Thrombosis Research 130 (2012) 253–258

Contents lists available at SciVerse ScienceDirect

Thrombosis Research

j ourna l homepage: www.e lsev ie r .com/ locate / thromres

Regular Article

Circulating tissue factor positive microparticles in patients with acute recurrent deepvenous thrombosis

Runyi Ye a,1, Caisheng Ye a,1, Yongbo Huang b, Longshan Liu c, Shenming Wang a,⁎a Department of Vascular Surgery, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, Chinab Zhongshan School of Medicine, Sun Yat-sen University, Guangzhou, Chinac Department of Surgical Laboratory, The First Affiliated Hospital of Sun Yat-sen University, Guangzhou, China

Abbreviations: DVT, Deep venous thrombosis; PTSMPs, microparticles; MPTF, tissue factor positive micropplatelet-free plasma; MoAbs, monoclonal antibodies.⁎ Corresponding author at: Department of Vascular Surg

of Sun Yat-sen University, 58 Zhongshan Road II, GuangzhFax: +86 20 87335856.

E-mail address: shenmingwang@vip.sohu.com (S. W1 These two authors contributed equally to this paper

0049-3848/$ – see front matter © 2011 Elsevier Ltd. Alldoi:10.1016/j.thromres.2011.10.014

a b s t r a c t

a r t i c l e i n f o

Article history:

Received 15 July 2011Received in revised form 24 September 2011Accepted 17 October 2011Available online 17 November 2011

Keywords:Deep venous thrombosisRecurrenceMicroparticlesTissue factor

Introduction: Circulating tissue factor positive microparticles (MPTF) were reported in a wide range of dis-eases with thrombotic tendency. Though D-dimer assay had a high negative predictive value for deep venousthrombosis (DVT) recurrence, there are currently no reliable positive predictors for recurrent DVT. We there-fore quantified MPTF in patients with acute recurrent DVT to determine whether MPTF levels could be usedto predict recurrent DVT.Materials andMethods:Microparticles (MPs) were isolated from plasma of initial DVT patients (n=25), recurrentDVT patients (n=25) and sex- and age-matched healthy individuals (n=25), stained with annexin V, cell-specific monoclonal antibodies (MoAbs) and a MoAb directed against tissue factor (TF), and analyzed by flowcytometry. We also determined the plasma procoagulant activity with a Human TF Chromogenic Activity AssayKit.

Results:We found total MPTF to be elevated in recurrent DVT patients versus normal individuals (P=0.001). Thenumber of monocyte-derived MPTF in both initial and recurrent DVT was higher than in normal individuals(Pb0.01, respectively). The platelet and endothelial cell derived MPTF in recurrent DVT were significantly in-creased relative to other MPTF (Pb0.05), although there was no difference between initial DVT patients and nor-mal individuals.We demonstrated elevated procoagulant activity of platelet-free plasma in DVT patients relativeto normal individuals, and a positive correlation with MPTF.Conclusions: The elevated MPTF could be a potentially predictor for DVT recurrence. Further studies are neededto validate its sensitivity and specificity.

© 2011 Elsevier Ltd. All rights reserved.

Introduction

Deep venous thrombosis (DVT) of the lower limbs is a commondisease with serious complications, including pulmonary thrombo-embolism and post-thrombotic syndrome (PTS). After stopping anti-coagulant treatment, individuals who have had a first episode ofDVT are at an increased risk of new events. In patients who receivedstandard anticoagulation for 3 to 6 months after an incident event,the cumulative rate of recurrence is 21% to 23% at 4 to 5 years, andas high as 40% at 10 years after diagnosis [1,2]. About 5% of recurrentepisodes are fatal, recurrent DVT markedly increases the risk of PTSand recurrent pulmonary embolism [3]. Although oral anticoagulanttreatment is highly effective, its use is hampered by its drawbacks

, post-thrombotic syndrome;articles; TF, tissue factor; PFP,

ery, The First Affiliated Hospitalou, Guangdong 510080, China.

ang)..

rights reserved.

such as frequent laboratory monitoring, dose adjustment and an in-creased risk of major hemorrhage. Consequently, identification ofrisk factors for recurrent DVT is important and could help to decidewhether to stop or to continue anticoagulant therapy. Risk factors as-sociated with increased recurrence were idiopathic or unprovokedDVT, proximal DVT, cancer, male gender, oral contraceptive use, andshorter duration of anticoagulation [4].

The objective test of D-dimer assay after vitamin K antagonist dis-continuation had a high negative predictive value for DVT recurrenceand could be used to guide the optimal duration of anticoagulation forpatients with a first unprovoked DVT [5]. However, there are currentlyno reliable positive predictors for recurrent DVT. The discovery of vari-ous tissue factor positive microparticles (MPTF) and their procoagulantactivity, makes them a potentially predictor alternative.

Giesen and colleagues [6] first demonstrated the existence of func-tionally intact microparticles (MPs) containing tissue factor in theblood of normal individuals. Tissue factor (TF) is an important initiatorof the extrinsic pathway of blood coagulation, and can be detected inthe vesselwall and the blood [7]. Themajority of circulating TF is presentin the form ofMPs, which are small membrane fragments released fromactivated or apoptotic vascular cells [8].MPs are highly heterogeneous in

254 R. Ye et al. / Thrombosis Research 130 (2012) 253–258

size, with dimensions of 0.1–1 μm, and with a composition dependentupon the stimulus and cellular origin [9]. MPTF may originate from leu-kocytes, endothelial cells, platelets or erythrocytes [10], and the propor-tion of different cellular origins may correlate with the underlyingdisease. Cardiovascular disease, diabetes, cancer, sickle cell anemia,and endotoxemia have elevated levels ofMPTF [11]. In cancer, hypercoa-gulability may be due in part to elevated levels of circulating MPTF,which could serve as biomarkers for a pre-thrombotic state [12,13]. Sim-ilarly, we askedwhetherMPTF levels could be used as risk biomarkers topredict recurrent DVT.

Materials and methods

Study subjects

The ethics committee of the First Affiliated Hospital of Sun Yat-senUniversity approved the study, and written informed consent wasobtained from all subjects. A test sample including 25 patients withacute initial DVT (group A) and 25 patients with recurrent DVT(group B) of the lower limbs were compared with 25 healthy individ-uals (group C) over the same time period from January 2009 to Janu-ary 2011. We selected acute DVT patients confirmed by Dopplerultrasound and that had not received any anticoagulant or thrombo-lytic therapy before admission. The criterion for initial DVT was thatvenous thrombosis occurred in the absence of an antecedent majorclinical risk factor such as surgery, trauma, active cancer, immobility,pregnancy or the puerperium of a thrombophilic blood abnormality.Recurrent DVT was defined as another episode of venous thrombosisafter a first episode of unprovoked DVT with standard anticoagulationfor 3 to 6 months and the withdrawal of oral anticoagulants at leastfor 3 months. Healthy individuals were unrelated to patients andwere also not related to each other. They were spouses or acquain-tances of patients, hospital staff, or friends or relatives of hospitalstaff. The control group was matched for sex and age to the patientgroups. We excluded subjects with any of the clinical risk factorsmentioned above or a family history of thrombosis. We recorded de-mographic data and past medical history for each subject, and Dopp-ler ultrasound results and the clinical state for DVT patients. Bloodcount and D-dimer tests were completed before plasma specimencollection. All the DVT patients also have thrombophilia factors testsincluding protein C and protein S.

Reagents and assays

MPs and MPTF were detected by flow cytometry using cell-specificmonoclonal antibodies (MoAbs) [14]. The MoAbs stained for monocytes(CD14), platelets (CD41a), endothelial cells (CD144), and erythrocytes(CD235a). Isotype control antibodies, respectively, were Mouse IgG2a, κ,Mouse IgG1, κ,Mouse IgG1, κ, andMouse IgG2b, κ. All secondary antibodyreagents were labeled with Phycoerythrin (PE) and obtained from BD-Pharmingen (San Diego, CA). Cy5-labeled Annexin V for phosphatidylser-ine (PS) was also obtained from BD-Pharmingen. Fluorescein isothiocya-nate (FITC)–labeled anti-human Tissue Factor MoAb (4507CJ) wasobtained from American Diagnostica Inc (Greenwich, CT), and isotypecontrol mouse immunoglobulin G1 (IgG1), labeled with carboxyfluores-cein(CFS) (11711) was obtained from R&D Systems (Minneapolis, MN).

Fluorescent microbeads of 0.1 μm (S37204) and 1 μm (S37498)were used for gating experiments and 7 μm nonfluorescent beadsfor enumeration of MPs were obtained from Molecular Probes.

The AssaySense Human Tissue Factor Chromogenic Activity AssayKit (CT1002b, AssayPro) was used to determine plasma Tissue Factor(TF) procoagulant activity. This assaymeasures the ability of lipoproteinTF/FVIIa to convert factor X (FX) to factor Xa. The amidolytic activity ofthe TF/FVIIa complex is quantitated by the amount of FXa producedusing a highly specific FXa substrate, which releases a yellow para-

nitroaniline (pNA) chromophore. The change in absorbance of pNA at405 nm is directly proportional to TF enzymatic activity.

Specimen collection and preliminary processing

Blood samples were drawn into buffered citrate using a 21-gaugeneedle (Becton Dickinson) and platelet-free plasma (PFP)was preparedin 30 minutes using a 2-step centrifugation procedure (1500 g for10 minutes at 20 °C to make platelet-rich plasma, followed by 13000 gfor 10 minutes at 20 °C). The PFP was immediately stored at −80 °C.

Preparation of microparticles

MPs were isolated as previously reported [14]. Briefly, wash buffer(10 mM HEPES (N-2-hydroxyethylpiperazine-N’-2-ethanesulfonicacid) , 140 mM NaCl, 4.5 mM KCl, 1% bovine serum albumin [BSA],0.1% Na azide, 2.5 mM CaCl2 , pH 7.4) was filtered using a 0.2 μm filterbefore washing the PFP. Washed PFP was ultracentrifuged at100,000 g for 120 minutes at 20 °C. Finally, MPs were resuspendedin the remainder wash buffer by gentle vortexing and pipetting, andused immediately for flow cytometry.

Flow cytometric detection of microparticles

We used a modification of standardmethods [14] to detect the mi-croparticles with a FACS Calibur flow cytometer, and CellQuest prosoftware (Becton Dickinson). MPs were quantified by spiking aknown quantity of 7 μm beads (23500 beads/25 μl) into each sample.Acquisition was stopped after 5000 beads and counted in R1 or thebeads gate (Fig. 1-A). We defined MPs as the events falling in the R2or MPs gate (based on size, Fig. 1-A) and the annexin V-positivegate (Fig. 1-B). To determine the cellular origin of MPTF, sampleswere triple labeled with Cy5-labeled annexin V, a cell type–specificPE-labeled MoAb, and a FITC-labeled MAb against TF (Fig. 1-C). Positivetriple labeled events were defined as cellular MPTF. Finally, total MPsand MPTF (per milliliter PFP) was calculated as reported [14] with theactual dilution factor.

Procoagulant activity of PFP assay

We detected PFP procoagulant activity on the basis of the Assay-Sense Human TF Chromogenic Activity Assay Kit (CT1002b, Assay-Pro). PFP and all reagents were brought to room temperature, and70 μl freshly prepared Assay Mix added (Assay Diluent 50 μl, FVII10 μl, FX 10 μl) to each well of the 96-well plate. To this was added10 μl of TF standards or samples per well with gentle mixing beforeincubating at 37 °C for 30 minutes. Next, 20 μl of FXa substrate wasadded to each well and absorbance at 405 nm by the MultifunctionMicroplate Reader (BioTek Instruments, USA) was recorded every2 minutes for 12 minutes. Finally, a standard curve was generatedby regression analysis using 4-parameters, from which the unknownsample concentration was determined.

Statistical analysis

Summary statistics were expressed as medians and Inter-QuartileRanges (IQR) since the results were not normally distributed. The me-dian value is the 50th percentile. The 25th and 75th percentiles of thedata specify the values covered by the IQR. The Kruskal-Wallis testwas used for an overall comparison between the groups. Significancedetermined by the Kruskal-Wallis test enabled subsequent intergroupcomparisons using the Mann–Whitney U test; Pb0.05 (2-tailed) wasconsidered significant. Bivariate correlations were determined usingSpearman's Rho test. Statistical analysis was performed using SPSS ver-sion 16.0 for Windows (Chicago, IL, USA).

Fig. 1. Flow cytometric analysis and quantification of MPs. (A) Bead size is reflected by forward scatter (FSC) and side scatter (SSC). The R1 or bead gate included 7-μm beads forenumerating MPs. The R2 or MP gate included 1.0 μm beads shown in the upper right corner, and contains all microparticles of 1.0 μm or less. (B) The R3 gate shows phosphati-dylserine (PS)–positive MPs in PFP by Annexin V–Cy5 labeling on the y-axis, in relation to SSC on the x-axis. (C) The cellular origin of MPTF are shown in the upper right quadrant.Shown is representative triple labeled sample for annexin V (not shown), TF-FITC on the x-axis, and CD144- PE (endothelial cells) on the y-axis.

255R. Ye et al. / Thrombosis Research 130 (2012) 253–258

Results

Basic characteristics of the subjects

Demographic, thrombosis clinical data, and medical test resultsare shown in Table 1. There were no significant statistical differenceswith respect to sex, average age, body mass index (BMI) and bloodcount, and no differences in the locations of the thrombosis betweenthe initial DVT and recurrent DVT group. In the recurrent DVT group,the first thrombosis occurred in the ipsilateral limbs in 13 of 25 cases.D-dimer was elevated in DVT groups relative to controls. The heredi-tary thrombophilia factors including protein C and protein S were allin the normal range.

Total MPs were elevated among recurrent DVT patients

Enumeration revealed normal individuals (group C) to have totalMPs of 223.6 (182.3-299.4)×103/ml PFP [Median (IRQ)]. In compari-son, total MPs were elevated for patients with DVT in group A [256.2(217.5-330.1)×103/ml] and in group B [340.1 (205.7-450.1)×103/ml]. Overall, the group showed a significant difference (P=0.025,Table 2), and pairwise comparisons indicated recurrent DVT patients(group B) had significantly elevated total MPs (P=0.012, versus nor-mal individuals, Fig. 2-A). However, there was no difference in group

Table 1Demographic and clinic data of the study subjects.

GROUP A GROUP B GROUP C P*

(N=25) (N=25) (N=25)

Age, Mean±SD 47.9±13.1 48.1±12.3 47.6±9.1 N.S.Gender, male/female 13/12 13/12 13/12 N.S.Smoking status, n(%) 3(12) 4(16) 3(12) N.S.BMI, kg/m2,mean±SD 25.1±1.8 24.4±1.4 24.4±1.6 N.S.LocalizationDistal DVT 6 5 -Proximal DVT 19 20 -Blood countErythrocyte(×1012/L) 4.36±0.38 4.03±0.69 4.56±1.95 N.S.Platelet(×109/L) 209±55 245±112 218±71 N.S.Leukocyte(×109/L) 8.89±2.86 10.2±4.5 9.56±3.4 N.S.Monocytes(×109/L) 0.63±0.25 0.54±0.20 0.53±0.15 N.S.Eosinophil(×109/L) 0.21±0.13 0.25±0.15 0.17±0.09 N.S.Basophil(×109/L) 0.046±0.03 0.073±0.05 0.063±0.05 N.S.FIB, g/L, mean±SD 2.85±1.00 3.15±1.4 3.59±1.8 N.S.D dimer(g/L) 1365±1160 1809±1789 140±65 b0.001Protein C deficiency 0 0 -Protein S deficiency 0 0 -

* N.S., not statistically significant (pb0.05).BMI , body mass index. FIB, Fibrinogen. SD, standard deviation.

A vs. group B or group A vs. group C (P=0.054 and P=0.327respectively).

Number and cell origin of MPs

The levels of MPs from different cells are shown in Table 2. Wefound no difference in erythrocyte origin (P=0.197) or in platlet or-igin (P=0.062), but a significant difference in monocyte cell origin(P=0.003, group B vs. group C, Fig. 3-A) and an higher level of endo-thelial cell MPs in recurrent DVT patients than initial DVT patientsand normal individuals (Fig. 3-A).

Total MPTF were elevated among recurrent DVT patients

Normal individuals (group C) had total MPTF of 36.2 (19.7-47.3)×103/ml PFP (Median (IRQ)). Total MPTF were elevated for pa-tients with DVT in group A [36.1 (31.1-73.1)×103/ml] and group B[97.7(34.2-116.5)×103/ml]. The overall group showed a significantdifference (P=0.001, Table 2), and pairwise comparisons showed re-current DVT patients (group B) had a significantly elevated totalMPTF (P=0.008, and P=0.001, versus group A and group C respec-tively, Fig. 2-B). However, there was no difference in group A vs.group C (P=0.114).

Number and cell origin of MPTF

The levels of MPTF from different cells from four different sourcesare shown (Table 2, Fig. 3-B). Monocyte derived MPTF were signifi-cantly elevated in recurrent DVT and initial DVT patients versushealthy controls, and a higher level for group B than group A. Plateletand endothelial cell derived MPTF from group B were significantly in-creased versus group A and group C, with no difference betweengroup A and group C. Erythrocyte derived MPTF were the same levelsin three groups.

Procoagulant activity of PFP in normal individuals and DVT patients

PFP procoagulant activity was estimated by measuring TF sampleconcentration. The median procoagulant activity of normal individ-uals was 45.95 pM (40.05-82.35 pM), but was significantly elevatedin group A [93.4 pM (82.25-99.98 pM)], and group B [99.35 pM(77.73-113.9 pM)]. There was no difference between group A andgroup B. (Fig. 4).

Table 2Flow cytometry results for MPs and MPTF levels among the three study groups.

Cell Origin Group A Group B Group C Z-value* P*

Median(IQR) Median(IQR) Median(IQR)

Monocyte MPs(×103/ml) 68.7(49.0-113.5) 108.2(58.0-136.7) 60.6(44.1-78.5) 9.476 0.009Platelet MPs(×103/ml) 60.2(48.3-74.8) 85.9(52.7-96.8) 54.3(34.7-77.2) 5.569 0.062EC MPs(×103/ml) 53.5(39.9-80.3) 81.6(54.5-109.1) 53.6(39.6-73.5) 9.027 0.011RBC MPs(×103/ml) 63.0(47.6-88.5) 79.6(54.8-104.1) 63.5(38.5-80.6) 3.25 0.197Total-MPs(×103/ml) 256.2(217.5-330.1) 340.1(205.7-450.1) 223.6(182.3-299.4) 7.347 0.025Monocyte MPTF(×103/ml) 13.2(9.61-31.8) 30.2(13.6-45.0) 8.46(5.25-10.8) 24.60 b0.001Platelet MPTF(×103/ml) 7.03(6.18-12.4) 16.2(6.52-28.5) 6.37(4.58-9.23) 15.19 0.004EC-MPTF(×103/ml) 7.49(6.17-11.2) 20.2(7.74-27.9) 8.44(4.76-10.7) 10.163 0.006RBC-MPTF(×103/ml) 9.22(7.08-13.6) 11.7(6.61-20.6) 9.36(6.11-14.4) 2.57 0.286Total-MPTF(×103/ml) 36.1(31.1-73.1) 97.7(34.2-116.5) 36.2(19.7-47.3) 13.165 0.001

Results are shown as medians and inter-quartile ranges (IQR). EC: Endothelial cells, RBC: Red Blood cells. * Kruskal-Wallis test.

150A a

256 R. Ye et al. / Thrombosis Research 130 (2012) 253–258

Procoagulant activity of PFP and MPs

The bivariate correlations between the procoagulant activity ofPFP and MPs are shown in Table 3. The data indicated that MPTFwas positively correlated with PFP procoagulant activity.

Discussion

Recent evidence indicates that MPs formation is a highly organizedprocess, leading to the shedding of distinct domains of the cellmembrane [15]. MPs behave as vectors of bioactive molecules ableto disseminate biological information in the vascular compartment.MPs contribute to hemostatic and inflammatory responses, vascularremodeling and angiogenesis, cancer, and apoptosis, well-knownprocesses involved in atherothrombosis [16–18]. Due to the expres-sion of both membrane phosphatidylserine, a procoagulant phos-pholipid necessary for the assembly of the blood clotting enzyme

GROUP A GROUP B GROUP C0

100

200

300

400

500A

*

MP

s (¡

Áth

ou

san

ds/

ml)

GROUP A GROUP B GROUP C0

25

50

75

100

125B

**

MP

TF

(¡Á

tho

usa

nd

s/m

l)

Fig. 2. Total blood MPs and MPTF. Data are expressed as number per milliliter ofplatelet-free plasma and shown as medians and IQR. (A) Total MP. (B) Total MPTF.Error bars represent IQR. *P=0.001, **P=0.001 compared with group C.

complex, and functional tissue factor, the major initiator of the co-agulation cascade, MPs are catalytic procoagulant surfaces involvedin thrombogenesis [8,19]. Decreased numbers of MPs are associatedwith bleeding syndromes, such as Scott Syndrome [20] and Casta-man Syndrome [21]. However, increased numbers of circulating MPshave been reported in awide range of diseaseswith thrombotic tenden-cy, including acute myocardial infarction, antiphospholipid syndrome,preeclampsia, rheumatoid arthritis, thrombotic thrombocytopenic pur-pura, vasculitis, heparin-induced thrombocytopenia (HIT), and parox-ysmal nocturnal hemoglobinuria (PNH) [22,23].

In our study, both the total number of MPs and the number of TFpositive MPs are increased in the acute recurrent DVT patients rela-tive to healthy individuals, but no difference between initial DVTand healthy individuals. Cihan Ay et al. had found no difference of

Monocyte Platelets Endothelial RBC0

50

100

£

£a

MP

s (¡

Áth

ou

san

ds/

ml)

Monocyte Platelets Endothelial RBC0

10

20

30

40

50B

GROUP A GROUP B GROUP C

£a£a

£a£a

££¦¤

MP

TF

(¡Á

tho

usa

nd

s/m

l)

Fig. 3. Cell origin of MPs and MPTF. Data are expressed as number per milliliter of PFPand shown as medians IQR. (A) Cell origin of MPs. (B) Cell origin of MPTF. Error barsrepresent IQR. *Pb0.05, compared with group C in Monocytes and Endothelial cellMPs respectively; **Pb0.01, compared with group C in Monocytes; ΔPb0.05, comparedwith other groups in Platelets; #P=0.006 compared with other groups in Endothelial.

**

TF

Act

ivit

y (p

M)

GROUP A GROUP B GROUP C

0

25

50

75

100

125

150

Fig. 4. A box-whisker pot that shows the TF activity (pM) of the PFP from DVT patients(group A: N=25, group B: N=25) and normal individuals (group C: N=25). The linein the box represents the median. The top and the bottom of the box represent the 75thand 25th percentile, respectively. *Pb0.01 compared with group C.

257R. Ye et al. / Thrombosis Research 130 (2012) 253–258

circulating procoagulant MPs between patients in the chronic stage ofrecurrent DVT and healthy controls [24]. This suggests that the MPsmay play a role in the hypercoagulable state of acute recurrent DVTbut not for chronic DVT.

Circulating MPs are derived from blood cells and the vascular wall,and the cellular origin of elevated MPs is variable based on the dis-ease. Atherosclerotic plaques contain highly MPTF of mainly mono-cytic and lymphocytic origin [25]. Lung cancer patients had a higherconcentration of monocyte and platelet derived MPs than normalsubjects [26], and sickle blood contains elevated numbers of TF posi-tive MPs derived from endothelial cells and monocytes [14]. In ourstudy, both MPs and MPTF derived from endothelial cells in recurrentDVT patients were elevated relative to other group, but no differencebetween normal individuals and initial DVT patients, suggesting acti-vation or damage of vascular endothelial cells after the first episode ofvenous thromboembolism. However, more important is the observa-tion that both initial and recurrent DVT patients had a higher level ofmonocyte derived MPTF than normal individuals. In addition, plateletderived MPTF from recurrent DVT patients was higher than initialDVT patients. Taken together, this suggests that monocyte derivedMPTF is the main difference between DVT and non-DVT patients,and that platelet and endothelial cell derived MPTF are the main dif-ference between recurrent and inital DVT patients. Some studies haveshown that both monocyte and platelet contained TF mRNA [11], andthat cancer patients exhibit an increased number of monocyte andplatelet derived MPTF [12,27,28]. Increased monocyte TF activity isassociated with higher risk of acute coronary syndrome [29] and dis-seminated intravascular coagulation (DIC) in endotoxemia and sepsis

Table 3Spearman's Rank Correlation test of MPs and TF activity of PFP.

Cases Correlation coefficient P-value*

Monocyte-MPs 75 0.393 b0.001Platelet-MPs 75 0.402 b0.001Endothelial-MPs 75 0.259 0.025Erythrocyte-MPs 75 0.354 0.002Total MPs 75 0.427 b0.001Monocyte-MPTF 75 0.457 b0.001Platelet-MPTF 75 0.396 b0.001Endothelial-MPTF 75 0.339 0.003Erythrocyte-MPTF 75 0.28 0.015Total MPTF 75 0.41 b0.001

*Pb0.05 was considered to be statistically significant.

[30]. Cardiac myocyte-specific overexpression of TF can restore he-mostasis in the hearts of low TF mice [11]. Thus, monocyte derivedMPTF may play an important role in thrombosis and could be a pre-dictor for recurrent DVT. Although there is limited prospectivestudy data demonstrating that MPTF levels can be considered a truemarker of thrombotic risk, an increase in monocyte associated TF ac-tivity after total knee arthroplasty suggests monocyte MPTF is a po-tential risk factor for DVT [31].

Some animal models experiments indicated that circulating MPTFwas an important mediator of thrombotic risk in disease states[32–34]. Other models explained that MPTF was recruited to a throm-bus and enhance its growth through the interaction of P-selectin gly-coprotein ligand 1 (PSGL-1) on the MPs with P-selectin expressed onthe surface of platelets [35,36]. Cancer patients with acute DVT havehigher levels of MPTF activity than normal individuals, cancer pa-tients without DVT, and subjects with idiopathic DVT [27]. Our datashow elevated TF activity of PFP in DVT patients and a positive corre-lation with MPTF. So the increased levels of MPTF may indicate an un-derlying prothrombotic state and levels of MPTF may be a usefuldeterminant of thrombotic episodes.

Something of our study need to be addressed. We performed ourspecimen collection from unprovoked DVT patients and healthy indi-viduals without medications to avoid any influence of trigger factorssuch as malignancy, surgery and pregnancy on circulating MPs, asthese condition are associated with the presumed mechanisms forMPs generation. We used a MPTF assay from several ways [37]. Themethodological variables that need to be considered and the com-plexity of our detection methods may limit use of MPTF as a predictorfor recurrent DVT in the clinic.

In summary, we have demonstrated that MPTF levels are signifi-cantly elevated in recurrent DVT patients from monocyte, plateletand endothelial cell. We believe that the levels of MPTF could helpto predict the risk of DVT recurrence, and combining the results ofD-dimer we could decide whether or when anticoagulant therapycan be safely withdrawn. Further studies are needed to validatetheir sensitivity and specificity.

Conflict of interest statement

None.

Acknowledgments

None.

References

[1] Hansson PO, Sorbo J, Eriksson H. Recurrent venous thromboembolism after deepvein thrombosis: incidence and risk factors. Arch Intern Med 2000;160(6):769–74.

[2] Prandoni P, Noventa F, Ghirarduzzi A, Pengo V, Bernardi E, Pesavento R, et al. Therisk of recurrent venous thromboembolism after discontinuing anticoagulation inpatients with acute proximal deep vein thrombosis or pulmonary embolism. Aprospective cohort study in 1,626 patients. Haematologica 2007;92(2):199–205.

[3] McRae S, Tran H, Schulman S, Ginsberg J, Kearon C. Effect of patient's sex on risk ofrecurrent venous thromboembolism: a meta-analysis. Lancet 2006;368(9533):371–8.

[4] Liem TK, Deloughery TG. First episode and recurrent venous thromboembolism:who is identifiably at risk? Semin Vasc Surg 2008;21(3):132–8.

[5] Palareti G, Legnani C, Cosmi B, Guazzaloca G, Pancani C, Coccheri S. Risk of venousthromboembolism recurrence: high negative predictive value of D-dimer per-formed after oral anticoagulation is stopped. Thromb Haemost 2002;87(1):7–12.

[6] Giesen PL, Rauch U, Bohrmann B, Kling D, Roque M, Fallon JT, et al. Blood-bornetissue factor: another view of thrombosis. Proc Natl Acad Sci U S A 1999;96(5):2311–5.

[7] Gross PL, Vaezzadeh N. Tissue factor microparticles and haemophilia. Thromb Res2010;125(Suppl. 1):S67–9.

[8] Morel O, Toti F, Hugel B, Bakouboula B, Camoin-Jau L, Dignat-George F, et al. Pro-coagulant microparticles: disrupting the vascular homeostasis equation? Arter-ioscler Thromb Vasc Biol 2006;26(12):2594–604.

[9] Thery C, Ostrowski M, Segura E. Membrane vesicles as conveyors of immune re-sponses. Nat Rev Immunol 2009;9(8):581–93.

258 R. Ye et al. / Thrombosis Research 130 (2012) 253–258

[10] Osterud B. Tissue factor expression in blood cells. Thromb Res 2010;125(Suppl. 1):S31–4.

[11] Mackman N, Tilley RE, Key NS. Role of the extrinsic pathway of blood coagulationin hemostasis and thrombosis. Arterioscler Thromb Vasc Biol 2007;27(8):1687–93.

[12] Tilley RE, Holscher T, Belani R, Nieva J, Mackman N. Tissue factor activity is in-creased in a combined platelet and microparticle sample from cancer patients.Thromb Res 2008;122(5):604–9.

[13] Vieira LM, Dusse LM, Fernandes AP, Martins-Filho OA, de Bastos M, Ferreira MF,et al. Monocytes and plasma tissue factor levels in normal individuals and pa-tients with deep venous thrombosis of the lower limbs: potential diagnostictools? Thromb Res 2007;119(2):157–65.

[14] Shet AS, Aras O, Gupta K, Hass MJ, Rausch DJ, Saba N, et al. Sickle blood containstissue factor-positive microparticles derived from endothelial cells and mono-cytes. Blood 2003;102(7):2678–83.

[15] Lechner D, Weltermann A. Circulating tissue factor-exposing microparticles;2008. S47-54 pp.

[16] Hugel B, Martinez MC, Kunzelmann C, Freyssinet JM. Membrane microparticles:two sides of the coin. Physiology (Bethesda) 2005;20:22–7.

[17] Morel O, Toti F, Hugel B, Freyssinet JM. Cellular microparticles: a disseminated stor-age pool of bioactive vascular effectors. Curr Opin Hematol 2004;11(3):156–64.

[18] VanWijk MJ, VanBavel E, Sturk A, Nieuwland R. Microparticles in cardiovasculardiseases. Cardiovasc Res 2003;59(2):277–87.

[19] Furie B, Furie BC. Thrombus formation in vivo. J Clin Invest 2005;115(12):3355–62.[20] Toti F, Satta N, Fressinaud E, Meyer D, Freyssinet JM. Scott syndrome, characterized

by impaired transmembrane migration of procoagulant phosphatidylserine andhemorrhagic complications, is an inherited disorder. Blood 1996;87(4):1409–15.

[21] Castaman G, Yu-Feng L, Battistin E, Rodeghiero F. Characterization of a novelbleeding disorder with isolated prolonged bleeding time and deficiency of plate-let microvesicle generation. Br J Haematol 1997;96(3):458–63.

[22] Nomura S, Ozaki Y, Ikeda Y. Function and role of microparticles in various clinicalsettings. Thromb Res 2008;123(1):8–23.

[23] Piccin A, Murphy WG, Smith OP. Circulating microparticles: pathophysiology andclinical implications. Blood Rev 2007;21(3):157–71.

[24] Ay C, Freyssinet JM, Sailer T, Vormittag R, Pabinger I. Circulating procoagulant mi-croparticles in patients with venous thromboembolism. Thromb Res 2009;123(5):724–6.

[25] Mallat Z, Hugel B, Ohan J, Leseche G, Freyssinet JM, Tedgui A. Shed membrane mi-croparticles with procoagulant potential in human atherosclerotic plaques: a rolefor apoptosis in plaque thrombogenicity. Circulation 1999;99(3):348–53.

[26] Kanazawa S, Nomura S, Kuwana M, Muramatsu M, Yamaguchi K, Fukuhara S.Monocyte-derived microparticles may be a sign of vascular complication in pa-tients with lung cancer. Lung Cancer 2003;39(2):145–9.

[27] Tesselaar ME, Romijn FP, Van Der Linden IK, Prins FA, Bertina RM, Osanto S.Microparticle-associated tissue factor activity: a link between cancer and throm-bosis? J Thromb Haemost 2007;5(3):520–7.

[28] Hron G, Kollars M, Weber H, Sagaster V, Quehenberger P, Eichinger S, et al. Tissuefactor-positive microparticles: cellular origin and association with coagulation ac-tivation in patients with colorectal cancer. Thromb Haemost 2007;97(1):119–23.

[29] Egorina EM, Sovershaev MA, Bjorkoy G, Gruber FX, Olsen JO, Parhami-Seren B,et al. Intracellular and surface distribution of monocyte tissue factor: applica-tion to intersubject variability. Arterioscler Thromb Vasc Biol 2005;25(7):1493–8.

[30] Mackman N. The many faces of tissue factor. J Thromb Haemost 2009;7(Suppl. 1):136–9.

[31] Johnson GJ, Leis LA, Bach RR. Tissue factor activity of blood mononuclear cellsis increased after total knee arthroplasty. Thromb Haemost 2009;102(4):728–34.

[32] Li YD, Ye BQ, Zheng SX, Wang JT, Wang JG, Chen M, et al. NF-kappaB transcriptionfactor p50 critically regulates tissue factor in deep vein thrombosis. J Biol Chem2009;284(7):4473–83.

[33] Day SM, Reeve JL, Pedersen B, Farris DM, Myers DD, Im M, et al. Macrovascularthrombosis is driven by tissue factor derived primarily from the blood vesselwall. Blood 2005;105(1):192–8.

[34] Himber J, Wohlgensinger C, Roux S, Damico LA, Fallon JT, Kirchhofer D, et al. Inhi-bition of tissue factor limits the growth of venous thrombus in the rabbit. JThromb Haemost 2003;1(5):889–95.

[35] Polgar J, Matuskova J, Wagner DD. The P-selectin, tissue factor, coagulation triad. JThromb Haemost 2005;3(8):1590–6.

[36] Del CI, Shrimpton CN, Thiagarajan P, Lopez JA. Tissue-factor-bearing microvesiclesarise from lipid rafts and fuse with activated platelets to initiate coagulation.Blood 2005;106(5):1604–11.

[37] Key NS. Analysis of tissue factor positive microparticles. Thromb Res 2010;125(Suppl. 1):S42–5.