Jump to
Abstract
Background—The Second Medicine, Angioplasty, or Surgery Study (MASS II) included patients with multivessel coronary artery disease and normal systolic ventricular function. Patients underwent coronary artery bypass graft surgery (CABG, n=203), percutaneous coronary intervention (PCI, n=205), or medical treatment alone (MT, n=203). This investigation compares the economic outcome at 5-year follow-up of the 3 therapeutic strategies.
Methods and Results—We analyzed cumulative costs during a 5-year follow-up period. To analyze the cost-effectiveness, adjustment was made on the cumulative costs for average event-free time and angina-free proportion. Respectively, for event-free survival and event plus angina-free survival, MT presented 3.79 quality-adjusted life-years and 2.07 quality-adjusted life-years; PCI presented 3.59 and 2.77 quality-adjusted life-years; and CABG demonstrated 4.4 and 2.81 quality-adjusted life-years. The event-free costs were $9071.00 for MT; $19 967.00 for PCI; and $18 263.00 for CABG. The paired comparison of the event-free costs showed that there was a significant difference favoring MT versus PCI (P<0.01) and versus CABG (P<0.01) and CABG versus PCI (P=0.01). The event-free plus angina-free costs were $16 553.00, $25 831.00, and $24 614.00, respectively. The paired comparison of the event-free plus angina-free costs showed that there was a significant difference favoring MT versus PCI (P=0.04), and versus CABG (P<0.001); there was no difference between CABG and PCI (P>0.05).
Conclusions—In the long-term economic analysis, for the prevention of a composite primary end point, MT was more cost effective than CABG, and CABG was more cost-effective than PCI.
Clinical Trial Registration Information—www.controlled-trials.com. Registration number: ISRCTN66068876.
- coronary artery disease
- coronary revascularization
- cost-effectiveness
- multivessel coronary artery disease
Introduction
Therapeutic strategies for multivessel coronary artery disease (CAD) with stable angina and preserved ventricular function are medical treatment (MT), percutaneous coronary intervention (PCI), and coronary artery bypass graft (CABG). All 3 treatment regimens have similar efficacy regarding prevention of myocardial infarction (MI) and death.1–4 The difference is that surgical patients have a lower necessity of further revascularization procedures. In this scenario of therapeutic equivalence, the economic consequences of therapeutic strategies are an important outcome. An initial cost of PCI is usually lower than that of CABG; however, additional procedures, even with the use of stents, increase the cost.5–8 Medical treatment is less costly, although it is usually less effective for alleviating symptoms.9 This apparent economic advantage can hide the costs applied to the long-term follow-up with repeated examinations, minor clinical interventions, clinical events involving hospitalization, or even percutaneous interventions or a surgical emergency. Thus, the costs of conservative treatment may hide higher costs that interventional treatments do not hide.
Currently, there is a lack of clinical trials comparing economic outcomes of these 3 therapeutic strategies for multivessel CAD. Trials analyze mainly the comparative costs between the surgical and percutaneous intervention or between PCI and MT.3,6–8 In fact, the prospective, randomized study we designed compared the hospital costs of CABG with and without cardiopulmonary bypass under certain conditions indicating advantages for one or another technique.10 In addition, initial data on comparative costs of short-term follow-up surgical, percutaneous, or clinical treatment revealed an initial advantage for the clinical treatment.11
This report is a post hoc analysis of a 5-year follow-up of the second Medicine, Angioplasty, or Surgery Study (MASS II) trial. This current investigation compares the economic outcome of the 3 therapeutic strategies for multivessel CAD and preserved ventricular function.
Methods
Subjects
Details of the MASS II design, study protocol, patient selection, inclusion criteria, and general results have been previously published.2,12 Briefly, the MASS II study was a prospective, randomized, controlled clinical trial designed to compare medical treatment, angioplasty/stent placement (PCI), and surgical myocardial revascularization (CABG) with cardiopulmonary bypass in patients with stable multivessel CAD and preserved left ventricular function. The predefined primary end point was the combined incidence of overall mortality, MI, or persistent angina requiring revascularization. All data were analyzed according to the intention-to-treat principle. In other words, all randomized patients were included in all analysis, including patients not receiving the treatment to which they were randomized. Patients with angiographically documented proximal multivessel coronary stenosis (>70% stenosis) by visual assessment and documented ischemia were considered for inclusion. Patients were enrolled and randomized if the surgeons, attending physicians, and interventional cardiologists agreed that revascularization could be attained by either strategy. Clinical exclusion criteria included unstable angina or acute MI requiring emergency revascularization, ventricular aneurysm requiring surgical repair, left ventricular ejection fraction of <40%, a history of PCI or CABG, single-vessel disease, and left main coronary artery stenosis ≥50%. Patients were also excluded if they had another coexisting condition that was a contraindication to CABG or PCI. In this trial, all patients were placed on an optimal medical treatment for CAD.
Patients were randomized to continue with optimal MT alone or to undergo PCI or CABG concurrently with optimal MT. All patients received optimal medical treatment; angiotensin-converting enzyme inhibitors, statins, β-blockers, calcium channel blockers, diuretics, and nitrates were available for all groups.
Patients were enrolled in a single clinical site at the Heart Institute (InCor) of the University of São Paulo Medical School in Brazil. Patients gave written, informed consent and were randomly assigned to a treatment group. The Ethics Committee of the Heart Institute approved the trial, and all procedures were performed in accordance with the Helsinki Declaration.
Clinical Follow-Up
Factors and adverse clinical events were recorded from the date the patient signed the consent form. The patients had outpatient follow-up quarterly in the first year and every 6 months for 5 years of follow-up. Additional outpatient visits were available if necessary. Routine laboratory tests were carried out every 6 months, and the resting electrocardiogram, stress test, and scintigraphic or angiographic studies were available if angina symptoms were difficult to control.
Angina symptoms were graded according to severity, from 1 to 4, as previously defined.13 Angina is considered unstable only when the patients received a full dose of anti-ischemic drugs without an effective response. The performance of a subsequent revascularization procedure (PCI or CABG) was determined by the presence of high-grade angina or MI. MI was considered present in accordance with new Q waves in at least 2 electrocardiographic leads or symptoms of chest pain characteristics associated with increased serum levels of creatine kinase-MB 3 times above the normal reference value. Stroke was diagnosed before the observation of the onset of neurological deficit associated with structurally compatible lesions identified by CT scan or MRI.
Economic Analysis
The economic analysis compared the cumulative costs of each therapeutic strategy in the 5-year follow-up period. The resources analyzed included: (1) for patients in the surgical or PCI group, the costs were applied for each patient based on the standard cost of these interventions. With the occurrence of a new intervention, the cost was added according to the standard cost of the particular intervention. For patients in the clinical group, charges were made for each kind of mechanical intervention observed during follow-up; (2) in-hospital complications of revascularization procedures; (3) outpatients visits; (4) medications; (5) cardiovascular tests (treadmill exercise test, echocardiogram, single photon emission CT, coronary arteriography); (6) subsequent hospitalization for cardiovascular disease (MI, unstable angina, stroke, and death); and (7) subsequent revascularization procedures and in-hospital complication.
Costs are expressed in US dollars. Costs of resources in our institution are: $10 294.00 for CABG; $2647.00 for angioplasty (PCI); $2282.00 for each bare metal stent; $5882.00 for hospitalization due to MI; $5294.00 for hospitalization due to unstable angina; $3529.00 for hospitalization due to stroke; $705.00 for a coronary angiogram; $362.00 for single photon emission CT; $85.00 for an echocardiogram; $47.00 for a treadmill exercise test; and $24.00 for outpatient visits. The estimated mean costs of in-hospital complications of a revascularization procedure are: $6470.00 for cardiogenic shock and use of intra-aortic counterpulsation balloon; $5882.00 for nosocomial pneumonia or acute respiratory distress syndrome; $2941.00 for acute renal failure and hemodialysis; $7058.00 for mediastinitis or sternum osteomyelitis; and $5000.00 for cardiac surgery because of bleeding and/or cardiac tamponade.
The 5-year medication costs were based on Arterial Revascularization Therapies Study (ARTS) trial. It was $5000.00 for PCI and CABG treatments and $6000.00 for MT group.8,14
Cost-Effectiveness
Cost-effectiveness analysis was performed by quality-adjusted life-year (QALY). It was combined evidence of effectiveness and the cost of the treatment expressed in dollars per event-free year of life gained from each treatment modality. An additional adjustment was made for the angina-free proportion at 5-year evaluation.
Thus, an adjustment for 5 years' event-free costs was made multiplying the actual costs by the ratio 5 over average time to the first event; further adjustment was made for angina alleviation by the ratio of event-free costs over angina free proportion. Thus, we introduced a QALY analysis in the present study for event-free survival and event and angina-free survival.
Statistical Analysis
Statistical analysis was carried out by using the t test and analysis of variance for normally distributed variables, nonparametric analysis of variance for non-Gaussian distribution, and χ2 for categorical variables. The pairwise probability values were calculated using the 2-sided method. The average time to first event was used to adjust costs and was based on Kaplan-Meier curves as previously published.2 Values of P<0.05 were considered statistically significant. Cumulative costs were compared on the intention-to-treat basis. Statistical analysis was performed with SAS 9.1 software (SAS Institute Inc, Chicago, IL).
Results
A total of 611 eligible patients who had all met inclusion criteria were randomly assigned to one of 3 therapeutic treatments: CABG (n=203), PCI (n=205), or MT (n=203). Randomization created balanced treatment groups with respect to important prognostic characteristics, as shown in Table 1. All patients received medical regimens according to a predefined approach. No patient was lost to follow-up. The minimal duration of follow-up was 5 years. Major adverse cardiac events at the 5-year follow-up are demonstrated in Table 2. As reported previously, there is a significant statistical difference in the incidence of the primary end point among these groups.2 The patients allocated to the CABG group had a lower incidence of the primary end point. No difference existed in overall mortality and nonfatal MI among the groups. The frequency of additional revascularization was the only component of the primary end point that was significantly different between the groups: 3.5% of patients undergoing CABG, 24% of medical patients, and 32% of stenting patients.
Characteristics of the 611 Patients Assigned to MT, PCI, and CABG in MASS II
Major Adverse Cardiac Events at 5-Y Follow-Up
Surgical Therapy
Of the 203 randomized patients, 198 underwent CABG with an average of 10 days of hospitalization. It was recorded, as in-hospital complications of revascularization procedures, 7 episodes of stroke, 4 episodes of cardiogenic shock with use of an intra-aortic balloon pump, 4 of acute renal failure and hemodialysis, one cardiac surgery for cardiac tamponade, 11 nosocomial pneumonias, and 2 episodes of wound infection. During the 5-year follow-up, 2 patients required another surgery, and 7 patients required PCI (mean, one stent per patient). During the follow-up, 11 episodes of unstable angina, 20 strokes, and 32 deaths were recorded. After randomization, 901 treadmill exercise tests, 310 echocardiograms, 40 single photon emission CTs, 67 coronary arteriographies, and 2350 outpatient visits were performed. The median cumulative cost of the surgical treatment, after 5 years of follow-up, was $16 072.00 per patient.
Medical Therapy
Two hundred three patients were randomized to MT. After 5 years of follow-up, 40 patients underwent surgical revascularization (mean 11.6 days of hospitalization). It was recorded, as in-hospital complications of revascularization procedures, 2 episodes of stroke, 2 nosocomial pneumonias, and one episode of mediastinitis. In addition, 28 PCIs were performed (average of 3.7 days of hospital stay and 1.1 stent per patient). After 5 years of follow-up, there were 8 episodes of unstable angina, 15 strokes, and 35 deaths. After randomization, 920 treadmill exercise tests, 355 echocardiograms, 24 single photon emission CTs, 87 coronary arteriographies, and 2458 outpatient visits were performed. The median cumulative cost of the medical group was $6876.00 per patient.
Percutaneous Coronary Intervention
Of the 205 randomized patients, 194 underwent PCI (average of 3.46 days of hospitalization and 1.32 stents per patient). It was recorded, as in-hospital complications of revascularization procedure, 2 episodes of stroke, 2 episodes of emergency CABG, and 2 additional emergency PCIs. After 5 years of follow-up, 64 additional PCIs were necessary (average of 3.1 days of hospitalization and 0.78 stent per patient). In addition, 21 CABGs were performed (mean 12 days of hospital stay). After the follow-up period, there were a total of 22 episodes of unstable angina, 12 strokes, and 28 deaths. Furthermore, 988 treadmill exercise tests, 311 echocardiograms, 123 single photon emission CTs, 138 coronary arteriographies, and 2593 outpatient visits were performed. The median cumulative cost of the treatment group angioplasty was $14 337.00 per patient.
The paired comparison of the cumulative costs showed that there was a significant difference favoring MT versus PCI (P<0.01) and versus CABG (P<0.01); there was no difference between CABG and PCI. The cumulative cost of each treatment group, with the 25% and 75% percentiles, and minimal and maximal values is demonstrated in Figure 1. The proportion of expenditure by the cumulative cost in each treatment group is demonstrated in Figure 2, whereas median of costs among 3 therapeutic strategies are shown in Figure 3.
Five-year costs among 3 therapeutics strategies. The medians of medical treatment and CABG are equal to the 25th percentile. Tukey comparison: MT<PCI and CABG (P<0.0001); PCI and CABG (P>0.05). MT indicates medical treatment; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft.
The proportion of expenditure by the cumulative cost. MT indicates medical treatment; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft.
Composition of Costs in MASS II, 5-year follow-up. MASS indicates Medicine, Angioplasty, or Surgery Study; ACS, acute coronary syndrome; CABG, coronary artery bypass graft; MT, medical treatment; PCI, percutaneous coronary intervention.
Cost-Effectiveness
To analyze the cost-effectiveness, as cited in the methodology, the average event-free time and angina-free proportion were calculated, as demonstrated in Table 3. The average time for first event was 3.79 years for MT, 3.59 years for PCI, and 4.4 years for CABG. There is a significant statistical difference favoring surgical patients (P=0.0026). At 5-year follow-up, 54.8% of the MT group was free of angina. On the other hand, the angina-free proportion was 77.3% for PCI and 74.2% for CABG with a significant statistical difference in comparison with medical group patients (P<0.001). Therefore, respectively, for event-free survival and event plus angina free survival, MT presented 3.79 QALYs and 2.07 QALYs; PCI presented 3.59 and 2.77 QALYs; and CABG demonstrated 4.4 and 2.81 QALYs.
Event-Free and Angina-Free Proportion
At the 5-year follow-up, the event-free costs were $9071.00 for MT; $19 967.00 for PCI; and $18 263.00 for CABG. All these figures represent the median cumulative cost for each patient during 5 years of follow-up. The paired comparison of the event-free costs showed that there was a significant difference favoring MT versus PCI (P<0.01) and versus CABG (P<0.01); a significant difference favored CABG versus PCI (P=0.01), as shown in Figure 4.
Event-free costs for each patient at 5-year follow-up. The median was equal to the 25th percentile in MT and CABG. Tukey multiple comparison: MT versus PCI and CABG; P<0.001; CABG versus PCI; P=0.04. MT indicates medical treatment; PCI, percutaneous coronary intervention; CABG, coronary artery bypass graft.
The event-free plus angina-free costs were $16 553.00 for MT; $25 831.00 for PCI; $24 614.00 for CABG. All these figures represent the median cumulative cost for each patient during 5 years of follow-up. The paired comparison of the event-free plus angina-free costs showed that there was a significant difference favoring MT versus PCI (P<0.01) and versus CABG (P<0.01); there was no difference between CABG and PCI, as shown in Figure 5. Refinements of comparative costs among event-free patients and those with crossover are shown in Table 4.
Event-free and angina-free for each patient at 5-year follow-up. Median is equal to the 25th percentile in MT and CABG. Tukey multiple comparison: MT versus PCI and CABG; P<0.0001; PCI versus CABG; P>0.05. MT indicates medical treatment; CABG, coronary artery bypass graft; PCI, percutaneous coronary intervention.
Five-Year Median Costs of the 3 Therapeutic Strategies
Discussion
The MASS II trial demonstrated a higher rate of repeat revascularizations in MT and PCI groups compared with the CABG group; the worst health-related quality of life also occurred in the MT patients secondary to the reduced angina-free proportion. However, regarding a hard event such as MI or death, there was no difference among these 3 regimens. This is a good scenario to compare cost-effectiveness, a randomized trial demonstrating a similar hard end point. Cost analysis is a challenging task with so many potential biases.15–18 To minimize such biases, this report performed an extensive cost analysis, trying to account for the global cost of the management of randomized patients with chronic coronary disease, in a long-term follow-up: outpatient visits, medication, cardiovascular tests, hospitalization for cardiovascular disease, revascularization procedures, and in-hospital complications of revascularization procedures.
This current analysis demonstrates that MT was more cost-effective than CABG, and CABG was more cost-effective than PCI for event-free survival. This result is consistent with the previous economic outcomes reported by other randomized clinical trials.3,4,19 On the other hand, MT was the most cost-effective approach; this advantage can be attributed to clinical stability and also to better ventricular function. Moreover, in this study, the angina-free status of the population was the driver of clinical outcomes.
Some other randomized clinical trials do compare PCI with MT. The second Randomized Intervention Treatment of Angina (RITA-2) trial demonstrated an overall mean additional cost of $4194.00 per patient over a 3-year period in patients randomized to PCI.19 The Clinical Outcomes Utilizing Revascularization and Aggressive Drug Evaluation (COURAGE) trial demonstrated that the addition of PCI to optimal MT in the early management of symptomatic patients is not a cost-effective measure. In this trial, the cumulative cost was significantly higher ($10 125.00) in the PCI group. The cost estimated for the improvement of angina was $154 580.00 for one patient.3 Recently, Hlatky et al4 reinforced the strategy of MT over prompt revascularization with PCI. Revascularization increases 4-year cost significantly, approximately $5700.00.
In the context of the comparison of CABG to MT, there is a lack of randomized clinical trials. The MASS II trial, in a 1-year follow-up, demonstrated a higher cost for CABG over MT. Furthermore, MT had the greatest increase in expected costs (317% versus 21%), because of its lowest ratio of alleviation of angina.11 Most recently, based on the 4-year Bypass Angioplasty Revascularization Investigation 2 Diabetes (BARI 2D) data, it was demonstrated that medical costs were higher for surgical revascularization than for medical therapy ($20 300.00; P=0.0001).4 It is interesting that, in our 5-year follow-up, this economic advantage of MT remains present in a population with more extensive anatomic disease and severe angina symptoms. Nearly 60% of patients had triple-vessel disease with proximal left anterior descending disease, and nearly 80% had Canadian Cardiovascular Society Class II or III angina.
Regarding CABG versus PCI, the medical care cost was compared in the Bypass Angioplasty Revascularization Investigation (BARI) trial. After a mean 11.4-year follow-up, there was no difference between the cumulative costs of both strategies. The hospital cost of surgical patients was higher. Patients in the PCI group had a greater necessity for reintervention and higher cost of drug therapy (P=0.009). The randomization of this study occurred between August 1988 and August 1991; therefore, the methodological limitation is the lack of availability of the stent, and only balloon angioplasty was performed. With the lack of this device, a contemporary economic analysis becomes difficult.7 In 2001, the cost-effectiveness of the ARTS was published.14 At the end of the index procedure, the total cost of PCI compared with CABG was significantly lower ($6441.00 versus $10 653.00; P<0.001), a total net difference of $4212.00. However, due to the high recurrence rate of revascularization in the PCI group compared with CABG (21% versus 3.8%), the difference declined to $2973.00 in the first year of follow-up. Considering the reduction of approximately 14% in the combined event rate provided by CABG (73.8% versus 87.8%), and the high incidence of further reintervention procedures, the time limitation of a 1-year follow-up of this study is a methodological constraint to achieving an economic analysis.8
Our result reinforces the fact that to minimize bias, it is necessary to make an economic analysis in a late follow-up. In our first year of analysis, initial average cost of treating a patient randomized to PCI was approximately 75% of that for the surgical therapy. Moreover, at 5-year follow-up, CABG becomes more cost-effective for the prevention of the composite primary end point and angina. A similar finding was obtained in the BARI trial; after the initial revascularization, the cost of PCI (without stents) was 35% lower than CABG.7 However, at the end of 5 years, this difference decreased to 5%. After 12 years, there was no difference between the average cumulative costs of both interventional strategies (PCI=$120 750.00 versus CABG=$123 000.00; P=0.55).
An important observation is the low cost-effectiveness for the improvement of angina in a population with extensive CAD and severe angina symptoms. Compared with the medical therapy group, PCI and CABG groups experienced significantly better angina relief. This fact decreased 82% of the cost-effectiveness of MT. As previously demonstrated, the cost estimated for the improvement of angina is very high.3
Final Considerations and Conclusion
In conclusion, this current analysis demonstrates that the adoption of prompt revascularization, with PCI or CABG, is more expensive and less cost-effective than providing an initial optimal medical treatment and performing a revascularization procedure only when necessary. Additionally, when revascularization becomes necessary, surgical therapy is the most cost-effective.
To our knowledge, this is the first randomized trial that compares, in a long-term economic outcome, these 3 therapeutic strategies for patients with multivessel CAD and preserved ventricular function. We believe that this report provides important information for public health, optimizing healthcare spending and sparing expenses.
On the other hand, assuming that all systems of healthcare financing can encompass a methodological bias, we think the most accurate model to reflect the actual costs are based on private payers. This model would be considered “real world.”
Sources of Funding
Financial support for the present study was provided in part by a research grant from the Zerbini Foundation, São Paulo, Brazil. Medical writing support was provided by Ann Conti Morcos during the preparation of this article, supported by the Zerbini Foundation.
Disclosures
None.
- © 2012 American Heart Association, Inc.
References
This Issue
Jump to
Article Tools
Share this Article
- Cost-Effectiveness Analysis for Surgical, Angioplasty, or Medical Therapeutics for Coronary Artery Disease
