Acute pulmonary embolism causes increasing right ventricular afterload and hypoxia by obstructing the pulmonary artery and triggering the release of vasoconstricting factors, especially in patients with massive or moderate-large pulmonary embolisms. Acute pulmonary embolism can cause sudden death in up to 10% of patients due to right ventricular dysfunction. We reviewed the records of 58 consecutive patients (mean age, 59.2 years) with pulmonary embolisms admitted at our institution from April 1994 to December 1999. Thirty-five patients were female. Predisposing factors for venous thromboembolism were postoperative state, immobilization, obesity, etc. Seven patients had caval thrombosis (free floating type in 4). The first choice clinical diagnostic methods were echocardiography and computed tomography for patients with massive or moderate-large PE. Echocardiography was a rapid, practical, and sensitive technique for the identification of right ventricular dysfunction due to pulmonary embolism. Treatments consisted of surgical embolectomy in 7, catheter-directed fragmentation in 3, heparin plus thrombolytic therapy alone in 20, anti-coagulation therapy alone in 22, and conservative therapy including inotropic agents in 6. Overall early mortality was 17.2%. Causes of death were shock in 2, right ventricular failure in 5, multiple organ failure with coagulopathy in 2, and postresuscitation syndrome in 1. Inferior vena cava filters were used in 9 patients (permanent type in 6, temporary type in 3). Neither recurrence of pulmonary embolism, nor complications caused by filters were clinically recognized. Multivariate analysis showed that modified Greenfield's classification and right ventricular dysfunction were significant incremental risk factors. If patients with massive pulmonary embolism can not improve severe right ventricular dysfunction only due to medical therapy, prompt catheter based therapy or surgical embolectomy are recommended. Moreover, combined surgical thrombectomy for free-floating type caval thrombosis and proximal massive pulmonary embolism was safe and effective. The use of IVC filter was effective to prevent the fatal recurrence of pulmonary embolism. If fatal pulmonary embolism could be predicted, first of all, the use of temporally IVC filter should be considered before treatments for pulmonary embolism or deep vein thrombosis.

View full abstract

Our algorithm for diagnosing pulmonary thromboembolism combines ventilation/perfusion scanning with clinical criteria. Our perfusion scanning criterion states that high probability defines 2 segmental perfusion defects without corresponding radiographic abnormality and indeterminate probability defines less than 2 segmental perfusion defects (low probability: less than one segmental perfusion defect; intermediate: perfusion defects between high and low probability). The clinical criterion is divided into 7 items related to symptoms and signs suggestive of pulmonary thromboembolism. More than 4 items are defined as a highly suspicious clinical manifestation (HSCM), and less than 4 are considered a low suspicious clinical manifestation (LSCM). In 31 cases of high probability, 18 of HSCM did not include pulmonary angiography (PAG), and 13 of LSCM included PAG (positive: 11; negative: 2). In 12 cases of indeterminate probability, 7 of LSCM were observed without PAG and 5 of HSCM with PAG (positive: 4; negative: 1). PAG performance thus decreased to 41.9%. The positive prediction of high probability is 93.5%, which is very high, compared to indeterminate probability at 33.3%.

View full abstract