| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
Ultrasonography |
1 From the Department of Radiology, Waikato Hospital, Hamilton, New Zealand (R.M.S., R.H., T.C., K.C., G.D., M.S.); and Departments of Radiology (R.M.S.)andMedicalEducation (R.M.S.), Waikato Clinical School, University of Auckland, Hamilton, New Zealand. Received July 25, 2004; revision requested September 29; revision received October 26; accepted November 11. Address correspondence to R.M.S., Department of Medical Imaging, Canberra Hospital and Australian National University, Garran, ACT 2605, Australia (e-mail: rathan.subramaniam{at}act.gov.au).
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
MATERIALS AND METHODS: Regional ethics committee approval and patient consent were obtained. A total of 542 consecutive ambulatory patients presented to the emergency department and were prospectively recruited from April 2001 to May 2003. Of these patients, 16 were excluded, and radiology residents and sonographers performed a complete lower limb US examination by means of compression and Doppler US in 526 patients. Patients with negative US findings received no anticoagulation therapy, and they were observed for occurrence of any thromboembolic event for 3 months. Patients with progressive or new symptoms that were indicative of thromboembolism within the follow-up period underwent objective testing with US, computed tomographic (CT) pulmonary angiography, or both.
RESULTS: There were 413 patients (78.5%) with US findings that were negative for DVT and 113 patients (21.5%) with findings that were positive. There were 64 patients (56.6%) with DVT isolated to the calf and 49 (43.4%) with proximal DVT. Of the 413 patients with negative initial US findings, 16 (3.9%) underwent a second US examination for new or progressive symptoms of DVT, one patient (0.25%) underwent CT pulmonary angiography for suspected pulmonary embolism, and one patient (0.25%) underwent both US and CT pulmonary angiography during the 3-month follow-up period. One of these patients (0.24%; 95% confidence interval: 0.01%, 1.3%) developed pulmonary embolism, which was diagnosed with CT pulmonary angiography. DVT was not diagnosed in any patient, and no patient died during follow-up. The negative predictive value of a complete single lower limb US examination to exclude clinically important DVT is 99.6% (95% confidence interval: 98.4%, 99.9%).
CONCLUSION: A single negative complete lower limb US examination is sufficient to exclude clinically important DVT, and it is safe to withhold anticoagulation therapy after negative complete lower limb US findings were obtained in patients suspected of having symptomatic lower limb DVT. New or progressive symptoms require further objective imaging.
© RSNA, 2005
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Ultrasonographic (US) examination limited to the common femoral vein, superficial femoral vein, and popliteal vein, with compressibility of the vein serving as the sole criterion for diagnosis, has been the reference standard for diagnosis of symptomatic lower limb DVT (12). This limited proximal US examination has nearly 100% sensitivity and specificity for proximal DVT, as measured with contrast venography; however, the accuracy of US in the detection of isolated calf DVT is lower (13). Since calf DVT can propagate proximally in one-third of cases (14), a second US examination is strongly recommended within 5–7 days to exclude DVT. Repeated US examination is time consuming, costly, and inconvenient to patients, and not all patients are able to return to undergo serial examinations (15–17). Furthermore, the yield from follow-up US examination is low (eg, less than 2%) (15,18,19).
It would be useful to determine the accuracy of negative findings of a complete lower limb US examination, on the basis of a prospective clinical outcome study, to justify withholding anticoagulation therapy. We performed a prospective cohort study of consecutive ambulatory patients who were seen in the emergency department and in whom symptomatic lower limb DVT was suspected to test the hypothesis that negative findings of a complete lower limb US examination safely exclude DVT and warrant no anticoagulation therapy.
|
MATERIALS AND METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Objective Testing for DVT
Complete real-time B-mode US examination with color Doppler flow analysis was performed in all patients by six 3rd- and 4th-year radiology residents and five sonographers with experience in day-to-day clinical practice. The results were interpreted by seven consultant radiologists (including G.D. and M.S.) who had 2–10 years of experience in the interpretation of vascular US studies as consultant radiologists, again reflecting day-to-day experience in clinical practice. High-spatial-resolution linear-array transducers with variable frequency (6–8 MHz) probes (Sequoia 512; Acuson, Mountain View, Calif) were used in all studies. Patients were lying in the supine position, with the symptomatic leg externally rotated and slightly flexed at the knee. They were examined from the level of the inguinal ligament to the medial malleolus. The common femoral vein, superficial femoral vein, popliteal vein and trifurcation, and all three deep calf vein sets were examined. Compressibility of these veins was assessed at 2–3-cm intervals in the transverse plane. Noncompressibility of a segment of the veins was the sole criterion for diagnosis of DVT. Doppler examination of these veins was performed for the purpose of acquiring supplemental information; these findings were used only as a road map, and they played no role in determining the result (whether DVT is present or absent) of the US examination. Anticoagulation therapy was withheld if the US findings were negative.
Follow-up
Patients were interviewed by a 4th-year resident and two research nurses (R.H., K.C.) after 3 months by telephone, and a questionnaire was completed for each patient. Patients were queried for symptoms of venous thromboembolism (eg, shortness of breath, chest pain, calf swelling and/or tenderness, and new onset ankle edema), hospitalization, surgical procedures, diagnostic testing, and any new medications. In instances where patients were unable to be contacted, their general practitioners were contacted. Clinical records, imaging records, and hospital admission records were systematically assessed by the senior resident (R.M.S.) and research nurses (K.C. and R.H.) to identify any evidence of an event of thromboembolism. The minimum follow-up period of 3 months was chosen on the basis of previous studies (16,20). There were three end points in this study: (a) the patient was alive and had experienced no thromboembolic event (eg, DVT or PE), (b) the patient was alive and had evidence of a thromboembolic event, and (c) the patient died within the follow-up period. If the patient had died, the cause of death was ascertained from the death certificate, which was determined by internal medicine and respiratory physicians who were blinded to the findings of this study.
Patients in whom DVT was suspected because of new or progressive symptoms were objectively assessed with a repeat complete lower limb US examination. Complete real-time B-mode US examination with color Doppler analysis was performed by using high-spatial-resolution, linear-array transducers with variable frequency (6–8-MHz) probes (Sequoia 512; Acuson). Patients were examined in the supine position from the level of the inguinal ligament to the medial malleolus on the symptomatic side of the body. The common femoral vein, superficial femoral vein, popliteal vein and trifurcation, and all three deep calf vein sets were examined. Compressibility of these veins was assessed at 2–3-cm intervals in the transverse plane.
In patients in whom PE was suspected, computed tomographic (CT) pulmonary angiography was performed by using a single-detector row helical CT scanner (HiSpeed CT/I; GE Medical Systems, Milwaukee, Wis). Contrast material–enhanced CT evaluation of pulmonary arteries was performed in a caudocranial direction from the top of the diaphragm to the level of the aortic arch. CT scans were acquired during a single breath hold or shallow breathing on the basis of each patient's ability to hold his or her breath during image acquisition. Scans were obtained with a collimation of 3 mm and a pitch of 1.6–1.8, and they were reconstructed at 1.5-mm intervals with a standard algorithm. Contrast material was injected in an antecubital vein with an 18–20-gauge cannula, and both arms were placed above the patient's head. A total volume of 150 mL of nonionic iohexol (Omnipaque 300; Nycomed, Auckland, New Zealand) was injected at a rate of 4.5 mL/sec. The examination began when contrast material was first seen in the pulmonary trunk by using Smart Prep (GE Medical Systems). Images were reviewed at a workstation with settings for pulmonary vasculature and lung parenchyma. The presence or absence of occlusive or nonocclusive thrombus in the main, lobar, segmental, and subsegmental arteries was recorded. The images were classified as positive for PE if thrombus was observed and negative for PE if no thrombus was observed. Images were interpreted by consultant radiologists with 4–5 years of experience in the interpretation of CT pulmonary angiograms. If the CT pulmonary angiogram was negative, then a complete lower limb US examination was performed, as described previously.
Statistical Analysis
We estimated that 380 (70%) patients in the original sample would have negative US findings and that the event rate would be 0.9% in the cohort with negative findings. A sample size of about 500 patients would be needed to yield a cohort of 350 patients with negative findings. With the hypothesized event rate, the upper limit of an exact two-sided 95% confidence interval (0%–2.29%) would exclude the published event rate of 2.6% (16). The exact two-sided 95% confidence interval was calculated with Excel, version 9.0 software (Microsoft, Redmond, Wash). We also estimated the negative predictive value of a complete compression US examination.
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
Positive for DVT
A total of 64 patients (56.6%) had DVT isolated to calf veins, and 49 patients (43.4%) had proximal DVT. All 113 patients in whom DVT was diagnosed were treated with anticoagulation therapy, in accordance with standard clinical practice. There were 59 men and 54 women, with an average age of 55.26 years ± 15.76 (age range, 18–88 years). A total of 26 patients had a history of previous DVT or PE, and 14 patients had a history of active malignancy.
Negative for DVT
Of the 413 patients in the negative cohort, there were 132 men and 281 women with an average age of 55.79 years ± 20.3 (age range, 21–86 years). A total of 36 patients had a previous history of DVT or PE, and 18 patients had a history of active malignancy. All 413 patients in the negative cohort completed 3-month follow-up. Of the 413 patients with negative findings at the initial US examination, 16 (3.9%; 95% confidence interval: 2.2%, 6.2%) underwent another US examination for new or progressive symptoms of DVT, one (0.25%) underwent CT pulmonary angiography for suspected PE, and one (0.25%) underwent both US and CT pulmonary angiography during the follow-up period. One of these patients had a PE that was diagnosed with CT pulmonary angiography. DVT was not diagnosed in any of the patients (Figure). The patient in whom PE was diagnosed during follow-up underwent the initial US examination 4 days earlier; superficial thrombophlebitis involving the greater saphenous vein to the midthigh level was diagnosed in this patient, and the sapheno-femoral junction was free of thrombus. This patient did not undergo anticoagulation therapy initially. The overall rate of a thromboembolic event in the negative cohort is 0.2% (95% confidence interval: 0.01%, 1.3%) within 3 months of initial complete lower limb US examination. None of the patients in the negative cohort died during follow-up. The negative predictive value of a complete single lower limb US examination to exclude clinically important DVT is 99.6% (95% confidence interval: 98.4%, 99.9%).
|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
If we use above-knee compression US, serial examination is necessary to detect propagation of calf vein thrombi proximally. Fatal complications have occurred before the second examination was performed when this strategy was applied (16,18). The failure rate with this strategy is 0.7% (95% confidence interval: 0.3%, 1.2%) (17). Because the prevalence of proximal DVT in ambulatory outpatients varies from 9% in the current study to 28% in a study performed by Bernardi et al (22); the rate of repeat US in up to 90% of patients is of low clinical efficiency, especially as the yield from these examinations is less than 2%. Though a complete lower limb US examination would take 10–15 minutes longer than a limited above-knee US examination, the second above-knee US study required would consume substantial resources, present an inconvenience to patients, and not always be practical (23).
Our study showed that it is safe to withhold anticoagulation therapy after a single negative complete lower limb US study, with only one patient (0.2%) experiencing a nonfatal PE 4 days after the initial US examination. The upper limit of a 95% confidence interval for a thromboembolic event within the 3-month follow-up period is 1.34%, which is favorable when compared with (a) the findings of normal above-knee serial compression US examinations (16,17,24) and combined above-knee single-compression US examinations, (b) normal results of a whole-blood D-dimer blood test (15,18), and (c) studies of combining pretest probability and serial US (two-point compression) with venography to exclude DVT (21). In these studies, the risk of a thromboembolic event in patients who did not undergo anticoagulant therapy after a negative test result is less than 3.3% (15–18,24). The risk of a thromboembolic event occurring within the 3-month follow-up period after a complete lower limb US examination varied between 0.3% and 0.8% in other studies (25–27), and the rate of a thromboembolic event occurring in our study compares favorably.
The proportion of isolated calf DVT is 56% (64 of 113 patients) in our study, which is greater than the rate observed in other studies (31%–45%), in which a complete single lower limb US examination (25–27) and a venography study (28) were performed. This may be because the general practitioners referred patients to the emergency department at an earlier stage of the disease spectrum than did physicians in other studies. Though isolated calf thrombi present a therapeutic dilemma (ie, whether to start anticoagulation therapy or not), the sixth American College of Chest Physicians consensus conference on antithrombotic therapy (29) recommends symptomatic isolated calf DVT be treated with anticoagulant medications. The 64 patients in whom isolated calf DVT was diagnosed represent 12% of the study population, and these patients underwent anticoagulation therapy. The sensitivity (30) and specificity of technically adequate duplex examination of calf veins is high (31–34), which makes risk of overdiagnosis less likely, and patients are not exposed to unnecessary anticoagulation therapy and complications.
Our results are applicable to adult ambulatory outpatients who are older than 18 years of age and in whom a technically adequate complete lower limb US examination could be performed. Patient body habitus, especially obesity, results in an inability to obtain technically adequate US examinations of calf veins (16,28), but only a few patients were excluded because of an inability to obtain adequate images of calf veins, both in other studies (25–27) and in our study. Our study was conducted at one institution, with experienced radiology residents and sonographers performing US examinations; thus, our findings are applicable in similar institutions. This strategy needs to be applied and validated prospectively in hospital inpatients, in whom this strategy would be useful. The D-dimer test results are rarely negative, with a 500-ng/mL (enzyme-linked immunosorbent assay D-dimer) conventional cutoff value (35) in inpatients, and it does not contribute to the diagnostic strategy of combining a negative D-dimer result with results of an above-knee compression US examination to rule out DVT.
Our study may be limited because we examined only the symptomatic leg rather than both legs with US, which reflects the local clinical practice. It has been debated whether bilateral US imaging should be performed in patients in whom unilateral DVT is suspected (36). There are very few clinical data to support such a practice (37). Although we may have missed small silent thrombus in the asymptomatic leg, this is considered less likely because our study population consisted of ambulatory outpatients who presented to the emergency department.
In summary, it is safe to exclude clinically symptomatic lower limb DVT with a single complete lower limb US examination, and anticoagulation therapy can be withheld, with a low failure rate. This diagnostic strategy has a favorable rate of a thromboembolic event occurring within the 3-month follow-up period when compared with the other diagnostic strategies currently used. Furthermore, this strategy would save time and be more convenient for imaging departments if it were used in ambulatory outpatients with suspected symptomatic lower limb DVT.
|
FOOTNOTES |
|---|
Abbreviations: DVT = deep venous thrombosis • PE = pulmonary embolism
Authors stated no financial relationship to disclose.
Author contributions: Guarantor of integrity of entire study, R.M.S.; 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, R.M.S., T.C., K.C.; clinical studies, R.M.S., K.C., M.S.; statistical analysis, R.M.S.; and manuscript editing, R.M.S., R.H., K.C., G.D., M.S.
|
References |
|---|
|
TOPABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONReferences |
|---|
This article has been cited by other articles:
|
|
 |
|
  W. Li, J. Salanitri, S. Tutton, E. E. Dunkle, J. R. Schneider, J. A. Caprini, L. N. Pierchala, P. M. Jacobs, and R. R. Edelman Lower Extremity Deep Venous Thrombosis: Evaluation with Ferumoxytol-enhanced MR Imaging and Dual-Contrast Mechanism--Preliminary Experience Radiology, March 1, 2007; 242(3): 873 - 881. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. M. Subramaniam, T. Chou, R. Heath, and R. Allen Importance of Pretest Probability Score and D-Dimer Assay Before Sonography for Lower Limb Deep Venous Thrombosis Am. J. Roentgenol., January 1, 2006; 186(1): 206 - 212. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Complete Leg Ultrasound Sufficiently Rules Out DVT Journal Watch (General), October 11, 2005; 2005(1011): 3 - 3. [Full Text]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| RADIOLOGY | RADIOGRAPHICS | RSNA JOURNALS ONLINE |
