| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Thoracic Imaging |
1 From the Department of Medical Imaging, University Hospital Sart-Tilman, Domaine Universitaire du Sart-Tilman B 35, B-4000 Liège, Belgium (B.G., C.T.N., R.F.D); and the Department of Medical Imaging, Clinique Saint Joseph, Liège, Belgium (A.N.). From the 2004 RSNA Annual Meeting. Received March 1, 2005; revision requested April 26; revision received July 4; accepted July 21; final version accepted September 8. Address correspondence to B.G. (e-mail: bghaye{at}chu.ulg.ac.be).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Materials and Methods: The institutional ethics committee approved this study; informed consent was not required. The authors retrospectively reviewed results of 1100 combined single–detector row CT pulmonary angiographic and indirect CT venographic examinations (542 men, 558 women; mean age, 61 years ± 17 [standard deviation]) (group 1) and 308 combined multi–detector row CT pulmonary angiographic and indirect CT venographic examinations (150 men, 158 women; mean age, 62 years ± 18) (group 2), performed in 1408 patients suspected of having pulmonary embolism (PE). Frequency of deep venous thrombosis (DVT), PE, and VTE, and the incremental value of indirect CT venography were recorded in both groups. Data were compared by means of the Student t test for continuous data and z statistics for independent proportions.
Results: VTE, PE, and DVT were found in 23.3% (n = 256), 19.9% (n = 219), and 18.3% (n = 201) of the 1100 patients in group 1, respectively, and in 23.7% (n = 73), 17.2% (n = 53), and 18.8% (n = 58) of the 308 patients in group 2, respectively (P values ranging from .273 to .876). The incremental value of indirect CT venography was 14.4% (37 of 256 patients) in group 1 and 27.4% (20 of 73 patients) in group 2.
Conclusion: Despite potential improved accuracy of multi–detector row CT pulmonary angiography for the diagnosis of PE, the addition of indirect CT venography increased the diagnosis of VTE in 27.4% of patients.
© RSNA, 2006
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Computed tomographic (CT) pulmonary angiography has progressively gained acceptance as the frontline imaging modality for the diagnosis of PE, replacing ventilation-perfusion lung scintigraphy and pulmonary angiography (7–9). Indirect CT venography, performed immediately after CT pulmonary angiography in patients suspected of having VTE (10), provides results similar to those of lower-limb ultrasonography (US) for the diagnosis of deep venous thrombosis (DVT) in the femoropopliteal veins (11–16). The incremental value of indirect CT venography is in the range of 8%–27% in patients with negative findings at 3- to 5-mm-collimation CT pulmonary angiography (12,15–19).
The advent of multi–detector row technology allows analysis of pulmonary arteries down to the sixth subdivision and substantially helps increase the detection rate of clots when 1-mm-collimation multi–detector row CT pulmonary angiography is used (20,21). Accuracy of CT pulmonary angiography has therefore increased from greater than 80% with single–detector row CT pulmonary angiography to greater than 90% with multi–detector row CT pulmonary angiography (7,22,23). Thus, the incremental value of indirect CT venography can be questioned if thin-collimation multi–detector row CT pulmonary angiography is being used (24).
The purpose of our study, therefore, was to compare retrospectively the incremental value of indirect CT venography performed after multi–detector row and single–detector row CT pulmonary angiography for the diagnosis of VTE.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Between 1998 and 2002, 1344 consecutive patients underwent combined single–detector row CT pulmonary angiography and indirect CT venography of the lower limbs (group 1). Pulmonary arteries were scanned from 2 cm below the diaphragm to the aortic arch with helical acquisition (PQ 5000; Philips, Eindhoven, the Netherlands) by using 2-mm collimation, 1-mm reconstruction increment, a pitch of 2, 1-second rotation time, 100 mAs, and 130 kVp, starting 20–25 seconds after the start of intravenous injection of 140 mL of 30% iodinated contrast material (Xenetix 300; Guerbet, Aulnay-sous-Bois, France) at a flow rate of 3 mL/sec. CT venography was performed 210 seconds after the start of injection, from calf to diaphragm, by using a sequential acquisition of 5-mm-thick sections at 20-mm intervals, 100–125 mAs, and 130 kVp. CT venography was started at 240–300 seconds in patients suspected of having low cardiac output (14,25).
Between 2002 and 2004, 329 consecutive patients underwent combined 16–detector row CT pulmonary angiography and indirect CT venography (group 2). Lungs were scanned from base to apex (Sensation 16; Siemens Medical Solutions, Forcheim, Germany) by using 0.75-mm-collimation helical acquisitions, 1-mm-thick reconstruction, 0.7-mm reconstruction increment, 0.5-second rotation time, 100–140 mAs, and 120 kVp, with the same protocol for contrast material injection as used in group 1. Lower-limb veins were scanned from calf to diaphragm by using 1.5-mm collimation, 5-mm-thick reconstruction, 5-mm reconstruction increment, 100 mAs, and 120 kVp, with the same delay as described in group 1.
Of these 1673 patients, 265 (15.8%) were excluded from the study for the following reasons: DVT documented before combined CT pulmonary angiography–indirect CT venography (118 patients in group 1 and 16 patients in group 2), follow-up examinations of documented VTE (100 patients in group 1 and four patients in group 2), and examinations unavailable for review (26 patients in group 1 and one patient in group 2). As a result, the final study population comprised 1408 patients; group 1 consisted of 1100 patients (542 men, 558 women; mean age, 61 years ± 17 [standard deviation]) and group 2 consisted of 308 patients (150 men, 158 women; mean age, 62 years ± 18).
Interpretation of CT Scans
Findings from all 1408 examinations had been interpreted on an independent workstation (Voxel Q; Philips) by two readers reaching a consensus. One of them was a chest radiologist with 10 or 20 years of experience in interpreting CT angiograms (B.G. or R.F.D.). The second reader was a staff radiologist with 3–8 years of experience in reading CT scans. The interpretations were retrospectively collected by A.N. or C.T.N., who reviewed the reports.
Interpretation was formulated according to a predefined protocol that included, along with positive, negative, or inconclusive findings for PE and DVT, the following parameters: The presence of an arterial clot was assessed from the main pulmonary artery down to the subsegmental level for single–detector row CT pulmonary angiography and as distal as possible for multi–detector row CT pulmonary angiography. Inconclusive CT pulmonary angiographic findings were those obtained from examinations without evidence of PE and in which pulmonary arteries were not assessable beyond the lobar level. Inconclusive indirect CT venographic findings were those obtained from examinations without evidence of venous thrombosis and in which one or more venous segments were not assessable. Reasons for inconclusive findings were systematically recorded. CT signs of PE and DVT were interpreted according to guidelines published in the literature and consisted of complete or partial intravascular filling defect surrounded by contrast material that presented the "polo mint" sign, in which sections were perpendicular to the long axis of the vessels, or the "railway" sign, in which sections were parallel to the long axis. The involved artery or vein is frequently dilated (Figure) compared with similar vessels on the other side (9,26).
|
Statistical Analysis
The software used for statistical analysis was Systat 9.0 (Systat Software, Richmond, Calif). Continuous data were expressed as mean ± standard deviation. The 95% confidence intervals for means and proportions were calculated by using the modified Wald method. Comparisons of continuous data were performed by means of the unpaired Student t test. Comparisons of independent proportions were performed with two-tailed z statistics. P < .05 was considered to indicate a significant difference.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
|
Inconclusive findings for PE were seen in 10.4% (146 of 1408) of patients. There were significantly fewer inconclusive CT pulmonary angiographic findings in group 2 (7.5% of patients [23 of 308]) than in group 1 (11.2% of patients [123 of 1100]) (P = .036). Inconclusive CT pulmonary angiographic findings were due to breathing motion artifacts (81% [100 of 123] of patients in group 1 and 74% [17 of 23] of patients in group 2), poor arterial opacification (9% [11 of 123] in group 1 and 13% [three of 23] in group 2), cardiac motion artifacts (6% [seven of 123] in group 1 and 4% [one of 23] in group 2), and low signal-to-noise ratio or arterial segments out of acquisition range (4% [five of 123] in group 1 and 9% [two of 23] in group 2).
Regarding inconclusive findings at CT pulmonary angiography and indirect CT venography, positive results at CT pulmonary angiography were found in 28% (30 of 109) and 33% (12 of 36) of the patients with inconclusive findings at indirect CT in groups 1 and 2, respectively (P = .495) (Table 3). Positive indirect CT venographic findings were seen in 35% (33 of 95) and 29% (five of 17) of patients with inconclusive CT pulmonary angiographic findings in groups 1 and 2, respectively (P = .710) (Table 3).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
Our incremental value of 14% with use of single–detector row CT pulmonary angiography is close to values in the largest, to our knowledge, study published to date (19). In a study of 1590 patients, Cham et al (19) reported a 16% incremental increase in VTE detection after the addition of 10-mm-collimation indirect helical CT venography to 3-mm single–detector row CT pulmonary angiography. On the other hand, the 27% incremental value we obtained with use of 16–detector row CT pulmonary angiography is higher than the 17% incremental increase found by Richman et al (17), who reported their experience with four–detector row CT in 800 outpatients suspected of having PE. Nevertheless, that study involved a greater reconstruction thickness than did our study: 3-mm-thick sections for CT pulmonary angiography and 10-mm-thick sections at 20-mm increments for indirect CT venography (17).
Schoepf et al (21) demonstrated that decreasing the section thickness from 3 to 1 mm resulted in a higher rate of detection of clots and in fewer inconclusive findings for PE. Indeed, we found a significant decrease in inconclusive findings for PE with use of multi–detector row compared with single–detector row CT pulmonary angiography (7.5% vs 11.2%; P = .036). Nevertheless, we found no significant difference in the rate of isolated PE (ie, patients with CT results positive for PE but without CT evidence of DVT) between patients who underwent testing with both technologies.
Similarly, a recent study (23) suggested that in routine practice, four–detector row CT pulmonary angiography may still have limitations in demonstrating small peripheral emboli in comparison with pulmonary angiography. On the other hand, the percentage of our patients with isolated DVT (ie, patients with CT results positive for DVT but without CT evidence of PE) was higher in the multi–detector row CT group (6.5% vs 3.4%; P = .038). Therefore, the increase in incremental value of indirect CT venography with multi–detector row CT in our study may be related to the use of contiguous thin-collimation (5-mm) sections for indirect CT venography. Indeed, short segmental DVT may be overlooked when sequential scanning is used because the risk increases with larger intersection gaps (12,27). When a helical acquisition is used, even small venous clots can be diagnosed during assessment of consecutive images (19). Furthermore, pseudo–filling defects (ie, flow-related artifacts or volume averaging around venous valves) are more confidently recognized with indirect helical CT venography (13,26). Finally, the addition of indirect CT venography still decreased, by 29%, the rate of inconclusive results for VTE that would have been obtained with CT pulmonary angiography alone when multi–detector row CT was used, compared with 34.7% when single–detector row CT was used.
Unlike with CT pulmonary angiography, the rate of indeterminate results for indirect CT venography was similar with both technologies. Inconclusive indirect CT venographic findings resulted mainly from poor venous enhancement, particularly at the level of the calves in most patients (97% in both groups). This may explain our higher rate of inconclusive findings at indirect CT venography (12.7%) compared with rates in other studies (1%–3%) that did not include the calf in the acquisition range (16,19,27,28).
So far, insufficient venous enhancement has not been compensated by the improvement in CT technology. The use of compressive elastic stockings or isosmolar contrast material (compared with low-osmolar contrast media, which are hyperosmolar to blood) has recently been shown to significantly increase delayed venous enhancement (29–31). Additional studies are needed to investigate whether such refinements can decrease the rate of inconclusive indirect CT venographic findings. Nevertheless, final and positive results for VTE were obtained with a positive result at CT pulmonary angiography in 28% and 33% of patients with inconclusive findings at indirect CT venography acquired with single– and multi–detector row CT, respectively.
We found a high frequency of DVT (74%; 95% confidence interval: 68.7%, 79.1%) with indirect CT venography in patients with PE. Despite being higher than the 36% frequency (95% confidence interval: 22%, 52%) reported in a meta-analysis by van Rossum et al (32), our results are in agreement with those of more recent studies reporting prevalence of DVT ranging from 66% to 83% with use of ascending venography (33,34), US (35), or indirect CT venography (12). To avoid selection bias, we excluded patients who were known to have venous thrombosis before indirect CT venography; thus, our results confirm that the clinical diagnosis of DVT is difficult and that all patients suspected of having VTE would require assessment of the lower-limb veins.
Although multi–detector row technology has greatly expanded our diagnostic capabilities in patients suspected of having VTE, the use of thin collimations often results in an increase in the radiation dose (21). The radiation dose delivered to the gonads and pelvis has to be cautiously considered when indirect CT venography is performed. With use of four–detector row indirect CT venography, researchers from two studies (28,36) reported median cumulative effective doses of 8.3–9.3 mSv and median effective gonadal doses of 3.4–4.4 mSv, with variations between individuals and according to sex. Nevertheless, VTE is a disease that primarily affects an elderly population, as confirmed by a mean age greater than 60 years in our patients and in most study populations (11–14,16–19,22,28). Furthermore, as a rule, we rarely perform indirect CT venography in patients younger than 40 years unless findings from previous US of the lower limb were inconclusive. Therefore, the risk-to-benefit ratio is rather low in the context of potential morbidity and mortality related to VTE (19,37). Dose modulation systems, which minimize the radiation dose required without compromising image quality, are being developed by all device manufacturers and could reduce the radiation dose by 35%–60% (38).
Our study had limitations. First, the retrospective analysis may lead to unintentional selection bias. In particular, we could not compare indirect CT venography with an independent reference test. However, indirect CT venography has been demonstrated to have results similar to those of US for the diagnosis of DVT (11–16,26). Second, patients were not classified according to symptoms of DVT. Nevertheless, patients who presented with symptoms of DVT as their chief complaint underwent US before CT and therefore were excluded from the study. Third, we did not specifically track the minority of patients who had clinical symptoms of PE but underwent other diagnostic tests, because CT pulmonary angiography has been the diagnostic modality of choice in our institution since 1995 (further refined by the addition of indirect CT venography in 1998).
Fourth, rare cases of isolated PE located distal to the subsegmental level may have been overlooked because we have restricted the interpretation to subsegmental pulmonary arteries with use of single-section CT pulmonary angiography. Nevertheless, authors of large studies have questioned the accuracy of positive findings determined at single-section CT pulmonary angiography at the subsegmental level (39). This may, however, have slightly increased the incremental value of single-section indirect CT venography. Finally, we did not examine interobserver agreement, which has been reported as good to excellent for CT pulmonary angiography (18,22,23) and as moderately good to excellent for indirect CT venography (15,27) in other studies.
Despite the theoretically improved accuracy of multi–detector row over single–detector row CT pulmonary angiography, the addition of multi–detector row indirect CT venography still results in an increase of VTE diagnosis in 27% of patients.
|
ADVANCES IN KNOWLEDGE |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
|
FOOTNOTES |
|---|
Abbreviations: DVT = deep venous thrombosis • PE = pulmonary embolism • VTE = venous thromboembolism
Authors stated no financial relationship to disclose.
Author contributions: Guarantors of integrity of entire study, B.G., A.N.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; manuscript final version approval, all authors; literature research, B.G., A.N.; clinical studies, B.G., C.T.N.; statistical analysis, A.N.; and manuscript editing, all authors
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONADVANCES IN KNOWLEDGEReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  H. Nazaroglu, C. A. Ozmen, H. O. Akay, I. Kilinc, and A. Bilici 64-MDCT Pulmonary Angiography and CT Venography in the Diagnosis of Thromboembolic Disease Am. J. Roentgenol., March 1, 2009; 192(3): 654 - 661. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 L. R. Goodman, H. D. Sostman, P. D. Stein, and P. K. Woodard CT Venography: A Necessary Adjunct to CT Pulmonary Angiography or a Waste of Time, Money, and Radiation? Radiology, February 1, 2009; 250(2): 327 - 330. [Full Text] [PDF]
|
|
|
|
|
|
G. Bierry, N. Holl, F. Kellner, S. Riehm, M.-N. Roedlich, M. Greget, and F. Veillon Venous Thromboembolism and Occult Malignancy: Simultaneous Detection During Pulmonary CT Angiography with CT Venography Am. J. Roentgenol., September 1, 2008; 191(3): 885 - 889. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
A. R. Hunsaker, K. H. Zou, A. C. Poh, B. Trotman-Dickenson, F. L. Jacobson, R. R. Gill, and S. Z. Goldhaber Routine Pelvic and Lower Extremity CT Venography in Patients Undergoing Pulmonary CT Angiography Am. J. Roentgenol., February 1, 2008; 190(2): 322 - 326. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
S. P. Kalva, J. P. Jagannathan, P. F. Hahn, and S. T. Wicky Venous Thromboembolism: Indirect CT Venography during CT Pulmonary Angiography--Should the Pelvis Be Imaged? Radiology, February 1, 2008; 246(2): 605 - 611. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Jouveshomme, I. Bohn, and A. Cazaban Diagnosis of pulmonary embolism in hospitalised patients: retrospective survey of an institutional standard Eur. Respir. J., December 1, 2007; 30(6): 1117 - 1123. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
L. R. Goodman, P. D. Stein, F. Matta, H. D. Sostman, T. W. Wakefield, P. K. Woodard, R. Hull, D. F. Yankelevitz, and A. Beemath CT Venography and Compression Sonography Are Diagnostically Equivalent: Data from PIOPED II Am. J. Roentgenol., November 1, 2007; 189(5): 1071 - 1076. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Remy-Jardin, M. Pistolesi, L. R. Goodman, W. B. Gefter, A. Gottschalk, J. R. Mayo, and H. D. Sostman Management of Suspected Acute Pulmonary Embolism in the Era of CT Angiography: A Statement from the Fleischner Society Radiology, November 1, 2007; 245(2): 315 - 329. [Full Text] [PDF]
|
|
|
|
|
|
L. R. Goodman, P. D. Stein, A. Beemath, H. D. Sostman, T. W. Wakefield, P. K. Woodard, and D. F. Yankelevitz CT Venography for Deep Venous Thrombosis: Continuous Images Versus Reformatted Discontinuous Images Using PIOPED II Data Am. J. Roentgenol., August 1, 2007; 189(2): 409 - 412. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Arakawa, T. Kohno, T. Hiki, and Y. Kaji CT Pulmonary Angiography and CT Venography: Factors Associated with Vessel Enhancement Am. J. Roentgenol., July 1, 2007; 189(1): 156 - 161. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
