Choice of reperfusion strategy at hospitals with primary percutaneous coronary intervention: a National Registry of Myocardial Infarction analysis.

BACKGROUND Many hospitals with percutaneous coronary intervention (PCI) capability also use fibrinolytic therapy in patients with ST-segment elevation myocardial infarction, but factors influencing the choice of reperfusion strategy at these hospitals are poorly understood. We examined clinical and system-related factors associated with choice of reperfusion strategy in patients with ST-segment elevation myocardial infarction at PCI-capable hospitals. METHODS AND RESULTS We analyzed patients with ST-segment elevation myocardial infarction who presented to PCI-capable hospitals between July 1, 2000, and December 31, 2006, in the National Registry of Myocardial Infarction. Hierarchical multivariable logistic regression was used to examine the association between choice of reperfusion strategy and patient-, hospital-, and system-related factors. We identified 25 579 patients who received primary PCI and 14 332 patients who received fibrinolytic therapy at 444 PCI-capable hospitals. Use of reperfusion strategies varied widely across hospitals, although primary PCI use increased over the study period. Among the key clinical factors that favored primary PCI, cardiogenic shock and delayed presentation were associated with greater use of primary PCI (adjusted odds ratios 2.14 [95% confidence interval 1.72 to 2.66] and 1.18 [95% confidence interval 1.09 to 1.27], respectively), whereas a Thrombolysis in Myocardial Infarction risk score >/=5 was not. In contrast, female gender, advanced age, and nonwhite race, all risk factors for intracranial hemorrhage after fibrinolytic therapy, were not associated with increased use of primary PCI. Off-hours presentation had the strongest association overall, with an approximately 70% lower likelihood of patients undergoing primary PCI (adjusted odds ratio 0.27, 95% confidence interval 0.25 to 0.29). CONCLUSIONS Use of primary PCI, although increasing over recent years, is not universal at PCI-capable hospitals, and optimization of its use at such hospitals represents a potential opportunity to improve outcomes in patients with ST-segment elevation myocardial infarction.