Bookmarkers beware: Bookmarks to pages other than the home page may not work after we change our server in April Editor's Choice | This Week in BMJ | Press releases BMJ No 7130 Volume 316 Editorial Saturday 14 February 1998 Imaging pulmonary embolism A new look with spiral computed tomography David M Hansell, Christopher D R Flower Pulmonary embolism is a common condition that often escapes diagnosis in life. The mortality attributable to pulmonary embolism depends on many factors but can be up to 30% in untreated patients, more than 10 times the annual mortality for patients treated with anticoagulant drugs (2.5%).(1) Balanced against the danger of underdiagnosis are the risks of treatment with anticoagulants.(2) Investigation is often haphazard and treatment empirical, reflecting the relative insensitivity and poor specificity of traditional non-invasive diagnostic tests such as ventilation/perfusion (V/Q) scintigraphy.(3) In an attempt to improve diagnostic precision, many studies have recommended combinations of tests.(4-6) Pulmonary arteriography is justly regarded as the final arbiter, but it is underused and is available in only a third of hospitals in the United Kingdom.(7) Spiral computed tomography images the pulmonary vessels directly. It is non-invasive and increasingly available. During a scan, which is performed in a single breath hold, intravenous contrast medium is given, and emboli are identified as filling defects within the opacified vessels down to a segmental level (figure). Less than 10% of these studies are inconclusive, compared with 50% of V/Q scans.(8,9) In addition, spiral computed tomography may reveal disease in the lungs, mediastinum, or pleura that mimics the symptoms of pulmonary embolism. Evidence supporting its accuracy has been thoroughly reviewed in the radiological literature(9,10); in contrast, the value and limitations of this technique have received little attention in the clinical literature.(11) Spiral computed tomography showing pulmonary emboli (arrows) within the opacified pulmonary arteries Compared with pulmonary arteriography, spiral computed tomography has a mean sensitivity of 90% for the detection of emboli in segmental and larger pulmonary arteries (range 75-100%(12,13)) and a mean specificity of 92% (range 76-100%(8)(12)). Within these apparently impressive figures is the important caveat that spiral computed tomography is currently unable to detect emboli reliably within subsegmental pulmonary arteries, although refinements in technique may allow the detection of these small emboli.(14) The prevalence of pulmonary emboli confined to the subsegmental arteries is small (probably in the region of 6%(15)), and their importance is unknown. However, although subsegmental embolism may have a negligible effect in otherwise normal individuals, the consequences may be serious in patients with pre-existing cardiopulmonary disease. The place of spiral computed tomography in the hierarchy of tests available for diagnosing pulmonary embolism has not been finalised. The danger is that it will be merely added to already unwieldy diagnostic algorithms. Goodman and Lipchik propose, radically, that spiral computed tomography should replace V/Q scintigraphy completely and that it should be the first test in patients with symptoms of pulmonary embolism.(16) For patients with symptoms of deep vein thrombosis and pulmonary embolism, ultrasound scanning of the lower limbs should be the first test. Pulmonary arteriography should be reserved for those patients in whom both computed tomography and ultrasound are negative and in whom clinical suspicion remains high or for patients who have limited cardiopulmonary reserve. This protocol has an appealing clarity but assumes that high quality spiral computed tomography and lower limb ultrasonography are readily available. Furthermore, spiral computed tomography exposes the patient to a relatively high dose of radiation. A practical modification is to take account of the pretest probability of pulmonary embolism, the appearance of the chest radiograph, and the high specificity of a normal perfusion scintigram. Patients with radiographic abnormalities (likely to result in an indeterminate V/Q scan) and a high pretest probability of pulmonary embolism can be investigated initially with spiral computed tomography; patients with a normal chest radiograph and a low pretest probability can be investigated with a perfusion scintigram. Most of the former will be inpatients, and most of the latter will be outpatients. Only one published study has investigated the cost effectiveness of spiral computed tomography for diagnosing pulmonary embolism.(17) In an analysis of 15 combinations of diagnostic tests (spiral computed tomography, lower limb ultrasonography, V/Q scintigraphy, pulmonary arteriography, and d-dimer plasma levels), the authors found that the five most effective strategies (least mortality at three months and lowest associated costs per life saved) all included spiral computed tomography, usually in combination with lower limb ultrasonography. In the clinical literature a study of the cost effectiveness of various other diagnostic strategies suggested that a combination of V/Q scintigraphy, serial impedance plethysmography, and pulmonary arteriography was the most cost effective; nowhere in the discussion was spiral computed tomography mentioned as an alternative or adjunctive test.(18) The diagnostic options are constantly shifting and, although most studies currently favour spiral computed tomography for identifying pulmonary embolism, magnetic resonance imaging is gaining favour(19) and may yet achieve parity with spiral computed tomography.(20) Meanwhile, selecting the most accurate and appropriate test for diagnosing pulmonary embolism remains the objective. David M Hansell Consultant radiologist Department of Radiology, Royal Brompton Hospital, London SW3 6NP Christopher D R Flower Consultant radiologist Department of Radiology, Addenbrooke's Hospital, Cambridge CB2 2QQ References 1 Carson J L, Kelley M A, Duff A, Weg J G, Fulkerson W J, Palevesky M, et al. The clinical course of pulmonary embolism. N Engl J Med 1992;326:1240-5. 2 Ginsberg J S. Management of venous thromboembolism. N Engl J Med 1996;335:1816-28. 3 Schluger N, Henschke C, King T, Russo R, Binkert B, Rackson M, et al. Diagnosis of pulmonary embolism at a large teaching hospital. J Thorac Imag 1994;9:180-4. 4 Kelley M A, Carson J L, Palevsky H I, Schwartz J S. Diagnosing pulmonary embolism: new facts and strategies. Ann Intern Med 1991;114:300-6. 5 Hull R D, Raskob G E, Coates G, Panju A A, Gill G J. A new non-invasive management strategy for patients with suspected pulmonary embolism. Arch Intern Med 1989;149:2549-55. 6 Hyers T H. Diagnosis of pulmonary embolism. Thorax 1995;50:930-2. 7 Cooper T J, Hayward M W, Hartog M. Survey on the use of pulmonary scintigraphy and angiography for suspected pulmonary thromboembolism in the UK. Clin Radiol 1991;43:243-5. 8 Remy-Jardin M, Remy J, Deschildre F, Artaud D, Borgi J P, Hossein-Foucher C, et al. Diagnosis of pulmonary embolism with spiral CT: comparison with pulmonary angiography and scintigraphy. Radiology 1996;200:699-706. 9 Hansell D M. Spiral computed tomography and pulmonary embolism: current state. Clin Radiol 1997;52:575-81. 10 Goodman L R, Lipchik R J, Kuzo R S. Acute pulmonary embolism: the role of computed tomographic imaging. J Thorac Imag 1997;12:83-6. 11 Fennerty T. The diagnosis of pulmonary embolism. BMJ 1997;314:425-9. 12 Teigen C L, Maus T P, Sheedy P F, Stanson A W, Johnson C M, Brear J, et al. Pulmonary embolism: diagnosis with contrast-enhanced electron-beam CT and comparison with pulmonary angiography. Radiology 1995;194:313-9. 13 Remy-Jardin M, Remy J, Wattinne L, Giraud F. Central pulmonary thromboembolism: diagnosis with spiral volumetric CT with the single breath-hold technique-comparison with pulmonary angiography. Radiology 1992;185:381-7. 14 Remy-Jardin M, Remy J, Artaud D, Deschildre F, Duhamel A. Peripheral pulmonary arteries: optimization of the spiral CT acquisition protocol. Radiology 1997;204:157-63. 15 The PIOPED investigators. Value of ventilation/perfusion scan in acute pulmonary embolism. JAMA 1990;263:2753-9. 16 Goodman L R, Lipchik R J. Diagnosis of acute pulmonary embolism: time for a new approach. Radiology 1996;199:25-7. 17 Van Erkel A R, van Rossum A B, Bloem J L, Kievit J, Pattynama P M T. Spiral CT angiography for suspected pulmonary embolism: a cost-effectiveness analysis. Radiology 1996;201:29-36. 18 Hull R D, Feldstein W, Stein P D, Pineo G F. Cost-effectiveness of pulmonary embolism diagnosis. Arch Intern Med 1996;156:68-72. 19 Meaney J F M, Weg J G, Chenevert T L, Stafford-Johnson D, Hamilton BH, Prince MR. Diagnosis of pulmonary embolism with magnetic resonance angiography. N Engl J Med 1997;336:1422-7. 20 Gefter W B, Hatabu H, Holland G A, Gupta K B, Henschke C I, Palevsky H I. Pulmonary thromboembolism: recent developments in diagnosis with CT and MR imaging. Radiology 1995;197:561-74. Home | Current issue | Past issues | Classified ads | Career Focus | Feedback Collections | About this site | About the BMJ | BMA | Medline