(Radiographics. 2002;22:S3-S19.)
© RSNA, 2002
Combined CT Venography and Pulmonary Angiography: A Comprehensive Review1
Douglas S. Katz, MD,
Peter A. Loud, MD,
Dennis Bruce, MD,
Adam M. Gittleman, MD,
Richard Mueller, LRT,
Donald L. Klippenstein, MD and
Zachary D. Grossman, MD
1 From the Department of Radiology, Winthrop-University Hospital, 259 First St, Mineola, NY 11501 (D.S.K., D.B., A.M.G., R.M.); and the Department of Radiology, Roswell Park Cancer Institute, Buffalo, NY (P.A.L., D.L.K., Z.D.G.). Presented as an education exhibit at the 2001 RSNA scientific assembly. Received February 6, 2002; revision requested March 26 and received May 7; accepted May 15. Address correspondence to D.S.K. (e-mail: dsk2928@pol.net).
 |
Abstract
|
|---|
The combination of computed tomographic (CT) venography and pulmonary angiography (CTVPA) was initially described in 1998 as a single comprehensive noninvasive imaging examination for suspected thromboembolic disease. It allowed the identification of pulmonary embolism as well as deep venous thrombosis (DVT) in the abdomen, pelvis, thighs, and calves. The venographic portion of CTVPA has now been studied by multiple researchers and has been shown to be an accurate imaging study for the thigh veins in comparison with lower extremity sonography. In contrast to sonography, however, CTVPA readily and rapidly permits evaluation of the inferior vena cava, the pelvic veins, the calf veins, and all of the superficial venous system. Complex venous anatomy can be surveyed, an additional sonographic study is not required, and only a few extra minutes and images are required over and above CT pulmonary angiography. A review of 957 recent cases of suspected pulmonary embolism examined with CTVPA revealed an overall 10.5% frequency of DVT, with a nearly equal distribution of thrombosis at the common femoral, superficial femoral, popliteal, and deep calf veins. Although a variety of protocols for CTVPA may be implemented, including a contiguous helical acquisition, obtaining 5- or 10-mm-thick images every 4 cm provides a high degree of accuracy and decreases overall radiation dose.
© RSNA, 2002
Index Terms: Computed tomography (CT), angiography, 93.12916, 98.12916 • Embolism, pulmonary, 60.72 • Thrombosis, CT, 93.12916, 98.12916 • Veins, CT, 93.12916, 98.12916 • Veins, thrombosis, 93.751, 98.751
|
LEARNING OBJECTIVES FOR TEST 1
|
|---|
After reading this article and taking the test, the reader will be able to:
- Describe the rationale for combined CT venography and pulmonary angiography.
- Comprehensively discuss the literature on combined CT venography and pulmonary angiography.
- Implement a protocol for CT venography and pulmonary angiography, as well as interpret the CT venographic portion of the study.
|
Introduction
|
|---|
In 1998, a combined technique of CT venography and CT pulmonary angiography (CTVPA) was initially described. Since then, several groups of investigators in the United States and Europe have reported high accuracy of the technique (compared with sonography) for the diagnosis or exclusion of deep venous thrombosis (DVT) in the inguinal regions and thighs. CTVPA provides a noninvasive method for imaging suspected embolic disease in the pulmonary arteries, as well as a comprehensive study of the venous system for DVT; the inferior vena cava, the pelvic veins, and the superficial and deep veins of the lower extremities, including the calves, are all routinely evaluated. In this article, we review the technical aspects of, rationale for, advantages of, and utility of CTVPA. We also demonstrate the broad spectrum of findings—venous and extravenous—that can be discovered in the CT venographic phase of the examination, based on our experience with the technique at our two institutions. The frequency and distribution of DVT found during the venous phase of almost 1,000 CTVPA studies recently performed at our institutions are discussed, as well as the significance of these findings in light of related controversies in the sonography literature. Emphasis is placed on DVT detection in areas that may be difficult to evaluate sonographically, including the deep femoral veins. Interpretive pitfalls also are illustrated and reviewed.
|
Rationale for CTVPA
|
|---|
Relationship of Pulmonary Embolism and DVT
Pulmonary embolism (PE) and deep venous thrombosis (DVT) are part of the same complex and common process (ie, venous thromboembolism), which has a wide spectrum of clinical presentations and consequences, ranging from no symptoms or sequelae to a lethal outcome (1). DVT usually starts in the deep calf veins and propagates superiorly, and it is associated with symptomatic PE in about one-third of untreated patients (2). DVT is commonly confused with other conditions such as cellulitis, leg edema, and chronic venous insufficiency, and PE is also notoriously difficult to diagnose (2).
According to Wolfe and Hartsell (3), the most important question may not be who has a PE, but who will have another embolism that may be life-threatening. A single episode of hemodynamically stable PE that may be undetectable by noninvasive means may be of less concern than a large, unknown proximal DVT that has not yet embolized (3).
Unfortunately, it has been estimated that half of the patients with proved PE do not have evidence of DVT at lower extremity sonography, so negative sonographic findings do not exclude PE (1,4) nor do positive sonographic findings reveal the extent of PE within the thorax. Some authors have therefore advocated the use of lower extremity sonography after noninvasive imaging studies of the thorax have produced negative findings in patients with suspected PE, as it is "reassuring to know that no large clots are lurking in the lower extremity veins" (4).
Role of CT Pulmonary Angiography
In the past decade, CT pulmonary angiography has been increasingly used as the primary diagnostic imaging examination after chest radiography for the evaluation of patients with suspected PE. CT pulmonary angiography has numerous advantages over both ventilation-perfusion scintigraphy and conventional pulmonary angiography. In patients without a contraindication for iodinated intravenous contrast medium, this readily available, rapid, and noninvasive study is well tolerated, allows direct demonstration of clot in the thorax, is much more accurate than scintigraphy, and reveals significant additional or alternative diagnoses (5–8).
Although the most important concern regarding the routine use of CT pulmonary angiography remains its accuracy at the subsegmental pulmonary arterial level, the significance of subsegmental PE continues to be controversial (7,8). However, in two recent series of patients with negative CT pulmonary angiographic studies who were followed up for several months, there were no deaths from subsequent venous thromboembolism (7,8). In addition, with the advent of multidetector helical CT scanners and the routine use of thoracic images as thin as 1–2 mm, even the subsegmental pulmonary arteries can now be evaluated routinely (9–11), and presumably subsegmental emboli can be more accurately detected (12).
Rationale for the Addition of CT Venography to CT Pulmonary Angiography
Unfortunately, even with state-of-the-art equipment and techniques, a relatively small percentage of CT pulmonary angiographic studies are still technically suboptimal, in the range of 3%–10% (6,13,14). Lower extremity sonography has high accuracy in symptomatic patients, involves no contrast medium or ionizing radiation, and is portable. However, sonography is also operator-dependent; is limited above the inguinal ligament, below the knee, and occasionally in the adductor canal; and is less accurate in asymptomatic patients (15–18). Also, sonographic identification of complex anatomy such as duplicated veins can be difficult, the study may cause some discomfort, and evaluation may be limited in obese persons (16).
Because the primary risk factor for recurrent PE is believed to be residual DVT in the proximal leg veins (15), a single comprehensive CT examination of the deep venous system and the thorax for venous thromboembolism would be desirable and presumably cost-effective (Fig 1) (4,5). It would serve as a valuable baseline for follow-up (9); might guide interventions, especially vena cava filter placement; and could potentially salvage the occasional suboptimal thoracic portion of the study.
View larger version (96K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a. Representative normal CT venographic portion of a combined CTVPA study (hereafter, CT venogram) in a 19-year-old woman with chest pain. CT venograms of the pelvis (a) and legs (b) demonstrate patent bilateral external iliac (arrows in a) and distal superficial femoral (large arrows in b) veins. The superficial femoral veins are posterolateral to the superficial femoral arteries (small arrows in b) at this level. Findings from CT pulmonary angiography were also normal (not shown).
|
|
View larger version (76K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b. Representative normal CT venographic portion of a combined CTVPA study (hereafter, CT venogram) in a 19-year-old woman with chest pain. CT venograms of the pelvis (a) and legs (b) demonstrate patent bilateral external iliac (arrows in a) and distal superficial femoral (large arrows in b) veins. The superficial femoral veins are posterolateral to the superficial femoral arteries (small arrows in b) at this level. Findings from CT pulmonary angiography were also normal (not shown).
|
|
|
Development of CTVPA
|
|---|
Identification of DVT at CT
Soon after the clinical introduction of CT, DVT in the inferior vena cava and iliac veins was described (19). Articles on the utility of CT specifically performed for suspected DVT in the pelvis and thighs subsequently emerged (20,21). These reports were followed by descriptions of CT venography performed with helical CT and two separate injections of intravenous contrast medium into each foot (22–24). Baldt et al (23) noted 100% sensitivity and 96% specificity of this latter technique in a comparison with conventional venography. These authors advocated the use of CT venography after CT pulmonary angiography, but this entailed the addition of a second separate CT study, as well as bilateral foot vein catheterization and additional intravenous administration of contrast medium (23).
Signs of DVT
In 1980, acute DVT was identified as venous dilatation caused by low-attenuation clot; the venous wall was sharply defined and enhanced in some cases, presumably because of the arterially supplied vasa vasorum. These observations remain true with regard to dedicated CT venography of the abdomen, pelvis, and lower extremities. Perivenous soft-tissue edema is also occasionally seen (23) (Figs 2–5). Chronic DVT, on the other hand, remains a problematic diagnosis in some cases at CTVPA (Figs 6, 7) as at sonography (17) but is clearly present in other cases (Figs 8–10).
View larger version (98K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a. Increasing shortness of breath in a 62-year-old woman with a history of lung cancer. (a) CT pulmonary angiographic portion of the CTVPA study (hereafter, CT pulmonary angiogram) shows a saddle pulmonary arterial embolus. A right peripheral pulmonary infarction (arrow) is evident even at this mediastinal window. (b) CT venogram obtained at the level of the pubic symphysis shows acute left common femoral venous thrombosis (arrow) and edema in the anterior fat (arrowheads). The clot extended from the left external iliac vein to the left calf.
|
|
View larger version (96K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b. Increasing shortness of breath in a 62-year-old woman with a history of lung cancer. (a) CT pulmonary angiographic portion of the CTVPA study (hereafter, CT pulmonary angiogram) shows a saddle pulmonary arterial embolus. A right peripheral pulmonary infarction (arrow) is evident even at this mediastinal window. (b) CT venogram obtained at the level of the pubic symphysis shows acute left common femoral venous thrombosis (arrow) and edema in the anterior fat (arrowheads). The clot extended from the left external iliac vein to the left calf.
|
|
View larger version (66K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3. Shortness of breath in a 53-year-old man with acute myelogenous leukemia. A right lower lobe PE was identified on the CT pulmonary angiogram (not shown). CT venogram shows acute right popliteal venous thrombosis. Localized edema is extensive, and enhancement of the venous wall (long arrow) is prominent, as are edema and subtle peripheral enhancement of the adjacent arterial wall (short arrow).
|
|
View larger version (92K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4 Chest pain and shortness of breath in a 39-year-old woman. CT venogram shows acute left common femoral venous thrombosis. Expansion of the vein (arrow) is significant compared with the normal right side, the venous wall is markedly thickened and enhanced, and the adjacent fat is edematous. PE was also present (not shown).
|
|
View larger version (62K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 5. Increasing shortness of breath and edema and erythema of the left leg in a 67-year-old woman with stage IV non-small-cell lung cancer who was hospitalized for dehydration and pneumonia. CT venogram shows acute left superficial femoral DVT (arrow), along with leg swelling and edema. PE was also identified (not shown).
|
|
View larger version (70K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 6. CT venogram shows DVT in the right superficial femoral vein (long arrow) and partial thrombosis of the right deep femoral vein (short arrow) in an 82-year-old woman. The findings were acute by clinical history; the right thigh exhibits relatively subtle swelling compared with the left.
|
|
View larger version (56K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7a. Shortness of breath in an 89-year-old woman. (a) CT venogram shows partial thrombosis of the left deep femoral vein (arrow). The chronicity of this finding is unclear from examination of this image alone. (b) CT pulmonary angiogram shows multiple segmental PEs (arrows), and the patient was acutely symptomatic.
|
|
View larger version (113K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 7b. Shortness of breath in an 89-year-old woman. (a) CT venogram shows partial thrombosis of the left deep femoral vein (arrow). The chronicity of this finding is unclear from examination of this image alone. (b) CT pulmonary angiogram shows multiple segmental PEs (arrows), and the patient was acutely symptomatic.
|
|
View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8a. Pneumonia and history of chronic right lower extremity DVT in a 77-year-old woman. (a) CT venogram shows chronic DVT (arrow) in a shrunken right common femoral vein. (b) CT venogram obtained immediately below a shows continued chronic right common femoral venous thrombosis (long arrow) and chronic superficial venous thrombosis in the saphenous veins bilaterally (short arrows).
|
|
View larger version (83K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 8b. Pneumonia and history of chronic right lower extremity DVT in a 77-year-old woman. (a) CT venogram shows chronic DVT (arrow) in a shrunken right common femoral vein. (b) CT venogram obtained immediately below a shows continued chronic right common femoral venous thrombosis (long arrow) and chronic superficial venous thrombosis in the saphenous veins bilaterally (short arrows).
|
|
View larger version (91K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9a. Pleuritic chest pain and shortness of breath in a 30-year-old woman. The patient had a history of prior pacemaker placement but no definitive history of DVT. (a, b) CT venograms reveal marked hypoplasia of the right external iliac vein (arrow in a) and atresia of the right common femoral vein associated with several tiny collateral veins (arrows in b). The right common femoral and superficial femoral veins were not visualized 4 years earlier at sonography (not shown). It is unclear whether this finding is an acquired or congenital abnormality in this patient.
|
|
View larger version (84K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 9b. Pleuritic chest pain and shortness of breath in a 30-year-old woman. The patient had a history of prior pacemaker placement but no definitive history of DVT. (a, b) CT venograms reveal marked hypoplasia of the right external iliac vein (arrow in a) and atresia of the right common femoral vein associated with several tiny collateral veins (arrows in b). The right common femoral and superficial femoral veins were not visualized 4 years earlier at sonography (not shown). It is unclear whether this finding is an acquired or congenital abnormality in this patient.
|
|
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10a. Myeloma, chest pain, and leg swelling in a 72-year-old woman. (a) CT venogram shows partial thrombosis of the inferior vena cava (arrow). The chronicity of this nonocclusive thrombus cannot be determined at this specific level. (b) CT venogram shows acute thrombosis of the left common femoral vein (arrow), which is expanded by the clot. (c) CT venogram shows both acute and chronic thrombosis in the distal left superficial femoral vein (arrow); subtle central venous calcification indicates chronic DVT, and expansion of the vein and soft-tissue edema indicates the acute component of the process. Thrombosis was also present in the left iliac, popliteal, and calf veins. The CT venographic findings led to the placement of an inferior vena cava filter via the right internal jugular vein.
|
|
View larger version (88K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10b. Myeloma, chest pain, and leg swelling in a 72-year-old woman. (a) CT venogram shows partial thrombosis of the inferior vena cava (arrow). The chronicity of this nonocclusive thrombus cannot be determined at this specific level. (b) CT venogram shows acute thrombosis of the left common femoral vein (arrow), which is expanded by the clot. (c) CT venogram shows both acute and chronic thrombosis in the distal left superficial femoral vein (arrow); subtle central venous calcification indicates chronic DVT, and expansion of the vein and soft-tissue edema indicates the acute component of the process. Thrombosis was also present in the left iliac, popliteal, and calf veins. The CT venographic findings led to the placement of an inferior vena cava filter via the right internal jugular vein.
|
|
View larger version (61K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 10c. Myeloma, chest pain, and leg swelling in a 72-year-old woman. (a) CT venogram shows partial thrombosis of the inferior vena cava (arrow). The chronicity of this nonocclusive thrombus cannot be determined at this specific level. (b) CT venogram shows acute thrombosis of the left common femoral vein (arrow), which is expanded by the clot. (c) CT venogram shows both acute and chronic thrombosis in the distal left superficial femoral vein (arrow); subtle central venous calcification indicates chronic DVT, and expansion of the vein and soft-tissue edema indicates the acute component of the process. Thrombosis was also present in the left iliac, popliteal, and calf veins. The CT venographic findings led to the placement of an inferior vena cava filter via the right internal jugular vein.
|
|
At sonography, chronic DVT is identified as a shrunken vein that may or may not have some flow within it; the venous wall is echogenic, and there are often associated dilated collateral veins (18). Findings of chronic DVT at CTVPA have been described in a small number of patients, although to our knowledge, they have not been formally studied in the literature to date. These findings include calcification of thrombi as well as of the venous walls, shrunken veins, and collateral vessels (25). However, if partial or complete DVT is present in a normal-sized vein, in the absence of other specific findings of acute (wall enhancement and perivenous edema) or chronic DVT, we believe the chronicity of the thrombus cannot easily be determined at CT venography alone.
Initial Experience with CTVPA
In 1998, Loud et al (26) first reported a combined technique of CT venography and pulmonary angiography that required no additional intravenous contrast medium or additional venipuncture and that permitted evaluation of the deep veins of the abdomen, pelvis, thighs, and calves. This comprehensive examination for PE and DVT added only a few minutes and images (26), yet revealed additional relevant pathologic findings in the abdomen, pelvis, and legs in a minority of patients (27). The initial protocol was as follows: 3–3
minutes after the start of the intravenous injection of approximately 150 mL of contrast medium (350–370 mg I/mL) for the CT pulmonary angiographic phase of the study, 10-mm axial images were obtained at every 5-cm interval from the diaphragm to the ankles (26). In a second report by Loud et al (28), there was 100% correlation between the CT venographic phase of the study and sonography of the thighs in 71 patients with suspected PE; extension of DVT into the pelvis was identified in six patients at CT venography.
Consistently high levels of enhancement in the deep venous system at CT venography have been reported for patients without DVT, whereas uniformly low-attenuation measurements have been noted in patients with DVT. These findings have usually permitted easy differentiation of normal from thrombosed veins in most patients (28–34). The mean attenuation of patent deep veins in the abdomen, pelvis, and legs is approximately 95–100 HU (28,30), with almost all attenuation measurements within the 60–140 HU range (31), whereas thrombosed veins usually measure approximately 30–50 HU (28,32).
Although the exact timing of CT venography has been debated, it is probably not critical, since high levels of enhancement have been obtained over a temporal window at least as wide as 2–4 minutes after the initiation of contrast medium administration (30,31,33,34). Although maximal venous enhancement appears to occur about 2 minutes after the start of intravenous injection of contrast medium, we prefer a delay of 3–3 minutes to minimize mixing artifacts (33,34).
|
Review of Recent Experience
|
|---|
Results of Two Multicenter Studies of CTVPA
Two large multicenter studies of CTVPA have recently been reported. One group, led by Cham et al (32), evaluated findings in 541 patients with suspected PE. The investigators obtained 10-mm images from the pelvis to the popliteal fossa in a single contiguous helical acquisition in each patient. DVT was present in 45 patients (8%), including 16 in whom no PE was found at CT. Of the 116 patients with sonographic correlation, 15 had DVT that was identified at both CT venography and sonography, and four other patients had clot correctly identified in the CT venographic phase of the study that was initially missed at sonography. CT venography helped identify clot in the inferior vena cava in one patient and revealed pelvic extension of DVT in seven patients. Significantly, there were no false-negative findings from the CT venographic studies (32).
The other group of investigators, led by Loud et al (35), studied CTVPA in 650 patients, 308 of whom had sonographic correlation within a 24-hour period. Eighteen percent of patients had PE or DVT, or both, with PE alone in 27 patients (4% of total), DVT alone in 31 (5%), and PE and DVT in 58 (9%). In the patients with sonographic correlation, CT venography was 97% sensitive and 100% specific for DVT in the thighs. In two patients, CT venographic studies yielded false-negative findings owing to undetected short-segment DVT in the common and superficial femoral veins, but findings from CT pulmonary angiography were positive in both patients. Additionally, four patients had initially negative results from sonography and positive findings from CT venography, but repeated sonography helped confirm the presence of DVT (35).
Frequency and Distribution of DVT in 957 Recent CTVPA Studies
For this article, we retrospectively reviewed the images of all positive-result CT venographic studies from a series of 957 recent and previously unpublished CTVPA examinations to determine the frequency and specific distribution of DVT. Most cases illustrated throughout this article are from this new series (eg, Figs 11–13).
View larger version (121K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 11. Right-sided chest pain in a 50-year-old man with acute myelogenous leukemia. CT venogram shows an isolated left iliac venous thrombosis (arrow). This finding, plus those of PE (not shown) and left external iliac venous thrombosis (not shown), led to inferior vena cava filter placement by means of a right inguinal approach. Although the nonocclusive left iliac venous thrombosis is not accompanied by findings of acute venous thrombosis such as wall enhancement, significant venous enlargement, or edema in the adjacent soft tissues, the PE and DVT were presumed to be acute on the basis of the patient’s history.
|
|
View larger version (117K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12a. Repeated complaints of syncope over the previous few months in a 30-year-old man. Left-sided chest pain and swelling of the right leg developed 9 days prior to imaging. Findings from sonography of the lower extremities, which included the usual survey of both common femoral, superficial femoral, and popliteal veins, were normal. CT pulmonary angiogram (a) shows a left PE (arrow), and CT venogram (b) reveals a right external iliac thrombosis (arrow). There was also right calf DVT, and the thighs were found not to have DVT.
|
|
View larger version (107K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 12b. Repeated complaints of syncope over the previous few months in a 30-year-old man. Left-sided chest pain and swelling of the right leg developed 9 days prior to imaging. Findings from sonography of the lower extremities, which included the usual survey of both common femoral, superficial femoral, and popliteal veins, were normal. CT pulmonary angiogram (a) shows a left PE (arrow), and CT venogram (b) reveals a right external iliac thrombosis (arrow). There was also right calf DVT, and the thighs were found not to have DVT.
|
|
View larger version (60K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 13. Six days of shortness of breath after abdominal surgery in a 55-year-old woman. CT venogram shows multiple thrombosed veins in the right calf (arrows). Both calves are edematous. DVT was isolated to the right calf, and PE was identified at the CT pulmonary angiographic phase of the study (not shown).
|
|
The distribution of DVT is summarized in the Table. Overall, 100 patients in our series (10.5%) had DVT, which was unilateral in 74 and bilateral in 26, and associated with PE in 58. Sites of particular interest, and potential controversy in some cases, are the superficial femoral, deep femoral, superficial, and calf veins.
Superficial Femoral Venous Thrombosis.—
There is controversy over the frequency of isolated superficial femoral venous thrombosis as the only site of DVT in the thigh, as well as whether this region can be skipped during sonographic examination and therefore during CT venography. Pezzullo et al (36) have advocated a limited sonographic protocol in which the superficial femoral vein is skipped. In our series and in the sonographic study reported by Frederick et al (37), only 5% of patients had isolated superficial femoral venous thrombosis in the thigh. However, in a different series of 2,704 serial studies (38), 22% of the sonographic examinations that were positive for DVT showed isolated superficial femoral venous thrombosis; thus, we do not recommend exclusion of the middle part of the thigh from CT venographic studies.
Deep Femoral Venous Thrombosis.—
The deep (profunda) femoral vein splits from the common femoral vein and runs lateral and posterior to the superficial femoral vein (Fig 14). This site is one of potential DVT that is difficult to evaluate at sonography (17) and that has received little attention in the CTVPA literature. Approximately one-quarter of the 100 positive cases in our series had evidence of deep femoral venous thrombosis. Deep femoral venous thrombosis was present in 26 cases (four bilateral) (Figs 15–17), and PE was present in 24 of these. There was a wide variety of patterns of concurrent DVT in these patients, most often complete DVT of the entire leg on that side. However, the deep femoral vein was never the only site of DVT. Only a few cases of deep femoral venous thrombosis have been specifically reported in the CTVPA literature to date (14,38,39). It is interesting that the first described case of deep femoral venous thrombosis at CT predated the introduction of CTVPA (40). The frequency of deep femoral venous thrombosis is probably underreported in the sonography literature. We could find few formal sonographic investigations regarding the status of the deep femoral vein; in one study of 64 patients with DVT at sonography, five had deep femoral venous thrombosis, but as with our patients, the deep femoral vein was never the only site of DVT (41).
View larger version (69K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 14. Acute shortness of breath after liver surgery in a 66-year-old man. CT venogram shows large but patent proximal deep femoral veins (solid arrows) bilaterally (with smaller deep femoral arteries immediately anterior to the veins), posterior to the superficial femoral veins (open arrows) and arteries. In this case, results of CTVPA were negative for both PE and DVT.
|
|
View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 16. Decreased oxygen saturation in an 80-year-old man. CT venogram shows bilateral superficial femoral (solid arrows) and deep femoral venous thrombosis (open arrows). CT pulmonary angiograms revealed PEs (not shown). Correct diagnosis of DVT is relatively difficult in this patient because the findings are bilateral, but the veins are expanded and decreased in attenuation compared with what is normally expected. In this case, sonography performed immediately after the CT study helped confirm bilateral DVT. Sonography is complementary in the small percentage of equivocal CT venographic studies.
|
|
View larger version (110K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 17a. Shortness of breath after sigmoid colonic resection in an 81-year-old woman. (a) CT pulmonary angiogram shows a right interlobar PE (arrow). (b) CT venogram reveals subtle left deep femoral venous thrombosis (solid arrow) and asymmetry with the normal right side (open arrow).
|
|
View larger version (77K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 17b. Shortness of breath after sigmoid colonic resection in an 81-year-old woman. (a) CT pulmonary angiogram shows a right interlobar PE (arrow). (b) CT venogram reveals subtle left deep femoral venous thrombosis (solid arrow) and asymmetry with the normal right side (open arrow).
|
|
Superficial Venous Thrombosis.—
Although superficial venous thrombosis can be identified at lower extremity sonography, especially in symptomatic patients, the status of the entire superficial venous system is easily identified at CT venography; both dilated patent veins and thrombosed veins may be discovered, with or without associated DVT (Figs 18, 19).
View larger version (68K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 18a. Shortness of breath and atrial fibrillation in an 81-year-old woman. (a) CT venogram shows a large acute left common femoral venous thrombus (solid arrow) and superficial thrombosis in the right saphenous vein (open arrow). (b) CT venogram shows DVT in a duplicated left distal superficial femoral vein (solid arrows) and in the medially located right saphenous vein (open arrow). Diagnosis of these complex venous thromboses is easy at CT venography but would be difficult at sonography.
|
|
View larger version (68K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 18b. Shortness of breath and atrial fibrillation in an 81-year-old woman. (a) CT venogram shows a large acute left common femoral venous thrombus (solid arrow) and superficial thrombosis in the right saphenous vein (open arrow). (b) CT venogram shows DVT in a duplicated left distal superficial femoral vein (solid arrows) and in the medially located right saphenous vein (open arrow). Diagnosis of these complex venous thromboses is easy at CT venography but would be difficult at sonography.
|
|
View larger version (58K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 19a. Left-sided chest pain in a 78-year-old woman. CT venograms show DVT in the right popliteal vein (arrow in a) and right calf (ie, peroneal) vein (solid arrow in b). Patent superficial varices in the right calf are also seen (open arrows in b).
|
|
View larger version (56K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 19b. Left-sided chest pain in a 78-year-old woman. CT venograms show DVT in the right popliteal vein (arrow in a) and right calf (ie, peroneal) vein (solid arrow in b). Patent superficial varices in the right calf are also seen (open arrows in b).
|
|
Calf DVT.—
There is controversy over the value of using sonography to examine calf veins, as well as the significance of isolated calf DVT (17,42). Gottlieb et al (42) noted that most sonographic examinations of the calves in 283 patients who had concurrent negative findings from thigh sonography were indeterminate with the use of strict criteria, and only three of these patients had adverse outcomes related to DVT. Fourteen percent of the 100 positive DVT studies in our current series had isolated calf DVT; eight of these 14 cases had concurrent PE, so such a frequency of isolated calf DVT was not trivial (Figs 13, 20). Although we have not formally compared CT venography of the calves with sonography and are not aware of any reported comparison, we have found that routine inclusion of the calves better reveals the overall clot burden and, in contrast to sonography, allows relatively easy evaluation of the calves.
View larger version (87K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 20a. Shortness of breath and right leg pain in a 43-year-old man. CT venograms show prominent but patent left medial thigh (a) and calf (b) superficial varices. A relatively subtle right calf DVT is also present (arrow in b). The patient had associated PE (not shown) and no evidence of DVT in the thighs. Analysis of complex venous anatomy in these types of cases is readily and rapidly accomplished with the CT venographic portion of the study.
|
|
View larger version (58K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 20b. Shortness of breath and right leg pain in a 43-year-old man. CT venograms show prominent but patent left medial thigh (a) and calf (b) superficial varices. A relatively subtle right calf DVT is also present (arrow in b). The patient had associated PE (not shown) and no evidence of DVT in the thighs. Analysis of complex venous anatomy in these types of cases is readily and rapidly accomplished with the CT venographic portion of the study.
|
|
Results of Additional Studies
Other investigators have also recently studied CTVPA (14,33,39,43,44). Garg et al (33) noted that the quality of 70 CT venographic examinations was satisfactory or good in 97%, although 19 patients had flow artifacts, beam hardening from orthopedic hardware, or less than ideal venous opacification. We have also occasionally encountered technical limitations in the CT venographic portions of CTVPA studies (Fig 21), as have Garg and others (39,45). Significantly, however, in the study by Garg et al, sonography was difficult to perform, uncomfortable, or inadequate in 25 patients (36%) (33). In a follow-up study by the same investigators (25), there was moderately good interobserver agreement between two radiologists for 146 studies. There was disagreement about the CT venographic results in 17 patients. Thirty-five of the patients had known severe cardiopulmonary disease, and 16 of these had only satisfactory levels of venous opacification. The authors therefore recommended a consensus reading in difficult cases (25). In addition, Ghaye et al (14) performed CTVPA in 209 patients and correlated the results with sonography in 199 patients. CT venographic results were positive in 99, negative in 103, and indeterminate in seven cases; in the latter seven cases, sonographic results were positive in only one. Sonographic findings were positive in 87, negative in 97, and indeterminate in 15 cases; in the latter 15, CT venographic findings were negative in all. There was excellent correlation (
= .84) between CT venography and sonography (14). It is interesting that the authors were able to identify DVT in the renal veins, portal vein, and ovarian veins in a small number of patients and were also able to identify lesions that contributed to DVT (eg, a lymphocele) as well as alternative diagnoses (14). We have also encountered similar findings in some of our patients (Figs 22–24).
View larger version (88K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 21. Shortness of breath after low anterior rectal resection in an 80-year-old woman. CT venogram shows significant streak artifacts caused by a right total hip prosthesis. Clot in the left common femoral vein is identified (arrow).
|
|
View larger version (71K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 23a. Shortness of breath in a 74-year-old woman with a history of malignancy. CT venograms show acute right proximal popliteal DVT (arrow in a), continuation of the clot in the more distal right popliteal vein (long arrow in b), and small bilateral popliteal arterial aneurysms (short arrows in b).
|
|
View larger version (68K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 23b. Shortness of breath in a 74-year-old woman with a history of malignancy. CT venograms show acute right proximal popliteal DVT (arrow in a), continuation of the clot in the more distal right popliteal vein (long arrow in b), and small bilateral popliteal arterial aneurysms (short arrows in b).
|
|
|
Currently Recommended Protocol for CTVPA
|
|---|
For our current routine protocol, we usually use a kVp of 120 and an mAs of 250 when acquiring the CT venographic images. We then review the images on a CT monitor or workstation, initially using routine abdominal windows (400 width, +40 center level) and then a narrower window width (typically a 70–100 width), which in our experience optimizes vessel contrast and improves differentiation between normal and thrombosed vessels (especially in the small percentage of cases in which relatively low-attenuation but otherwise patent veins are present)
We believe that the use of at least 150 mL of a dense contrast agent (ideally 300 mg I/mL), given by means of a rapid intravenous bolus, optimizes routine CTVPA. Although the contrast medium dose could be decreased if multidetector helical CT pulmonary angiography is being performed alone, we are reluctant to decrease the dose for CTVPA, as it would probably decrease the venous attenuation on the CT venographic portion of the study and therefore possibly decrease diagnostic accuracy.
At present, we also recommend obtaining either 5- or 10-mm axial CT venographic images at 4-cm intervals from the diaphragm to the ankles. We believe the minimally increased yield of DVT that might be detected with a contiguous helical acquisition in the abdomen, pelvis, and legs—although now relatively easily performed with a multidetector CT scanner—is not justified compared with the markedly increased number of images that would need to be reviewed and stored. In addition, using body and leg phantoms, we calculated the average dose on our current multidetector CT scanner (Mx8000; Philips Medical Systems, Highland Heights, Ohio) to be 12 mGy for the abdomen and pelvis and 19 mGy for the legs; these doses would increase (eg, to roughly 22 mGy for the abdomen, pelvis, and thighs, on the basis of a calculation by Rademaker et al [46]), if a contiguous helical acquisition of the abdomen, pelvis, and legs was performed. Also, isolated short segment DVT is reportedly rare (35,47).
Finally, our two institutions have not and do not charge patients for the additional abdominal, pelvic, and extremity images, because relatively few extra images are generated during the CT venographic portion of the study with the use of our protocol compared with multidetector CT angiography of the thorax. We are not aware of any precedent in the literature for doing otherwise.
|
Conclusions
|
|---|
Combined CTVPA is the recommended standard protocol when CT is performed in patients with suspected PE, unless sonography of the legs has recently been performed. Multiple investigations have validated the high accuracy of CT venography compared with lower extremity sonography and have shown the advantages of a combined noninvasive imaging study for PE and DVT. Only a few extra minutes are required, and the additional radiation dose and number of images can be reduced if the CT venographic sections are acquired at several-centimeter intervals. In contrast to lower extremity sonography, with CTVPA the deep veins of the abdomen, pelvis, and calves are routinely and easily examined; complex venous (and extravenous) anatomy can be surveyed; and additional or alternative diagnoses in the abdomen, pelvis, and legs can be suggested or established. Combined CTVPA represents a "one-stop shopping" imaging examination for venous thromboembolism, serves as a baseline study for follow-up imaging, may be used to plan patient therapy, may salvage the occasional suboptimal CT pulmonary angiographic examination, and can increase the confidence of the interpreting radiologist in situations in which equivocal findings of subsegmental PE are present if DVT is detected.
|
Footnotes
|
|---|
Abbreviations: CTVPA = computed tomographic venography and pulmonary angiography,
DVT = deep venous thrombosis,
PE = pulmonary embolism
See the commentary by Mayo and Ketai
following this article.
|
References
|
|---|
- Goldhaber SZ. Pulmonary embolism. N Engl J Med 1999; 339:93-104.[Free Full Text]
- Hyers TM. Venous thromboembolism. Am J Respir Crit Care Med 1999; 159:1-14.[Free Full Text]
- Wolfe TR, Hartsell SC. Pulmonary embolism: making sense of the diagnostic evaluation. Ann Emerg Med 2001; 37:504-514.[CrossRef][Medline]
- Goodman LR. CT diagnosis of pulmonary embolism and deep venous thrombosis. RadioGraphics 2000; 20:1201-1205.[Free Full Text]
- Fishman EK, Horton KM. CT of suspected pulmonary embolism: study design optimization. AJR Am J Roentgenol 2000; 175:1003-1004.
- Mayo JR, Remy-Jardin M, Muller NL, et al. Pulmonary embolism: prospective comparison of spiral CT with ventilation-perfusion scintigraphy. Radiology 1997; 205:447-452.[Abstract/Free Full Text]
- Lomis NNT, Yoon HC, Moran AG, Miller FJ. Clinical outcomes of patients after a negative spiral CT pulmonary arteriogram in the evaluation of acute pulmonary embolism. J Vasc Interv Radiol 1999; 10:707-712.[Medline]
- Goodman LR, Lipchik RJ, Kuzo RS, Liu Y, McAuliffe TL, O’Brien DJ. Subsequent pulmonary embolism: risk after a negative helical CT pulmonary angiogram—prospective comparison with scintigraphy. Radiology 2000; 215:535-542.[Abstract/Free Full Text]
- Ghaye B, Szapiro D, Mastora I, et al. Peripheral pulmonary arteries: how far in the lung does multi–detector row spiral CT allow analysis? Radiology 2001; 219:629-636.[Abstract/Free Full Text]
Top
Abstract
LEARNING OBJECTIVES FOR TEST...
