Background
Coronary artery disease (CAD) remains a major public health problem in Australia and globally. A variety of imaging techniques allow for both anatomical and functional assessment of CAD and selection of the optimal investigation pathway is challenging. Cardiovascular magnetic resonance (CMR) is not widely used in Australia, partly due to perceived cost and lack of Federal Government reimbursement compared to the alternative techniques. The aim of this study was to estimate the cost-effectiveness of different diagnostic strategies in identifying significant CAD in patients with chest pain suggestive of angina using the evidence gathered in the Clinical Evaluation of Magnetic Resonance Imaging in Coronary Heart Disease 2 (CE-MARC trial), analysed from the Australian health care perspective.
Methods
A decision analytic model coupled with three distinct Markov models allowed eight potential clinical investigation strategies to be considered; combinations of exercise electrocardiogram stress testing (EST), single-photon emission computed tomography (SPECT), stress CMR, and invasive coronary angiography (ICA). Costs were from the Australian health care system in Australian dollars, and outcomes were measured in terms of quality-adjusted life-years. Parameter estimates were derived from the CE-MARC and EUropean trial on Reduction Of cardiac events with Perindopril in patients with stable coronary Artery disease (EUROPA) trials, and from reviews of the published literature.
Results
The most cost-effective diagnostic strategy, based on a cost-effectiveness threshold of $45,000 to $75,000 per QALY gained, was EST, followed by stress CMR if the EST was positive or inconclusive, followed by ICA if the stress CMR was positive or inconclusive; this held true in the base case and the majority of scenario analyses.
Conclusions
This economic evaluation shows that an investigative strategy of stress CMR if EST is inconclusive or positive is the most cost-effective approach for diagnosing significant coronary disease in chest pain patients within the Australian health care system.
Keywords
Introduction
Coronary artery disease (CAD) remains a major public health problem in Australia and around the globe. A variety of imaging techniques allow for both anatomical and functional assessment of CAD and, as a result, the selection of the optimal investigation pathway is increasingly challenging. In general, there is the need for investigative pathways that are feasible, safe, cost-effective and of high diagnostic accuracy.
Physicians traditionally choose between one or a combination of tests that provide anatomical information regarding the degree of atheroma and stenosis (e.g. computed tomography coronary angiography, [CTCA]) and/or functional tests to detect ischaemia or reduced myocardial blood flow (e.g. exercise electrocardiogram stress testing [EST], stress echocardiography, stress single-photon emission computed tomography [SPECT], or stress perfusion cardiovascular magnetic resonance [stress CMR]) in order to exclude significant CAD. SPECT is the most common non-invasive ischaemia test worldwide, however it exposes patients to ionising radiation and its diagnostic accuracy is variable. In comparison, CMR with myocardial stress perfusion imaging has no radiation exposure and has high diagnostic performance in the identification of significant coronary ischaemia, with studies showing stress CMR has a sensitivity, specificity, and diagnostic accuracy of 87–89%, 84–93%, and 89–91%, respectively [
[1]
,[2]
]. Compared to the gold standard invasive fractional flow reserve (FFR), stress CMR with adenosine has a high sensitivity of 91% and specificity of 94% for identification of lesion-specific ischaemia [[3]
]. In addition, stress CMR delivers important prognostic information in the intermediate risk chest pain setting [[4]
], is a stronger predictor of prognosis than SPECT [[5]
], and provides late gadolinium enhancement imaging with additional information on segmental viability.- Greenwood J.P.
- Herzog B.A.
- Brown J.M.
- Everett C.C.
- Nixon J.
- Bijsterveld P.
- et al.
Prognostic value of cardiovascular magnetic resonance and single-photon emission computed tomography in suspected coronary heart disease: long-term follow-up of a prospective, diagnostic accuracy cohort study.
Ann Intern Med. 2016; 165: 1-9
From a health economics point of view, stress CMR has also been shown to be cost-effective in the diagnosis and management of CAD in a number of countries [
6
, 7
, - Moschetti K.
- Petersen S.E.
- Pilz G.
- Kwong R.Y.
- Wasserfallen J.B.
- Lombardi M.
- et al.
Cost-minimization analysis of three decision strategies for cardiac revascularization: results of the “suspected CAD” cohort of the European Cardiovascular Magnetic Resonance Registry.
J Cardiovasc Magn Reson. 2016; 18: 3
8
, 9
, 10
]—it has a high diagnostic yield, allowing a reduction in false positive studies (and thus a reduction in costly invasive investigations), and offers better use of resources. Two (2) recent large-scale, multi-centre, randomised comparative effectiveness trials have shown that first line investigation of stable chest pain by stress CMR significantly reduced the rate of invasive coronary angiography (ICA) where no obstructive disease was found as compared to UK 2010 National Institute for Health and Care Excellence (NICE) guidelines [[11]
], and that stress CMR compared to a strategy of ICA±FFR resulted in lower overall rates of revascularisation with no difference in major adverse cardiovascular events [- Greenwood J.P.
- Ripley D.P.
- Berry C.
- McCann G.P.
- Plein S.
- Bucciarelli-Ducci C.
- et al.
Effect of care guided by cardiovascular magnetic resonance, myocardial perfusion scintigraphy, or NICE Guidelines on subsequent unnecessary angiography rates: the CE-MARC 2 randomized clinical trial.
JAMA. 2016; 316: 1051-1060
[12]
].Despite these major advantages of stress CMR, it is not publicly funded in Australia, which reduces patient access. Consequently, it is infrequently used in Australia compared to other diagnostic testing strategies. One impediment to public funding of stress CMR in Australia is a lack of local cost-effectiveness data to assist decision makers and physicians in weighing up the benefits of improved diagnostic accuracy, patient safety and experience, and savings on further downstream investigations, versus the isolated monetary cost of the magnetic resonance imaging (MRI) scan.
The Clinical Evaluation of MAgnetic Resonance imaging in Coronary heart disease (CE-MARC) trial was the largest prospective diagnostic accuracy evaluation to date of stress CMR compared with the reference standard of ICA for patients referred to cardiologists for the investigation of symptoms consistent with angina pectoris [
[1]
,[13]
]. All patients underwent EST if physically able, and were scheduled for both SPECT and CMR (in random order) followed by ICA irrespective of clinical intention. Subsequent cost-effectiveness analyses based on this trial demonstrated lower costs with a stress CMR based strategy, both in the UK and other health care environments [[6]
,[9]
].The aim of this study was to estimate the cost-effectiveness of different diagnostic testing strategies, including stress CMR, ICA, EST, and SPECT, in identifying significant CAD in patients with chest pain suspicious of angina using the evidence gathered in the CE-MARC study but viewed from the Australian health care perspective.
Methods
The methods used in this analysis are consistent with those detailed by the NICE and were based on the UK cost-effectiveness analysis of the CE-MARC trial [
[6]
]. The model included a decision analytic model coupled with three distinct Markov models, comparing stress CMR, ICA, EST, and SPECT. Parameter estimates were derived from the CE-MARC [[1]
], and EUropean trial on Reduction Of cardiac events with Perindopril in patients with stable coronary Artery disease (EUROPA) [[14]
] trials, and from reviews of the published literature (refer to CE-MARC cost-effectiveness study [[6]
] for full details of all parameters). Costs were from the Australian health care system perspective, and outcomes were measured in terms of quality-adjusted life-years (QALYs). The time horizon of the analysis was 50 years, and costs and QALYs were discounted at 5% per annum, consistent with Australian guidelines [[15]
]. Stress echocardiography and CTCA were not assessed in these models as they were not part of the original CE-MARC derivation cohort.Medical Services Advisory Committee
Technical guidelines for preparing assessment reports for the Medical Services Advisory Committee – service type: investigative (Version 3.0).
Technical guidelines for preparing assessment reports for the Medical Services Advisory Committee – service type: investigative (Version 3.0).
http://www.msac.gov.au/internet/msac/publishing.nsf/Content/0BD63667C984FEEACA25801000123AD8/$File/InvestigativeTechnicalGuidelines-December-2016-Version-3.0.pdf
Date: 2017
Date accessed: November 19, 2019
Base Case Analysis
For the base case analysis, the case of a 60-year-old male with grade 2 symptoms on the Canadian Cardiovascular Society (CCS) angina scale and pre-test likelihood of significant stenosis requiring revascularisation of 39.5% was used; 15.9% of patients were considered to have CAD but not significant coronary artery stenosis (based on CE-MARC data) [
[1]
]. The choice of characteristics was based on real patient data from the CE-MARC trial and clinical opinion. Alternative scenarios for age, sex, CCS grade, varying pre-test likelihood of disease, and costs of the diagnostic tests in the Australian health care system were also modelled.Diagnostic Strategies and Pathways
Eight (8) possible diagnostic strategies were derived from the CE-MARC trial for comparison:
- 1.ICA only
- 2.EST, followed by ICA if EST positive or inconclusive
- 3.EST, followed by CMR if EST positive or inconclusive, followed by ICA if the CMR positive or inconclusive
- 4.EST, followed by SPECT if EST positive or inconclusive, followed by ICA if the SPECT positive or inconclusive
- 5.CMR, followed by ICA if CMR positive or inconclusive
- 6.SPECT, followed by ICA if SPECT positive or inconclusive
- 7.EST, followed by ICA if positive, or followed by CMR if EST inconclusive, followed by ICA if CMR positive or inconclusive
- 8.EST, followed by ICA positive, or followed by SPECT if EST inconclusive, followed by ICA if SPECT positive or inconclusive.
A major aim of diagnostic testing is to identify high risk patients with significant coronary artery stenosis who may benefit from revascularisation (either percutaneous coronary intervention or coronary artery bypass surgery). For the model, it was assumed that all patients who are suspected of having significant coronary stenosis must undergo a coronary angiogram as a definitive test before revascularisation. Invasive coronary angiography was regarded as the ‘reference standard’ test (i.e., assumed sensitivity and specificity of 100%, no false positives), so it was assumed that no patients would receive an inappropriate revascularisation procedure. However, as the non-invasive diagnostic tests (EST, SPECT and stress CMR) are not 100% accurate (i.e., sensitivities and specificities below 100%), some patients with clinically significant stenosis requiring revascularisation will not progress across the diagnostic pathways to ICA, as a consequence of a false negative (FN) test. These patients incorrectly identified as not having significant stenosis will not receive an appropriate revascularisation procedure and, as a result, may experience less relief from their angina symptoms until their disease is subsequently correctly managed. It was assumed, however, that these ‘false negative’ cases would have their ischaemia treated by optimal medical therapy. Furthermore, some patients without clinically significant stenosis will progress to ICA, as a consequence of a false positive test, with its associated cost and morbidity/mortality risk. There will also be patients without clinically significant stenosis and hence who do not require revascularisation, but who do suffer from angina, and these patients were assumed to receive risk factor modification and optimal medical management.
Health Outcomes
Health outcomes were reported in terms of QALYs, which combine survival and health related quality of life (HRQoL) into a single measure. Australian all-cause mortality rates and cardiovascular mortality rates by age and sex were applied to estimate survival [
[16]
]. HRQoL was estimated using the EuroQol 5 Dimension (EQ-5D) instrument. Australian population norms for the EQ-5D [Australian Bureau of Statistics
3303.0 - Causes of death Australia 2017 2018.
3303.0 - Causes of death Australia 2017 2018.
https://www.abs.gov.au/AUSSTATS/[email protected]/DetailsPage/3303.02017?OpenDocument
Date accessed: November 19, 2019
[17]
], by age and sex, were combined with other sources and assumptions to estimate QALYs by initial CCS grade and treatment status (whether the patient had received a revascularisation procedure or medical management). Full details of the sources and methods used are previously published [[6]
].Resource Use and Costs
Costs were reported in 2016–2017 Australian dollars (AU). Diagnostic testing, revascularisation and ongoing treatment costs were based on Medicare Benefit Schedule (MBS) fees, the National Hospital Cost Data Collection, and published Australian observational studies (Table 1). Costs were inflated to 2016–2017 values using the Australian Institute for Health and Welfare health price index where required [
[18]
]. Of note, at the time of publication there was no MBS item for stress CMR imaging, therefore, in the base case a cost of $900 was assumed for stress CMR, as proposed in a recent assessment report for the Medical Services Advisory Committee (MSAC) [[19]
]. However, analyses were conducted to consider the cost differential between SPECT and stress CMR imaging. Costs applied in the model are summarised in Table 1.Table 1Costs applied in the model in Australian dollars.
| Parameter | Mean Value (95% CI) | Source | Comments/Distribution |
|---|---|---|---|
| Imaging costs | |||
| EST | $152.15 | MBS item 11712 [ [25] ] | Fixed value |
| SPECT | $834.90 | MBS item 61307 [ [25] ] | Fixed value |
| Stress CMR | $900.00 | Assumed and Morona (2016) [ [19] ] | Fixed value. Threshold analysis conducted |
| ICA | $5,127.25 | NHCDC (public sector) Tier 2, item 1005, NHCDC (public and private sector) AR-DRG F42B, and MBS item 57360, weighted by separations and inflated to 2016–2017 values [ [18] ,[25] ,[26] ] | Fixed value |
| ICA complications | $7,411.12 | Difference between NHCDC (public sector) AR-DRG F42A and F42B, and NHCDC (private sector) AR-DRG F42A and F42B, weighted by separations and inflated to 2016–2017 values [ [18] ,[25] ,[26] ] | Fixed value |
| Procedure costs | |||
| CABG | $41,810.66 | NHCDC (public and private sector) AR-DRG F06A and F06B, and MBS items 38500, 51303, 17615, 20560 and 23121, weighted by separations and inflated to 2016–2017 values [ [18] ,[25] ,[26] ]. Assumes pre-anaesthesia consultation of 15–30 minutes and patient under anaesthesia for 3:51 to 4 hours. | Fixed value |
| PCI | $8,977.28 | NHCDC (public sector) Tier 2, item 1005, NHCDC (public and private sector) AR-DRG F15A and F15B, and MBS items 38306 and 51303, weighted by separations and inflated to 2016–2017 values [ [18] ,[25] ,[26] ]. Assumes 1 segment stented. | Fixed value |
| Ongoing costs | |||
| Background cost per quarter with CHD | $1,280.89 | Turkstra et al. (2013), excluding hospitalisations for MI/angina/CHF, and inflated to 2016–2017 values [ [18] ,[27] ] | Gamma distribution (SE = $488) |
| Non-fatal CHD event | $9,648.02 | Turkstra et al. (2013), hospitalisations for MI/angina/CHF, and inflated to 2016–2017 values [ [18] ,[27] ] | Gamma distribution (SE = $1,797) |
| Cardiac fatal event | $5,864.39 | Kardamanidis et al. (2007), hospitalisation costs in last 1 month of life for people who died of cardiovascular disease, and inflated to 2016–2017 values [ [18] ,[28] ] | Gamma distribution (SE = mean) |
| Non-cardiac fatal event | $8,055.05 | Kardamanidis et al. (2007), hospitalisation costs in last 1 month of life for people who died for causes other than cardiovascular disease, and inflated to 2016–2017 values [ [18] ,[28] ] | Gamma distribution (SE = mean) |
Abbreviations: CA, coronary angiography; CABG, coronary artery bypass grafting; CHD, coronary heart disease; CHF, chronic heart failure; CMR, cardiovascular magnetic resonance; EST, exercise electrocardiogram stress testing; MBS, Medicare Benefits Schedule; MI, myocardial infarction; PCI, percutaneous coronary intervention; SE, standard error; SPECT, single-photon emission computed tomography; NHCDC, National Hospital Cost Data Collection.
Analysis
Standard decision rules were used to identify the most cost-effective diagnostic strategy for CAD based on a given set of patient characteristics. This involved ranking strategies in terms of their expected costs and QALYs, removing strategies which result in fewer QALYs and higher costs than one or more other strategies (i.e. dominated) and thus would never be funded, then removing strategies which would also result in fewer QALYs and higher costs if patients received a combination of the other strategies (i.e. extendedly dominated). Incremental cost-effectiveness ratios (ICERs) were calculated for all remaining strategies as follows:
To assess which option was potentially cost-effective, the range of cost-effectiveness thresholds for Australia ($45,000–$75,000 per QALY, midpoint $60,000) [
[20]
] was used, such that the most effective option with an ICER below the threshold was considered the cost-effective strategy.To reflect the uncertainty in the evidence used in the model, input parameters were entered as probability distributions. Probabilistic sensitivity analysis was then used to calculate the mean costs and QALYs for each strategy and the probability that a strategy was cost-effective for the cost-effectiveness threshold.
Results
Cost effectiveness results for the base case are presented in Table 2, with strategies that result in fewer QALYs and higher costs (dominated and extendedly dominated) excluded. In the base case, three of the eight strategies were not excluded (Strategies 3, 5 and 7).
Table 2Base case 60-year-old male, CCS 2, prior likelihood of disease 39.5%.
| Strategy | Expected Costs | Expected QALYs | ICER | Probability Cost-effective at AUD $45,000 per QALY | Probability Cost-effective at AUD $60,000 per QALY | Probability Cost-effective at AUD $75,000 per QALY |
|---|---|---|---|---|---|---|
| Strategy 3- EST- CMR-ICA | $45,279 | 11.0051 | 0.9922 | 0.9826 | 0.9658 | |
| Strategy 5- CMR-ICA | $45,416 | 11.0054 | $440,325 | 0.0078 | 0.0174 | 0.0342 |
| Strategy 7- EST then ICA if +ve, EST- CMR - ICA if EST uncertain | $45,876 | 11.0065 | $449,208 | 0 | 0 | 0 |
Abbreviations: EST, exercise electrocardiogram stress test; CMR, cardiovascular magnetic resonance; ICA, invasive coronary angiogram; ICER, incremental cost-effectiveness ratio; AUD$, Australian dollars; QALY, quality adjusted life years.
The least costly strategy that was included was Strategy 3 (EST, followed by CMR if EST positive or inconclusive, followed by ICA if CMR positive or inconclusive). When the next more costly strategy, Strategy 5 (CMR, followed by ICA if CMR positive or inconclusive), was compared with Strategy 3, an ICER of $440,325 per QALY gained was estimated. When the next more costly strategy, Strategy 7 (EST, followed by ICA if positive, or followed by CMR if EST inconclusive, followed by ICA if CMR positive or inconclusive), was compared with Strategy 5, an ICER of $449,208 per QALY gained was estimated. Strategy 3 was therefore the most cost-effective diagnostic strategy based on a cost-effectiveness threshold of $45,000 to $75,000 per QALY gained.
Different scenarios related to sex, age, severity of symptoms (CCS), prior likelihood of disease, and stress CMR costs are displayed in Table 3. Changing the sex of the base case from male to female, reducing or increasing the age to 50 or 70 year old respectively (compared to 60 year old in the male base case), and increasing the severity of symptoms from CCS grade 2 to grade 4 had minimal impact on the results, with Strategy 3 remaining the most cost-effective diagnostic strategy.
Table 3Scenario analyses.
| Strategy | Expected Costs | Expected QALYs | ICER (AUD$ per QALY Gained) |
|---|---|---|---|
| A) 60-yr-old female, CCS 2, prior likelihood of disease 39.5% | |||
| Strategy 3- EST- CMR-ICA | $46,594 | 11.3086 | |
| Strategy 5- CMR-ICA | $46,729 | 11.3089 | $529,401 |
| Strategy 7- EST then ICA if +ve, EST- CMR -ICA if ETT uncertain | $47,178 | 11.3097 | $553,223 |
| B) 60-yr-old male, CCS 4, prior likelihood of disease 39.5% | |||
| Strategy 3- EST- CMR-ICA | $44,843 | 10.7291 | |
| Strategy 7- EST then ICA if +ve, EST- CMR - ICA if EST uncertain | $45,418 | 10.7311 | $278,331 |
| C) 50-yr-old male, CCS 2, prior likelihood of disease 39.5% | |||
| Strategy 3- EST- CMR-ICA | $50,060 | 13.1080 | |
| Strategy 5- CMR-ICA | $50,198 | 13.1082 | $556,933 |
| Strategy 7- EST then ICA if +ve, EST- CMR - ICA if EST uncertain | $50,661 | 13.1090 | $621,598 |
| D) 70-yr-old male, CCS 2, prior likelihood of disease 39.5% | |||
| Strategy 3- EST- CMR-ICA | $37,786 | 8.2884 | |
| Strategy 7- EST then ICA if +ve, EST- CMR - ICA if EST uncertain | $38,392 | 8.2899 | $398,706 |
| E) 60-yr-old male, CCS 2, prior likelihood of disease 20% | |||
| Strategy 3- EST- CMR-ICA | $23,416 | 11.2291 | Dominant |
| F) 60-yr-old male, CCS 2, prior likelihood of disease 60% | |||
| Strategy 3- EST- CMR-ICA | $67,359 | 10.7832 | |
| Strategy 7- EST then ICA if +ve, EST- CMR - ICA if EST uncertain | $67,496 | 10.7867 | $39,905 |
| Strategy 2- EST- CA | $67,912 | 10.7869 | $1,519,544 |
| G) 60-yr-old male, CCS 2, prior likelihood of disease 80% | |||
| Strategy 7- EST then ICA if +ve, EST- CMR - ICA if EST uncertain | $89,020 | 10.5658 | |
| Strategy 2- EST- ICA | $89,119 | 10.5671 | $77,827 |
| H) 60-yr-old male, CCS 2, prior likelihood of disease 39.5%, EST inappropriate | |||
| Strategy 5- CMR-ICA | $44,947 | 11.0123 | Dominant |
| I) 60-yr-old male, CCS 2, prior likelihood of disease 39.5%, CMR costs same as SPECT | |||
| Strategy 3- EST- CMR-ICA | $44,753 | 11.0120 | |
| Strategy 5- CMR-ICA | $44,881 | 11.0123 | $440,967 |
| Strategy 7- EST then ICA if +ve, EST- CMR - ICA if EST uncertain | $45,384 | 11.0133 | $512,090 |
| J) 60-yr-old male, CCS 2, prior likelihood of disease 39.5%, CMR costs $100 more than SPECT | |||
| Strategy 3- EST- CMR-ICA | $44,840 | 11.0120 | |
| Strategy 5- CMR-ICA | $44,981 | 11.0123 | $484,140 |
| Strategy 7- EST then ICA if +ve, EST- CMR - ICA if EST uncertain | $45,418 | 11.0133 | $444,017 |
| K) 60-yr-old male, CCS 2, prior likelihood of disease 39.5%, CMR costs $250 more than SPECT | |||
| Strategy 4- EST-SPECT-ICA | $44,967 | 11.0065 | |
| Strategy 5- CMR-ICA | $44,971 | 11.0120 | $698 |
| Strategy 7- EST then ICA if +ve, EST- CMR - ICA if EST uncertain | $45,467 | 11.0133 | $389,368 |
| L) 60-yr-old male, CCS 2, prior likelihood of disease 39.5%, CMR costs $500 more than SPECT | |||
| Strategy 4- EST-SPECT-ICA | $44,967 | 11.0065 | |
| Strategy 5- CMR-ICA | $45,190 | 11.0120 | $40,088 |
| Strategy 7- EST then ICA if +ve, EST- CMR - ICA if EST uncertain | $45,550 | 11.0133 | $282,955 |
Abbreviations: EST, exercise electrocardiogram stress test; CMR, stress cardiovascular magnetic resonance; ICA, coronary angiogram; ICER, incremental cost-effectiveness ratio; AUD$, Australian dollars; QALY, quality adjusted life years; SPECT, single-photon emission computed tomography; CCS, Canadian Cardiovascular Score for angina.
Reducing the prior likelihood to 20% (compared with 39.5% in the base-case) led to Strategy 3 dominating all other strategies (i.e., lower costs and more QALYs). Increasing the prior likelihood to 60% led to Strategy 7 (EST, followed by ICA if positive, or followed by CMR if EST inconclusive, followed by ICA if CMR positive or inconclusive) being the most cost-effective. Further increasing the prior likelihood to 80% led to Strategy 2 (EST, followed by ICA if EST positive or inconclusive) being the most cost-effective given the upper threshold value of $75,000 per QALY.
If EST was deemed inappropriate, i.e. the patient cannot exercise on a treadmill, then this led to Strategy 5 (CMR, followed by ICA if CMR positive or inconclusive) dominating all other strategies.
Because there was no set price for stress CMR at the time of this study, the cost increment of stress CMR compared with SPECT was varied to assess the impact on cost-effectiveness in the base case. If CMR costs were the same as SPECT or if the cost was $100 more than SPECT, the results were similar. If the CMR cost was increased to $250 and $500 more than SPECT then Strategy 3 became more costly and less effective than the other strategies, and thus no longer appeared among the non-dominated strategies. In these scenarios Strategy 5 (CMR, followed by ICA if CMR positive or inconclusive) became the most cost-effective strategy at the lower threshold value of $45,000 per QALY.
Discussion
The results of this economic evaluation based on strategies and modelling from the CE-MARC trial, show that Strategy 3 (EST, followed by CMR if EST is positive or inconclusive, followed by ICA if the CMR is positive or inconclusive) is the most cost-effective diagnostic strategy in diagnosing significant coronary disease in patients with stable chest pain at the Australian cost-effectiveness thresholds. These findings are consistent with the extensive published trial literature on stress CMR showing high diagnostic accuracy [
[1]
], high prognostic discrimination [[5]
,- Greenwood J.P.
- Herzog B.A.
- Brown J.M.
- Everett C.C.
- Nixon J.
- Bijsterveld P.
- et al.
Prognostic value of cardiovascular magnetic resonance and single-photon emission computed tomography in suspected coronary heart disease: long-term follow-up of a prospective, diagnostic accuracy cohort study.
Ann Intern Med. 2016; 165: 1-9
[21]
], clinical utility in acting as an effective gate-keeper for diagnostic ICA [[11]
], and as an effective first line strategy for guiding revascularisation decisions in patients with stable angina [- Greenwood J.P.
- Ripley D.P.
- Berry C.
- McCann G.P.
- Plein S.
- Bucciarelli-Ducci C.
- et al.
Effect of care guided by cardiovascular magnetic resonance, myocardial perfusion scintigraphy, or NICE Guidelines on subsequent unnecessary angiography rates: the CE-MARC 2 randomized clinical trial.
JAMA. 2016; 316: 1051-1060
[12]
]. These data support the wider use of stress CMR in Australian clinical practice.At present, however, there remain issues surrounding the perceived cost and accessibility of stress CMR in Australia. It is important to consider this because imaging tests for cardiovascular diseases contribute a large proportion of health care costs. Although an argument is often made that “simple” tests, such as EST, may cost less on face value, careful evaluation is needed about the cumulative cost of downstream investigations that may occur because of false-positive results.
Stress CMR is a mature imaging modality for the investigation of coronary artery disease, which is acknowledged in international clinical guidelines. The United Kingdom (UK) leads development in this field and Australia is in the early stages of adoption. Cardiac magnetic resonance trends in Australia are difficult to measure and currently there is limited reimbursement available for CMR scans. In 2008 there were 20,597 CMR scans performed in the UK and 38,485 in 2010, an increase of over 85% in only 2 years [
[22]
]. This increase in demand is being driven predominantly by the most common indications for CMR including the assessment of ischaemia and viability, and cardiomyopathy and heart failure [[22]
]. In contrast, within Australia, there is currently no reimbursement for CMR to investigate and manage most of these pathologies. Also, of note, Australia has different CMR related infrastructure compared to the UK—although it has more MRI units per million population than the UK [[23]
], there is a disproportionately lower utilisation rate of MRI [[24]
]. Not all magnets in Australia are suitably equipped to perform CMR (e.g. they need specific cardiac coils and cardiac software), and both adequate staffing (trained physicians) and appropriate reimbursement remain challenges to the expansion of this technology in Australia.The secondary analyses evaluating different scenarios also showed that stress CMR remains cost-effective at higher incremental costs than SPECT. However, in the scenario analyses, when the incremental cost of CMR was $250–$500 more than SPECT, Strategy 5 became dominant (CMR, followed by ICA if CMR positive or inconclusive). Changing the prior likelihood of disease in the scenarios analyses supported the use of stress CMR in the low-to-intermediate risk populations. Increasing the prior likelihood of disease to high risk (>80%) pushed stress CMR out and led to Strategy 2 (EST followed by ICA if positive) being the most cost-effective. This information may be helpful in terms of setting future restrictions on access to stress CMR in Australia.
Limitations
The approach of using costs based on the MBS excludes any potential out-of-pocket patient co-payments above the MBS fee that may occur at certain institutions and service providers in Australia, particularly in private health care.
This study doesn't include strategies and costs for stress echocardiography and CT coronary angiography, which are both frequently utilised in Australia. However, these investigations were not part of the CE-MARC trial upon which this cost effectiveness analysis was based and their inclusion would have necessitated the adoption of multiple assumptions, rather than relying on robust randomised clinical trial outcomes data.
Conclusions
This economic evaluation shows that an investigative strategy using stress CMR if EST is inconclusive or positive followed by ICA if the stress CMR is positive or inconclusive is the most cost-effective approach for diagnosing significant coronary disease in chest pain patients within the Australian health care system. These findings support more widespread adoption of stress CMR for CAD assessment in Australian clinical practice, and are consistent with the high diagnostic accuracy of stress CMR and cost effectiveness of this modality demonstrated in other health care systems.
Funding Sources
This research did not receive any specific grant funding.
Declarations of Interest
We report no competing interest associated with the work reported in this manuscript.
Conflicts of Interest
There are no conflicts of interest to disclose.
Acknowledgements
None.
Appendices. Supplementary Data
- Appendix Tables 1 and 2
References
- Cardiovascular magnetic resonance and single-photon emission computed tomography for diagnosis of coronary heart disease (CE-MARC): a prospective trial.Lancet. 2012; 379: 453-460
- Comparison of stress cardiovascular magnetic resonance imaging (CMR) with stress nuclear perfusion for the diagnosis of coronary artery disease.J Nucl Cardiol. 2016; 23: 287-297
- Validation of magnetic resonance myocardial perfusion imaging with fractional flow reserve for the detection of significant coronary heart disease.Circulation. 2009; 120: 2207-2213
- Prognostic value of adenosine stress perfusion cardiac MRI with late gadolinium enhancement in an intermediate cardiovascular risk population.Int J Cardiol. 2013; 167: 2055-2060
- Prognostic value of cardiovascular magnetic resonance and single-photon emission computed tomography in suspected coronary heart disease: long-term follow-up of a prospective, diagnostic accuracy cohort study.Ann Intern Med. 2016; 165: 1-9
- Cost-effectiveness of cardiovascular magnetic resonance in the diagnosis of coronary heart disease: an economic evaluation using data from the CE-MARC study.Heart. 2013; 99: 873-881
- Cost-minimization analysis of three decision strategies for cardiac revascularization: results of the “suspected CAD” cohort of the European Cardiovascular Magnetic Resonance Registry.J Cardiovasc Magn Reson. 2016; 18: 3
- Cost-effectiveness of cardiovascular magnetic resonance and single-photon emission computed tomography for diagnosis of coronary artery disease in Germany.J Cardiovasc Magn Reson. 2013; 15: 30
- Cost-effectiveness of functional cardiac imaging in the diagnostic work-up of coronary heart disease.Eur Heart J Qual Care Clin Outcomes. 2016; 2: 201-207
- Cost-effectiveness analysis for imaging techniques with a focus on cardiovascular magnetic resonance.J Cardiovasc Magn Reson. 2013; 15: 52
- Effect of care guided by cardiovascular magnetic resonance, myocardial perfusion scintigraphy, or NICE Guidelines on subsequent unnecessary angiography rates: the CE-MARC 2 randomized clinical trial.JAMA. 2016; 316: 1051-1060
- Magnetic resonance perfusion or fractional flow reserve in coronary disease.N Engl J Med. 2019; 380: 2418-2428
- Clinical evaluation of magnetic resonance imaging in coronary heart disease: the CE-MARC study.Trials. 2009; 10: 62
- Cost effectiveness of perindopril in reducing cardiovascular events in patients with stable coronary artery disease using data from the EUROPA study.Heart. 2007; 93: 1081-1086
- Technical guidelines for preparing assessment reports for the Medical Services Advisory Committee – service type: investigative (Version 3.0).(Available at:)http://www.msac.gov.au/internet/msac/publishing.nsf/Content/0BD63667C984FEEACA25801000123AD8/$File/InvestigativeTechnicalGuidelines-December-2016-Version-3.0.pdfDate: 2017Date accessed: November 19, 2019
- 3303.0 - Causes of death Australia 2017 2018.(Available at:)https://www.abs.gov.au/AUSSTATS/[email protected]/DetailsPage/3303.02017?OpenDocumentDate accessed: November 19, 2019
- Health-related quality of life measured using the EQ-5D-5L: South Australian population norms.Health Qual Life Outcomes. 2016; 14: 133
- Health Expenditure Australia 2016-17. Health and Welfare Expenditure Series No. 64.(Cat. no. HWE 74) AIHW, Canberra, ACT2018
- Cardiac MRI for Myocardial Stress Perfusion and Viability Imaging in Patients with Known or Suspected Coronary Artery Disease. MSAC Application Number 1237, Assessment Report.Commonwealth of Australia, Canberra, ACT2016
- Estimating the reference incremental cost-effectiveness ratio for the Australian health system.Pharmacoeconomics. 2018; 36: 239-252
- Prognostic value of vasodilator stress cardiac magnetic resonance imaging: a multicenter study with 48000 patient-years of follow-up.JAMA Cardiol. 2019; 4: 256-264
- Cardiovascular magnetic resonance activity in the United Kingdom: a survey on behalf of the British Society of Cardiovascular Magnetic Resonance.J Cardiovasc Magn Reson. 2011; 13: 57
- MRI units, 2015 (or nearest year).in: Health Care Activities. OECD Publishing, Paris2017
- MRI exams, 2015 (or nearest year).in: Health care activities. OECD Publishing, Paris2017
- Australian Government Department of Health. Medicare Benefits Schedule Book, Operating from 01 May 2017.DoH, Canberra2017
- Independent Hospital Pricing Authority. National Hospital Cost Data Collection Cost Report: Round 20, Financial Year 2015-16.IHPA, Canberra2018
- Cost-effectiveness of a coronary heart disease secondary prevention program in patients with myocardial infarction: results from a randomised controlled trial (ProActive Heart).BMC Cardiovasc Disord. 2013; 13: 33
- Hospital costs of older people in New South Wales in the last year of life.Med J Aust. 2007; 187: 383-386
Article info
Publication history
Published online: August 08, 2020
Accepted:
July 6,
2020
Received in revised form:
June 20,
2020
Received:
December 7,
2019
Identification
Copyright
Crown Copyright © 2020 Published by Elsevier B.V. on behalf of Australian and New Zealand Society of Cardiac and Thoracic Surgeons (ANZSCTS) and the Cardiac Society of Australia and New Zealand (CSANZ). All rights reserved.
ScienceDirect
Access this article on ScienceDirectLinked Article
- Letter to the Editor Regarding: “Cost-Effectiveness of Cardiovascular Magnetic Resonance in Diagnosing Coronary Artery Disease in the Australian Health Care System”, Heart Lung Circ. 2021;30(3):380–387Heart, Lung and CirculationVol. 30Issue 10
- PreviewWe read with interest the article by Kozor et al. on the cost effectiveness of stress perfusion cardiovascular magnetic resonance (stress CMR) in the assessment of stable coronary artery disease (CAD) [1]. In this hypothetical, analytical model, the authors concluded that exercise electrocardiogram stress testing (EST), followed by stress CMR then invasive coronary angiography (ICA) was the most cost-effective diagnostic strategy for CAD in the Australian health care system [1]. It is important to acknowledge that this study only assessed strategies utilising EST, stress CMR, stress single-photon emission computed tomography (SPECT) and ICA.
- Full-Text
- Preview
