© 2004 American Thoracic Society
Labeling the ThrombusThe Future of Nuclear Medicine for Venous Thromboembolism?Geneva University Hospital Geneva, Switzerland In this issue of the Journal (pp. 987–993), Morris and coworkers (1) report the results of an experimental study on a new nuclear imaging technique to diagnose pulmonary embolism and deep venous thrombosis in an established canine model of acute pulmonary embolism. They induced the formation of bilateral thrombi by placing a double-balloon catheter in the femoral veins of five dogs and injecting between the balloons a mix of human fibrinogen (because the humanized murine anti-fibrin antibody they selected had a poor reactivity with canine fibrin) and thrombin to initiate clotting. After 3 hours they confirmed the presence of a thrombus by venography and provoked the embolization of one of the thrombi by deflating the balloons in one hind leg of the animal and inducing passive motion. They then imaged the clots using single photon emission computed tomography (SPECT) by the intravenous injection of a 99mTc-labeled humanized murine anti-fibrin antibody. The best quality images were obtained 4 hours after antibody injection and allowed the correct identification and localization of the pulmonary embolus in four of the five dogs and of the femoral vein thrombosis in three of the five animals. There were no false positives and the technique detected all clots weighing more than 0.4 g, which, according to the authors, is a detection threshold lower than that of even the most sensitive techniques (such as pulmonary angiography in humans). Although their claim of a 100% sensitivity and specificity of this new imaging method should not be taken too literally because of the small number of animals and the experimental nature of the study, the results are clearly encouraging. What then is novel in this approach? Labeling the clot with a radioisotope-labeled element of the clot or an antibody reacting with one of the clot components has already been tried in numerous studies. 125Iodine-labeled fibrinogen scintigraphy was a popular method for detecting deep vein thrombosis in the '80s and was used as a diagnostic criterion in the only randomized trial that compared anticoagulant treatment with placebo in distal deep vein thrombosis (2). It was progressively abandoned, however, because of a low sensitivity and specificity (3) and replaced by venous compression ultrasonography. Since then, investigators have evaluated various radiolabeled clot elements or anti-fibrin antibodies to image deep vein thrombosis and/or pulmonary embolism: 123iodine-tissue plasminogen activator (4), indium-111–labeled anti-fibrin antibody (5, 6), and, more recently, an antibody directed against the ß-chain of fibrin labeled with 123iodine (7). The results, however, were uniformly disappointing, probably because the fibrin epitope targeted by the antibody was available for antibody binding only during the active phase of fibrin formation. Thus, there was little binding to constituted clots, a major drawback especially for pulmonary emboli, and therefore the clot-to-blood signal ratios were low. In their study, Morris and coworkers (1) used a novel antibody directed against the D-dimer region of crosslinked fibrin, which remains accessible even after thrombus formation has been stopped by anticoagulant treatment. That target selection resulted in high signal-to-noise ratios in the range of 10 to 20 and clear identification of the leg and lung thrombi in the experimental animals. Because of the amount of signal originating from the blood vessels, however, this would not have allowed a sufficient spatial resolution with conventional planar imaging. To circumvent that difficulty, Morris and coworkers (1) used the SPECT technique, a method that uses radionuclides that emit a single photon of a given energy and can provide transaxial, sagittal, and coronal reconstructions from the 3-dimensional distribution of radionuclides in the organ. Ventilation–perfusion SPECT-scintigraphy has already been compared with planar imaging both in animal models (8) and in human subjects (9). Indeed, the main weakness of conventional ventilation–perfusion scintigraphy is the high rate of nondiagnostic results (50 to 70% depending on the population studied) (10), and enhanced spatial resolution by SPECT-scintigraphy has the potential for improving the discrimination between nonspecific perfusion defects secondary to pulmonary arterial hypoxic vasoconstriction and those caused by pulmonary embolism. A recent small series showed a higher specificity of SPECT ventilation–perfusion scans for pulmonary embolism in comparison with planar imaging (9). Nevertheless, the combination of clot labeling and SPECT imaging presented by Morris and coworkers (1) is clearly original. Will that approach prove fruitful in the clinical setting? Dutch investigators have proposed that the steps for developing a new diagnostic test should be as clearly identified as those for therapeutic agents (11). In that framework, the elaboration of the technical aspects of a new test corresponds to a phase I study, and the technique described by Morris and coworkers (1) is clearly at that stage. Further validation requires phase II studies in which the new test's characteristics are compared with an accepted diagnostic criterion (the previously called "gold standard") to define its sensitivity and specificity, studies that should adhere to rigorous methodological criteria (12). In phase III studies, the new test is used for clinical decision-making in a real-life setting and the outcomes are compared with those of patients managed according to the reference diagnostic test. Is labeled anti-fibrin antibody SPECT imaging (1) worth all the effort? Several combinations of clinical assessment, plasma D-dimer measurement, lower limb venous compression ultrasonography, and either conventional ventilation–perfusion scintigraphy or helical CT scan have been recently validated in large-scale outcome studies (13–15) and shown cost-effective for diagnosing both deep venous thrombosis and pulmonary embolism. Such algorithms limit the requirement for phlebography and pulmonary angiography to a small fraction of patients suspected of the disease. Because they are based on morphologic criteria, however, traditional imaging instruments have distinct limitations for diagnosing recurrent thromboembolic events because leg or pulmonary thrombi do not resolve in about 30% of patients and there are no validated criteria to distinguish between residual and new clots. Recurrent venous thromboembolism after interruption of the initial course of anticoagulants is a frequent event in patients with an unprovoked first acute event (around 8% per year [16]). This is certainly the most promising indication for techniques aimed at imaging the thrombus, such as radiolabeled anti-fibrin antibody SPECT scintigraphy (1). FOOTNOTES Conflict of Interest Statement: A.P. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. REFERENCES
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
