Detection of Pulmonary Embolism by D-Dimer Assay, Spiral Computed Tomography, and Magnetic Resonance Imaging Julia H. Indik and Joseph S. Alpert Pulmonary embolism (PE) remains difficult to diag- nose. Ventilation perfusion lung scan, the standard diagnostic test for PE, has poor overall sensitivity. The gold standard examination, pulmonary angiog- raphy, is invasive and has some risk, making clini- cians reluctant to refer patients. In recent years, new diagnostic modalities have been investigated, includ- ing D-Dimer assays, spiral computed tomography (CT), and magnetic resonance imaging (MRI). The authors reviewed the literature and noted that the D-Dimer assays by ELISA or rapid ELISA design are approximately 90% to 95% sensitive, but are not specific for the diagnosis of pulmonary embolism. Spiral CT has been studied with conflicting results; however, in the largest studies the reported sensitivi- ties are greater than approximately 85%. Electron beam CT is an alternative technique, which has not been as extensively studied. MRI is also useful for imaging the pulmonary arterial vasculature, but remains experimental. Although a more accurate assessment of the sensitivity of these new modali- ties will need to wait until a large angiographically controlled study, such as the planned PIOPED II, can be done, D-Dimer assay and spiral CT are often useful in the detection of pulmonary embolism. The authors make recommendations for their use in a diagnostic algorithm, as alternatives to the standard ventilation perfusion lung scan. Copyright  2000 by W.B. Saunders Company I n the United States, pulmonary embolism (PE) and deep venous thrombosis (DVT) are diag- nosed in 300,000–600,000 people each year, leading to as many as 50,000 deaths. 1 Left un- treated,theriskofdeathfromPEisabout30%. 2,3 Symptoms that are suggestive of PE include suddenonsetofdyspnea,pleuriticchestpain,and hemoptysis. Patients also may be anxious. The physical examination may be notable only for tachypnea and tachycardia; evidence of right ventricular failure may be present if PE is severe. Initialtestssuchasarterialbloodgases,electrocar- diography, and chest radiograph may indicate the severity of PE, but are often normal. The clinical suspicionforPEisraisedifriskfactorsforvenous thrombosis are present. A patient may have a hypercoagulable state, either inherited, such as the factor V Leiden mutation, or acquired, as can occur with malignancy or medications such as estrogen.Otherrisksincludesurgery,immobiliza- tion,sepsis,trauma,burns,andsevereventricular dysfunction. The Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) investigators 4 looked at the accuracy of the ventilation-perfu- sion (V/Q) scan in diagnosing PE as compared withtheacceptedgoldstandard,pulmonaryangi- ography. Pulmonary embolism was present in 88% of patients with a high probability scan. If clinical suspicion was high, the likelihood of PE increased to 96%. Only 41% of patients with PE, however, had a high probability scan. Alterna- tively, if the scan was low probability, the likeli- hood of PE was 14%, but decreased to 4% if the clinical suspicion was low. Thus, a diagnosis of reasonablecertaintycanbemadeforpatientswith high-clinicalsuspicionandhighprobabilityscans or else low clinical suspicion and low probability From the Department of Medicine, University of Arizona Health Sciences Center, Tucson, AZ. Address reprint requests to Joseph S. Alpert, MD, Department of Medicine, Arizona Health Sciences Center, 1501 N Campbell Ave, Tucson, AZ 85724-5035. Copyright  2000 by W.B. Saunders Company 0033-0620/00/4204-0003$10.00/0 Progress in Cardiovascular Diseases, Vol. 42, No. 4 (January/February), 2000: pp 261-272 261