Background. - Multicentre registries of myocardial infarction management show a steady improvement in prognosis and greater access to myocardial revascularization in a more timely manner. While French registries are the standard references, the question arises: are data stemming solely from the activity of French cardiac intensive care units (ICUs) a true reflection of the entire French population with ST-segment elevation myocardial infarction (STEMI)? Aim. - To compare data on patients hospitalized for STEMI from two French registries: the French registry of acute ST-elevation or non-ST-elevation myocardial infarction (FAST-MI) and the Echantillon generaliste des beneficiaires (EGB) database. Methods. - We compared patients treated for STEMI listed in the FAST-MI 2010 registry (n=1716) with those listed in the EGB database, which comprises a sample of 1/97th of the French population, also from 2010 (n=403). Results. - Compared with the FAST-MI 2010 registry, the EGB database population were older (67.2 +/- 15.3 vs 63.3 +/- 14.5 years; P < 0.001), had a higher percentage of women (36.0% vs 24.7%; P < 0.001), were less likely to undergo emergency coronary angiography (75.2% vs 96.3%; P < 0.001) and were less often treated in university hospitals (27.1% vs 37.0%; P = 0.001). There were no significant differences between the two registries in terms of cardiovascular risk factors, comorbidities and drug treatment at admission. Thirty-day mortality was higher in the EGB database (10.2% vs 4.4%; P < 0.001). Conclusions. - Registries such as FAST-MI are indispensable, not only for assessing epidemiological changes over time, but also for evaluating the prognostic effect of modern STEMI management. Meanwhile, exploitation of data from general databases, such as EGB, provides additional relevant information, as they include a broader population not routinely admitted to cardiac ICUs.