| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Clinical Observations |
1 Department of Diagnostic Radiology, Yale University School of Medicine and
Yale New Haven Hospital, 333 Cedar Street, New Haven, CT 06504.
2 Yale New Haven Hospital, New Haven, CT.
3 Institute for Neurodegenerative Disorders, New Haven, CT.
Received March 10, 2004;
accepted after revision September 23, 2004.
Presented in part at the 2002 annual meeting of the Radiological Society of
North America, Chicago, IL.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CONCLUSION. Chest radiograph can be a valuable triage tool in deciding an appropriate technique for imaging PE, and can yield more definitive diagnoses.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Traditionally, a combination of ventilation-perfusion scintigraphy and lower-extremity venous Doppler examinations has been used to assess the risk for pulmonary thromboembolic disease [3, 4]. Patients with a high clinical suspicion for PE or with intermediate or indeterminate results from the ventilation-perfusion scintigraphy or lower-extremity venous Doppler studies were further evaluated with pulmonary angiography. Perfusion abnormalities in patients with underlying nonthromboembolic cardiopulmonary disease have been described previously in detail; although the addition of ventilation imaging has improved the specificity of the test, approximately 30-40% of patients still have intermediate probability studies [3, 5, 6].
The addition of pulmonary CT angiography (CTA) has been a valuable tool in the radiologist's arsenal for diagnosing PE [7], and certain studies suggest that this is an adequate test for excluding PE [7-10]. Studies in patients with cardiopulmonary disease have suggested a 1.8-4.9% recurrence rate of PE after negative CTA, depending on whether the patients received anticoagulation [11]. Further, in patients with an underlying lung disease, the negative predictive value of CTA is not significantly affected [12]. The advent of MDCT has resulted in a decrease in the number of CTAs that are limited by motion artifact or contrast bolus deficiencies. However, there still appears to be inadequate filling to the subsegmental level in approximately 13% of subjects [13]. The ability of pulmonary CTA to diagnose central PE is relatively high (> 90%), while it is less reliable in smaller, more distal vessels (50-60%) [14].
Recent studies have suggested more definitive results when a combination of ventilation-perfusion scintigraphy and pulmonary CTA is used [15]. Because of the high radiation burden of CTAs and the additional risks of IV contrast media, it has been suggested that ventilation-perfusion scintigraphy scan be used more frequently and to limit the use of CTAs to those situations in which ventilation-perfusion scinitigraphy results are indeterminate [16]. However, this process would result in an unnecessary delay and duplication of studies in approximately 39% of patients who would have intermediate probability studies [3].
The purpose of our study was to determine if the chest radiograph could be used as a triage tool in determining the appropriate study of choice to evaluate for the presence of PE, thus expediting diagnosis and therapy.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We performed a retrospective review of reported results for ventilation-perfusion scintigraphy scans and pulmonary CTAs from July 1 to October 31 in 1996 and 2001, before and after the introduction of this policy. The study was performed at a tertiary care medical center with the approval of the human investigation committee.
None of the studies, including radiographs, ventilation-perfusion scintigraphy scans, or CTAs, were reinterpreted; rather, the data were collected from the computerized radiology report system from the finalized reports in the system.
All ventilation-perfusion scintigraphy scans were performed using xenon-133 ventilation and technetium-99m microaggregated albumin (MAA). All CTAs were performed using 4-MDCT with 1.3-mm axial sections and sagittal and coronal reformats.
The chest radiographs with large opacities, infiltrates, lobar atelectasis, chronic obstructive pulmonary disease, advanced interstitial lung disease with honeycombing, lobectomy with prominent postoperative changes, fibrothorax, moderate effusions, moderate pulmonary edema, or mediastinal adenopathy/tumor were designated as abnormal. Abnormalities such as cardiomegaly and minimal pulmonary vascular congestion were considered to be normal.
Ventilation-perfusion scintigraphy studies interpreted as high, low, very low probability, or normal were considered to be definitive diagnoses, while intermediate or inderterminate studies were considered to be nondefinitive. Studies were interpreted using the modified Prospective Investigation of Pulmonary Embolism Diagnosis (PIOPED) criteria [3]. Pulmonary CTAs that were interpreted as positive or negative were considered definitive diagnoses, while those limited by motion artifact, sub-optimal vessel opacification, or body habitus were considered nondefinitive.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
In 2001, while there were 221/370 patients with normal chest radiographs, only 157 ventilation-perfusion scintigraphy scans were performed. The remaining 64 patients under-went CTAs because they were worked up after 11 pm, when ventilation-perfusion scintigraphy scans are generally not available unless there is a contraindication to CTA. Similarly, in 2001, 43 patients with abnormal chest radiographs received ventilation-perfusion scintigraphy scans rather than CT scans. The main reason for this was a contraindication to contrast media (i.e., chronic renal insufficiency or history of contrast allergy).
The number of patients evaluated with ventilation-perfusion scintigraphy decreased from 214 in 1996 to 157 in 2001, representing a 46.7% decrease. Definitive interpretations increased from 70% (150/214) in 1996 to 87% (137/157) in 2001.
Overall, definitive diagnoses increased from 70% (150/214) in 1996 to 91% (336/370) in 2001 after the change in triage method.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
With the advent of MDCT, pulmonary CTA has rapidly become an important technique in the noninvasive diagnosis of PE. CT is readily available even after hours and is less susceptible to underlying lung abnormality. Several studies have shown that patients with negative CTA do not need further evaluation even if they have an underlying cardiopulmonary disease [12, 13, 19, 20]. The frequency of indeterminate/inadequate results for this technique has been reported as about 6% [21].
With the availability of ventilation-perfusion scintigraphy scans and CTA, the role of each individual technique in the workup of pulmonary embolism is evolving. Clinicians use both depending on individual preference and availability. Some prefer CTA because the number of indeterminate or inadequate studies tends to be fewer. Instead of relegating ventilation-perfusion scintigraphy scanning to the background, we adopted a novel triage method toward ventilation-perfusion scintigraphy scans or CTA, playing on the inherent strengths and limitations of both techniques. The triage decision was essentially based on the chest radiograph. If focal radiographic abnormalities were present, the patient received CTA, the technique that is less affected by background lung disease. However, if the radiograph was normal, a ventilation-perfusion scintigraphy scan was performed. This protocol has worked well and we continue to follow it in our daily clinical practice.
In the two time periods compared in our study, a 70% increase occurred in the number of patients evaluated for PE. This may be due to the increased availability of both CTA and ventilation-perfusion scintigraphy scans to clinicians. This has resulted in a more definitive role for imaging in the diagnosis of PE. In 1996, definitive diagnosis with ventilation-perfusion scintigraphy scanning was made 70% of the time. In 2001, this number had increased to 91% (ventilation-perfusion scintigraphy scans definitive 87% of the time versus CTA, 93% of the time).
Although a rigorous analysis is difficult to perform given that the time periods represent two different patient populations, the increased percentage of definitive results in 2001 is striking. Some of this can be attributed to the CTA, in which a result is usually positive or negative (or inadequate study); however, there has also been a simultaneous decrease in the number of indeterminate ventilation-perfusion scintigraphy scans—30% in 1996, down to 13% in 2001. The concomitant increase in definitive ventilation-perfusion scintigraphy scans (70% in 1996 and up to 87% in 2001) was observed mostly in the category of low-probability ventilation-perfusion scintigraphy scans in which the proportion increased from 60% to 85%. It appears that using a normal chest radiograph results in a 17% decrease in the number of indeterminate ventilation-perfusion scintigraphy scans with a corresponding 15% increase in low-probability scans (considered a definitive result for the purposes of this study). CTA results in a 93% definitive rate of interpretation versus 87% for the ventilation-perfusion scintigraphy scans, a benefit of 6%.
Limitations of this study included the absence of clinical follow-up or pulmonary angiograms on these patients to confirm the presence or absence of PE. A definitive result in this study was not deemed equal to a final result, as the majority of these patients were not subjected to pulmonary angiography (the gold standard). In addition, patient selection was not based on the d-dimer assay, a serum test with a high negative predictive value in excluding PE. The latter may lead to significant changes in the number of patients imaged. At present, the PIOPED II study is underway, along with plans to use a more composite gold standard for confirming the presence or absence of PE and modify current criteria for interpreting ventilation-perfusion scintigraphy.
In conclusion, using the normal chest radiograph when deciding to proceed with a ventilation-perfusion scintigraphy scan instead of CTA has resulted in an increase in definitive diagnoses compared with the use of ventilation-perfusion scintigraphy scan alone and provides a viable triage tool to optimize imaging workup of patients with suspected PE.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  A. F. Scarsbrook and F. V. Gleeson Investigating suspected pulmonary embolism in pregnancy BMJ, February 24, 2007; 334(7590): 418 - 419. [Full Text] [PDF]
|
|
|
|
 |
|
 M R Jones and J H Reid Emergency chest radiology: thoracic aortic disease and pulmonary embolism Imaging, September 1, 2006; 18(3): 122 - 138. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. N. C. Milne Comment on Chest Radiograph as a Triage Tool Am. J. Roentgenol., April 1, 2006; 186(4): 1198 - 1198. [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
