COURAGE

15 US VA 19 US Non-VA Hospitals 1,355 patients 16 Canadian Hospitals 932 patients 50 Hospitals 2,287 patients enrolled between 6/99-1/04 A North American Trial

Design 2287 Patients Randomized to PCI + Optimal Medical Therapy vs Optimal Medical Therapy alone Intensive, guideline-driven medical therapy and lifestyle intervention in both groups Primary Outcome: Death + MI Secondary Outcomes include health care economics, HRQOL 2.5 to 7 year (median 4.6 year) follow-up

Definition of MI In patients with a clinical presentation c/w an acute ischemic syndrome and who have 1 of the following: – New Q Waves >0.03sec in > 2 contiguous leads as assessed by ECG Core Laboratory reading – For Spontaneous MI: CK/CK-MB > 1.5X UNL or (+) Troponin > 2.0X UNL – For Peri-PCI MI: CK/CK-MB > 3.0X UNL or (+) Troponin > 5.0X UNL (only if CK not available) – For Post-CABG MI: CK-MB > 10.0X UNL or (+) Troponin > 10.0xUNL (only if CK not available)

Inclusion/Exclusion Criteria Inclusion Men and Women 1, 2, or 3 vessel disease (> 70% visual stenosis of proximal coronary segment) Anatomy suitable for PCI CCS Class I-III angina Objective evidence of ischemia at baseline, ECG or imaging ACC/AHA Class I or II indication for PCI Exclusion Uncontrolled unstable angina Complicated post-MI course Revascularization within 6 months Ejection fraction <30% Cardiogenic shock/severe heart failure History of sustained or symptomatic VT/VF

Coronary Intervention Best practice May use all FDA or Health Canada approved devices Completeness of revascularization as clinically appropriate

Risk Factor Goals VariableGoal SmokingCessation Total Dietary Fat / Saturated Fat<30% calories / <7% calories Dietary Cholesterol<200 mg/day LDL cholesterol (primary goal)60-85 mg/dL HDL cholesterol (secondary goal)>40 mg/dL Triglyceride (secondary goal)<150 mg/dL Physical Activity30-45 min. moderate intensity 5X/week Body Weight by Body Mass indexInitial BMI Weight Loss Goal BMI <25 > % relative weight loss Blood Pressure<130/85 mmHg DiabetesHbAlc <7.0%

Optimal Medical Therapy Pharmacologic Anti-platelet: aspirin; clopidogrel in accordance with established practice standards Statin: simvastatin ± ezetimibe or ER niacin ACE Inhibitor or ARB: lisinopril or losartan Beta-blocker: long-acting metoprolol Calcium channel blocker: amlodipine Nitrate: isosorbide 5-mononitrate Lifestyle Smoking cessation Exercise program Nutrition counseling Weight control Applied to Both Arms by Protocol and Case-Managed

Enrollment and Outcomes 3,071 Patients met protocol eligibility criteria 2,287 Consented to Participate (74% of protocol-eligible patients) 1,149 Were assigned to PCI group 46 Did not undergo PCI 27 Had a lesion that could not be dilated 1,006 Received at least one stent 784 Did not provide consent Did not receive MD approval Declined to give permission - 97 Had an unknown reason 107 Were lost to follow-up 1,149 Were included in the primary analysis 1,138 Were assigned to medical-therapy group 97 Were lost to follow-up 1,138 Were included in the primary analysis

Baseline Clinical and Angiographic Characteristics CharacteristicPCI + OMT (N=1149)OMT (N=1138)P Value Age – yr.62 ± ± Sex %0.95 Male85 % Female15 % Race or Ethnic group %0.64 White86 % Non-white14 % CLINICAL Angina (CCS – class) % and I42 %43 % II and III59 %56 % Median angina duration5 (1-15) months Median angina episodes/week 3 (1-6)

Baseline Clinical and Angiographic Characteristics CharacteristicPCI + OMT (N=1149)OMT (N=1138)P Value CLINICAL History – % Diabetes32 %35 %0.12 Hypertension66 %67 %0.53 CHF 5 % 4 %0.59 Cerebrovascular disease 9 % 0.83 Myocardial infarction38 %39 %0.80 Previous PCI15 %16 %0.49 CABG11 % 0.94

Baseline Clinical and Angiographic Characteristics CharacteristicPCI + OMT (N=1149)OMT (N=1138)P Value CLINICAL Stress test0.84 Total patients - %85 %86 % Treadmill test57 % 0.84 Pharmacologic stress43 % Nuclear imaging - %70 %72 %0.59 Single reversible defect22 %23 %0.09 Multiple reversible defects65 %68 %0.09 ANGIOGRAPHIC Vessels with disease – %0.72 1, 2, 331, 39, 30 %30, 39, 31 % Disease in graft62 %69 %0.36 Proximal LAD disease31 %37 %0.01 Ejection fraction60.8 ± ±

Long-Term Improvement in Treatment Targets (Group Median ± SE Data) Treatment TargetsBaseline60 Months PCI +OMTOMTPCI +OMTOMT SBP131 ± ± ± ± 0.92 DBP74 ± ± ± 0.65 Total Cholesterol mg/dL172 ± ± ± ± 1.64 LDL mg/dL100 ± ± ± ± 1.21 HDL mg/dL39 ± ± ± ± 0.75 TG mg/dL143 ± ± ± ± 4.70 BMI Kg/M²28.7 ± ± ± ± 0.31 Moderate Activity (5x/week)25% 42%36%

Angiographic Outcomes PCI was attempted on 1,688 lesions (in 1,077 patients), of whom 1,006 received at least 1 stent 590 patients (59%) received 1 stent and 416 (41%) received 2 or more stents Stenosis diameter was reduced from a mean of 83 ± 14% to 31 ± 34% in the 244 balloon PTCA lesions, and from 82 ± 12% to 1.9 ± 8% in the 1,444 stented lesions Angiographic success (<20% residual stenosis by visual assessment) post-PCI was 93% and clinical success was 89% post-PCI.

Need for Subsequent Revascularization At a median 4.6 year follow-up, 21.1% of the PCI patients required an additional revascularization, compared to 32.6% of the OMT group who required a 1 st revascularization 77 patients in the PCI group and 81 patients in the OMT group required subsequent CABG surgery Median time to subsequent revascularization was 10.0 mo in the PCI group and 10.8 mo in the OMT group

Survival Free of Death from Any Cause and Myocardial Infarction Number at Risk Medical Therapy PCI Years PCI + OMT Optimal Medical Therapy (OMT) Hazard ratio: % CI ( ) P =

Overall Survival Number at Risk Medical Therapy PCI Years PCI + OMT OMT 7 Hazard ratio: % CI ( ) P = 0.38

Freedom From Hospitalization for ACS Number at Risk Medical Therapy PCI Years PCI + OMT OMT 7 Hazard ratio: % CI ( ) P = 0.56

Freedom From Myocardial Infarction Number at Risk Medical Therapy PCI Years PCI + OMT OMT 7 Hazard ratio: % CI ( ) P = 0.33

Primary and Secondary Outcomes OutcomeHazard Ratio (95% Cl)Number of EventsP Value PCI+OMTOMT Death and nonfatal MI ( )0.62 Death6874 Periprocedural MI359 MI Death, MI, and stroke ( )0.62 Hospitalization for ACS ( )0.56 Death ( )0.38 Total nonfatal MI ( )0.33 Periprocedural MI359 MI Revascularization (PCI or CABG) ( )<0.001

Freedom from Angina By CCS Class During Long-Term Follow-up CharacteristicPCI + OMTOMT CLINICAL Angina free – no. Baseline12%13% 1 Yr66%58% 3 Yr72%67% 5 Yr74%72% The comparison between the PCI group and the medical-therapy group was significant at 1 year ( P<0.001) and 3 years (P=0.02) but not at baseline or 5 years.

Quality of Life Data: QuestionnairesSeattle Angina Questionnaire (5 scales, all 0-100) Rand 36 (8 scales, all 0-100) Utility by Standard Gamble (Scaled 0 – 1.0) Data Collection:Baseline, 1, 3, 6, 12 months, then annually. Quality of Life Data

SAQ Data – Physical Limitation Follow-upPCI+OMTOMT OnlyUnadjusted P Value Baseline Month Months Months Months Months Months

SAQ Data – Angina Frequency Follow-upPCI+OMTOMT OnlyUnadjusted P Value Baseline Month8276< Months8580< Months Months Months Months

SAQ Data – Quality of Life Follow-upPCI+OMTOMT OnlyUnadjusted P Value Baseline Month6862< Months7368< Months7570< Months Months Months

Angina Free by SAQ Angina Frequency Score

* p<0.01

SAQ Data – Angina Frequency Clinically Significant Improvement Follow-upPCI+OMTOMT OnlyUnadjusted P Value 1 Month39%30%< Months47%40% Months50%44% Months52%46% Months54%47% Months57%50%0.045 NNT 17 for 1 to improve at 6 months Wyrwich KW et al. Clinically important differences in health status for patients with heart disease: an expert consensus panel report. Am Heart J 2004;147

SAQ Data – Angina Frequency Tercile Scores (Mean Interaction P=0.008, Clinically Significant Improvement Interaction P<0.0001) PCI + OMTOMTClinically Significant Improvement TercileFollow-upMean P ValuePCI+OMTOMTP Value 1stBaseline months %73% months %81% months %84% months %88%0.14 2ndBaseline months8679< %50% months %56% months %58% months %64%0.11 3rdBaseline months months months months

Sensitivity and Subgroups The SAQ data were quite similar for the analysis of patients with complete data up to three years, with SAQ scores set to zero for deaths, for an analysis by initial treatment received, and imputation of missing values. Similar results were noted for subgroups defined by gender, age and severity of angina at baseline.

Rand -36 Physical Functioning Follow-upPCI+OMTOMT OnlyUnadjusted P Value Baseline Month Months Months Months Months Months

* p<0.01

Conclusions PCI as an initial management strategy in the setting of stable CAD has not been shown to reduce the incidence of Death or MI PCI has not been shown to prolong life expectancy PCI+OMT does offer better control of angina than OMT alone Most patients will have improvement in anginal status whether treated initially with PCI+OMT or OMT alone