Published online before print July 30, 2007, 10.1148/radiol.2443061693
(Radiology 2007;244:883-889.)
© RSNA, 2007
Discordance between CT and Angiography in the PIOPED II Study1
Conrad Wittram, MB, ChB,
Arthur C. Waltman, MD,
Jo-Anne O. Shepard, MD,
Elkan Halpern, PhD, and
Lawrence R. Goodman, MD
1 From the Division of Thoracic Radiology, Department of Radiology, Massachusetts General Hospital and Harvard Medical School, Founders Bldg 202, 55 Fruit St, Boston, MA 02114 (C.W., A.C.W., J.A.O.S., E.H.); and Department of Radiology, Medical College of Wisconsin, Milwaukee, Wis (L.R.G.). Received September 30, 2006; revision requested December 12; revision received January 4, 2007; accepted February 1; final version accepted March 14.
Address correspondence to C.W. (e-mail: cwittram{at}partners.org).
 |
ABSTRACT
|
|---|
Purpose: To retrospectively evaluate the causes of discordant computed tomographic (CT)–angiographic readings from the Prospective Investigation of Pulmonary Embolism Diagnosis, or PIOPED, II study.
Materials and Methods: Institutional review board approval was obtained for this HIPAA-compliant study. Of 1036 patients suspected of having pulmonary embolism who were examined with CT, 226 underwent angiography; 206 patients had concordant results and 20 had discordant results according to two independent readers. Of these 20 patients, 10 were men and 10 were women (mean age, 49 years). Among the 20 studies with discordant results, central readers identified seven cases as negative and 13 as positive for pulmonary embolism at CT; these findings were reversed at angiography. Side-by-side comparisons of discordant studies were performed in consensus. The time between CT and angiography and all locations of pulmonary embolism vascular territory were recorded. The McNemar binomial test was used.
Results: One patient had false-positive findings at angiography, 13 patients had false-negative findings at angiography, and two patients had false-negative findings at CT. Four patients had true-negative findings at CT; however, findings were positive for thrombus at angiography. The sensitivity for the detection of pulmonary embolism was 87% for CT and 32% for angiography (P = .007). The largest missed thrombus at angiography was subsegmental in eight patients, segmental in two patients, and lobar in three patients; at CT it was subsegmental in two patients. The mean time between CT and angiography was 40 hours ± 21 (standard deviation) (range, 10–97 hours).
Conclusion: In the interval between CT and angiography, thrombi can remain the same, resolve, develop, or result from angiography.
© RSNA, 2007
|
INTRODUCTION
|
|---|
Pulmonary embolism is the third most common acute cardiovascular disease after myocardial infarction and stroke. It leads to thousands of deaths each year because it often goes undetected (1). The potential of computed tomographic (CT) pulmonary angiography to aid in the diagnosis of pulmonary embolism has been realized; this modality has become the test of choice and thus the de facto standard of care in many institutions (2). Study findings have shown that for the detection of pulmonary emboli to the level of subsegmental arteries, the sensitivity of multidetector CT pulmonary angiography ranges from 90% to 100% and the specificity ranges from 89% to 94%, with pulmonary angiography used as the reference standard (3,4). The findings of a much larger multicenter study were recently published: the Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) II study (5). In this study, investigators used a composite reference standard, and the results showed that CT pulmonary angiography had a sensitivity of 83% and a specificity of 96% in the detection of pulmonary embolism. The investigators also found that combined CT pulmonary angiography and CT venography had a sensitivity of 90% and a specificity of 95% for the detection of venous thromboembolic disease. The purpose of our study was to retrospectively evaluate the causes of discordant CT angiographic readings from the PIOPED II study.
|
MATERIALS AND METHODS
|
|---|
Patients
The Massachusetts General Hospital institutional review board approved this retrospective study and waived the informed consent requirement. Our study was compliant with requirements of the Health Insurance Portability and Accountability Act (HIPAA). All patients gave written informed consent for the original PIOPED II study, which was also compliant with HIPAA requirements and had institutional review board approval at all participating centers. The PIOPED II trial was a prospective multicenter study sponsored by the National Heart, Lung, and Blood Institute. Between September 2001 and July 2003, 1036 patients suspected of having pulmonary embolism underwent CT, and the results were compared with those obtained with a composite reference standard. A positive result at a reference test fulfilled one of the following conditions: (a) high-probability ventilation-perfusion lung scintigram in a patient with no history of pulmonary embolism, (b) abnormal pulmonary angiogram, or (c) abnormal finding at venous ultrasonography in a patient who had not previously had deep venous thrombosis at that site and who had a nondiagnostic result at ventilation-perfusion scintigraphy (5). A total of 226 of these patients underwent conventional angiography; of these, 206 had concordant readings and 20 had discordant readings. Among the 20 patients with discordant readings, central readers identified seven as negative and 13 as positive for pulmonary embolism at CT; the angiographic findings were reversed. Ten of the 20 patients were men and 10 were women (average age, 49 years; age range, 26–80 years). Patients underwent heparin therapy if a local diagnosis of pulmonary embolism was assigned at CT pulmonary angiography.
CT and Angiographic Techniques
CT pulmonary angiography was performed with four-, eight-, or 16-detector row scanners. Low-osmolar nonionic contrast material (135–150 mL) was injected through an arm vein at a rate of 4 mL/sec. The delay between injection and scanning was 20–28 seconds; alternatively, the delay was determined by bolus tracking. Scanning was performed from the diaphragm to the apex of the lung. For patients who weighed less than 250 pounds and were examined with a four-detector row scanner, collimation was 1.25 mm, table speed was 7.5 mm per rotation, pitch was 1.5 (usually between 1.0 and 2.0), voltage was 120 kVp, current was 400 mA, and rotation time was approximately 0.8 second. For CT venography, the deep veins were scanned from the level of the iliac crest to the popliteal veins, with collimation of 7.5 mm, reconstruction of 7.5 mm, table speed of 30 mm per rotation, pitch of 1.5, current of 180 mA, voltage of 120 kVp, rotation time of 1 second, and delay between injection and scanning of 3 minutes. Minor protocol modifications were made for heavier patients and for newer scanners. All pulmonary angiograms were obtained by experienced angiographers using digital subtraction angiographic equipment. The full PIOPED II CT pulmonary angiographic and angiographic techniques have been described elsewhere by Gottschalk et al (6).
Data Collection
The studies of the 20 patients were read by two highly experienced readers who had independently agreed on a CT reading and by two highly experienced readers who had independently agreed on an angiographic reading; however, the results were discordant. Therefore, a side-by-side comparison was made by two PIOPED II study readers in consensus (one CT reader [C.W.] who had 7 years of experience reading CT pulmonary angiograms and one angiographer [A.C.W.] who had more than 20 years of experience reading pulmonary angiograms). The comparison was arbitrated by a PIOPED II reader (J.A.O.S., who had 9 years of experience reading CT pulmonary angiograms and who was a trained angiographer) with experience in both techniques. The CT venograms were also read in consensus. The CT pulmonary angiograms, CT venograms, and angiograms were classified as readable or indeterminate. Readable images were defined as those in which vessels were adequately depicted with contrast material, without motion or image noise obscuring vessel evaluation. Indeterminate images were defined as those in which vessels were inadequately depicted with contrast material or one in which motion or image noise obscured vessel evaluation.
All thromboembolic locations were recorded in one of 33 possible pulmonary vascular territories for CT and angiography. The vascular territories consisted of the main pulmonary artery; the right pulmonary artery to the right middle lobe take-off; the right upper lobe artery; the apical, posterior, and anterior segments of the right upper lobe; any subsegment of the right upper lobe; the right middle lobe artery; the lateral and medial segments of the right middle lobe; any subsegment of the right middle lobe; the right lower lobe artery; the superior, medial-basal, anterior-basal, lateral-basal, and posterior-basal segments of the right lower lobe; any subsegment of the right lower lobe; the left pulmonary artery to the lingula take-off; the left upper lobe artery; the apicoposterior and anterior segments of the left upper lobe; any subsegment of the left upper lobe; the lingula artery; the lingula superior and inferior segments; any lingula subsegment; the left lower lobe artery; the superior, anterior-medial-basal, lateral-basal, and posterior-basal segments of the left lower lobe; and any subsegmental artery of the left lower lobe.
The attenuation values of the main pulmonary arteries were recorded in Hounsfield units. These measurements were obtained by drawing a region of interest with an area equal to half the cross-sectional area of the main pulmonary artery, such that the region of interest also covered vessel and contrast material on the immediately superior and inferior adjacent images. The time difference between the CT scan and the angiogram was recorded in hours.
Diagnostic Criteria for Pulmonary Embolism
Both acute and chronic pulmonary embolism were identified as intraluminal filling defects that demonstrated a sharp interface with intravascular contrast material for both imaging modalities. The diagnostic criteria for acute pulmonary embolism were (a) complete arterial occlusion with failure to opacify the entire lumen (the artery may be enlarged in comparison to pulmonary arteries of the same order of branching) (6–8), (b) central arterial filling defect surrounded by intravascular contrast material (6), and (c) peripheral intraluminal filling defect that made acute angles with the arterial wall (7,8).
The diagnostic criteria for chronic pulmonary embolism were (a) complete occlusion of a vessel that was smaller than the pulmonary arteries of the same order of branching (7,8), (b) peripheral eccentric filling defect that made obtuse angles to the vessel wall (7,8), (c) contrast material flowing through apparent thick-walled arteries that were smaller because of recanalization (7,8), and (d) a band or web within a contrast material–filled artery (7,8).
Diagnostic Criteria for Deep Venous Thrombosis
Both acute and chronic pulmonary embolism were identified as intraluminal filling defects that demonstrated a sharp interface with intravascular contrast material. Criteria for acute deep venous thrombosis were as follows: For the occlusive form, a complete filling defect with failure to opacify the entire lumen because of a central filling defect was seen. The vessel may enlarge compared with the opposite vein, and the vessel wall may demonstrate enhancement. For the nonocclusive form, a partial filling defect surrounded by contrast material was noted. Criteria for deep venous thrombosis of unknown age were as follows: (a) complete filling defect with vein smaller than its peers; (b) contrast material flowing through a thickened often smaller vein; and (c) a secondary sign, such as increased collateral vessels (6).
Statistical Analysis
The McNamara binomial test (SAS, version 9.1; SAS Institute, Cary, NC) was used to compare the performances of angiography and CT pulmonary angiography in the detection of pulmonary embolism. A P value less than .05 was considered to indicate a statistically significant difference.
|
RESULTS
|
|---|
Quality
All discordant angiograms and CT pulmonary angiograms were considered readable.
Per-Patient Analysis
At angiography, findings of one examination were false-positive and findings of 13 examinations were false-negative; findings of two examinations were false-negative at CT. Four studies had findings that were true-negative at CT but had become positive for thrombus by the time of angiography (Fig 1, Table 1). Sensitivities for detection of pulmonary embolism were 87% for CT and 32% for angiography (P = .007). The specificity could not be calculated for either investigation.
View larger version (122K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1a: (a) Anteroposterior pulmonary angiogram demonstrates segmental thrombus in the apicoposterior segment of the left upper lobe (arrow). This finding was not present at CT. (b) Transverse CT pulmonary angiogram obtained 19 hours before angiography demonstrates no evidence of thrombus in the apicoposterior segment (arrow) of the left upper lobe. (c) Coronal reformatted image of CT pulmonary angiography performed 19 hours before angiography demonstrates no evidence of thrombus in the apicoposterior segment (arrow) of the left upper lobe.
|
|
View larger version (59K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1b: (a) Anteroposterior pulmonary angiogram demonstrates segmental thrombus in the apicoposterior segment of the left upper lobe (arrow). This finding was not present at CT. (b) Transverse CT pulmonary angiogram obtained 19 hours before angiography demonstrates no evidence of thrombus in the apicoposterior segment (arrow) of the left upper lobe. (c) Coronal reformatted image of CT pulmonary angiography performed 19 hours before angiography demonstrates no evidence of thrombus in the apicoposterior segment (arrow) of the left upper lobe.
|
|
View larger version (50K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 1c: (a) Anteroposterior pulmonary angiogram demonstrates segmental thrombus in the apicoposterior segment of the left upper lobe (arrow). This finding was not present at CT. (b) Transverse CT pulmonary angiogram obtained 19 hours before angiography demonstrates no evidence of thrombus in the apicoposterior segment (arrow) of the left upper lobe. (c) Coronal reformatted image of CT pulmonary angiography performed 19 hours before angiography demonstrates no evidence of thrombus in the apicoposterior segment (arrow) of the left upper lobe.
|
|
Per–Vascular Territory Analysis
Angiographic findings were false-positive for one study and false-negative for 24 studies; in 15 vascular territories, pulmonary embolism was seen at CT but not at angiography. With CT, findings at two vascular territories were false-negative. Among the four CT studies with true-negative results, findings for six vascular territories were positive at angiography (Table 2).
Thrombi, Attenuation, and Time
The largest thrombus missed at angiography was subsegmental in eight patients (Fig 2), segmental in two patients (Fig 3), and lobar in three patients (Fig 4); at CT, the largest missed thrombus was subsegmental in two patients. The mean attenuation of the main pulmonary artery was 335 HU ± 77 (standard deviation) (range, 230–531 HU). The mean time difference between CT and angiography was 40 hours ± 21 (range, 10–97 hours); pulmonary angiography was performed after CT pulmonary angiography in all 20 cases.
View larger version (81K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2a: (a) Transverse CT pulmonary angiogram demonstrates subsegmental acute pulmonary embolism (arrow) within the posterior basal segment of the right lower lobe. (b) On anteroposterior angiogram obtained 45 hours after a, the thromboembolus (arrow) is again seen in the subsegmental posterior basal segment artery. There is also poor perfusion peripheral to the thrombus that was missed at original evaluation.
|
|
View larger version (153K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 2b: (a) Transverse CT pulmonary angiogram demonstrates subsegmental acute pulmonary embolism (arrow) within the posterior basal segment of the right lower lobe. (b) On anteroposterior angiogram obtained 45 hours after a, the thromboembolus (arrow) is again seen in the subsegmental posterior basal segment artery. There is also poor perfusion peripheral to the thrombus that was missed at original evaluation.
|
|
View larger version (79K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3a: (a) Transverse CT pulmonary angiogram demonstrates segmental acute pulmonary embolism (arrow) within the anterior basal segment of the right lower lobe. (b) Right anterior oblique angiogram obtained 42 hours after a. The thromboembolus (arrow) is again seen in the anterior basal segmental artery, and there is also complete occlusion of one of the more peripheral subsegmental arteries (arrowhead) that was missed at the original evaluation.
|
|
View larger version (142K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 3b: (a) Transverse CT pulmonary angiogram demonstrates segmental acute pulmonary embolism (arrow) within the anterior basal segment of the right lower lobe. (b) Right anterior oblique angiogram obtained 42 hours after a. The thromboembolus (arrow) is again seen in the anterior basal segmental artery, and there is also complete occlusion of one of the more peripheral subsegmental arteries (arrowhead) that was missed at the original evaluation.
|
|
View larger version (78K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4a: (a) Transverse CT pulmonary angiogram demonstrates lobar acute pulmonary embolism (arrow) within the right lower lobe artery. (b) Anteroposterior angiogram obtained 32 hours after a. The thromboembolus (arrow) is seen as a large filling defect within the right lower lobe artery that was missed at the original evaluation.
|
|
View larger version (106K):
[in this window]
[in a new window]
[Download PPT slide]
|
Figure 4b: (a) Transverse CT pulmonary angiogram demonstrates lobar acute pulmonary embolism (arrow) within the right lower lobe artery. (b) Anteroposterior angiogram obtained 32 hours after a. The thromboembolus (arrow) is seen as a large filling defect within the right lower lobe artery that was missed at the original evaluation.
|
|
CT Venography
All 20 CT venograms were considered readable. Three patients had acute deep venous thrombosis. One patient with thrombus in the left popliteal vein had false-negative findings at CT pulmonary angiography and positive angiographic results. The second patient, whose thrombus was in the left deep and superficial femoral veins, had positive findings at CT pulmonary angiography and false-negative results at angiography. The third patient, who had thrombus in the left common femoral vein, had positive findings at CT pulmonary angiography and a false-negative angiogram.
|
DISCUSSION
|
|---|
Our findings showed that CT was significantly more sensitive than angiography in the detection of pulmonary embolism. Previous clinical studies that have compared multidetector CT with angiography have demonstrated trends to support this assertion. Qanadli et al (3) reviewed 157 cases that were imaged with multidetector CT, involving effective section thicknesses of 2.7 mm, and with angiography. The sensitivity and specificity of CT pulmonary angiography were 90% and 94%, respectively. One hundred forty-five studies had concordant results, six CT pulmonary angiographic studies had inconclusive results, and six studies had discordant results (three positive and three negative for pulmonary embolism at CT). Upon further review in consensus, two of the discordant readings were considered to be true-positive at CT pulmonary angiography and false-negative at angiography, one was considered false-positive at CT pulmonary angiography, two were considered false-negative at CT pulmonary angiography, and one was considered false-positive at angiography (because of flow artifact) and true-negative at CT pulmonary angiography. In the study by Qanadli et al (3), 92 thromboemboli were seen at CT pulmonary angiography and 56 were seen at angiography.
In a more recent study by Winer-Muram et al (4), 93 patients suspected of having pulmonary embolism were imaged by using multidetector CT pulmonary angiography with a section thickness of 3.2 mm; compared with angiography, the sensitivity and specificity of CT pulmonary angiography were 100% and 89%, respectively. There were 85 concordant results and eight discordant readings, all of which yielded positive results at CT. Further review in consensus indicated that of the discordant results, four studies had true-positive findings at CT pulmonary angiography and false-negative findings at angiography, three had false-positive findings at CT pulmonary angiography, and one study had inconclusive findings. In addition, there were 71 locations for pulmonary embolism at CT pulmonary angiography versus 50 locations at angiography (4). As in the studies by Qanadli et al (3) and Winer-Muram et al (4), we also found more thromboemboli at CT (39 vascular territories) than at angiography (30 vascular territories). This observation may be the result of clot lysis and dissolution, which allow the thromboembolus to move to a more peripheral location in the time between the two studies. However, this may also be a function of the greater sensitivity of cross-sectional imaging compared with angiography. The use of CT pulmonary angiography with 1.25-mm section thickness throughout PIOPED II may also have increased the sensitivity of CT pulmonary angiography in the detection of pulmonary embolism compared with previous studies.
Our results demonstrate that in the time between CT and angiography, thromboemboli can remain the same, resolve, develop, or result as a complication of angiography. Thromboembolic complications that occur during and after endovascular procedures (because of associated vascular injury and the thrombogenic characteristics of vascular catheters and contrast materials) are easier to detect in more sensitive body systems than in the lungs. Reports of such phenomena abound within the neurointerventional and cardiac interventional literature. Qureshi et al (9) reviewed the incidence of thromboembolic and ischemic events associated with diagnostic and therapeutic cerebral angiography. Among 1547 patients who underwent detachable coil treatment, thromboembolic events were observed in 127 (8.2%); in 86 patients, these events were due to stroke. Of 834 patients who underwent carotid angioplasty and stent placement, 73 (8.8%) experienced a thromboembolic event; of these patients, 47 had a stroke (9). In a study in which coronary angioplasty was compared with excisional atherectomy, Harrington et al (10) randomized patients to undergo percutaneous coronary angioplasty (n = 500) or atherectomy (n = 512). Seventy-eight (15.2%) myocardial infarctions were demonstrated in the atherectomy group versus 34 (6.8%) in the angioplasty group (10).
As the results of these studies demonstrate, the complication of thrombus formation as a result of catheterization is well known in interventional radiology (9,10). To our knowledge, thrombus formation as a result of catheterization has not been previously reported as a complication of pulmonary angiography. This is probably because small thromboembolism in this vascular territory often does not have such catastrophic consequences. The known complication of thrombus formation resulting from conventional lower-extremity venography in a small percentage of cases supports this theory. We believe that the four cases that were true-negative at CT and became positive for thromboembolus by the time of angiography resulted from formation of thrombus between the studies or as a complication of the catheter investigation.
A limitation of this study was the side-by-side comparison of the angiographic and CT pulmonary angiographic images; this type of study design introduces reader bias. However, the point of this study was not to repeat a reader variability study but to understand why two highly experienced CT readers independently agreed on a CT reading and two highly experienced angiography readers independently agreed on an angiography reading but the results were discordant. Usually, we can rely on criteria from radiologic-pathologic correlative studies. However, in vascular imaging, pathologic findings are more often unavailable, and the thrombus disappears when the patient is treated appropriately. Experimental studies, observational studies, and CT–angiographic correlations of pulmonary embolism have all been performed (11–15). With angiography and CT pulmonary angiography, the thromboembolus, or pathologic abnormality, is directly imaged. However, as with all radiologic examinations, the radiologist must be cognizant of the causes of false-positive and false-negative results (16). It is important to review all the data available and read in consensus to understand what is being missed and why.
A similar study design was used by Kakinuma et al (17) to clarify the CT findings and the progression of minute lung cancers that were missed at early spiral CT screening (17). As discussed earlier in this article, the method of a consensus reading was also used by Qanadli et al (3) and Winer-Muram et al (4) to elucidate the causes of discordant angiographic and CT pulmonary angiographic results. We believe that this was the most appropriate method for the important questions raised from this interesting group of patients.
In conclusion, the results of this ancillary PIOPED II study have shown that CT pulmonary angiography was significantly more sensitive than angiography in the detection of pulmonary embolism. These findings support the paradigm shift in the imaging of pulmonary embolism toward CT pulmonary angiography.
|
ADVANCE IN KNOWLEDGE
|
|---|
- The results of this study show that CT pulmonary angiography was significantly more sensitive than angiography in the detection of pulmonary embolism.
|
IMPLICATION FOR PATIENT CARE
|
|---|
- CT pulmonary angiography can be used with confidence as a first-line investigative tool in the diagnosis of pulmonary embolism.
|
ACKNOWLEDGMENTS
|
|---|
We acknowledge all the PIOPED II study staff for their work and dedication with the PIOPED II study and for approving our retrospective study.
|
FOOTNOTES
|
|---|
Abbreviations: PIOPED = Prospective Investigation of Pulmonary Embolism Diagnosis
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, C.W.; 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, C.W., L.R.G.; clinical studies, C.W., A.C.W., J.A.O.S., L.R.G.; experimental studies, A.C.W.; statistical analysis, C.W., E.H.; and manuscript editing, all authors
|
References
|
|---|
- Horlander KT, Mannino DM, Leeper KV. Pulmonary embolism mortality in the United States, 1979–1998: an analysis using multiple-cause mortality data. Arch Intern Med 2003;163:1711–1717.[Abstract/Free Full Text]
- Wittram C, Meehan MJ, Halpern EF, Shepard JO, McLoud TC, Thrall JH. Trends in thoracic radiology over a decade at a large academic medical center. J Thorac Imaging 2004;19:164–170.[CrossRef][Medline]
- Qanadli SD, Hajjam ME, Mesurolle B, et al. Pulmonary embolism detection: prospective evaluation of dual-section helical CT versus selective pulmonary arteriography in 157 patients. Radiology 2000;217:447–455.[Abstract/Free Full Text]
- Winer-Muram HT, Rydberg J, Johnson MS, et al. Suspected acute pulmonary embolism: evaluation with multi-detector row CT versus digital subtraction pulmonary arteriography. Radiology 2004;233:806–815.[Abstract/Free Full Text]
- Stein PD, Fowler SE, Goodman LR, et al. Multidetector computed tomography for acute pulmonary embolism. N Engl J Med 2006;354:2317–2327.[Abstract/Free Full Text]
- Gottschalk A, Stein PO, Goodman LR, Sostman HD. Overview of prospective investigation of pulmonary embolism diagnosis II. Semin Nucl Med 2002;32:173–182.[CrossRef][Medline]
- Ghaye B, Remy J, Remy-Jardin M. Non-traumatic thoracic emergencies: CT diagnosis of acute pulmonary embolism—the first 10 years. Eur Radiol 2002;12:1886–1905.[Medline]
- Washington L, Goodman LR, Gonyo MB. CT for thromboembolic disease. Radiol Clin North Am 2002;40:751–771.[CrossRef][Medline]
- Qureshi AI, Luft AR, Sharma M, Guterman LR, Hopkins LN. Prevention and treatment of thromboembolic and ischemic complications associated with endovascular procedures. II. Clinical aspects and recommendations. Neurosurgery 2000;46:1360–1375.
- Harrington RA, Lincoff AM, Califf RM. Characteristics and consequences of myocardial infarction after percutaneous coronary intervention: Insights from the Coronary Angioplasty Versus Excisional Atherectomy Trial (CAVEAT). J Am Coll Cardiol 1995;25:1693–1699.[Abstract]
- Dalen JE, Mathur VS, Evans H, et al. Pulmonary angiography in experimental pulmonary embolism. Am Heart J 1966;72:509–520.[CrossRef][Medline]
- Sagel SS, Greenspan RH. Nonuniform pulmonary arterial perfusion: pulmonary embolism? Radiology 1971; 99: 541–548.[Medline]
- Dalen JE, Brooks HL, Johnson LW, et al. Pulmonary angiography in acute pulmonary embolism: indications, techniques, and results in 367 patients. Am Heart J 1971;81:175–185.[CrossRef][Medline]
- Auger WR, Fedullo PF, Moser KM, Buchbinder M, Peterson KL. Chronic major-vessel thromboembolic pulmonary artery obstruction: appearance at angiography. Radiology 1992;182:393–398.[Abstract/Free Full Text]
- Wittram C, Kalra MK, Maher MM, Greenfield A, McLoud TC, Shepard JO. Acute and chronic pulmonary emboli: angiography–CT correlation. AJR Am J Roentgenol 2006; 186(6 suppl 2):S421–S429.[Abstract/Free Full Text]
- Wittram C, Maher MM, Yoo AJ, Kalra MK, Shepard JO, McLoud TC. CT angiography of pulmonary embolism: diagnostic criteria and causes of misdiagnosis. RadioGraphics 2004;24:1219–1239.[Abstract/Free Full Text]
- Kakinuma R, Ohmatsu H, Kaneko M, et al. Detection failures in spiral CT screening for lung cancer: analysis of CT findings. Radiology 1999;212:61–66.[Abstract/Free Full Text]
eLetters:
Read all eLetters
- "Sensitivity of discordance?"
- David Buccigrossi, MD
- Radiology Online, 26 Sep 2008
[Full text]
TOP
ABSTRACT
INTRODUCTION
