Lower limb contrast venography - BIR Publications

The British Journal of Radiology, 80 (2007), 859–865

REVIEW ARTICLE

Lower limb contrast venography: a modified technique for use in thromboprophylaxis clinical trials for the accurate evaluation of deep vein thrombosis 1

P S SIDHU,

BSc, MRCP, FRCR,

2

R ALIKHAN,

MD, MRCP,

1

T AMMAR,

MBBS, MRCP

and 1D J QUINLAN,

MBBS

1

Department of Radiology, King’s College Hospital and 2Department of Haematology, John Radcliffe Hospital, Oxford, UK

ABSTRACT. Lower limb venography remains the imaging modality of choice for detection of asymptomatic deep vein thrombosis (DVT) in clinical trials of anticoagulant agents. A variety of techniques of venography have been described. Here, we describe a modified technique (the ‘‘King’s’’ technique) developed to increase the overall adequacy of identification of lower limb veins and detection of small asymptomatic DVT. Essential elements include proper preparation of patients prior to their arrival in the radiology department, intermittent use of tourniquets to ensure complete and adequate deep vein filling, use of a consistent image acquisition sequence and visualization of all veins in at least two different planes. Use of this technique minimizes technical difficulties, provides improved patient through-put in ‘‘busy’’ fluoroscopy units and, ultimately, improves ‘‘off-site’’ levels of adjudication. Contrast venography has traditionally been the accepted reference imaging examination for the diagnosis of deep vein thrombosis (DVT) [1, 2]. Although now commonly replaced by ultrasound for the initial diagnosis of symptomatic DVT, venography remains widely used in clinical trials investigating anticoagulant therapies [3]. Owing to its high sensitivity for the detection of small asymptomatic lower limb DVT, venography enables clinical trials to be conducted with smaller sample sizes than those evaluating clinical outcomes [3–7]. With the development of B-mode and colour Doppler compression ultrasound as a non-invasive imaging modality for symptomatic DVT, the technique of venography is now practised less often and, more importantly, taught to fewer radiology trainees [2]. Doppler ultrasound has limitations for the detection of asymptomatic DVT, particularly calf vein thrombi [8, 9]. These asymptomatic (silent) DVTs are often non-occlusive, small (1–2 cm) and are usually not associated with venous distension or impaired blood flow, factors important for the ultrasonographic diagnosis of symptomatic DVT [10, 11]. Anatomical variations in venous anatomy are well documented and are often not apparent on an ultrasound examination [12]. In addition, Doppler ultrasound is heavily examiner dependent, and does not lend itself to ‘‘off-site’’ blinded adjudication of thrombotic events. Although ultrasound has both a sensitivity and specificity approaching 95% for the detection of symptomatic proximal DVT, this sensitivity Address correspondence to: P S Sidhu, Department of Radiology, King’s College Hospital, London, UK. E-mail: paul.sidhu@kingsch. nhs.uk

The British Journal of Radiology, November 2007

Received 2 May 2006 Revised 3 December 2006 Accepted 7 January 2007 DOI: 10.1259/bjr/15041517 ’ 2007 The British Institute of Radiology

falls to 26–40% for asymptomatic DVT [6, 9, 13–15]. Conversely, contrast venography maps out and records venous anatomy in a form readily interpretable off-site and has a far higher rate of detection of asymptomatic distal DVT. Although various techniques of venography have been described, the most widely used remains the technique described by Rabinov and Paulin over 30 years ago [16], representing a modification of the technique described by Greitz in 1954 [17]. A further technique described by Thomas differs in its use of tourniquets, smaller contrast volumes and the degree of semi-upright tilt on the fluoroscopy table [18, 19]. Having participated in a large number of clinical trials involving the use of venography for detection of asymptomatic DVT, we developed a modified technique in an attempt to improve overall standards, minimize technical difficulties, improve patient through-put in ‘‘busy’’ fluoroscopy units and, ultimately, improve ‘‘offsite’’ levels of adjudication.

The ‘‘King’s’’ technique Using electronic databases (MEDLINE and EMBASE, 1980–1996) and clinical trial venography protocols, a review of various venography techniques was undertaken. A protocol for a modified venography technique, described here as the ‘‘King’s’’ technique, was developed and then implemented (from 1997 onwards) in our department. Since then, the technique has been used in clinical trials involving patients undergoing joint replacement (hip/knee) [20–23], surgery for hip fracture [24, 25], surgery for cancer [26], abdominal surgery [27] and 859

P S Sidhu, R Alikhan, T Ammar and D J Quinlan

non-surgical patients who were acutely immobilized [28, 29]. All trials had received ethics committee approval for bilateral venography and patients gave written informed consent for the procedure. The modified technique is described below.

Preparation of the patient Prior to venography, patients are reviewed either on the ward or in the out-patient department. Adequate preparation of patients prior to arrival in the radiology department allows greater time for the identification of suitable venous access in an environment that is more comfortable for the patient. Swollen oedematous feet are wrapped in firm compression bandages and then elevated for at least one hour before venous cannulation is attempted prior to arrival in the fluoroscopy unit. This assists in identification of large easily accessible veins, thereby reducing the possibility of failed venous access. The majority of patients undergoing venography are elderly with thin skin, but in those with thicker skin the use of EmlaH cream (AstraZeneca, Luton, UK) may be beneficial to reduce patient discomfort upon venous cannulation. Before cannulation, the feet are bathed in warm water for 5 min, allowing the veins to become dilated and more prominent. The warm water has the additional effect of softening the skin and may also provide a mild analgesic effect. For venous cannulation, the patient then sits on the edge of the bed or chair with the legs dependent. Venepuncture is performed via the dorsal vein of the great toe using a size 22-gauge cannula (VenflonH; Ohmeda, Helsingborg, Sweden). The cannula is secured with adhesive tape. If a patient has undergone an orthopaedic procedure, then cannulation of the operated leg is attempted first, as 80% of thrombi usually occur in this limb [30]. The dorsal vein of the great toe has the advantage of being easily located, even in an oedematous foot, by gently squeezing the dorsum of the toe with both thumbs. Compared with a ‘‘butterfly needle’’, the advantage of using a cannula is that extravasation of contrast into the subcutaneous tissues is rare and can be detected promptly because of the thin layer of skin and adipose tissue in the area of cannulation. If it is not possible to cannulate the dorsal vein, then an attempt at cannulation is made at the most distal and medial vein on the dorsum of the foot. Personal experience has shown that a more proximally and laterally located venous cannulation results in inadequate filling and opacification of the deep venous system; with contrast more likely to pass into the dorsal venous arch, which then drains into the long and short saphenous veins. If distal veins are not available for cannulation, more proximally located veins may be utilized, although the venous cannula is now sited in an antegrade direction, i.e. pointing towards the toes (Figure 1). Contrast medium is then injected in a downstream direction, where it communicates with the deep plantar arch of the foot before flowing in a retrograde direction into the posterior tibial veins [18, 31]. In patients with very small foot veins, a 23-gauge ‘‘butterfly needle’’ or a 25-gauge cannula may need to be used; however, warm contrast is required for prompt and adequate injection volumes. 860

Figure 1. Antegrade cannulation in the dorsum of the left foot and dorsum of the right great toe. A Foley catheter is shown in situ above the left malleolus with a tourniquet above the left knee.

Contrast medium A non-ionic iodinated contrast medium is routinely used in venography. Our practice is to use OmnipaqueH (Iohexol; Nycomed Amersham Imaging AS, Norway) with an iodine content of 350 mg ml21. Two 50 ml syringes containing 35 ml of OmnipaqueH diluted with 15 ml of 0.9% saline, for each limb under investigation, are prepared. A large volume (100 ml) of dilute contrast results in better deep venous filling and improved image quality [16, 18, 31].

Radiographic equipment A fluoroscopy unit with a real-time television monitor and tilting table is required for adequate lower limb venography. Although other venography techniques without fluoroscopy, e.g. the long leg technique, have been described [8, 32, 33], visualization of specific veins is not guaranteed, as it is not possible to accurately predict contrast flow along the deep veins in order to calculate image acquisition. Furthermore, it is often impossible to distinguish between artefacts resulting from flow or underfilling, which may be confused with thrombosis or post-thrombotic syndrome [19]. Our venography studies are performed on a digital fluoroscopic imaging system (Siemens Medical Engineering, Erlangen, Germany). Tube voltages of 65–70 kV are used for the lower leg, 80–90 kV for the thigh, and 100 kV for the pelvic region. At the end of the procedure, all images are displayed on a monitor, which allows for immediate assessment of venography quality and diagnosis of venous thrombus. Digital imaging has the advantage of post-procedure manipulation, such as windowing, edge enhancement and correction using the video-enhancer [34].

Positioning of patient Ideally, the procedure should be carried out with the patient in the erect position to allow maximum mixing of contrast and blood, with subsequent filling of the deep The British Journal of Radiology, November 2007

Review article: Lower limb contrast venography

venous system. To attain the fully erect position, specialized equipment is required and many patients, especially the elderly and those who have had recent surgery, are unable to tolerate or sustain this position for any length of time. A compromise is to allow a 30–40 ˚ tilt on the fluoroscopy table. The patient’s head and body are kept flat against the table in order to prevent a hypotensive or vasovagal circulatory reaction. Support is provided by handgrips on both sides. The limb not being examined is allowed to weight bear on a specially constructed box (measuring 20610610 cm), allowing for free movement and rotation of the contralateral limb under investigation (Figure 2). It is important that the limb under investigation is kept relaxed without excessive strain or calf muscle contraction during the procedure, which might result in contrast being expelled from the region of interest. To prevent superficial vein filling and to aid deep vein filling, two tourniquets are used. The first tourniquet is placed tightly above the ankle malleoli, whereas the second is placed 5–10 cm above the patella, where it is useful in delaying the emptying of contrast from the calf veins, thereby allowing improved deep venous filling. The best tourniquet for this purpose is a rubber tube, such as a size 16 Foley catheter (Bard Limited, Crawley, UK), which allows a greater degree of force per unit area; this is held in place with a pair of artery forceps.

Sequence of image acquisition The examination is most efficient when performed by a radiologist and an assistant; the assistant injects the contrast and rotates the limb, leaving the radiologist free to check the position of the limb and obtain the relevant images at the optimum time. The first syringe containing contrast is connected to the cannula via a 50 cm infusion line (Connecta Plus 3; Becton Dickinson Infusion Therapy AB, Sweden) and is injected by hand as rapidly as possible (approximately 50 ml at 2 ml s21) to produce a dense bolus. During injection, the left hand of the assistant is placed over the injected vein to reduce mechanical over-distension and also to palpate for contrast extravasation.

The radiologist obtains a screening image at the level of the knee joint and allows time for the distal deep calf veins to fill. The posterior tibial veins, which arise from the deep plantar foot vein, are usually the first to fill, followed by the peroneal veins arising from the lateral region of the foot. The anterior tibial veins originating from the dorsal vein of the foot are commonly the last and most difficult veins to fill. If the anterior tibial vein does not fill with contrast following administration of the first 50 ml syringe, the table is then tilted to 50–60 ˚, the ankle tourniquet is removed and sufficient time allowed for these veins to fill. If filling still does not occur, a further 25 ml of contrast is administered, with additional time allowed for filling. 12 images are routinely obtained. When the three pairs of calf veins have filled, and with the leg in the neutral position, the first image obtained is of the upper calf, including the knee joint (Figure 3a). The assistant then rotates the leg, and images are obtained with the leg in internal (Figure 3b) and then external rotation (Figure 3c). It is important to take these three images, as the deep veins often overlie one another and can make separation and interpretation difficult. A screening image is then obtained above the ankle, making sure that there is overlap with the previous images. Similar to the procedure above, three images are obtained: in neutral, internal rotation and external rotation. Removal of the ankle tourniquet at this stage allows further filling of the more distal veins without vein compression caused by the tourniquet. The radiologist obtains two images at the level of the popliteal fossa, where the junction of the anterior and posterior tibial veins forms the popliteal vein. The first image is taken in the neutral position (Figure 4a) and the second in external rotation. In those patients with knee prostheses, it may be necessary to take a third image with the patient rolled onto his or her side to fully visualize the popliteal vein (Figure 4b). If the image intensifier has a rotating C-arm, this can be used to achieve proper visualization of the popliteal vein. The leg is then kept in external rotation and the aboveknee tourniquet is removed as the second syringe of

(a)

(b)

(c)

Figure 2. Fluoroscopy table tilted to 30 ˚ with contrast

Figure 3. Venography images of the upper calf in three

injected by the assistant who also compresses the calf prior to the tilting the table into the horizontal position.

separate views: (a) neutral position; (b) internal rotation; and (c) external rotation.


861


(a)

(b)

Figure 4. Venography images of the popliteal fossa of a patient with a total knee replacement: (a) neutral position; and (b) external rotation.

contrast is slowly administered (approximately 35 ml at 1 ml s21). The superficial femoral vein is the direct continuation of the popliteal vein and passes upwards and medially across the lower third of the femur through the adductor canal towards the groin. Below the inguinal ligament, it receives the profunda femoris (deep femoral) vein, which is often visualized in its proximal segment only when there is retrograde contrast flow. The common femoral vein is formed by the confluence of the superficial femoral and profunda femoris veins and continues above the inguinal ligament as the external iliac vein. As with the knee, a hip prosthesis (especially when surgical cement is used) may obscure a full view of the vein, and therefore an image of the common femoral vein is obtained by either rolling the patient onto his or her side or by the use of a rotating C-arm. Tilting the table towards the horizontal position at this stage enables a good bolus of contrast to pass through the common femoral vein. For the final image, the image intensifier is placed over the abdomen. A fast bolus of the remaining contrast is injected (approximately 15 ml at 3 ml s21). As the table is lowered back to the horizontal position, the assistant gently squeezes the calf with the palms of both hands, enhancing the filling of the external iliac vein, common iliac vein and inferior vena cava. This can be followed by passive elevation of the legs, which assists in visualization of the iliac veins and inferior vena cava (Figure 5). Sufficient attention to detail is required in visualization of the common femoral and iliac veins, as both are subject to flow defects that produce false visualization of a thrombus. If poor deep venous filling occurs in the proximal veins, the valsalva manoeuvre may be attempted by asking the patient to take a deep breath in and to blow 862

Figure 5. Venography images of the common iliac vein and inferior vena cava. The image intensifier is placed over the abdomen and the remaining 15 ml of contrast injected as a fast bolus, while the calf is gently squeezed and the lower leg elevated.

out with the mouth closed and nose pinched. The effect of blowing out against a closed glottis causes a rise in intra-thoracic pressure that is transmitted to the inferior vena cava, producing a reflux of blood down towards the limb veins and an increase in opacification with contrast. If this proves to be difficult, a similar effect can be obtained if the patient forcefully coughs. To decrease any potential for post-venography thrombosis, the venography study is performed in a swift manner in order to decrease the amount of time contrast remains in contact with the vein wall. Once all of the images have been obtained, the infusion line is flushed with 25 ml of 0.9% saline and the limb is elevated to help clear the veins of contrast material. A screening image of each limb is obtained at the end of the examination to ensure dispersal of contrast from the deep veins. In the clinical diagnosis of deep vein thrombosis, it may be possible to exclude some of the images to produce a diagnostic venography study of the lower limb. In order to confirm the presence of a thrombus, an intra-luminal filling defect or vein occlusion must be visible in at least two images obtained in different planes. Therefore, in the context of a clinical trial, at least 12 images are taken to be confident of providing a complete examination suitable for evaluation by an independent ‘‘off-site’’ adjudication committee.

Discussion Here, we describe a modified venography technique that may potentially improve the adequacy of visualizing veins within the deep venous system of the lower limbs. Although this technique is most suited to use in clinical trials where bilateral venography is performed to detect The British Journal of Radiology, November 2007

Review article: Lower limb contrast venography

asymptomatic DVT, adoption of the technique can have advantages whenever venography must be performed, especially in difficult patients in whom ultrasound provides an inconclusive result. In clinical trials of anticoagulant agents, the overall adequacy of venography is reported to range between 70% and 90% [4, 5, 35, 36], despite apparent strict adherence to established venography protocols. In order to assess the efficacy of an anticoagulant, it is of paramount importance to visualize the entire deep venous system so that it is possible to identify or exclude the presence of thrombus. An inadequate venography examination invalidates the data for that individual patient and, by implication, has subjected the individual patient to unnecessary investigation and treatment. Furthermore, as venography is included in the primary efficacy outcome of many clinical trials, this results in up to 30% of patients being non-evaluable for the primary efficacy outcome [36]. Indeed, clinical trial protocols incorporate this venogram failure rate in their sample size calculations. These issues led to our development of a modified technique to improve venography examination adequacy rates. Ka¨lebo et al [35] highlighted the two most common reasons for inadequate venography examinations in multicentre clinical trials: insufficient contrast filling (16%) and unilateral examination (4%). Other reasons include obscuring of veins by metallic materials, a missing part on films, poor exposure, missing films and incomplete labelling [35]. These findings are consistent with our experience, with the most common reason for an inadequate venography examination being insufficient contrast filling of veins, especially of the anterior tibial veins. Many clinical trial venography protocols allow for the non-visualization of the anterior tibial vein [37–39], as isolated thrombi in this location are thought to be extremely rare and usually occur in combination with other calf vein thrombi [40–42]. Rose et al [11], however, demonstrated that two out of five asymptomatic anterior tibial vein thrombi detected by venography were isolated to this segment. Performing a unilateral venogram may occur as a result of patient refusal, failure of adequate venous cannulation, blood vessel rupture during contrast injection or movement of a butterfly needle [35]. Therefore, proper preparation of patients before their arrival in the radiology department remains an essential component of a successful examination. Despite having a reputation as a painful and difficult procedure, application of the preparation phase we describe will ensure the procedure can take place in a timely fashion with reduced patient discomfort. In those patients undergoing venography following lower-limb joint replacement, cannulation should be first attempted on the operated limb, as the majority of thrombi will occur ipsilateral to the surgical site [10]. If a patient then refuses to have a cannula placed in the other foot, at least the side most likely to be affected by thrombus will have been assessed. In our experience, with adequate preparation, successful cannulation can be achieved in 90–95% of patients. The other advantage of preparing patients in advance is that only patients in whom a successful procedure can be performed are brought to the fluoroscopy unit, thereby reducing the pressure of obtaining venous access in the The British Journal of Radiology, November 2007

fluoroscopy unit, which should in turn allow for ease of imaging and rapid movement of patients through the examination without disruption to working practices of the radiology department. Metallic prostheses, which can obscure the opacified vein, are another cause of inadequate venography examinations [35]. Being alert to this problem and preempting it by placing the patient onto his or her side and taking extra images or using a rotating C-arm will avoid this problem. In line with other interventional procedures, non-ionic contrast medium is used, which eliminates the majority of complications and contraindications, including anaphylactic reactions and discomfort along the injection line of the vein [43, 44]. In addition, the non-ionic medium has less thrombogenic effect on vessel walls, such that thrombosis related to endothelial damage is virtually eliminated [45]. Although it is appreciated that contrast-induced nephrotoxicity has been associated with non-ionic contrast, especially in patients with diabetes and pre-existing renal insufficiency, adherence to adequate patient hydration significantly reduces any potentially harmful effects caused by these disorders. The use of tourniquets remains a contentious issue, with disagreement between authors and practitioners of venography [5, 19]. We have found tourniquets to be useful in ensuring the initial opacification of the deep veins without superficial vein opacification. Opacification of superficial veins often obscures images of the deep veins, and causes difficulties in interpretation and assessment. However, application of an ankle tourniquet may result in failure to opacify the anterior tibial vein [5]. A compromise would therefore be to allow images of the deep veins of the calf to be obtained with the ankle tourniquet in situ. If the anterior tibial veins do not fill, the tourniquet is removed with further tilt of the fluoroscopy table in conjunction with administration of further contrast. So far, this technique has been performed only by a select group of individuals interested in obtaining good examinations for clinical trials. More widespread evaluation of the technique is required with further validation in clinical trials. Similarly, blinded comparison of the technique with traditional venography methods requires further evaluation. In summary, venography will remain the imaging modality of choice for accurate visualization of the deep calf veins and detection of asymptomatic DVT in clinical trials. We describe a successful modified technique that demonstrates a high venography adequacy rate for visualization of the veins. In addition, this technique allows for the rapid processing of patients with minimum disruption to the daily routine of the fluoroscopy unit.

Acknowledgments P Sidhu and D Quinlan developed the technique and wrote the protocol; P Sidhu, R Alikhan and D Quinlan refined the technique and performed the venograms; P Sidhu and T Ammar implemented the technique within the department; P Sidhu, T Ammar, R Alikhan and D Quinlan contributed to the writing of the paper. We thank all the radiographers and study coordinators 863


involved in performing this technique and caring for the patients undergoing these procedures.

References 1. Tapson VF, Carroll BA, Davidson BL, Elliott CG, Fedullo PF, Hales CA, et al. The diagnostic approach to acute venous thromboembolism. Clinical practice guideline. American Thoracic Society. Am J Respir Crit Care Med 1999;160:1043–66. 2. Rolfe MW, Solomon DA. Lower extremity venography: still the gold standard. Chest 1999;116:853–4. 3. Eriksson BI, Quinlan DJ. Oral anticoagulants in development: focus on thromboprophylaxis in patients undergoing orthopaedic surgery. Drugs 2006;66:1411–29. 4. Kalebo P, Ekman S, Lindbratt S, Eriksson BI, Pauli U, Zachrisson BE, et al. Percentage of inadequate phlebograms and observer agreement in thromboprophylactic multicenter trials using standardized methodology and central assessment. Thromb Haemost 1996;76:893–6. 5. Kalebo P, Anthmyr BA, Eriksson BI, Zachrisson BE. Optimization of ascending phlebography of the leg for screening of deep vein thrombosis in thromboprophylactic trials. Acta Radiol 1997;38:320–6. 6. Kearon C, Julian JA, Newman TE, Ginsberg JS. Noninvasive diagnosis of deep venous thrombosis. Ann Intern Med 1998;128:663–77. 7. Points to consider on clinical investigation of medicinal products for prophylaxis of intra- and post-operative venous thromboembolic risk. CPMP/EWP/707/98. London: The European Agency for the Evaluation of Medicinal Products; 2000. 8. Lensing AW, Buller HR, Prandoni P, Batchelor D, Molenaar AH, Cogo A, et al. Contrast venography, the gold standard for the diagnosis of deep-vein thrombosis: improvement in observer agreement. Thromb Haemost 1992;67:8–12. 9. Jongbloets LM, Lensing AW, Koopman MM, Buller HR, ten Cate JW. Limitations of compression ultrasound for the detection of symptomless postoperative deep vein thrombosis. Lancet 1994;343:1142–4. 10. Kalebo P, Anthmyr BA, Eriksson BI, Zachrisson BE. Phlebographic findings in venous thrombosis following total hip replacement. Acta Radiol 1990;31:259–63. 11. Rose SC, Zwiebel WJ, Miller FJ. Distribution of acute lower extremity deep venous thrombosis in symptomatic and asymptomatic patients: imaging implications. J Ultrasound Med 1994;13:243–50. 12. Quinlan DJ, Alikhan R, Gishen P, Sidhu PS. Variations in lower limb venous anatomy: implications for US diagnosis of deep vein thrombosis. Radiology 2003;228:443–8. 13. Borris LC, Christiansen HM, Lassen MR, Olsen AD, Schott P. Comparison of real-time B-mode ultrasonography and bilateral ascending phlebography for detection of postoperative deep vein thrombosis following elective hip surgery. The Venous Thrombosis Group. Thromb Haemost 1989;61:363–5. 14. Rose SC, Zwiebel WJ, Murdock LE, Hofmann AA, Priest DL, Knighton RA, et al. Insensitivity of color Doppler flow imaging for detection of acute calf deep venous thrombosis in asymptomatic postoperative patients. J Vasc Interv Radiol 1993;4:111–17. 15. Schellong SM, Beyer J, Kakkar AK, Halbritter K, Eriksson BI, Turpie AGG, et al. Ultrasound screening for asymptomatic deep vein thrombosis after major orthopaedic surgery: the VENUS study. J Thromb Haemost 2007;5:1431–7. 16. Rabinov K, Paulin S. Roentgen diagnosis of venous thrombosis in the leg. Arch Surg 1972;104:134–44. 17. Greitz T. The technique of ascending phlebography of the lower extremity. Acta Radiol 1954;42:421–41.

864

18. Thomas ML. Phlebography. Arch Surg 1972;104:145–51. 19. Thomas ML. Techniques of phlebography: a review. Eur J Radiology 1990;11:125–30. 20. Eriksson BI, Bergqvist D, Kalebo P, Dahl OE, Lindbratt S, Bylock A, et al. Ximelagatran and melagatran compared with dalteparin for prevention of venous thromboembolism after total hip or knee replacement: the METHRO II randomised trial. Lancet 2002;360:1441–7. 21. Lassen MR, Bauer KA, Eriksson BI, Turpie AG. Postoperative fondaparinux versus preoperative enoxaparin for prevention of venous thromboembolism in elective hip-replacement surgery: a randomised doubleblind comparison. Lancet 2002;359:1715–20. 22. Eriksson BI, Agnelli G, Cohen AT, Dahl OE, Mouret P, Rosencher N, et al. Direct thrombin inhibitor melagatran followed by oral ximelagatran in comparison with enoxaparin for prevention of venous thromboembolism after total hip or knee replacement. Thromb Haemost 2003;89:288–96. 23. Eriksson BI, Agnelli G, Cohen AT, Dahl OE, Lassen MR, Mouret P, et al. The direct thrombin inhibitor melagatran followed by oral ximelagatran compared with enoxaparin for the prevention of venous thromboembolism after total hip or knee replacement: the EXPRESS study. Thromb Haemost 2003;1:2490–6. 24. Eriksson BI, Bauer KA, Lassen MR, Turpie AGG. Fondaparinux compared with enoxaparin for the prevention of venous thromboembolism after hip-fracture surgery. New Engl J Med 2001;345:1298–304. 25. Eriksson BI, Lassen MR. Duration of prophylaxis against venous thromboembolism with fondaparinux after hip fracture surgery: a multicenter, randomized, placebo-controlled, double-blind study. Arch Intern Med 2003;163:1337–42. 26. Bergqvist D, Agnelli G, Cohen AT, Eldor A, Nilsson PE, Moigne-Amrani A, et al. Duration of prophylaxis against venous thromboembolism with enoxaparin after surgery for cancer. N Engl J Med 2002;346:975–80. 27. Agnelli G, Bergqvist D, Cohen AT, Gallus AS, Gent M. Randomized clinical trial of postoperative fondaparinux versus perioperative dalteparin for prevention of venous thromboembolism in high-risk abdominal surgery. Br J Surg 2005;92:1212–20. 28. Samama MM, Cohen AT, Darmon JY, Desjardins L, Eldor A, Janbon C, et al. A comparison of enoxaparin with placebo for the prevention of venous thromboembolism in acutely ill medical patients. Prophylaxis in Medical Patients with Enoxaparin Study Group. N Engl J Med 1999;341: 793–800. 29. Cohen AT, Davidson BL, Gallus AS, Lassen MR, Prins MH, Tomkowski W, et al. Efficacy and safety of fondaparinux for the prevention of venous thromboembolism in older acute medical patients: randomised placebo controlled trial. BMJ 2006;332:325–9. 30. Zachrisson BE, Jansen H. Phlebographic signs in fresh postoperative venous thrombosis of the lower extremity. Acta Radiol Diagn (Stockh) 1973;14:82–96. 31. Halliday P. Phlebography of the lower limb. Br J Surg 1968;55:220–6. 32. Rampton JB, Armstrong JD Jr. Bilateral venography of the lower extremities. Radiology 1977;123:802–4. 33. Couson F, Bounameaux C, Didier D, Geiser D, Meyerovitz MF, Schmitt HE, et al. Influence of variability of interpretation of contrast venography for screening of postoperative deep venous thrombosis on the results of a thromboprophylactic study. Thromb Haemost 1993;70:573–5. 34. Lee KR, Templeton AW, Cox GG, Dwyer SJ, III, McClure CB. Digital venography of the lower extremity. AJR Am J Roentgenol 1989;153:413–7.


Review article: Lower limb contrast venography 35. Ka¨lebo P, Eriksson BI, Zachrisson BE. Central assessment of bilateral phlebograms in a major multicentre thromboprophylactic trial. Reasons for inadequate results. Acta Radiol 1999;40:29–32. 36. Leizorovicz A, Kassai B, Becker F, Cucherat M. The assessment of deep vein thromboses for therapeutic trials. Angiology 2003;54:19–24. 37. Heit JA, Colwell CW, Francis CW, Ginsberg JS, Berkowitz SD, Whipple J, et al. Comparison of the oral direct thrombin inhibitor ximelagatran with enoxaparin as prophylaxis against venous thromboembolism after total knee replacement: a phase 2 dose-finding study. Arch Intern Med 2001;161:2215–21. 38. Francis CW, Davidson BL, Berkowitz SD, Lotke PA, Ginsberg JS, Lieberman JR, et al. Ximelagatran versus warfarin for the prevention of venous thromboembolism after total knee arthroplasty. A randomized, double-blind trial. Ann Intern Med 2002;137:648–55. 39. Francis CW, Berkowitz SD, Comp PC, Lieberman JR, Ginsberg JS, Paiement G, et al. Comparison of ximelagatran with warfarin for the prevention of venous thromboembolism after total knee replacement. N Engl J Med 2003; 349:1703–12.


40. Kerr TM, Cranley JJ, Johnson JR, Lutter KS, Riechmann GC, Cranley RD, et al. Analysis of 1084 consecutive lower extremities involved with acute venous thrombosis diagnosed by duplex scanning. Surgery 1990;108:520–7. 41. Mattos MA, Melendres G, Sumner DS, Hood DB, Barkmeier LD, Hodgson KJ, et al. Prevalence and distribution of calf vein thrombosis in patients with symptomatic deep venous thrombosis: a color-flow duplex study. J Vasc Surg 1996;24: 738–44. 42. Masuda EM, Kessler DM, Kistner RL, Eklof B, Sato DT. The natural history of calf vein thrombosis: lysis of thrombi and development of reflux. J Vasc Surg 1998;28:67–73. 43. Lea TM, Keeling FP, Piaggio RB, Treweeke PS. Contrast agent induced thrombophlebitis following leg phlebography: iopamidol versus meglumine iothalamate. Br J Radiol 1984;57:205–7. 44. Katayama H, Yamaguchi K, Kozuka T, Takashima T, Seez P, Matsuura K. Adverse reactions to ionic and nonionic contrast media. A report from the Japanese Committee on the Safety of Contrast Media. Radiology 1990;175: 621–8. 45. Albrechtsson U, Olsson CG. Thrombotic side-effects of lower-limb phlebography. Lancet 1976;1:723–4.

865