fdg-pet/ct enables the detection of recurrent same-site deep

DOI: 10.1161/CIRCULATIONAHA.114.008902

1

FDG-PET/CT Enables the Detection of Recurrent Same-Site Deep Vein

Thrombosis by Illuminating Recently Formed, Neutrophil-Rich Thrombus

Running title: Hara et al.; FDG-PET detection of recurrent DVT

Tetsuya Hara, MD, PhD1; Jessica Truelove, BS2; Ahmed Tawakol, MD3;

Gregory R. Wojtkiewicz, MS2; William J. Hucker, MD, PhD3; Megan H. MacNabb, BA3;

Anna-Liisa Brownell, PhD4; Kimmo Jokivarsi, PhD4; Chase W. Kessinger, PhD1;

Michael R. Jaff, DO3; Peter K. Henke, MD5; Ralph Weissleder, MD, PhD2;

Farouc A. Jaffer, MD, PhD1,3

1Cardiovascular Research Center; 2Center for Systems Biology; 3Cardiology Division; 4Martinos

Biomedical Imaging Center, Massachusetts General Hospital, Harvard Medical School, Boston,

MA; 5Section of Vascular Surgery, Dept of Surgery, University of Michigan, Ann Arbor, MI

Address for Correspondence:

Farouc Jaffer, MD, PhD

Cardiovascular Research Center

Massachusetts General Hospital

Simches Research Building, Room 3206

Boston, MA 02114

Tel: 617-724-9353

Fax: 617-860-3180

E-mail: [email protected]

Journal Subject Codes: Diagnostic testing:[32] Nuclear cardiology and PET, Thrombosis:[173] Deep vein thrombosis, Basic science research:[130] Animal models of human disease

Anna-Liisa Brownell, PhD ; Kimmo Jokivarsi, PhD ; Chase W. Kessingerr,, PhPhP DD ;;

Michael R. Jaff, DO3; Peter K. Henke, MD5; Ralph Weissleder, MD, , PhPhPhDDD22;;

Farouc A. Jaffer, MD, PhD1,3

1CCaCarrdrdioiovaascscuuularr RReseseaearcrch h CeC ntnterer;;; 2CeCentnterer fforor SSysyssteeemss BBioiolologyy;;; 333CCCardrdioiolol gygy DDivivisisioion;n; 4MaM rtrtininos

BBioomomedical IImaaagiiingg CCeeenteterrr, MMMasassasasachchhuussettsss GGGeneeerraal HoHoHospspspiiitaaal, HHaHarrvardd d MeMeMedididicacal l SScSchhohooool,, BoBoststoon,

MAMAMA;;; 5SeSectctctioiion n ofoff VVaaasccuculal r r SuSuSurgrgrgerereryy,y, DDDepepeptt t oofof SSururrgeeeryryy,, UnUnUniivivererersisittyty ooof f f MMMichchigigiganann,, AnAnAnnnn AAArbbobor,r, MMMII

AdAddrdresesss foforr CoCorrrresespopondndenencece::

by guest on March 25, 2018

http://circ.ahajournals.org/D

ownloaded from

by guest on M

arch 25, 2018http://circ.ahajournals.org/

Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from



ownloaded from

by guest on M


Dow

nloaded from

http://circ.ahajournals.org/











2

Abstract

Background—Accurate detection of recurrent same-site deep vein thrombosis (DVT) is a

challenging clinical problem. As DVT formation and resolution are associated with a

preponderance of inflammatory cells, we investigated whether noninvasive 18F-

fluorodeoxyglucose (FDG)-PET imaging could identify inflamed, recently formed thrombi and

thereby improve the diagnosis of recurrent DVT.

Methods and Results—We established a stasis-induced DVT model in murine jugular veins and

also a novel model of recurrent stasis DVT in mice. C57BL/6 mice (n=35) underwent ligation of

the jugular vein to induce stasis DVT. FDG-PET/CT was performed at DVT timepoints of day 2,

4, 7, 14, or 2+16 (same-site recurrent DVT at day 2 overlying a primary DVT at day 16).

Antibody-based neutrophil depletion was performed in a subset of mice prior to DVT formation

and FDG-PET/CT. In a clinical study, 38 patients with lower extremity DVT or controls

undergoing FDG-PET were analyzed. Stasis DVT demonstrated that the highest FDG signal

occurred at day 2, followed by a time-dependent decrease (p<0.05). Histological analyses

demonstrated that thrombus neutrophils (p<0.01), but not macrophages, correlated with

thrombus PET signal intensity. Neutrophil depletion decreased FDG signals in day 2 DVT

compared to controls (p=0.03). Recurrent DVT demonstrated significantly higher FDG uptake

than organized day 14 DVT (p=0.03). The FDG DVT signal in patients also exhibited a time-

dependent decrease (p<0.01).

Conclusions—Noninvasive FDG-PET/CT identifies neutrophil-dependent thrombus

inflammation in murine DVT, and demonstrates a time-dependent signal decrease in both murine

and clinical DVT. FDG-PET/CT may offer a molecular imaging strategy to accurately diagnose

recurrent DVT.

Key words: deep vein thrombosis, fluorodeoxyglucose, inflammation, recurrent event, positron emission tomography, recurrent DVT, inflammation, neutrophils, molecular imaging

Antibody-based neutrophil depletion was performed in a subset of mice prior too DDDVTVTVT fffororormamamatitit oono

and FDG-PET/CT. In a clinical study, 38 patients with lower extremity DVT or r cocoontnntroroolslsls

undergoing FDG-PET were analyzed. Stasis DVT demonstrated that the highest FDG signal

occcucucurrrrrededed aaatt t dadad y 2,2,2 ffolo lowed by a time-dependentntt ddeecrease (p<0.005)5)5 . HiHiistststoological analyses

dededemmomonstratededd tthahahat ththrororombmbmbususs nnneueueutrtrtropopphihihillsls ((ppp<0<0<0.0.0. 11),,, bbbut nononott t mamamacrcrropopo hahahagegegess, ccorrrererelalalateteted dd wiwiwiththth

hhhrooombm us PETET siigignnnal iintttensnsiityyy. NNeuuutrtrt oopophhih l dddepppletiioionn dedeecrcrcreaeaaseeed FDFDFDG siiignnnalals inini dayyy 222 DDDVVVT

cocompmpparararededed tttoo cocoontntntrororolslsls (((p=p=p 00.0 0333)).). RRReececururrereentntnt DDDVTVTVT dddemememonononstststrararateetedd d sisigngngnififificicicananantltltlyy y hihihighghghererer FFFDGDGDG uuuptptptakakakeee

han organizzededd dddayayay 111444 DVDVD T T T (p(p( =0=00.0.0. 3)3)3 . ThThThee e FDFDFDG G G DVDVDVTTT ssigigignananal l l ininn pppatatieieientntntsss alalalsososo eexhxhxhibibibitititedede a time-



ownloaded from



3

Introduction

Deep vein thrombosis (DVT) and subsequent pulmonary embolism is a major cause of

cardiovascular death.1, 2 The incidence of DVT in industrialized countries is 1–3 individuals per

1,000 per year.1, 3 Furthermore, the risk of recurrent DVT is up to 10% per year in patients with

idiopathic or unprovoked DVT.4 In addition, after DVT treatment, patients may experience

recurrent leg pain due to a recurrent DVT, the post-thrombotic syndrome, or other etiologies.

Failure to diagnose a recurrent DVT places the patient at risk of fatal pulmonary embolism.5

Therefore the accurate diagnosis of a recurrent DVT is critical to preserve health, as well as to

justify prolonged, potentially life-long, anticoagulation and its attendant bleeding risks.

Unfortunately, the diagnosis of a recurrent DVT often poses a significant diagnostic challenge

for anatomical imaging methods such as duplex ultrasonography, computed tomography (CT)

venography, or magnetic resonance venography, as a prior residual thrombus confounds the

diagnosis of acute superimposed on remote thrombus.5, 6 Thus accurate approaches to identify

recurrent DVT are urgently needed.

DVT formation and resolution are time-dependent inflammatory processes that involve

neutrophils and macrophages.7, 8 Since 18-fluorine-fluorodeoxyglucose (FDG) uptake is

upregulated in lesions containing pro-inflammatory myeloid cells such as neutrophils9-11 and

macrophages,12, 13 we hypothesized that FDG-positron emission tomography (PET)-CT could

allow noninvasive imaging of DVT-induced inflammation, and allow the identification of

recurrent DVT even in sites with pre-existing older thrombi. In this study, we systematically

investigated the imaging profile of FDG-PET in murine DVT, and then explored cellular

mechanisms of FDG uptake in DVT, focusing on thrombus neutrophils and macrophages. Then

we developed and validated a novel animal model of recurrent stasis DVT to investigate whether

Unfortunately, the diagnosis of a recurrent DVT often poses a significant diagnososstiiccc chchchalalallelelengngngeet

for anatomical imaging methods such as duplex ultrasonography, computed tomography (CT)

veenononogrgrgraapaphyhyhy, , oor mmmaagagnetic resonance venographyyy,, asas a prior residuauaal thhroroommmbus confounds the

ddidiagggnon sis of aacucuttet supuppererrimimimpopopossesed d ononon rreemmoote ththrrommbmbuus.55,5, 666 TThThuusus aaaccccururatatte apappprprooaoachchhesese ttoo idddeenentitit fyfyy

eecucucurrrrrenene tt DVDVDVTT aarree ururgggenntntlyly nnneeeedededed.d.d.

DVTT fffororormamamatititiononn aanddd rrresee olololututu ioioon n arara e e e titiimememe-d-ddepepepenenendeeentntnt iiinfnfn lalalammmmmmatata ororory yy prprprococo esese seseses s thththata involve



ownloaded from



4

FDG-PET could specifically identify recurrent DVT in the same site as the primary DVT.

Additionally, in a blinded clinical study, we investigated the temporal profile of FDG uptake in

DVT patients undergoing FDG-PET.

Methods

Mouse model of stasis-induced DVT in jugular vein

Animal studies were approved by the MGH Subcommittee on Research Animal Care. Male

C57BL/6 mice (14-20 weeks) were anesthetized using a mixture of intraperitoneal

ketamine/xylazine. The right main jugular vein and a reliably present large side branch were

surgically ligated with 6-0 nylon sutures to induce stasis DVT, analogous to murine inferior vena

cava (IVC) stasis DVT models.14 As a sham control, the left jugular vein was surgically exposed

and tied loosely without constriction (Fig. 1A). At various timepoints after ligation from day 2

up to day 16, PET/CT imaging was performed, followed by sacrifice and ex vivo analyses (n=4-

6 at each timepoint). A subset of mice (n=3) underwent three-timepoint serial PET/CT imaging

and were sacrificed after the final imaging timepoint (Supplemental Fig. S1). For the recurrent

DVT model (n=6), we removed the suture of the ligation at day 2 to allow recanalization of

thrombosed jugular vein, followed by re-ligation 12 days later to induce recurrent same-site

stasis-induced DVT (Supplemental Fig. S1). We also performed complete ligation of the

contralateral jugular vein at day 2 to enhance recanalization of the first DVT, and to compare a

fresh recurrent DVT and an older DVT in the same mouse.

PET/CT imaging

Following an overnight fast, mice underwent tail vein injection of 15-25 Ci/g of 18F-FDG

(PETNET Solutions, Woburn, MA). One hour later, microPET/CT imaging was performed

urgically ligated with 6-0 nylon sutures to induce stasis DVT, analogous to muririineee infnfnferrerioioior r r vevena

cava (IVC) stasis DVT models.14 As a sham control, the left jugular vein was surgically exposed

annd d d tititiededed lllooooooseseely wwwititi hoh ut constriction (Fig. 1A). AAAt vvarious timepoooini tss afafaftteter ligation from day 2

uup tttoo day 16, PEPEET///CTCTT imamamaggigingngng wwasasa ppeerrffoformmmeddd, fooolllowowededed bbyyy ssasacrcrrifificice e aannddd exexx vvvivivo o anannalallyysyseeses (((n=n=n=4-

6 atatat eeeacacachh titiimemem popooinint)t)t .. AAA ssusubsbsetete ooofff mimimicece ((n=n=n=3)3)3) uuunnndedeerwwwennnt t thththrreree-e--titit mmemepopopoininnt sseseririialala PPPETETT//C/CTT T immmagagginng g g

and were saccririrififificeceed d d afafafteteter r thhhee e fiff nananal l l imimmagagginining g g tititimememepopopoinini tt t ((SuSuupppppplelelememementntalala FFFigigig. . S1S1S ).).. FFFororo tttheheh recurrent



ownloaded from



5

using a small animal scanner (Inveon, Siemens Medical Solutions, Inc., Malvern, PA). CT

imaging preceded PET imaging, with acquisition of 360 cone beam projections using a source

power and current of 80 keV and 500 A, respectively. Mice were injected with an iodinated

contrast agent (Isovue-370) at 20 L/min during the CT scan. Projections were reconstructed

into 3-dimensional volumes containing 512 x 512 x 768 voxels with the dimension of 0.11 x 0.11

x 0.11 mm (Amira, San Diego, CA). The PET image voxel size was 0.797 x 0.861 x 0.861 mm.

Data were calculated as mean and max standardized uptake values (SUVs) and TBR (target-to-

background ratio; DVT/sham).

Induction of neutropenia

A subset of mice (n=4) was treated with anti-neutrophil antibody (Accurate Chemical, NY) to

induce neutropenia.15, 16 The antibody was initially injected (500 g/mouse, i.p.) one day prior to

surgery. Additional doses of the antibody were i.v. administered (250 g/mouse) on

postoperative days 1 and 2 (Supplemental Fig. S1C).

Ex vivo gamma counting and histopathology

See Supplement for full details. After sacrifice, mice were perfused with 0.9% saline via the left

ventricle. Radioactivity of excised jugular veins was measured by a gamma radiation counter

(Wizard, PerkinElmer). In a subset of resected DVT (day 2 timepoint, n=5), the vein wall and

thrombus were gently separated followed by gamma radioactivity measurements. Next, jugular

veins were fixed overnight with paraformaldehyde (PFA) and embedded in optimal cutting

temperature compound (Sakura Finetek, Torrance, CA). Serial 6-�m cryostat sections were

obtained for H&E, Carstairs’ fibrin staining, Masson trichrome, and immunohistochemistry. The

number of thrombus neutrophils and macrophages per 5 high power fields (HPF; 1000x) was

quantified as previously reported.17

A subset of mice (n=4) was treated with anti-neutrophil antibody (Accurate Cheemmmicacc ll,, NNNY)Y)Y) ttto o

nduce neutropenia.15, 16 The antibody was initially injected (500 g/mouse, i.p.) one day prior to

uurgrgrgeereryy. AAAddddd iitioonanal l doses of the antibody y were ii.v.vv. aadministered (((2522 00 g/g/mouse) on

poop ststtooperative dadaysyss 11 aaanddd 22 (((SuSuSupplellememeennttal FFFiigg. SS1CC).

ExEx vvvivivivooo gagagammmmmma aa cococounununtititingngng andndnd hhhisisisttotopapaathththololologogogyy y

See Supplememementntnt fffororor fffululu l ll dedeetatatailili s. AfAfAftetet r r sasasacrcrcrififificicce,e,e mimimicecece wwweereree pepeperfrfrfusu ededd wwwititith hh 0.0.0 9%9%% ssalala ininineee via the leftftt



ownloaded from



6

Protein extraction and immunoblotting

See supplement.

Clinical FDG-PET study

See Supplement for full details. Access of patient records and analysis of patient data were

approved by the Partners Institutional Review Board. Participants for the clinical imaging sub-

study were consecutively identified from a database of patients who had undergone 18F-FDG-

PET/CT imaging for oncological evaluation at Massachusetts General Hospital. From a clinical

database of 437 patients who underwent PET/CT imaging and also had a clinical diagnosis of

DVT, we consecutively identified 19 patients with iliofemoral DVT and whose FDG-PET

imaging was performed within 6 months after the onset of DVT. Matched control patients were

then identified for each individual with a DVT by: age, sex, malignancy status, chemotherapy

exposure, and steroid immunosuppressive therapy within 6 months. An individual region-of-

interest (ROI) was placed around the area just superior to the popliteal vein and extended to the

inferior of the iliac bifurcation in order to obtain a maximum standardized uptake value

(SUVmax). The average venous blood uptake of the right atrium was used to derive a TBR.18

Patients were then analyzed to assess the time-dependent changes in FDG uptake in DVT and

corresponding vein of matched control patients without a DVT.

Statistics

See Supplement for full details. For animal data, results are expressed as median [25%-75%

quartiles]. Statistical comparisons between two groups were evaluated by the Mann-Whitney U

test, and by the Kruskal-Wallis test for multiple groups followed by the Dunn’s post-test. For

comparison between two groups within the same animal, the Wilcoxon matched-pairs signed

rank test was used. Continuous variables at multiple timepoints were compared by the Friedman

maging was performed within 6 months after the onset of DVT. Matched controolll papapatiiienentststs wwwerere

hen identified for each individual with a DVT by: age, sex, malignancy status, chemotherapy

exxpopoposususurere, anana ddd stterererooioid immunosuppressive therapppy yy wwwithin 6 monththhs.s AAnn n iinindividual region-of-

nnteeerer st (ROI)) wwasas pplalalaceceeddd aararouououndnd ttthhhee aaareeea juuusttt suppeperriorrr ttto o ththhee popoplplititeaeal ll vveeinin aanndnd eeextxtxtenene dded d d d toto thhhe

nnfefefeririiororor ooff thththee ililiiaaccc bibifffurrcrcaatioonon iiinnn ororordederr r totoo ooobbbtaiaiain n n aaa mmamaxixix mumumum m ststs aanandadadardrdr iiizeeded uuuptptptakaka e e e vavvaluuuee

SUVmax). Thehee aaaveveerararagegege vvvennnououous blblb ooooood dd upupupttakakakeee ofofof tthehehe rrrigigi hththt aaatrtrtriuiuium m m wawawasss usususededed ttto o o dedeeriririveveve aaa TTBR.181818



ownloaded from



7

test. FDG uptake measurements were correlated with histological findings by use of the two-

tailed Spearman method. Statistical comparisons were performed with GraphPad Prism (La Jolla,

CA). Mice that did not develop jugular DVT (histologically negative, n=3) were excluded from

analyses. For clinical data, values are expressed as mean SEM. The Wilcoxon signed-rank test

was used for single comparisons, and after confirming normality of distribution, linear regression

analysis was used to determine the association between FDG uptake and DVT age (SPSS 22,

IBM, Chicago, IL). A value of P<0.05 was considered statistically significant.

Results

Development of a murine stasis-induced DVT model in the jugular vein suitable for FDG-

PET imaging

The surgical ligation-induced stasis model of murine DVT is well established in the IVC.14, 15, 17

However, preliminary experiments with FDG-PET imaging of acute IVC DVT did not allow

thrombus detection due to high FDG background in the kidneys and spine, as well as the

surrounding bowel, an organ with high glucose utilization.19 Therefore we established a new

murine stasis-induced DVT model in the jugular vein, an area of lower background FDG uptake

and a common site of clinical DVT.5 Jugular DVT were readily detected as obstructions on CT

venography, in contrast to sham-operated contralateral jugular veins (Fig. 1A-B). Histological

analyses confirmed fibrin, red blood cells, and white blood cells, similar to the ligation IVC

stasis model (Fig. 1C-E).14, 17, 20

Noninvasive imaging of DVT inflammation by FDG-PET/CT

In vivo 18F-FDG PET/CT imaging was performed from day 2 up to day 16 after jugular vein

ligation. Elevated FDG signal was observed in the right thrombosed jugular vein, with

Development of a murine stasis-induced DVT model in the jugular vein suittabaablell fffororor FFFDGDGD -

PET imaging

ThThee e sususurrgicicicalalal lligatatatioioion-n induced stasis model of mururrinnne DVT is wellll l estatablblbliisished in the IVC.f 14, 15, 17

HHowwwever, prelilimmminnnaryry exxpxpererimimimenentststs wwwiitthh FDDG-G-PEETT immagagagininggg oofof aacuuute IIVCVCVC DDDVTVVT ddididid nnnotott aaalllllowoww

hhroroombmbmbususus dddetetetecectititionnn dduuue ttto o hiighghgh FFFDGDGDG bbbacacackgkgkgrorr ununundd d innn tthhehe kkkidididneneeysysy aandndnd ssppipinnene, , asasas wwwelelell l l aasas ttthehee

urrounding g bobobowewewel,l,l aaan n n ororo gagaannn wiwiiththth hhhigigi h h h glglglucucucososose e e ututu ilililizizzata ioioon.n.n.19 TTThehehererefofoforerere wwweee esese tataablblblisisi hehehed dd a new



ownloaded from



8

significantly less signal detected in the sham-operated jugular vein (Fig. 2A, FDG-PET).

Thrombosed jugular veins induced a filling defect on CT venography (Fig. 2B, CT) that

colocalized well with FDG-PET signals (Fig. 2C, PET/CT).

18F-FDG accumulation diminishes over time in experimental DVT

To determine whether the FDG-PET signal was modulated by DVT age, we analyzed DVT

signals as a function of time after jugular vein ligation. We observed a significant decrease in the

18F-FDG DVT signal over time (SUV (median [quartiles]) = 1.98 [1.87-2.70] day 2, 1.20 [1.04-

1.30] day 4, 0.88 [0.82-0.96] day 7, 0.68 [0.57-1.22] day 14, p<0.001, Fig. 3A). The DVT

SUVmax and TBR values were highest at day 2 and decreased thereafter (SUVmax = 2.31 [2.20-

3.23] day 2, 1.35 [1.17-1.58] day 4, 1.07 [0.92-1.27] day 7, 0.93 [0.66-1.54] day 14, p=0.01;

TBR = 1.85 [1.79-2.09] day 2, 1.49 [1.45-1.79] day 4, 1.19 [1.10-1.32] day 7, 1.13 [1.08-1.23]

day 14, p=0.002, Fig. 3B-C). Gamma radiation counts of resected DVT demonstrated similar

findings, with a time-dependent decrease in radioactivity (%IDGT = 7.81 [6.47-19.2]% day 2,

7.72 [7.14-13.1]% day 4, 6.47 [5.25-9.36]% day 7, 3.04 [2.48-5.44]% day 14, vs. 0.735 [0.458-

1.75]% (pooled sham), p<0.0001, Fig. 3D).

Three mice underwent serial PET/CT imaging at day 2, 4, and 14 after the ligation.

Serial imaging revealed a time-dependent decrease in the FDG signal in individual DVT

(p=0.03, Fig. 3E-F).

Neutrophil appearance associates with FDG uptake in DVT

We compared 18F-FDG uptake measurements and histological profiles at the study timepoints of

2, 4, 7, and 14 days after induction of stasis DVT. Neutrophils infiltrated DVT abundantly at the

earlier timepoints of day 2 and 4, with fewer neutrophils observed at day 7, and minimal

neutrophils at day 14 (Fig. 4). Thrombus macrophages were evident at day 7 and resided in the

3.23] day 2, 1.35 [1.17-1.58] day 4, 1.07 [0.92-1.27] day 7, 0.93 [0.66-1.54] day y 141414,,, p=p=p=0.0.0 010101;;;

TBR = 1.85 [1.79-2.09] day 2, 1.49 [1.45-1.79] day 4, 1.19 [1.10-1.32] day 7, 1.13 [1.08-1.23]

daay y y 141414,, p=p==0.0.0.00002,2, FFiFig. 3B-C). Gamma radiation n cococ uunts of resectededd DVTVTVT ddemonstrated similar

fifiindddini gs, withh aa tttimmme-e-deded pepependndndenennt t dededeccrcreeaassee in raaadiooacacctiviviitytyty ((%I%I%IDGDGGTTT == 7.7.7 811 [[66.6.44747-1-119.9.9.2]2]2]%%% dadaday y 2,2,, n

7..727272 [[[7.7.7 1414-1-1-13.33.1]1]]%% % dadayyy 4,4,4, 66.444777 [5[5[5 2.2.25-5-5 9.9.36366]%]%]% dddayayay 777, 3.3 00404 [[[22.2.48488-5-5- ..444]4]4]% % % dadaay y y 141414, vsvsv .. 0.0.0.737355 [0[0[ .4.445888--

1.75]% (poololededed sshahaham)m)m), , p<p<p 0.0.0 00000 010101, , FiFiFig.g. 333DDD).)..



ownloaded from



9

outer DVT edge (Fig. 4). These histological findings in our developed jugular DVT stasis

ligation model recapitulate both the IVC full stasis DVT and the partial flow IVC DVT

models.14, 20 The number of neutrophils in DVT correlated with the FDG DVT uptake (SUV;

r=0.41, p=0.004, TBR; r=0.62, p<0.001, Fig. 5A). The macrophage FDG relationships were

analyzed in day 7 and day 14 DVT, as day 2 and day 4 DVT showed few macrophages and also

substantial confounding neutrophils. In the day 7 and 14 cohort, macrophages were not

significantly associated with the FDG DVT signal (SUV; r=-0.15, p=0.47, TBR; r=0.26, p=0.23,

Fig. 5B). To avoid confounding neutrophil-based FDG signal at day 7, we further assessed only

the day 14 macrophage association with FDG signals. This correlation remained nonsignificant

(SUV; r=0.04, p=0.91, TBR; r=0.30, p=0.37, Fig. 5C).

To further examine the relationship of FDG-PET and thrombus inflammation, we

assessed the protein expression of matrix metalloproteinase-9 (MMP-9), a key inflammatory

mediator of DVT.8 Immunoblot analyses showed a time-dependent decrease of MMP-9

expression in DVT (Supplemental Fig. S2A-B, p=0.01), similar to the in vivo FDG signals.

To evaluate the relative contributions of FDG signal from the thrombus and from the vein

wall, we performed ex vivo gamma radiation measurements after carefully separating the

thrombus and the vein wall components (day 2 DVT, n=5). We found that 66.7±5.9% of the

radioactivity localized in thrombus. We further analyzed the distribution of neutrophils using

immuno-histological images. We observed 86.4±1.4% of neutrophils (NIMP-R positive area)

were localized to thrombi, with the remaining 14% localized to the vein wall. These data indicate

that the majority of the FDG signal arises from thrombus itself.

Neutropenia markedly reduces FDG-PET signal generation in DVT

Histological analyses above demonstrated that thrombus neutrophils, but not macrophages,

SUV; r=0.04, p=0.91, TBR; r=0.30, p=0.37, Fig. 5C).

To further examine the relationship of FDG-PET and thrombus inflammation, we f

assseseessssssededed ttthehehe ppprooteteteiinin expression of matrix metalllopopo rrooteinase-9 (MMMPMM -9-99)),), aa key inflammatory

mmeddidiator of DVDVTT.T 888 IImmmmmm nununobobobloloot t anannaalalyysseess shhoowwwed a a timmmee--dedeepepeendndenennt t dedecccreeaeasesee ooof f MMMMMMP-P-P 99

exxprprpresesessisis onon iiin nn DVDVVTTT ((SSuuppppplel mmementntntalalal FFigigg. S2S2S2AA-A-BBB, , p=p=p=0.0 00101),)), sssimimimilililararr ttooo ththt eee ininn vvivivi o oo FDFDDGGG sisiignnnalals.ss.

To evvalalaluauau tetete ttthehee rrrelatatativivive cococontnttrirr bububutitiionononss ofofof FFFDGDGDG sigigignananall frfrromomom ttheheh ttthrhrhromomombubuusss ananandd d frfrf om the veiinn



ownloaded from



10

provided the basis for elevated FDG signals in DVT. To further determine whether neutrophils

could directly modulate the FDG signal in DVT, we induced systemic neutropenia in a subgroup

of mice prior to DVT induction. Pre-treatment with an anti-neutrophil antibody decreased

circulating neutrophils in day 2 DVT mice (0.6 [0.2-0.9] x103/ L vs. control day 2 DVT mice,

1.7 [1.2-2.1] x103/ L, p=0.03). On day 2, imaging demonstrated a significant decrease in FDG-

DVT signal in neutropenic mice, compared with normal mice (SUV = 1.15 [1.06-1.51] vs. 1.98

[1.87-2.70], p=0.03, Fig. 6). Histological assessment confirmed diminished neutrophil

accumulation in DVT (neutropenia 21.0 [14.0-59.0] vs. control 102 [86.0-126], # neutrophils per

5 high-power-fields, p<0.001). Neutropenic mice also showed decreased expression levels of

MMP-9 in DVT in parallel to FDG signal (Supplemental Fig. S2C).

Establishment of novel animal model of recurrent DVT

To our knowledge, this is the first report demonstrating an animal model of recurrent DVT. The

conventional stasis-induced DVT model is not suitable for inducing recurrent DVT due to the

absence of adequate blood flow and blood volume after complete ligation,14 precluding

generation of a second, same location, stasis-induced DVT. To create an environment suitable

for a second stasis DVT, at day 2 we de-ligated (cut the suture) the initial jugular vein ligation, to

spur partial thrombus recanalization and restoration of blood flow. In addition we ligated the

contralateral jugular vein to induce an occlusive stasis DVT. Twelve days later (at day 14), the

de-ligated vein was then re-ligated at same location (Fig. 7A). This procedure successfully

generated a second new venous thrombus at day 2 directly overlying the 16 days old primary

thrombus. Histological staining clearly distinguished the second thrombus at day 2 (increased

fibrin, increased neutrophils, and less collagen) from the organized primary thrombi at day 16

(Fig. 7B-D). The vein wall also exhibited DVT-induced scarring (increased thickness and

MMP-9 in DVT in parallel to FDG signal (Supplemental Fig. S2C).

Establishment of novel animal model of recurrent DVT

ToTo oouurur kknonowlwlwleddgege, tthis is the first report demonsstrtrt aaating an animaall l moodeded ll l of recurrent DVT. The

coonnvnventional sstatassisss-ininduduuceed d DVDVDVT T mmmodddelll is nnnoott suuiitaaableee ffofor rr ininduducccinnng reeccuuurrereentntt DDVVTVT dduuee tttoo ththhe

ababseseencncnceee ofofof aaaddedequququatatte e e bblbloooood dd fllowowow aandndnd bblololoododod vvvolollumumume e afafafteteer r cococompmpmplelel tetee llligigigatatatioion,n,n 14144 ppprerereclclcludududininng g g

generation oof f f aaa seses cococondndnd,, ssamememe llococcatata ioioon,n, ssstatatasisisis-s-s-ininindududucececed d d DVDVVT.T.T ToToTo cccreeeatatate e e ananan eeenvnvn iririronononmemementnn suitable



ownloaded from



11

collagen-rich), which is observed in patients with post-thrombotic syndrome (PTS).21 A number

of neutrophils were present in day 2 recurrent DVT, while macrophages were abundant in day 16

DVT, similar to day 14 DVT (Fig.7 E-F).

FDG-PET/CT enables the identification of recurrent, same site DVT

As our data revealed that FDG signal is neutrophil-dependent, we examined whether FDG-PET

could identify recently formed, neutrophil-rich recurrent DVT in the same site as the primary DVT.

As shown in Fig. 7G-I, recurrent DVT was successfully identified by FDG-PET/CT. FDG signal

from recurrent DVT (recurrent DVT at day 2 + primary DVT at day 16) was significantly higher

than old day 14 DVT (SUV; 1.65 [1.42-1.83] vs. 0.962 [0.892-1.07], SUVmax; 2.01 [1.63-2.17] vs.

1.06 [0.973-1.16], p=0.03 respectively, Fig. 7J-K). Ex vivo gamma radiation measurement also

showed a significant increase in the activity of recurrent DVT (day 2+16) compared to older DVT

at day 14 (%IDGT = 20.46 [16.4-26.3]% vs. 4.60 [2.46-7.75]%, p=0.02, Fig 7L).

18F-FDG activity in DVT decreases over time in patients

We retrospectively analyzed 38 individuals who underwent PET/CT scanning for clinical

indications: 19 patients with DVT (10/9 male/female, mean age, 64.1 years) and 19 matched

control patients (Supplemental Table). We observed a significantly increased signal within

recently formed DVT (Fig. 8A-B). Across the entire population, the FDG signal in DVT was

significantly greater than in the matched vein of controls (SUVmax 1.87 0.15 vs. 1.32 0.10,

P=0.02; TBR 1.62 0.19 vs. 1.10 0.05, P=0.01). Furthermore, we observed a time-dependent

decrease in the FDG signal within DVT. When the DVT age was stratified by tertiles, both the

SUVmax and TBR of DVT diminished over time (p=0.002 for SUVmax, p=0.004 for TBR, Fig.

8C-D). The relationship between DVT age (by tertiles) and DVT FDG uptake (either as SUVmax

or TBR) remained significant after multivariable adjustments for: demographic factors, (TBR:

1.06 [0.973-1.16], p=0.03 respectively, Fig. 7J-K). Ex vivo gamma radiation meaasasururremmmennent tt alalalsoso

howed a significant increase in the activity of recurrent DVT (day 2+16) compared to older DVT y

att dddayayay 111444 (%(%(%IDIIDGTGTGT == 20.46 [16.4-26.3]% vs. 4.60 0 0 [2[2..46-7.75]%, p==0.0 022, , FFiFig 7L).

888F---FDF G actiiviviittyt in n DVDVDVTTT dededeccrcreaeaasesees s ovovver tttimmme inn ppatttieieentntsss

WeWeWe rretetetroror spspspececectitiveveelyyy aanannalylyyzezed d 38388 iiindndndiivivididduauaualslsls whwhho o o ununundederwrwrwenenenttt PEPEPETTT/CTCTCT ssccacannnnninining g foforr r clclclinniiicaalal

ndications: 1199 9 papapatititienene tstst wwititth h h DVDVDVTTT ((10101 /9/9/9 mmmalalale/e/e/fefefemamamalelee, , mememeananan aaagegege,, 64644.1.1.1 yyyeaeaearsrsrs) )) ananand d d 191919 mmmatched



ownloaded from



12

=-0.209, p=0.007; SUVmax: =-0.233, p=0.003); factors that could impact systemic

inflammation (TBR: =-0.228, p=0.002; SUVmax: =-0.237, p=0.003); and oncologic history

(TBR: =-0.225, p=0.003; SUVmax: =-0.245, p=0.001).

Discussion

In this experimental and clinical study, we demonstrate that the intense inflammatory response

produced by DVT can be systematically imaged, serially assessed, and quantified with

noninvasive PET/CT. The FDG inflammatory signal exhibited a time-dependent and neutrophil-

dependent decay in murine DVT. We also established that FDG-PET/CT could differentiate

recurrent DVT from primary DVT utilizing a novel animal model of recurrent DVT.

Furthermore, in a clinical study, we observed a similar time-dependent decrease in the FDG-PET

signal in human DVT. The overall findings demonstrate that FDG-PET/CT provides an accurate

approach to assess neutrophil-dependent and age-dependent DVT inflammation, and can

specifically diagnose recurrent DVT.

Recurrent DVT occurs up to 10% per year in patients with unprovoked DVT 4 and is a

highly morbid condition, increasing the risk of the post-thrombotic syndrome, pulmonary

embolism, and death.2, 22, 23 Accurate diagnosis of recurrent DVT is therefore important in

initiating timely anticoagulant therapy, determining the duration of treatment, and recognizing

whether a failure of anticoagulation has occurred, with implications for considering IVC filter

placement to reduce the risk of pulmonary embolism.2 However, the diagnosis of a recurrent

ipsilateral same-site DVT, defined as a new DVT occurring in the same location as a previous

DVT, poses a significant diagnostic challenge. Although recurrent DVT is common as an

endpoint in observational and therapeutic clinical venous thromboembolism (VTE) trials,2 there

ecurrent DVT from primary DVT utilizing a novel animal model of recurrent DVDVVT.T.T

Furthermore, in a clinical study, we observed a similar time-dependent decrease in the FDG-PET

iigngnnalalal iiinn huhuhummman n DVDVDVT. The overall findings demomomonsnsstrate that FDGGG-PETETT/C/C/CT provides an accurate

appprproao ch to asssesessss nneueuutrrropopophihihil-l-l-ddedepepepenndndeenentt andndd aaage--ddeepeennddedennnt DDDVVTT iinfnfflalalammmmmaatatioioion,n, aaandndnd ccanann

ppecececififificicicalallylyly dddiaiaggngnoososee rreccucurrrenennt t DVDVDVT.T.T

Recuurrrrrenenent t DVDVDVTTT oco cucucursrsr uuuppp totoo 10%0%0% ppperere yeyeyearara iin n n papapatititienenentststs wwwititith h unununprprprovovovokokokededd DDDVTVTVT 444 and is a



ownloaded from



13

is no gold standard for its diagnosis, including ultrasound, MRI, CT, or venography.5, 6, 24 Venous

compression ultrasound, the gold standard for initial DVT diagnosis,25 is limited in cases of

recurrent DVT where residual thrombus or vein wall scarring is present after the initial DVT.

The current findings suggestive of prior DVT on duplex ultrasonography include retraction of

the vein, the presence of collateral veins, and recanalization. These signs are often subtle and are

frequently absent. Diagnosis of an acute DVT in the milieu of a prior DVT is particularly

difficult, especially if large changes in thrombus length or compressed vein diameter are not

apparent.5, 6 The situation is even more complicated if previous ultrasound images are

unavailable, and some patients may require invasive venography to attempt to establish a

diagnosis, which is also limited in its accuracy.5 In the past, studies have attempted to distinguish

acute from organized thrombi with ultrasonography,26 MRI,27, 28 and SPECT imaging.29, 30

However, no established imaging technique is routinely utilized in the clinic.5 The current

results herein demonstrate that FDG-PET/CT can distinguish newly formed, neutrophil-rich

thrombus from older thrombus and can specifically identify recurrent DVT.

Whereas a single case report suggested the potential for FDG-PET/CT to detect recurrent

DVT, the authors did not specify if DVT recurred in the same location as the original DVT, nor

systematically assess the age of the thrombus, nor obtain histological assessment.31 In addition,

while other clinical reports have demonstrated that FDG-PET can identify DVT (but not

specifically recurrent DVT),32, 33 our experimental results extend these prior studies, as we

precisely controlled the DVT age and timing of subsequent FDG-PET/CT imaging studies,

performed quantitative ex vivo radioactivity and detailed histology assessments to corroborate in

vivo FDG-PET signals, and induced neutropenia to modulate the FDG-PET signal.

As we established that the FDG DVT signal is time-dependent, FDG-PET offers a

diagnosis, which is also limited in its accuracy.5 In the past, studies have attemptptedeed tto ddidistststininingugug ish

acute from organized thrombi with ultrasonography,26 MRI,27, 28 and SPECT imaging.29, 30

HoHowewewevvever,r, nnnoo estataabblblisi hed imaging technique is rooouutu iinely utilized iin nn the e clclcliinic.5 The current

eesuuultl s herein ddeemmononststraaatetete ttthahahatt t FDFDDGG-G-PPEPETT/CCTT cann ddistiiningguguiisshhh nenewlwly y fofoformrmmededd, nneneututtroror phphphilii -rrricicich hh

hhroroombmbmbusus fffrororomm oololddederr tththrrorombmbuusu aaandndnd ccanann ssspepepecccififificacac lllly y ididdenenentititifyfyfy rrrecececuuurrererentntn DDDVVTVT.

Wherreaeaeasss a a a sisisingngnglelee ccasasa e e e reeepopoportrtt suguguggegegestststededed tthehehe pppototo enenntititialala fffororor FFFDGDGDG-P-PPETETET/C/C/CTTT totoo dddetetetecee t recurrenntt



ownloaded from



14

noninvasive approach to assess DVT age in vivo. This finding has implications for fibrinolytic

therapy of DVT using catheter-directed therapy or pharmacomechanical therapy, two emerging

clinical strategies used to treat large iliofemoral DVT.34, 35 This is because intravascular-based

fibrinolysis of DVT appears more effective for earlier stage thrombi < 10 days old.36 As the

precise age of DVT is not often possible to assess clinically due to the insidious onset of flow-

limiting DVT, FDG-PET assessment of DVT age might help predict the success of catheter-

directed therapy or pharmacomechanical therapy.

In this study, the observation that the inflammatory FDG signal was more closely

associated with neutrophils than macrophages in DVT is not unexpected. FDG uptake occurs in

cells exhibiting increased glycolysis, which is higher in myeloid cells, particularly activated

macrophages12, 37 and neutrophils.38 We determined that the FDG-PET signal was neutrophil-

dependent, and accordingly higher in early, neutrophil-rich DVT. Subsequently, in resolving

subacute DVT, we hypothesized that infiltrating macrophages might be of an M2-polarized

(reparative) phenotype, as M2-macrophages do not exhibit up-regulated glucose transporter

protein type 1 (GLUT-1) expression needed to concentrate FDG.12, 39 Histological analyses

supported this hypothesis, as macrophages within day 14 DVT in fact displayed little GLUT-1

expression and little iNOS expression, an M1 marker (Supplemental Fig. S3). This is in

distinction to chronic inflammatory conditions such as in atherosclerosis, where macrophages

account for the FDG signal,13, 37, 40-43 and demonstrate an M1-polarized (pro-inflammatory) and

GLUT1-upregulated phenotype.12 Moreover, the neutrophil-based FDG findings in DVT are

similar to studies of other acute, neutrophil-rich lesions such as acute lung injury, where

neutrophils but not macrophages serve as the main cellular source of FDG uptake.9-11 As there

was a mild FDG signal in day 14 DVT without neutrophils, other non-macrophage cells that may

cells exhibiting increased glycolysis, which is higher in myeloid cells, particularrlylyly aaactctivivvatatatededed

macrophages12, 37 and neutrophils.38 We determined that the FDG-PET signal was neutrophil-

deepepependndndenentt,t, aannnd aaaccccccoordingly higher in early, neutttrrropopphil-rich DVTT. Sububbseseseqquently, in resolving

uubabaacute DVTT, wewewe hhyyppototothehehesisiizezzed d thththaatat iinfnffiltraaatiinng mmmaaacrooopphphagagageess mmmiggghtht bbbee ofof aaan n M2M22-p-p- olololaarrizzz deded

rrepeppararratata ivive)e)e) ppphehennonotytyt ppepe, asas MMM2-2-2-mamamacrcrcropopphahahagegegesss dododo nnootot exexxhihiibibibit tt upupp-r-r- eegegulululatata eeed glglg ucucucososose e trtrtraananspsppororrteter r

protein type 11 (((GLGLGLUTUTUT-1-11) ) exexexprprpressssisisiononn neeeeeedededed d d tototo cconononceceentntrararatetee FFFDGDGDG...1212,12, 393939 HHHisisistototololol gigigicacacal l l anananala yses



ownloaded from



15

have accumulated FDG include activated myofibroblasts, which are present in the remodeling

vein wall after DVT7, 44 and ingest FDG in experimental arthritis models.45 Another

consideration regarding lower FDG signal in the later-stage DVT could be due to greater

thrombus organization, limiting access of FDG to interior thrombus cells.

Limitations of this study are present. As stasis-induced DVT was generated by surgical

ligation of the jugular vein, surgery-induced inflammation is inevitable especially at day 2,

resulting in FDG uptake at wound healing region around the DVT and likely in the venous wall.

While experiments demonstrate that the majority of FDG signal arises from the thrombus, the

vein wall also contributes modestly to the FDG signal. As the ligation suture is necessary to

induce stasis thrombi, we cannot precisely resolve intrinsic vein wall inflammation that occurs

during DVT resolution from ligation-induced injury. Second, our clinical cohort did not contain

any patients with a recurrent DVT who underwent FDG-PET imaging. Third, the clinical cohort

was derived from individuals undergoing imaging for oncologic indications, and was a

retrospective study. Accordingly, generalization of the findings may be constrained, and

prospective clinical FDG-PET studies of recurrent DVT and serial DVT imaging will be needed

to validate our findings. Finally, while it is known that 18F-FDG uptake is upregulated in lesions

containing pro-inflammatory myeloid cells such as neutrophils9-11 and macrophages,12, 13

histological analysis of clinical DVTs was not possible to perform in this study. Future FDG-

PET DVT clinical studies that include a DVT biopsy will be able to elucidate specific

inflammatory cells underlying FDG uptake in human DVTs.

Clinical Implications

Our clinical retrospective study also demonstrated a time-dependent decrease in the FDG signal

within DVT, extending the main experimental findings from the animal study. DVT showed

nduce stasis thrombi, we cannot precisely resolve intrinsic vein wall inflammatitionoon tthahaat t t ocococcucucursrs

during DVT resolution from ligation-induced injury. Second, our clinical cohort did not contain

anny y y papapatititienenntststs wwwitthh h aaa rrecurrent DVT who underwenenent FFDG-PET imagagaging.g. TTThird, the clinical cohort

wwass s ded rived frfromomm iiindnddivivi ididduauaalslss uuundnddererergogog iningg immmaagginggg fffor onononcocololoogigicc innndidicacaatiiononss,s, aaandnd wwwaasas aaa

eetrtrtrososospepepectctivivve ee ststuuddyy.y. AAcAccococ rdrdinininglglglyyy, gggenenerere alalalizizizatatioioion n ofoff thhehe fffininindddingngngs s mamamay y y bbbe ccononnstststrarar ininnededd, aanandd d

prospective clcllinininicicalalal FFDGDGDG-PPPETETET ssstututudidiiesee ooof f rererecucucurrrrrrenennt t DVDVD TTT ananand d d seseseriririala DDDVTVTVT iiimamam gigigingngng wwwililll l be neededd



ownloaded from



16

significantly greater FDG signal compared to the matched vein of control patients. By analyzing

greater numbers of DVT patients and with clot ages up to 21 weeks old, our findings extend a

smaller study that was restricted to < 10 week old thrombi.32

The findings of this study support the concept that FDG-PET imaging might be useful for

determining the inflammatory activity of DVT in living subjects. It is possible that PET/CT

could be employed when there is uncertainty regarding whether recurrent DVT exists, or to what

degree there is a substantial inflammatory component to an initial DVT, which might also inform

the risk of the post-thrombotic syndrome.46, 47 Ultimately, prospective clinical studies will need

to be conducted to evaluate if FDG-PET imaging of recurrent DVT provides additive diagnostic

value, and enables better outcomes.

Conclusions

Noninvasive FDG-PET/CT enables assessment of thrombus age and inflammation in

experimental and clinical DVT, and enables the specific detection of recurrent murine DVT.

Thrombus neutrophils are a major cellular basis of FDG signal in early murine DVT. Elevated

FDG-PET signal indicates a recently formed, neutrophil-rich thrombus and thereby offers an

imaging strategy to accurately diagnose recurrent same-site DVT.

Funding Sources: This study was supported by NIH HL108229 (FJ), Wagner-Torizuka Society

of Nuclear Medicine Fellowship (TH), American Heart Association Founders postdoctoral

fellowship #13POST14640021 (TH), and Grant-in-Aid #13GRNT17060040 (FJ).

Conflict of Interest Disclosures: None.

value, and enables better outcomes.

CoConncnclululu isisionononss

NNonnininvasive FDFDDGG--PEPEET/T//CTCTCT eeennanablbleseses aassseeessmmmennnt offf tthhrooommbmbuus aagege aandnd iiinnfn llaammmmmmatatioioon nn ininin

exxpepeperiririmemem ntnttalalal aandndd ccclilinniicaaal l DVDVVT,T,T, aaandndnd eennanablblbleseses thhhe ee spspspeececiffficicc dddeteetececctitit oonon ooof f f rereecucuurrrreenenttt mumuuriririnee DDVTVT..

Thrombus neueuutrtrtropopophihihilslss aaarer aaa mmmajajjororor cccelee luuulalalarr r bababasisis sss ofofof FFFDGDGG ssigigignananall l ininin eeearararlylyly mmmurururinineee DVDVDVT.T.T. Elevated



ownloaded from



17

References:

1. Go AS, Mozaffarian D, Roger VL, Benjamin EJ, Berry JD, Borden WB, Bravata DM, Dai S, Ford ES, Fox CS, Franco S, Fullerton HJ, Gillespie C, Hailpern SM, Heit JA, Howard VJ, Huffman MD, Kissela BM, Kittner SJ, Lackland DT, Lichtman JH, Lisabeth LD, Magid D, Marcus GM, Marelli A, Matchar DB, McGuire DK, Mohler ER, Moy CS, Mussolino ME, Nichol G, Paynter NP, Schreiner PJ, Sorlie PD, Stein J, Turan TN, Virani SS, Wong ND, Woo D, Turner MB. Heart disease and stroke statistics--2013 update: a report from the american heart association. Circulation. 2013;127:e6-e245.

2. Jaff MR, McMurtry MS, Archer SL, Cushman M, Goldenberg N, Goldhaber SZ, Jenkins JS, Kline JA, Michaels AD, Thistlethwaite P, Vedantham S, White RJ, Zierler BK. Management of massive and submassive pulmonary embolism, iliofemoral deep vein thrombosis, and chronic thromboembolic pulmonary hypertension: a scientific statement from the American Heart Association. Circulation. 2011;123:1788-1830.

3. Kyrle PA, Eichinger S. Deep vein thrombosis. Lancet. 2005;365:1163-1174.

4. Baglin T, Luddington R, Brown K, Baglin C. Incidence of recurrent venous thromboembolism in relation to clinical and thrombophilic risk factors: prospective cohort study. Lancet. 2003;362:523-526. 5. Bates SM, Jaeschke R, Stevens SM, Goodacre S, Wells PS, Stevenson MD, Kearon C, Schunemann HJ, Crowther M, Pauker SG, Makdissi R, Guyatt GH. Diagnosis of DVT: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest. 2012;141:e351S-418S. 6. Schellong SM. Diagnosis of recurrent deep vein thrombosis. Hamostaseologie. 2013;33:195-200. 7. Saha P, Humphries J, Modarai B, Mattock K, Waltham M, Evans CE, Ahmad A, Patel AS, Premaratne S, Lyons OT, Smith A. Leukocytes and the natural history of deep vein thrombosis: current concepts and future directions. Arterioscler Thromb Vasc Biol. 2011;31:506-512. 8. Wakefield TW, Myers DD, Henke PK. Mechanisms of venous thrombosis and resolution. Arterioscler Thromb Vasc Biol. 2008;28:387-391. 9. Zhou Z, Kozlowski J, Goodrich AL, Markman N, Chen DL, Schuster DP. Molecular imaging of lung glucose uptake after endotoxin in mice. Am J Physiol Lung Cell Mol Physiol. 2005;289:L760-768. 10. Locke LW, Williams MB, Fairchild KD, Zhong M, Kundu BK, Berr SS. FDG-PET Quantification of Lung Inflammation with Image-Derived Blood Input Function in Mice. Int J Mol Imaging. 2011;2011:356730. 11. de Prost N, Tucci MR, Melo MF. Assessment of lung inflammation with 18F-FDG PET

4. Baglin T, Luddington R, Brown K, Baglin C. Incidence of recurrent venous thrhrrommmboboboememembobobolilismn relation to clinical and thrombophilic risk factors: prospective cohort study. Laancncnceetet.

2003;362:523-526.

5.. BBBatatateeses SSSM,MM, Jaeeescscschkh e R, Stevens SM, Goodacreree S,, Wells PS, Stteveve ennsosoonn n MD, Kearon C,ScScchuhuunemannnn HHJ,J,, CCCrorowtwtwtheheer rr M,M,, PPPauauaukekerr SGSG, , MaMakdkdiiissssi RR, , GuGuGuyayay ttt GGGHH.H. DDiaiaagngg ososisiss oooff f DVDVT:T:: AAAnttititht romboticic TTThhherarapypyy aaandndnd PPPrerevevevenntntiooonn of TTThhhrommmbbosiiis,s,s, 99ththh eeed:d: AAmemeriiricacaan n CoCoCollllegeggeee ofofof CCChehehestst PhPhPhysysysicians EEviv deeennce-BaBaBased d ClClinici alall PPraraccticee GGGuidddeelelinineseses.. ChCC eeest. 20012;1144141:eee33551S1 -4441188SSS.

6. SSchchelllolongng SSM.M DDiaiagnossisi ooff rerecuc rrent t dedeepep vveiein thhrorombmbososisis. HaHamomosts asseoeolologigiee. 200113;333:3:191 5-5-200.



ownloaded from



18

during acute lung injury. AJR Am J Roentgenol. 2010;195:292-300.

12. Satomi T, Ogawa M, Mori I, Ishino S, Kubo K, Magata Y, Nishimoto T. Comparison of Contrast Agents for Atherosclerosis Imaging Using Cultured Macrophages: FDG Versus Ultrasmall Superparamagnetic Iron Oxide. J Nucl Med. 2013;54:999-1004.

13. Tawakol A, Migrino RQ, Bashian GG, Bedri S, Vermylen D, Cury RC, Yates D, LaMuraglia GM, Furie K, Houser S, Gewirtz H, Muller JE, Brady TJ, Fischman AJ. In vivo 18F-fluorodeoxyglucose positron emission tomography imaging provides a noninvasive measure of carotid plaque inflammation in patients. J Am Coll Cardiol. 2006;48:1818-1824 14. Diaz JA, Obi AT, Myers DD, Jr., Wrobleski SK, Henke PK, Mackman N, Wakefield TW. Critical review of mouse models of venous thrombosis. Arterioscler Thromb Vasc Biol. 2012;32:556-562. 15. Varma MR, Varga AJ, Knipp BS, Sukheepod P, Upchurch GR, Kunkel SL, Wakefield TW, Henke PK. Neutropenia impairs venous thrombosis resolution in the rat. J Vasc Surg. 2003;38:1090-1098.

16. Eliason JL, Hannawa KK, Ailawadi G, Sinha I, Ford JW, Deogracias MP, Roelofs KJ, Woodrum DT, Ennis TL, Henke PK, Stanley JC, Thompson RW, Upchurch GR, Jr. Neutrophil depletion inhibits experimental abdominal aortic aneurysm formation. Circulation. 2005;112:232-240. 17. Henke PK, Pearce CG, Moaveni DM, Moore AJ, Lynch EM, Longo C, Varma M, Dewyer NA, Deatrick KB, Upchurch GR, Jr., Wakefield TW, Hogaboam C, Kunkel SL. Targeted deletion of CCR2 impairs deep vein thombosis resolution in a mouse model. J Immunol. 2006;177:3388-3397.

18. Subramanian S, Tawakol A, Burdo TH, Abbara S, Wei J, Vijayakumar J, Corsini E, Abdelbaky A, Zanni MV, Hoffmann U, Williams KC, Lo J, Grinspoon SK. Arterial inflammation in patients with HIV. JAMA. 2012;308:379-386.

19. Heijink DM, Kleibeuker JH, Nagengast WB, Oosterhuis D, Brouwers AH, Koornstra JJ, de Jong S, de Vries EG. Total abdominal 18F-FDG uptake reflects intestinal adenoma burden in Apc mutant mice. J Nucl Med. 2011;52:431-436. 20. Nosaka M, Ishida Y, Kimura A, Kondo T. Time-dependent appearance of intrathrombus neutrophils and macrophages in a stasis-induced deep vein thrombosis model and its application to thrombus age determination. Int J Legal Med. 2009;123:235-240.

21. Deatrick KB, Eliason JL, Lynch EM, Moore AJ, Dewyer NA, Varma MR, Pearce CG, Upchurch GR, Wakefield TW, Henke PK. Vein wall remodeling after deep vein thrombosis involves matrix metalloproteinases and late fibrosis in a mouse model. J Vasc Surg. 2005;42:140-148.

2003;38:1090-1098.

16. Eliason JL, Hannawa KK, Ailawadi G, Sinha I, Ford JW, Deogracias MP, RoRoelelelofofofs s KJKJKJ, aWoodrum DT, Ennis TL, Henke PK, Stanley JC, Thompson RW, Upchurch GR, Jr. Neutrophildepletion inhibits experimental abdominal aortic aneurysm formation. CiC rculation. 200050505;1;1;1121212:2:2:2323232-22404040.

1177. HHenke PKK, PPeP aaarcecee CCCGGG, MMMooaoaveveeninini DDDMMM, MMMooore AJAJ, LyLyLyncnchh h EMEMEM, LLoLongnggo C,C, VVVararmamama MMM, , DeDeewywyyererr NNANA,,, Deatrickck KBBB, Uppcchhhurcchhh GGRGR, JrJrr.,., WWWaakefffieeeld TTTWW,W, HHHogogogabaa oooammm CCC, Kunununkekel SSLSL. TTaTarrggeetteedd dedeleleletitiionono ooff f CCCCCCRR22 iiimpmpmpaiairsrsr deeeeepp veveveininin tthhohombmbmbooosisisis rreeesoololututtiooonnn iinin aaa mmmououusesese mmmoododelell... J J J ImImmmumum nonool..200606;1;17777:3:3383 8-8-3339797..

1818 SuSubrbramamananiaiann SS TTawawakakolol AA BuBurdrdoo THTH AbAbbabarara SS WeWeii JJ VVijijayayakakumumarar JJ CoCorsrsininii EE



ownloaded from



19

22. Kahn SR, Shrier I, Julian JA, Ducruet T, Arsenault L, Miron MJ, Roussin A, Desmarais S, Joyal F, Kassis J, Solymoss S, Desjardins L, Lamping DL, Johri M, Ginsberg JS. Determinants and time course of the postthrombotic syndrome after acute deep venous thrombosis. Ann Intern Med. 2008;149:698-707. 23. Lecumberri R, Alfonso A, Jimenez D, Fernandez Capitan C, Prandoni P, Wells PS, Vidal G, Barillari G, Monreal M. Dynamics of case-fatalilty rates of recurrent thromboembolism and major bleeding in patients treated for venous thromboembolism. Thromb Haemost. 2013;110:834-843. 24. van Langevelde K, Tan M, Sramek A, Huisman MV, de Roos A. Magnetic resonance imaging and computed tomography developments in imaging of venous thromboembolism. JMagn Reson Imaging. 2010;32:1302-1312.

25. Kory PD, Pellecchia CM, Shiloh AL, Mayo PH, DiBello C, Koenig S. Accuracy of ultrasonography performed by critical care physicians for the diagnosis of DVT. Chest. 2011;139:538-542.

26. Geier B, Barbera L, Muth-Werthmann D, Siebers S, Ermert H, Philippou S, Mumme A. Ultrasound elastography for the age determination of venous thrombi. Evaluation in an animal model of venous thrombosis. Thromb Haemost. 2005;93:368-374. 27. Westerbeek RE, Van Rooden CJ, Tan M, Van Gils AP, Kok S, De Bats MJ, De Roos A, Huisman MV. Magnetic resonance direct thrombus imaging of the evolution of acute deep vein thrombosis of the leg. J Thromb Haemost. 2008;6:1087-1092.

28. Saha P, Andia ME, Modarai B, Blume U, Humphries J, Patel AS, Phinikaridou A, Evans CE, Mattock K, Grover SP, Ahmad A, Lyons OT, Attia RQ, Renne T, Premaratne S, Wiethoff AJ, Botnar RM, Schaeffter T, Waltham M, Smith A. Magnetic resonance T1 relaxation time of venous thrombus is determined by iron processing and predicts susceptibility to lysis. Circulation. 2013;128:729-736.

29. Bates SM, Lister-James J, Julian JA, Taillefer R, Moyer BR, Ginsberg JS. Imaging characteristics of a novel technetium Tc 99m-labeled platelet glycoprotein IIb/IIIa receptor antagonist in patients With acute deep vein thrombosis or a history of deep vein thrombosis. Arch Intern Med. 2003;163:452-456.

30. Brighton T, Janssen J, Butler SP. Aging of acute deep vein thrombosis measured by radiolabeled 99mTc-rt-PA. J Nucl Med. 2007;48:873-878. 31. Chang KJ, Zhuang H, Alavi A. Detection of chronic recurrent lower extremity deep venous thrombosis on fluorine-18 fluorodeoxyglucose positron emission tomography. Clin Nucl Med. 2000;25:838-839. 32. Rondina MT, Lam UT, Pendleton RC, Kraiss LW, Wanner N, Zimmerman GA, Hoffman JM, Hanrahan C, Boucher K, Christian PE, Butterfield RI, Morton KA. (18)F-FDG PET in the

26. Geier B, Barbera L, Muth-Werthmann D, Siebers S, Ermert H, Philippou S, MuMuMummmmmeee A.A.A. Ultrasound elastography for the age determination of venous thrombi. Evaluatioon n ininin aann n anananimimimalalal model of venous thrombosis. Thromb Haemost. 2005;93:368-374.

277. . WeWeWe tststerererbebebeeek RRREE,E, Van Rooden CJ, Tan M, Vananan GGGils AP, Kok S,S,S De e BBaBats MJ, De Roos A, HuHuHuisissman MVMVV. MaMaagngneteticicic rrresese ononananancecece ddirirecect t ththrorombmbuusus immagagaginining g g off ttthehehe eevovolululutit onon ooofff acaca ututee dededeepepp vvveiein hhhrooombm osis of f ththhee lleg.g. J J ThThThrororombmbm HHHaaeaemmomosst. 22200008;666:1110887-7-7-1010929292.

2888. SaSaSahahah PP,, AnAnA didiia MEMEE, MMoModadaararaiii B,B,B, BBBluluumememe UUU, HuHuH mmmpphphrrrieseses JJJ,, , PaPaPatetel l ASASAS, , PhPhPhininikikkararrididououou AA,, Evvvananss CCECEmmMaattttocockk K,K, Grorover r SPSP, AhAhmamadd AA, LLyonsns OOT,T, AAttttia RRQ,Q RRenennene TT, PrPrememarrata nene SS, WiWietethooffff AAJ, Botnar RM,, SSSchchchaeaeeffffffteteer rr T,TT WWWalaa thhhamamam MMM,, , SmSmSmititith h h A.A.A MMMagagagneneetitit ccc rereresososonanan ncncnceee T1T1T1 rrrelelelaxaxxatatatioioion n n titit mem of vevenonousus tthrhromombubuss isis ddeteterermiminenedd byby iiroronn prprococesessisingng aandnd ppreredidictctss sususcscepeptitibibililityty ttoo lylysisiss



ownloaded from



20

evaluation of acuity of deep vein thrombosis. Clin Nucl Med. 2012;37:1139-1145. 33. Kikuchi M, Yamamoto E, Shiomi Y, Nakamoto Y, Fujiwara K, Watanabe F, Shinohara S. Case report: internal and external jugular vein thrombosis with marked accumulation of FDG. BrJ Radiol. 2004;77:888-890. 34. Enden T, Haig Y, Klow NE, Slagsvold CE, Sandvik L, Ghanima W, Hafsahl G, Holme PA, Holmen LO, Njaastad AM, Sandbaek G, Sandset PM. Long-term outcome after additional catheter-directed thrombolysis versus standard treatment for acute iliofemoral deep vein thrombosis (the CaVenT study): a randomised controlled trial. Lancet. 2012;379:31-38. 35. Vedantham S, Goldhaber SZ, Kahn SR, Julian J, Magnuson E, Jaff MR, Murphy TP, Cohen DJ, Comerota AJ, Gornik HL, Razavi MK, Lewis L, Kearon C. Rationale and design of the ATTRACT Study: a multicenter randomized trial to evaluate pharmacomechanical catheter-directed thrombolysis for the prevention of postthrombotic syndrome in patients with proximal deep vein thrombosis. Am Heart J. 2013;165:523-530 e523. 36. Mewissen MW, Seabrook GR, Meissner MH, Cynamon J, Labropoulos N, Haughton SH. Catheter-directed thrombolysis for lower extremity deep venous thrombosis: report of a national multicenter registry. Radiology. 1999;211:39-49. 37. Tawakol A, Migrino RQ, Hoffmann U, Abbara S, Houser S, Gewirtz H, Muller JE, Brady TJ, Fischman AJ. Noninvasive in vivo measurement of vascular inflammation with F-18 fluorodeoxyglucose positron emission tomography. J Nucl Cardiol. 2005;12:294-301. 38. Borregaard N, Herlin T. Energy metabolism of human neutrophils during phagocytosis. JClin Invest. 1982;70:550-557. 39. Rodriguez-Prados JC, Traves PG, Cuenca J, Rico D, Aragones J, Martin-Sanz P, Cascante M, Bosca L. Substrate fate in activated macrophages: a comparison between innate, classic, and alternative activation. J Immunol. 2010;185:605-614. 40. Ogawa M, Ishino S, Mukai T, Asano D, Teramoto N, Watabe H, Kudomi N, Shiomi M, Magata Y, Iida H, Saji H. (18)F-FDG accumulation in atherosclerotic plaques: immunohistochemical and PET imaging study. J Nucl Med. 2004;45:1245-1250. 41. Vucic E, Dickson SD, Calcagno C, Rudd JH, Moshier E, Hayashi K, Mounessa JS, Roytman M, Moon MJ, Lin J, Tsimikas S, Fisher EA, Nicolay K, Fuster V, Fayad ZA. Pioglitazone modulates vascular inflammation in atherosclerotic rabbits noninvasive assessment with FDG-PET-CT and dynamic contrast-enhanced MR imaging. JACC. Cardiovascular imaging. 2011;4:1100-1109. 42. Hyafil F, Cornily JC, Rudd JH, Machac J, Feldman LJ, Fayad ZA. Quantification of inflammation within rabbit atherosclerotic plaques using the macrophage-specific CT contrast agent N1177: a comparison with 18F-FDG PET/CT and histology. J Nucl Med. 2009;50:959-965.

36. Mewissen MW, Seabrook GR, Meissner MH, Cynamon J, Labropoulos N, Haaugugughthth onon SSH.H. Catheter-directed thrombolysis for lower extremity deep venous thrombosis: repopoortrt off f aa a nananatititiononal multicenter registry. Radiology. 1999;211:39-49.

37. Tawakol A, Miggrino RQ, Hoffmann U, Abbara S, Houser S, Gewirtz H, Muller JE, Brady ffTJJ,, FiFiFiscscschmhmhmananan AAJ.J.J NNNoninvasive in vivo measuremmemennnt of vascular iinfnn laammmmmmation with F-18flfluouuorrroded oxygyggluucocooseses pppososositittrororon n ememmisisissisionon ttomomogoggraraphphhyy. J NuNuuclclcl CCarrdididioolol.. 2020050505;12:2:2929294-4-4 30301.1.

38388. BBoB rregaaaardr NNN, Herrrliinn TT. EEnEnerrgyyy mmetetaabollisssm oofof hhumummananan nnneeeutrropopphils duururininng g php agggooccyyttoossis. JJClCllininin IIInvnvn esesttt.. 1911982822;7770:0:555500-0-55557.7.7

39. Rodrigueez-z-z PrPrP adadadososo JJC,CC TTTrararaveees s s PGPGG, CuCuCuenenencacaca JJJ, , RiRiRicococo DDD,,, ArArAragagagonononese JJJ, , MaMaMartrtrtininin-S-SSanananz zz P,P,P, CCascante MM BBososcaca LL SuSubsbstrtratatee fafatete iinn acactitivavatetedd mamacrcropophahageges:s: aa ccomompaparirisosonn bebetwtweeeenn ininnanatete clclasassisicc aandnd



ownloaded from



21

43. Hag AM, Pedersen SF, Christoffersen C, Binderup T, Jensen MM, Jorgensen JT, Skovgaard D, Ripa RS, Kjaer A. (18)F-FDG PET imaging of murine atherosclerosis: association with gene expression of key molecular markers. PLoS One. 2012;7:e50908. 44. Wakefield TW, Linn MJ, Henke PK, Kadell AM, Wilke CA, Wrobleski SK, Sarkar M, Burdick MD, Myers DD, Strieter RM. Neovascularization during venous thrombosis organization: a preliminary study. J Vasc Surg. 1999;30:885-892. 45. Matsui T, Nakata N, Nagai S, Nakatani A, Takahashi M, Momose T, Ohtomo K, Koyasu S. Inflammatory cytokines and hypoxia contribute to 18F-FDG uptake by cells involved in pannus formation in rheumatoid arthritis. J Nucl Med. 2009;50:920-926. 46. Roumen-Klappe EM, Janssen MC, Van Rossum J, Holewijn S, Van Bokhoven MM, Kaasjager K, Wollersheim H, Den Heijer M. Inflammation in deep vein thrombosis and the development of post-thrombotic syndrome: a prospective study. J Thromb Haemost. 2009;7:582-587. 47. Bouman AC, Smits JJ, Ten Cate H, Ten Cate-Hoek AJ. Markers of coagulation, fibrinolysis and inflammation in relation to post-thrombotic syndrome. J Thromb Haemost. 2012;10:1532-1538. Figure Legends:

Figure 1. Induction of stasis-induced DVT model in the murine jugular vein. (A) Surgical

ligation of the right jugular vein was performed to induce deep vein thrombosis (DVT). As a

sham control, the left jugular vein was surgically exposed and loosely tied without constriction.

(B) Contrast-enhanced computed tomography (CT) venography shows a filling defect only in the

right jugular vein with DVT (dark signal, yellow arrow). Surgery-induced air artifact was

observed in front of ligated jugular vein (asterisk). (C) Resected red-white thrombus in the

ligated jugular vein. (D) Hematoxylin-eosin stain. (E) Carstairs’ stain (red = fibrin, yellow =

erythrocytes, blue = collagen). Representative DVT images at day 0 (A) and day 4 (B-E). Scale

bars, 500 μm.

47. Bouman AC, Smits JJ, Ten Cate H, Ten Cate-Hoek AJ. Markers of coagulationonn,,, fifif brbrininololysysy is and inflammation in relation to post-thrombotic syndrome. J Thromb Haemost. 2202012121 ;110:0:0:151515323232--1538.

FiFiiguuure Leggenenendsds:::

FiFigugugurerere 11.. InInIndudductcttioion n ooff sstatasis s-s--inindududucececed d DVDVDVTTT mmmodododeell iinn tththe ee mmumuririinenen jjjugugugululu aaar vvveieinnn. (A(A(A))) SSSurgrgrgicccalal

igation of thhe e e ririighghght t t jujujugugugulaar r r veveveinnn wwwasasas pperererfofoformrmrmededed ttto o o ininnduducecece dddeeeepepep vvveie n n n thththrororombmbmbososisisis (((DVDVDVT)TT . As a



ownloaded from



22

Figure 2. FDG-PET enhances acute DVT in mice. Representative images of (A) FDG-PET (B)

contrast-enhanced CT venography, and (C) fused PET/CT at various timepoints. DVT (yellow

arrow) induced filling defects in contrast CT venography not seen in the contralateral sham-

operated vein (white arrow). Surgery-induced air artifact (dark signal) is observed anterior to

each vein on CT (white asterisk). FDG-PET DVT signal was elevated in acute DVT, and

diminished over time. Mild surgery-induced inflammation at wound healing region was observed

in the sham surgery left jugular vein region at day 2.

Figure 3. Time-course quantitative changes of FDG signal in DVT. (A) Standard uptake value

(SUV), (B) SUVmax , (C) target-to-background ratio (TBR), and (D) %injected dose per gram of

tissue (%IDGT) showed a similar trend (p<0.05) in that values were highest at day 2 and

significantly decreased over time (4-5 mice per group). (E and F) Subset of mice underwent

serial PET imaging at day 2, 7, and 14. FDG accumulation in DVT (yellow arrow, sham control

white arrow) demonstrated a time-dependent significant decrease (p=0.03). *p<0.05, **p<0.01,

***p<0.001. (Box-and-Whisker plot) Middle line represents median value, box indicates

interquartile range (25th–75th percentiles), and range bars show maximum and minimum.

Figure 4. Recruitment of inflammatory cells into DVT. (A) Representative immunostaining of

neutrophil and macrophage from various DVT timepoints. Neutrophils are abundant and

predominate in early day 2-4 DVT. Thrombus macrophages were evident from day 7 and resided

at the outer DVT edge. (B) The number of neutrophils (black) and macrophages (white) per 5

HPF (high power field) were shown. (*p<0.0001). Scale bar, 500 m. n.s.=not significant.

SUV), (B) SUVmax , (C) target-to-background ratio (TBR), and (D) %injected dodooseee pperere gggrararam m m oof

issue (%IDGT) showed a similar trend (p<0.05) in that values were highest at day 2 and

iigngnnififificicicananntltltly yy ddecrcrreeaeassed over time (4-5 mice per gggrorouup). (E and F)) SSubbsesesett t ofo mice underwent

eeriiaal PET imamagigiginggg aat t dadaay y 2,2,2, 777, ananndd d 1414. FFDGGG accumumumullatattioionnn ininn DDDVTVT ((yyeelllowoww aaarrrrowoww, shshshamamam ccoononttrtrol

whwhhititi e ee arara rorow)w)w) ddememmoononststtraratetet d d a a titimememe-d-d-depepenenendededennnt sssigigi nnniffificacaantntt dddeececrereeasasa ee (p(p(p=0=0=0.0003)3). *p*p*p<0<0.0.0.055,5, ***ppp<0<0.0.01,1,1 t

***p<0.001.. (((BBBoxoxx-a-andnd-WhWhWhisisiskekek rrr pp MMididdldlee lilinene pplolot)t)rr rerereprprp esessenenentststs mememedidd ananan tesvvalalueue, , boboxx inindidicacates



ownloaded from



23

Figure 5. FDG signal associations with thrombus neutrophils and macrophages. (A) The

number of thrombus neutrophils correlated with the FDG uptake in DVT (SUV; r=0.41, p=0.004,

TBR; r=0.62, p<0.001). (B) Thrombus macrophages pooled from day 7 and day 14 did not

significantly correlate with the FDG-DVT signal (SUV; r=-0.15, p=0.47, TBR; r=0.26, p=0.23).

(C) Sub-analysis of macrophages at day 14 also did not significantly correlate with FDG-DVT

signals (SUV; r=0.04, p=0.91, TBR; r=0.30, p=0.37). MAC= macrophage.

Figure 6. Experimental neutropenia reduces FDG signal within DVT. (A) Representative

images of contrast-enhanced CT and fused PET/CT from neutropenic mice (left) and uninjected

control (right) are shown. DVT (yellow arrow) and sham-operated jugular vein (white arrow) are

indicated by arrows. (B) SUV, SUVmax, and TBR of FDG signal of DVT in neutropenic mice are

significantly lower than control mice (*p<0.05). N=4 per group. Median SUV and SUVmax value

of sham-operated contralateral jugular veins are shown as respective dotted lines.

Figure 7. Establishment of a novel recurrent DVT model, and specific detection of recurrent

DVT by FDG-PET/CT. (A) Two days after the initial ligation, the suture was de-ligated and

removed. At day 14, re-ligation was performed at previously ligated site to induce a recurrent

DVT. Histological images of resected recurrent DVT (day 2 recurrent DVT (outlined with black

dotted line) overlying the day 16 DVT (outlined with blue dotted line)) are shown. (B)

Hematoxylin-eosin, (C) Carstairs’ staining, and (D) Masson Trichrome staining shows red blood

cell-rich zones in recurrent DVT (red in H&E and yellow in Carstairs’) and collagen-rich zones

in older DVT (pink in H&E and blue in Carstairs’ and Masson). (E and F) Immunostaining of

neutrophils (E) and macrophages (F) shows the newly formed recurrent DVT is neutrophil-rich,

control (right) are shown. DVT (yellow arrow) and sham-operated jugular vein ((whwhwhititee arararrororow)ww) aare

ndicated by arrows. (B) SUV, SUVmax, and TBR of FDG signal of DVT in neutropenic mice are

iigngnnififificicicananntltltlyyy llowewewerrr tht an control mice (*p<0.05). NN==4 per group. MMMeddiaiaiann n SUV and SUVmax value

off ssshham-operatatededed cccononttralalalatataterereralalal jjuguggululularar vveeinsss aaare sshhooownnn aasas rrressspepeccctiivve e dodod ttttededd lllininineses.

Figure 7. Eststababablilishshshmemementntnt oof f f a a a nonoovevevell rerer cucuurrrrrenenent t DVDVDVT T T momom dededel,l,l, aaandndnd ssspepp cicicififific c c dededeteteectctioioion n n ofofof rrrece urrent



ownloaded from



24

while the older primary DVT shows macrophage predominance. Representative (G) PET, (H)

CT, and (I) PET/CT images are shown. FDG enhanced recurrent DVT substantially more than

older DVT (G and I). (J-L) FDG signal was significantly higher in recurrent DVT (day 2

overlying the day 16 DVT) than in older day 14 DVT in the contralateral vein. N=6 per group.

Scale bar, 500 μm.

Figure 8. 18F-FDG DVT signal is elevated in patients and diminishes over time. Representative

PET/CT images from (A) a patient with DVT and (B) a matched control patient without DVT.

Elevated FDG signal was observed in the thrombosed femoral vein (A, yellow arrow). (C) The

SUVmax and (D) TBR of the DVT exhibited a time-dependent decrease (p=0.002 for SUVmax,

p=0.004 for TBR, n=6-7 per DVT group). FDG uptake in the DVT vein was shown for

individuals grouped according to age of the DVT, divided into tertiles. Values from the matched

vein from control patients are shown in the No DVT group (n=19 patients).

SUVmax and (D) TBR of the DVT exhibited a time-dependent decrease (p=0.0022 ffororr SSSUVUVUVmaxmaxax, ,

p=0.004 for TBR, n=6-7 per DVT group). FDG uptake in the DVT vein was shown for

nndidiivivividududualalalsss grgrrouupepeped d according to age of the DVTVTT,, dddivided into tererrttiless. VVaValues from the matched

vveinnn from contntrorool ppapatiitienenntststs aaarerere sshohownwnwn iinn tthe NNNooo DVVVTT grgrroououpp p (nnn=1=199 papatitiieenentsts).).



ownloaded from


Figure 1



ownloaded from


Figure 2



ownloaded from


Figure 3



ownloaded from


Figure 4



ownloaded from


Figure 5



ownloaded from


Figure 6



ownloaded from


Figure 7



ownloaded from


Figure 8



ownloaded from


Peter K. Henke, Ralph Weissleder and Farouc A. JafferMegan H. MacNabb, Anna-Liisa Brownell, Kimmo Jokivarsi, Chase W. Kessinger, Michael R. Jaff,

Tetsuya Hara, Jessica Truelove, Ahmed Tawakol, Gregory R. Wojtkiewicz, William J. Hucker,Illuminating Recently Formed, Neutrophil-Rich Thrombus

FDG-PET/CT Enables the Detection of Recurrent Same-Site Deep Vein Thrombosis by

Print ISSN: 0009-7322. Online ISSN: 1524-4539 Copyright © 2014 American Heart Association, Inc. All rights reserved.

is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231Circulation published online July 28, 2014;Circulation.

http://circ.ahajournals.org/content/early/2014/07/28/CIRCULATIONAHA.114.008902World Wide Web at:

The online version of this article, along with updated information and services, is located on the

http://circ.ahajournals.org/content/suppl/2014/07/28/CIRCULATIONAHA.114.008902.DC1Data Supplement (unedited) at:

http://circ.ahajournals.org//subscriptions/

is online at: Circulation Information about subscribing to Subscriptions:

http://www.lww.com/reprints Information about reprints can be found online at: Reprints:

document. Permissions and Rights Question and Answer available in the

Permissions in the middle column of the Web page under Services. Further information about this process isOnce the online version of the published article for which permission is being requested is located, click Request

can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.Circulation Requests for permissions to reproduce figures, tables, or portions of articles originally published inPermissions:



ownloaded from

http://circ.ahajournals.org/content/early/2014/07/28/CIRCULATIONAHA.114.008902

http://circ.ahajournals.org/content/suppl/2014/07/28/CIRCULATIONAHA.114.008902.DC1

http://www.ahajournals.org/site/rights/

http://www.lww.com/reprints

http://circ.ahajournals.org//subscriptions/


1

SUPPLEMENTAL MATERIAL

Supplemental Methods

Ex vivo gamma counting and histopathology

After sacrifice, mice were perfused with 0.9% saline via the left ventricle. Radioactivity

of excised jugular veins was measured by a gamma radiation counter (Wizard,

PerkinElmer). In a subset of resected DVT (day 2 timepoint, n=5), the vein wall and

thrombus were gently separated followed by gamma radioactivity measurements. Next,

jugular veins were fixed overnight with paraformaldehyde (PFA) and embedded in

optimal cutting temperature compound (Sakura Finetek, Torrance, CA). Serial 6-µm

cryostat sections were obtained for H&E, Carstairs’ fibrin staining, Masson trichrome,

and immunohistochemistry. Immunohistochemical detection of thrombus neutrophils

(NIMP-R14, Santa Cruz Biotechnology, TX), macrophages (CD68, AbD Serotec,

Oxford, UK), iNOS (Abcam, MA), and glucose transporter protein type 1 (GLUT-1,

Santa Cruz) was performed. The number of thrombus neutrophils and macrophages per

5 high power fields (HPF; 1000x) was quantified as previously reported.17 As a positive

controls for GLUT-1 and iNOS expression, we stained macrophage-rich atheroma

sections from the aortic sinus of cholesterol-fed ApoE knockout mice.

Protein extraction and immunoblotting

Resected tissues were homogenized in 100-200 µl of RIPA buffer (Sigma-Aldrich)

supplemented with protease inhibitor. Extracted proteins were separated by SDS PAGE

2

and transferred to PVDF membranes. Anti- matrix metalloproteinase-9 (MMP-9)

antibody (Abcam) and anti-beta actin antibody (Sigma) were used.

Clinical FDG-PET study. The Partners Institutional Review Board approved access to

patient records and analysis of patient data (Protocol #: 2011-P-000521). Participants

for the clinical imaging sub-study were consecutively identified from a database of

patients who had undergone 18-fluorine-fluorodeoxyglucose-positron emission

tomography (18F-FDG-PET) and computed tomography (CT) imaging for oncological

evaluation at the Massachusetts General Hospital. We used the clinical research

database to generate an initial cohort of patients (n=437) with the following inclusion

criteria: age >18, diagnosis of DVT by billing code, and receiving an 18F-FDG-PET and

CT scan between 2004 through 2011 at Massachusetts General Hospital. Exclusion

criteria included: deep vein thrombosis (DVT) outside the area of PET imaging, a DVT

outside of the iliac-femoral veins, or a confounding reason for inflammation around the

site of DVT (i.e. recent catheterization, vasculitis, nearby infection or tumor). This cohort

was then restricted to patients with DVT occurring within 6 months after PET, leaving a

total of 20 subjects. In one of these patients however an appropriate matched control

could not be identified; therefore 19 patients, and 19 matched controls were included in

the final analysis. Age of the DVT was determined as accurately as possible by clinical

(n=17) and imaging evidence (n=2). Control subjects (n=19) without DVT who had

undergone FDG imaging for clinical purposes were consecutively identified and

matched 1:1 by gender, age (±10 years), active cancer and chemotherapy treatment

within six months of imaging, and steroid immunosuppressive therapy at the time of

3

PET imaging, if present (Supplemental Table). Additionally, individuals with DVT were

grouped according to the clinical age of their DVTs at time of imaging (divided into

tertiles). Thereafter, the group-mean FDG signal was compared to the age of the DVT

using linear regression analysis.

PET/CT protocol. Whole body FDG-PET imaging was performed per clinical protocol

using a Biograph 64 Scanner (Siemens, Forcheim Germany) or similar system. FDG

was administered ~370 MBq (~10 mCi) intravenously after an overnight fast. PET

images were acquired in 3D mode ~60 minutes later. Patients were imaged in the

supine position and images were obtained over 15-20 minutes. A low-dose, non-gated,

non-contrast enhanced CT (120 keV, 50 mAs) was employed for attenuation correction

prior to the PET scan. FDG-PET/CT image analysis. PET-CT images were analyzed

by an investigator blinded to the patients’ clinical information. 18F-FDG uptake was

measured within the right and left legs of both control and DVT subjects to assess

uptake from the common femoral vein through the external iliac vein. An individual

region-of-interest (ROI) was placed around the area just superior to the popliteal vein

and extended to the inferior of the iliac bifurcation in order to obtain a maximum

standardized uptake value (SUVmax). The average venous blood uptake of the right

atrium was used to derive a target-to-background ratio (TBR) from the SUVmax.1

Patients were then analyzed to assess the time-dependent changes in FDG uptake in

DVT and corresponding vein of matched control patients without a DVT.

Statistics

4

For animal data, results are expressed as median [25%-75% quartiles]. Statistical

comparisons between two groups were evaluated by the Mann-Whitney U test, and by

the Kruskal-Wallis test for multiple groups followed by the Dunn’s post-test. For

comparison between two groups within the same animal, the Wilcoxon matched-pairs

signed-rank test was used. Continuous variables at multiple timepoints were compared

by the Friedman test. FDG uptake measurements were correlated with histological

findings by use of the two-tailed Spearman method. Statistical comparisons were

performed with GraphPad Prism (La Jolla, CA). Mice that did not develop jugular DVT

(histologically negative, n=3) were excluded from analyses. For clinical data, values are

expressed as mean±SEM. Wilcoxon signed ranks test was used for single comparisons,

and after confirming normality of distribution, linear regression analysis was used to

determine the association between FDG uptake and DVT age (SPSS 22, IBM, Chicago,

IL). In the multivariable models, FDG uptake within the DVT was entered as the

dependent variable, while DVT age was entered as an independent variable.

Subsequently, clinical factors were added to the model as additional independent

variables to assess their potential impact on the relationship between DVT age and

FDG uptake. In order to limit the number of variables entered into the regression model

at one time (given the modest sample size), the clinical variables were added to the

multivariable model in three separate groupings: (1) demographic factors (patient age

and gender); (2) factors that might impact systemic inflammation (immunosuppressive

steroid use, history of recent infection, or statin use); and (3) oncologic history (history

of cancer, cancer subtype, and history of chemotherapy). A value of P<0.05 was

considered statistically significant.

5

SUPPLEMENTAL TABLE

All Control

Subjects

All Subjects

with DVT

DVT (<17

days)

DVT (18-58

days)

DVT

(>59days)

Number of Patients 19 19 7 6 6

Age (years) 62±15.1 63.2±15.8 71.0±12.6 53.7±19.7 65.2±14.8

Male 10 10 3 5 2

BMI (kg/m2) 26.6±4.7 26.5±5.6 28.6±3.1 23.9±2.5 28.7±7.9

Malignancy 18 18 6 6 6

gastrointestinal 5 5 1 2 2

lung 5 4 2 1 1

hematologic 3 4 1 2 1

others 5 5 2 1 2

Chemotherapy 18 18 6 6 6

Statin 4 4 1 2 1

Steroid 1 1 1 0 0

Infection 3 5 2 2 1

Supplemental Table. Clinical characteristics of patients in the FDG-PET/CT study.

Values are presented as mean±SD. BMI=body mass index; DVT= deep venous

thrombosis.

6

SUPPLEMENTAL FIGURES and FIGURE LEGENDS

Supplemental figure S1

Supplemental figure S1. Schematic experimental design. (A) Mice underwent FDG-

PET imaging at various timepoints, followed by ex vivo analyses and histology. (B)

Serial FDG-PET imaging at day 2, 7, and 14 was performed in a subset of three mice.

Mice were sacrificed after the final PET imaging at day 14. (C) To assess the role of

neutrophils in FDG signal, antibody-based neutrophil depletion was performed in subset

of four mice before DVT formation and imaging. (D) For the recurrent DVT model (n=6

mice), we removed the suture of the ligation at day 2 to spur recanalization of the

thrombosed right jugular vein, followed by re-ligation 12 days later to induce recurrent

same-site stasis-induced DVT. We also performed complete ligation of the left

contralateral jugular vein at day 2 to enhance recanalization of the initial right jugular

DVT, and to compare a fresh recurrent DVT and an older DVT in the same mouse.

7

Supplemental figure S2.

Supplemental figure S2. The time-dependent decrease in matrix metalloproteinase-9

(MMP-9) expression in DVT parallels the time-dependent decrease in the in vivo DVT

FDG signal. (A and B) Immunoblotting of MMP-9 demonstrates time-dependent

decrease. (C) DVTs from neutropenic mice shows decreased MMP-9 expression.

8

Supplemental figure S3.

Supplemental figure S3. Little glucose transporter protein type 1 (GLUT-1) and iNOS

expression is present in macrophages within day 14 DVT (top row, asterisk).

Representative images of immunostaining (CD68-macrophage, iNOS-M1 macrophage

marker, and GLUT-1) of day 14 DVT (top row), as well as of an aortic plaque on an

ApoE knock-out mice showing higher expression of GLUT-1 and iNOS (bottom row,

arrowhead). Asterisk=DVT. Scale bar, 100 µm.

SUPPLEMENTAL REFERENCE

1. Subramanian S, Tawakol A, Burdo TH, et al. Arterial Inflammation in Patients with HIV. JAMA. 2012;308:379-86.

fdg-pet/ct enables the detection of recurrent same-site deep

Documents