The rate of resolution of a first episode of pulmonary embolism (PE) is uncertain. A baseline test indicating any residual PE is pivotal in aiding a more accurate diagnosis of recurrent PE. This study aimed to assess the rate and risk factors of residual PE with either multidetector computed tomography imaging (MDCT) or lung perfusion scan (LPS) using a cross-sectional study in which consecutive patients were enrolled with a first objectively documented episode of symptomatic PE, and who were considered for possible treatment withdrawal after at least 3 months of anticoagulation. A first cohort of patients (n = 80) underwent MDCT, while the subsequent cohort (n = 93) underwent LPS. The two cohorts had similar characteristics, and 98.3% of patients had non high-risk index PE. MDCT detected residual PE in 15% of subjects (12/80, 95% CI 8-25%) after a mean of 9 months of anticoagulation. No clinical characteristics were significantly associated with residual PE at MDCT. LPS detected residual PE in 28% (26/93, 95% CI 19-38%) of patients after a period of a mean of 9 months of anticoagulation with a significant association with increasing age and known pulmonary disease. Resolution of PE was high after a first episode of non high-risk PE treated with heparin followed by at least 3 months of anticoagulation. Age and coexistent pulmonary disease influence the presence of residual PE detected by LPS, but not by MDCT. Further studies are warranted in which the presence of residual embolism is detected by repetition of the same test that had been initially carried out.