(en) In 2013, pulmonary embolism remains a diagnostic challenge for emergency physicians. As neither clinical signs nor symptoms taken individually are specific or sensitive, further exams are necessary for each clinical suspicion. With the advent of new sensitive exams such as D-dimer and multi-detector computed tomographic pulmonary angiography the overtesting problem has appeared, in particular with an overall increase in negative further exams for PE with potentially harmful consequences. The diagnostic strategy is a probability approach, based on the estimation of the risk of error taken when excluding or confirming diagnosis. The current strategy is based on 3 levels: assessment of clinical probability, D-dimer measurement and computed tomographic pulmonary angiography. In this work, we aimed to optimize suspicion of pulmonary embolism in the emergency department and to rationalize the recourse to exams. Therefore, we analyzed and explored each step of the diagnostic strategy in order to better understand the suspicion and the disease.