| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1
Department of Radiology, Memorial Sloan-Kettering Cancer Center, 1275 York
Ave., New York, NY 10021.
2
Present address: Women's Diagnostic and Wellness Center, Nyack Hospital, 160
N. Midland Ave., Nyack, NY 10960-1998.
Received July 24, 2000;
accepted after revision October 5, 2000.
Address correspondence to C. S. Giess.
Abstract
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
MATERIALS AND METHODS. Reports from 433 bilateral and 619 unilateral lower extremity Doppler sonograms obtained over an 18-month period in patients with cancer were retrospectively reviewed, and clinical indication and findings were determined.
RESULTS. Overall, 228 (22%) of 1052 examinations revealed deep venous thrombosis (DVT): 83 (19%) of 433 bilateral and 145 (23%) of 619 unilateral. Among studies performed for unilateral symptoms (pain, edema, or postorthopedic procedure), 23% (135/581) of unilateral and 27% (44/162) of bilateral studies revealed DVT. Among these 44 bilateral studies with positive findings performed for unilateral symptoms, there were 30 DVT in the symptomatic side, 12 bilaterally, and two in the asymptomatic side alone. Ten percent (11/110) of the bilateral studies performed for bilateral symmetric symptoms revealed DVT. Among studies performed for bilateral asymmetric symptoms, 13% (1/8) of the unilateral and 8% (2/25) of the bilateral studies revealed DVT; both bilateral studies showed positive findings in the more symptomatic side. Among studies performed for suspected or proven pulmonary embolus, 20% (23/113) of bilateral and 54% (7/13) of unilateral studies had positive findings.
CONCLUSION. In a high-risk cancer population, the incidence of DVT in patients with unilateral symptoms is more than twice that of patients with bilateral symptoms. Because DVT isolated to an asymptomatic lower extremity is rare (1%), bilateral sonographic examination is generally unnecessary with unilateral lower extremity symptoms.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
There has been controversy in the literature [6,7,8,9,10,11] over the need to perform sonographic examination of both lower extremities in patients with unilateral lower extremity symptoms, as well as over the necessity for sonographic examination of both lower extremities in patients with bilateral symptoms. Sheiman et al. [6] found no DVT in patients with bilateral symptoms considered suggestive of DVT. Among another group of patients with unilateral symptoms who underwent bilateral examination, Sheiman and McArdle [7] found no DVT in the asymptomatic limb. Conversely, Naidich et al. [8] found DVT in 23% of their patients with bilateral symptoms undergoing bilateral examination. In addition, these researchers found that 9% of patients with unilateral symptoms had DVT in the contralateral leg, of which 1% of DVT was isolated to the asymptomatic side. On the basis of these studies, one author has concluded that the necessity for bilateral sonographic examination of patients with bilateral symptoms seems to depend on the presence of risk factors for DVT [9].
To our knowledge, the need for bilateral lower extremity venous sonography has not been specifically addressed in a population of patients with cancer who are known to be at high risk for the development of DVT. In our tertiary care cancer center, DVT is a frequent clinical concern; 20-30% of sonogram requests are for lower extremity duplex sonography to exclude clinically suspected thromboembolism. The purpose of this study was to determine the diagnostic yield of unilateral versus bilateral lower extremity venous sonography in our high-risk cancer population, as correlated with clinical indication for the examination.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
The diagnosis of cancer was obtained from radiology or pathology records. If more than one malignancy was reported, the first one listed was recorded. Distribution of types of malignancy in our population is detailed in Table 1.
|
Four hundred thirty-three examinations were bilateral and 619 examinations were unilateral. The decision to perform unilateral versus bilateral lower extremity sonography was based on clinician request. When a unilateral examination is performed and shows positive findings for DVT, we routinely examine the contralateral common femoral vein for the presence of DVT, to facilitate inferior vena cava filter placement. For the purpose of this study, these examinations were still considered unilateral because only a small portion of the contralateral femoropopliteal system was assessed.
Examinations were performed on an XP128 or Sequoia (Acuson, Mountain View, CA) or ATL 3000 (Advanced Technology Laboratories, Bothell, WA) unit using a 5-MHz linear array transducer; a 3.5-MHz transducer was used when necessary for patients with marked edema. Examinations were performed in a standardized fashion from the saphenofemoral junction to the popliteal bifurcation using compression in the transverse scanning plane, supplemented with color flow Doppler sonography in the longitudinal and transverse planes and spectral Doppler sonography of the common femoral and popliteal veins. Examinations were performed by a sonographer or radiology fellow. Findings were confirmed by an attending radiologist or radiology fellow.
The major diagnostic criteria for the presence of DVT were noncompressibility or incomplete compressibility of the vein or visualization of intraluminal thrombus; lack of phasic venous blood flow and lack of flow augmentation were used as secondary criteria. During this study period calf veins were not routinely examined.
Reports were retrospectively reviewed and coded for unilateral versus bilateral examination, clinical indication, and DVT findings. It is standard protocol at our institution to include clinical information in the radiology report, and this was used to obtain specific patient symptoms and clinical indication for the examination. Clinical indication was classified as unilateral symptoms (including recent orthopedic procedure), bilateral symptoms (specified as symmetric or asymmetric), suspected or proven pulmonary embolus, or other indication. When the specified clinical indication was bilateral lower extremity symptoms, these were assumed to be symmetric unless an asymmetric distribution was indicated. Patients who had undergone recent orthopedic surgery were classified into the unilateral symptom group, because they had undergone unilateral lower extremity or hemipelvic surgery with resultant postoperative edema and pain. During this study interval, all orthopedic patients at our institution underwent routine postoperative lower extremity venous sonography to exclude DVT. Study indications for bilateral and unilateral sonography are detailed in Table 2.
|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
In studies performed for unilateral symptoms considered suspicious for DVT, 135 (23%) of 581 unilateral studies and 44 (27%) of 162 bilateral studies revealed DVT. In the 44 bilateral studies with positive findings that were performed for unilateral symptoms, 30 patients revealed DVT in the symptomatic lower extremity, 12 revealed DVT bilaterally, and two revealed DVT in the asymptomatic lower extremity alone.
In studies performed for bilateral symmetric symptoms considered suspicious for DVT, 11 (10%) of 110 bilateral studies revealed DVT. In the 11 bilateral studies with positive findings, four patients had bilateral DVT and seven patients had unilateral DVT. In studies performed for bilateral asymmetric symptoms suspicious for DVT, one (13%) of eight unilateral studies and two (8%) of 25 bilateral studies showed DVT. Both bilateral studies with positive findings, which were performed for bilateral asymmetric symptoms, showed DVT in the more symptomatic lower extremity.
In studies performed to identify an embolic source due to suspected or proven pulmonary embolus, 30 (24%) of 126 revealed DVT, which included 23 (20%) of 113 bilateral studies and seven (54%) of 13 unilateral studies. In these 23 bilateral studies with positive findings, 21 DVT were unilateral and two were bilateral. In the seven unilateral studies that revealed DVT, two patients (29%) also had unsuspected DVT in the contralateral common femoral vein, which had undergone limited assessment to facilitate inferior vena cava filter placement.
In the small number of studies performed for other indications, three (13%) of 23 bilateral studies and two (12%) of 17 unilateral studies revealed DVT. Indications in these 40 studies included possible intraluminal filling defect detected on contrast-enhanced pelvic CT performed for other reasons (n = 20), recent postoperative status (n = 11), lower extremity fracture or tumor (n = 6), immobility (n = 2), and ascites (n = 1).
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
DVT was present in 10% of our studies performed for bilateral lower extremity symptoms. This is somewhat lower than the results of Naidich et al. [8], who found DVT in 35 (23%) of 149 patients with bilateral symptoms. As in our patient population, most patients in that study had predisposing risk factors for the development of DVT. The reasons for the lower incidence of DVT among our patients with bilateral symptoms compared with those in the study of Naidich et al. are uncertain. Some patients with cancer may have bilateral lower extremity edema caused by neoplastic lymphatic obstruction or as a side effect of treatment. In contradistinction, Sheiman et al. [6] found no DVT in their patients with bilateral symptoms, but only 16 of 50 patients in that study had predisposing risk factors.
In a study correlating lower extremity symptoms with the presence of DVT in patients with cancer, Loud and Klippenstein [12] reported a 17% incidence of DVT in patients with bilateral symptoms undergoing bilateral examination. They found that DVT was significantly more common when bilateral symptoms were asymmetric rather than symmetric. Although we did not find such a large discrepancy in the incidence of DVT in patients with asymmetric compared with symmetric lower extremity symptoms, in their series and ours, this subset of patients was relatively small. These studies, as well as our own, support the conclusion of Cronan [9] that the likelihood of finding DVT in patients with bilateral symptoms is related to predisposing risk factors, and in these patients bilateral lower extremity venous sonography is indicated.
DVT isolated to an asymptomatic lower extremity was quite rare in our study, present in only two (1%) of 162 bilateral examinations performed for unilateral symptoms. This is similar to the results of Sheiman and McArdle [7], who found no DVT isolated to an asymptomatic leg, and Naidich et al. [8], who found 3 (1%) of 245 patients with unilateral symptoms to have isolated contralateral DVT. These two studies reached different conclusions regarding the necessity of performing a bilateral examination when unilateral symptoms are present. Sheiman and McArdle concluded that bilateral examination was unnecessary and Naidich et al. concluded that a 1% yield was worth the additional time and expense of a bilateral examination.
At our institution, we have traditionally allowed the decision to perform a unilateral versus a bilateral examination to be determined by the requesting clinician, based on medical factors. Therefore, a secondary purpose of our study was to test the validity of this approach. We believe that our results support continuing this clinical practice. Bilateral lower extremity symptoms merit a bilateral examination, but unilateral examination is sufficient for unilateral symptoms, because the incidence of isolated contralateral DVT is 0-1%. We continue to routinely examine the contralateral common femoral vein in patients with positive findings on unilateral lower extremity venous studies to facilitate possible inferior vena cava filter placement. Documenting the extent of DVT in both lower extremities, even in patients with unilateral symptoms, may be a clinical concern in certain patients, and, therefore, in this setting we perform bilateral examination when requested by the referring clinician.
During our study period, calf veins were not routinely included in the sonographic assessment. We have subsequently added routine evaluation of calf veins to sonography of the lower extremities. At our institution, calf vein thrombosis is generally not treated with anticoagulation therapy, but patients undergo follow-up sonographic examination to exclude propagation of thrombus into the femoropopliteal system.
Overall, 126 (12%) of 1052 studies were performed for clinical suspicion of pulmonary embolus, and 30 (24%) of 126 revealed an embolic source. This incidence is higher than that reported by others [13,14,15]. Fowl et al. [13] found a 10.9% incidence of acute DVT in patients in whom there was clinical suspicion of pulmonary embolism. Matteson et al. [14] found a 13% incidence of DVT and Lipski et al. [15] found a 9% incidence of DVT in patients referred for venous sonography of suspected pulmonary embolus. Our higher diagnostic yield in the subgroup of studies performed for this clinical indication undoubtedly reflects the high-risk nature for the development of DVT in patients with cancer.
The general impact of venous sonography on the diagnosis of suspected pulmonary embolism has been evaluated by a number of authors [13,14,15,16,17,18]. In the setting of high clinical suspicion and lung scan (ventilation-perfusion scan) revealing an intermediate probability of pulmonary embolus, the diagnostic yield of lower extremity venous sonography has ranged from 10% to 55% in various studies [14,15,16,17,18]. In this situation venous sonography can help to guide clinical management, without subjecting the patient to pulmonary angiography or high-resolution chest CT. However, the usefulness of performing lower extremity venous sonography in patients with ventilation-perfusion scans revealing a high probability of pulmonary embolus and in whom there is a high clinical suspicion of pulmonary embolus has been questioned, because anticoagulation therapy generally does not alter the identification of an embolic source [15]. In fact, the study by Lipski et al. [15] found that treatment decisions for their patients were overwhelmingly (97%) based on the results of the ventilation-perfusion scans or subsequent pulmonary angiography rather than venous sonography. Matteson et al. [14] reported that 29% of venous sonograms could have been avoided in their population on the basis of ventilation-perfusion scans showing either normal findings or a high probability of pulmonary embolism. In the setting of ventilation-perfusion scans showing normal findings or a low probability of pulmonary embolism, Matteson et al. reported an 8% incidence of DVT and Lipski et al. reported a 0% incidence of DVT. Contrary to these results, Smith et al. [16] found thrombotic disease in the lower extremity in 21% of patients with normal findings on lung scans compared with 25% of patients with abnormal findings on lung scans.
We did not attempt to correlate the diagnostic yield in studies performed for suspected or proven pulmonary embolism with the results of ventilation-perfusion scintigraphy. However, our clinical physicians are often reluctant to institute anticoagulation unless there is definitive evidence of thromboembolism, because many of our patients are undergoing complex medical, surgical, and chemotherapeutic treatments. In addition, because our patient population is at recurring risk for development of DVT, determining the extent of acute thrombosis and its subsequent resolution is often believed to be of clinical importance. Both factors are likely to contribute to the frequency of requests for sonography for suspected or proven pulmonary embolism in our patient population.
We did not attempt to correlate specific lower extremity symptoms of edema or pain with the presence of DVT. Specific lower extremity symptoms were not always specified in the clinical history and, therefore, were not always known in our retrospective study. This was particularly true for studies in which the primary clinical indication was suspected pulmonary embolism. Other authors have correlated specific lower extremity symptoms of edema and pain with the presence of DVT. Fowl et al. [13] found the relative risk of DVT to be four times greater in patients with lower extremity edema than in patients without edema. In that study, patients presenting with lower extremity pain in whom associated edema was found had an incidence of DVT that was four times greater than that for patients who presented with pain alone. Similarly, Loud and Klippenstein [12] found that patients with cancer with unilateral lower extremity symptoms of edema, with or without associated pain, had a much higher incidence of DVT compared with patients with only lower extremity pain. Results of studies correlating specific symptoms with the presence of DVT may help to heighten clinical suspicion for DVT in the appropriate clinical setting.
In conclusion, in a high-risk cancer population, the incidence of DVT is more than twice as high in patients with unilateral symptoms as those with bilateral symptoms. However, patients with bilateral lower extremity symptoms have a significant (10%) incidence of unilateral or bilateral DVT, and, therefore, bilateral examination in these patients is warranted. DVT isolated to an asymptomatic lower extremity in patients with cancer is rare, and bilateral examination in the presence of unilateral symptoms is usually unnecessary. An embolic source is frequently (24%) identified in high-risk patients with suspected or proven pulmonary embolus.
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Reddit 
Technorati What's this?
This article has been cited by other articles:
|
|
 |
|
  C. S. Giess, H. Thaler, A. M. Bach, and L. E. Hann Clinical Experience With Upper Extremity Venous Sonography in a High-Risk Cancer Population J. Ultrasound Med., December 1, 2002; 21(12): 1365 - 1370. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
