This investigation evaluated the changes of pulmonary perfusion at 4 different points of follow-up within 1 y in patients with pulmonary embolism (PE) and the factors predictive of complete or incomplete recovery of pulmonary perfusion. Methods: Patients with symptomatic PE underwent perfusion lung scintigraphy and blood gas analysis within 48 h from clinical presentation, after 1 wk, and after 1, 6, and 12 mo; echocardiography was performed at baseline and after 6 and 12 mo. All perfusion lung scintigraphy scans were examined by 2 expert nuclear medicine physicians with a scoring method that attributed a score of 0, 0.5, or 1 for extension (maximum score, 18) to the presence of perfusion defects (PD), both at baseline and on each follow-up scan. Results: Among 183 patients who completed 1 y of follow-up, the median baseline PD score was 8.2; it decreased significantly at each follow-up time point until 6 mo (P , 0.001). Median baseline alveolar-arterial difference in oxygen partial pressure (PA-aO 2) was 50.9 and decreased significantly up to 1 mo (P , 0.001); median pulmonary artery systolic pressure (PAsP) was 45.9 mm Hg and decreased significantly until 12 mo (P , 0.001). A correlation was found between PD and both PA-aO 2 (P , 0.05) and PAsP (P , 0.05). We found a correlation between PD ≠ 0 and PAsP $ 40 mm Hg at 12 mo (P , 0.05); in 6 (3.3%) of these patients such a correlation was still present after 24 mo, suggesting they could develop chronic thromboembolic pulmonary hypertension. Low baseline PD (odds ratio, 0.80; P , 0.0001) and high 1-wk percent recovery (odds ratio, 1.04; P , 0.0001) were predictive factors of complete 6-mo recovery. Conclusion: Perfusion scintigraphy may be useful to follow patients with PE. The follow-up should consist of 3 steps: the baseline examination, which reflects the severity of PE; the scan at 1 wk, which indicates the early amount of reperfusion; and the scan at 6 mo, which demonstrates the maximum attainable recovery. Patients with incomplete recovery and persistence of pulmonary hypertension on the 24-mo control should be further studied for possible development of chronic thromboembolic pulmonary hypertension.