The use of cardiovascular magnetic resonance to identify adverse cardiac prognosis: An important step in reducing image related heathcare expenditures

Advances in cardiovascular (CV) imaging have contributed to the marked reductions observed in CV mortality over the past forty years.(1) Importantly however, expenses for CV imaging have outpaced the growth in expenses for other physicians’ services over the same timeframe.(2) Recently, both the government and private insurers have reacted to this increase in CV imaging related expenditures by instituting processes to curb imaging procedure utilization.(3) Specifically, the implementation of radiology benefit managers (RBMs) have been utilized to pre-certify CV related imaging procedures in the fields of CV computed tomography (CCT) and magnetic resonance (CMR).(4) Often, to obtain qualifying payment for an imaging procedure requires a pre-certification phone call from a physician, which takes 30 minutes and delays healthcare delivery. Although RBMs may disallow up to 26% of imaging requested cases, additional concerns have been raised that as many as 20 to 30% of those performing the screening procedures may be unfamiliar with both the diagnostic accuracy and patient suitability for CMR.(5)

Physicians are also concerned with growth in health care expenditures related to CV imaging. Importantly, however, physicians have advocated different means of limiting CV imaging health related expenditures by focusing on quality of care.(6) Quality is a broad term including several processes that when utilized together may be used to limit costs by selecting patients for the correct imaging procedure. Quality processes include: 1) identifying the appropriateness of the referral of a patient in an imaging procedure; 2) insuring adequacy of the performance of a procedure; 3) reporting procedural results in a timely fashion, and often; 4) linking results to patient related outcomes.(6) Importantly, physician societies, such as the American College of Cardiology, rely on evidence-based materials to develop criteria, guidelines, and standards that direct quality based initiatives.(6) Quality measures can be implemented in a fashion to select patients’ suitability for procedures without lengthy work flow delays.

Research is utilized to establish the processes for guiding the development of quality based initiatives.(7) In regards to CV imaging, this research has several stages including establishment of the technical feasibility and safety, determining the diagnostic accuracy, defining the prognostic utility, and finally, understanding both the cost benefit and patient satisfaction aspects of the procedure. Not performing all of these steps, can, however, lead to increases in expenditures. Potentially contributing to escalating health care expenses related to CV imaging is the fact that this process is often short changed such that the latter stages (prognosis and cost-benefit) of development are not accomplished.(6) If quality is to be used to facilitate control of health care expenditures related to CV imaging (particularly CCT and CMR), further research is necessary across large study populations that helps to define the prognostic utility as well as cost benefit of these modalities.

To this end, two studies presented in this issue of the Journal, from Korosoglov, et al.,(8) and Aquaro, et al.,(9) address the prognostic utility of CMR results in patients with chest pain syndromes and individuals with >1,000 premature ventricular contractions (PVCs) in a left bundle-branch block (LBBB) pattern on 24 hour telemetry monitoring. In the first study, Korosoglov et al. studied 1,493 patients that underwent dobutamine CMR stress testing incorporating both left ventricular (LV) wall motion and myocardial perfusion analyses. They then followed these individuals for 2 years and identified 53 major CV events (including 14 cardiac deaths and 39 myocardial infarctions) within the study population. In this relatively large single center study, dobutamine induced LV wall motion and perfusion abnormalities predicted future adverse CV events after accounting for established risk factors of CV disease.

Features of this study allow practitioners to draw several important conclusions regarding the use of CV stress testing to assess cardiac prognosis. Because of the versatility of CMR, the investigators were able to assess LV wall motion and perfusion in all the patients during a single exam. This allowed the investigators to compare the 2 techniques and determine when one (wall motion or perfusion) should or should not be used to supplement the other. Also, the relatively large number of study participants allowed the investigators to reach important conclusions regarding patient subgroups and associated comorbidities.

Important conclusions drawn from this study include: first, the identification of dobutamine induced wall motion abnormalities forecasted CV prognosis in those with or without myocardial perfusion deficits. The converse, however, was not true as perfusion assessments only provided incremental prognostic information in individuals that did not experience an inducible wall motion abnormality during intravenous dobutamine. Moreover, the prognostic utility of these additional perfusion images occurred only in those with existing LV wall motion abnormalities at rest, those with known coronary disease, or those with LV hypertrophy. Since perfusion assessments require the addition of gadolinium contrast, and contrast is associated with incremental expense and minor risks to participants, the results of this large study indicate that if individuals with normal resting LV function experience an inducible wall motion abnormality during testing, then contrast perfusion need not be administered since the determinations are not useful in providing any increment prognostic information regarding CV risks. The same is true in patients without inducible LV wall motion abnormalities if individuals have a normal LV ejection fraction, no LV hypertrophy, and no existing wall motion abnormalities at rest.

The data from Korosoglou et al. also indicate when the addition of contrast will provide incremental benefit for assessing cardiac prognosis. As shown in their study, the implementation of gadolinium contrast is helpful for identifying adverse prognosis if a patient does not exhibit a wall motion abnormality at peak dobutamine, but exhibits a resting LV wall motion abnormality, known coronary disease, or LV hypertrophy. The data from this study and others(10) raises an interesting question as to whether dobutamine CMR stress perfusion studies should be considered as a first-line dobutamine stress modality (rather than echocardiography alone with wall motion) in appropriately equipped and credentialed centers when patients exhibit resting LV wall motion abnormalities, coronary artery disease, or LV hypertrophy, and there is no contraindication to contrast.

Second, the presence of dobutamine induced LV wall motion abnormalities forecasts cardiac prognosis in individuals regardless of the pretest probability of coronary artery disease (low, intermediate, or high). The absence of inducible LV wall motion abnormalities only confers a favorable CV prognosis in those that are at low or intermediate risk.

Finally, the results (positive or negative for ischemia) of either wall motion or perfusion stress tests do not add incremental information regarding CV prognosis in individuals with a severely reduced LV ejection fraction at rest. Thus, for individuals with a resting LV ejection fraction of <35%, dobutamine stress testing will only be useful in identifying myocardial ischemia or viability when selecting individuals that may be candidates for coronary artery revascularization procedures to relieve symptoms.

The study by Aquaro, et al., was different in that it was a resting CMR exam, and is focused on the right ventricle as opposed to the left ventricle. The study population included relatively young (average age of 33 years) individuals with >1000 PVCs in 24 hours of a LBBB morphology with inferior axis on the twelve lead electrocardiogram. Four hundred and forty subjects, of which 396 were included in the analysis, underwent a resting CMR exam in which right ventricular (RV) regional wall motion, volumes, ejection fraction, and tissue characterization (using determinations of myocardial fat content without gadolinium contrast) were performed. All individuals had a normal maximal exercise stress test, a negative family history for sudden cardiac death, normal echocardiogram, no hypertension, diabetes, or coronary artery disease, and otherwise, were not suspected to be at risk for CV events.

Major (RV dyskinesis, severe RV dilation, or a RV ejection fraction <40%) and minor (RV dilation that was above 2 standard deviations but below 4 standard deviations of normal, RV hypokinesis, or individuals with a RV ejection fraction between 40% and 50%) criteria for arrhythmogenic RV cardiomyopathy were identified in all participants.(11) Importantly, in this study, individuals with ≥2 of these abnormalities (major or minor) experienced a significant increase in the incidence of a major CV event including sudden cardiac death, resuscitated cardiac arrest, or subsequent defibrillator shocks for ventricular tachycardia or fibrillation recorded by the defibrillators implanted after the CMR procedure.

In addition, individuals with none of these factors, even if there was suspected fatty infiltration of the RV wall, had a favorable prognosis. The results of this latter study indicate that individuals with frequent PVCs of a LBBB morphology over the age of 20 years should likely undergo a thorough CMR examination of the right ventricle as these individuals may be at risk for developing RV cardiomyopathy that could lead to an adverse CV event. It is important to recognize that these abnormalities were not observed on transthoracic echocardiography; thus one should not assume a normal echocardiogram will serve as a surrogate for a CMR study in these patients.

What are the limitations to the studies of Aquaro et al. and Korosoglov et al.? First, studies such as these require ancillary and physician staffing with expertise in performance and interpretation of relatively advanced CMR. This includes management of ventricular arrhythmias, the capability to administer and monitor the effects of cardiac stress agents, and the ability to react to adverse CV events during the procedure. Thus, studies should be performed in centers with appropriate credentialing and expertise.(12) To this end, the Society of Cardiovascular Magnetic Resonance (http://SCMR.org), an international society that includes cardiologists, radiologists, physicists, and biomedical engineers, can provide reference and guidance on training and protocols necessary to perform the types of procedures described in these manuscripts.

Secondly, in both studies, most subjects possessed a relatively regular heart rhythm during their exam. In fact, in Aquaro et al.’s study, some individuals (number not disclosed) received anti-arrhythmic therapy 1 week prior to the CMR procedure for the purpose of minimizing PVCs. It is important to recognize that not all patients may be eligible for this type of treatment (particularly those in an acute hospital setting). Recently, high temporal resolution “real time” (no cardiac gating required) imaging has been made available on some vendor platforms.(13) Whether “real-time” imaging would be useful in these situations is uncertain. Finally, certain patients are not well suited for these procedures. Currently those include individuals with implanted devices, such as a pacemaker or defibrillator, or ferro-magnetic intracranial metal.(14) Also, in patients with severe renal dysfunction, one avoids administration of gadolinium contrast (for myocardial stress perfusion) to avert the risk of nephrogenic systemic fibrosis.(14)

In summary, these 2 relatively large single center studies demonstrate the effective use of CMR in identifying CV prognosis in 2 groups of patients: those with chest pain syndromes, and relatively younger individuals with multiple premature ventricular contractions of a LBBB morphology. Results from large prognostic studies such as these provide important evidence for directing the appropriate use of advanced CV imaging techniques such as CMR, and can direct planning of additional research that addresses cost-benefit related issues. Korosoglov et al. and Aquaro et al. are to be commended regarding their research initiatives because it is results such as these that help to provide the foundation for high quality healthcare initiatives that can be used to direct individual patient-care imaging in an increasingly resource limited environment.

Abbreviations

CCT

cardiovascular computer tomography

CMR

cardiovascular magnetic resonance

CV

cardiovascular

LBBB

left bundle-branch block

LV

left ventricular

PVC

premature ventricular contractions

RBM

radiology benefit managers

RV

right ventricular