HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by King, V. Articles by Ginsberg, M. S. Search for Related Content PubMed PubMed Citation Articles by King, V. Articles by Ginsberg, M. S.

D-Dimer Assay to Exclude Pulmonary Embolism in High-Risk Oncologic Population: Correlation with CT Pulmonary Angiography in an Urgent Care Setting¹

¹ From the Department of Radiology (V.K., M.S.G.), Department of Medicine, Urgent Care Service (A.A.V.), Department of Epidemiology and Biostatistics (C.S.M.), and Clinical Laboratories (L.J.S.), Memorial Sloan-Kettering Cancer Center, 1275 York Ave, New York, NY 10021. Received June 1, 2007; revision requested July 27; revision received September 26; final version accepted January 4, 2008. Address correspondence to M.S.G. (e-mail: ginsberm{at}mskcc.org).

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 
Purpose: To prospectively evaluate (a) the diagnostic performance of D-dimer assay for pulmonary embolism (PE) in an oncologic population by using computed tomographic (CT) pulmonary angiography as the reference standard, (b) the association between PE location and assay sensitivity, and (c) the association between assay results and clinical factors that raise suspicion of PE.

Materials and Methods: This HIPAA-compliant study had institutional review board approval; informed consent was obtained. Five hundred thirty-one consecutive patients were clinically suspected of having PE; 201 were enrolled (72 men, 129 women; median age, 61 years) and underwent CT pulmonary angiography and D-dimer assay. Relevant clinical history, symptoms, and signs were recorded. CT images were interpreted, and the location of emboli was recorded. The negative predictive value (NPV), positive predictive value (PPV), sensitivity, specificity, and diagnostic likelihood ratios of the D-dimer assay results were calculated.

Results: Forty-three patients (21%) had pulmonary emboli at CT. D-Dimer results were positive in 171 patents (85%). The NPV and sensitivity were 97% and 98%, respectively. The specificity and PPV were 18% and 25%, respectively. No association was shown between clinical history, symptoms, or signs and NPV, PPV, sensitivity, or specificity or between location of PE and sensitivity.

Conclusion: D-Dimer results have high NPV and sensitivity for PE in oncologic patients and, if negative, can be used to exclude PE in this population. Combining the assay with clinical symptoms and signs did not substantially change NPV, PPV, sensitivity, or specificity.

© RSNA, 2008

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 
Computed tomographic (CT) pulmonary angiography has become a commonly used noninvasive modality for the diagnosis of pulmonary embolism (PE) (1). This is attributed to the fact that it is readily available and is a quick study to perform, with a relatively high sensitivity and specificity for PE (2,3). The widespread use of CT pulmonary angiography and the high incidence of patients suspected of having PE have led to a high volume of studies performed to rule out PE, which in turn have resulted in substantial costs to hospitals (4,5). CT is also not without risks, such as adverse reaction to intravenous contrast material, renal failure due to contrast material nephrotoxicity, and exposure to radiation (1,6).

In many emergency departments, the D-dimer assay is used in conjunction with imaging studies and clinical signs and symptoms to exclude PE. Some studies (7–11) have suggested that in patients with a low clinical probability, the D-dimer assay may be used to reliably exclude PE or deep venous thrombosis (DVT). Another study (12), however, indicated that it may be inadvisable to rely solely on this assay to exclude this serious and potentially fatal disease.

D-Dimer is a marker for the presence of stabilized fibrin. When present in the circulation, D-dimer serves as an indirect indicator of thrombotic activity. Immunoturbidimetric assays are a newer generation of rapid, automated, quantitative D-dimer tests based on the agglutination of microlatex particles coated with monoclonal antibodies specific for D-dimer. These assays have been reported to have a negative predictive value (NPV) of 94% and a sensitivity of 97% for PE (13). The purpose of our study was to prospectively evaluate the diagnostic performance of the D-dimer assay for PE in an oncologic population by using CT pulmonary angiography as the reference standard, to assess the association between the location of the PE and the sensitivity of the assay, and to examine the association between the assay and clinical factors that raise suspicion of PE.

	MATERIALS AND METHODS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 
The Institutional Review Board granted approval for our prospective Health Insurance Portability and Accountability Act–compliant study. Informed consent was obtained from all participants.

Patients
Between June 24, 2005, and June 12, 2006, 531 consecutive adult (minimum age, 21 years) outpatients presenting to the Urgent Care Center of a tertiary care cancer center were suspected of having PE and were referred for CT pulmonary angiography (Figure). The Urgent Care Center is a 24-hour acute and semiacute care center for outpatients being treated by physicians at the hospital. The need for a CT pulmonary angiogram was determined by the Urgent Care Center physician (including A.A.V.) by using a standard practice-based evaluation, which included clinical history, symptoms, and vital signs. From this group, 214 nonconsecutive patients were enrolled in our study.

View larger version (16K): [in this window] [in a new window] [Download PPT slide]   Flow diagram shows number of patients who met inclusion criteria and underwent reference standard and index tests. Numbers of patients with positive and negative CT findings and positive D-dimer result do not sum to 171 because there was one patient who had an equivocal CT pulmonary angiogram and a positive D-dimer result. CTPA = CT pulmonary angiography, IV = intravenous.

CT Protocol and Image Evaluation
Axial CT was performed with four- or 16-section multidetector CT scanners (Qx/i and Lightspeed; GE Medical Systems, Milwaukee, Wis). On the four-section scanner, images were acquired from the lung bases to the apices, with 2.5-mm section thickness and 2.5-mm spacing. Images from the four-section scanner were reconstructed to 1.25-mm section thickness with 1.0-mm spacing from 2 cm below the diaphragm to the aortic arch. On the 16-section scanner, images were acquired from the lung bases to the apices, with 5-mm section thickness and 5-mm spacing. Images from the 16-section scanner were reconstructed to 1.25-mm section thickness with 0.8-mm spacing from 2 cm below the diaphragm to the aortic arch. All images were sent to a picture archiving and communication system (GE PACS; GE Medical Systems).

Imaging and contrast agent protocols at our institution were periodically modified for reasons unrelated to the study. Patients were imaged according to the following protocols: From June 24, 2005, to September 28, 2005, (n = 73 patients), CT of the chest or chest, abdomen, and pelvis was performed by using 100–150 mL of iohexol (Omnipaque 300; GE Healthcare, Princeton, NJ). From September 30, 2005, to January 21, 2006, (n = 56), CT of the chest or chest, abdomen, and pelvis was performed by using 100–150 mL of Omnipaque 300 and an 80–120-mL saline bolus. From January 23, 2006, to June 12, 2006, (n = 72), CT of the chest was performed by using 40 mL of saline, then 80 mL of iohexol (Omnipaque 350; GE Healthcare), and finally 80 mL of saline; CT of the chest, abdomen, and pelvis was performed by using 40 mL of saline, then 150 mL of Omnipaque 300, and finally 80 mL of saline. Scans performed on the four-section scanner were done by using an injection rate of 4 mL/sec with an empiric 2-second injection-to-scan delay. On the 16-section scanner, a pseudosmart prep with a cursor placed on the main pulmonary artery was used to determine the appropriate scanning delay; the injection rate used was 4 mL/sec.

CT pulmonary angiograms were interpreted by 29 attending radiologists (including M.S.G., with 1 to more than 30 years experience interpreting chest CT images) who were blinded to the results of the D-dimer tests. Results were designated as positive, negative, or equivocal. An angiogram was considered positive for PE if there was an intraluminal filling defect on more than one contiguous axial section, with expansion of the vessel and/or an abrupt termination of the opacified vessel peripherally. An angiogram was considered negative for PE if neither of these findings was present. Equivocal angiograms were those in which there was a questionable filling defect on one or two images. Emboli were classified by their most central location as central (main pulmonary artery, left pulmonary artery, right pulmonary artery), lobar (interlobar and lobar), or small vessel (segmental and subsegmental pulmonary artery) by two authors (V.K., M.S.G.). Whenever present, factors limiting interpretation of the CT pulmonary angiogram were recorded (V.K., M.S.G.); these included suboptimal timing of the contrast agent bolus, artifacts due to beam hardening and motion, and substantial confounding pathologic findings. The results of additional imaging studies (eg, repeat CT pulmonary angiography, lower extremity ultrasonography [US], ventilation-perfusion scan) performed within 2 days of the initial D-dimer assay were also recorded (V.K., M.S.G.).

D-Dimer Assay
D-Dimer sample analysis was performed by using the STA Liatest D-dimer assay kit (Diagnostica Stago, Asnieres, France). Blood was collected in 0.109 mol/L (3.2%) trisodium citrate anticoagulant following National Committee for Clinical Laboratory Standards guidelines H3-A3 and H21-A4. Samples were centrifuged for 15 minutes at 2500g. All samples were processed and analyzed within 1–2 hours of collection by using either the STA-Compact or STA-R coagulation analyzer (Diagnostica Stago). These utilized an immunoturbidimetric method to perform a quantitative determination of D-dimer in plasma. D-Dimer values were expressed as micrograms per milliliter of D-dimer units (DDU). A normal D-dimer value was less than 0.21 µg/mL DDU, and a positive D-dimer value was greater than or equal to 0.21 µg/mL DDU. Our laboratory used a lower standard cutoff value for the D-dimer than that suggested by the manufacturer because our numbers were validated to within 2 standard deviations when our normal range was established, rather than the 3 standard deviation range typically used by the manufacturer. The D-dimer assay was performed within 24 hours of CT pulmonary angiography (range, 23 hours 47 minutes before–23 hours 59 minutes after). The median time difference between the CT pulmonary angiogram and D-dimer assay was 3 hours 6 minutes. The reader of the D-dimer assay was blinded to the CT pulmonary angiogram results and other clinical information.

Data Reference Standard and Statistical Analysis
Urgent Care Center data sheets and patient records were reviewed to compile clinical and historical data (A.A.V.), imaging results (V.K., M.S.G.), and D-dimer results (V.K., M.S.G.) into a coded database for analysis. The NPV, positive predictive value (PPV), sensitivity, specificity, and positive and negative diagnostic likelihood ratios (LRs) of the D-dimer assay for detecting PE were calculated (C.S.M.) by using CT pulmonary angiography as the reference standard for PE. Confidence intervals (CIs) for proportions were estimated (C.S.M.) by using exact binomial CIs. Sensitivities of the D-dimer assay for PE in central, lobar, and small-vessel locations were calculated (V.K., M.S.G.).

The planned accrual for this study was 500 patients. The target enrollment was calculated on the basis of the expected number of patients needed to have 80% power to demonstrate that the D-dimer assay has an NPV of greater than or equal to 94% by using a one-sided .05-level test. The study was stopped for reasons that were independent of the outcomes we were studying. Basic likelihood principles state that the only components that go into estimation and inferential procedures are events that are related to the parameter(s) of interest (14). Since the reason for stopping the study was independent of these parameters, there was no adjustment in the estimation or inferential process. To test whether factors were significantly associated with assay accuracy, marginal generalized linear models were used (C.S.M.), as described by Pepe (15). Associated Wald tests were used to evaluate statistical significance, with a P value of less than .05 considered to indicate a significant result. All analyses were performed (C.S.M.) by using software (Stata 9.0 for Windows, 2005; StataCorp, College Station, Tex).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 
CT Pulmonary Angiography Results
Of the 201 patients included, 43 (21%) had a CT angiogram that was positive and 157 (78%) had a CT angiogram that was negative for PE. Only one CT angiogram (<1%) yielded an equivocal finding. In this patient, repeat CT pulmonary angiography yielded a second equivocal result; subsequently the patient was treated for PE.

Forty (20%) of the 201 angiograms were classified as being limited for the evaluation of PE. Reasons cited included poor contrast agent bolus timing on 12 (6%), beam hardening artifact on nine (4%), respiratory motion on four (2%), and other reasons (including a combination of the above factors or adjacent lung or pleural pathologic findings) on 15 (7%). The single equivocal case was limited due to respiratory motion. All other cases could be interpreted as positive or negative despite limitations.

Nineteen (9%) patients underwent additional imaging studies within 2 days of the initial CT pulmonary angiogram; three underwent repeat CT pulmonary angiography, and 16 underwent lower extremity US. No patient underwent a ventilation-perfusion scan or conventional pulmonary angiography during that period. The three CT angiograms that were repeated were classified as limited owing to poor contrast agent opacification of the pulmonary arteries. Of the 19 patients who underwent additional imaging, six had initial angiograms that were interpreted as positive; 12, as negative; and one, as equivocal. In the three patients who underwent repeat CT pulmonary angiography, the initial and repeat CT angiograms were both negative in two patients and were both equivocal in the third. In no patient did the results of the additional imaging studies contradict the original CT pulmonary angiography findings.

Of the 43 patients who had PE detected at CT pulmonary angiography, 10 (23%) had emboli in central pulmonary arteries; 15 (35%), in lobar arteries; and 18 (42%), in small arteries.

D-Dimer Assay Results
The D-dimer assay results were positive in 171 patients (85%) and negative in 30 (15%) patients. For the analysis, the patient with equivocal results, who was treated for PE, was included in the positive group. NPV, PPV, sensitivity, and specificity were not substantially affected by the inclusion of this case in the analysis. The assay had an estimated NPV of 97% (95% CI: 83%, 100%) (Table 2). The lower bound of the 95% CI indicates that the NPV is at least 83%. While the sensitivity of the assay was high (98%), its specificity was low (18%). Of the 157 patients without PE at CT pulmonary angiography, 128 (82%) had false-positive D-dimer assay results. The LRs suggest that the odds of a patient having a PE were increased by an LR of 1.2 (from 0.27 to 0.32) with knowledge of a positive D-dimer assay result. The odds of a patient not having a PE were decreased by the LR of 0.12 (from 0.27 to 0.03) with knowledge of a negative D-dimer assay result. Of the 30 patients with negative D-dimer assay results, one patient had positive CT pulmonary angiography findings. In this case, emboli were found only in small arteries. In the one case in which the angiogram was equivocal, the D-dimer assay result was positive.

Clinical Factors
Of the 201 patients, 118 (59%) presented with chest pain; 163 (82%), with dyspnea; and nine (5%), with hemoptysis. Several patients presented with more than one clinical symptom. Twenty-six (13%) patients had a history of DVT or prior PE. There was no substantial difference in the frequency of these symptoms or of historical factors between patients with and those without PE (Table 3). For example, the prevalence of chest pain was 59% in the total population, 61% in patients with PE, and 59% in patients without PE. At the time of the study, 147 (73%) patients were undergoing treatment for cancer or had been treated for cancer within the past 6 months, and nine (4%) patients were receiving palliative treatment. One hundred forty-one (70%) patients were receiving treatment for cancer at the time of our study. By using marginalized linear models, we found no significant difference (P > .05) in the NPV, PPV, sensitivity, or specificity of the D-dimer assay for PE in patients on the basis of whether they did or did not present with chest pain, dyspnea, hemoptysis, or a history of previous DVT or PE (Table 4). For example, the NPV, PPV, sensitivity, and specificity of the D-dimer assay were 96% (95% CI: 78%, 100%), 27% (95% CI: 18%, 37%), 96% (95% CI: 81%, 100%), and 24% (95% CI: 16%, 34%), respectively, for patients with chest pain, and 100% (95% CI: 59%, 100%), 23% (95% CI: 14%, 34%), 100% (95% CI: 81%, 100%), and 11% (95% CI: 4%, 21%), respectively, for patients without chest pain.

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

Our results show that if a negative D-dimer assay result was used as the sole criterion to exclude PE, only one case of PE in the 201 patients we evaluated would have been missed. These results approach those found for a nononcologic population of patients, where the NPV and sensitivity of the D-dimer assay results were reported to both be 100% (16). Owing to the low specificity of the assay, however, the majority of patients in our population would have had positive assay results and therefore would have been required to undergo further testing. Because the assay results have a low specificity and PPV, only 15% of patients in this study would not have required CT pulmonary angiography. Irani et al (17) similarly found that in high-risk elderly patients, D-dimer assay results may not be clinically effective for the exclusion of PE because of low specificity (17%).

A major factor that distinguishes the patient population in our study from that in others is that 99% of our patients had a diagnosis of cancer, a known risk factor for PE. The prevalence of PE in our study population, 21%, was similar to that found in another study (18) that had primarily oncologic patients. Overall PE and DVT prevalence rates of 10%–16% were reported in investigations (7,11,16) that studied nononcologic populations. The influence of malignancy on the results of the STA Liatest D-dimer assay is unknown. Findings of ten Wolde et al (19) suggest that the NPV of the SimpliRED D-dimer assay, an autologous red cell agglutination assay, for DVT is not significantly different between cancer and noncancer patients. Results of our study suggest that the STA Liatest D-dimer assay has a high NPV for the exclusion of PE, despite being evaluated in an oncologic population.

According to recommendations of the Prospective Investigation of Pulmonary Embolism Diagnosis II (PIOPED II) investigators (20), who used a quantitative rapid enzyme-linked immunosorbent assay for D-dimer, D-dimer assay results can be used in combination with clinical assessment to determine the need for further testing in patients suspected of having PE. In the recommendations of the PIOPED II investigators, the clinical probability for PE was based on clinical probability scoring indices (eg, the Wells Score, which stratifies patients into low, moderate, and high probability on the basis of factors such as clinical signs and symptoms of DVT; hemoptysis; heart rate; history of cancer, PE, or DVT; and whether an alternative diagnosis might be more likely than PE) (20–22). According to that study (20), the combination of a normal D-dimer assay result with objective clinical assessment yields a posttest probability for PE of 0.7%–2% in patients with a low clinical probability assessment and of up to 5% in patients with a moderate clinical probability assessment. Consequently, the PIOPED II recommendation is that no further testing is required if the D-dimer assay result is normal in a patient with a low or moderate clinical probability assessment. A normal or negative D-dimer assay result is not as useful for the exclusion of PE in patients with a high clinical probability assessment (20).

Given that 99% of the patients in our study had a history of cancer, the clinical probability of PE may be higher in our group than that in the general population. Therefore, D-dimer assay results may not be as useful for the exclusion of PE in such a high-risk population. However, our results demonstrate that PE can be excluded, with 97% certainty, by using negative immunoturbidimetric D-dimer assay results in an oncologic population.

Clinical history, symptoms, relevant vital signs, and oxygen saturation did not substantially raise or lower the NPV, PPV, sensitivity, or specificity of STA Liatest D-dimer assay results for PE. In this respect, the results here differ from those in some other studies (eg, PIOPED II), which suggest that the presence of certain clinical factors should influence the accuracy of the D-dimer assay results for PE. Burkill et al (10) found that combining the D-dimer assay result with pulse oximetry findings improved the NPV from 94% to 97%. Kline et al (23), who derived the PE rule-out criteria (PERC rule), assert that malignancy is not a substantial risk factor for PE, but that other factors, some of which we evaluated (ie, history of DVT or PE, elevated heart rate, low oxygen saturation, and hemoptysis), would increase the pretest probability of PE and likely would lower the NPV of D-dimer assay results. Given the oncologic population in our study, the clinical symptoms, relevant vital signs, and oxygen saturation levels that raised the suspicion for PE in the Kline et al study may have been attributable to other factors related to the tumor or chemotherapy treatment, thereby confounding the possible relationship between the presence of these factors and the likelihood of PE in our study.

In cases that were positive for PE, the sensitivity of the D-dimer assay for PE showed no substantial difference on the basis of the location of the embolus. It has been suggested (24) that D-dimer assay results are less sensitive for emboli in subsegmental pulmonary arteries. In the one patient in our study who had a false-negative D-dimer assay result, the pulmonary emboli were in small pulmonary arterial branches.

A limitation of our study was its relatively small sample size. Although the initial recruitment goal was 500 subjects, the study was terminated after a year of data collection when only 201 patients had been accrued. As a result, it is unclear whether the absence of a significant association between clinical symptoms or vital signs and the accuracy of D-dimer assay results resulted from a true lack of association or from a lack of power to detect a difference. Enrollment of patients in our study presented a challenge for several reasons. The acuteness of disease in these patients varied tremendously, as did the emotions of the patients and families. Some patients were simply too sick to give consent; others were so distraught over their condition that they did not want to hear about research or a study. Some Urgent Care Center physicians did not attempt to obtain consent from every potential patient because they thought it was inappropriate to do so in a clinically unstable or emotionally distressed patient. In addition, the Urgent Care Center was so busy at times that the staff thought there was insufficient time to obtain consent for the study. Undoubtedly these factors introduced selection bias into the study and limit our ability to generalize the results to the entire oncologic population. It is possible that, had a greater number of sicker patients been enrolled, the NPV of the assay results would not have been as high.

The results presented here show that, at least in oncologic patients, STA Liatest immunoturbidimetric D-dimer assay results have a high NPV and sensitivity for PE. This suggests that the assay may be used for the exclusion of PE, even in an oncologic patient population. The specificity and PPV were lower than those shown for the general population. Owing to the high number of false-positive results, only 15% of patients would be spared CT angiography. As pointed out by Burkill et al (10) and the PIOPED II investigators (20), the potential for combining clinical data (eg, patient history, signs, symptoms) with the D-dimer assay result to increase the NPV of the assay results or to decrease the posttest clinical probability of PE is promising. Owing to the small sample size of our study, further analyses of larger sample populations would be useful to resolve some of these unanswered questions.

	ADVANCES IN KNOWLEDGE

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 

• In an oncologic population, D-dimer assay results have a high sensitivity (98%) and negative predictive value (NPV) (97%) for pulmonary embolism (PE) but a low specificity (18%) and positive predictive value (PPV) (25%).
  
• Sensitivity of D-dimer assay results does not change substantially on the basis of the location of the pulmonary emboli.
  
• Clinical symptoms (eg, chest pain, dyspnea) and signs (eg, heart rate, oxygen saturation) do not significantly change the NPV, PPV, sensitivity, or specificity of the D-dimer assay results for PE (P > .05).
  

	IMPLICATIONS FOR PATIENT CARE

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 

• In an oncologic population, negative results on the STA Liatest immunoturbidimetric D-dimer assay can be used to exclude PE, but this assay has limited clinical utility owing to its low specificity and PPV.
  
• Clinical symptoms and signs cannot be reliably used in conjunction with D-dimer assay results to improve NPV, PPV, sensitivity, or specificity for the exclusion of PE.
  

	FOOTNOTES

 

Abbreviations: CI = confidence interval • DDU = D-dimer unit • DVT = deep venous thrombosis • LR = diagnostic likelihood ratio • NPV = negative predictive value • PE = pulmonary embolism • PIOPED II = Prospective Investigation of Pulmonary Embolism Diagnosis II • PPV = positive predictive value

Author contributions: Guarantors of integrity of entire study, V.K., M.S.G.; study concepts/study design or data acquisition or data analysis/interpretation, all authors; manuscript drafting or manuscript revision for important intellectual content, all authors; approval of final version of submitted manuscript, all authors; literature research, V.K., M.S.G.; clinical studies, V.K., A.A.V., L.J.S., M.S.G.; statistical analysis, C.S.M.; and manuscript editing, all authors

Authors stated no financial relationship to disclose.

	References

TOP ABSTRACT INTRODUCTION MATERIALS AND METHODS RESULTS DISCUSSION ADVANCES IN KNOWLEDGE IMPLICATIONS FOR PATIENT CARE References

 

1. Schoepf UJ, Costello P. CT angiography for diagnosis of pulmonary embolism: state of the art. Radiology 2004;230(2):329–337.[Abstract/Free Full Text]
2. 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(3):806–815.[Abstract/Free Full Text]
3. Fedullo PF, Tapson VF. The evaluation of suspected pulmonary embolism. N Engl J Med 2003;349(13):1247–1256.[Free Full Text]
4. Perrier A, Nendaz MR, Sarasin FP, Howarth N, Bounameaux H. Cost-effectiveness analysis of diagnostic strategies for suspected pulmonary embolism including helical computed tomography. Am J Respir Crit Care Med 2003;167(1):39–44.[Abstract/Free Full Text]
5. Perrier A, Bounameaux H. Cost-effective diagnosis of deep venous thrombosis and pulmonary embolism. Thromb Haemost 2001;86(1):475–487.[Medline]
6. Wiest P, Roth P. Introduction. In: Fundamentals of emergency radiology. Philadelphia, Pa: Saunders, 1996; 1.
7. Kearon C, Ginsberg JS, Douketis J, et al. An evaluation of D-dimer in the diagnosis of pulmonary embolism. Ann Intern Med 2006;144(11):812–821.[Abstract/Free Full Text]
8. Kruip MJ, Slob MJ, Schijen JH, van der Heul C, Buller HR. Use of a clinical decision rule in combination with D-dimer concentration in diagnostic work-up of patients with suspected pulmonary embolism. Arch Intern Med 2002;162(14):1631–1635.[Abstract/Free Full Text]
9. British Thoracic Society Standards of Care Committee Pulmonary Embolism Guideline Development Group. British Thoracic Society guidelines for management of suspected acute pulmonary embolism. Thorax 2003;58(6):470–483.[Free Full Text]
10. Burkill GJ, Bell JR, Chinn RJ, et al. The use of a D-dimer assay in patients undergoing CT pulmonary angiography for suspected pulmonary embolus. Clin Radiol 2002;57(1):41–46.[CrossRef][Medline]
11. Wells PS, Anderson DR, Rodger M, et al. Evaluation of D-dimer in the diagnosis of suspected deep vein thrombosis. N Engl J Med 2003;349(13):1227–1235.[Abstract/Free Full Text]
12. Farrell S, Hayes T, Shaw M. A negative SimpliRED D-dimer assay result does not exclude the diagnosis of deep vein thrombosis or pulmonary embolus in emergency department patients. Ann Emerg Med 2000;35(2):121–125.[CrossRef][Medline]
13. Quinn DA, Fogel RB, Smith CD, et al. D-Dimer in the diagnosis of pulmonary embolism. Am J Respir Crit Care Med 1999;159(5 pt 1):1445–1449.[Abstract/Free Full Text]
14. Casella G, Berger RL. Statistical inference. 2nd ed. New York, NY: Duxbury Advanced Series, 1990.
15. Pepe MS. The statistical evaluation of medical tests for classification and prediction. Oxford, England: Oxford University Press, 2003.
16. Abcarian PW, Sweet JD, Watabe JT, Yoon HC. Role of a quantitative D-dimer assay in determining the need for CT angiography of acute pulmonary embolism. AJR Am J Roentgenol 2004;182(6):1377–1381.[Abstract/Free Full Text]
17. Irani N, Nguyen AP, Lupetin AR. D-Dimer assay utility in the workup of pulmonary embolism in an elderly population (abstr). In: Radiological Society of North America scientific assembly and annual meeting program. Oak Brook, Ill: Radiological Society of North America, 2006; 199.
18. Ginsberg MS, King V, Panicek DM. Comparison of interpretations of CT angiograms in the evaluation of suspected pulmonary embolism by on-call radiology fellows and subsequently by radiology faculty. AJR Am J Roentgenol 2004;182(1):61–66.[Abstract/Free Full Text]
19. ten Wolde M, Kraaijenhagen RA, Prins MH, Büller HR. Clinical usefulness of D-dimer testing in cancer patients with suspected deep venous thrombosis. Arch Intern Med 2002;162(16):1880–1884.[Abstract/Free Full Text]
20. Stein PD, Woodard PK, Weg JG, et al. Diagnostic pathways in acute pulmonary embolism: recommendations of the PIOPED II investigators. Radiology 2007;242(1):15–21.[Free Full Text]
21. Stein PD, Fowler SE, Goodman LR, et al. Multidetector computed tomography for acute pulmonary embolism. N Engl J Med 2006;354(22):2317–2327.[Abstract/Free Full Text]
22. Wells PS, Anderson DR, Rodger M, et al. Derivation of a simple clinical model to categorize patients probability of pulmonary embolism: increasing the models utility with the SimpliRED D-dimer. Thromb Haemost 2000;83(3):416–420.[Medline]
23. Kline JA, Mitchell AM, Kabrhel C, Richman PB, Courtney DM. Clinical criteria to prevent unnecessary diagnostic testing in emergency department patients with suspected pulmonary embolism. J Thromb Haemost 2004;2(8):1247–1255.[CrossRef][Medline]
24. De Monye W, Sanson BJ, Mac Gillavry MR, et al. Embolus location affects the sensitivity of a rapid quantitative D-dimer assay in the diagnosis of pulmonary embolus. Am J Respir Crit Care Med 2002;165(3):345–348.[Abstract/Free Full Text]

This article has been cited by other articles:

				S. S. Adam, N. S. Key, and C. S. Greenberg D-dimer antigen: current concepts and future prospects Blood, March 26, 2009; 113(13): 2878 - 2887. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Submit a response Alert me when this article is cited Alert me when eLetters are posted Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by King, V. Articles by Ginsberg, M. S. Search for Related Content PubMed PubMed Citation Articles by King, V. Articles by Ginsberg, M. S.