We sought to evaluate the rates, time course, and causes of death in the long-term follow-up of unselected patients with acute coronary syndromes (ACS). We enrolled 2046 consecutive patients hospitalized from January 2004 to December 2005 with an audited final diagnosis of ACS. The primary study end point was 5-year all-cause mortality. In our series, 896 patients had ST-segment elevation (STE) and 1,150 non–ST-segment elevation (NSTE). Mean age of the study population was 71.6 years. Primary percutaneous coronary intervention was performed in 86% of STE-ACS, and 70% of NSTE-ACS was managed invasively. The 5-year all-cause mortality was 36.4% for STE-ACS and 42.0% for NSTE-ACS, with patients with STE-ACS showing a trend boarding statistical significance toward a lower risk of mortality (hazard ratio [HR] = 0.88, 95% confidence interval [CI] 0.76 to 1.02, p = 0.08). Landmark analysis demonstrated that patients with STE-ACS had a higher risk of 30-day mortality (STE-ACS vs NSTE-ACS HR = 1.53, 95% CI 1.16 to 2.06, p = 0.003) whereas the risk of NSTE-ACS increased markedly after 1 year (STE-ACS vs NSTE-ACS HR = 0.67, 95% CI 0.53 to 0.84, p = 0.001). The contribution of noncardiovascular (CV) causes to overall mortality increased from 3% at 30 days to 34% at 5 years, with cancer and infections being the most common causes of non-CV death both in STE-ACS and NSTE-ACS. In conclusion, long-term mortality after ACS is still too high both for STE-ACS and NSTE-ACS. Although patients with STE-ACS have a higher mortality during the first year, the mortality of patients with NSTE-ACS increases later, when non-CV co-morbidities gain greater importance.