Chest pain units (CPUs) have been established to optimize treatment of patients with acute coronary syndrome (ACS) and to early and accurately discharge patients with non-coronary chest pain. The aim of this analysis was to elucidate whether treatment of ACS patients in the CPU versus emergency department (ED) has prognostic implications.

Patients presenting with suspected ACS to either the ED between August 2004 and June 2005 or the CPU between July 2005 and May 2006 were retrospectively analyzed. Of 1,796 included patients, 483 had the discharge diagnosis ACS. When compared to patients with exclusion of ACS they had more cardiovascular risk factors and higher troponin, creatinine and C-reactive protein levels (P < 0.001) at admission. Within 1 year, 37 patients of the ACS group suffered an event. Treatment in the ED compared with the CPU showed a significant increase in hazard ratio of 2.1 (P = 0.034) for the combined endpoint death, myocardial infarction and stroke, remaining unchanged after adjusting for confounders. Event-free 1-year survival was higher in CPU patients for the combined endpoint (P _logrank = 0.02).

These results demonstrate a better 1-year prognosis for ACS patients treated in the CPU instead of the ED, therefore, supporting the idea to establish CPUs in Europe.