Professional Documents
Culture Documents
Thrombosis Research
journal homepage: www.elsevier.com/locate/thromres
A R T I C L E I N F O A B S T R A C T
Keywords: The diagnostic approach to suspected symptomatic deep-vein thrombosis of the lower extremities is usually
Venous thrombosis based on non-invasive methods, including the estimation of clinical probability, the measurement of D-dimer
Diagnosis levels, and ultrasonography. The present review discusses the evidence available from the literature about the
Clinical decision rule management of the first episode of suspected deep-vein-thrombosis.
Fibrin fibrinogen degradation products
Ultrasonography
Doppler
Color
Recurrence
1. Introduction Therefore, the main challenge for any diagnostic approach is to rule-out
DVT safely (i.e., low incidence of thromboembolic events at follow-up
For descriptive, as well as diagnostic and prognostic purposes, the in patients left untreated on the basis of normal findings, otherwise
deep veins network of the lower limbs is classically divided into two “cleared” from DVT) and efficiently (i.e., the proportion of patients in
regions: the proximal and the distal territory. Noteworthy, proximal whom a given strategy may be safely applied). The commonly accepted
DVT is more frequently associated with pulmonary embolism (PE) [1], safety threshold is below 2%, corresponding to the follow-up pre-
and recurrence [2] than isolated distal DVT. The former includes the valence of DVT in patients with a normal venography [8].
femoral (common, superficial, deep or profunda) veins and the popli- Venography, the official “gold standard” for the diagnosis of DVT, is
teal vein; the latter comprises the paired anterior and posterior tibial seldom used in everyday practice, being invasive, costly, technically
veins, and the peroneal veins, cumulatively known as axial, plus the demanding, painful, contraindicated in case of allergy or renal in-
muscular (gastrocnemius, soleal) veins [3]. The calf “trifurcation”, sufficiency, and difficult to interpret, with considerable inter- and intra-
formed by the joining of the tibial and peroneal veins, though formally observer variability [5]. Alternative (invasive) imaging approaches, i.e.:
belonging to the distal venous district, is usually screened when prox- CT- and MR-venography, not only share many of the same limitations of
imal CUS is performed. Finally, it is now commonplace to include in the venography, but also do not possess adequate accuracy to be used as
proximal territory also the last 3 cm of the superficial veins close to the gold standard [5].
saphenous junctions. Current non-invasive diagnostic algorithms to rule-out suspected
In symptomatic patients, as shown by classic venographic studies, symptomatic DVT include pretest probability estimation, D-dimer, and
DVT invariably develops in the venous-valves sinuses of the distal ultrasonography [5,9,10]. We will discuss the relevant literature con-
network, extending to the proximal system, in the absence of prophy- cerning those different strategies in the following sections. Noteworthy,
laxis or treatment, in 5–20% of the patients [4,5]. Conversely, in only ultrasonography may be used a stand-alone test to rule-in or rule-
asymptomatic patients, DVT may arise anywhere in the deep-vein out DVT.
system [3]. This observation, along with the more common finding of
small and non-occlusive thrombi, may account for the lower sensitivity 1.1. Pretest probability
of ultrasonography for asymptomatic DVT [6].
In recent studies, the prevalence of DVT in symptomatic patients Although useful to raise the clinical suspicion of DVT, individual
was around 10–15%, suggesting a low referral-threshold, as compared clinical features; such as, calf pain or swelling, warmth, tenderness,
with older (venographic) studies reporting figures as high as 35% [7]. erythema, oedema, difference in calf diameter, Homan's sign, history of
⁎
Corresponding author.
E-mail addresses: enrico.bernardi@aulss2.veneto.it (E. Bernardi), giuseppe.camporese@aopd.veneto.it (G. Camporese).
http://dx.doi.org/10.1016/j.thromres.2017.10.006
Received 21 April 2017; Received in revised form 17 September 2017; Accepted 6 October 2017
0049-3848/ © 2017 Elsevier Ltd. All rights reserved.
Please cite this article as: Bernardi, E., Thrombosis Research (2017), http://dx.doi.org/10.1016/j.thromres.2017.10.006
E. Bernardi, G. Camporese Thrombosis Research xxx (xxxx) xxx–xxx
Table 1 Table 2
The modified Wells DVT rule [12]. The primary care rule [19].
2
E. Bernardi, G. Camporese Thrombosis Research xxx (xxxx) xxx–xxx
Table 3
Performance of different strategies for the exclusion of deep-vein thrombosis in symptomatic patients.
Geersing 2014 [9] Meta-analysis 3.114b Primary care Wells ≤ 1 + normal D-dimer 1.2 (0.7 to 1.8) 29 (20 to 40)
Buller 2009 [21] Prospective 1.002 Primary care Oudega ≤ 3 1.4 (0.6 to 2.9) 49 (46 to 52)
Van der Velde 2011 [22] Prospective 495b Primary care Oudega ≤ 3 1.4 (0.6 to 3.0) 57 (54 to 61)
447b Wells ≤ 1 + normal D-dimer 1.6 (0.7 to 3.3) 70 (66 to 73)
Schouten 2012 [28] Retrospective 2.086 Primary care, all patients Wells ≤ 1 + normal D-dimerc 0.5 (0.01 to 1.3) 48 (44 to 52)
Age ≥ 80 years 0 (0.0 to 15.0) 35 (24 to 49)
Schutgens 2003 [40] Prospective 812 Wells ≤ 2 + normal D-dimer 0.6 (0.02 to 3.1) 22 (19 to 25)
Geersing 2014 [9] Meta-analysis 3225b Outpatients Wells ≤ 1 + normal quantitative D-dimer 0.4 (0.3 to 0.6) 21 (13 to 32)
4400b Wells ≤ 1 + normal qualitative D-dimer 0.6 (0.4 to 0.9) 46 (35 to 57)
Silveira 2015 [18] Prospective 1.135 Inpatients Wells ≤ 0 5.9 (3.0–11.3) 12 (10 to 14)
Cogo 2005 [44] Prospective 1.741 Outpatients Normal serial CUS 0.7 (0.3 to 1.2) NA
Bernardi 1998 [45] Prospective 946 Outpatients Normal serial CUS + normal D-dimer 0.4 (0 to 0.9) 72 (69 to 75)
Kraijenhagen 2002 [46] Prospective 1.739 Outpatients Normal serial CUS + normal D-dimer 0.7 (0.3 to 1.6) 48 (45 to 50)
Wells 1997 [47] Prospective 593 Outpatients Wells ≤ 0 + normal single CUS 0.3 (0 to 1.7) 54 (50 to 58)
Schellong 2003 [3] Prospective 1.646 Outpatients Normal WLUS 0.3 (0.1–0.8) NA
Subramaniam 2005 [44] Prospective 542 Outpatients Normal WLUS 0.2 (0.01 to 1.3) NA
Gottlieb 2003 [48] Randomized 261b Outpatients Selective WLUSd 0.8 (0.1 to 2.7) 42 (36 to 49)
Johnson 2010 [49] Meta-analysis 10.090 Outpatients with Wells ≥ 3 Normal WLUS 2.5 (0.9 to 6.1) NA
Stevens 2013 [50] Prospective 183 Outpatients with Wells ≥ 2 Normal WLUS 0.6 (0.02 to 3.3) NA
Ageno 2015 [53] Prospective 410b Outpatients Wells ≥ 2 + abnormal D-dimer + normal WLUS 1.5 (0.5 to 4.3) 51 (46 to 56)
DVT: deep-vein thrombosis; CUS: compression ultrasonography; WLUS: whole-leg ultrasonography; NA: not applicable.
a
Safety: proportion of patients left untreated on the basis of normal testing with the proposed strategy who develop thromboembolic events during follow-up Efficiency: proportion of
patients in whom DVT can be excluded on the basis of the proposed strategy. Expressed as: % (95% confidence interval).
b
Subset of the data is reported.
c
Age-dependent D-dimer cut-off: if > 50 years, cut-off = age × 10 mcg/L.
d
WLUS was only performed if calf symptoms/signs were present, otherwise patients underwent CUS.
tative assays possess higher sensitivity and lower specificity while the Diagnostic strategies Serial CUS Single Serial CUS Single WLUS
contrary is true for qualitative, and point-of-care tests [6,32,38,39]. + D-dimer WLUS
Furthermore, D-dimer assays are less sensitive for isolated distal as
compared with proximal DVT, and for asymptomatic as compared with Patient n. 1045 1053 257 264
Initial DVT 23% 26% 23% 38%
symptomatic DVT [6,39]; finally, the overall performance of D-dimer
prevalence
assays decreases with increasing DVT prevalence [36]. VTE at follow-upb 0.9 (0.3 to 1.2 (0.5 to 2.0 (0.6 to 1.2 (0.2 to
Thus, D-dimer should not be used as a stand-alone test for to rule-in 1.8) 2.2) 5.1) 4.3)
b
or rule-out DVT [5,10]. Instead, there is sound clinical evidence that Absolute difference 0.3 (− 1.4 to 0.8) 0.8 (− 2.3 to 2.9)
suspected symptomatic DVT can be safely excluded in patients with a
DVT: deep-vein thrombosis; CUS: compression ultrasonography; WLUS: whole-leg ultra-
non-high, or an unlikely probability according to Wells, and normal D-
sonography; VTE: venous thromboembolism.
dimer results, both with a qualitative or a quantitative assay, without a
Patients with Wells DVT score of ≤1 did not undergo ultrasonography.
the need for further testing (Table 3) [40,41]. A recent meta-analysis of b
Expressed as: % (95% confidence interval).
13 studies, including > 10.000 patients, reported that in patients with
an unlikely probability according to the Wells' rule, DVT may be safely the common femoral vein at the groin and the popliteal vein in the
excluded on the basis of normal D-dimer results, irrespective if the popliteal fossa, including the so-called trifurcation (i.e.: the most
latter is either a qualitative point-of-care or a quantitative laboratory- proximal segments of the tibial and peroneal veins), are investigated
based assay (Table 3) [9]; instead, in patients with a likely or a high [42]. It is a simple, quick, bedside approach that does not require
probability level, ultrasound testing should be performed [5,10,12]. specialized operators or high-range technical equipment, widely im-
In conclusion, a normal D-dimer result in combination with an plemented in Emergency Departments, and hospital wards. Hence, CUS
unlikely, or non-high pretest probability according to Wells, allows for is usually available after-hours and during weekends. With WLUS, the
the exclusion of suspected symptomatic DVT. To the contrary, ab- whole deep-vein network of the leg is scanned in a continuous fashion,
normal D-dimer levels mandate further testing; namely, ultra- from the femoral veins, through the popliteal vein, to the distal veins. It
sonography. is usually performed by consultants in vascular medicine or vascular
surgery; thus, it is only available during office hours. Furthermore, it
1.3. Ultrasonography demands dedicated high-level machines and requires the patients to be
moved from the ED/ward to be evaluated [3].
Two distinct ultrasound approaches are available to rule in/out DVT Both approaches are based on the single diagnostic criterion of vein
in symptomatic patients (Table 4): compression ultrasonography (CUS; compressibility; broadly, if the veins are compressible, then DVT is
also, “2-point ultrasonography”, or “limited-US”), and whole-leg ul- ruled-out; if the veins are not compressible, then DVT is ruled-in [43].
trasonography (WLUS; also, “echo-color-Doppler”). Notably, while DVT can be safely excluded after a single normal WLUS
With CUS only two distinct spots of the proximal-vein system, i.e., [3,44], this is not the case with CUS. Indeed, as the distal veins are not
3
E. Bernardi, G. Camporese Thrombosis Research xxx (xxxx) xxx–xxx
investigated, repeat testing within one week is mandatory to identify A recent prospective, multicentre, cohort study devised a new se-
proximal extension of undetected distal DVT, the relative timing being quential strategy, including the evaluation of pretest probability with
mandated by the clinical picture; that is, the patients' symptoms are the Wells rule, D-dimer, CUS and WLUS, as follows [53]. All patients
either stable or waning, then repeat CUS may be done after 7 days, underwent D-dimer testing; patients with unlikely probability and
otherwise CUS should be performed as soon as possible (Table 3) [42]. normal D-dimer were spared further testing; those with either a likely
Although safe, repeat testing of all patients with a normal baseline CUS probability or abnormal D-dimer underwent a single CUS test; and
to detect only 3–5% extending distal DVT is inconvenient and expensive those with both a likely probability and abnormal D-dimer were eval-
[42]. Unnecessary repeat testing may be safely reduced (from 70 to uated with WLUS [53]. Only 35% of the patients required WLUS, and in
80% to 15–30%) by combining a single normal CUS test with either a half (49%) of them DVT was confirmed, indicating a high dis-
normal (quantitative, or qualitative) D-dimer, or with a low pretest criminatory power of the algorithm. The overall prevalence of throm-
probability according to Wells (Table 3) [45–47]. On the other side, boembolic events at follow-up in 950 untreated patients was < 1%,
selective use of WLUS, triggered by the presence of pain, redness or confirming the safety of the sequential strategy; however, in high-risk
swelling in the calf, was compared to routine WLUS in a randomized patients with both a likely Wells score and abnormal D-dimer, the upper
prospective study (Table 3) [48]. A < 1% prevalence of events was confidence limit for the prevalence was > 4% (Table 3).
observed in both groups, and selective use of WLUS spared calf-vein The results reported by the latter study and by the cited meta-
evaluation in > 40% of patients, statistically significantly reducing the analysis [49], although likely due to the small sample included, suggest
mean evaluation time (23 ± 14 min, vs 32 ± 14 min). the hypothesis that a single WLUS might not be sufficiently safe in high-
A meta-analysis of 7 studies, including > 10.000 patients, con- risk patients. Interestingly, in a classic accuracy study of the Wells rule,
firmed the safety of a single normal WLUS for the exclusion of DVT, but in which venography was planned for all cases of discrepancy between
reported surprisingly high (up to 3-fold the expected proportion) rates pretest clinical probability and CUS, the negative predictive value of a
of venous thromboembolism during follow-up in patients with a high normal CUS was only 82% (95% CI, 59.7 to 94.8) in patients with high
Wells score (Table 3) [49]. Conversely, both a recent small prospective pretest probability [47]. To our knowledge, a study evaluating WLUS
cohort study and a randomized clinical trial, investigating the safety of against venography in high-risk patients is not available from the lit-
withholding anticoagulation from patients with a likely or not-low erature, and it's very unlikely to be performed based on the con-
Wells score, reported a low rate of thromboembolic complication after a siderations stated in the introduction. None the less, we believe the
single normal WLUS, ranging between 0.6 and 1.2% (Tables 3, 4) signal coming from this subgroup of high-risk patients might be further
[50,51]. investigated in a properly designed trial, perhaps using a combined
Two randomized studies compared CUS and WLUS in symptomatic (instrumental and clinical) end-point.
outpatients with suspected DVT [51,52]. The first allocated all patients In conclusion, venous ultrasonography allows for both safe con-
to either serial CUS plus D-dimer, or a single (“one-shot”) WLUS; the firmation and exclusion of suspected symptomatic DVT. Compression
second randomized patients with a likely Wells score or an abnormal D- ultrasonography is widely available, quick, but needs to be repeated in
dimer, to either serial CUS or one-shot WLUS. Both studies failed to all patients with normal baseline testing, unless combined with either a
record statistically significant differences between the two ultrasound low pretest probability or a normal D-dimer. Whole-leg ultra-
approaches, despite a statistically significantly higher initial prevalence sonography allows for “one-shot” testing, but is rarely available outside
of DVT in patients managed by WLUS, due to distal-vein testing working-hours and requires highly skilled operators; furthermore, it
(Table 4). The authors concluded either strategy is safe to exclude raises the point of isolated distal DVT, neglected by CUS, though with
suspected symptomatic DVT; however, the debate on the optimal equivalent safety. Never the less, investigating the whole leg allows also
management of isolated distal vein thrombosis is on-going [5]. prompt identification of other pathologic conditions (such as, muscle-
4
E. Bernardi, G. Camporese Thrombosis Research xxx (xxxx) xxx–xxx
tear, joint-effusion, erysipelas, phlebitis, arthrosis, arthritis, chronic [13] S. Goodacre, A.J. Sutton, F.C. Sampson, Meta-analysis: the value of clinical as-
sessment in the diagnosis of deep venous thrombosis, Ann. Intern. Med. 143 (2005)
venous insufficiency, lymphatic stasis, etc.), in up to 50% of patients in 129–139.
whom suspected DVT has already been ruled out [48,54,55]. [14] D.R. Anderson, P.S. Wells, I. Stiell, B. MacLeod, M. Simms, L. Gray, K.S. Robinson,
J. Bormanis, M. Mitchell, B. Lewandowski, G. Flowerdew, Thrombosis in the
emergency department: use of a clinical diagnosis model to safely avoid the need for
2. Conclusions urgent radiological investigation, Arch. Intern. Med. 159 (5) (1999) 477–482.
[15] R.P. Engelberger, D. Aujesky, L. Calanca, P. Staeger, O. Hugli, L. Mazzolai,
Invasive imaging techniques such as venography, albeit formally Comparison of the diagnostic performance of the original and modified Wells score
in inpatients and outpatients with suspected deep vein thrombosis, Thromb. Res.
regarded as “gold standard” for the diagnosis of DVT, are infrequently 127 (2011) 535–539, http://dx.doi.org/10.1016/j.thromres.2011.02.008.
used – if at all, in clinical practice. Ultrasonography is the most vali- [16] R. Oudega, A.W. Hoes, K.G. Moons, The Wells rule does not adequately rule out
dated and accurate non-invasive imaging method in symptomatic pa- deep venous thrombosis in primary care patients, Ann. Intern. Med. 143 (2005)
100–107.
tients, and may be performed either in a simplified fashion (CUS), in
[17] M. Sartori, B. Cosmi, C. Legnani, E. Favaretto, L. Valdré, G. Guazzaloca,
which only two proximal venous spots are scanned; or in an extended G. Rodorigo, M. Cini, G. Palareti, The Wells rule and D-dimer for the diagnosis of
mode (WLUS), in which both the proximal and the distal veins are in- isolated distal deep vein thrombosis, J. Thromb. Haemost. 10 (11) (2012)
vestigated. Ultrasonography can be safely spared in patients with un- 2264–2269, http://dx.doi.org/10.1111/j.1538-7836.2012.04895.
[18] P.C. Silveira, I.K. Ip, S.Z. Goldhaber, G. Piazza, C.B. Benson, R. Khorasani,
likely pretest probability for DVT and normal D-dimer levels. Performance of Wells score for deep vein thrombosis in the inpatient setting, JAMA
Otherwise, if CUS is performed, all patients with normal baseline Intern. Med. 175 (7) (2015) 1112–1117, http://dx.doi.org/10.1001/
findings need to undergo either a repeat CUS within one week, unless jamainternmed.2015.1687.
[19] R. Oudega, K.G. Moons, A.W. Hoes, Ruling out deep venous thrombosis in primary
they are at low pretest probability, or have normal D-dimer results. If care. A simple diagnostic algorithm including D-dimer testing, Thromb. Haemost.
WLUS is chosen, no further testing is foreseen in patients with normal 94 (2005) 200–205.
baseline findings. Fig. 1 illustrates a suggested algorithm for the eva- [20] D.B. Toll, R. Oudega, R.J. Bulten, A.W. Hoes, K.G. Moons, Excluding deep vein
thrombosis safely in primary care, J. Fam. Pract. 55 (2006) 613–618.
luation of suspected DVT in symptomatic patients, based on the current [21] H.R. Buller, A.J. Ten Cate-Hoek, A.W. Hoes, M.A. Joore, K.G. Moons, R. Oudega,
review. et al., Safely ruling out deep venous thrombosis in primary care, Ann. Intern. Med.
150 (2009) 229–235.
[22] E.F. Van der Velde, D.B. Toll, A.J. Ten Cate-Hoek, R. Oudega, H.E. Stoffers,
Conflicts of interest
P.M. Bossuyt, et al., Comparing the diagnostic performance of 2 clinical decision
rules to rule out deep vein thrombosis in primary care patients, Ann. Fam. Med. 9
None. (2011) 31–36.
[23] J. Kelly, A. Rudd, R.R. Lewis, B.J. Hunt, Plasma D-dimers in the diagnosis of venous
thromboembolism, Arch. Intern. Med. 162 (2002) 747–756.
Role of funding source [24] P. Raimondi, O. Bongard, P. de Moerloose, G. Reber, F. Waldvogel, H. Bounameaux,
D-Dimer plasma concentration in various clinical conditions: implication for the use
This research did not receive any specific grant from funding of this test in the diagnostic approach of venous thromboembolism, Thromb. Res.
69 (1993) 125–130.
agencies in the public, commercial, or not-for-profit sectors. [25] G. Sodeck, H. Domanovits, M. Schillinger, M.P. Ehrlich, G. Endler, H. Herkner,
A. Laggner, D-dimer in ruling out acute aortic dissection: a systematic review and
References prospective cohort study, Eur. Heart J. 28 (2007) 3067–3075.
[26] C. Haase, M. Joergensen, C. Ellervik, M.K. Joergensen, L. Bathum, Age- and sex-
dependent reference intervals for D-dimer: evidence for a marked increase by age,
[1] M. Meignan, J. Rosso, H. Gauthier, et al., Systematic lung scans reveal a high fre- Thromb. Res. 132 (6) (2013) 676–680, http://dx.doi.org/10.1016/j.thromres.2013.
quency of silent pulmonary embolism in patients with proximal deep venous 09.033.
thrombosis, Arch. Intern. Med. 160 (2000) 159–164. [27] R.A. Douma, M. Tan, R.E.G. Schutgens, S.M. Bates, A. Perrier, C. Legnani,
[2] J.D. Douketis, C. Kearon, S. Bates, E.K. Duku, J.S. Ginsberg, Risk of fatal pulmonary D.H. Biesma, J.A. Ginsberg, H. Bounameaux, G. Palareti, M. Carrier, G.C. Mol, G. Le
embolism in patients with treated venous thromboembolism, JAMA 279 (1998) Gal, P.W. Kamphuisen, M. Righini, Using an age-dependent D-dimer cut-off value
458–462. increases the number of older patients in whom deep vein thrombosis can be safely
[3] S.M. Schellong, T. Schwarz, K. Halbritter, J. Beyer, G. Siegert, W. Oettler, excluded, Haematologica 97 (10) (2012) 1507–1513, http://dx.doi.org/10.3324/
B. Schmidt, H.E. Schroeder, Complete compression ultrasonography of the leg veins haematol.2011.060657.
as a single test for the diagnosis of deep vein thrombosis, Thromb. Haemost. 89 (2) [28] H.J. Schouten, H.L. Koek, R. Oudega, G.J. Geersing, K.J. Janssen, J.J. van Delden,
(2003) 228–234. K.G. Moons, Validation of two age dependent D-dimer cut-off values for exclusion of
[4] M. Righini, H. Bounameaux, Clinical relevance of distal deep vein thrombosis, Curr. deep vein thrombosis in suspected elderly patients in primary care: retrospective,
Opin. Pulm. Med. 14 (5) (2008) 408–413. cross sectional, diagnostic analysis, BMJ 344 (2012) e2985, , http://dx.doi.org/10.
[5] S.M. Bates, R. Jaeschke, S.M. Stevens, S. Goodacre, P.S. Wells, M.D. Stevenson, 1136/bmj.e2985.
C. Kearon, H.J. Schunemann, M. Crowther, S.G. Pauker, R. Makdissi, G.H. Guyatt, [29] P.M. Baker, D.M.A. Keeling, Single dose of Low-molecular-weight Heparin (LMWH)
American College of Chest Physicians, Diagnosis of DVT: antithrombotic therapy invalidates the use of D-dimer as part of a Deep Vein Thrombosis (DVT) diagnostic
and prevention of thrombosis, American College of Chest Physicians Evidence- algorithm, Int. J. Lab. Hematol. 39 (1) (2017) e17–e18, http://dx.doi.org/10.1111/
based Clinical Practice Guidelines, 9th ed, Chest, 141(2 Suppl) 2012, pp. ijlh.12579.
e351S–418S, , http://dx.doi.org/10.1378/chest.11-2299. [30] F. van der Graaf, H. van den Borne, M. van der Kolk, P.J. de Wild, G.W. Janssen,
[6] S. Goodacre, F. Sampson, M. Stevenson, A. Wailoo, A. Sutton, S. Thomas, et al., S.H. van Uum, Exclusion of deep venous thrombosis with D-dimer testing: com-
Measurement of the clinical and cost-effectiveness of non-invasive diagnostic parison of 13 D-dimer methods in 99 outpatients suspected of deep venous
testing strategies for deep vein thrombosis, Health Technol. Assess. 10 (15) (2006). thrombosis using venography as reference standard, Thromb. Haemost. 83 (2)
[7] A.J. Ten Cate-Hoek, M.H. Prins, Management studies using a combination of D- (2000) 191–198.
dimer test result and clinical probability to rule out venous thromboembolism: a [31] R. Anastasio, L. Quartero, C. POrta, F. FAlaschi, S. Baratto, M.A. Bressan,
systematic review, J. Thromb. Haemost. 3 (2005) 2465–2470. S. Siragusa, Diagnostic accuracy of D-dimer test is affected by thrombus site,
[8] R. Hull, J. Hirsh, D.L. Sackett, D.W. Taylor, C. Carter, A.G. Turpie, P. Powers, duration of symptoms, and heparin treatment, J. Thromb. Haemost. (Suppl. 1)
M. Gent, Clinical validity of a negative venogram in patients with clinically sus- (2001) 748.
pected venous thrombosis, Circulation 64 (3) (1981) 622–625. [32] P.D. Stein, R.D. Hull, K.C. Patel, R.E. Olson, W.A. Ghali, R. Brant, R.K. Biel,
[9] G.J. Geersing, N.P.A. Zuithoff, C. Kearon, D.R. Anderson, A.J. ten Cate-Hoek, V. Bharadia, N.K. Kalra, D-dimer for the exclusion of acute venous thrombosis and
J.L. Elf, S.M. Bates, A.W. Hoes, R.A. Kraaijenhagen, R. Oudega, R.E.G. Schutgens, pulmonary embolism: a systematic review, Ann. Intern. Med. 140 (8) (2004)
S.M. Stevens, S.C. Woller, P.S. Wells, Moons KGM, Exclusion of deep vein throm- 589–602.
bosis using the Wells rule in clinically important subgroups: individual patient data [33] G.J. Geersing, K.J. Janssen, R. Oudega, L. Bax, A.W. Hoes, J.B. Reitsma,
meta-analysis, BMJ 348 (2014) g1340, , http://dx.doi.org/10.1136/bmj.g1340. K.G. Moons, Excluding venous thromboembolism using point of care D-dimer tests
[10] A. Quaseem, W. Snow, P. Barry, et al., Current diagnosis of venous thromboem- in outpatients: a diagnostic meta-analysis, BMJ 339 (2009) b2990, , http://dx.doi.
bolism in primary care: a clinical practice guideline from the American Academy of org/10.1136/bmj.b2990.
family physicians and the American College of Physicians, Ann. Fam. Med. 5 (2007) [34] D. Neale, C. Tovey, A. Vali, S. Davies, K. Myers, M. Obiako, V. Ramkumar, A. Hafiz,
57–62, http://dx.doi.org/10.1370/afm.667. Evaluation of the simplify D-dimer assay as a screening test for the diagnosis of deep
[11] S.M. Stevens, W. Ageno, Review: the wells rule is more useful than individual vein thrombosis in an emergency department, Emerg. Med. J. 21 (6) (2004)
clinical features for predicting risk of deep venous thrombosis, Evid. Based Med. 11 663–666.
(2) (2006) 56. [35] K. Boeer, R. Siegmund, D. Schmidt, T. Deufel, M. Kiehntopf, Comparison of six D-
[12] P.S. Wells, C. Owen, S. Doucette, D. Fergusson, H. Tran, Does this patient have dimer assays for the detection of clinically suspected deep venous thrombosis of the
deep-vein thrombosis? JAMA 295 (2006) 199–207. lower extremities, Blood Coagul. Fibrinolysis 20 (2) (2009) 141–145, http://dx.doi.
5
E. Bernardi, G. Camporese Thrombosis Research xxx (xxxx) xxx–xxx