Printer Friendly

CT of pulmonary thromboembolism.

Pulmonary embolism (PE) remains a leading cause of morbidity and mortality in the United States, with an incidence of approximately 500,000 to 650,000 cases annually. As many as one-third of victims of PE die from a future embolic event. Pulmonary embolism and its precursor, deep venous thrombosis (DVT), are manifestations of numerous conditions that lead to hypercoagulability, stasis, or endothelial damage, particularly in elderly, hospitalized, and oncology patients. Despite improvements in clinical algorithms and the introduction of sensitive serologic tests for excluding PE (such as the enzyme-linked immunosorbent assay [ELISA] d-dimer), the clinical evaluation of PE remains problematic, with the minority of patients presenting with "typical" signs or symptoms.

Advances in multidetector computed tomography (MDCT) scanners (a technology that is now widely available) and an increased awareness of the importance of timely and accurate diagnosis of PE have resulted in increased use of MDCT for emergency department and hospitalized patients. The increased utilization of MDCT for possible PE has been associated with a decrease in the incidence of PE as detected by CT in the period 1998 to 2003. This result supports the concept that CT pulmonary angiography, by providing a global evaluation of the patient with acute chest disease, has become a widely utilized tool in these patients. This trend in MDCT utilization, which parallels its use in the evaluation of acute abdominal pain, is largely a result of its overall accuracy, cost-effectiveness, high negative predictive value, and safety and utility in the evaluation of PE (Figure 1). With its high accuracy and its ability to identify those conditions that account for acute chest symptomatology in the 75% to 90% of patients whose CT pulmonary angiograms are negative for PE, MDCT has largely replaced ventilation-perfusion scanning and pulmonary angiography in the evaluation of PE. (1-4) In as many as 4% of hospitalized patients and 1% of outpatients, unsuspected PE has been identified on routine contrast-enhanced MDCT scans of the chest not necessarily optimized for PE detection. This has reinforced the relatively high incidence of this often asymptomatic condition in the patient population and the diagnostic capabilities of the newest generation of MDCT scanners (Figure 2). (5)


This article reviews the techniques, imaging findings, limitations, accuracy, clinical utility, and potential adverse effects of CT for the evaluation of pulmonary thromboembolism. The additional performance of indirect CT venography of the lower extremities and pelvic veins to improve the sensitivity for the detection of thromboembolic disease will also be addressed.

Technical parameters

Pulmonary CT angiography (CTA) was initially implemented using a single-detector CT (SDCT) scanner, with limitations of minimal slice collimation width and anatomic coverage achievable in a single breath-hold. (6) Despite a lower (as compared with MDCT) sensitivity of SDCT for the detection of subsegmental emboli, the negative predictive value of single-detector pulmonary CTA at 6-month follow-up after withholding of anticoagulation following a negative CT study has been high, similar to that of MDCT. (7) However, a number of studies have reported an increase in the sensitivity and specificity of MDCT in the detection of emboli, particularly subsegmental emboli, and this is now the preferred technique where available. It is presumed that the increased detection of smaller emboli by MDCT will eventually translate into improved patient outcomes (Figure 3). (8-10) At our institution over the last 6 years, we have performed pulmonary CTA sequentially on single-, 4-, 16-, 40-, and 64-detector units, but we now preferentially scan patients on either the 40- or 64-detector scanner, each with a 0.50-second gantry rotation time, to provide optimal temporal, in-plane, and z-axis resolution throughout the thorax.





The most important aspect in optimizing MDCT pulmonary angiography is maximizing pulmonary arterial opacification, with a number of different techniques available to achieve this goal. These include using a fixed scanning delay from the time of initiating contrast injection, usually 15 to 20 seconds, or using a small 15 to 25-mL test bolus of contrast and calculating a time-density curve in the main pulmonary artery to determine the optimal scan delay. Bolus tracking software can also be used since it measures sequential attenuation values in the main pulmonary artery and, once a predetermined threshold has been reached, automatically triggers scan acquisition. We have employed each of these methods with varying success, and we currently use a fixed scanning delay time of 15 seconds on 40- and 64-detector scanners with consistent opacification of the pulmonary arterial vasculature. Table 1 summarizes our scanning parameters for a 64-detector CT. In large patients, the slice thickness and peak voltage can be increased (1.4-mm thickness and 140 kV, respectively) to improve the signal-to-noise ratio and optimize scan quality. An 18-gauge intravenous (IV) catheter placed in a left antecubital vein is preferred for contrast injection, with 100 mL of nonionic contrast (Isovue 370, Bracco Diagnostics Inc., Princeton, NJ) injected at a rate of 5 mL/sec via a dual injector, followed by a 20-mL saline bolus. The use of a saline bolus helps optimize contrast utilization and decreases the attenuation of contrast in the superior vena cava (SVC), thereby lessening beam-hardening artifact in the medial right-upper-lobe vessels (Figure 4). The use of isosmolar contrast agents does not appear to increase attenuation as compared with the use of low-osmolarity contrast agents, although this may improve patient comfort during injection. (11) Typical breath-holding for a 40- or 64-detector CT is 6 to 10 seconds, which allows for scanning of the entire chest even in dyspneic patients. Additionally, pharmacological paralysis and brief interruption of respirators in mechanically ventilated intensive-care-unit patients is unnecessary in most cases.


Poor-quality (ie, nondiagnostic) CT examinations occur in 5% to 10% of patients evaluated for suspected PE. At our institution, we prefer to repeat the study with careful attention to rectifying factors (such as the scan parameters and image quality) that rendered it nondiagnostic (Figure 4). In most cases, inaccurate scan acquisition timing relative to the contrast bolus can be addressed by altering the time delay used, by using an automated trigger program, or by calculating the proper delay with a small timing bolus. In patients with normal renal function, this second contrast injection is unlikely to result in renal nephrotoxicity, assuming adequate patient hydration and normal urine output. Respiratory artifact in dyspneic patients can be reduced by hyperventilating the patient prior to the scan acquisition and with the use of supplemental oxygen.

Electrocardiographic (ECG) gating of scan acquisition is not performed routinely, as it does not appear to improve scan quality and interpretation and it is associated with an incremental increase in radiation dose. (12,13) However, ECG gating can be of use in repeat scanning to clarify equivocal filling defects in the left lower lobe or lingual arteries, which are affected by cardiac pulsation artifact, or for the detection of intracardiac thrombi (Figure 5). Table 2 summarizes common technical problems in pulmonary CTA and proposed solutions.

Patient factors

In patients who require pulmonary CTA and have borderline renal insufficiency (creatinine >1.7 mg/dL), or those with identifiable risk factors for contrast-induced nephrotoxicity (CIN)--such as diabetes (especially in association with renal insufficiency), dehydration, and chronic renal failure--pre- and postcontrast CT hydration is encouraged. The discontinuation of other medications associated with nephrotoxicity prior to the CT scan is also suggested. Intravenous hydration and bicarbonate infusion have been shown to reduce the risk of CIN, whereas N-acetylcysteine administration has shown variable results. (14) There is limited data at present to support the routine use of isosmolar nonionic contrast agents to prevent CIN. Decisions to scan patients who have more severe renal insufficiency are made in consultation with the ordering physician after factoring in the risk/benefit ratio of performing the contrast CT study and considering alternative methods of investigation, including lower-extremity ultrasound, ventilation-perfusion scintigraphy, magnetic resonance angiography, and gadolinium-enhanced pulmonary CTA. (15-17) The risk of nephrogenic systemic sclerosis with the use of gadolinium-based agents in this patient population must also be considered.

Contrast allergies are managed with premedication using corticosteroids, with the optimum benefit seen with administration 10 to 12 hours prior to contrast injection. For those who have prior moderate or severe allergic reactions to contrast or in those whose prior reactions included a respiratory component, the addition of preprocedure oral H1 and H2 blockers provides an additional benefit. Pregnant patients can be safely scanned with overall low risk and less radiation exposure to the fetus and with greater accuracy for detecting PE as compared with ventilation-perfusion scanning performed at routine radionuclide doses. (18,19)




CT findings of pulmonary embolism

The CTA findings of acute PE are analogous to those seen on conventional pulmonary angiography. In acute PE, an embolus is most often seen as an intraluminal filling defect of soft tissue attenuation with the proximal convex aspect of the clot outlined by contrast and the contrast column forming an acute angle with the vessel wall (Figure 6). Thromboemboli that completely occlude a pulmonary artery branch often produce vessel dilatation, with the vessel seen as larger than the accompanying bronchus (Figure 7). This finding is best appreciated by assessing the caliber of the vessel proximal to the embolus and by noting gradual dilatation to the level of the embolus on coronal and sagittal reformations of the axial data set. Nonocclusive emboli may also be central within the lumen, producing the "polo mint" sign and the railway track sign (Figure 8). (20) Large emboli to the main pulmonary artery--so-called saddle emboli--are seen as sausage-shaped filling defects that straddle the main pulmonary arterial bifurcation (Figure 9). Rarely, pulmonary emboli can be seen on noncontrast CT scans as low- or high-attenuation filling defects within the unenhanced central pulmonary arteries. (21)

In addition to the direct findings of intraluminal filling defects, there are several indirect findings that can be of use in the detection of PE on CT. Pulmonary infarcts, though rare, are an important ancillary finding of acute PE. These appear as wedge-shaped, peripheral masslike opacities (Hampton's hump) that have a broad base on the pleural surface with an apex that is oriented toward the hilum (Figure 10). An area of surrounding ground-glass opacity is often seen in association with pulmonary infarction and pathologically reflects alveolar hemorrhage. Occasionally, the peripheral embolus responsible for the pulmonary infarct is not visualized directly, but the presence of a characteristic Hampton's hump should prompt further investigation for PE. Other CT findings associated with PE include pulmonary hemorrhage (seen as peripheral areas of consolidation or ground-glass opacity; Figure 11), subsegmental atelectasis, and pleural effusion. Mosaic lung attenuation is another ancillary sign of acute PE (22) (Figure 12), but it is seen with greater frequency in patients with chronic thromboembolic pulmonary hypertension. Additional important ancillary CT findings are signs of right ventricular strain that may signal the need for more aggressive therapy, including thrombolytics or mechanical thrombectomy. Signs of right ventricular strain include right ventricular dilatation (right ventricular/left ventricular area ratio >1), reflux of contrast material into the hepatic veins, and deviation of the interventricular septum toward the left ventricle (Figure 13). (23)






Approximately 50% of PEs resolve within 1 month of treatment, and there is complete resolution of the abnormality in approximately two thirds of patients. (Figure 14). (24) In the remaining one third of patients, findings of chronic PE are present. These include complete occlusion or stenosis, a web or flap within the artery, an intraluminal filling defect with occasional calcification that forms obtuse angles with the vessel wall, and thickening or calcification of the vessel wall (Figure 15). Secondary signs of chronic PE include a pattern of mosaic perfusion within the lung and enlargement of systemic (usually bronchial) collateral vessels. In patients with chronic PE who have residual intraluminal filling defects, the thrombi show higher attenuation than is seen in acute embolism. (24,25) In up to 4% of patients with acute PE who develop chronic thromboembolic pulmonary hypertension, the findings of associated pulmonary hypertension, including central pulmonary arterial and right-sided cardiac dilatation, may be seen.

Other than thromboemboli, the only differential considerations for intraluminal filling defects as seen on pulmonary CTA are primary or metastatic tumors of the pulmonary artery and the rare macroscopic fat embolus. Metastatic tumor emboli from extrathoracic tumors can spread to the lung by gaining access to the hepatic or renal venous system and then embolizing to the lung via the inferior vena cava and right heart, or by secondary invasion of the pulmonary arterial vasculature following hematogenous dissemination to the lung. Therefore, the most common tumors that produce large-vessel pulmonary arterial emboli are hepatoma and renal cell carcinoma (Figure 16). Primary sarcoma of the pulmonary artery are rare malignant neoplasms that can be seen as filling defects and may be difficult to distinguish from thromboemboli. These tumors can be distinguished by differences in clinical presentation and, on imaging, by enhancement of the embolus that occupies the entire width of the main or proximal right or left pulmonary artery and that frequently extends beyond the confines of the artery into the surrounding tissues. (26) Fat embolism is a clinical syndrome that is caused by microscopic fat emboli, usually following a long bone fracture, and results in neurological, cutaneous, and pulmonary disease with the development of noncardiogenic pulmonary edema radiologically. There are rare reports of macroscopic fatty emboli that are seen as low-attenuation intraluminal filling defects in the pulmonary arteries on MDCT, although the emboli are typically not visible on CT.



Technical artifacts

There are a number of commonly encountered artifacts that can reduce the diagnostic quality of pulmonary CTA and can produce findings that could be misinterpreted as PE. The most common of these is inadequate opacification of the pulmonary arterial vasculature. This is most often the result of improper timing of the image acquisition relative to the contrast bolus, an inability to achieve adequate IV contrast flow rates because of a small-gauge IV catheter access or thoracic inlet venous inflow compromise, or an excessively large patient body habitus with low relative intravascular contrast concentrations and high image noise level (Figure 17). To accommodate for patients >250 pounds, we increase the nominal collimation width of the scans from 1 to 1.4 mm to decrease relative image noise. Another cause of dilute intrapulmonary arterial contrast levels can result from unopacified blood in the inferior vena cava drawn into the right heart and pulmonary arteries during the deep inspiratory maneuver performed by the patient immediately prior to image acquisition. This so-called "transient interruption of contrast" can result in an abrupt, bilateral segmental loss of pulmonary arterial opacification, which makes it difficult to detect emboli. (27,28)

Dense contrast in the SVC as a result of the rapid injection of IV contrast can result in beam-hardening artifacts, particularly in the right upper lobe, which may be misinterpreted as intra-arterial filling defects. These artifacts are best recognized by noting the extension of the artifact from the SVC across the lung beyond the confines of the vessel, a finding that is sometimes best appreciated on coronal reformatted images (Figure 18). This artifact can be limited by utilizing a dual-chamber power injector that injects saline immediately following the contrast bolus, or by scanning the patient from caudal to cranial, thereby imaging the upper lungs and SVC just following the completion of the contrast injection. (6) Similarly, a central venous catheter or a pacemaker lead may produce streak artifacts that mimic acute PE. A high spatial frequency reconstruction algorithm used to improve spatial resolution for lung parenchymal evaluation can render the vessel walls as having artificially high attenuation, thereby mimicking an embolism. For this reason, a standard or soft tissue reconstruction algorithm should be used to view the pulmonary arterial vasculature, with a high spatial frequency algorithm reserved for review of lung disease.

Cardiac and respiratory motion within the adjacent lung may produce an apparent pulmonary arterial filling that is usually the result of partial volume averaging of the moving vessel (Figure 19). This is most easily recognized by viewing lung windows alongside soft tissue windows, as respiratory misregistration of vessels is more readily apparent at lung windows.

Partial volume averaging can produce an apparent filling defect, as the vessel wall with the adjacent lung is depicted as a mural filling defect on axial CT images. While this artifact had been relatively common in middle-lobe and lingular vessels that are oriented obliquely to the scan plane on single-detector CT scanners, the current use of MDCT with 0.625- to 1.0-mm nominal collimation width and the display of high-quality oblique reconstructions from isotropic volumetric data sets has reduced the frequency of this artifact.




An uncommon but often overlooked artifact that can result in an apparent intraluminal filling defect is a localized increase in pulmonary arterial resistance in vessels supplying consolidated regions of the lung that produces poor opacification of peripheral pulmonary artery branches during the rapid MDCT acquisition (Figure 20). This is usually recognized as relative underfilling of the artery branch as opposed to the identification of a discrete intraluminal filling defect, although repeat examination with a longer scan delay or, occasionally, with catheter pulmonary angiography may be necessary for clarification.

Finally, small incompletely obstructing clots can be obscured during CT interpretation when an improperly narrow window width and level are used to view the CT data set. While the use of an adequately wide window width that is equal to the mean attenuation value within the main pulmonary artery plus 2 standard deviations (29) is recommended when interpreting pulmonary CTA, most radiologists manually window and level the scans individually for each patient on a workstation in an effort to accurately detect intraluminal filling defects.

Anatomic and pathologic mimicks of PE

Various normal structures and disease entities can be mistaken for acute PE. (20) Hilar lymph nodes, particularly those situated in the right hilum between the upper and lower divisions of the right pulmonary artery, are usually easily recognized on cine viewing or coronal and sagittal reconstructions of MDCT data sets. Similarly, lobar, segmental, and subsegmental nodes are now often seen as normal peribronchovascular soft tissue densities within the substance of the lung but are readily distinguished from intraluminal filling defects on sequential images (Figure 21). Rarely, poorly opacified pulmonary veins can be mistaken for arterial emboli. This is more often seen with the faster gantry rotation and rapid scan acquisition times of current 16- through 64-detector CT scanners, where the scan acquisition can occur during the early phase of pulmonary venous contrast return with peripheral contrast and central unopacified blood within the vein as a result of laminar flow mimicking a filling defect of PE (Figure 22).



Mucus plugs within vertically oriented airways in patients with bronchitis or asthma are usually readily distinguished from emboli by noting the accompanying opacified artery and by following the mucus-filled bronchi proximally to the better-aerated central bronchi (Figure 23). As discussed previously, primary or metastatic tumors of the pulmonary arterial vasculature can, on occasion, mimic the appearance of acute PE.

Accuracy and clinical utility of pulmonary CTA

Early reports describing the use of spiral pulmonary CTA for PE diagnosis focused on the accuracy of the test as compared with other diagnostic imaging studies that until the mid-1990s were the primary methods of evaluating patients with suspected PE: ventilation-perfusion (V/Q) lung scanning and conventional pulmonary angiography. Studies that compared single-detector spiral CT pulmonary angiography with V/Q scanning showed that CT was a more accurate examination with higher specificity, particularly in those patients with underlying cardiopulmonary disease in whom the vast majority (90% in the PIOPED 1 study) have indeterminate V/Q scan results. (30) Furthermore, pulmonary CTA has consistently shown superior interobserver agreement for study interpretation as compared with V/Q scanning. (30) Subsequent studies compared CT pulmonary angiography with conventional pulmonary angiography and found that single-detector spiral CT had high sensitivity for segmental or larger PE but was limited in the detection of subsegmental emboli. More recent studies using MDCT technology with 1-mm collimation describe the routine visualization of subsegmental (4th to 5th order) branches, and this has translated into improved sensitivity for subsegmental emboli. (31-33) In addition, it is likely that isolated subsegmental emboli are uncommon in patients with PE, occurring in less than 10% of all patients. (34,35) Furthermore, there is considerable controversy regarding the clinical significance of subsegmental emboli, as not all patients with subsegmental emboli require treatment. (36,37) This, combined with the fact that the detection of subsegmental emboli is difficult on conventional pulmonary angiography (an invasive test not widely utilized for definitive diagnosis and with significant interobserver variability), has supported the use of spiral CT for PE diagnosis.


Recognizing that pulmonary CTA may miss very small emboli but noting that the significance and frequency of such emboli are uncertain, investigators in the last 5 years have focused less on the diagnostic accuracy of CT and more on the negative predictive value of spiral CT. Multiple studies that assessed the utility of MDCT have shown a very high (98% to 99%) negative predictive value for a diagnostic-quality negative study (albeit when interpreted by experts), suggesting that anticoagulation can be safely withheld in such patients. (37,38) Most recently, several investigators have evaluated the use of multidetector pulmonary CTA within a diagnostic algorithm that included the use of pretest probability for PE, D-dimer blood assay, and lower-extremity ultrasound for detection of concomitant DVT. The focus of these papers was not limited to the role of MDCT in the diagnostic evaluation of patients with suspected PE but attempted to make use of clinical probability scores such as the Wells criteria, which reliably risk-stratify patients as to the pretest likelihood of PE, along with a sensitive, rapidly available D-dimer assay that, when negative, confidently excludes PE in low-risk patients. A recent prospective study of pulmonary CTA in the diagnosis of PE utilized a dichomotous version of the Wells criteria, with those patients categorized as "unlikely" to have PE having no further diagnostic testing, and with those determined "likely" to have PE undergoing single- or multidetector CT. The main outcome measure was symptomatic or fatal venous thromboembolism (VTE). During a 3-month follow-up period for untreated patients, the rate of recurrent VTE was 0.5% for those classified as unlikely to have PE (with normal D-dimer tests), while those who were deemed clinically likely to have PE (with abnormal D-dimers and subsequent negative CT examinations) had a 1.3% incidence of recurrent VTE. (39) The authors of this study concluded that a diagnostic algorithm that incorporates simple clinical decision stratification, D-dimer testing, and pulmonary CTA is effective in the evaluation and management of patients with clinically suspected PE, with a low risk of recurrent VTE in untreated patients. The results of this and similar studies suggest that the use of nonimaging criteria to identify low-risk patients, with the judicious use of MDCT in moderate-to-high-risk patients, can achieve high diagnostic accuracy in the diagnosis of PE while limiting unnecessary testing in low-risk patients.



In addition to its accuracy in the assessment of patients with suspected PE, MDCT has great utility in the assessment of patients with acute chest symptoms in whom PE is but one of several diagnostic considerations. The identification of alternate conditions that may account for the patients' symptoms is perhaps the greatest strength of multidetector pulmonary CTA in the evaluation of suspected PE, as up to 90% of patients studied have no embolism identified on CT. In this regard, CT offers diagnostic advantages over alternative imaging modalities that do not provide diagnostic information beyond the status of lung perfusion (and ventilation in V/Q scans). Some of the conditions that clinically mimic PE (including pneumothorax, rib fracture, pneumonia, and pleural effusion) are readily detected on conventional chest radiography, which therefore should be obtained prior to CT in all patients. (40) Nevertheless, multiple studies have shown that findings that account for patients' symptoms and signs of acute chest disease can be seen in a majority of patients in whom pulmonary CTA is negative for PE. In 1 study, 67% of patients with negative spiral pulmonary CTA showed findings that were suggestive or confirmatory of an alternate clinical diagnosis. (41) Some of the entities commonly detected on multidetector pulmonary CTA include rib, cartilage, and other chest wall abnormalities, pneumomediastinum, pneumothorax, pneumonia (Figure 24), pleural effusion, or empyema, malignancy, aortic dissection, esophageal conditions, postoperative changes, and acute myocardial infarction (Figure 25).

Adverse effects

As discussed above, current evidence suggests that CT is best used selectively in the evaluation of potential PE, as there are numerous studies confirming its utility and cost-effectiveness when used within an appropriate clinical algorithm. (42,43) An underappreciated risk to patients who undergo MDCT for evaluation of possible PE is the radiation exposure from the examination. Particular caution must be exercised in pediatric patients because of the increased lifetime risk of radiation-induced malignancy and in female patients because of the radiation dose to the female breast. In a recent study, a breast dose of 20 mGy was calculated for pulmonary CTA as compared with a dose of 3 mGy for standard 2-view screening mammography. (44)

Other than the selective use of MDCT for the evaluation of patients with suspected PE, other efforts aimed at reducing radiation exposure from the MDCT examination itself include low-dose exposure techniques, dose modulation with automated dose reduction in the axial and craniocaudal scan planes, and the use of radiation attenuation devices affixed to the chest wall to reduce absorbed radiation dose to the female breasts. A recent study that compared standard and simulated low-dose pulmonary multidetector CTA showed comparable diagnostic parameters in patients evaluated for PE with a ninefold reduction in mAs. (45)

CT venography

Pulmonary embolism represents one end of the spectrum of venous thromboembolic disease, with untreated DVT a major risk factor for PE. For this reason, several investigators have developed and studied the feasibility and utility of performing indirect CT venography as an adjunct to pulmonary CTA in patients with suspected PE to increase the overall diagnostic yield and negative predictive value of CT for venous thromboembolic disease (Figure 26). In addition to the detection of lower-extremity DVT, indirect CT venography provides evaluation of the pelvic and abdominal veins, which is particularly important in patients with a history of gynecological malignancy or prior surgery.




The sensitivity of CT venography compares favorably with duplex ultrasound evaluation of the lower extremities, with >95% sensitivity for lower-extremity DVT and abdominopelvic extension of thrombus or isolated caval or pelvic thrombus seen in as many as one third of patients in some series. (46) In several studies of combined pulmonary CTA with CT venography, DVT was detected in the absence of PE in as many as 20% of patients, suggesting a significant incremental gain in sensitivity for the detection of venous thrombembolic disease with the combined technique. (47) In the subset of patients with non-diagnostic-quality pulmonary CTA, particularly those in the critical care setting, the CT venogram can help salvage the diagnostic utility of the combined examination by detecting evidence of DVT.

CT venography is performed 2.5 to 3 minutes after the initiation of contrast injection for pulmonary CTA, with scans obtained using contiguous axial or helical 5- to 10-mm collimated scans from the popliteal fossa to the diaphragms. While no additional contrast beyond that administered for the pulmonary CTA is needed, there is an increase in radiation dose and cost to the patient; therefore, the technique has not been uniformly adopted by all operators. Because of variation in venous return or arterial inflow problems due to atherosclerosis, care must be taken in distinguishing acute embolism from flow or early contrast mixing artifact.


Pulmonary CTA is a robust, noninvasive technique that is highly accurate and cost-effective in identifying PE and those conditions that mimic PE clinically. The increased availability of MDCT with greater spatial and temporal resolution should further improve the sensitivity, negative predictive value, and overall clinical utility of this technique. Importantly, radiologists should be aware of the various technical and interpretive pitfalls encountered on pulmonary CTA and should use the study judiciously according to accepted clinical algorithms, particularly given the associated radiation exposure to the patient. CT venography can incrementally increase the sensitivity for detection of thromboembolic disease and may be particularly useful in ICU patients with poor-quality pulmonary CTA and in postoperative patients with a higher risk of abdominopelvic DVT.


(1.) Schoepf UJ. Diagnosing pulmonary embolism: Time to rewrite the textbooks. Int J Cardiovasc Imaging. 2005;21:155-163.

(2.) Quiroz R, Schoepf UJ. Spiral CT for pulmonary embolism: The paradigm has shifted. Am Heart Hosp J. 2003;1:281-288.

(3.) Quiroz R, Kucher N, Zou KH, et al. Clinical validity of negative computed tomography scan in patients with suspected pulmonary embolism: A systematic review. JAMA. 2005:293:2012-2017.

(4.) Prologo JD, Gilkeson RC, Diaz M, Asaad J. CT pulmonary angiography: A comparative analysis of the utilization patterns in emergency department and hospitalized patients between 1998 and 2003. AJR Am J Roentgenol. 2004;183:1093-1096.

(5.) Storto ML, Di Credico A, Guido F, et al. Incidental detection of pulmonary emboli on routine MDCT of the chest. AJR Am J Roentgenol. 2005;184:264-267.

(6.) Patel S, Kazerooni EA. Helical CT for the evaluation of acute pulmonary embolism. AJR Am J Roentgenol. 2005;185:135-149.

(7.) Prologo JD, Gilkeson RC, Diaz M. The effect of single detector CT versus MDCT on clinical outcomes in patients with suspected acute pulmonary embolsm and negative results on CT pulmonary angiography. AJR Am J Roentgenol. 2005; 184: 1231-1235.

(8.) Winer-Muram HT, Rydberg J, Johnson MS. Suspected acute pulmonary embolism: Evaluation with multi-detector row CT versus digital subtraction pulmonary arteriography. Radiology. 2004; 233: 806-815.

(9.) Russo V, Piva T, Lovato L, et al. Multidetector CT: A new gold standard in the diagnosis of pulmonary embolism? State of the art and diagnostic algorithms [in Italian and English]. Radiol Med (Torino). 2005; 109(1-2):49-61.

(10.) Patel S, Kazerooni EA, Cascade PN. Pulmonary embolism: Optimization of small pulmonary artery visualization at multi-detector CT. Radiology. 2003; 227: 455-460.

(11.) Bedard JP, Blais C, Patenaude YG, Monga E. Pulmonary embolism: Prospective comparison of iso-osmolar and low-osmolarity nonionic contrast agents for contrast enhancement at CT angiography. Radiology. 2005;234:929-933.

(12.) Marten K, Engelke C, Obenauer S, et al. Diagnostic performance of retrospectively ECG-gated multislice CT of acute pulmonary embolism [in German]. Rofo. 2003;175:1490-1495.

(13.) Marten K, Engelke C, Funke M, et al. ECG-gated mutlislice spiral CT for diagnosis of acute pulmonary embolism. Clin Radiol. 2003;58:862-868.

(14.) Morcos SK. Prevention of contrast mediainduced nephrotoxicity after angiographic procedures. J Vasc Interv Radiol. 2005;16:13-23.

(15.) Costello P, Gupta KB. Pulmonary embolism: Imaging modalities-V/Q scan, spiral (helical) CT, and MRI. Semin Vasc Med. 2001;1:155-164.

(16.) Kanne JP, Lalani TA. Role of computed tomography and magnetic resonance imaging for deep venous thrombosis and pulmonary embolism. Circulation. 2004;109(suppl I):I15-I21.

(17.) Remy-Jardin M, Dequiedt P, Ertzbischoff O, et al. Safety and effectiveness of gadolinium-enhanced multi-detector row spiral CT angiography of the chest: Preliminary results in 37 patients with contraindications to iodinated contrast agents. Radiology. 2005; 235:819-826.

(18.) Huda W. When a pregnant patient has a suspected pulmonary embolism, what are the typical embryo doses from a chest CT and a ventilation/perfusion study? Pediatric Radiol. 2005; 35: 452-453.

(19.) Coche E, Verschuren F, Keyeux A, et al. Diagnosis of acute pulmonary embolism in outpatients: Comparison of thin collimation multi-detector row spiral CT and planar ventilation-perfusion scintigraphy. Radiology. 2003;229:757-765.

(20.) Wittram C, Maher MM, Yoo AJ, et al. CT angiography of pulmonary embolism: Diagnostic criteria and causes of misdiagnosis. RadioGraphics. 2004; 24:1219-1238.

(21.) Cobelli R, Zompatori M, De Luca G, et al. Clinical usefulness of computed tomography without contrast injection in the evaluation of acute pulmonary embolism. J Comput Assist Tomogr. 2005; 29:6-12.

(22.) Arakawa H, Kurihara Y, Sasaka K, et al. Air trapping on CT of patients with pulmonary embolism. AJR Am J Roentgenol. 2002;178:1201-1207

(23.) Mansencal N, Joseph T, Vieillard-Baron A, et al. Diagnosis of right ventricular dysfunction in acute pulmonary embolism using helical computed tomography. Am J Cardiol. 2005;95:1260-1263.

(24.) Kearon C. Natural history of venous thromboembolism. Semin Vasc Med. 2001;1(1):27-37.

(25.) Wittram C, Maher MM, Halpern E, Shepard JA. Attenuation of acute and chronic pulmonary emboli. Radiology. 2005;235:1050-1054.

(26.) Yi CA, Lee KS, Choe YH, et al. Computed tomography in pulmonary artery sarcoma: Distinguishing features from pulmonary embolic disease. J Comput Assist Tomogr. 2004;28:34-39.

(27.) Gosselin MV, Rassner UA, Thieszen SL, et al. Contrast dynamics during CT pulmonary angiogram: Analysis of an inspiration associated artifact. J Thorac Imaging. 2004;19:1-7.

(28.) Tay KL, Ridley LJ. Contrast column interruption artefact in computed tomography pulmonary angiography. Australas Radiol. 2005;49:338-341.

(29.) Bae KT, Mody GN, Balfe DM, et al. CT depiction of pulmonary emboli: Display window settings. Radiology. 2005;236:677-684.

(30.) 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.

(31.) Coche E, Pawlak S, Dechambre S, Maldague B. Peripheral pulmonary arteries: Identification at multislice spiral CT with 3D reconstruction. Eur Radiol. 2003;13:815-822.

(32.) 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.

(33.) 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.

(34.) Stein PD, Henry JW, Gottschalk A. Reassessment of pulmonary angiography for the diagnosis of pulmonary embolism: Relation of interpreter agreement to the order of the involved pulmonary arterial branch. Radiology. 1999;210:689-691.

(35.) Diffin DC, Leyendecker JR, Johnson SP, et al. Effect of anatomic distribution of pulmonary emboli on interobserver agreement in the interpretation of pulmonary angiography. AJR Am J Roentgenol. 1998; 171:1085-1090.

(36.) Eyer BA, Goodman LR, Washington L. Clinicians' response to radiologists' report of isolated subsegmental pulmonary embolism or inconclusive interpretation of pulmonary embolism using MDCT. AJR Am J Roentgenol. 2005;184:623-628.

(37.) Swensen SJ, Sheedy PF 2nd, Ryu JH, et al. Outcomes after withholding anticoagulation from patients with suspected acute pulmonary embolism and negative computed tomographic findings: A cohort study. Mayo Clin Proc. 2002; 77:130-138.

(38.) Quiroz R, Kucher N, Zou KH, et al. Clinical validity of a negative computed tomography scan in patients with suspected pulmonary embolism: A systematic review. JAMA. 2005;293:2012-2017.

(39.) van Belle A, Buller HR, Huisman MV, et al; Christopher Study Investigators. Effectiveness of managing suspected pulmonary embolism using an algorithm combining clinical probability, D-dimer testing, and computed tomography. JAMA. 2006; 295:172-179.

(40.) Daftary A, Gregory M, Daftary A, et al. Chest radiograph as a triage tool in the imaging-based diagnosis of pulmonary embolism. AJR Am J Roentgenol. 2005;185:132-134.

(41.) Kim KI, Muller NL, Mayo JR. Clinically suspected pulmonary embolism: Utility of spiral CT. Radiology. 1999;210:693-697.

(42.) Quiroz R, Schoepf UJ. CT pulmonary angiography for acute pulmonary embolism: Cost-effectiveness analysis and review of the literature. Semin Roentgenol. 2005;40:20-24.

(43.) van Erkel AR, van den Hout WB, Pattynama PM. International differences in health care costs in Europe and the United States: Do these affect the cost-effectiveness of diagnostic strategies for pulmonary embolism? Eur Radiol. 1999:9: 1926-1931.

(44.) Parker MS, Hui FK, Camacho MA, et al. Female breast radiation exposure during CT pulmonary angiography. AJR Am J Roentgenol. 2005; 185:1228-1233.

(45.) Tack D, De Maertelaer V, Petit W. Multi-detector row CT pulmonary angiography: Comparison of standard-dose and simulated low-dose techniques. Radiology. 2005;236:318-325.

(46.) Loud PA, Katz DS, Klippenstein DL, et al. Combined CT venography and pulmonary angiography in suspected thromboembolic disease: Diagnostic accuracy for deep venous evaluation. AJR Am J Roentgenol. 2000;174:61-65.

(47.) Cham MD, Yankelevitz DF, Henschke CI. Thromboembolic disease detection at indirect CT venography versus CT pulmonary angiography. Radiology. 2005;234:591-594.

Dr. Gentchos is an Assistant Professor and Dr. Klein is a Professor, Director of Thoracic Radiology, Department of Radiology, Fletcher Allen Health Care and the University of Vermont College of Medicine, Burlington, VT.

George Gentchos, MD, and Jeffrey S. Klein, MD
Table 1. 64-row CT pulmonary
angiography protocol

Parameter Value

Scan delay 15 seconds
Field of view Widest rib to
 widest rib

Syringes (2) 100 mL contrast ([dagger])
 20 mL saline
Injection rate 5 mL/sec
Slice thickness 0.9 mm
Increment 0.45 mm
kV 120
mAs 300
Collimation 64 x 0.625 mm
Pitch 1.172
Rotation time 0.50 sec
Filter B (soft tissue
Matrix 512 x 512

* Protocol is based on the use of the Brilliance
64 Slice CT scanner (Philips Medical
Systems, Bothell, WA).

([dagger]) Contrast is Isovue 370 (GE Healthcare,

Princeton, NJ).

Table 2. Common technical problems encountered
in CT pulmonary angiography

Technical problems Solutions

Poor opacification of 1. Consider an automated timing
pulmonaryvasculature bolus or test injection
 2. Use 18 gauge IV line
 3. Flow rate to 5 mL/sec

SVC streak artifact 1. Saline flush in second auto injector
 2. Scan caudal to cranial

Cardiac pulsation artifact ECG gating
Respiratory motion Hyperventilate and administer
 supplemental oxygen to patient priorto
Obese patient or increased 1. Increase kV to 140
image noise 2. Increase collimation to 1.4 mm

IV = intravenous; SVC = superior vena cava; ECG = electrocardiographic
COPYRIGHT 2007 Anderson Publishing Ltd.
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2007 Gale, Cengage Learning. All rights reserved.

 Reader Opinion




Article Details
Printer friendly Cite/link Email Feedback
Title Annotation:computed tomography
Author:Gentchos, George; Klein, Jeffrey S.
Publication:Applied Radiology
Article Type:Clinical report
Geographic Code:1USA
Date:Oct 1, 2007
Previous Article:Radiological case of the month: C. Frank Gould, MD, Brian J. Fortman, MD, Justin Q. Ly, MD, Scot E. Campbell, MD, and Douglas P. Beall, MD.
Next Article:Accessory ossicles and sesamoid bones: spectrum of pathology and imaging evaluation.

Related Articles
Mediastinal staging of non-small cell lung carcinoma using computed and positron-emission tomography.
Do all patients with suspected appendicitis benefit from CT imaging in community-based emergency departments?
Getting burned: radiation exposure from CT scans.
Use of combined PET/CT imaging in evaluation of the solitary pulmonary nodule: principles, techniques, and pitfalls.
Things that go bump in the night: neuroradiology emergencies.
Some aspects of computed tomography in Russia.

Terms of use | Copyright © 2015 Farlex, Inc. | Feedback | For webmasters