Previous small clinical trials have suggested that treatment with nitric oxide donors in suspected myocardial infarction can reduce mortality by 30-35%. To confirm this finding in a large-scale trial, we compared molsidomine and its active metabolite linsidomine (a nitric oxide donor) with placebo in 4017 patients with acute myocardial infarction. In our trial, patients without signs of overt heart failure (Killip III/IV) were randomly assigned in a double-blind design within 24 h of symptom onset to receive linsidomine 1 mg/h intravenously for 48 h, followed by 16 mg molsidomine by mouth daily for 12 days (n = 2007), or an identical placebo (n = 2010). All other treatments could be used at the responsible physician's discretion with the exception of systematic vasodilator treatment. The molsidomine and placebo groups showed similar all-cause 35-day mortality (168 [8.4%] vs 176 [8.8%] deaths, p = 0.66), and adjustment for baseline variables in a Cox model had no effect. Similarly, we found no difference for long-term mortality (mean follow-up 13 months; 294 [14.7%] vs 285 [14.2%] deaths, p = 0.67). The two groups showed similar frequencies of major and minor adverse events; only headache was significantly more common in the molsidomine group. Changes in treatment practices and the lower risk profile of our study subjects than of participants in previous trials may explain the results. It is still not clear whether nitric oxide donors can improve survival in higher-risk myocardial infarction patients.