Diagnosing deep vein thrombosis: the case for compression ultrasonography limited to the proximal veins

Abstract

In studies using phlebography as the gold standard, lower limb venous compression ultrasonography (CUS), an entirely non‐invasive test, has a sensitivity of 97%[95% confidence interval (CI) 96–98%] and a specificity of 98% for symptomatic proximal deep vein thrombosis (DVT) [1Kearon C. Ginsberg J.S. Hirsh J. The role of venous ultrasonography in the diagnosis of suspected deep venous thrombosis and pulmonary embolism.Ann Intern Med. 1998; 129: 1044-9Crossref PubMed Google Scholar]. It has become the cornerstone of DVT diagnosis in clinically suspected individuals [2American Thoracic Society. The diagnostic approach to acute venous thromboembolism: clinical practice guideline.Am J Respir Crit Care Med. 1999; 160: 1043-66Crossref PubMed Google Scholar, 3Keeling D.M. Mackie I.J. Moody A. Watson H.G. The diagnosis of deep vein thrombosis in symptomatic outpatients and the potential for clinical assessment and d‐dimer assays to reduce the need for diagnostic imaging.Br J Haematol. 2004; 124: 15-25Crossref PubMed Scopus (0) Google Scholar, 4Kearon C. Julian J.A. Math M. Newman T.E. Ginsberg J.S. Noninvasive diagnosis of deep venous thrombosis.Ann Intern Med. 1998; 128: 663-77Crossref PubMed Google Scholar]. The single well validated diagnostic criterion for deep vein thrombosis on CUS is absence of full compressibility of the deep vein when applying gentle pressure through the ultrasound probe. The extensiveness of the exam (particularly the inclusion of calf veins or not in the diagnostic procedure) is heavily debated among experts. The 3‐month thromboembolic risk in patients with a negative CUS of the proximal veins only is low: in management studies, it is around 1% in series using serial CUS (CUS repeated after 1 week in patients with an initially negative CUS [5Cogo A. Lensing A.W.A. Koopman M.M.W. Piovella F. Siragusa S. Wells P.S. Villalta S. Büller H.R. Turpie A.G.G. Prandoni P. Compression ultrasonography for diagnostic management of patients with clinically suspected deep vein thrombosis: prospective cohort study.BMJ. 1998; 316: 17-20Crossref PubMed Google Scholar, 6Bernardi E. Prandoni P. Lensing A.W. Agnelli G. Guazzaloca G. Scannapieco G. Piovella F. Verlato F. Tomasi C. Moia M. Scarano L. Girolami A. d‐Dimer testing as an adjunct to ultrasonography in patients with clinically suspected deep vein thrombosis: prospective cohort study: the Multicentre Italian d‐dimer Ultrasound Study Investigators Group.BMJ. 1998; 317: 1037-40Crossref PubMed Google Scholar, 7Wells P.S. Ginsberg J.S. Anderson D.R. Kearon C. Gent M. Turpie A.G. Bormanis J. Weitz J. Chamberlain M. Bowie D. Barnes D. Hirsh J. Use of a clinical model for safe management of patients with suspected pulmonary embolism.Ann Intern Med. 1998; 129: 997-1005Crossref PubMed Google Scholar, 8Kraaijenhagen R.A. Piovella F. Bernardi E. Verlato F. Beckers E.A. Koopman M.M. Barone M. Camporese G. Potter Van Loon B.J. Prins M.H. Prandoni P. Buller H.R. Simplification of the diagnostic management of suspected deep vein thrombosis.Arch Intern Med. 2002; 162: 907-11Crossref PubMed Google Scholar]) and about 2% in the study that studies a single CUS [9Perrier A. Desmarais S. Miron M.J. De Moerloose P. Lepage R. Slosman D. Didier D. Unger P.F. Patenaude J.V. Bounameaux H. Non‐invasive diagnosis of venous thromboembolism in outpatients.Lancet. 1999; 353: 190-5Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar] (Table 1, upper panel). Table 1 also shows that the 3‐month thromboembolic risk would also have been around 2% in the serial CUS series if the repeat CUS had not been performed due to the very low prevalence of proximal DVT at 1 week in patients with an initially negative CUS (this is assuming that all patients in whom a DVT was shown by the repeat CUS would have had a thromboembolic event during the 3‐month follow‐up if left untreated). Those data obviously question the need for diagnosing so‐called distal DVT, at least in non‐high clinical probability patients. As a matter of fact, the 3‐month thromboembolic risk in patients with clinically suspected DVT who had a negative venogram was found to be as high as 1.9% (95% CI 0.4–5.4%) [10Hull R. Hirsh J. Sackett D.L. Clinical validity of a negative venogram in patients with clinically suspected venous thrombosis.Circulation. 1981; 64: 622-5Crossref PubMed Google Scholar].Table 1Performance and safety of proximal only or proximal and distal CUS for diagnosing DVTSeriesPatients (n)Prevalence of DVT (%)Proportion of distal DVTs (%)Number of CUS performed per 100 patients (n)Three‐month thromboembolic risk, % (95% CI)*During 3‐month follow‐up in patients left untreated.Serial CUSSingle initial CUS****This figure is obtained by adding the patients in whom the repeat CUS at 1 week showed a DVT to the observed 3‐month thromboembolic risk in the serial CUS series (see text). N.A., not applicableProximal CUS onlyCogo et al. [5Cogo A. Lensing A.W.A. Koopman M.M.W. Piovella F. Siragusa S. Wells P.S. Villalta S. Büller H.R. Turpie A.G.G. Prandoni P. Compression ultrasonography for diagnostic management of patients with clinically suspected deep vein thrombosis: prospective cohort study.BMJ. 1998; 316: 17-20Crossref PubMed Google Scholar]170224N.A.1760.7 (0.3–1.2)1.6 (1.0–2.6)Bernardi et al. [6Bernardi E. Prandoni P. Lensing A.W. Agnelli G. Guazzaloca G. Scannapieco G. Piovella F. Verlato F. Tomasi C. Moia M. Scarano L. Girolami A. d‐Dimer testing as an adjunct to ultrasonography in patients with clinically suspected deep vein thrombosis: prospective cohort study: the Multicentre Italian d‐dimer Ultrasound Study Investigators Group.BMJ. 1998; 317: 1037-40Crossref PubMed Google Scholar]94628N.A.1760.7 (0.3–1.2)1.2 (0.5–2.3)Wells et al. [7Wells P.S. Ginsberg J.S. Anderson D.R. Kearon C. Gent M. Turpie A.G. Bormanis J. Weitz J. Chamberlain M. Bowie D. Barnes D. Hirsh J. Use of a clinical model for safe management of patients with suspected pulmonary embolism.Ann Intern Med. 1998; 129: 997-1005Crossref PubMed Google Scholar]59316N.A.1090.4 (0–0.9)1.2 (0.5–2.7)Kraaijenhagen et al. [8Kraaijenhagen R.A. Piovella F. Bernardi E. Verlato F. Beckers E.A. Koopman M.M. Barone M. Camporese G. Potter Van Loon B.J. Prins M.H. Prandoni P. Buller H.R. Simplification of the diagnostic management of suspected deep vein thrombosis.Arch Intern Med. 2002; 162: 907-11Crossref PubMed Google Scholar]175622N.A.1280.6 (0.1–1.8)2.5 (1.8–3.5)Perrier et al. [9Perrier A. Desmarais S. Miron M.J. De Moerloose P. Lepage R. Slosman D. Didier D. Unger P.F. Patenaude J.V. Bounameaux H. Non‐invasive diagnosis of venous thromboembolism in outpatients.Lancet. 1999; 353: 190-5Abstract Full Text Full Text PDF PubMed Scopus (0) Google Scholar]47424N.A.1210.7 (0.3–1.6)2.6 (0.2–4.9)Proximal and distal CUSElias et al. [11Elias A. Mallard L. Elias M. Alquier C. Guidolin F. Gauthier B. Viard A. Mahouin P. Vinel A. Boccalon H. A single complete ultrasound investigation of the venous network for the diagnostic management of patients with a clinically suspected first episode of deep venous thrombosis of the lower limbs.Thromb Haemost. 2003; 89: 221-7Crossref PubMed Google Scholar]623364573N.A.0.5 (0.1–1.8)Schellong et al. [12Schellong S.M. Schwarz T. Halbritter K. Beyer J. Siegert G. Oettler W. Schmidt B. Schroeder H.E. Complete compression ultrasonography of the leg veins as a single test for the diagnosis of deep vein thrombosis.Thromb Haemost. 2003; 89: 228-34Crossref PubMed Google Scholar]16461756100N.A.0.5 (0.1–1.8)Stevens et al. [13Stevens S.M. Elliott C.G. Chan K.J. Egger M.J. Ahmed K.M. Withholding anticoagulation after a negative result on duplex ultrasonography for suspected symptomatic deep venous thrombosis.Ann Intern Med. 2004; 140: 985-91Crossref PubMed Google Scholar]4451431100N.A.0.8 (0.2–1.3)* During 3‐month follow‐up in patients left untreated.** **This figure is obtained by adding the patients in whom the repeat CUS at 1 week showed a DVT to the observed 3‐month thromboembolic risk in the serial CUS series (see text). N.A., not applicable Open table in a new tab Three recent large series (Table 1, lower panel [11Elias A. Mallard L. Elias M. Alquier C. Guidolin F. Gauthier B. Viard A. Mahouin P. Vinel A. Boccalon H. A single complete ultrasound investigation of the venous network for the diagnostic management of patients with a clinically suspected first episode of deep venous thrombosis of the lower limbs.Thromb Haemost. 2003; 89: 221-7Crossref PubMed Google Scholar, 12Schellong S.M. Schwarz T. Halbritter K. Beyer J. Siegert G. Oettler W. Schmidt B. Schroeder H.E. Complete compression ultrasonography of the leg veins as a single test for the diagnosis of deep vein thrombosis.Thromb Haemost. 2003; 89: 228-34Crossref PubMed Google Scholar, 13Stevens S.M. Elliott C.G. Chan K.J. Egger M.J. Ahmed K.M. Withholding anticoagulation after a negative result on duplex ultrasonography for suspected symptomatic deep venous thrombosis.Ann Intern Med. 2004; 140: 985-91Crossref PubMed Google Scholar]) conclude that complete examination of the leg deep vein system without any other exam is safe and effective in managing patients with clinically suspected DVT. However, although the 3‐month thromboembolic risk appears to be 1.5% lower than for the CUS limited to the proximal veins, detecting clots in the posterior tibial or peroneal veins or even in calf muscle veins may be double‐edged: on one hand, the potential of reducing the 3‐month thromboembolic risk is small (because it is already quite low), and, on the other hand, the risk of false‐positive findings and subsequent unnecessary anticoagulant treatment in patients who could be left untreated, is quite high. As 31–56% of the DVTs diagnosed in those recent series resorting on complete examination of the lower limb veins were distal, such an approach entails a substantial risk of overdiagnosis and overtreatment that may outweigh the apparent small difference in terms of 3‐month thromboembolic risk. In addition, the exam protocols that include a study of the distal veins are quite cumbersome and require more specialized skills. Interestingly, Gottlieb et al. [14Gottlieb R.H. Voci S.L. Syed L. Shyu C. Fultz P.J. Rubens D.J. Strang J.G. Carson N. DiGrazio W.J. Francis C.W. Randomized prospective study comparing routine versus selective use of sonography of the complete calf in patients with suspected deep venous thrombosis.AJR Am J Roentgenol. 2003; 180: 241-5Crossref PubMed Google Scholar] randomized more than 500 patients clinically suspected of DVT to undergo routine complete US of the calf veins or selective exam in the area of calf symptoms if present. The rate of isolated calf DVT detected was very low and similar in the two groups (1.3 and 1.5%, respectively), and the 3‐month thromboembolic risk was below 1% with no difference between the groups. Again, these findings question the pertinence of the systematic complete calf veins exam, especially in view of a positive predictive value of at best 50% if we assume a specificity of distal CUS of 99% (quite an optimistic assumption) in a population with such a low prevalence of the disease. As long as convincing hard data on the need for treating isolated distal DVT are not available, we strongly feel that CUS should be limited to the proximal veins in clinically suspected DVT, except perhaps in patients with a high clinical probability in whom a lower rate of false‐positive CUS may be anticipated. Strategies with a so‐called complete CUS are associated with a high risk of overdiagnosis and, hence, potentially dangerous overtreatment without obvious clinical benefit.