In patients following acute pulmonary embolism, little is known about long-term prognosis attributed to individual risk factors and entities of pulmonary embolism. The aim of our study was to identify predictors of long-term mortality and morbidity, taking into account precipitating causes and entities of pulmonary embolism, especially in cases of idiopathic pulmonary embolism. We reviewed the records of 257 consecutive patients presenting with acute pulmonary embolism in the emergency room of the University Hospital in Frankfurt between 1997 and 2006. The median of the follow-up time in this study population was 670 days (n = 236; interquartile range: 378-1397 days). The patients were divided into three groups depending on the cause of pulmonary embolism: idiopathic pulmonary embolism (n = 83; 33%), neoplasm (n = 44; 17%) and other causes (n = 130; 50%). As expected, patients with neoplastic-associated pulmonary embolism demonstrated the poorest prognosis in the Kaplan-Meier analysis. Surprisingly though, patients with idiopathic pulmonary embolism demonstrated a markedly worse long-term survival compared with other causes (log-rank P < 0.001). When Cox-regression analysis was restricted to patients discharged alive without neoplasm, idiopathic pulmonary embolism was the strongest predictor of long-term mortality (hazard ratio: 3.29; P = 0.017). Patients with acute pulmonary embolism and known malignancy demonstrated the highest mortality compared with those with other causes. However, idiopathic pulmonary embolism is a common diagnosis and is associated with an unfavorable long-term prognosis compared with other entities of pulmonary embolism.