Abstract
Objective: An economic evaluation was performed, using modelling techniques, to compare 1-year total costs of four revascularisation procedures in patients with multivessel disease: on-pump coronary artery bypass grafting (CABG); off-pump CABG; percutaneous coronary intervention (PCI) with bare-metal stents (BMS); and PCI with drug-eluting stents (DES).
Methods: Clinical data were derived from four randomised clinical trials comparing CABG versus PCI, as well as from literature reviews. Resource use and unit cost estimates were modelled to reflect current Canadian practice.
Results: This study demonstrated that 1 year after the initial revascularisation, PCI with BMS is the least costly procedure, followed by off-pump CABG, PCI with DES and on-pump CABG. DES became the most costly procedure if 3.5 or more DES were used or if staged PCI was performed.
Introduction
Revascularisation techniques for the treatment of multivessel coronary artery disease (MVD) have evolved rapidly over the last decadeCitation1–4. Coronary stents, and more recently drug-eluting stents (DES), now provide a percutaneous option for treating patients with MVD. Simultaneously with the development of percutaneous coronary intervention (PCI), conventional coronary artery bypass grafting (CABG) using cardiopulmonary bypass (on-pump CABG) has evolved with the introduction of the less invasive approach of beating heart surgery (off-pump CABG)Citation5–8. These alternative techniques have led to a drastic change in the approach to coronary artery revascularisation. According to the Canadian Institute for Health Information and Statistics Canada, across Canada there was a downward trend in the total number of CABG surgeries performed, but at the same time a sharp increase in the total number of PCI proceduresCitation9; a similar trend has been observed in other developed countriesCitation10,Citation11.
Healthcare systems are facing tremendous financial pressures to accommodate an increasing demand for PCI as well as the growing utilisation of DESCitation12,Citation13. For example, it has been estimated that implementing exclusive use of DES instead of bare-metal stents (BMS) could cost the Canadian healthcare system an additional $67–127 million dollars annually, even after allowing for savings from a reduction in repeat revascularisation. This additional cost approximated 3–6% of Canada's new healthcare funding in the 2005–2006 fiscal yearCitation12,Citation14–16.
With this major paradigm shift in coronary artery revascularisation, it is important to re-examine the role of each revascularisation strategy and to understand its economic impact. Although several clinical studies have compared CABG versus PCI, most were done in the era of on-pump CABG and BMSCitation17–20. No randomised clinical trial (RCT) has examined all four revascularisation strategies prospectively. Many recent cost-effectiveness studies have compared various revascularisation strategies6,15,17,21–26, but there has been no direct evaluation comparing the costs of all four revascularisation strategies in patients with MVD. The objective of this study was to conduct an economic evaluation to assess and compare, over a 1-year period, the total costs of these four revascularisation strategies in a MVD patient population with stable angina.
Methods
Study design overview
From the perspective of the Canadian healthcare system, a decision model was constructed to estimate and compare up to 1-year total costs and clinical event rates of four revascularisation procedures, using decision analytic software (DATA 3.5; TreeAge Software Inc., Williamstown, MA). Literature reviews were performed and unit costs were estimated to obtain the model input. Sensitivity analyses were conducted to account for uncertainties. The model outcomes were the expected costs, measured as direct medical costs only, and the expected effectiveness, defined as the combined rate of death, myocardial infarction (MI) and stroke. All costs were reported in 2006 Canadian dollars and the consumer price index was applied when necessaryCitation27
Model structure and key assumptions
The decision tree structure () contained four study arms: on-pump CABG; off-pump CABG; PCI with BMS; and PCI with DES. Five mutually exclusive clinical events were modelled in each arm: event free; all-cause mortality; MI; stroke; and repeat revascularisation (re-do CABG or repeat PCI). The time horizon of the model was 1 year, starting from the indexed intervention (initial hospitalisation for revascularisation procedure) and ending with a 1-year follow-up.
Figure 1. Decision tree model.
Five key assumptions were made. First, repeat revascularisation was included in the model as a cost consideration only; major clinical adverse endpoints were death, MI and stroke. Second, it was assumed, based on pooled RCT trial data, that BMS and DES had similar clinical event rates at 1 year of follow-up; however, follow-up repeat revascularisation rates clearly differed between these two groupsCitation28,Citation29. Third, based on pooled RCT data, the number of grafts used was 2.6 per CABG surgery and the number of stents used was 2.8 per PCI procedureCitation17–20. Fourth, the cost formulae for a repeat revascularisation were assumed to be the same as for a typical baseline procedure. If CABG was required following the index CABG or PCI procedure, a multivessel on-pump CABG procedure was assumed. If PCI was required after an index PCI or CABG procedure, a single-vessel PCI with DES was assumed. Fifth, in the Monte Carlo simulation model, uniform distributions were assumed for the uncertain parameters, except for the number of stents where a gamma distribution was assumed.
Model clinical input
All clinical event rates were pooled from contemporary randomised trial data whenever possible. Pooled data (meta-analysis) of RCTs were preferred if available. Four large RCTs, comparing CABG versus PCI with BMS in a MVD patient population (Arterial Revascularization Therapy Study (ARTS), Stent or Surgery (SoS) trial, Angina with Extremely Serious Operative Mortality Evaluation (AWESOME) and Argentine Randomised Study: Coronary Angioplasty with Stenting versus Coronary Bypass Surgery in Patients with Multiple-Vessel Disease (ERACI-II))Citation17, were pooled to give clinical endpoints for the four arms of the model. Since this dataset represented the clinical outcomes of on-pump CABG and BMS, the clinical event rates in the off-pump arm and DES arm were adjusted with relative risk reductions derived from two separate meta-analyses: off-pump versus on-pump CABG; and BMS versus DESCitation28,Citation30. displays the pooled clinical event rates imputed into the model.
Table 1. Decision model clinical input, based on pooled data from four randomised clinical trials and two meta-analyses.
Model cost input
Costs are the product of healthcare resource utilisation and corresponding unit costs. Resource use was categorised into five major cost components: (1) revascularisation procedures (CABG and PCI); (2) procedural devices, including stent type, number of stents, disposable portion of cardiopulmonary bypass for on-pump CABG, and disposable or reusable stabiliser/retractor for off-pump CABG; (3) hospital length of stay (LoS), including intensive care unit, cardiac care unit and cardiac ward; (4) adjunctive pharmacological agents, restricted to glycoprotein 2b/3a inhibitors (Gp2b3a) during PCI and clopidogrel post PCI; and (5) cardiac-related events during the study period. Whenever available, resource datasets were derived from published literature. Resource data that were not available from the literature were based on experts’ opinions and were modelled to reflect current Canadian practice patterns. The later dataset included: percentage use of Gp2b3a inhibitors (41%); duration of post-PCI clopidogrel use (1 month for BMS and 12 months for DES); physician's visits (one pre-procedure specialist visit, post-procedure ward visit, and additional diagnostic catheterisation and cardiologist visit for patients needing repeat revascularisation). presents the resource use information and data sources.
Table 2. Decision model resource input and data sources (base model).
Corresponding unit costs were derived from various sources. Unit cost estimates for CABG/PCI procedures, hospital LoS and complications were derived from the Hamilton Health Sciences Cardiac Case-Costing systemCitation35–37. Procedural device costs (cardiopulmonary bypass, stabiliser/retractor, stents) were based on 2006 average Canadian manufactures’ wholesale price (personal communication with Medtronic Canada and Boston Scientific Canada, June 2006). Costs of pharmacological agents and physician fees were available from the Ontario Drug Benefit Formulary and Ontario Physician Schedule of BenefitsCitation38,Citation39
Sensitivity analyses
Various types of sensitivity analyses were performed to examine the uncertainties associated with the model assumptions and underlying input parametersCitation40. In the base model, 10 cost, resource use and clinical parameters were tested in a traditional one-way sensitivity analysis, where the value of the parameters was altered within a plausible range based on pooled RCTs, recent literature and clinical practice. These parameters included: unit cost of stents (BMS and DES); number of stents per PCI; hospital LoS for all four procedures; percentage Gp2b3a use during PCI procedure; mixed use of abciximab and eptifibatide; and average duration of post-PCI clopidogrel use for BMS. A sensitivity analysis was also performed on the follow-up target vessel revascularisation (TVR) rate to account for uncertainty with the model's assumptionsCitation28,Citation29. Finally, a scenario sensitivity analysis was carried out in the base model. The expected costs were re-computed in a scenario where PCI patients had the procedure done in a two-step fashion (staged PCI)Citation18,Citation41. In addition, a contemporary probabilistic sensitivity analysis using the Monte Carlo simulation technique was conducted. A total of 1,000 replications were performed to explore the joint distribution and simultaneous effects of all the uncertain parameters on each revascularisation strategy.
Results
Base model results
The expected costs and clinical events of the four revascularisation procedures are summarised in . During the initial hospitalisation, BMS was the least costly procedure at $7,938 per patient, followed by $12,614 for off-pump CABG, $12,661 for DES and $14,265 for on-pump CABG. The expected clinical event rate was 3.2% for both PCI procedures and 5.3 and 6.8% for off-pump CABG and on-pump CABG, respectively. One-year after the initial intervention, total expected costs were $10,055 for BMS, $13,395 for off-pump CABG, $13,827 for DES and $15,103 for on-pump CABG. The 1-year expected clinical event rate was 9.8% for both PCI groups and 9.6 and 12.4% for off-pump CABG and on-pump CABG, respectively.
Table 3. Base model results.
Compared with the cost of initial hospitalisation, at the end of 1-year follow-up both PCI procedures had a sizeable cost increase: 27 and 9% for BMS and DES, respectively. The cost increase for both CABG procedures was 6%. Despite a large cost increase for BMS, owing to a higher follow-up TVR rate, BMS remained the least costly procedure at 1 year. Compared with BMS, DES follow-up costs were greatly reduced because of fewer TVR procedures during the follow-up period.
Sensitivity analyses
The results of the sensitivity analyses are displayed in . Overall, the model was not sensitive to all uncertain parameters because BMS remained the least costly procedure throughout the analyses. More specifically, the model 1-year results were not at all sensitive to 7 of the 10 tested parameters. shows that the variation of the cost of BMS ($500–700), hospital LoS for PCI (1–1.4 days) and on-pump CABG (7–10.6 days), percentage use of Gp2b3a during PCI procedure (30–80%), mixed use of abciximab to eptifibatide (ranging from a 100:0 ratio to a 50:50 mixed ratio), duration of clopidogrel use in the BMS arm (1–6 months) and follow-up relative risk reduction for TVR between BMS and DES (30–74%) did not significantly alter the ranking of the expected total 1-year costs for the four procedures. However, the ranking of total costs between DES and the two CABG procedures was affected by the variation of three parameters: unit cost of DES; LoS for off-pump CABG; and the number of stents per PCI procedure. At 1-year, DES would cost less than off-pump CABG if the unit cost of DES was reduced from $2,000 to $1,800, or if the average off-pump CABG LoS increased to >9 days; PCI with DES would become the most costly procedure when the average number of DES stents per PCI procedure exceeded 3.5. Scenario sensitivity analyses suggested that 1-year total costs of BMS and both CABG procedures were not affected by staged PCI; however, if as few as 6% of patients underwent staged PCI with DES, PCI with DES became the most costly procedure. When staged PCI reached 30% of DES patients in the study population, the model predicted that 1 year after the initial intervention DES would cost an extra $3,574, $2,884 and $1,489 per patient compared with BMS, off-pump CABG and on-pump CABG, respectively.
Table 4. Base model sensitivity analyses: value range and results.
The results of the probabilistic sensitivity analysis are presented in . After taking account of joint uncertainties simultaneously, the Monte Carlo model yielded similar results to the base model. One year after the initial intervention, the simulated total cost was $11,177 for BMS, $13,853 for off-pump CABG, $16,796 for DES and $15,297 for on-pump CABG. Simulation models predicted that at 1 year BMS was still the least costly procedure and DES was the most costly procedure in the studied patient population.
Table 5. Monte Carlo simulation model results (probabilistic sensitivity analysis).
Discussion
To the author's knowledge, this is the first study to assess and directly compare the costs of four alternative revascularisation procedures in patients with stable angina and MVD undergoing any one of the four revascularisation options. The focus was on the cost assessment of these four commonly utilised coronary revascularisation techniques, not on their various clinical efficacies, impacts on quality of life or on the patient's preference in their treatment choices. Meta-analyses of RCTs comparing DES versus BMS as well as off-pump CABG versus on-pump CABG have not demonstrated a statistically significant difference in rates of death, MI or stroke during relatively short follow-up periods; in the current literature, there is still a lack of convincing clinical evidence to support fully one procedure over the other.
DES technology has made impressive gains in the market share of stent utilisation, mainly due to a significant decrease in TVR. To date, most published cost-effectiveness analyses comparing alternative revascularisation strategies have included TVR as part of the efficacy outcomes in their studiesCitation14,Citation15,Citation23. However, in this study, efficacy was measured as the combined rate of death, MI and stroke only; TVR was purposely excluded from this composite. Although the authors agree that TVR is a clinical outcome of significant importance, they believe that TVR is not an outcome comparable with death, stroke or MI. Repeat PCI is often a short procedure with minimal correlation with death or permanent damage to end organs. Furthermore, they believe that choosing death, MI and stroke as the efficacy outcome measure created a more level field for comparison of the four revascularisation procedures in patients with MVD. Nevertheless, repeat revascularisation was incorporated into the model cost formula and, therefore, the costs of associated in-hospital outcomes (death, MI, stroke and revascularisation) and follow-up sequelae were captured and attributed to the initial strategy.
The model has demonstrated that total 1-year DES costs were not compensated by a reduced repeat revascularisation rate on account of higher DES unit costs and added costs of adjunctive medications. It could be argued that this base model overestimated the unit costs of DES (average $2,000 based on the range of $1,800–2,200). Estimating unit costs of new technology is challenging and the method is often controversial. Hospital acquisition costs for BMS and DES are often opaque and vary from hospital to hospital based on negotiations over complex purchasing matrices according to the volumes of stents and other products purchased. For instance, the price of BMS could be artificially increased to subsidise the acquisition cost of DES, according to a marketing strategy known as exit tactics. Guiding catheters, balloons, other equipment used in PCI as well as fringe benefits may be packaged into a global purchase agreement. Therefore, the authors believe that it was reasonable and perhaps less biased to use the average Canadian market value in this base model. This estimation was similar to other Canadian cost analysesCitation15,Citation42. The sensitivity analyses showed that when the unit cost of DES was at its lower range ($1,800), DES would cost less than off-pump CABG, but BMS still remained the least costly strategy.
Whilst lower DES unit costs as a result of market competition may favourably impact on the total cost of DES, increased numbers of DES per patient would clearly have an unfavourable effect on the total costs of DES. Factors promoting more DES use would include a more aggressive approach with operators treating lesions that are more diffusely diseased or treating additional side branches. Other economic studies comparing BMS versus DES have suggested that DES were more effective in higher-risk patients such as diabetes patients and patients with small vessels and longer segments of disease. If these were indeed the case, more stents and longer stents per patient would be expected, which would lead to an increase in the total cost of DESCitation23.
Although the estimated total costs for the four revascularisation procedures are similar to other published studiesCitation15,Citation42, this model is very conservative in terms of resource estimates for the PCI procedure, which might underestimate the total costs of PCI procedures despite the use of extensive sensitivity analyses. First, the initial RCTs examining DES versus BMS were performed in relatively low-risk patients, with relatively low rates of TVR even in the BMS ‘control’ populations. Therefore, the model inputs may underestimate the ‘real’ rates of TVR with DES in contemporary practice. Second, in the base model an average of 2.8 stents was assumed both for BMS and DES, whereas DES may lead to a more aggressive PCI approach given the expected lower rate of restenosis and TVR. In the ongoing Future Revascularization in Patients with Diabetes Mellitus: Optimal Management of Multivessel Disease (FREEDOM) and Synergy Between PCI with Taxus and Cardiac Surgery (SYNTAX) trials, enrolling patients with diabetes and multivessel disease or patients undergoing left main procedures with multivessel disease, respectively, the reported number of stents utilised was 4.2 and 4.7, respectivelyCitation41,Citation43. Third, the model did not separately include late stent thrombosis events per se, although any such events would be captured by death, MI or TVR in this model. The rate of late stent thrombosis is low and may differ between DES and BMS; however, this complication frequently leads to a serious adverse event and therefore may be costlyCitation44,Citation45. Fourth, the results were limited by a 1-year time frame owing to data availability; CABG might well prove more effective and less costly than multivessel PCI, but long-term data are lacking. Furthermore, it is unknown whether indefinite use of clopidogrel would increase DES efficacy, mitigate late stent thrombosis and change the cost effectiveness of a DES strategyCitation13. Finally, staged PCI was not included in the base model analysis owing to lack of data. However, several recent published studies have suggested that, in real practice, patients with MVD were routinely scheduled for a second procedure 4–8 weeks after the index procedure for various reasons, including: the risk of contrast nephropathy for patients with renal insufficiency; risk of prolonged fluoroscopic exposure; informed consent for diabetic patients with MVD; patient and physician fatigue; and economic consideration due to reimbursement policyCitation46–49. The scenario sensitivity analysis has indicated that DES would be the most costly procedure when the percentage of staged PCI was as low as 6%.
The optimal timing for PCI has been evaluated in various populations of patients at various degrees of risk for short-term cardiovascular mortality. In the relatively high-risk acute ST-elevation myocardial infarction (STEMI) population, there is a very large reduction in both death and recurrent MI with urgent, timely primary PCICitation50. However, there is a decreased relative mortality benefit with a lengthening time to the initial balloon inflation; the recommended ideal for emergency room door to first balloon inflation is 90 minutes or lessCitation51. Multiple RCTs have also demonstrated a reduction in death and MI rates in the relatively lower risk acute coronary syndrome (ACS) population (unstable angina and non-STEMI) with urgent in-hospital PCI, usually performed within 24–72 hours of admissionCitation52. A large ongoing RCT is currently evaluating the optimal timing for PCI in a typical moderate-risk ACS populationCitation53. In contrast to the PCI population, the optimal timing for CABG surgery after an acute MI remains uncertainCitation54. There are, in fact, no randomised trials examining the optimal timing for CABG following acute MI. Multiple retrospective studies examining this issue have given divergent opinions on the risks associated with early CABG after MICitation54. Data from the National Registry of Myocardial Infarction-2 (NRMI-2) in the US demonstrated a mortality rate of 6.0% (95% confidence interval 5.9–6.1%) amongst 82,055 patients undergoing inpatient CABG following an acute MICitation55. This mortality rate is similar to data from several other smaller retrospective series and is two- to three-fold higher than for a clinically stable outpatientCitation54,Citation55. Logistic regression modelling from the NRMI-2 population identified 14 predictive factors for death; amongst the strongest predictors were Q-wave versus non-Q-wave MI, female sex, age 66–75 years versus <65 years, history of congestive heart failure and Killip classes 2–4. Individually, these risk factors had odds ratios ranging from 1.5 to 7.0. Furthermore, in today's current ACS practice paradigm, the use of Gp2b3a inhibitors and clopidogrel have lead to small but real increases in major bleeding rates. The timing of CABG may be influenced by the use of such medications, plus the perioperative bleeding rates are clearly higher for those patients who have been treated recently with these medications. Periprocedural mortality rates following PCI, in both STEMI and non-STEMI populations, are comparatively lower than those undergoing urgent CABG. Enthusiasm for urgent catheter-based interventions should not necessarily be extended to urgent CABG, for the safety and efficacy have not been fully established. Thus, the results should be best applied to a stable population with multivessel disease.
Study limitations
The present study has certain limitations inherent to all economic evaluations using a modelling approach. First, a common and major limitation of modelling is the quality and reliability of the model input data. Owing to lack of direct head-to-head RCTs, the model input was synthesised from various sources that may introduce uncertainty and potential biases. Although this method of restricting model input data to RCTs as much as possible may minimise certain biases, it may introduce other concerns. For example, the patients from the CABG versus BMS studies may not be identical in their risk for periprocedural or follow-up adverse clinical events. Differences in healthcare systems may affect the average resource use rates, namely adjunctive drugs during PCI as well as LoS after CABG. In addition, patients that fit strict inclusion and exclusion criteria of RCTs may well differ from patients in the ‘real world’ who may be at higher overall risk, more complicated and have more co-morbidity. Such patients would not only affect the rates of TVR in the BMS/DES groups, but could also impact the rates of stroke and hospital LoS in the surgical groups.
The second limitation associated with this modelling approach is the set of assumptions and their underlying uncertainties. Following the guidelines of economic evaluations, efforts were made to make sure that model structure, assumptions and analysis process were transparent and reproducibleCitation40. Extensive sensitivity analyses were performed and the results suggested that this model is fairly robust.
A third limitation is the relatively short 1-year follow-up period. It is not clear whether the major clinical adverse event rates would diverge further over time and favour one strategy over another. Clearly, as clinical events accrue disproportionately, cost effectiveness will change. For these reasons, this model may underestimate the total costs of both stents and possibly even the cost of CABG strategiesCitation56,Citation57
Conclusions
This model predicted that 1 year after the initial intervention in a stable patient population, BMS is the least costly procedure despite a higher rate of stent restenosis requiring repeat revascularisation in a population of stable angina patients with MVD, who are amendable by any one of four alternative revascularisation options. Off-pump CABG is the second least costly procedure. On-pump CABG and DES had similar 1-year total costs, but expected total DES costs were sensitive to certain parameters and clinical scenarios. DES became the most costly procedure if an average of 3.5 stents or more were used or if staged PCI was performed. Further analysis may be warranted should better trial data or longer-term data become available, and the results are best applied to a stable outpatient population.
Acknowledgement
Declaration of interest: The authors have declared no conflict of interest and have received no payment in the preparation of this manuscript. The authors would like to thank Mr Barry Kent for his technical assistance.
Notes
References
- Holmes DR, Jr, Firth BG, Wood DL. Paradigm shifts in cardiovascular medicine. Journal of the American College of Cardiology 2004; 43: 507–512
- Rose EA. Off-pump coronary-artery bypass surgery. New England Journal of Medicine 2003; 348: 379–380
- Yang YM, Moussa I. Percutaneous coronary intervention and drug-eluting stents. Canadian Medical Association Journal 2005; 172: 323–325.
- Serruys PW, Kutryk MJ, Ong AT. Coronary-artery stents. New England Journal of Medicine 2006; 354: 483–495
- Angelini GD, Taylor FC, Reeves BC, et al. Early and midterm outcome after off-pump and on-pump surgery in Beating Heart Against Cardioplegic Arrest Studies (BHACAS 1 and 2): a pooled analysis of two randomised controlled trials. Lancet 2002; 359: 1194–1199
- Nathoe HM, Van Dijk D, Jansen EW, et al. A comparison of on-pump and off-pump coronary bypass surgery in low-risk patients. New England Journal of Medicine 2003; 348: 394–402
- Moses JW, Leon MB, Popma JJ, et al. Sirolimus-eluting stents versus standard stents in patients with stenosis in a native coronary artery. New England Journal of Medicine 2003; 349: 1315–1323
- Stone GW, Ellis SG, Cox DA, et al. A polymer-based, paclitaxel-eluting stent in patients with coronary artery disease. New England Journal of Medicine 2004; 350: 221–231
- Sibbald B. Declining CABG rate means fewer jobs for surgeons. Canadian Medical Association Journal 2005; 173: 583–584.
- Ulrich MR, Brock DM, Ziskind AA. Analysis of trends in coronary artery bypass grafting and percutaneous coronary intervention rates in Washington state from 1987 to 2001. American Journal of Cardiology 2003; 92: 836–839
- Poyen V, Silvestri M, Labrunie P, et al. Indications of coronary angioplasty and stenting in 2003: what is left to surgery?. Journal of Cardiovascular Surgery 2003; 44: 307–312
- Brophy JM. The dollars and sense of drug-eluting stents. Canadian Medical Association Journal 2005; 172: 361–362
- Ong AT, Serruys PW. Drug-eluting stents: current issues. Texas Heart Institute Journal 2005; 32: 372–377
- Shrive FM, Ghali WA, Lewis S, et al. Moving beyond the cost per quality-adjusted life year: modelling the budgetary impact and clinical outcomes associated with the use of sirolimus-eluting stents. Canadian Journal of Cardiology 2005; 21: 783–787
- Mittmann N, Brown A, Seung SJ, et al. Economic Evaluation of Drug Eluting Stents., Technology Report 53. Ottawa, Canada: Canadian Coordinating Office for Health Technology Assessment, 2005.
- Budget 2004: Health Care, Learning and Communities. Canada: Department of Finance, 2006
- Serruys PW, Unger F, Sousa JE, et al. Comparison of coronary-artery bypass surgery and stenting for the treatment of multivessel disease. New England Journal of Medicine 2001; 344: 1117–1124
- SoS Investigators. Coronary artery bypass surgery versus percutaneous coronary intervention with stent implantation in patients with multivessel coronary artery disease (the Stent or Surgery trial): a randomised controlled trial. Lancet 2002; 360: 965–970
- Rodriguez A, Bernardi V, Navia J, et al. Argentine randomized study: coronary angioplasty with stenting versus coronary bypass surgery in patients with multiple-vessel disease (ERACI II): 30-day and one-year follow-up results. ERACI II Investigators. Journal of the American College of Cardiology 2001; 37: 51–58
- Morrison DA, Sethi G, Sacks J, et al. Percutaneous coronary intervention versus coronary artery bypass graft surgery for patients with medically refractory myocardial ischemia and risk factors for adverse outcomes with bypass: a multicenter, randomized trial. Investigators of the Department of Veterans Affairs Cooperative Study #385, the Angina With Extremely Serious Operative Mortality Evaluation (AWESOME). Journal of the American College of Cardiology 2001; 38: 143–149
- Ascione R, Lloyd CT, Underwood MJ, et al. Economic outcome of off-pump coronary artery bypass surgery: a prospective randomized study. Annals of Thoracic Surgery 1999; 68: 2237–2242
- Puskas JD, Williams WH, Mahoney EM, et al. Off-pump vs conventional coronary artery bypass grafting: early and 1-year graft patency, cost, and quality-of-life outcomes: a randomized trial. Journal of the American Medical Association 2004; 291: 1841–1849
- Cohen DJ, Bakhai A, Shi C, et al. Cost-effectiveness of sirolimus-eluting stents for treatment of complex coronary stenoses: results from the Sirolimus-Eluting Balloon Expandable Stent in the Treatment of Patients With De Novo Native Coronary Artery Lesions (SIRIUS) Trial. Circulation 2004; 110: 508–514
- Weintraub WS, Mahoney EM, Zhang Z, et al. One year comparison of costs of coronary surgery versus percutaneous coronary intervention in the stent or surgery trial. Heart 2004; 90: 782–788
- Eefting F, Nathoe H, Van Dijk D, et al. Randomized comparison between stenting and off-pump bypass surgery in patients referred for angioplasty. Circulation 2003; 108: 2870–2876
- Stroupe KT, Morrison DA, Hlatky MA, et al. Cost-effectiveness of coronary artery bypass grafts versus percutaneous coronary intervention for revascularization of high-risk patients. Circulation 2006; 114: 1251–1257
- Statistics Canada. Consumer Price Index. CANSIM Catalogue No. 62-001-XPE, Vol. 86, No. 8. 2007.
- Babapulle MN, Joseph L, Belisle P, et al. A hierarchical Bayesian meta-analysis of randomised clinical trials of drug-eluting stents. Lancet 2004; 364: 583–591.
- Tu JV, Bowen J, Chiu M, et al. Effectiveness and safety of drug-eluting stents in Ontario. New England Journal of Medicine 2007; 357: 1393–1402
- Cheng DC, Bainbridge D, Martin JE, et al. Does off-pump coronary artery bypass reduce mortality, morbidity, and resource utilization when compared with conventional coronary artery bypass? A meta-analysis of randomized trials. Anesthesiology 2005; 102: 188–203
- Serruys PW, Ong AT, Morice MC, et al. Arterial revascularization therapies study party II—sirolimus-eluting stents for the treatment of patients with multivessel de novo coronary artery lesions. EuroInterv 2005; 1: 147–156.
- Bakhai A, Stone GW, Grines CL, et al. Cost-effectiveness of coronary stenting and abciximab for patients with acute myocardial infarction: results from the CADILLAC (Controlled Abciximab and Device Investigation to Lower Late Angioplasty Complications) trial. Circulation 2003; 108: 2857–2863
- Cowper PA, Udayakumar K, Sketch MH, Jr, et al. Economic effects of prolonged clopidogrel therapy after percutaneous coronary intervention. Journal of the American College of Cardiology 2005; 45: 369–376
- Beinart SC, Kolm P, Veledar E, et al. Long-term cost effectiveness of early and sustained dual oral antiplatelet therapy with clopidogrel given for up to one year after percutaneous coronary intervention: results from the Clopidogrel for the Reduction of Events During Observation (CREDO) trial. Journal of the American College of Cardiology 2005; 46: 761–769
- Lamy A, Yusuf S, Pogue J, et al. Cost implications of the use of ramipril in high-risk patients based on the Heart Outcomes Prevention Evaluation (HOPE) study. Circulation 2003; 107: 960–965
- Lamy A, Wang X, Kent R, et al. Economic evaluation of the MEDENOX trial: a Canadian perspective. Medical Patients with Enoxaparin. Canadian Respiratory Journal 2002; 9: 169–177.
- Smith KM, Lamy A, Arthur HM, et al. Outcomes and costs of coronary artery bypass grafting: comparison between octogenarians and septuagenarians at a tertiary care centre. Canadian Medical Association Journal 2001; 165: 759–764
- Ontario Ministry of Health and Long-Term Care. Ontario Drug Benefit Formulary. No. 38. Toronto, Canada: Ministry of Health, 2003.
- Ontario Ministry of Health and Long-Term Care. Schedule of Benefits: physician Services under the Health Insurance Act. Toronto, Canada: The Ministry, 2003.
- Gold MR, Siegel JE, Russell LB, et al. Cost-Effectiveness in Health and Medicine. Oxford University Press, New York, NY 1996
- Kini AS. Revascularization in diabetics: FREEDOM Trial Update. 2007.
- Shrive FM, Manns BJ, Galbraith PD, et al. Economic evaluation of sirolimus-eluting stents. Canadian Medical Association Journal 2005; 172: 345–351.
- Dawkins KD. Multivessel DES or CABG SYNTAX Trial. 2007.
- Kaul S, Diamond GA. An ounce of prevention for a pound of flesh?. CardioSource 11 October, 2006
- Eisenstein EL, Anstrom KJ, Kong DF, et al. Clopidogrel use and long-term clinical outcomes after drug-eluting stent implantation. Journal of the American Medical Association 2007; 297: 159–168
- Griffiths H, Bakhai A, West D, et al. Feasibility and cost of treatment with drug eluting stents of surgical candidates with multi-vessel coronary disease. European Journal of Cardiothoracic Surgery 2004; 26: 528–534
- Ong AT, van der Giessen WJ. Drug-eluting stents for interventional revascularization of coronary multivessel disease. Journal of Interventional Cardiology 2005; 18: 447–453
- Kahn JK, O'Neill WW. Two vessels or not two vessels? That is the question. American Heart Journal 2002; 143: 948–949
- Nikolsky E, Halabi M, Roguin A, et al. Staged versus one-step approach for multivessel percutaneous coronary interventions. American Heart Journal 2002; 143: 1017–1026
- Keeley EC, Boura JA, Grines CL. Comparison of primary and facilitated percutaneous coronary interventions for ST-elevation myocardial infarction: quantitative review of randomised trials. Lancet 2006; 367: 579–588
- Antman EM, Hand M, Armstrong PW. 2007 focused update of the ACC/AHA 2004 guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Journal of the American College of Cardiology 2008; 51: 210–247
- Mehta SR, Cannon CP, Fox KA, et al. Routine vs selective invasive strategies in patients with acute coronary syndromes: a collaborative meta-analysis of randomized trials. Journal of the American Medical Association. 2005; 293: 2908–2917
- Metra M, et al. Early vs. delayed timing of intervention in patients with acute coronary syndromes (TIMACS), ClinicalTrials.govNCT00552513.
- Boden WE. Is it time to reassess the optimal timing of coronary artery bypass graft surgery following acute myocardial infarction?. American Journal of Cardiology 2002; 90: 35–38
- Zaroff JG, di Tommaso DG, Barron HV. A risk model derived from the National Registry of Myocardial Infarction 2 database for predicting mortality after coronary artery bypass grafting during acute myocardial infarction. American Journal of Cardiology 2002; 90: 1–4
- Hannan EL, Racz MJ, Walford G, et al. Long-term outcomes of coronary-artery bypass grafting versus stent implantation. New England Journal of Medicine 2005; 352: 2174–2183
- Malenka DJ, Leavitt BJ, Hearne MJ, et al. Comparing long-term survival of patients with multivessel coronary disease after CABG or PCI: analysis of BARI-like patients in northern New England. Circulation 2005; 112((9 Suppl):)I371–I376