The microbiome and the chance of recurrent venous thrombosis determined by the DASH-score (MIDAS-study)

In around half of patients who suffer a venous thrombotic event (VTE), no clear
cause can be found. The main clinical dilemma in these patients is to decide
whether to continue anticoagulant treatment, which carries a significant risk
of major bleeding, or to withdraw treatment, which puts patients at risk of a
recurrent venous thrombotic event. Several models have been developed to more
accurately predict the risk of recurrent VTE, but these have been shown to be
too inaccurate for clinical decision-making. One of these models is the
DASH-score. By contrast, there has been less attention in recent research for
identifying novel, modifiable risk factors for a recurrent VTE beyond the
classic risk factors (surgery, cancer, pregnancy/puerperium,
estrogen/progestagens containing birth control methods). Identifying such risk
factors and treating them would be an ideal strategy to solve the
abovementioned clinical dilemma; by doing so, the individual risk of recurrence
is lowered and thereby the necessity to expose patients to a risk of bleeding
is removed. A possible risk factor for recurrent VTE might be inflammation
associated with permeability of the gastrointestinal tract. Diseases in which
permeability of this tract (i.e. periodontitis, inflammatory bowel disease) is
elevated have already been shown to be associated with an elevated risk of
thrombotic events, both arterial and venous. The inflammation associated with
this elevated permeability is mainly caused by bacteria leaking into the
bloodstream- these induce an inflammatory response with downstream changes in
metabolic and coagulation status. This process has been termed endotoxemia. It
has already been shown that administration of probiotics is effective for
reversing endotoxemia. Disturbances of the human gut and oral microbiome have
been associated with increased endotoxemia as well as with systemic low-grade
inflammation. We intend to investigate whether such disturbances can be
identified in patients who have a high risk of recurrent venous thrombosis, as
determined by the DASH score.
Additionally, basic research has shown common pathways between the complement
system and the coagulation cascade. Complement has already been implicated in
several diseases that have a distinct thrombotic presentation, such as
paroxysmal nocturnal hemoglobinuria, for which an inhibitor of complement
(Eculizumab) has proven to be a promising therapeutic agent. To our knowledge,
complement activity has not yet been explored by epidemiological methods as a
risk factor for venous thrombosis.
Finally, skin autofluorescence (SAF) measurement of Advanced glycation end
products (AGEs) has been validated as a non-invasive method of measuring skin
AGEs. This measurement method has been shown to be predictive of of arterial
cardiovascular disease. Epidemiologic research has shown that arterial disease
and venous thromboembolism probably show common risk factors than previously
thought.

We want to explore whether disturbances in the oral and gut microbiome can be
identified in patients with a high risk of recurrence (determined by a DASH
score greater than 1) compared to the oral and gut microbiome of patients with
a low risk of recurrent venous thrombosis. Second, we will determine whether
pathologic changes in the microbiome are associated with a procoagulant state,
measured by factor VIII:C, D-dimer, clot lysis time and endogenous thrombin
potential, or with activated complement pathway, or with increased load of
AGEs. Finally, we will follow-up on patients for three years to assess the
incidence of venous thrombosis. We will then compare microbiome composition
between subjects who suffered a recurrent venous thrombotic event and those who
havent.

cross-sectional study and prospective cohort study

Subjects will visit the UMCG outside of routine care once. Before this visit,
subjects must collect a faeces sample and store this according to instructions.
During the study visit subjects will undergo invasive procedures which may be
painful (measurement of PISA). Formally, measurement of PISA is not a
predefined outcome parameter in this study, but evaluation of periodontal
status is necessary before selecting the appropriate site of sampling for our
analysis of the oral microbiome: the oral microbiome sample will be collected
from the deepest gingivocrevicular pockets in each quadrant.
Additionally, patients will undergo venous sampling once. We will draw
approximately 28 ml of blood (2x 9ml citrate, 1x 10 ml EDTA) for measurements
of coagulation parameters and complement levels