Abstract
We evaluated the influence of the expertise of the technician on the accuracy of compression ultrasonography as a screening test to detect postoperative deep venous thrombosis in patients who had had primary or revision arthroplasty of the hip or the knee. The study was performed in two phases on two groups of patients who were evaluated with both compression ultrasonography and conventional venography. All of the patients received prophylaxis for deep venous thrombosis.In the first phase of the study, adequate venograms and ultrasonograms were made for 126 extremities (121 patients). Venography demonstrated seven thrombi in the proximal veins. Only three ultrasonograms were considered positive. Ultrasonography had a sensitivity of 0 per cent because none of the seven proximal thrombi identified with venography were detected with ultrasonography. Three ultrasonograms were false-positive (that is, the ultrasonogram was positive and the venogram was negative), so the specificity was 97 per cent. The accuracy of compression ultrasonography was 92 per cent.Both tests were performed on a second group of eighty-seven extremities (eighty-four patients). Proximal clots were detected with venography in five extremities. The ultrasonogram was positive for seven extremities, including the five in which a clot had been identified with venography. There were no false-negative and two false-positive results. In this phase of the study, ultrasonography had a sensitivity of 100 per cent, a specificity of 98 per cent, and an accuracy of 98 per cent.The only difference that was identified between the two groups of patients was the experience of the ultrasonography technician. It appears that the reliability of compression ultrasonography is directly dependent on the experience of the technician.
Patients who have had a total hip or knee arthroplasty have been shown to be at risk for thromboembolic complications, and a simple and accurate method of early diagnosis of deep venous thrombosis has been sought for decades. Ascending venography has been the standard with which other methods of evaluation have been compared1,2,4-6,10,14,16,17,24-27. However, problems related to ascending venography, such as discomfort, allergy to the contrast medium, and anaphylaxis, have prompted the search for non-invasive tests. Initially, plethysmography, radioactive fibrinogen uptake, and Doppler testing appeared to be useful for the evaluation of patients who were clinically suspected of having deep venous thrombosis. However, because of the lack of accuracy, these modalities have limited use for the screening of patients who do not have clinical symptoms. Ultrasonography has been reported to be highly accurate in the detection of thrombi both in patients for whom there is a clinical suspicion of deep venous thrombosis and in high-risk patients6.
The purpose of this study was to compare ultrasonography with ascending venography as a screening method for deep venous thrombosis in patients who have had a total joint arthroplasty of the hip or the knee and to determine factors that influence the accuracy of ultrasonography. We performed the study in two phases that differed only with regard to the experience of the technician. We then compared the results to determine if that experience affected the reliability of compression ultrasonography.
*One or more of the authors has received or will receive benefits for personal or professional use from a commercial party related directly or indirectly to the subject of this article. In addition, benefits have been or will be directed to a research fund, foundation, educational institution, or other non-profit organization with which one or more of the authors is associated. Funds were received in total or partial support of the research or clinical study presented in this article. The funding sources were Wyeth-Ayerst, Philadelphia, Pennsylvania, and Zimmer, Warsaw, Indiana.
†Department of Orthopaedic Surgery, Hospital of the University of Pennsylvania, 3400 Spruce Street, Philadelphia, Pennsylvania 19104.
‡3525 Prytania Street, New Orleans, Louisiana 70115.
This study was performed in two phases. Phase I included 125 patients (130 extremities) who had had a total hip or knee arthroplasty performed between September 1989 and February 1991 at the Hospital of the University of Pennsylvania. The operations included forty-one primary and eleven revision total hip replacements as well as sixty-six primary and twelve revision total knee replacements. Five patients had a second arthroplasty during the course of the study. The average age of the patients was sixty-eight years (range, twenty-six to eighty-eight years).
Because of our involvement in several research protocols for the prevention of deep venous thrombosis following total joint replacement, the patients were managed with a variety of prophylactic regimens during hospitalization. Twenty-one patients received fifty units of low-molecular-weight heparin (ardeparin sodium) per kilogram of body weight subcutaneously the evening after the operation and then ninety units per kilogram of body weight once a day, twenty-four patients received fifty units of low-molecular-weight heparin per kilogram of body weight subcutaneously twice a day, fifty-two patients received 650 milligrams of aspirin twice a day, and twenty-nine patients received low-dose Coumadin (warfarin). One patient, who had a second arthroplasty, was managed with two prophylactic regimens. All of the patients had ascending venography and duplex ultrasonography of the treated extremity, performed an average of six and a half days (five, six, or seven days) postoperatively. The ultrasonographic examination was performed on the fifth or sixth postoperative day, before the venogram. Both studies were completed within a twenty-four-hour period, and the results were evaluated by radiologists who were unaware of the prophylactic regimen.
The results of ultrasonography for four of the 130 extremities were excluded because the study was suboptimum (two extremities) or the proximal veins could not be visualized (two extremities). The venograms showed normal findings for these four extremities. Therefore, 126 extremities (121 patients) from phase I were included in the study.
Phase II included eighty-nine patients who had had ninety-two total hip or knee replacements between April 1992 and October 1992. The operations included twenty-nine primary and seven revision total hip replacements as well as forty-eight primary and eight revision total knee replacements. Three patients received fifty units of low-molecular-weight heparin per kilogram of body weight subcutaneously the evening after the operation and then ninety units per kilogram of body weight once a day, two patients received fifty units of low-molecular-weight heparin per kilogram of body weight subcutaneously twice a day, eighty-six patients received 650 milligrams of aspirin twice a day, and one patient received low-dose Coumadin. Three patients, who had a second arthroplasty, were managed with two prophylactic regimens.
The venograms were inadequate for two of the ninety-two extremities and were therefore excluded; the ultrasonograms for these two extremities were negative. Three ultrasonograms were considered inadequate and were excluded; the venograms for these extremities were interpreted as negative. Therefore, eighty-seven extremities (eighty-four patients) from phase II were included in the study.
The ultrasonography was performed with use of an ATL Mark 9 (Advanced Technology Laboratories, Bothell, Washington) or a Quantum 2000 (Quantum-Medical Systems, Issaquah, Washington) sonograph with color-enhancement capabilities. Both machines have a linear-array transducer with a five-megahertz imaging component and a five-megahertz Doppler device. The imaging was performed with the patient in a 20-degree reverse Trendelenberg position. The veins of the calf were evaluated with venography but not with ultrasonography. The results of Valsalva and Doppler testing were noted but were considered to be adjunctive to the compression test. During the compression test, the technician presses the transducer to the skin overlying the vein with enough force to cause the walls of the vessel to collapse, obliterating the lumen21. Inability to compress the vein completely is considered diagnostic of a thrombus. In addition, the technician observes the lumen of the vein while the patient performs a Valsalva maneuver; dilatation of the vein suggests the absence of a thrombus.
The scan was considered to be negative if the lumen of the proximal veins could be completely occluded with compression. An echogenic thrombus was noted when present, but the study was considered positive only when a vein could not be compressed completely. The final interpretation of the result of ultrasonography was made at the time of the examination, although a hard copy of the image was available for subsequent review.
Venography was performed with use of the modified method of Rabinov and Paulin. In this study, the portion of the popliteal vein proximal to the knee as well as the superficial and common femoral veins are referred to as the proximal veins. Real-time imaging of the proximal veins and simultaneous Doppler-flow examination were performed by a technologist who was supervised by a radiologist. The patient was placed in the reverse Trendelenberg position with a tourniquet about the ankle; fifty milliliters of contrast medium (Omnipaque) was injected into a vein on the dorsum of the foot. After fluoroscopic screening, several roentgenograms were made in multiple planes. An additional injection was performed proximal to the knee under tourniquet control if there was poor visualization of the proximal veins. A thrombus had to be visualized on multiple images to be considered present. The size of the thrombus was measured on the venogram.
The radiologist who interpreted the venogram and the ultrasonographer were unaware of each other's findings. The results of the two examinations were compared, with use of the venographic results as the standard. Although the veins distal to the knee were evaluated with standard venography, screening for clots in the distal veins with ultrasonography was not part of this study.
No clinical signs or symptoms of deep venous thrombosis or pulmonary embolism developed, and there were no postoperative deaths. There were no complications associated with either venography or ultrasonography.
Phase I
Of the 126 ultrasonograms (121 patients) included in the study, three were interpreted as being positive because the superficial femoral vein could not be compressed. The remaining 123 were negative and included those of the seven extremities in which a clot was detected with venography.
The venous system in all 126 limbs was adequately visualized on venography. Fifty-five extremities (44 per cent) had an isolated thrombus distal to the knee. A thrombus was present in the proximal veins of seven extremities (six patients); the operations associated with the seven clots included a unilateral total hip arthroplasty (four patients), a unilateral total knee arthroplasty (one patient), and a bilateral total knee arthroplasty (one patient). There was no association between the type of arthroplasty and either the size or the location of the proximal clot. Four of the proximal thrombi were large: two patients had a clot in the superficial femoral vein that measured two and three centimeters, and one patient had a fifteen-centimeter clot in the right superficial femoral vein and a three-centimeter clot in the left popliteal vein. The remaining three thrombi were small, non-occlusive clots that were less than one centimeter long.
Phase II
Eighty of the eighty-seven ultrasonograms were considered to be negative. Seven were positive and included those of the five extremities for which the venogram was positive.
Venography demonstrated an isolated clot in the calf in thirty-eight extremities (44 per cent) and a thrombus in the proximal veins in five extremities. The operations associated with these five extremities were a total hip arthroplasty for three, a total knee arthroplasty for one, and a revision total hip arthroplasty for one. As in phase I, there was no association between the type of arthroplasty and the size or the location of the proximal clot. One patient had complete occlusion of the popliteal and superficial femoral system with extension into the external iliac vein. Extensive collateral circulation suggested that the patient had chronic deep venous thrombosis, although an acute component could have been present. The length of the clot in the superficial femoral vein was fourteen centimeters in two patients and seven centimeters in two.
Comparison of Ultrasonography with Venography (Table I)
Phase I
The ultrasonogram was negative for all seven extremities in which a thrombus was visualized in the proximal veins on the venogram. The patient who had a large proximal thrombus bilaterally had a decreased response to the Doppler augmentation test in both extremities; however, the ultrasonogram was negative on the basis of a negative compression test. The results of ultrasonography were considered false-negative, and the sensitivity was 0 per cent. The false-negative ultrasonograms were distributed throughout the course of the study and were not confined to the initial period.
The venograms were negative for the three extremities that had a positive ultrasonogram; thus, with three false-positive ultrasonograms, the specificity of the test was 97 per cent. No proximal thrombi were detected with ultrasonography or venography in 116 extremities. The positive predictive value of ultrasonography was 0 per cent, the negative predictive value was 94 per cent, and the over-all accuracy for this phase was 92 per cent. The deceptively high accuracy is primarily due to the negative reading for 116 extremities.
Phase II
The results dramatically improved in phase II in that the ultrasonogram was positive for each of the five extremities for which the venogram demonstrated a proximal clot. In phase I, the sensitivity of the duplex Doppler testing was 0 per cent. A complete reversal was achieved in phase II, in which the sensitivity was 100 per cent.
Although three of the Doppler studies were excluded because of an indeterminate result, as mentioned earlier, none of the remaining eighty-seven examinations demonstrated false-negative findings. Therefore, the specificity was 98 per cent. This is comparable with the 97 per cent specificity attained in phase I. Two ultrasonograms of extremities for which the venogram was negative were inaccurately interpreted as positive. These two studies were considered false-positive, yielding an over-all accuracy of ultrasonography of 98 per cent in phase II, compared with 92 per cent in phase I.
Effect of Type of Arthroplasty and Anticoagulation Regimen
Neither the type of arthroplasty nor the anticoagulation regimen was related to the prevalence of clots in the proximal veins. Thrombi distal to the knee were present in 30 per cent (twenty-six) of eighty-eight extremities following total hip arthroplasty and in 71 per cent (ninety-five) of 134 extremities following total knee arthroplasty; however, screening for clots in the distal veins was not part of this study.
Several studies have shown duplex ultrasonography to be extremely accurate in the diagnosis of deep venous thrombosis, and the authors of those studies have recommended that ultrasonography replace ascending venography as the standard method for detection of deep venous thrombosis12,13,23. The reported sensitivity and accuracy of ultrasonography in patients who have clinical signs and symptoms of deep venous thrombosis has ranged from 88 to 100 per cent1,2,8,9,14,18,21,22. Aitken and Godden reported a sensitivity of 94 per cent and a specificity of 100 per cent for forty-six patients. Appelman et al. reported a sensitivity and specificity of 96 and 97 per cent, respectively, for 121 extremities. Other investigators, who used venography as the standard, have reported the sensitivity of ultrasonography to be between 63 and 100 per cent in patients who were thought to be at high risk for deep venous thrombosis4-6,10,12,25,26 (Table II). The finding of a positive ultrasonogram in association with an apparently negative venogram may be due to a failure of the venography to reveal the thrombus. Several investigators have shown, by repeating the venography, that the first venogram was in fact false-negative4,24,25. The rate of false-negative venograms has been reported to be very low4,27. We did not repeat the venography in our study.
The terms used to describe ultrasonography are sometimes confusing. B-mode ultrasonography is a method of processing ultrasonic information as a two-dimensional image19. Real-time ultrasonography generates dynamic images that are viewed instantaneously. The latest ultrasonography scanners are of the real-time B-mode type and can visualize the venous anatomy without the use of contrast medium. Large organized thrombi are usually echogenic and can be visualized easily; however, a fresh thrombus that develops within the first week after a total hip or knee arthroplasty may not be echogenic and may not be visualized directly on the real-time scan2,11,13,23. A compression test must be performed to detect fresh clots1,2,4-6,8,10,13,21,23,26.
Pulse-gated Doppler ultrasonography employs the Doppler effect to generate an audiological and graphic representation of blood flow and venous function. Manual compression of the calf increases the blood flow through the veins of the thigh. Partial venous occlusion is suspected when this maneuver does not augment blood flow. Doppler studies are reportedly less reliable than real-time scans with compression23. The combination of real-time ultrasonography and Doppler ultrasonography (termed duplex ultrasonography) provides information about the anatomy and function of the venous system and thereby has a potential for greater accuracy in the diagnosis of deep venous thrombosis23. Recently, color-flow enhancement has been added to the duplex scan and has been reported to increase the accuracy of the test15.
As with any diagnostic study, ultrasonography has its limitations and pitfalls. The three most common causes of false-negative ultrasonograms are a small clot, the location of the clot, and inexperience on the part of the technician1,5,18,25,27. Borris et al. reported that only two of eleven thrombi less than one centimeter in length were detected with the compression test in a study of sixty patients who had had a total hip arthroplasty. In phase I of our study, we failed to identify, with ultrasonography, three thrombi that were less than one centimeter long (Fig. 1).
The adductor hiatus of the thigh has been described as a blind spot for examination with ultrasonography because it is difficult to compress the vein at that location21,27 (Fig. 2). Comerota et al.6,7 recommended the use of adjunctive ultrasonography tests, such as Doppler flow studies, to examine the region of the adductor hiatus; however, other authors have not found such studies to be useful1. In the present study, one of the three false-positive ultrasonograms in phase I and both of the false-positive ultrasonograms in phase II could be attributed to this anatomical pitfall. The interpretation of an echogenic area as a thrombus is a known cause of false-positive ultrasonograms2,14; therefore, we did not use echogenicity as a criterion for the diagnosis of deep venous thrombosis in the present study.
A lack of experience on the part of the technician is believed to be the most common cause of false-negative ultrasonograms4,9,10,13,22,25,27. Acute clots are rarely visualized on real-time scans, which forces the technician to depend on the compression test. The examination may be technically difficult because acute clots are soft and are partially compressible, a finding that may deceive an inexperienced technician11,13. The accuracy of the test increases with experience. Woolson and Pottorff25 were able to increase the sensitivity of the ultrasonography test from 67 to 83 per cent, over a four-year period, for evaluation of the proximal veins in patients who had had a total knee arthroplasty. They believed that there was a time-dependent learning curve for ultrasonography, as they were able to detect the thrombi on a second ultrasonogram for two of the five extremities for which the ultrasonogram had initially been interpreted as negative. Our results were similar. In the present study, three large clots in the proximal veins were missed on the ultrasonographic examination (Fig. 3), with the compression test for each of these three extremities being believed to reveal normal findings. Venography, performed within twenty-four hours, revealed the thrombus on multiple views of each extremity. In the patient who had a large thrombus bilaterally, compression testing performed one week later clearly demonstrated the lesion. We believe that lack of experience on the part of the technician was the most likely cause of the false-negative finding for these patients.
Our impressive results for specificity and accuracy are deceptive. Because accuracy is a function of the percentage of correctly diagnosed extremities ([number of true-positive results + number of true-negative results]/total number of extremities), the inability to identify accurately a small number of extremities that had a clot in phase I (no true-positive results) is mathematically trivialized because of the correct screening of the extremities that did not have a clot (166 true-negative results). Since seven clots were missed and three clots were misdiagnosed, ten extremities were inaccurately screened with ultrasonography. With a total of 126 extremities in phase I, the accuracy of ultrasonography was (0 + 116)/126 = 0.92, or 92 per cent. This high accuracy initially leads to the assumption that ultrasonography is a reasonably useful screening test; however, it is misleading as all seven clots identified with venography were missed with ultrasonography.
Methods to improve the sensitivity of ultrasonography must be considered, especially if this modality is to be used for routine screening purposes. As with any new test or procedure, there is a learning curve when duplex ultrasonography is used in the diagnosis of deep venous thrombosis. The need for experience has become better appreciated by ultrasonographers, and the accreditation process requires the performance of 100 studies3. Our results suggest that, until the ultrasonographer obtains sufficient experience, venograms should be used concomitantly with ultrasonography when screening for deep venous thrombosis.
White et al.23 reported that a second ultrasonographic examination revealed proximal thrombi in six of sixty patients who had sustained a pelvic fracture. One possibility is to repeat ultrasonography postoperatively, but this would increase the over-all cost of hospitalization. Any analysis of a cost-containment strategy should evaluate not only the actual costs of the studies but also their over-all accuracy; false-negative and false-positive results may prove more costly than repeat examinations in the long run. The use of ultrasonography scanners with higher resolution may also improve the sensitivity of the test5.
Duplex ultrasonography is potentially valuable for screening for deep venous thrombosis in the thigh in patients who have had an orthopaedic procedure because it is non-invasive and is easily repeated. The compression test is the crucial component of the duplex examination, although the compressibility of a clot changes with its size, location, and age. The variation in sensitivity between institutions is probably due to dissimilar levels of technician experience and to differences in equipment. Therefore, at this time, we do not recommend that duplex ultrasonography replace venography as the standard for screening for deep venous thrombosis.
Initially, concomitant use of venography may provide an effective means by which Doppler ultrasonography can be used while an experience base is developed. In addition, since therapeutic anticoagulation is associated with a risk of bleeding, it may be prudent to make venograms for patients who have positive findings on ultrasonography before anticoagulation treatment is begun. The present study suggests that experience, supervision, and positive feedback can improve the value of this non-invasive study.
NOTE: The authors thank Dr. Peter Arger for his assistance with the interpretation of the ultrasonographic data and his input regarding this study.
Aitken, A. G., and |and |Godden, D. J.: Real-time ultrasound diagnosis of deep vein thrombosis: a comparison with venography. Clin. Radiol.,38: 309-313, 1987.38309
1987
[PubMed][CrossRef]
Appelman, P. T.; De Jong, T. E.; and |and |Lampmann, L. E.: Deep venous thrombosis of the leg: US findings. Radiology,163: 743-746, 1987.163743
1987
[PubMed]
Arger, P.: Personal communication, 1995.
Barnes, R. W.; Nix, M. L.; Barnes, C. L.; Lavender, R. C.; Golden, W. E.; Harmon, B. H.; Ferris, E. J.; and |and |Nelson, C. L.: Perioperative asymptomatic venous thrombosis: role of duplex scanning versus venography. J. Vasc. Surg.,9: 251-260, 1989.9251
1989
[PubMed][CrossRef]
Borris, L. C.; Christiansen, H. M.; Lassen, M. R.; Olsen, A. D.; and |and |Schott, P.: Comparison of real-time B-mode ultrasonography and bilateral ascending phlebography for detection of postoperative deep venous thrombosis following elective hip surgery. Thromb. and Haemost.,61: 363-365, 1989.61363
1989
Comerota, A. J.; Katz, M. L.; Greenwald, L. L.; Leefmans, E.; Czeredarczuk, M.; and |and |White, J. V.: Venous duplex imaging: should it replace hemodynamic tests for deep venous thrombosis?. J. Vasc. Surg.,11: 53-59, 1990.1153
1990
[PubMed][CrossRef]
Comerota, A. J.; Katz, M. L.; Grossi, R. J.; White, J. V.; Czeredarczuk, M.; Bowman, G.; DeSai, S.; and |and |Vujic, I.: The comparative value of noninvasive testing for diagnosis and surveillance of deep vein thrombosis. J. Vasc. Surg.,7: 40-49, 1988.740
1988
[PubMed][CrossRef]
Cronan, J. J.; Dorfman, G. S.; and |and |Grusmark, J.: Lower-extremity deep venous thrombosis: further experience with and refinements of US assessment. Radiology,168: 101-107, 1988.168101
1988
[PubMed]
Dauzat, M. M.; Laroche, J. P.; Charras, C.; Blin, B.; Domingo-Faye, M. M.; Saintes Luce, P.; Domergue, A.; Lopez, F. M.; and |and |Janbon, C.: Real-time B-mode ultrasonography for better specificity in the noninvasive diagnosis of deep venous thrombosis. J. Ultrasound Med.,5: 625-631, 1986.5625
1986
[PubMed]
Froehlich, J. A.; Dorfman, G. S.; Cronan, J. J.; Urbanek, P. J.; Herndon, J. H.; and |and |Aaron, R. K.: Compression ultrasonography for the detection of deep venous thrombosis in patients who have a fracture of the hip. A prospective study. J. Bone and Joint Surg.,71-A: 249-256, Feb. 1989.71-A249
1989
Gerlock, A. J., Jr.; Giyanani, V. L.; Krebs, C.: Applications of Noninvasive Vascular Techniques, pp. 224-246. Philadelphia, W. B. Saunders, 1988.
Grady-Benson, J. C.; Oishi, C. S.; Hanson, P. B.; Colwell, C. W., Jr.; Otis, S. M.; and |and |Walker, R. H.: Postoperative surveillance for deep venous thrombosis with duplex ultrasonography after total knee arthroplasty. J. Bone and Joint Surg.,76-A: 1649-1657, Nov. 1994.76-A1649
1994
Langsfeld, M.; Hershey, F. B.; Thorpe, L.; Auer, A. I.; Binnington, H. B.; Hurley, J. J.; and |and |Woods, J. J.: Duplex B-mode imaging for the diagnosis of deep venous thrombosis. Arch. Surg.,122: 587-591, 1987.122587
1987
[PubMed]
Lensing, A. W.; Prandoni, P.; Brandjes, D.; Huisman, P. M.; Vigo, M.; Tomasella, G.; Krekt, J.; Wouter Ten Cate, J.; Huisman, M. V.; and |and |Buller, H. R.: Detection of deep-vein thrombosis by real-time B-mode ultrasonography. New England J. Med.,320: 342-345, 1989.320342
1989
[CrossRef]
Leutz, D. W., and |and |Stauffer, E. S.: Color duplex Doppler ultrasound scanning for detection of deep venous thrombosis in total knee and hip arthroplasty patients. Incidence, location, and diagnostic accuracy compared with ascending venography. J. Arthroplasty,9: 543-548, 1994.9543
1994
[PubMed][CrossRef]
O'Leary, D. H., and Kane, R. A.: Venous ultrasonography of the lower extremities. In Clinical Applications of Doppler Ultrasound, pp. 338-353. Edited by K. J. W. Taylor, P. N. Burns, and P. N. T. Wells. New York, Raven Press, 1988.
O'Leary, D. H.; Kane, R. A.; and |and |Chase, B. M.: A prospective study of the efficacy of B-scan sonography in the detection of deep venous thrombosis in the lower extremities. J. Clin. Ultrasound,16: 1-8, 1988.161
1988
[PubMed][CrossRef]
Persson, A. V.; Jones, C.; Zide, R.; and |and |Jewell, E. R.: Use of the triplex scanner in diagnosis of deep venous thrombosis. Arch. Surg.,124: 593-596, 1989.124593
1989
[PubMed]
Price, R.; Kremkau, F.; Fleischer, A.; and James, A.: Introduction. In Diagnostic Sonography: Principles and Clinical Applications, pp. 1-28. Edited by A. Fleischer and A. James. Philadelphia, W. B. Saunders, 1989.
Rabinov, K., and |and |Paulin, S.: Roentgen diagnosis of venous thrombosis in the leg. Arch. Surg.,104: 134-144, 1972.104134
1972
[PubMed]
Raghavendra, B. N.; Horii, S. C.; Hilton, S.; Subramanyam, B. R.; Rosen, R. J.; and |and |Lam, S.: Deep venous thrombosis: detection by probe compression of veins. J. Ultrasound Med.,5: 89-95, 1986.589
1986
[PubMed]
Rollins, D. L.; Semrow, C. M.; Friedell, M. L.; Calligaro, K. D.; and |and |Buchbinder, D.: Progress in the diagnosis of deep venous thrombosis: the efficacy of real-time B-mode ultrasonic imaging. J. Vasc. Surg.,7: 638-641, 1988.7638
1988
[PubMed][CrossRef]
White, R. H.; McGahan, J. P.; Daschbach, M. M.; and |and |Hartling, R. P.: Diagnosis of deep-vein thrombosis using duplex ultrasound. Ann. Intern. Med.,111: 297-304, 1989.111297
1989
[PubMed]
White, R. H.; Goulet, J. A.; Bray, T. J.; Daschbach, M. M.; McGahan, J. P.; and |and |Hartling, R. P.: Deep-vein thrombosis after fracture of the pelvis: assessment with serial duplex-ultrasound screening. J. Bone and Joint Surg.,72-A: 495-500, April 1990.72-A495
1990
Woolson, S. T., and |and |Pottorff, G. T.: Venous ultrasonography in the detection of proximal vein thrombosis after total knee arthroplasty. Orthop. Trans.,16: 115, 1992.16115
1992
Woolson, S. T.; McCrory, D. W.; Walter, J. F.; Maloney, W. J.; Watt, J. M.; and |and |Cahill, P. D.: B-mode ultrasound scanning in the detection of proximal venous thrombosis after total hip replacement. J. Bone and Joint Surg.,72-A: 983-987, Aug. 1990.72-A983
1990
Wright, D. J.; Shepard, A. D.; McPharlin, M.; and |and |Ernst, C. B.: Pitfalls in lower extremity venous duplex scanning. J. Vasc. Surg.,11: 675-679, 1990.11675
1990
[PubMed][CrossRef]