Diagnosis of venous thrombosis
and the post-thrombotic syndrome
Cover design: Renee Bosma
Printed by :
© 2008, L.W. Tick
The study described in this thesis was supported by a grant of the Netherlands
Heart Foundation (NHS-98.113)
Gildeprint, Enschede, the Netherlands
Diagnosis of venous thrombosis
the post-thrombotic syndrome
ter verkrijging van
de graad van Doctor aan de Universiteit Leiden,
volgens besluit van het College voor Promoties
te verdedigen op woensdag 24 september 2008
klokke 15.00 uur
Lidwine Winnifred Tick
geboren te Gouda
op gezag van Rector Magnificus prof. mr. P.F. van der Heijden,
Prof. dr F.R. Rosendaal
Prof. dr M.H.H. Kramer, Vrije Universiteit, Amsterdam
Co-promotores: Dr C.J.M. Doggen
Dr M.V. Huisman
Referent: Prof. dr M.M. Levi, Universiteit van Amsterdam, Amsterdam
Lid: Prof. dr W.R. Faber, Universiteit van Amsterdam, Amsterdam
Aan Renke, Meike, Julia en mijn ouders
Part I – Diagnosis of venous thrombosis
Chapter 2 Practical diagnostic management of patients with clinically
suspected deep vein thrombosis by clinical probability test,
compression ultrasonography, and D-dimer test.
Chapter 3 Effectiveness of managing suspected pulmonary embolism
using an algorithm combining clinical probability, D-dimer
testing, and computed tomography.
Chapter 4 Excluding pulmonary embolism without imaging tests; can
our diagnostic algorithm be optimized?
Chapter 5 High D-dimer levels increase the likelihood of pulmonary
Part II – The post-thrombotic syndrome
Chapter 6 Risk factors for the post-thrombotic syndrome in patients with
a first deep venous thrombosis.
Chapter 7 Predictors of the post-thrombotic syndrome with non-invasive
venous examinations in patients six weeks after a first
episode of deep venous thrombosis.
List of Publications
Venous thrombosis is a common disorder with an incidence in the general population of 1 to 3 in
1000 individuals per year1.
Deep venous thrombosis is caused by pathological thrombus formation in the leg. When part of the
thrombus is dislodged and migrates through the venous system to the pulmonary arteries, a
pulmonary embolism arises. Hence deep venous thrombosis and pulmonary embolism are often
regarded as different expressions of a single clinical entity called venous thrombosis.
In 1846 Virchow was the first to recognize that blood clots in the pulmonary artery originate as
venous thrombi2. He proposed three pathophysiologic concepts that contribute to thrombosis, namely
vessel wall injury, blood stasis, and changes in the composition of blood (hypercoagulability).
Virchow´s famous triad has withstood the test of time and still contributes to our understanding of the
pathophysiology of venous thrombosis3.
Venous thrombosis, if not treated, is associated with high morbidity and mortality4. Treatment with
anticoagulants can prevent mortality. However, anticoagulant treatment carries a substantial risk of
major hemorrhage. The risk of bleeding is 2.7 major bleeds per 100 treatment-years5. Therefore it is
important to confirm or exclude the diagnosis in patients with clinically suspected venous thrombosis.
Diagnosis of Venous Thrombosis
Imaging tests are necessary to diagnose venous thrombosis. The test of choice for clinically
suspected deep venous thrombosis is venous ultrasonography. For pulmonary embolism, computed
tomography is replacing ventilation perfusion scanning. Despite the accuracy of imaging tests, the
post-test probability of disease is highly dependent on pretest probability6.
The clinical appearance of venous thrombosis is heterogeneous and for a long time the clinical
parameters have been considered to be useless. The introduction in the nineties of a standardized
clinical probability test enabled physicians to stratify patients with clinically suspected venous
thrombosis into clinical probability categories with concomitant low and high risk of venous
thrombosis7,8. This clinical probability test accurately categorizes patients’ risk prior to diagnostic
Another strategy to determine the pretest probability in patients with clinically suspected venous
thrombosis is to incorporate D-dimer test in the diagnostic algorithm. D-dimer is a degradation
product of a cross-linked fibrin blood clot. The levels of D-dimer increase in the presence of
coagulation activation and subsequent fibrinolysis. D-dimer tests have a high sensitivity in excluding
thrombosis9. It has been suggested that a normal D-dimer level can be used to exclude thrombosis,
where high D-dimer levels necessitate further investigation with diagnostic imaging.
The pretest probability determination can be optimized by combining clinical probability assessment
and D-dimer test. The integration of clinical probability and D-dimer in diagnostic algorithms for
venous thrombosis led to a reduced need for imaging techniques10-15.
In patient with clinically suspected deep vein thrombosis, serial repeated testing with ultrasonography
is required to detect the few patients with progression of calf vein thrombosis to the proximal
veins16,17. However, routine serial testing is inefficient, inconvenient and not cost-effective18,19. To
reduce the need for repeat ultrasonography, we developed a new diagnostic strategy introducing D-
dimer test after ultrasonography in patients with an intermediate or high clinical probability test.
Pulmonary angiography is regarded as the gold standard test for the diagnosis of pulmonary
embolism. This procedure is invasive, expensive and requires a skilled radiologist and a cooperative
patient20. Ventilation perfusion scanning has been the imaging procedure of choice. The major
disadvantage of this procedure is that further testing is needed in 40% to 60% of patients due to non-
diagnostic test results. As a consequence computed tomography replaced ventilation perfusion
scanning. The position of computed tomography was unclear and complex and impractical algorithms
were used in the diagnosis of pulmonary embolism21-24. We developed a novel simplified diagnostic
algorithm using a dichotomized clinical decision rule, D-dimer testing and computed tomography.
The Post-Thrombotic Syndrome
The most common complication of venous thrombosis is the post-thrombotic syndrome (PTS) and
PTS develops in up to one half of patients after symptomatic deep vein thrombosis. PTS becomes
established within 1 to 2 years after deep vein thrombosis25.
The clinical manifestations of PTS are probably due to high walking venous pressure. Venous
hypertension occurs as a consequence of venous valvular incompetence with diminished calf muscle
pump function and persistent obstruction. This results in alterations of the skin microcirculation and
morphological skin changes26. Typical features of PTS include symptoms such as heaviness and pain
of the leg and signs such as oedema, hyperpigmentation and new venous ectasia. In severe cases
venous ulcers may develop27.
As PTS reduces quality of life28, and is costly to society29, it is important to identify patients at risk for
PTS at an early stage. However, there is no ‘gold standard’ test that establishes the diagnosis of
PTS. Venous hypertension will be present before clinical symptoms are manifest. Non-invasive
venous examinations, such as duplex scanning and strain gauge plethysmography, can be useful to
predict the development of PTS. With duplex scanning it is possible to measure the extend of the
initial thrombus, residual thrombosis and valvular reflux, while strain gauge plethysmography
quantifies venous outflow resistance and calf muscle pump function.
In contrast to the many identified risk factors for deep venous thrombosis30, the only identified risk
factors for the development of PTS are recurrent, ipsilateral deep venous thrombosis and an
increased body mass index. Other factors such as age, sex, idiopathic thrombosis, localization of
thrombosis, duration of anticoagulant therapy, factor V Leiden or prothrombin 20210A mutation
residual vein thrombosis, valvular reflux, venous outflow resistance and calf muscle pump function
show inconsistent results in previous studies31-36.
Elastic compression stockings are not only effective in preventing deep venous thrombosis but also
play an important role in the prevention of PTS. These stockings assist the calf muscle pump function
and reduce venous hypertension and reflux, thereby reducing edema and improving tissue
microcirculation. Randomized controlled trials have shown that daily use of elastic compression
stockings after deep venous thrombosis reduces the risk of PTS by approximately 50%33,37.
Although PTS is a common condition, it received little attention within the field of venous
thromboembolism research. The start of the Multiple Environmental and Genetic Assessment of risk
factors for venous thrombosis (MEGA) study provided us with the opportunity to assess the
cumulative incidence of PTS and to study risk factors in the development of PTS. At the same time
we designed a follow-up study to elucidate the functional hemodynamic changes that lead to PTS
and to identify patients at risk for PTS at an early stage.
Outline of the Thesis
The aim of the first part of this thesis is to investigate new diagnostic strategies for patients with
suspected venous thrombosis. For this purpose two multicenter studies addressing different aspects
of the diagnostic work up have been performed.
Chapter 2 describes the safety of ruling out deep vein thrombosis in patients with clinically suspected
thrombosis, using a management strategy, which combines clinical probability test, compression
ultrasonography, and D-dimer measurements. It also reports the reduced need for repeat
ultrasonogaphy. This management strategy is notable for introducing the D-dimer test after the
ultrasonography, and only selectively in patient with intermediate or high probability on the clinical
prediction rule. Chapter 3 presents the findings of a large clinical follow-up study in patients with
suspected pulmonary embolism. In this study the safety of excluding pulmonary embolism and
withholding anticoagulant therapy in patients with either the combination of an unlikely clinical
decision rule score and a normal D-dimer level or a normal computed tomography is evaluated.
Whether the cut-off levels of the clinical decision rule as well as the D-dimer test should be varied to
increase the clinical utility in excluding pulmonary embolism is discussed in Chapter 4. The clinical
consequences of strongly elevated D-dimer levels combined with a clinical probability score in the
management strategy in patients with suspected pulmonary embolism are addressed in Chapter 5.
The second part concerns the incidence, risk factors and early predictors of the PTS. Chapter 6
reports the cumulative incidence of PTS after a first deep vein thrombosis and the contribution of risk
factors in the development of PTS. The analyses are performed in the Multiple Environmental and
Genetic Assessment (MEGA) study of risk factors for venous thrombosis, a large population-based
study. Patients aged 18 to 70, with a first episode of deep venous thrombosis of the leg, are included
from six participating anticoagulation clinics in the Netherlands, between March 1999 and June 2002.
Chapter 7 describes the predictive value of non-invasive venous examination for the development of
PTS, assessed in a 2-year follow-up study.