BACKGROUND: We sought to describe the long-term prognosis after routine primary percutaneous coronary intervention (pPCI) in a contemporary consecutive population of patients with presumed ST-segment elevation myocardial infarction, compare it with similar results from the landmark DANAMI-2 trial, and to identify a possible impact of time of presentation and referral pattern. METHODS AND RESULTS: Long-term prognosis in 1019 presumed ST-segment elevation myocardial infarction patients, treated according to modern routine pPCI during the year 2004, was analyzed and compared with similar data from the DANAMI-2 trial. Furthermore, we analyzed the impact of patient presentation to the angioplasty center during "off hours" (4 pm to 8 am plus weekends and holidays) and the impact of being referred from noninvasive hospitals. At 3 years, 20.4% in the routinely treated population versus 19.6% in the DANAMI-2 trial reached the combined end point of death, reinfarction, or stroke (P=0.68), whereas the all-cause mortality was 13.0% and 13.7%, respectively (P=0.65). Patients admitted during off hours had the same risk of reaching the combined end point of death, reinfarction, or stroke compared with patients admitted during office hours (hazards ratio, 1.04; 95% CI, 0.8 to 1.5; P=0.81). Door-to-balloon times of less than 90 minutes were achieved in 60% among patients admitted directly to an invasive center but only in 40% among transferred patients (P<0.001). Despite this difference, no difference in unadjusted or adjusted long-term prognosis was found between the 2 groups. CONCLUSIONS: This study shows that ST-segment elevation myocardial infarction patients treated with contemporary routine pPCI achieve a similar long-term prognosis as patients in the landmark randomized pPCI trial (DANAMI-2). Furthermore, the long-term prognosis was the same regardless of whether the pPCI was performed during off hours or office hours. Thus, pPCI including transportation of patients from noninvasive centers can be applied successfully in a real-life population.