Eur J Echocardiography (2006) 7, 268e273

POSITION PAPER

The future of cardiovascular imaging

and non-invasive diagnosis*

A joint statement from the European Association

of Echocardiography, the Working Groups on
Cardiovascular Magnetic Resonance, Computers
in Cardiology, and Nuclear Cardiology of the
European Society of Cardiology, the European
Association of Nuclear Medicine and the
Association for European Paediatric Cardiology
A.G. Fraser a,*, P.T. Buser b, J.J. Bax d, W.R. Dassen c,
P. Nihoyannopoulos a, J. Schwitter b, J.M. Knuuti d,
M. Höher c, F. Bengel e, A. Szatmári f

a
European Association of Echocardiography, ESC Office for Associations, The European Heart House,
2035 Route des Colles, B.P. 179 e Les Templiers, FR-06903 Sophia Antipolis, France
b
Working Group on Cardiovascular Magnetic Resonance
c
Working Group on Computers in Cardiology
d
Working Group on Nuclear Cardiology
e
European Association of Nuclear Medicine
f
Association for European Paediatric Cardiology

Received 5 April 2006; accepted 4 May 2006

Available online 4 July 2006

*
This paper is published simultaneously in the European Heart Journal, vol. 27 no. 14, pp. 1750e1753 (doi:10.1093/eurheartj/
ehl031) and in the European Journal of Nuclear Medicine and Molecular Imaging (doi:10.1007/s00259-006-0201-8).
* Corresponding author. Tel.: þ334 9294 7600; fax: þ334 9294 7608.
E-mail address: fraserag@cf.ac.uk (A.G. Fraser).

1525-2167/$32 ª 2006 The European Society of Cardiology. Published by Elsevier Ltd. All rights reserved.
doi:10.1016/j.euje.2006.05.001
The future of cardiovascular imaging and non-invasive diagnosis 269

Abstract Advances in medical imaging now make it possible to investigate any

KEYWORDS patient with cardiovascular disease using multiple methods which vary widely in
Cardiovascular their technical requirements, benefits, limitations, and costs. The appropriate
imaging; use of alternative tests requires their integration into joint clinical diagnostic
Clinical services; services where experts in all methods collaborate. This statement summarises
Training; the principles that should guide developments in cardiovascular diagnostic services.
Guidelines; ª 2006 The European Society of Cardiology. Published by Elsevier Ltd. All rights
Research reserved.

Introduction for patients in whom abnormal haemodynamic

function or ischaemia has already been confirmed.
Continuing developments in medical technology Invasive tests of perfusion reserve can aid clinical
and clinical research constantly expand the range decision-making during a coronary procedure, but
of imaging tests and diagnostic measurements. with this exception precise haemodynamic inves-
Nowhere is this more true than in cardiology, where tigation of cardiovascular pathophysiology is now
the choice of investigations includes different performed during cardiac catheterisation less fre-
technologies with overlapping capabilities. How quently than before.
should clinicians, hospital managers, and funding The selection of which test to use to establish
bodies respond and adapt to these changes? How a diagnosis of abnormal function should be based
should the demands of different approaches be on a broad perspective and expert knowledge of
reconciled to the benefit of patients? What are the what each available technique can offer. This is
implications for the education and training of possible only if colleagues experienced in each
future cardiologists, cardiac surgeons, and other imaging modality collaborate fully, so that choices
specialists in cardiovascular imaging? are not constrained by knowledge or practice
limited to a single technology.
It follows that experts in different imaging
subspecialties should produce joint recommenda-
Collaboration between imaging tions and guidelines, from which shared diagnostic
subspecialties protocols can be developed and promulgated.
These may require the development of new crite-
Important information for the diagnosis and man- ria for summarising the outcome of diagnostic
agement of patients with cardiovascular disease studies that vary from those now in common usage
can be provided by ultrasound examination in- for the assessment of therapeutic studies.1 In addi-
cluding echocardiography, by scintigraphy using tion, when there is a diagnostic component of gen-
single photon and positron emitting radiopharma- eral clinical guidelines, then an expert in all the
ceuticals, by magnetic resonance with or without relevant imaging modalities should be included in
a paramagnetic imaging agent, and by X-ray the writing group.
computed tomography or cardiac catheterisation Collaboration should of course also embrace
and angiography with the injection of an iodinated non-invasive and invasive services since these
contrast agent. These alternatives encompass approaches are complementary rather than com-
a wide spectrum from non-invasive investigations petitive. This can be achieved when there are
with no associated risks, through non-invasive appropriate funding and resources for both. Col-
tests that require exposure to ionising radiation, leagues who specialise in either approach should
to invasive procedures with a small risk of major have some education and experience in all imaging
complications. modalities.
Diagnostic cardiac catheterisation is increas-
ingly conducted by cardiologists who also perform
interventions, and this trend is likely to continue.
The progression to therapy by percutaneous in- Evidence-based diagnostic practice
tervention makes coronary arteriography a unique
imaging modality because of its direct and imme- Best diagnostic practice should be based on im-
diate link to treatment, but the limitations of partial professional advice. Some investigations
arteriography, well demonstrated by intravascular such as the exercise ECG were widely implemented
ultrasound, mean that it may be appropriate in before they had been adequately assessed, and
non-urgent cases to reserve the invasive approach some new technologies are evolving so fast that
270 A.G. Fraser et al.

thorough assessment is very difficult. Nonetheless, the systematic evaluation and meta-analysis of
investment in new imaging technologies for rou- diagnostic tests, analogous to the Cochrane coll-
tine practice can only be justified when there is aborations that now overview therapeutic trials2
clear scientific evidence that the new modality is (www.cochrane.org).
substantially better and preferably also more cost-
effective than prior alternatives.
Talk of ‘‘winners’’ and ‘‘losers’’, and confron- Joint imaging strategies
tational debates, are inappropriate. Advice should
be based on clear data from well-conducted re- The clinical use of diagnostic imaging technologies
search establishing accuracy (preferably against an available in a cardiac unit and hospital should be
independent reference criterion), reproducibility, coordinated through a joint service (Fig. 1). This
and safety. Most importantly, diagnostic tests should be managed by a group of specialists includ-
should be evaluated and compared for their ing clinicians and radiologists and chaired by an ex-
impact on clinical outcomes, not just on their pert in several cardiovascular imaging modalities.
attainment of more precise diagnoses or more There is much more to ‘‘imaging’’ than pictures e
impressive images. Relevant outcomes might in- so colleagues with clinical and technical expertise
clude a reduction in complications or hospital together can exploit the potential of new tech-
admissions, or an improvement in quality of life, niques for studying both anatomy and pathophys-
as well as survival. iology. Rather than ‘‘imaging’’, perhaps a new
Different imaging technologies and tests may be term will emerge as current subspecialties con-
appropriate for screening asymptomatic individ- verge and a new type of multi-modality diagnostic
uals or for identifying a disease and then monitor- specialist evolves.
ing its response to treatment. The utility of It is established that echocardiography and
applying a test may vary between populations, coronary arteriography are performed within
and within populations between individuals with cardiology departments under the supervision of
different pre-test probabilities of disease. Some cardiologists. Nuclear cardiology is usually man-
tests may be very effective in the controlled aged by a team of experts including physicists or
conditions of a clinical research study, but in- nuclear medicine physicians and cardiologists.
efficient in routine clinical practice. Feasibility The integration of these technologies into daily
and costs can vary greatly between technologies practice has greatly enhanced their diagnostic
that are capable of demonstrating the same di- impact and indirectly their therapeutic potential.
agnoses. When alternative tests are equally useful, The newer tomographic imaging techniques such
their safety and convenience to the patient may as magnetic resonance and X-ray computed to-
differ. All these issues need to be assessed and mography have been developed by radiological
acknowledged in recommendations. An organisa- and cardiological research groups, and it is ap-
tional framework could usefully be established for propriate and timely now to integrate them also

Cardiology
Nuclear
Medicine Radiology

Joint Cardiac
Imaging Group

Ultrasound
Common
Patients requiring diagnostic SCINT / PET
Clinical impact
cardiovascular protocols CT and end-points
investigation and CMR
pathways
Angiography

Figure 1 Suggested organisation of joint, multidisciplinary diagnostic services. SCINT, nuclear scintigraphy; PET,
positron emission tomography; CT, X-ray computed tomography; CMR, cardiovascular magnetic resonance.
The future of cardiovascular imaging and non-invasive diagnosis 271

into clinical practice. This process will be helped be referred to a joint diagnostic team; the
by a shift of diagnostic expertise, from special- investigation will then be performed either by
ising in a particular technique that is applied by a cardiologist or by a cardiovascular radiologist as
cross-sectional imaging to multiple organs, to an part of the combined service. Ideally, reports
organ- or system-based approach where the di- should be issued jointly by cardiologists and
agnostic expert is more concerned with function, radiologists, but in principle a study can be
the integration of results into clinical decision- performed and interpreted by any appropriately
making, and the impact of diagnostic imaging on trained specialist.
clinical outcomes.
For logistic and physical reasons, it may not be
feasible to locate tomographic imaging machines
geographically within the cardiological depart- Priorities for research
ments of many hospitals. This need not prevent
planning for such an arrangement in future, The prevailing trend is to emphasise the impor-
when resources and local circumstances or new tance of basic biomedical, molecular and genetic
buildings allow it. What is much more important research3 and to allocate the largest share of
than the location of equipment or the back- resources to these fields of enquiry. These pro-
ground and specialty of the staff who perform grammes have significantly advanced our under-
examinations, however, is that all disciplines standing of the mechanisms of disease, but in
work together and that services are coordinated. order to translate developments into any major
A joint department is not necessary, but a joint clinical impact on the prevalence, morbidity or
service where patients follow agreed common mortality of common cardiovascular diseases such
clinical investigative pathways, should become as atherosclerosis, hypertension, diabetes and
the norm. These pathways should be reviewed heart failure, it is necessary also to develop more
regularly and modified in response to feedback sophisticated tests for the precise measurement
from their impact on outcomes (see Fig. 1), and of vascular and cardiac function and the effects
also to take account of the preferences of of new treatments. Clinical diagnostic expertise
patients. with sensitive and accurate characterisation of
A common channel for processing requests for early disease and its progression is required to as-
sophisticated diagnostic tests is preferred, since it sist further progress, and cardiovascular imaging
allows recommendation or selection by experts has a key role to play in meeting this challenge.
of the most appropriate test in any specific The best basic science needs to be combined
circumstance. It is important to avoid duplication with the best imaging methods.
of tests and to prioritise locally according to Governmental and European Union grant-funding
available facilities and expertise. Selection of bodies should allocate funds to joint cardiovas-
a single test in a patient, when it has been shown cular research initiatives that encompass clinical
to have clinical value, is preferred to a succession research in diagnostic imaging in conjunction with
of tests based on different technologies, used in research in mechanical and electronic engineer-
turn after each previous test has given ambiguous ing, informatics and biostatistics, and epidemiol-
or uncertain results. ogy. Medical equipment companies cannot be
Ideally, algorithms for the investigation of expected to fund such research; collaboration
cardiac patients by tomographic imaging should between universities and research engineers in
be established jointly by cardiologists and radi- industry is very important, but major clinical
ologists. If new methods are to be properly research should be conducted without specific
evaluated and implemented, then access should company sponsorship.
be unhindered. The cardiological indications for The increasing precision of non-invasive diagno-
best use and cost-effectiveness of magnetic sis paradoxically makes it more difficult to de-
resonance and X-ray computed tomographic im- termine whether or not a patient has subclinical
aging need to be established by clinical research disease, because the influences or associations of
in departments where there are no major finan- risk factors and age on arterial and myocardial
cial barriers to their use, and where equivalent structure and function can now be identified. This
expertise is available for all modalities that are makes it necessary to establish very large data-
compared. bases of normal subjects which are relevant to
When direct access to diagnostic services is each country or major region. Population studies
offered to non-cardiovascular specialists or to with precise phenotypic definition by new diag-
primary care physicians, then patients should nostic techniques, undertaken in conjunction
272 A.G. Fraser et al.

with genetic investigations, may yield important more closely involved in a clinical service should
insights into the interaction of genetic and obtain some experience during their training
environmental influences and thereby offer di- within that clinical specialty, both of mechanisms
rections for new epidemiological approaches of disease and of the clinical utility of the di-
to the prevention of common diseases. Such agnostic tests.
projects will need to be organised as multicentre Acquiring theoretical and practical expertise in
collaborations. several diagnostic techniques presents major chal-
Academic cardiologists who organise large clin- lenges, but there is a need for imaging specialists
ical trials of new drugs with support from pharma- who have multiple skills. The goal of joint educa-
ceutical companies should take expert advice tional programmes should be to train cardiologists
about recent diagnostic advances whenever they or cardiovascular radiologists who have the know-
design a new study. This will avoid any time-lag ledge, skills and experience to assume clinical
between testing new drugs and implementing new responsibility for organising, supervising, and re-
tests for detecting changes with optimal sensitivity porting diagnostic tests in at least two different
and accuracy. modalities (and ideally, in all current modalities).
Universities should recruit, encourage and sup- Physicians who have clinical responsibility for non-
port academic cardiologists to develop and retain invasive diagnostic imaging should have obtained
expertise in clinical physiology and diagnosis. national or European accreditation in their chosen
These are necessary and legitimate interests for subspecialties4; cardiologists in this role should
an academic department that should be developed have completed a special fellowship in advanced
in conjunction with expertise in basic science and diagnostic methods in addition to general cardio-
genetics. logical training. Accreditation should be main-
tained through participation in continuing
medical education, which should be supported
Implications for training financially by employers.
Cardiac departments need at least as many non-
There should be joint educational programmes invasive experts as colleagues who perform coro-
that rotate fellows who are training in cardio- nary interventions, if advances in diagnosis and
vascular medicine, between echocardiography, monitoring of disease are to be implemented fully
cardiovascular magnetic resonance, nuclear car- for the benefit of patients. Doctors in training who
diology, and X-ray computed tomography, with want to choose non-invasive cardiology as their
optional experience for example in positron emis- special interest should be rewarded similarly to
sion tomography or vascular ultrasound if it is those working in other subspecialties. Recruitment
available. Training time should be timetabled and should be based on interest and not skewed by
protected. For prospective general cardiologists, disincentives caused by extraneous non-clinical
this experience is perhaps more important than factors.
time allocated for training in diagnostic cardiac Clinical cardiologists should not personally per-
catheterisation, since the non-invasive diagnostic form or report investigations such as echocardiog-
department is now the place where fellows raphy or scintigraphy unless they keep up-to-date,
observe and learn in detail about cardiovascular for example by participating in hospital reporting
function, myocardial perfusion, and pathophysiol- sessions and case-review conferences and by
ogy. Such experience provides a sound under- attending national or international courses and
standing of mechanisms of disease, which forms conferences. Cardiac surgeons in training should
the essential foundation for logical clinical undertake specific education in non-invasive car-
practice. diovascular imaging as well as in the interpre-
Cardiovascular radiologists have their own tation of arteriography, since they need to
training programme, which includes particular understand the potential and limitations of any
experience of radiation safety, medical physics technique that is used to establish a preoperative
and technological factors required to perform, diagnosis. They should also be familiar with
optimise and interpret tomographic imaging. Ex- techniques that are used to assess the outcome
pertise in these aspects is important for the safe of surgery, such as intraoperative transesophageal
and appropriate clinical implementation of ad- echocardiography. A component of continuing
vanced diagnostic imaging techniques, and there- medical education for all cardiologists and cardiac
fore radiologists and physicists have a particular surgeons should relate to advances in diagnostic
contribution to make to joint diagnostic services. imaging and their clinical implementation and
At the same time, radiologists who wish to become value.
The future of cardiovascular imaging and non-invasive diagnosis 273

Practitioners who undertake limited diagnostic Acknowledgement

imaging, for example with hand-held echocardio-
graphic machines, should also complete approved This document has been reviewed and endorsed by
training and participate in appropriate continuing the members of the Board of the European
medical education. Association of Echocardiography, by the members
of the Nuclei of the Working Groups on Cardiovas-
Conclusions cular Magnetic Resonance, Computers in Cardiol-
ogy, and Nuclear Cardiology of the European

Experts in different imaging modalities should Society of Cardiology, and by members of the
collaborate not compete. Cardiovascular Committee of the European Asso-

Joint clinical services and common diagnostic ciation of Nuclear Medicine. We thank all our
pathways should be developed. colleagues for their helpful comments. We also

Future diagnostic specialists should be trained thank Professor Matthijs Oudkerk, President of the
in several imaging modalities. European Society of Cardiac Radiology, for his

Diagnostic tests should be evaluated by their helpful advice.
impact on clinical outcomes.

Diagnostic guidelines should compare all
methods that can be applied to a particular References
clinical question.

New criteria should be developed for judging 1. Gibbons RJ, Smith S, Antman E. American College of Cardio-
the quality of diagnostic research. logy/American Heart Association Clinical Practice Guide-

Expertise in imaging should be encouraged lines: part I. Where do they come from? Circulation 2003;
and funded as an integral component of basic, 107:2979e86.
2. Ebrahim S. The Cochrane collaboration: an international ef-
epidemiological, and clinical collaborative
fort for preparing, maintaining and disseminating systematic
research networks. reviews of the effects of health care in cardiology. Cardio-
vasc Risk Factors 1999;7:216e22.
Our professional groups will now pursue these 3. Martin J, Nimmesgern E, editors. The future of cardiovascu-
objectives jointly, both within the European Soci- lar research in Europe. Brussels: European Commission; 2005.
4. Fox KF, Flachskampf F, Luis Zamorano J, Badano L,
ety of Cardiology and also through collaborations
Fraser AG, Pinto FJ. Report on the first written exam held
between the medical professional societies in as part of the European Association of Echocardiography ac-
Europe that are concerned with diagnostic imaging creditation process in adult transthoracic echocardiography.
in cardiovascular disease. Eur J Echocardiogr 2004;5:320e5.