicians. Secondly, there are challenges on a level of national health care system organisation and health care reimbursements. Accordingly, as implementation of initial CTCA for patients with CCS will divert patient flows away from traditional testing, intensified collaboration between the departments of cardiology and radiology is recommended.
Another challenge, specific for CTCA is the difference in image quality and the absence of standard operation procedures for CTCA-examinations across hospitals. To address this problem, the SCCT has reported guidelines for the performance and acquisition of coronary computed tomographic angiography [
10]. Although all hospital organisations perform CTCA on CT-scanners that meet minimum standards (64 slice detector width), variation in acquisition protocols among hospitals remains substantial, and this is inextricably linked to differences in image quality and clinical utility. For example, recommended beta-blocker medication is administered in all hospitals whereas recommended nitro-glycerine is not administered in a substantial percentage of hospitals (11.1%) [
10]. Adequate CTCA equipment and protocols will increase the accuracy of stenosis evaluation, thereby reducing the number of a false positive CTCA-examinations [
16‐
19]. These factors also will affect the accuracy of new frontiers in CTCA such as CT-FFR and CT myocardial perfusion [
20,
21]. Besides image quality, CTCA equipment and protocols have major effects on radiation dose of CTCA which should also be taken into account [
22]. In the Netherlands, national coverage of modern CT-scanners is already high, with 71.4% CT-scanner that have 256 slices or more.
Lastly, there is increasing demand for standardised reporting of CTCA findings. The CAD-RADS or ESCR smart reporting tool are standardised communication and reporting methods that link CTCA findings to further management recommendations and provide additional prognostic information of future CAD events [
11,
23,
24]. Although the majority of hospital organisations (76.2%) use a standard report, only a minority (14.3%) used the CAD-RADS reporting system. Furthermore, different hospitals use different cut-off values for a clinically relevant coronary stenosis by either stenosis measurement or visual assessment. This variation in clinical practice is already recognised by the NVvR and NVVC and resulted in a national initiative that aims to optimise the quality of CTCA and reduce the variation in clinical practice by developing uniform image acquisition, post-processing, interpretation and reporting protocols.
The British Society of Cardiovascular Imaging assessed the provision and capability of CTCA within the UK in 2016 and described similar findings with modeled increase in CTCA production of ~ 700% [
25]. We, however, report a lower modeled increase in CTCA production of ~ 300%. These differences might be explained by the different recommendations by the NICE for the UK and the ESC for the Netherlands. The 2016 British NICE guidelines recommend CTCA for all chest pain patients, whereas the current ESC 2019 recommend either CTCA or non-invasive functional imaging for myocardial ischaemia as initial tests and only recommend diagnostic testing in patients with a pre-test probability of > 15% [
1,
26]. Besides, the UK performed a mean number of 592.5 CTCA-examinations per million inhabitants in 2016 which is substantially lower compared with 2085 CTCA-examinations per million inhabitants in the Netherlands in 2019. Compared to the British model, we also included patients in our prediction model with an indication for coronary evaluation in the diagnostic work-up for ventricular tachycardia, cardiomyopathy and heart failure and in the work-up for non-coronary cardiac surgical and transcatheter procedures. This group represents a significant number of patients in which CTCA is able to safely exclude obstructive CAD [
16,
17].
Limitations
The current survey explored the provision and utilisation of CTCA-services in the Netherlands. With this survey, we cannot evaluate how hospitals use diagnostic modalities for individual patients and to what extent the recent guidelines are already followed. A patient specific assessment is necessary to decide which diagnostic work-up is indicated for each individual patient, depending on specific characteristics such as heart rhythm and frequency, kidney function, implanted cardiac devices, obesity. Secondly, the results of this survey only comprised the total annual number of performed CTCA-examinations. Consequently, we are not able to differentiate between the numbers of performed CTCA-examinations per indication and are unable to report numbers for different pre test probability categories, numbers of (preceding) coronary artery calcium score scans or presence of risk-modifiers. Thirdly, we did not include information on the number of non-invasive and invasive downstream diagnostic tests after CTCA (i.e., perfusion MRI, Single-photon emission computed tomography, ICA). Adding this information would have been valuable to better capture the current use of CTCA in clinical practice and interplay with other tests (coronary artery calcium score, non-invasive functional test and invasive test). Furthermore, this information would be valuable to better understand differences in diagnostic approaches between hospitals.
Lastly, we did not inquire CTCA radiation exposure. Despite a substantial reduction in radiation exposure (78% from 2007 to 2017), a large inter-site variation in radiation exposure persists (factor 37, 57–2090 mGy*cm) [
22]. Information about CTCA radiation exposures in The Netherlands in 2018 would add valuable information about the potential of protocol optimization for further radiation exposure reduction in the future.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.