BackgroundPulmonary embolism which is an mysterious and difficult disease to diagnose is the third most common cause of death from cardiovascular disease. Despite recent clinical studies and technological development, pulmonary embolism diagnosing is hard and complicated. Diagnosis of pulmonary embolism starts with physcians suspicion. Firstly, assessment of clinical pre-test probability is important. Clinical pre-test probability is based on assessment of whether symptoms and signs are typical for pulmonary embolism. Geneva score and  Wells score are most known pre-tests. The revised Geneva score, a standardized clinical decision rule in the diagnosis of pulmonary embolism (PE). The Wells score is widely used for the assessment of pretest probability of pulmonary embolism (PE). The revised Geneva score is a fully standardized clinical decision rule. We compared the predictive accuracy of these two pre-test probabilities. Methods: In 119 consecutive patients, the clinical probability of PE was assessed prospectively by the Wells rule and retrospectively using the revised Geneva score. Patients comprised a random sample from a single center, participating in a large prospective multicenter diagnostic study. Results: The overall prevalence of PE was 31%. The prevalence of PE in the low-probability, intermediate-probability and high-probability categories are grouped by the revised Geneva score.  After three months of followup, any patient categorised into the low or intermediate clinical probability category by the revised Geneva score. Normal D-dimer result was diagnosed with acute venous thromboembolism. When we compare Geneva and Wells scores, their predictive values were similar for PE group. Predictive values of Geneva scores for PE and non-PE were not importantly different (p=0.169), but Wells scores were importantly higher for PE group (p=0.006).Conclusions: This study recommends that the performance of the revised Geneva score is equal to that of the Wells score. In addition, it appears safe to exclude PE in patients by the combination of a low or intermediate clinical probability by the revised Geneva score and a normal D-dimer. Determining clinical probability is the main step for diagnosis. High D-dimer value is important for clinical suspicion, but low values can’t eliminate the diagnosis. Also we showed that Wells score’s predictive value was higher than Genova score but predictions of mortality were similar. Both clinical risk classification and laboratory results must be evaluated  together with Genova score in suspected PE cases. In our population, the Wells score appeared to be more accurate than the simplified revised Geneva score. Patient outcomes should be examined in a prospective study.