Journal of the American College of Cardiology Vol. 45, No. 4, by the American College of Cardiology Foundation ISSN /05/$30.

Transcription

1 Journal of the American College of Cardiology Vol. 45, No. 4, by the American College of Cardiology Foundation ISSN /05/$30.00 Published by Elsevier Inc. doi: /j.jacc Primary Angioplasty Is Cost-Minimizing Compared With Pre-Hospital Thrombolysis for Patients Within 60 Min of a Percutaneous Coronary Intervention Center The Comparison of Angioplasty and Pre-Hospital Thrombolysis in Acute Myocardial Infarction (CAPTIM) Cost-Efficacy Sub-Study Jacques Machecourt, MD,* Eric Bonnefoy, MD, Gérald Vanzetto, MD,* Pascal Motreff, MD, Stéphanie Marlière, MD,* Alain Leizorovicz, MD, PHD, Benoit Allenet, PHD, Jean Michel Lacroute, MD, Jean Cassagnes, MD, Paul Touboul, MD Grenoble, Lyon, and Clermont Ferrand, France OBJECTIVES BACKGROUND METHODS RESULTS CONCLUSIONS This ancillary study of the Comparison of Angioplasty and Pre-hospital Thrombolysis in Acute Myocardial Infarction (CAPTIM) trial sought to assess the cost-efficacy ratio of primary coronary angioplasty (PCA) and pre-hospital thrombolysis (PHT) in patients suffering from an acute myocardial infarction (AMI) ( 6 h) close to ( 60 min journey) a percutaneous coronary intervention (PCI) center. In the CAPTIM study, at 30 days follow-up PCA was as equally effective as PHT with rescue angioplasty if needed. The cost efficacy of these two strategies has not yet been compared. Data were prospectively collected for 299 patients in three centers. The efficacy analysis was extended at one-year follow-up for those patients. Direct fixed and variable actual costs were assessed with a piggyback data collection. The one-year primary end point event-rate (death, non-fatal myocardial infarction, and stroke) was not different after PCA or PHT (14% vs %, p NS). Costs were lower in the PCA group either during the in-hospital period (8,287 vs. 9,170 $, p ) and after one-year follow-up, in relation to a higher rate of subsequent revascularizations in the PHT group (49% vs. 23%, p 0. 01), leading to a longer hospital stay (10 vs. 9.1 days, p 0. 03). After AMI in patients less than 1 h from a PCI center, PCA is as effective and less costly than a combined strategy of PHT followed by rescue angioplasty. (J Am Coll Cardiol 2005;45: ) 2005 by the American College of Cardiology Foundation The aim of reperfusion therapy in acute myocardial infarction (AMI) is to achieve complete and sustained patency of the infarct-related coronary artery as soon as possible. When patients are located more than a 60- to 90-min journey from a catheterization facility, thrombolysis is recommended as the first-line approach (1,2). In comparison to in-hospital administration, pre-hospital thrombolysis (PHT) is particularly useful in such patients as it is associated with an average gain of 60 min or more in some situations, resulting in a 17% reduction in death rate (3,4). For patients within a 90-min journey of a percutaneous coronary intervention (PCI) center, data are more controversial: primary coronary angioplasty (PCA) rather than in-hospital thrombolysis is often advocated as a higher reperfusion rate is obtained with angioplasty without an excessive delay in reperfusion. Consequently the mortality rate is lower (5) even in patients who From *Coronary Care Unit and Pharmacy Department, CHU Grenoble, Grenoble, France; Coronary Care Unit, Hopital Louis Pradel, Lyon, France; Coronary Care Unit, Centre Hospitalo Universitaire Clermont Ferrand, Clermont Ferrand, France; Clinical Pharmacology Unit, Laennec University, Lyon, France; and Service d Aide Médicale Urgente (SAMU) 38, Grenoble, France. The CAPTIM trial was supported by public funding (Programme Hospitalier de Recherche Clinique, from the French Ministry of Health). Manuscript received May 10, 2004; revised manuscript received October 30, 2004, accepted November 2, have to be transferred from a community hospital to a PCI center (6 8). However, when PHT is available, the choice of optimal reperfusion therapy remains more difficult, as a faster reperfusion obtained by this technique is being balanced against a higher reperfusion rate obtained with primary angioplasty. The main result of the recent Comparison of Angioplasty and Pre-hospital Thrombolysis in Acute Myocardial Infarction (CAPTIM) study comparing PCA and PHT with rescue angioplasty if needed is that both strategies lead to equivalent results in terms of major events at 30-day follow-up (9). Little data is available assessing the cost-efficacy of primary angioplasty versus in-hospital thrombolysis (10 13), and no study to our knowledge has compared PHT with primary angioplasty in patients presenting with AMI. The cost-efficacy analysis of the CAPTIM study was planned to prospectively collect data concerning efficacy and costs during the initial hospitalization period, and at one-year follow-up in every patient included by three university hospitals participating in the CAPTIM study. This ancillary study aimed to determine whether one of these two reperfusion strategies is dominant (more effective and cost-saving), cost-effective (more effective with a moderate incremental cost), or

2 516 Machecourt et al. JACC Vol. 45, No. 4, 2005 The CAPTIM Cost-Efficacy Sub-Study February 15, 2005: Abbreviations and Acronyms AMI acute myocardial infarction CAPTIM Comparison of Angioplasty and Pre-hospital Thrombolysis in Acute Myocardial Infarction PCA primary coronary angioplasty PCI percutaneous coronary intervention PHT pre-hospital thrombolysis PTCA percutaneous transluminal coronary angioplasty TIMI Thrombolysis In Myocardial Infarction cost-minimizing (both strategies are equally effective but one is cost-saving). METHODS Study design. The methodology of the CAPTIM study has been previously described (9). Patients within 6 h from onset of a ST-segment elevation AMI were randomly assigned at the site of initial management by the mobile emergency care unit (Service d Aide Médicale Urgente [SAMU]) to PHT with intravenous infusion of alteplase or PCA. Patients were excluded if the duration of transfer to the PCI center was expected to exceed 1 h, that is, if first care-to-balloon time was more than 90 min. Patients with cardiogenic shock, or any other contraindication to thrombolysis at presentation were also excluded. For the cost-efficacy sub-study, 299 patients from the three participating centers were consecutively enrolled between June 1997 and September 2000; of these, 289 completed follow-up. Ten patients were lost to follow-up after the initial hospitalization period, as they had moved to another area (six in the PCA group, five in the PHT group). None had presented with a Killip class 2 or experienced an end point during the initial hospitalization. Mean age and duration of hospitalization were not different from the whole group. Baseline characteristics and clinical presentation of the patients included in the cost-efficacy analysis were not different from those of the whole CAPTIM population. There was no difference between patients assigned to PHT and those assigned to primary angioplasty (Table 1). However, patients in the cost-efficacy study were non-significantly older (59.5 vs. 58 years) and presented more often with diabetes mellitus (14.3% vs. 12.6%, p 0.05). Consistent with the results of the main study, cardiogenic shock occurring between first care and admission to the PCI center or to the intensive care unit was more often present in the primary angioplasty arm. The ethics committee of one of the three hospitals approved this study, and all eligible patients provided a written informed consent. The analysis was performed from a societal perspective with a time horizon of one year. The primary clinical end point for efficacy as defined in the CAPTIM study was a composite of death, non-fatal re-infarction, and non-fatal disabling stroke within 30 days and was extended to one-year follow-up. The secondary clinical end point extended the efficacy definition to end points that included emergent or delayed revascularization (angioplasty or coronary artery bypass graft surgery) and frequency of severe bleeding. All events were assessed up to 120 min after hospital admission, at discharge, and after one-year followup. Immediate angioplasty was defined as any revascularization performed at the patient s admission (primary angioplasty for patients included in the PCA group and rescue angioplasty for patients included in the PHT group); rescue angioplasty was performed if suitable for any patients with no clear clinical evidence of reperfusion (persisting chest pain and/or significant ST-segment elevation) 60 to 120 min after the bolus injection of alteplase. Emergent angioplasty was defined as any revascularization (except primary Table 1. Baseline Characteristics Cost-Efficacy Analysis Whole CAPTIM Study Parameters PCA (n 149) PHT (n 150) PCA (n 421) PHT (n 419) Age (yrs) Male 115 (77.1) 125 (83.3) 343 (81.5) 345 (82.5) Current smokers 62 (58.4) 68 (54.7) 205 (49.2) 216 (52.6) Diabetes 25 (16.7) 18 (12) 57 (13.5) 46 (11.1) Dyslipidemia 88 (59.1) 90 (60) 215 (51.4) 212 (51.1) Previous MI 23 (15.4) 32 (21.3) 51 (12) 56 (13) Anterior infarction 79 (46.9) 81 (46) 178 (42.7) 166 (40.2) Heart rate at admission 75 [64 89] 78 [69 88] 75 [66 88] 75 [64 84] (beats/min)* Systolic blood (mm Hg)* 130 [ ] 130 [ ] 128 [ ] 125 [ ] TIMI risk score* 2 [1 4] 2 [1 4] 2 [1 4] 2 [1 4] Killip 4 at admission 3 (2.1) 0 9 (2.1) 0 Killip 4 in-hospital 8 (5.6) 4 (2.7) 20 (4.9) 10 (2.5) Time to treatment (min) 184 [ ] 123 [87 176] 190 [ ] 130 [95 180] Results are number (%) except as otherwise indicated. *Median [interquartile range]. Student test p 0.07 versus PCA in the cost-efficacy analysis. Student test p and student test p 0.09 versus PCA in the whole CAPTIM study. CAPTIM Comparison of Angioplasty and Pre-hospital Thrombolysis in Acute Myocardial Infarction; MI myocardial infarction; PCA primary coronary angioplasty; PHT pre-hospital thrombolysis; TIMI Thrombolysis In Myocardial Infarction.

3 JACC Vol. 45, No. 4, 2005 February 15, 2005: Machecourt et al. The CAPTIM Cost-Efficacy Sub-Study 517 or rescue angioplasty) performed during the initial hospitalization period. Emergent angioplasty was driven by the occurrence of ischemia (chest pain or electrocardiographic changes) or performed at the discretion of the investigator (elective angioplasty). Elective angioplasty was encouraged when the PCI of the infarct-related artery was Thrombolysis In Myocardial Infarction (TIMI) flow grade 0, 1, or 2, or when ischemia has been shown on non-invasive tests but angioplasty of a non-infarct-related artery was not recommended. Delayed revascularization was defined as the need for revascularization owing to recurrent ischemia from the end of the initial hospitalization period up to one year. Recurrent infarction and post-angioplasty infarction were defined as recurrent chest pain with a new ST-segment elevation and an associated increase in creatine kinase and creatine kinase-mb fraction. Creatine kinase, creatine kinase-mb fraction, and troponins were measured every 6 h the first day and at least once a day during the following days, and routinely 6 to 12 h after an angioplasty. Computed tomographic or magnetic resonance imaging brain scans were requested for all patients suspected of stroke. Severe bleeding was defined as any intracranial hemorrhage or any bleeding that required blood transfusion. A clinical events committee adjudicated all events. Sources of cost data. Resource use was identified using a piggyback data collection. Only direct costs (fixed and variable) were considered. Initial hospitalization and oneyear costs were assessed. An ancillary micro-costing study was developed in two of the three hospitals (211 of 299 patients); the unit cost of each procedure was extended to the third hospital. The current cost estimation was performed in the clinical pharmacy unit of the Grenoble University hospital. Valuation of resources for services and procedures was performed from actual costs derived from the accounting data system of the hospital, and for drugs and devices on market prices. All costs are expressed in U.S. dollars using the 2000 exchange rate for the French franc. A 4% discounting rate was applied. Costs of pre-hospital care by the mobile emergency-care unit were calculated according to the duration of the intervention and the duration of helicopter flight if needed. Costs of hospitalization were assessed in the cardiology intensive care unit, in the general ward, and in the surgical intensive care unit. The mean cost for one day spent in these different units was obtained by adding fixed costs (facilities, equipment, maintenance) variable costs (generic caregiver time, catering, supplies except drugs, laboratory tests, and interventions). Labor costs for physicians, nurses, or technicians were derived from overall number of hours worked per year during normal working hours, weekends, or nights multiplied by the hourly rate of salary divided by the number of patient-days that year. Concerning revascularization procedures, we first evaluated the basal cost of coronary angiography and percutaneous transluminal coronary angioplasty (PTCA) (which included physicians and nurses charges, the cost of disposables, the cost of contrast agent, and the depreciation of the catheterization laboratory equipment), and subsequently, the cost of single-use treatments (stents, balloons, glycoprotein IIb/IIIa receptor antagonists, intra-aortic balloon pump) needed for each patient was added. The cost of coronary bypass was calculated in the same way. Concerning drugs, the unit price for the thrombolytic agent was added to the price of the other cardiac (aspirin, clopidogrel, heparin, beta-blockers, angiotensin-converting enzymes) or non-cardiac (statins, anti-diabetic drugs) treatments. We also estimated a median cost per day for all medical examinations, including standard biologic tests, electrocardiography, and pulmonary radiography. The number of echocardiographies, exercise stress tests, and nuclear cardiology studies were assessed for each patient during the in-hospital period and then up to one year. To determine whether our results can be extended to other health care systems, a sensitivity analysis was performed: U.S. and U.K. unit costs were estimated by applying U.S. and U.K. unit costs to resources used for our patients. For the U.K., National Health and social care reference costs were taken. Hospitalization and revascularization procedure costs (PCI, coronary artery bypass graft surgery) were assigned on the basis of previous studies (14). The costs of drugs were obtained from the British National Formulary. Accordingly, for the U.S. costs, hospitalization costs in the coronary care unit, normal ward, revascularization procedure costs, and drug costs were derived from previous studies (15 17). In these studies involving a large number of centers, the charges were obtained from the UB92 Medicare Uniform Bill, and reduced to costs using the Medicare cost to charge ratio of these centers. Physician fees were assessed using the Medicare Fee Schedule (North Carolina version), taking into account the daily follow-up of the patients, cardiac catheterization, and revascularizations of the patients. Patient rehabilitation and pre-hospital transportation costs were assessed using the Medicare reference coverage. Statistical analysis. All analyses were performed on the basis of the intention-to-treat principle. For the CAPTIM study, a sample of 1,200 patients was chosen to ensure the detection of an absolute reduction of 5% of the primary end point in one group; for the cost-efficacy study, 300 patients were chosen to enable the detection of a 10% absolute reduction of the in-hospital total cost between the groups with an alpha error of 0.05 and a beta error of As finally only 289 patients have been studied, a post hoc analysis of the statistical power of the study was performed: with the same hypotheses for standard deviation, a difference of 10.25% for the in-hospital cost between the two groups was needed in order to be detected. Discrete data are reported as number (frequencies) and continuous data as mean SD. Costs are reported as mean SD and [median] values. Normally distributed continuous variables (as assessed by graphic comparison to the theoretical normal function of same mean and standard

4 518 Machecourt et al. JACC Vol. 45, No. 4, 2005 The CAPTIM Cost-Efficacy Sub-Study February 15, 2005: Table 2. Hospital Stay and Frequency of Cardiac Revascularizations and Testing Initial Hospitalization One-Year Follow-Up PCA (n 143) PHT (n 146) p Value PCA (n 143) PHT (n 146) p Value* Hospitalizations ICCU (days) 3.4 [3] 3.7 [3.5] [3] 4 [3.5] 0.03 Medical ward (days) 5.7 [5] 6.3 [6] [5] 9.2 [6] 0.03 Other ICU (days) 0.4 [0] 0.5 [0] NS 0.6 [0] 0.6 [0] NS Total in-hospital stay (days) 9.5 [9] 10.5 [10] [9] 13.8 [10] 0.01 Cumulative re-hospitalization (days) 2.9 [0] 3.3 [0] NS Rehabilitation (days) 7.9 [0] 9.9 [0] NS Revascularizations procedures (number of patients) Coronary angiographies 140 (98) 121 (83) (98) 125 (86) Immediate PCI (primary or rescue) 126 (88) 51 (35) Emergent PCI 14 (10) 61 (42) PCI total 129 (90) 90 (62) (90) 99 (68) Stent 103 (72) 74 (51) (72) 80 (55) GP IIb/IIIa inhibitors 43 (30) 5 (3.4) (30) 5 (3.4) Thrombolysis 7 (5) 146 (100) (5) 146 (100) Intra-aortic balloon pump 11 (8) 6 (4) (8) 6 (4) 0.19 CABG 1 (0.7) 0 NS 8 (5.6) 10 (6.4) NS Testing Exercise stress tests 77 (54) 67 (46) (76) 106 (73) NS Nuclear cardiology studies 34 (24) 45 (31) (40) 65 (45) NS Echocardiography 110 (77) 118 (81) NS 119 (83) 125 (86) NS Data are numbers (%) except as otherwise indicated. Days of hospitalization are mean [median]. *Mann-Whitney p values. Chi-square p values. CABG coronary artery bypass graft surgery; GP glycoprotein; ICCU intensive coronary care unit; ICU intensive care unit; PCA primary coronary angioplasty; PCI percutaneous coronary intervention; PHT pre-hospital thrombolysis. deviation) were compared by Student t test and discrete data by a chi-square test. Costs and non-normally distributed data were compared using a Mann-Whitney nonparametric test. Analyses were performed in some subgroups of patients, related to the initial clinical presentation (age, gender, Killip class, TIMI risk score, location of the AMI), the occurrence of the primary end point, and the need of an immediate or an emergent angioplasty. The statistical power for these subgroup analyses was calculated taking into account the number of comparisons performed. The cost-efficacy ratio of the two strategies was assessed; the cost difference between PHT and PCA was compared with the difference in efficacy (in terms of primary and secondary clinical end points as previously defined). RESULTS Primary and secondary clinical end points. The occurrence of a 30-day primary clinical end point was not different for PCA and PHT (7.7% vs. 12.3%, p NS). There was also no difference after one-year follow-up (14% vs. 16.4%, p NS). The trend towards a better outcome for the primary angioplasty group was driven by a decrease in the 30-day re-infarction rate (1.4% vs. 5.5% for PHT, p 0.11). During the in-hospital period, major bleeding occurred in five patients in the PCA group (groin hematoma with significant deglobulinization) versus none in the PHT group. The secondary clinical end point (primary clinical end point major bleeding revascularizations) after one year was significantly lower in the PCA group than in the PHT group (34% vs. 61%, p ). Mechanical revascularization (Table 2). During the inhospital period, in the PCA group all patients except three underwent an immediate coronary angiography as planned, followed by immediate PCI in 88% of cases. Pre-procedure, 13% of these patients were TIMI flow grade 3 and 5% had non-significant ( 30%) stenosis. Post-procedure, 84% of patients were TIMI flow grade 3, 11% TIMI flow grade 2, and 5% TIMI flow grade 0 to 1. At the end of the initial hospitalization period as well as after one-year follow-up, the revascularization rate did not change significantly, 90% of patients being revascularized (some patients had two revascularizations or more). In the PHT group, 83% of the patients required a coronary angiography during the initial hospitalization period leading to 35% of patients with a rescue angioplasty, 62% of patients with a rescue or an emergent angioplasty, and 68% of patients revascularized after one year. Emergent angioplasty was driven by recurrence of ischemia for 12% of patients and was elective in 18% of patients. Consequently, fewer patients underwent a PCI after PHT than after PCA, but more patients underwent an emergent or delayed PCI (49% vs. 23%, p 0.001). Revascularization by bypass surgery was performed in 5.6% in the PCA group versus 6.4% in the PHT group (p NS) mostly between day 30 and one year.

5 JACC Vol. 45, No. 4, 2005 February 15, 2005: Machecourt et al. The CAPTIM Cost-Efficacy Sub-Study 519 Costs analysis. Reference costs of all resources used are shown in Table 3, left column. Mean costs of initial hospitalization were $883 significantly lower in the PCA group than in the PHT group ( 11%, p 0.008). The $372 significantly higher cost of hospitalization in the PHT group was the result of a longer hospital stay, either in the coronary care unit or in the general ward (Table 2). The number of re-hospitalization days was not different in the two groups. For the revascularization procedures, as expected, catheterization and angioplasty were more costly in the PCA group compared with the PHT group. The mean total cost of angioplasty was $1,997 in the PCA group (including an average cost of $300 per patient for glycoprotein IIb/IIIa antagonist used in one-third of patients before or during angioplasty) versus $1,243 in the PHT group. On the other hand, the additional cost of the thrombolytic agent in the PHT group more than offset the higher cost of PCI in the PCA group, and revascularization was $283 significantly more costly in the PHT group than in the PCA group. Mobile care unit pre-hospital transportation was also more costly for the PHT patients because of the preparation time of alteplase. At one-year follow-up, costs remained $1,224 significantly lower in the PCA group than in the PHT group (p 0.04) with no further significant difference between the end of the initial hospitalization period (accounting for about 70% of the overall cost) and one year. Rehospitalization rate, costs of drugs, and fees for exercise stress tests, echocardiography, or nuclear studies were not different between the two groups (Tables 2 and 3). Sensitivity analysis. Applying U.S. and U.K. costs to our patients (Table 4), total average costs at the end of the initial hospitalization were, respectively, 1.9 and 1.6 higher. Prehospital care was the only cost remaining higher in our study. Revascularization procedures (PTCA, coronary artery bypass graft surgery, and thrombolysis) were twice as expensive as in the U.S. or U.K. system. After one-year follow-up, figures were not different. Concerning the cost comparison of the two groups of patients, when applying U.K. and U.S. costs, PHT remained more expensive than PCA, whether at the end of the initial hospitalization ( 7.5% with U.K. costs, 9% with U.S. costs vs. 11% in our study) or at one-year follow-up ( 9% with U.K. costs, 10.5% with U.S. costs vs. 10% in our study). Subsets analysis for cost comparison (Fig. 1). Results were not different whether based on gender, age, or location of the AMI, PCA being less expensive than PHT. However, they were influenced by the presence of clinical risk factors at presentation (the Killip class, heart rate, systolic blood pressure, alone or combined). Primary coronary angioplasty was less costly than PHT only for non-high-risk patients and there was no difference in cost between the two strategies for high-risk patients, these patients being significantly more expensive than nonhigh risk patients whatever the treatment used. Similarly, PCA was less costly than PHT for patients free of a primary clinical end point, whereas no difference was seen in patients who experienced at least one event, and PCA was less costly than PHT whether or not an immediate or emergent PCI was performed. However, the difference in favor of PCA was particularly large for patients who did not need an immediate PCI ($2,000 saved) as the initial coronary angiography selected low-risk patients, with an early discharge of these patients. The PHT patients who did not need a rescue angioplasty were costlier than PHT patients followed with rescue angioplasty (p 0.01), but as costly as PCA patients. Results were not modified using U.S. or U.K. costs. Cost-efficacy ratio. At one-year follow-up when the primary clinical end point for efficacy is considered, PCA is a cost-minimizing strategy with a non-significant difference in efficacy but 10% less costly than PHT, and with the same relative difference when applied to U.S. and U.K. costs. When the secondary clinical end point for efficacy is considered (including further revascularizations), PCA was both significantly more effective and less costly than PHT and is consequently a dominant strategy in the population studied. DISCUSSION An ideal strategy is a treatment that can both improve clinical outcome and save costs and resource consumption. Moreover, it must be applicable to the majority of diseased patients. Over the last 10 years, PCA has shown better clinical outcome and lower cost for several subsets of patients than hospital thrombolysis in skilled high-volume catheterization centers (5,6,12), whereas results from registries during the same period have been less favorable (11). The CAPTIM study was designed to compare PHT (combined with rescue angioplasty if needed) to PCA in patients within a short distance of a catheterization center. A centralized triage system allowed direct transportation of the patient to the catheterization laboratory for the PCA arm or to the cardiology intensive care unit for the PHT arm. The main result of the CAPTIM study indicates that the rate of major cardiovascular events (death re-infarction stroke) is no different after PHT than after PCA. Moreover, about one-third of the patients treated by thrombolysis needed rescue angioplasty. The CAPTIM cost-efficacy sub-study, performed prospectively for all patients included in three of the participating centers, also shows a similar clinical outcome between the two strategies and highlights several important issues concerning resource consumption: the rate of immediate mechanical revascularization was respectively 88% for the PCA group and 35% for the PHT group (rescue angioplasty), whereas the rate of subsequent revascularization was higher after PHT, either during the initial hospitalization period (42% vs. 10%) or after one-year follow-up (49% vs. 23%). Overall, less patients underwent revascular-

6 Table 3. Reference Costs of Resources and Costs Per Patient by Reperfusion Group Reference Cost ($) Per day PCA (n 143) In-Hospital Cost Per Patient PHT (n 146) Mann-Whitney p Value PCA (n 143) One-Year Cumulative Cost Per Patient PHT (n 146) Mann-Whitney p Value Hospitalizations* ICCU 602 2,041 1,228 2,214 1, ,233 1,339 2,407 1, Other ICU , ,822 NS 490 4, ,018 NS Medical ward 196 1, , ,567 1,038 1, NS Rehabilitation 201 1,595 3,138 2,038 5,749 NS Subtotal 3,550 4,208 3,922 3, ,885 5,023 6,492 7, Revascularization procedures Per unit PCI procedure 971 1, , Balloon Stent GP IIb/IIIa inhibitors (/unit) Intra-aortic balloon pump NS NS Overall PCI cost 1, , ,299 1,472 1,413 1, Overall CABG cost 3, NS NS Thrombolysis (/unit) 1, , , Subtotal 2, , ,552 1,449 2,774 1, Mobile care unit (/h) 866 1, , , , Other medications (/day) NS NS Testing and fees Lab X-ray ECG (/day) 124 1, , , , Echocardiography (/unit) NS NS Nuclear cardiology NS NS Exercise stress tests NS Subtotal 1, , NS 1, , NS Total 8,287 5,160 9,170 4, ,132 7,671 13,356 9, ,440 [6,236 8,719] 8,553 [7,181 9,490] 9,872 [8,483 14,689] 11,216 [8,816 14,555] All costs are in U.S.$. *Fixed costs (facilities, equipment, maintenance) variable costs (generic caregiver time, catering, supplies except drugs, laboratory tests, and interventions). Caregiver time, contrast agents, sterile supplies, maintenance (excluding stent and balloon costs). Fixed costs (facilities, equipment, maintenance) variable costs (labor costs, drugs, and supplies). Data are expressed as mean SD or median [25 75 percentile]. ECG electrocardiogram; GP glycoprotein; PCA primary coronary angioplasty; PHT pre-hospital thrombolysis; other abbreviations as in Table Machecourt et al. JACC Vol. 45, No. 4, 2005 The CAPTIM Cost-Efficacy Sub-Study February 15, 2005:515 24

7 JACC Vol. 45, No. 4, 2005 February 15, 2005: Machecourt et al. The CAPTIM Cost-Efficacy Sub-Study 521 Table 4. Sensitivity Cost Analysis Using U.K and U.S. Costs U.K. Costs U.S. Costs Present Study Costs PCI PHT PCA PHT PCA PHT Pre-hospital ,036 1,239 Hospitalization 8,325 9,202 10,069 11,129 5,128 5,521 PCI 3,628 2,479 4,476 3,083 1,699 1,209 CABG Thrombolysis/GP IIb/IIIa inhibitors 513 1, , ,151 Other treatments Total initial hospitalization 13,368 14,324 16,220 17,644 8,287 9,170 Rehabilitation 2,488 3,118 2,804 3,515 1,594 2,038 Re-hospitalizations 1,414 1,537 3,169 3, New revascularizations ,417 1, Initial to 1 yr 4,754 5,397 7,390 8,459 2,809 2,982 Total 1 yr 18,122 19,717 23,610 26,103 11,096 12,152 All costs are in U.S. $. Hospitalization costs include cost in intensive coronary care unit and normal ward, laboratory and X-ray testing, echocardiography, electrocardiograms, radionuclide imaging, and professional fees. Outpatients follow-up costs are not included. CABG coronary artery bypass graft; GP glycoprotein; PCI percutaneous coronary intervention; PHT pre-hospital thrombolysis. ization in the PHT group after one-year follow-up (68% vs. 90% for the PCA group). Overall, direct costs were 11% significantly lower at the end of the initial hospitalization period after PCA and 10% lower after one-year follow-up. The higher costs of mechanical revascularization in the PCA group, including physician and paramedic charges and use of glycoprotein IIb/IIIa inhibitors, more than offset the PHT group by the cost of the fibrinolytic agent and the cost associated with a longer hospital stay. This longer hospital stay in the PHT group was the result of a higher rate of Figure 1. Subgroup analysis of difference of cost between pre-hospital thrombolysis (PHT) and primary coronary intervention (PCI), according to the baseline presentation, the presence of an immediate or emergent revascularization, the occurrence of a major cardiac event (composite of death, myocardial infarction [MI], or stroke) or bleeding. White bars PHT group; black bars PCI group. For a global risk alpha 5%, the significance level for each analysis was p SBP systolic blood pressure.

8 522 Machecourt et al. JACC Vol. 45, No. 4, 2005 The CAPTIM Cost-Efficacy Sub-Study February 15, 2005: re-infarction and a higher rate of recurrent ischemia as assessed by the recurrence of symptoms or stress tests, leading to a higher rate of emergent angioplasties. As to the efficacy criterion, the Primary Angioplasty in Myocardial Infarction (PAMI) study (18), the Global Use of Strategies to Open Occluded Coronary Arteries (GUSTO) IIb substudy (19), the Air PAMI study (6), the PRimary Angioplasty in patients transferred from General community hospitals to specialized PTCA Units with or without Emergency thrombolysis (PRAGUE)-2 study (7), the DANish Multicenter Randomized Study on Fibrinolytic Therapy versus Acute Coronary Angioplasty in Acute Myocardial Infarction (DANAMI-2) (8), and the Canadian Stenting versus Thrombolysis in Acute Myocardial Infarction Trial (STAT) (13) all demonstrated a better clinical outcome with PCA. (All the studies used in-hospital thrombolysis.) Our different results (same efficacy between the two groups) can be explained by three factors: the beneficial effect of the pre-hospital triage of our patients, the pre-hospital administration of the thrombolytic agent, and the liberal use of rescue angioplasty after PHT. Pre-hospital triage of the patients allows a very short time to reperfusion, particularly for the PHT group (130 min), and more patients were treated 2 h from the onset of pain, a situation where the fibrinolytic agents are more effective (20 22). Rescue angioplasty was performed in 35% of our patients and in only 2% to 6% of patients in the other recent studies (7,8). The mortality rate in our PHT group is low (3.8%), and much higher in these studies (10.3% and 7.3%, respectively). Concerning resource consumption, most previous studies found that primary angioplasty was less costly than hospital thrombolysis (6,10,12,13), and our study extends the result to PHT. As in the present study, the PAMI sub-study also found lower costs for primary angioplasty, because of earlier patient discharge and lower recurrence of ischemia (12). The Canadian STAT study shows higher costs in the tissue plasminogen activator group compared with the angioplasty ones, this resulting from a longer hospital stay (13). De Boer et al. (10) found no difference in costs but a better clinical outcome after angioplasty. However, streptokinase, which is less expensive and less effective than alteplase. was used for this study. Total costs reported in most studies for treatment of acute coronary syndromes were higher than in the CAPTIM study (10,12,14 17), and in a few studies (13) total costs and unit costs were close to our results. Total costs in Stent-PAMI were $15,004 for PTCA without stent implantation versus $16,959 for PTCA with stent implantation (23); in the Treat Angina with Aggrastat and Determine Cost of Therapy with an Invasive or Conservative Strategy (TACTICS) study, total costs were $15,714 for the invasive approach versus $14,047 for the conservative one compared with $8,287 and $9,170 in CAPTIM and $6,354 and $7,893 in the STAT study. These differences in total costs may be explained by different factors: As noticed by Taira et al. (24), the method used in cost estimates can lead to significant differences in results. The methods using hospital charges converted to costs that are applied in most studies (10,12,14 17) lead to higher figures than the bottom-up method that used itemized costs from the local hospital accounting system (13,23). Differences in health care systems may be another important factor. For instance, the median costs of drugs were very low in CAPTIM as statins, beta-blockers, angiotensinconverting enzyme inhibitors, or clopidogrel were not charged to the hospital, because of negotiation mechanisms between pharmaceutical firms and hospital pharmacy departments. Physician fees are absent from our estimate, and physician labor costs were included only on the basis of the number of hours worked per patient (as patients were treated in public non-profit facilities). This also leads to an underestimation of our real costs, and it partially explains a number of discrepancies with some studies where physician fees ranged from $1,298 (13) to $3,372 (17,18) per patient. The costs of medical or surgical supplies such as stents or the cost of thrombolytic agents greatly vary from one system to another. Some cost inputs such as maintenance, technician labor, or house-keeping inputs may have been missed or underestimated in our micro-costing analysis. Hence, for the purpose of checking if our results can be extrapolated, a sensitivity analysis was performed whereby U.S. and U.K. costs (based on charges reduced to costs) were applied to our patients. Total costs measured were, at the end of the first in-hospital period, respectively 1.9 and 1.6 times higher than using our local costs. This estimate using U.S. costs was slightly higher than the costs assessed in Stent-PAMI, this being the result of a longer hospital stay after AMI in our study than in the U.S. Every cost component (hospitalization, revascularization procedures, pharmacy) was 1.6 to 2.3 times higher in the U.S., except for pre-hospital care (because of the particular organization of our mobile care unit allowing PHT with a physician on board the ambulance). The cost difference between the two strategies in favor of PCA remains unchanged, PCA being 7.5% to 11% less expensive than PHT regardless of the health care and accounting system chosen. Another parameter largely varying from one country to another is the rate of revascularization during the inhospital period. It was 62% for our PHT group (including 35% of rescue angioplasty). In comparison with other studies, this rate is high, ranging from 42% (13) to60% (18,25) in most of them, but is as low as 15% and 17% in some countries (25) or studies (8) where rescue angioplasty after thrombolysis was rarely performed. From our subgroup analysis, PHT patients with no rescue or emergent PCA remained 4% non-significantly costlier than PCA patients, whereas PHT patients with PCI were 15% higher in cost. Consequently, the revascularization rate only mildly influences the costs difference between the two strategies. Adjusting for other different variables did not significantly alter the cost difference in favor of the PCA arm. However, the lower cost of PCA over PHT was large for low-risk AMI

9 JACC Vol. 45, No. 4, 2005 February 15, 2005: Machecourt et al. The CAPTIM Cost-Efficacy Sub-Study 523 patients at presentation for those who did not require a PTCA after the initial coronary angiography and for eventfree patients during the in-hospital period. For high-risk patients at presentation, there was no cost difference between the two strategies, but for this subset of patients PHT has advantages in terms of efficacy: cardiogenic shock occurred less often, and mortality had a trend to be lower for AMI patients treated early (20). Interestingly, the total hospital cost in the fibrinolytic group of the CAPTIM study was comparable or even lower in constant currency terms than the cost of myocardial infarction patients treated with alteplase in the 1990s in the same institutions, before the era of primary or rescue angioplasty (26). In this earlier study, the mortality rate and the hospital stay were far higher. Study limitations. This prospective sub-study was performed only on a subset of 299 patients from the 840 enrolled in the CAPTIM trial in the three university hospitals that initiated the main trial. Baseline data are well balanced between the whole group of patients and the sub-study patients, as the randomization of patients was stratified in each center. These three institutions are within 100 miles of each other, permitting the efficient and accurate collection of all the resources used by the two fellows and the research nurse. The micro-costing evaluation was performed in two of the three hospitals and then extrapolated to the third. These three structures have similar clinical standards and the same financial and accounting system. Extrapolation to other health care systems may be questioned as details of costing differ from one country to another. Clinical decisions are conducted through the same guidelines; however, the rate of catheterization after thrombolysis can differ from one country to another. Nevertheless, our sensitivity analyses show that the results would not have been drastically different. This study is, to our knowledge, the first comparing PCA with PHT and liberal use of rescue angioplasty. However, the comparison between primary angioplasty and PHT followed by systematic 60- to 90-min coronary angiography and angioplasty if necessary ( facilitated angioplasty ) also needs to be explored in terms of efficacy and costs. Our results suggest that PHT and PCA are as effective for AMI patients close to an available PCI center, PHT being probably more effective than PCA for very early presenters ( 2 h). These conclusions cannot be extrapolated to other AMI patients, especially when the patient cannot be transported within 1 h to a PCI center. Future directions. In the light of our results and the other recent studies, patients more than 60 min from the catheterization laboratory and those early presenters within 2 to 3 h from onset of symptoms should benefit from PHT. Conversely, patients presenting later and within 1 h of the PCI center should be referred directly to the catheterization laboratory without performing PHT. Importantly, we will reinforce the pre-hospital triage of every AMI patient via the mobile care unit to optimize the network of care for these patients. Some questions remain open to discussion in the following setting: Should immediate transfer to a PCI center be applied to all AMI patients after PHT, or should it be reserved only for high-risk patients with failed reperfusion or in hemodynamically unstable condition? Because 10% to 15% of our patients treated with PTCA were hospitalized for a second procedure, will active stents further reduce the re-hospitalization rate by reducing the restenosis rate? Ongoing studies should provide answers to these important points. Reprint requests and correspondence: Dr. Jacques Machecourt, Service Cardiologie et Urgences Cardiologiques, Centre Hospitalo Universitaire Grenoble BP 217 X 38043, Grenoble, France. JMachecourt@chu-grenoble.fr. REFERENCES 1. Ryan TJ, Antman EM, Brooks NH, et al update: ACC/AHA guidelines for the management of patients with acute myocardial infarction: executive summary and recommendations: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Acute Myocardial Infarction). Circulation 1999;100: Bertrand ME, Simoons ML, Fox KA, et al., Task Force on the Management of Acute Coronary Syndromes of the European Society of Cardiology. Management of acute coronary syndromes in patients presenting without persistent ST-segment elevation. Eur Heart J 2002;23: The European Myocardial Infarction Project Group (EMIP). Pre hospital thrombolytic therapy in patients with suspected acute myocardial infarction. N Engl J Med 1993;329: Morrison LJ, Verbeek PR, McDonald AC, et al. Mortality and pre-hospital thrombolysis for acute myocardial infarction: a metaanalysis. JAMA 2000;283: Keeley EC, Boura JA, Grines CL. Primary angioplasty versus intravenous thrombolytic therapy for acute myocardial infarction: a quantitative review of 23 randomised trials. Lancet 2003;361: Grines CL, Westerhausen DR Jr, Grines LL, et al. Air PAMI Study Group. A randomized trial of transfer for primary angioplasty versus on-site thrombolysis in patients with high-risk myocardial infarction: the Air Primary Angioplasty in Myocardial Infarction study. J Am Coll Cardiol 2002;39: Widimsky P, Budesinsky T, Vorac D, et al., for the PRAGUE Study Group Investigators. Long distance transport for primary angioplasty vs immediate thrombolysis in acute myocardial infarction: final results of the randomized national multicentre trial PRAGUE-2. Eur Heart J 2003;24: Andersen HR, Nielsen TT, Rasmussen K, et al., for the DANAMI-2 Investigators. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med 2003;349: Bonnefoy E, Lapostolle F, Leizorovicz A, et al. Primary angioplasty versus pre hospital fibrinolysis in acute myocardial infarction: a randomised study. Lancet 2002;360: De Boer MJ, Van Hout BA, Liem AL, et al. A cost-effective analysis of primary coronary angioplasty versus thrombolysis for acute myocardial infarction. Am J Cardiol 1995;76: Every NR, Parsons LS, Hlatky M, et al., for the Myocardial Infarction Triage and Intervention (MITI) Investigators. A comparison of thrombolytic therapy with primary coronary angioplasty for acute myocardial infarction. N Engl J Med 1996;335: Stone GW, Grines CL, Rothbaum D, et al. Analysis of the relative costs and effectiveness of primary angioplasty versus tissue-type plasminogen activator: the Primary Angioplasty in Myocardial Infarction (PAMI) trial. J Am Coll Cardiol 1997;29:901 7.

10 524 Machecourt et al. JACC Vol. 45, No. 4, 2005 The CAPTIM Cost-Efficacy Sub-Study February 15, 2005: Le May MR, Davies RF, Labinaz M, et al. Hospitalization costs of primary stenting versus thrombolysis in acute myocardial infarction: cost analysis of the Canadian STAT Study. Circulation 2003;108: 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: Mark DB, Talley JD, Topol EJ, et al. Economic assessment of platelet glycoprotein IIb/IIIa inhibition for prevention of ischemic complications of high-risk coronary angioplasty. EPIC Investigators. Circulation 1996;94: Lincoff AM, Mark DB, Tcheng JE, et al. Economic assessment of platelet glycoprotein IIb/IIIa receptor blockade with abciximab and low-dose heparin during percutaneous coronary revascularization: results from the EPILOG randomized trial. Circulation 2000;102: Mahoney EM, Jurkovitz CT, Chu H, et al., for the TACTICS-TIMI 18 Investigators. Treat Angina with Aggrastat and Determine Cost of Therapy with an Invasive or Conservative Strategy Thrombolysis in Myocardial Infarction. Cost and cost-effectiveness of an early invasive vs conservative strategy for the treatment of unstable angina and non ST-segment elevation myocardial infarction. JAMA 2002;288: Grines CL, Browne KF, Marco J, et al., for the Primary Angioplasty in Myocardial Infarction (PAMI) Study Group. A comparison of immediate angioplasty with thrombolytic therapy for acute myocardial infarction. N Engl J Med 1993;328: The Global Use of Strategies to Open Occluded Coronary Arteries in Acute Coronary Syndromes Angioplasty Sub Study Investigators. A clinical trial comparing primary coronary angioplasty with tissue plasminogen activator for acute myocardial infarction. N Engl J Med 1997;336: Steg PG, Bonnefoy E, Chabaud S, et al. Impact of time to treatment on mortality after prehospital fibrinolysis or primary angioplasty: data from the CAPTIM randomized clinical trial. Circulation 2003;108: Wallentin L, Goldstein P, Armstrong PW, et al. Efficacy and safety of tenecteplase in combination with the low-molecular-weight heparin enoxaparin or unfractionated heparin in the prehospital setting: the Assessment of the Safety and Efficacy of a New Thrombolytic Regimen (ASSENT)-3 PLUS randomized trial in acute myocardial infarction. Circulation 2003;108: Nallamothu BK, Bates ER. Percutaneous coronary intervention versus fibrinolytic therapy in acute myocardial infarction: is timing (almost) everything? Am J Cardiol 2003;92: Cohen DJ, Taira DA, Berezin R, et al. Cost-effectiveness of coronary stenting in acute myocardial infarction: results from the Stent Primary Angioplasty in Myocardial Infarction (Stent-PAMI) trial. Circulation 2001;104: Taira DA, Seto TB, Siegrist R, Cosgrove R, Berezin R, Cohen DJ. Comparison of analytic approaches for the economic evaluation of new technologies alongside multicenter clinical trials. Am Heart J 2003; 145: Gandjour A, Kleinschmit F, Lauterbach KW, and the INTERCARE International Investigators. European comparison of costs and quality in the treatment of acute myocardial infarction ( ). Eur Heart J 2002;23: Machecourt J, Dumoulin J, Calop J, et al. Cost effectiveness of thrombolytic treatment for myocardial infarction: comparison of anistreplase, alteplase and streptokinase in 270 patients treated within 4 hours. Eur Heart J 1993;14:75 83.