Relative Cost Comparison of Treatments for Coronary Artery Disease: The First Year Follow-Up of MASS II Study
Background— Prior comparisons of costs following CABG and PTCA have demonstrated higher initial costs after CABG but following PTCA, recurrent symptoms and repeat revascularization result in increased late costs and over time their costs equilibrate. The MASS II trial provides an opportunity to compare the costs of CABG and PTCA in addition to a strategy of medical therapy.
Methods— We studied the 611 patients of MASS II [Medical (203), Angioplasty (205), or Surgery (203) Study], a randomized study to compare treatments for multivessel CAD and preserved left ventricle function. The costs were: CABG US$ 10 650.00; PTCA US$ 6400.00; new AMI hospitalization AMI U$ 2550; angiography not followed-up of PTCA US$ 1900.00; and medication US$ 1200.00 for medical, and US$ 1000.00 for the other groups. We did adjustment for average event-free time, and angina-free proportion. The statistical analysis carried out was chi-square, t test, and analysis of variance.
Results— After 1 year, 49% Medical, 79% PTCA, and 88% CABG became angina-free; P<0.0001. There were 26 coronary angiograms (5 medical, 17 PTCA, and 4 CABG), 23 AMI (8 medical, 17 PTCA, and 6 CABG; P=0.03); PTCA was performed in 7 Medical, 17 PTCA, and 1 CABG, (P=0.0003), CABG was performed in 15 Medical, 8 PTCA, and zero CABG; P=0.002. The event-free and event and angina-free-costs in the first year were US$ 2453.50 and US$5006.32 for Medical; US$ 10348,43; and US$ 13 099.31 for PTCA; and US$ 12 404.21 and US$ 14 095.09 for CABG group. An increase from expected costs of 317%, 77%, and 21%, respectively.
Conclusion— PTCA effective costs were similar to CAGB costs, Medical treatment presented the lowest cost, and however, the greatest increment, and CABG presented the most stable costs.
The survival rate of patients with stable coronary and preserved left ventricular function is similar under the 3 modality of treatment, Medical, Coronary Angioplasty (PTCA), or Coronary Aorta Bypass Graft surgery (CABG).1–3 However, usually CABG is associated to lower necessity of further revascularization procedures, mainly because of its higher proportion of patients turned asymptomatic.1,3–5
Initial costs of PTCA is usually lower to the CABG, however, additional procedures, even with the use of stents, elevate the cost so that ultimately costs are similar to CABG.2,3,5
Medical treatment is less costly, however, is usually less effective for alleviating symptoms.1
We hypothesized that the use of full medication including statins, the advent and widespread utilization of stents and the use of arterial grafts would alter the cost/benefice ratio in the treatment of stable coronary disease with preserved left ventricle function.
Our objective in this study was to compare costs of the 3 treatment modalities for CAD: Medical treatment, PTCA-Stent, and CABG in the first year of these treatments.
From 1995 to 2000, 2076 candidates to myocardial revascularization in the Heart Institute (InCor) were evaluated, and 611 patients (203 Medical, 205 PTCA, and 203 CABG) were enrolled in MASS II, a randomized study designed to compare Medical treatment, PTCA-Stent, and CABG in multivessel CAD with preserved left ventricular function.
Randomization process was done by generation of random numbers before the beginning of study, and only the statistical board was aware to the sequence of randomization number and the linked treatment code before the patient enrollment in the study. Ethical committee of the Hospital approved the study and all patients signed an informed consent.
The inclusion criteria included symptomatic multivessel coronary disease (2 or more epicardic coronary with ≥70% lesion), preserved left ventricular function, that is, normal ejection fraction. Additionally, the coronary lesions have to be amenable to angioplasty.
The primary objective of study was to compare the incidence of the major cardiac events: acute myocardial infarction (AMI), necessity of revascularization procedures and death, and symptom relief. Control coronary angiogram was not included for cost calculation.
Additional evaluations for quality of life by SF-36 and cost/efficiency analysis were performed.
We expressed the costs in US$ dollar prices of the ARTS study5. PTCA-stent procedure was charged US$ 6400.00; US$ 10 650.00 for CABG; US$ 2550.00 for one hospitalization because of AMI, and US$ 1900.00 for 1 coronary angiogram; 1-year medication costs were US$ 1000.00 for PTCA and CABG treatments, and US$ 1200.00 for Medical treatment group.
Perioperative infarction was defined as all acute myocardial infarction occurring in first 30 days of revascularization procedures. The classic diagnostic criteria for AMI were used: prolonged typical chest pain; ST displacement or new Q waves on ECG, and elevation of marker of myocardial injure, doubling of CKMB, except in the postoperative CBAG when the latter has to be over 20 ng/mL
The cost analysis was made by adding up the value of each AMI, diagnostic or revascularization procedures to the expected initial costs, thus, obtaining the actual costs. Cost/efficiency analysis was performed by combining the evidence of effectiveness and cost of the treatment and was expressed in dollars per event free-year life gained from each treatment modality.6,7 Additional adjustment was performed for angina-free proportion at 1-year evaluation.6,7
Thus, eventually we got the event-free costs (US$/year of event-free follow-up) by division of the mean actual costs by the relation of event-free time over 12 months. Further we also adjusted for angina-free proportion by division of event-free costs by proportion of angina-free patients at the end of the first year of follow-up (US$/angina and event free follow-up).
Statistical analysis was carried out by t test and ANOVA for normally distributed variable and chi-square for categorical nongaussian distributed variables. We chose 5% probability for statistical significance. All statistical analysis was done by the intention to treat principle.
The initial characteristics were presented in Table 1. The three groups were homogenous in age, male/female ratio, prevalence of risk factor, and proportion of positive exercise tests, however, PTCA group presented a greater prevalence of previous acute myocardial infarction (P=0.03), and CABG a greater frequency of class III or IV angina pectoris (P=0.05).
Clinical Status at 12-Month Evaluation
The incidence of death was similar among the 3 groups: 3 (1.9%) in Medical, 9 (4.4%) in PTCA, and 6 (3.9%) in CABG. Operative deaths, death within first 30 days of procedure, occurred in 6 (3%) for CABG group, and 5 (2.4%) in PTCA group.
There were a greater proportion of angina-free patients in CABG group (88%), PTCA group (79%); followed by Medical (49%); the differences were statistically significant (P<0.0001).
Acute Myocardial Infarction Hospitalization and Additional Procedures
Acute myocardial infarction (AMI) occurred in 8 (3.9%) patients in Medical group, 17 (8.3%) in PTCA, and in 6 CABG (3.0%); P=0.03. The incidence of perioperative AMI was similar for PTCA and CABG, 4 (1.5%) and 3 (2%), respectively.
Coronary angiographies that were not part of PTCA procedures occurred in 5 (2.5%) patients of Medical group, 17 (8.3%) in PTCA patients, and 6 (3.0%) in CABG group (P=0.002)
The number of angioplasties in the follow-up period were 7 (3.5%) in the Medical group, 17 (8.3%) in PTCA group, and 1 (0.5%) in CABG; P=0.0003). The other procedures were surgical revascularization performed in 15 (7.4% in medical group, 8 (3.9%) in PTCA group, and none of CABG group; P=0.0007
There were similar rates of revascularization procedures in Medical and PTCA groups, 22 (10.8%) and 25 (12.2%), respectively.
The average time to first event was 4.6±3.5 months for Medical treatment; 4.65±3.4 months fir PTCA group; and 3.7±2.7 months for CABG. The average time to event were not statistically different; however, the average time of survival without event did reach statistically significant difference, it was 11.18±2.60 months for medical treatment, 10.06±3.72 months for PTCA, and 11.41±2.33 months for CABG group; P<0.0001. The pair comparison showed that there was no difference between CABG and medical treatment (P=0.17), however, PTCA event-free time proved to be lower than medical treatment (P=0.0002), and CABG group (P<0.0001).
Cost and Efficiency Analysis
The actual unitary costs have suffered elevation from the predicts ones; the elevation was most pronounced in Medical (90.4%), and PTCA group (17%), than in CABG group (1%), the mean costs were US$ 2285.47±2990.90, US$ 8675.85±2797.24, and US$11 794.33±811.14, for Medical, PTCA, and CABG, respectively.
We adjusted the costs to event-free patient-year follow-up for each treatment for comparing efficiency, than the figures turned to the following: the 1 year event-free cost for medical treatment became US$ 2453.50±3210.27, PTCA US$ 10 348.93±3336.67, and CABG US$ 12 404.21±926.71.
Further adjusting for angina-free proportion gave following for patient-year event-free and angina-free costs: the Medical treatment increased to US$ 5006.32±6551.52; US$ 13 099.31±4223.62 for PTCA group, and US$ 14 ,095.69±1053.08 for CABG group, an increase of expected costs of 317.2%, 77%, and 21%, respectively (Figure 1)
We observed no difference in the incidence of death, however, the greater occurrence of the necessity of latter revascularization procedures in PTCA group elevated its costs to that of CABG in the first year of MASS II study.
The observed no difference in the incidence of death was already expected because the patients presented preserved left ventricular function, and this subset of patients presented comparable survival rate at CASS trial1 and in other studies with PTCA, like RITA, BARI, and EAST.2–4
The observed incidence of additional revascularization procedures in the PTCA group, 11.7% was similar to other PTCA-stent studies.5
Our findings, the progressive equalization of costs between PTCA treatment and CABG, reproduced those of previous randomized studies. The RITA trial showed that initial average cost of treating a patient randomized to PTCA was about 52% of that of CABG, but after 2 years this increased to about 80% because of the greater need for subsequent interventions.2 Similar finding was obtained in BARI trial, which costs for PTCA increased from 65% of CABG costs to 95% of this latter treatment in 3-year follow-up. The initial difference diminished with time because the greater incidence of additional revascularization procedures in the PTCA treatment.3 The EAST study showed again the initial financial advantage of PTCA was lost in the following up.4
The greatest number of event and revascularization procedures in PTCA group shorted the event-free time in that treatment when compared with Medical and CABG treatment, since the average time to first event was similar in all three treatments.
Ours first year follow-up results confirm the low mortality of stable coronary disease patient with preserved left ventricle function and the superiority of the revascularization treatment by CABG or PTCA for alleviating symptoms, as observed by others.1,6
An interesting finding was the similar rate of revascularization procedures in Medical and PTCA groups, 22 (10.8%) and 25 (12.2%), respectively. The difference was the preferential procedure, new PTCA for PTCA group, while CABG was preferred for Medical treatment failure.
The low incidence of events was responsible by the stability of costs of CABG when compared both to Medical and PTCA treatment strategies.
Study Limitations and Implications
The main limitations of our cost analysis were that we did not take in account the individual duration of hospitalization of each patient, and the cost of outpatient clinical medical appointments because the persistence of symptoms.
The implications are that current PTCA-stent treatment, although being similar to CABG treatment for angina alleviation and incidence of perioperative infarction, presents a higher rate of additional revascularization procedures.
This resulted a smaller cost difference between these procedures at end of the first year of following-up. Those facts should be taken in account to decide the revascularization treatment for symptomatic patients with multivessel disease and preserved left ventricle function. Again, it was confirmed that revascularization procedures are superior to medical treatment for alleviation of symptoms in this subset of patients with preserved left ventricle function.
The effective PTCA costs increased to those of CABG within the first year of follow-up of MASS II study. Medical treatment presented the greatest increase in costs because its lowest ratio of alleviation of angina. The greatest stability of costs was observed in CABG group.
Financial support was provided, in part, by a research grant from the Zerbini Foundation, SP, Brazil.