1. Cost-Effectiveness in Combination with Other Imaging Modalities
Regardless of such challenges, several studies have explored the cost-effectiveness and clinical utility of Cardiac Magnetic Rresonance (CMR). One such study conducted a cost–benefit analysis on a modeled population of patients with acute myocardial infarction (MI) to compare the cost of standard therapy (dual antiplatelet therapy and/or aspirin for life) versus CMR-guided management over a 10-year period
[1]. Researchers used a hypothetical population consisting of 2000 patients with acute MI and normal coronary angiography
[1]. Half of the patients were assigned to standard therapy and the other half to CMR-guided management. Analysis was performed based on United Kingdom costs with outcomes varying based on the time period
[1]. Within the first year, the routine use of CMR to identify patients with a true MI increased spending by 14% per patient. By seven years, CMR-guided practice was cost-neutral. After 10 years, CMR-guided management was found to reduce costs by 3% per patient
[1]. These findings suggested that while CMR-guided management of acute MI may have increased costs in the first few years, it is cost-effective when managing long-term patients. Another similar study that collected data on a cohort of cardiology referrals in a European multicenter registry concluded that over two-thirds of the 11,040 patients (with indications of myocarditis/cardiomyopathies (32%), suspected coronary artery disease (CAD) (31%), myocardial viability assessment (15%)) experienced appropriate changes in therapeutic management based on CMR results, and in a minority of patients (16%), an entirely new diagnosis was discovered
[2]. For instance, the study described a patient who presented to the emergency department with heart failure symptoms and was diagnosed with severe aortic stenosis with a valve replacement recommendation
[2]. The patient was later found to have amyloidosis, as evidenced by subendocardial enhancement, with mild aortic stenosis on CMR evaluation
[2]. Before undergoing CMR, most patients (64.1%) in the registry had undergone transthoracic ECHO with the remaining cohort undergoing coronary angiography (25.1%), cardiac computed tomography (CT) (1.8%), and single-photon emission computed tomography (SPECT) (0.8%), prompting further clinical evaluation with CMR due to uncertainty of prior imaging results. Despite these findings, the applicability of these results in North American and Asian countries remains relatively unknown, due to major differences in healthcare systems, cultures, disease prevalence, and social status which predispose to challenges in interpreting the transference of economic effectiveness
[2].
2. The Role of Diagnostic Accuracy
The diagnostic accuracy of CMR for cardiac diseases drives its cost-effectiveness. One study aimed to identify key determinants for the cost-effectiveness of CMR in the United Kingdom in patients with multivessel CAD and unobstructive coronary arteries
[3]. The patients were divided into two models of multivessel CAD and unobstructed coronary arteries using an index angiogram and underwent initial percutaneous coronary intervention. The group with multivessel CAD underwent one of three diagnostic and treatment pathways: CMR, fractional free flow (FFR), or stress ECHO. Although there was a small QALYs increase for CMR and FFR compared to stress ECHO, with an increased prevalence of ischemia, it was predicted that FFR would have reduced costs due to simultaneous revascularization during ischemia testing
[3]. The reasoning for this prediction was the low ischemia rate of 35% in the FFR group where most patients did not require additional testing or treatment
[3]. Despite FFR and CMR having the same QALY increase in this model, CMR was less costly than FFR (£5431 vs. £5855) in terms of overall costs over 1 year. The second model for unobstructed coronary arteries had two diagnostic pathways which were CMR and stress ECHO, or stress ECHO alone. Through the second model, the addition of CMR resulted in a decrease in costs due to treating fewer patients with MI, but this decrease in cost only partly compensated for the additional cost of CMR
[3]. However, the two models indicated that the sensitivities and specificities of the imaging modality played a significant role in determining the cost-effectiveness of the overall strategy.
3. Cost-Effectiveness of CMR in Conjunction with Guideline Recommendations
The use of CMR has also shown to be cost-effective when used in conjunction with guidelines from different cardiology organizations including the American Heart Association (AHA) and the American College of Cardiology (ACC), as well as radiology committees such as the Society for Cardiovascular Magnetic Resonance (SCMR)
[4][5][6]. A retrospective study of 361 patients in the United States compared CMR with other imaging modalities
[7]. The study had 350 patients who met the AHA and ACC criteria for CMR imaging based on having a limited ECHO (27%), valvular disease (26%), cardiomyopathy (20%), tissue viability (16%), and aortic or vascular disease (11%). There were 11 cases of inappropriately ordered CMR, including eight for left ventricular hypertrophy and three with inadequate justification
[7]. Among the 361 patients, 353 had conclusive results based on the CMR findings, and there was a major change in the plan of care for 256 patients. Herein, CMR had an overall net healthcare saving of $833,037 and a per-patient cost saving of $2308. The study concluded that CMR was cost-effective when used in conjunction with the AHA and ACC criteria
[7].
Furthermore, a cohort study of 1158 patients from 1 January 2003, to 31 December 2004, in Germany examined the utilization of CMR in stable CAD patients
[8]. The inclusion criteria required a clinical presentation of stable CAD with controls for demographic parameters such as age, gender, and cardiovascular risk factors. The exclusion criteria included prior cardiac transplant, left ventricular ejection fraction ≤40%, and known CAD based on angiography. Eventually, of the 1158 patients, 502 patients enrolled in the study with 209 and 293 patients allocated to the CMR and coronary angiogram groups, respectively. Direct CMR was used in accordance with the SCMR recommendations. Due to coronary angiography being a morphological modality and CMR being a functional study (e.g., exercise-induced abnormalities), different rates of CAD diagnoses were measured in the two groups with CMR having the lower prevalence
[8]. Patients undergoing CMR had a savings of 12,466€ in hospital costs per life year. These savings could be associated with the CMR group managed in a mainly ambulatory setting compared to the coronary angiography group managed in an inpatient setting. The CMR group also had shorter hospital stays and decreased coronary angiogram interventions, leading to overall savings
[8].
Similar results were also found in the prospective study known as Stress Cardiac Magnetic Resonance Versus Computed Tomography Coronary Angiography for the Management of Symptomatic Revascularized Patients (STRATEGY). Herein, coronary CT angiography (CCTA) was compared with CMR in terms of overall cumulative costs over one year and cost-effectiveness of the index examination in 600 symptomatic CAD patients with a history of revascularization. In this study, the CMR group not only had lower CAD spending and downstream costs using both invasive and noninvasive imaging but also had lower radiation exposure compared to the CCTA group
[9]. Over the course of one year, the cumulative costs (with cost of index test included) for the CCTA group averaged 2012 ± 2888€ while the CMR group averaged 1516 ± 2464€. A comparison of costs for each examination revealed that CCTA (218 ± 298 €/y) had a higher cost-effective ratio compared to CMR (119 ± 250 €/y)
[9]. One of the explanations for the higher cumulative costs in the CCTA group is the higher rate of additional noninvasive imaging and invasive coronary angiography when abnormal findings arise. With increased invasive coronary angiography, there were increased stent interventions due to the oculostenotic reflex, which is the reflexive revascularization upon visualizing stenotic coronary vasculature even if the vessel is unlikely to cause problems in the future
[9][10]. It was highlighted that the effect of disease progression and treatment history on selecting the appropriate imaging modality; it is more cost-effective for patients with known CAD and a history of revascularization to undergo CMR instead of CCTA as this is more likely to avoid unnecessary revascularization. In terms of major cardiac adverse events, the CMR group had a lower rate of 5% compared to the 10% in the CCTA group. Taking these three studies together, guidelines used in conjunction with CMR demonstrated overall net savings and cost-effectiveness compared to other imaging modalities.
4. Stepwise Testing with Other Imaging Modalities
Imaging strategies with stepwise testing may result in higher cost-effectiveness and increased QALYs compared to a single imaging modality alone or starting with the most definitive examination
[11][12]. One study conducted in the United Kingdom investigated eight different diagnostic strategies with combinations of exercise tolerance test (ETT), CMR, SPECT, and coronary angiography in a patient population from the CE-MARC study that was referred to cardiologists with suspicion of angina pectoris
[12]. In the study population, a pre-test probability of 40% (requiring revascularization due to significant stenosis) was used for cost-effectiveness analysis (outcomes measured in QALYs), with 15.9% of the patients suspected to have CAD without significant stenosis. With the lower cost-effective threshold of £20,000, the best strategy was the exercise stress test followed by CMR, then coronary angiography. The sequence of tests only proceeded if the prior test was positive or inconclusive. With a higher cost-effectiveness threshold of £30,000, the best strategy was CMR followed by coronary angiography
[12]. Both strategies utilized CMR, with the study concluding that the use of CMR is likely a component of the most cost-effective strategy, especially in patients with CAD when the incremental cost of CMR compared with SPECT is not too large (cost increment < £90 at a threshold of £20,000 per QALY, and <£115 at a threshold of £30,000 per QALY)
[12]. In contrast, in the strategies that used SPECT followed by coronary angiography, or the use of ETT followed by coronary angiography, CMR had lower cost-effectiveness at both £20,000 and £30,000 per QALY.
5. Regional Applicability of CMR
Another investigation produced similar results when utilizing evidence from the CE-MARC study and applying it to the Australian healthcare system
[13]. It was also investigated that eight potential clinical strategies using different combinations of electrocardiogram stress testing (EST), SPECT, stress CMR, and coronary angiography by using a decision analytical model coupled with three distinct Markov models
[13]. Based on a cost-effectiveness threshold of $45,000 to $75,000 per QALY gained, the most cost-effective strategy was initial EST, followed by stress CMR if EST was positive or inconclusive, followed by coronary angiography if the stress CMR was positive or inconclusive
[13]. A similar trial was conducted in Switzerland with the same eight strategies
[11]. However, the conclusion differed as the most cost-effective strategy was ETT followed by CMR, then coronary angiography if CMR was positive or inconclusive. The limitation herein was that the cost difference between coronary angiography and CMR is smaller in Switzerland compared to the United Kingdom
[11]. With different costs for imaging studies including CMR and coronary angiography among various thresholds per QALY, a single cost-effective strategy for moderate to high pretest probability groups may not be uniformly used.
One study compared the use of CMR with invasive coronary angiography from the European registry data in Germany, the United Kingdom, Switzerland, and the United States healthcare systems
[14]. The study recruited 2717 patients for the evaluation of a diagnostic workup for CAD. In their study, a cost analysis was conducted on a multicenter European registry where CMR was used as an initial diagnostic modality to assess for myocardial ischemia, with CAD-positive patients being referred to coronary angiography. This was compared to another hypothetical strategy that utilized coronary angiography as a single diagnostic test. By correcting for different healthcare models with outpatient/inpatient procedure coding in these countries, the study determined the cost-reduction of using CMR in a non-emergency setting as compared to invasive coronary angiography. Specifically, when comparing the use of coronary angiography (inpatient) in all these countries, the CMR strategy was associated with an average of 53.5% lower costs. In addition, all tests (coronary angiography, SPECT, CT, CMR, ECHO) conducted as an outpatient procedure in Germany, the United Kingdom, and Switzerland were associated with 50%, 25%, and 23% lower costs with the CMR-driven strategy, respectively. In contrast, when all tests were conducted in an outpatient setting in the United States, the CMR strategy was 8% more costly. The study concluded that the use of CMR could be a useful diagnostic tool to screen for myocardial ischemia for patients with suspected CAD and lead to better resource allocation for healthcare institutions outside of the United States
[14]. In addition, some of the alternative options to screen myocardial viability such as positron emission tomography (PET) require further testing whereas CMR results are conclusive prompting better treatment and cost saving through decreased testing
[7].
6. The Role of Pretest Probability
One combination of imaging strategies investigated was CMR accompanied by invasive coronary angiography, which was compared with coronary angiography followed by FFR in the diagnostic workup of CAD. Although both coronary angiography and FFR were cost-effective for varying prevalence in some countries, CMR and coronary angiography combination was the most effective strategy across most countries
[15]. Specifically, a lower prevalence of CAD was associated with higher cost-effectiveness for the CMR and coronary angiography strategy
[15]. Higher cost-effectiveness was noted below the CAD prevalence of 62%, 65%, 83%, and 82% for Switzerland, Germany, United Kingdom, and United States healthcare systems, respectively
[15]. In contrast, the coronary angiography and FFR strategy showed increasing cost-effectiveness in higher prevalence of CAD. Another study comparing the same three diagnostic pathways for CAD had similar supporting results
[16]. In 3647 patients with suspected CAD, CMR plus coronary angiography minimized costs compared to coronary angiography with or without FFR
[16]. In patients with typical angina, costs were reduced by 11.6 to 12.8% in the United States and Switzerland, respectively, and by 18.9% in the United Kingdom while having minimal savings in Germany (2.3%). The pretest probability, however, still guided the cost-effectiveness as cost savings in all four countries were higher for low pretest probabilities and decreased as pretest probabilities increased
[16]. Both studies highlight the higher cost-effectiveness of CMR in low prevalence populations for the diagnosis of CAD by assessing the degree of cardiac ischemia and determining eligibility for revascularization.
A study in Germany comparing CMR, SPECT, and coronary angiography determined an inverse exponential curve between cost per diagnosis of CAD and prevalence of CAD for SPECT, angiography, and CMR
[17]. The relation of this curve occurs because CMR and SPECT were able to rule out CAD at low prevalence rates, but with higher prevalence rates of CAD, patients needed additional testing or interventions
[17]. Consequently, there would be a smaller number of invasive angiograms needed at a lower prevalence, suggesting CMR and SPECT are more cost-effective at a lower prevalence of CAD. However, with additional interventions and testing needed at a higher prevalence and increased severity of CAD for CMR and SPECT, there would be increased costs due to complications, and increased mortality favoring invasive angiography with more cost-effective results
[17]. The study determined that at a low prevalence of CAD (<60%), CMR was the most cost-effective strategy followed by SPECT, but at a high prevalence of CAD (>60%), angiography was determined to be the most cost-effective strategy
[17].
7. Evidence against Cost-Effectiveness of CMR
A few studies have shown CMR was not a cost-effective imaging technique at lower-to-intermediate pretest probabilities. One study conducted on 60-year-old patients with a low-to-intermediate pretest probability of CAD using a microsimulation model measured lifetime costs, QALYs, and incremental cost-effectiveness ratios (ICERs) to evaluate the most cost-effective imaging strategy in stable chest pain
[18]. The study indicated that with low-to-intermediate probability of CAD, stress CMR and SPECT had less efficacy and were more expensive than stress ECHO. The study combined stress imaging with CT angiography and showed the best combination to be CT angiography with stress ECHO
[18]. It was demonstrated that prevalence plays a significant role in determining the appropriate approach to a patient with CAD. Another study produced similar results when assessing the diagnosis of CAD in a European population with a low-to-intermediate prevalence of CAD from the Evaluation of Integrated Cardiac Imaging in Ischemic Heart Disease (EVINCI) study
[19]. Cost-effectiveness analysis was performed in 350 patients with symptoms of CAD undergoing CCTA combined with one other cardiac imaging stress test, including stress ECHO, SPECT, PET, or stress CMR
[19]. Effectiveness was defined as the percentage of correct diagnosis (cd) (i.e., obstructive CAD with >50% stenosis at quantitative coronary angiography in at least one major coronary vessel), and costs were calculated using country-specific reimbursements. Strategies combining stress CMR followed by CCTA or CCTA followed by stress ECHO, SPECT, or PET were all cost-effective
[19]. ICERs were calculated using “no imaging” as a reference and indicated cost savings of −969 €/cd for CMR-CCTA, −1490 €/cd for CCTA-PET, −3092€ for CCTA-SPECT, and −3776 €/cd for CCTA-ECHO
[19]. It was confirmed that there is no self-standing non-invasive imaging modality that is superior, but rather, the combination of CCTA with stress imaging is a cost-effective means to diagnose CAD and identify revascularization candidates prior to coronary angiography.
8. Pretest Probability in the United States
There has been limited available data from studies done in the United States regarding the use of CMR and cost-effectiveness analysis
[20], besides contributions from the heterogeneity of payment and reimbursement systems in the United States. Notably, the Stress CMR Perfusion Imaging in the United States (SPINS) study was a multicenter cohort study that assessed the prognostic values of stress CMR and the associated costs of care in patients initially presenting with chest pain syndromes
[21]. Utilizing data from the SPINS registry to assess the use of CMR as compared to SPECT, coronary angiography, and CCTA might be one of the first in its nature conducted in the United States
[20]. SPECT has thus far been one of the principal non-invasive imaging methods performed prior to invasive angiography in the United States
[20]. The results herein suggest that prior to utilizing coronary angiography, CMR may be a better cost-effective modality for obstructive CAD
[20]. In patients with a 32.4% probability of obstructive CAD, a CMR-based assessment of CAD was considered the best approach based on the $100,000/QALYs threshold in the United States
[20]. These results were applicable to a wide range of patient populations adjusted for age, rate of revascularization, and cardiac events with different comorbidities. However, these results are applicable to the patient population presenting with stable chest pain syndromes with intermediate CAD prevalence of under 60%, beyond which coronary angiography becomes the preferred imaging modality. This further endorses that disease prevalence plays a large role in determining the cost-effectiveness of CMR.
9. Relationship with Morise Scores
Another investigation focused on the use of CMR preceding the need for invasive procedures
[22]. This German study included 218 participants who were matched to a comparison group of the same size using age, gender, body mass index, diabetes, hypertension (HTN), and dyslipidemia. Furthermore, the study utilized relative value units (RVUs) to determine cardiac catheterization costs in Germany and CMS reimbursement to determine the costs of CMR. As the study reported that CMR was not reimbursed in Germany at the time, the cost was estimated using the ratio of CMR to catheterization with CMS data. These costs were compared with each other at different pretest probabilities for CAD and Morise scores, which is a validated score that stratifies for CAD by accounting for age, sex, symptoms, estrogen status, diabetes, and other pertinent pieces of patient history
[23]. The two groups did not significantly differ in CAD risk factors and Morise scores, and the investigation found that CMR was associated with net savings of £90 per patient, being inversely correlated to the Morise score. Those with lower pre-test probability and Morise scores for CAD had higher catheterization avoidance rates
[22]. The study also concluded that the lowest Morise score was associated with the highest cost savings indicating that CMR is more cost-effective for mild cases of CMR with better myocardial viability while catheterization is more cost-effective for more severe cases, as there is a higher chance that revascularization will be necessary.
This entry is adapted from the peer-reviewed paper 10.3390/jcdd9100357