| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
PLENARY SESSION |
1 From the Department of Radiology, Medical College of Wisconsin, 9200 W Wisconsin Ave, Milwaukee, WI 53226-3596. From the Plenary Session, Friday Imaging Symposium: Algorithmic Controversies, at the 1999 RSNA scientific assembly. Received January 28, 2000; revision requested February 23 and received March 3; accepted March 8. Address correspondence to the author (e-mail: lgoodman@mcw.edu).
Index Terms: Embolism, pulmonary, 60.721 • Pulmonary angiography, 60.12116 • Veins, extremities, 935.751 • Veins, thrombosis, 935.751
Learning Objective
Discuss the relative merits of various imaging modalities in the diagnosis of pulmonary embolism and deep venous thrombosis.
Introduction
The use of helical computed tomographic (CT) pulmonary angiography followed by axial CT of the inferior vena cava and the lower-extremity veins to the knee is a powerful alternative to conventional imaging in thromboembolic disease. In addition to having high sensitivity and specificity, this combination frequently provides an alternative diagnosis when the patient's signs and symptoms are not due to pulmonary emboli.
This article discusses and illustrates the use of a variety of imaging modalities in the detection of pulmonary embolism with emphasis on CT.
Diagnosis of Thromboembolic Disease
Thromboembolic disease is difficult to diagnose. Only 25%–30% of patients with suspected pulmonary embolism prove to have the disease at imaging, and in the majority of patients with pulmonary embolism, the disease is never suspected clinically (1,2). In a review of 21,000 autopsies, fatal pulmonary embolism was diagnosed in 9% of cases; however, an antemortem diagnosis of pulmonary embolism was made in only 18% (3). Similarly, evidence of deep venous thrombosis is seen at imaging in only half of patients with proved pulmonary embolism. In the 70%–75% of patients in whom thromboembolism is excluded, an alternative source for the patient's signs and symptoms must be found (4–7).
Ventilation-Perfusion Scanning and Pulmonary Angiography
In 1990, the PIOPED [Prospective Investigation of Pulmonary Embolism Diagnosis] Trial, a multi-institutional study of ventilation-perfusion scanning and pulmonary angiography, found that a normal ventilation-perfusion scan virtually excluded pulmonary embolism and that a high-probability scan was virtually diagnostic for the disease (7). Unfortunately, three-fourths of patients had neither high-probability nor normal scans and required additional imaging such as pulmonary angiography or lower-extremity venous studies. In this inconclusive group, 22% had pulmonary embolism. Recent improvements in scintigraphy, such as the revised PIOPED criteria for interpreting ventilation-perfusion scans and the use of technetium-99m pyrophosphate aerosol, have decreased the number of low-probability and indeterminate ventilation-perfusion scans at our institution to approximately 50% (8), which is still an unacceptably high percentage. Especially troubling is the PIOPED finding that in patients with underlying cardiopulmonary disease, and in particular chronic obstructive pulmonary disease, the likelihood of a definitive scintigraphic diagnosis decreases to 23% and 10%, respectively (9).
Pulmonary angiography is the recommended study in patients with inconclusive scintigraphic findings but is used infrequently. Therefore, the majority of patients undergo anticoagulation therapy or simple observation with no definitive diagnosis being made (10,11).
Ultrasonography
Doppler and compression ultrasonography (US) of the lower extremities has proved to be a reliable alternative to venography. When venous US findings are positive, anticoagulation therapy is appropriate and imaging of the lungs for pulmonary embolism becomes optional. Unfortunately, approximately one-half of patients with proved pulmonary emboli do not have evidence of deep venous thrombosis at US. Thus, a negative US scan does not exclude pulmonary embolism. Some authors have advocated that patients with a low-probability ventilation-perfusion scan, a low clinical probability of pulmonary embolism, and a negative US scan be followed up with clinical evaluation and serial US (12,13).
Helical CT Pulmonary Angiography and Axial Lower-Extremity CT
Like conventional angiography, helical CT pulmonary angiography provides direct visualization of the blood clot rather than indirect evidence. Multiple studies from the early 1990s that were performed with 5-mm axial sections, a 1:1 pitch, a 24–30-second breath hold, and overlapping reconstructions showed a sensitivity and specificity of approximately 90% for central vessel emboli. However, the majority of subsegmental vessels were not visualized with that technique, and most subsegmental pulmonary emboli were not diagnosed (5,14). It is difficult to know how frequently pulmonary embolism is limited to the subsegmental arteries and whether all subsegmental emboli require treatment. Best estimates indicate that clots limited to the subsegmental vessels are found in 10%–15% of patients with suspected pulmonary emboli (5,7,15).
There has been considerable improvement in helical CT equipment over the past decade. Helical scanners are now faster, allowing images to be obtained with thinner sections and shorter scan times. The new multidetector scanners are considerably faster, allowing thin-section (1.25-mm) helical CT pulmonary angiography to be performed during a shorter breath hold (15–17 seconds). The segmental and subsegmental pulmonary vessels are better demonstrated and easier to interpret with this technique (Fig 1). Remy-Jardin et al (16) demonstrated that decreasing the section thickness from 3 mm to 2 mm and the scan time from 1 second to .75 seconds increases the percentage of visible subsegmental vessels from 37% to 60%. Multidetector scanning with a section thickness of 1.25 mm displays almost 70% of subsegmental vessels (Figs 2a, 3a–3c) (17). Thinner collimation also improves interobserver agreement in the evaluation of subsegmental vessels. The faster scan time also diminishes respiratory motion, decreasing the number of suboptimal CT studies (18).
|
|
|
|
|
|
|
Another major advantage of using helical CT pulmonary angiography that has not been adequately stressed is the ability to make alternative diagnoses in the 70% of patients who do not have pulmonary embolism or additional diagnoses in those who do have the disease (Fig 3b). Although estimates vary widely, a conservatively estimated 20%–30% of patients with pulmonary embolism have other clinically significant findings at helical CT pulmonary angiography (4,21,22). In a recent editorial, Rogers (23) stressed the expanding role of helical CT pulmonary angiography in many diseases: "We have gone from potentially supporting the clinician in a diagnosis, to showing the clinician, with reasonable certainty, what's actually wrong...a decided, genuine, real upgrading in our capabilities."
Diagnostic Algorithm
The foregoing considerations have led us to adopt the following diagnostic algorithm:
1. Patients with normal chest radiographic findings are evaluated with a perfusion scan and, if necessary, an aerosol ventilation scan. In this group of patients, the likelihood of a definitive scintigraphic diagnosis is high and the radiation dose is relatively low. Patients with normal or very low probability scintigraphic findings are presumed not to have pulmonary emboli and rarely undergo further testing or anticoagulation therapy. Patients with a high-probability scan usually undergo anticoagulation therapy. All other patients should be evaluated with helical CT pulmonary angiography, conventional pulmonary angiography, or lower-extremity US, depending on the clinical situation (5).
2. Patients with abnormal chest radiographic findings, especially chronic obstructive pulmonary disease, are unlikely to have definitive scintigraphic findings. These patients undergo helical CT pulmonary angiography as well as axial CT of the inferior vena cava and the iliac, femoral, and popliteal veins. If the findings at helical CT pulmonary angiography are equivocal or technically inadequate (5%–10% of cases) or clinical suspicion remains high despite negative findings, additional imaging is required.
3. Patients who have symptoms of deep venous thrombosis but not of pulmonary embolism initially undergo US, which is a less expensive alternative. If the findings are negative, imaging is usually discontinued; if they are positive, the patient is evaluated for pulmonary embolism at the discretion of the referring physician.
We have recently investigated the effects of using this algorithm on ventilation-perfusion scan interpretation over selected 6-month periods. In 1994, prior to the introduction of helical CT pulmonary angiography for pulmonary embolism, 129 of 269 scintigrams (48%) were definitive (ie, normal, very low probability, or high-probability). After the introduction of CT pulmonary angiography, the percentage of definitive scintigrams rose to 63% (107 of 170) in 1997 and 70% (149 of 213) in 1999 (Chen H, Liu Y, Goodman LR, unpublished data, 1999).
If some subsegmental pulmonary embolism is missed at helical CT pulmonary angiography and the patient does not undergo anticoagulation therapy, how often does he or she return with clinical evidence of new pulmonary embolism? Over a 25-month period, we closely monitored each of 812 untreated patients for 3 months following negative helical CT pulmonary angiography (n = 285) or negative or low-probability ventilation-perfusion scanning (n = 527) to determine the prevalence of subsequent clinically apparent pulmonary embolism. Eighty-seven of the patients who underwent CT (31%) and 177 of the patients who underwent ventilation-perfusion scanning (34%) were lost to follow-up or were undergoing anticoagulation therapy for other reasons, leaving 198 patients with negative CT findings and 350 patients with negative or low-probability scintigraphic findings. In the latter group, 188 patients had negative ventilation-perfusion scans and 162 had low-probability scans. Subsequent pulmonary emboli were found in two (1%) of the 198 patients who had undergone CT and in five (1.4%) of the 350 patients who had undergone scintigraphy, all five of whom had low-probability scans (24). Two recent studies have demonstrated almost identical findings (25,26). This indicates that although some small emboli may be missed at helical CT pulmonary angiography, it does not appear that the subsequent morbidity from pulmonary embolism is high or that the rate of subsequent pulmonary embolism diagnosis is higher with helical CT than with negative or low-probability scintigraphy.
Conclusions
CT is a practical, efficient alternative to conventional imaging in pulmonary emboli. The results can only improve as helical and multidetector CT equipment improves. The addition of CT venography further enhances the value of CT in thromboembolic disease. The triage of patients between scintigraphy and helical CT pulmonary angiography on the basis of chest radiographic findings more evenly distributes the demand on resources and improves the chances for definitive scintigraphy at a lower radiation dose and for definitive diagnosis of pulmonary embolism or deep venous thrombosis in patients with underlying cardiopulmonary disease. In addition, CT often provides alternative explanations for cardiopulmonary symptoms when pulmonary embolism is not present.
Although technical aspects and potential pitfalls of helical CT pulmonary angiography are not discussed in this article, the reader is referred to recent articles by Remy-Jardin et al (27) and Kuzo and Goodman (28).
Footnotes
Abbreviation: PIOPED = Prospective Investigation of Pulmonary Embolism Diagnosis
References
This article has been cited by other articles:
|
|
 |
|
  J. P. Kanne and T. A. Lalani Role of Computed Tomography and Magnetic Resonance Imaging for Deep Venous Thrombosis and Pulmonary Embolism Circulation, March 30, 2004; 109(12_suppl_1): I-15 - I-21. [Abstract] [Full Text]
|
|
|
|
 |
|
 M. E. Schuster, J. E. Fishman, J. F. Copeland, H. Hatabu, and P. M. Boiselle Pulmonary Embolism in Pregnant Patients: A Survey of Practices and Policies for CT Pulmonary Angiography Am. J. Roentgenol., December 1, 2003; 181(6): 1495 - 1498. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 A. J. Ravin, R. P. Edwards, M. A. Krohn, J. R. Kelley, W. A. Christopherson, and J. M. Roberts The Factor V Leiden Mutation and the Risk of Venous Thromboembolism in Gynecologic Oncology Patients Obstet. Gynecol., December 1, 2002; 100(6): 1285 - 1289. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. S. Katz, P. A. Loud, D. Bruce, A. M. Gittleman, R. Mueller, D. L. Klippenstein, and Z. D. Grossman Combined CT Venography and Pulmonary Angiography: A Comprehensive Review RadioGraphics, October 1, 2002; 22(90001): S3 - 19. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
J. R. Mayo, L. H. Ketai, P. A. Loud, Z. D. Grossman, D. S. Katz, and A. M. Gittleman Invited Commentary * Authors' Response RadioGraphics, October 1, 2002; 22(90001): S20 - 24. [Full Text] [PDF]
|
|
|
|
 |
|
 P. A. Loud, D. S. Katz, D. A. Bruce, D. L. Klippenstein, and Z. D. Grossman Deep Venous Thrombosis with Suspected Pulmonary Embolism: Detection with Combined CT Venography and Pulmonary Angiography Radiology, May 1, 2001; 219(2): 498 - 502. [Abstract] [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOGRAPHICS | RADIOLOGY | RSNA JOURNALS ONLINE |
