Deep vein thrombosis and pulmonary embolism are a major health problem in the United States. Due to their nonspecific clinical manifestations and the risks associated with effective therapy, imaging tests have a crucial role in the management of patients suspected of having venous thromboembolism. It now is often necessary to perform multiple imaging studies to evaluate such patients fully, although information sufficient to begin therapy often can be gained from more limited testing. Imaging procedures In common use, including ascending venography, venous ultrasound, ventilation-perfusion scintigraphy and pulmonary arteriography, all have limitations which Include complications, Imperfect accuracy and limitation to specific anatomic regions. Preliminary reports suggest that magnetic resonance (MR) imaging may have advantages for diagnosing deep vein thrombosis and pulmonary embolism. The goal of this project is to extend the promising preliminary experience with MR venous Imaging to a prospective clinical trial, and to further develop and evaluate clinically MR methods for detecting pulmonary emboli. The hypotheses to be tested are that MR imaging with current technique is accurate for detecting deep vein thrombosis; that MR imaging with further development will detect pulmonary emboli accurately within main, lobar and segmental pulmonary arteries; and that these methods can be combined when necessary into a single clinical examination which can screen the caudal veins for thrombi and the lungs for emboli. Accordingly, the specific aims of this project are to test prospectively the accuracy of MR venous imaging; to further develop MR methods for imaging pulmonary emboli and test their accuracy; and to assess the feasibility and effectiveness of combined MR venous and pulmonary imaging. MR venous imaging will be performed with a standard technique in patients who have ascending venography requested clinically. Further development of MR pulmonary imaging techniques will include methods which accentuate and those which eliminate the MR signal from flowing blood. Studies in phantoms and a dog model will evaluate clot: blood contrast achieved and degree of motion artifact. Pulmonary MR imaging, when developed suitably, will be performed in patients who have pulmonary arteriography requested clinically. The sensitivity and specificity of MR for detecting deep vein thrombosis and pulmonary embolism will be determined by comparison with the corresponding contrast study and by outcome analysis. Finally, both venous and pulmonary MR imaging will be performed in patients and the accuracy and cost of MR compared with the studies obtained clinically. If successful, this work could lead to more accurate and in some cases less expensive diagnosis of venous thromboembolism.
