OBJECTIVE: Coronary revascularization methods are among the most common major medical procedures performed in industrialized countries. The short- and long-term comparative effectiveness of different techniques remains undetermined. METHODS AND RESULTS: The study population included all adult patients (94,864 subjects) admitted for acute myocardial infarction (AMI) in 2004 in Italy, as recorded in the National Hospital Discharge Records. Invasive reperfusion therapy (coronary artery bypass graft, percutaneous coronary intervention), and standard medical treatments were compared. End points were short-term (30-day) in-hospital mortality and any combination of new revascularizations, re-AMI, stroke, or death (MACCE) occurring within 6 and 12 months from initial treatment. Risk factors and comorbidities were used to define patients' health status. The Cox model was applied to evaluate risk-adjusted hazard ratios (HR) for different approaches. Medical treatment was used as the reference category. Propensity score matching was performed to evaluate selection bias in the allocation to CABG or percutaneous coronary intervention (PCI). Both short-term mortality and 6- and 12-month MACCE risk for invasive reperfusion therapy was significantly lower than the reference (HR: 0.50, P < 0.001; 0.54, P < 0.001; 0.64, P < 0.001, respectively). In the matched population, while short-term mortality was significantly lower for PCI (HR 0.36, P < 0.001), long-term MACCE risk was lower for CABG (6-month HR 2.93, P < 0.001; 12-month HR 3.01, P < 0.001). CONCLUSIONS: In spite of adjustments for patients' health status, medical treatment resulted in a higher mortality and increased MACCE risk. While PCI gave the best short-term results, after 1 year the risk of re-interventions or further events was significantly lower for patients undergoing CABG.
OBJECTIVE: Coronary revascularization methods are among the most common major medical procedures performed in industrialized countries. The short- and long-term comparative effectiveness of different techniques remains undetermined. METHODS AND RESULTS: The study population included all adult patients (94,864 subjects) admitted for acute myocardial infarction (AMI) in 2004 in Italy, as recorded in the National Hospital Discharge Records. Invasive reperfusion therapy (coronary artery bypass graft, percutaneous coronary intervention), and standard medical treatments were compared. End points were short-term (30-day) in-hospital mortality and any combination of new revascularizations, re-AMI, stroke, or death (MACCE) occurring within 6 and 12 months from initial treatment. Risk factors and comorbidities were used to define patients' health status. The Cox model was applied to evaluate risk-adjusted hazard ratios (HR) for different approaches. Medical treatment was used as the reference category. Propensity score matching was performed to evaluate selection bias in the allocation to CABG or percutaneous coronary intervention (PCI). Both short-term mortality and 6- and 12-month MACCE risk for invasive reperfusion therapy was significantly lower than the reference (HR: 0.50, P < 0.001; 0.54, P < 0.001; 0.64, P < 0.001, respectively). In the matched population, while short-term mortality was significantly lower for PCI (HR 0.36, P < 0.001), long-term MACCE risk was lower for CABG (6-month HR 2.93, P < 0.001; 12-month HR 3.01, P < 0.001). CONCLUSIONS: In spite of adjustments for patients' health status, medical treatment resulted in a higher mortality and increased MACCE risk. While PCI gave the best short-term results, after 1 year the risk of re-interventions or further events was significantly lower for patients undergoing CABG.