These data we* CorrespondE-mail addr

http://dx.doi.o1760-2734/ª 2

Journal of Veterinary Cardiology (2015) 17, S306eS317

www.elsevier.com/locate/jvc

Secondary prevention of cardiogenicarterial thromboembolism in the cat:the double-blind, randomized,positive-controlled feline arterialthromboembolism; clopidogrel vs. aspirintrial (FAT CAT)

Daniel F. Hogan, DVM a,*, Philip R. Fox, DVM, MS b,Kristin Jacob, DVM c, Bruce Keene, DVM, MS d,Nancy J. Laste, DVM e, Steven Rosenthal, DVM c,Kimberly Sederquist, RVT, VTS-Cardiology f,Hsin-Yi Weng, BVM, MPH, PhD g

a Purdue University, College of Veterinary Medicine, Department of Veterinary ClinicalSciences, 625 Harrison Street, West Lafayette, IN 47907-2026, USAbCaspary Institute, The Animal Medical Center, 510 East 62nd Street, New York, NY10065, USAcCVCA-Cardiac Care for Pets, 1209 Cromwell Bridge Road, Towson, MD 21286, USAdNorth Carolina State University, College of Veterinary Medicine, Department ofClinical Sciences, 1060 William Moore Drive, Raleigh, NC 27607, USAeMSPCA-Angell Animal Medical Center, 350 South Huntington Ave., Boston, MA 02130,USAf Purdue University, College of Veterinary Medicine, Veterinary Teaching Hospital, 625Harrison Street, West Lafayette, IN 47907-2026, USAg Purdue University, College of Veterinary Medicine, Department of ComparativePathobiology, 625 Harrison Street, West Lafayette, IN 47907-2026, USA

Received 17 November 2014; received in revised form 9 September 2015; accepted 7 October 2015

re presented at the American College of Veterinary Internal Medicine Scientific Forum, 2013.ing author.ess: hogandf@purdue.edu (D.F. Hogan).

rg/10.1016/j.jvc.2015.10.004015 Elsevier B.V. All rights reserved.

FAT CAT study

KEYWORDSThromboprophylaxis;Thrombosis;Antithrombotics;Infarction;Cardioembolic

Abbreviations

ADP adenosine diphosphALP alkaline phosphatasALT alanine aminotransfCATE cardiogenic arterialCHF congestive heart faiDCM dilated cardiomyopaGGT gamma-glutamyl traHCM hypertrophic cardioHOCM hypertrophic

cardiomyopathyRCM restrictive cardiomySAM systolic anterior mo

valveUCM unclassified cardiom

Abstract Objectives: To determine if clopidogrel administration is associatedwith a reduced likelihood of recurrent cardiogenic arterial thromboembolism(CATE) in cats compared to aspirin administration. Secondary aims were to deter-mine if clopidogrel administration had an effect on the composite endpoint of re-current CATE and cardiac death and to identify adverse effects of chronicclopidogrel or aspirin therapy.Animals: Seventy-five cats that survived a CATE event.Methods: Multicenter, double-blind, randomized, positive-controlled study. Catswere assigned to clopidogrel (18.75 mg/cat PO q 24 h) or aspirin (81 mg/cat PO q72 h). KaplaneMeier survival curves were created for each endpoint and the logrank test performed to compare treatment groups with respect to time to eventand the likelihood of the event occurring.Results: The mean age of all cats was 8.0 � 3.5 yr and 57/75 (76%) were male(p < 0.001); 62/75 (83%) were mixed breed with the remainder including Persian,Abyssinian, American Shorthair, Bengal, Birman, Himalayan, Maine Coon, Ragdoll,Snowshoe, and Sphynx breeds. Only 15% (11/75) of cats had a history of heart dis-ease recorded prior to the CATE event. Clopidogrel administration was associatedwith significantly reduced likelihood of recurrent CATE compared to aspirin(p ¼ 0.024) and had a longer median time to recurrence [443 (95% CI 185e990) daysvs. 192 (95% CI 62e364) days, respectively]. Clopidogrel was also associated with asignificantly reduced likelihood of the composite endpoint of recurrent CATE or car-diac death (p ¼ 0.033) with a longer median time to event [346 (95% CI 146e495)days vs. 128 (95% CI 58e243) days].Conclusions: Clopidogrel administration significantly reduces the likelihood of re-current CATE compared with aspirin in cats; both drugs were well tolerated.ª 2015 Elsevier B.V. All rights reserved.

S307

ateeerasethromboembolismlurethynspeptidasemyopathy

obstructive

opathytion of the mitral

yopathy

Introduction

Cardiogenic arterial thromboembolism (CATE) is awell-recognized and devastating clinical morbiditywith feline cardiomyopathy. When fragments of anintracavitary thrombus gain access to the systemiccirculation, they may obstruct distant arterialsegments which can result in tissue injury or organ

infarction. Distal aortic occlusion or “saddlethrombus” occurs in approximately 90% of cases1

while right brachial, renal, splanchnic, and cere-bral infarction occur less commonly.1,2 Theunderlying mechanisms of CATE are distinctly dif-ferent from arterial thrombosis with the mostappropriate human corollary being cardioembolicdisease secondary to atrial fibrillation.3,4 Previousretrospective studies have reported that 6%e17%5e7 of cats with underlying cardiac disease goon to develop CATE with mortality rates of 61%e67%.1,8e10 Euthanasia, similar in frequency tonatural death, is often selected in part because oflack of evidence for management strategies thatmight prevent recurrent CATE.

Antithrombotic drugs are the standard of carefor cardioembolic prevention in humans11,12 andthis has been incorporated into clinical protocolsfor cats. However, there have been no prospectiveclinical trials in cats evaluating antithrombotictherapy for primary (initial CATE) or secondary(recurrent CATE) prevention. A number of retro-spective studies have reported recurrent CATErates of 17%e75%1,8e10,13 in cats receiving antith-rombotic therapy with one year recurrence ratesof 25%e61%.9,10 Aspirin, the most commonly usedantithrombotic in cats, is an antiplatelet drug

S308 D.F. Hogan et al.

which exerts a mild antithrombotic effect throughinhibition of cyclooxygenase with reduction ofthromboxane A2 production. Clopidogrel is anirreversible antagonist of the platelet adenosinediphosphate (ADP)2Y12 receptor which inhibits pri-mary and secondary platelet aggregation as well asinhibiting the ADP-induced conformational changeof the glycoprotein IIb/IIIa receptor complex.14,15

Clopidogrel has been shown to induce a morepotent antiplatelet effect than aspirin inhumans,16 but the comparative effect betweenclopidogrel and aspirin to limit CATE recurrencehas not been reported in the cat.

The primary objective of this study was todetermine if clopidogrel administration is asso-ciated with a reduced likelihood of recurrent CATEin cats compared to aspirin administration. Sec-ondary aims were to determine if clopidogreladministration had an effect on the compositeendpoint of recurrent CATE and cardiac death andto identify adverse effects of chronic clopidogrelor aspirin therapy. Our hypothesis was that catsreceiving clopidogrel would have a decreasedlikelihood of recurrent CATE compared to catsreceiving aspirin.

h Research Randomizer, www.randomizer.org.i PCCA, Houston, TX USA.

Animals, materials, and methods

The study protocol was approved by the IACUC at allparticipating institutions and informed client con-sent was achieved using a standard form used by allenrolling veterinarians. This double-blind,randomized, positive-controlled multicenter studyrecruited investigators from throughout the world.Enrollment was accomplished through an onlinestudy website (www.vin.com/fatcat). The studywas advertised to veterinarians and cat ownersthrough websites, conferences, and social media.Entry criteria required a complete medical historyand physical examination, arterial thromboembo-lism that developed 1e3 months previously, echo-cardiographic confirmation of underlying cardiacdisease that could support the formation of anintracardiac thrombus, clinical pathology resultsthat included complete blood count, biochemicalpanel, and, in cats >7 years of age, total serumthyroxine concentration. Exclusion criteria inclu-ded no evidence of cardiac disease on the echo-cardiogram, presence of an intracardiac thrombus,platelet count <50,000/ml, clinical evidence ofbleeding, unstable congestive heart failure orpresence of concurrent non-cardiac diseasethought to independently make one year survivalunlikely. Cats were permitted to receive anytreatment, including antithrombotic drugs, during

the 1e3 month run-in period deemed necessary bythe attending veterinarian who did not have to be aboard-certified specialist. The medical treatmentduring this run-in periodwas continued until the daythe study began which was the first day the studydrug (clopidogrel or aspirin) was administered.

Study enrollment

When a cat was enrolled through the study web-site, the principal investigator was notified and theenrolling veterinarian was contacted directly bythe principal investigator to verify that an arterialevent had likely occurred and to determine otherconcurrent medical conditions. The principalinvestigator (DFH) reviewed all cases submitted forenrollment to verify that they conformed to entrycriteria. There was no attempt to standardize theechocardiographic technique or measurementprotocol. Baseline data collection included phys-ical examination findings, complete blood count,serum chemistry panel, and echocardiogram aswell as demographic data, estimated clinicalrecovery from the CATE event (percent improve-ment from initial presentation), site of infarction,categorization of cardiac disease, and drug ther-apy prior to and following the CATE event. Theclinical pathology tests were performed by labo-ratories used by the enrolling veterinarian.

Study group allocation

Enrolled cases were randomly assignedh to receiveeither clopidogrel (18.75 mg PO q 24 h) or aspirin(81 mg tablet PO q 72 h) using an online randomallocation program where group allocation waspre-determined for 100 study animals. Only thestudy coordinator was not blinded to study groupallocation and did not participate in any studydecisions. The principal investigator, enrollingveterinarian, and cat owner were kept blinded togroup allocation through the use of color-coded,#3 opaque gelatin capsules.i Cats in the clopidog-rel group were given a solid yellow capsule everythird day which contained the 18.75 mg of clopi-dogrel along with lactose powder while the othertwo days they received a yellow and gray capsulewhich also contained the 18.75 mg of clopidogreland lactose powder. Cats in the aspirin group weregiven a solid pink capsule every third day whichcontained the 81 mg aspirin tablet along withlactose powder while the other two days theyreceived a pink and gray capsule which contained

j SPSS Statistics for Windows, Version 22.0, IBP Corp., Armonk,NY USA.

FAT CAT study S309

lactose powder only. Capsules were compoundedat one site by the study coordinator for cats fromNorth America and New Zealand while the capsuleswere compounded at another study site for catsfrom Europe. The study code and group allocationwere broken after the final data analysis wasperformed.

Concurrent cardiac therapy

There was an effort to standardize concurrentcardiac therapy based on the presence or absenceof congestive heart failure (CHF) and morphologicclassification of cardiac disease as follows. All catswith a history of CHF, or who developed CHF duringthe study, were treated with furosemide and anangiotensin-converting enzyme inhibitor. In theabsence of CHF, it was suggested that cats withhypertrophic cardiomyopathy (HCM) receive nocardiac therapy, cats with restrictive (RCM) orunclassified cardiomyopathy (UCM) receive anangiotensin-converting enzyme inhibitor, and catswith dilated cardiomyopathy (DCM) receive anangiotensin-converting enzyme inhibitor anddigoxin. The use of atenolol therapy in cats withsystolic anterior motion of the mitral valve was leftto the discretion of the veterinarian providingdirect care. No antithrombotic therapy other thanthe study drug was allowed during the studyperiod.

Study monitoring

During the study, owners completed a daily logsheet that recorded appetite, mental attitude,stool consistency, development of skin lesions, andthe presence or absence of vomiting or bleeding.Clinical follow up of the cats was done at 1 month,3 months, 6 months, and 12 months after beginningthe study. A physical exam, complete blood countand serum chemistry panel were repeated at 1month, a physical exam was repeated at 3 months,and the owners were contacted through phonecalls at 6 and 12 months and every 6 monthsthereafter. If the health status changed at a timepoint outside of these scheduled follow up visits,an unscheduled visit could be registered throughthe study website by the attending veterinarian;this entry would notify the principal investigatorwho would contact the veterinarian and together,would make a joint decision about whether the catshould continue on the study.

The primary study endpoint of recurrent CATEwas defined prospectively and determined byclinical history, physical examination and whennecessary, nuclear imaging, angiography, or

necropsy. Once the primary endpoint was reached,the cat was removed from the study. Secondaryendpoints included cardiac death or euthanasiadue to cardiovascular disease, and sudden death.In the event of sudden death, an attempt wasmade to obtain a necropsy to determine if suddendeath was due to a CATE event. In the absence of aknown CATE event, sudden death was attributed tocardiac disease. A composite endpoint of recurrentCATE or death attributed to cardiac disease wasdeveloped prospectively because cardiac deathhas been reported to be the second most commoncause of death in cats following a CATE event17 andthe investigators wanted to see if the greaterantiplatelet effects of clopidogrel compared toaspirin would have an identifiable effect on theprogression of cardiac disease. The developmentof or worsening of congestive heart failure was notan endpoint in the study. Each attending veter-inarian was permitted to add or modify CHF ther-apy as judged necessary for effective patientmanagement to allow the cat to continue on thestudy. The number of study days was calculatedfrom the first day of study drug administration to astudy endpoint. Cats that reached a secondaryendpoint were right-censored for the primaryendpoint of CATE.

Statistical analysis

Statistical analyses were performed by commer-cially available computer software.j Baselinecomparisons between the two treatment groupswere performed on the key covariates that wereknown to be associated with the outcomes ofinterest including age, weight, gender, underlyingcardiac disease category, and echocardiographicvariables. T-test or ManneWhitney test (if outlierswere identified) was used to analyze continuouscovariates while Pearson’s chi-square test orFisher’s exact test was used for categorical cova-riates, and the binomial test was used for genderproportions. Paired t-test or Wilcoxon test (ifoutliers were identified) was used to compareclinicopathological data at baseline and after 1month of study drug administration within eachtreatment group. KaplaneMeier method was usedto estimate cumulative study continuation rateand median time to event. Cox proportional haz-ards regression was used to adjust for any cova-riates found to be significantly different betweenthe two treatment groups at baseline and alsosignificant in the Cox regression. Adjusted hazard

S310 D.F. Hogan et al.

rate ratio and 95% confidence intervals werederived from the Cox regression. If no significantcovariates were found, log rank test was per-formed to compare cumulative study continuationrates between the groups. Statistical significancewas defined as p < 0.05.

An original power analysis suggested that 20cats in each group would be sufficient to identify asignificant difference between groups if the fre-quency of recurrent CATE decreased from anestimated 50% to 15% at a 95% level of confidenceand 80% power. After approximately 75% of theoriginal 40 cats had been enrolled, it was clearthat the reduction in frequency of recurrent CATEbetween groups was not as great as assumed in theoriginal power analysis. Therefore, the decisionwas made to continue the study with an aim todouble the original study population from 40 catsto 80 cats. The principal investigator, enrollingveterinarian, and cat owner were kept blinded togroup allocation until final data analysis had beencompleted.

Results

Eighty-five cats were enrolled through the studywebsite with 42 allocated to the clopidogrel groupand 43 allocated to the aspirin group (Fig. 1). Catswere enrolled at a total of 52 sites, 50 of whichwere from North America and 1 of each from NewZealand and Europe. Three cats were excludedfrom analysis in the clopidogrel group: two fornoncompliance prior to study initiation and onedue to death from recurrent CATE prior to studydrug administration. Seven cats were excludedfrom the aspirin group prior to starting the studydrug: three due to owner wishes (one deemedmedically unstable, one due to recurrent vomiting,one due to inability to administer study medi-cation), one due to death from recurrent CATE,one cat ran away, one euthanized due to non-cardiac disease, and one who died from cardiacdisease. There were 39 and 36 cats that remainedin the clopidogrel and aspirin groups, respectivelywhich were included in the final data analysis. Themedian time on study for all cats was 182 days(4e2034 days).

The demographic data for the cats enrolled inthe study are summarized in Table 1. Mean age was8.0� 3.5 yr and there was a strongmale gender bias(57/75, 76%; p < 0.001). The majority of cats wereof mixed breed (62/75; 83%) but pure breed catswere also represented including Persian (2), Hima-layan (2), Abyssinian, American Shorthair, Bengal,Birman, Maine Coon, Ragdoll, Snowshoe, and

Sphynx. Eleven out of 75 cats (15%) had a knownhistory of heart disease prior to the initiating CATEevent with 10/11 (91%) receiving cardiac drugtherapy (Table 2). Therapy included beta-blocker(4/10, 40%), angiotensin-converting enzyme inhib-itor (4/10; 40%), furosemide (3/10; 30%), aspirin (2/10; 20%), calcium channel blocker (2/10; 20%),spironolactone (2/10; 20%), clopidogrel (1/10;10%), and low molecular-weight heparin (1/10;10%); 7/10 (70%) received multiple drug therapy. Inmost cats (70/75; 93%), the initiating CATE wastheir first vascular event. However, 3/75 (4%) hadtwo prior CATE events, while 2/75 (3%) had fourprior CATE events. The majority of cats presentedfor bilateral, symmetrical pelvic limb infarction(45/75; 60%) but right thoracic limb (14/75; 19%),unilateral pelvic limb (14/75; 19%), unilateralthoracic and unilateral pelvic limb (1/75; 1%), andleft thoracic limb infarction (1/75; 1%) were alsoreported. Owner-reported percent clinical recov-ery from the CATE event was high with a mean of87.5% � 16.1% (range, 30%e100%). A number ofunderlying cardiac diseases were representedincluding HCM (46/75; 61%), UCM (12/75; 16%),hypertrophic obstructive cardiomyopathy (HOCM)(8/75; 11%), RCM (7/75; 9%), DCM (1/75; 1%), andcongenital cardiac disease (mitral valve stenosis)(1/75; 1%).

There was no significant difference in age,weight, gender, or underlying cardiac disease(HCM vs. non-HCM) between the clopidogrel andaspirin groups. Echocardiographic measurementswere recorded in 31/39 (80%) cats in the clopi-dogrel group and 28/36 (78%) cats in the aspiringroup (Table 1). There were significantly more catswith reported systolic anterior motion of themitral valve (SAM) in the clopidogrel treated group(10/31, 32%) than the aspirin treated group (2/28,7%) (p ¼ 0.024). The presence of SAM was includedas a covariate in the Cox regression model forevent with regard to the primary endpoint ofrecurrent CATE, the composite endpoint of recur-rent CATE or cardiac death, and all-cause mortal-ity. However, SAM did not have a significant effectin this model so it was not included in the finaldata analysis. The median follow up period was185 days (4e2034 days) for the clopidogrel treatedgroup and 122 days (6e883 days) for the aspirintreated group.

With regard to the primary endpoint of recur-rent CATE, clopidogrel was associated with a sig-nificantly reduced likelihood (p ¼ 0.024) andlonger median time to event than aspirin [443 (95%CI 185e990) days vs. 192 (95% CI 62e364) days,respectively] (Fig. 2). Recurrent CATE occurred in19/39 (49%) cats receiving clopidogrel with 14/39

FAT CAT study S311

(36%) occurring within the first year. In compar-ison, recurrent CATE occurred in 27/36 (75%) catsreceiving aspirin with 23/36 (64%) occurring withinthe first year. Most of the cats that experienced arecurrent CATE event did so within the first year ofstarting the study (14/19; 74% and 23/27; 85% forclopidogrel and aspirin, respectively) with amedian time to recurrence of 146 days (7e990days) for clopidogrel and 83 days (6e883 days) foraspirin. The cumulative frequency for recurrentCATE increased most rapidly over the first 6months of the study in cats receiving clopidogrelwith 8/19 (42%) cats experiencing a recurrent CATEevent within the first 3 months of the study.In comparison, the cumulative frequency forrecurrent CATE increased most rapidly over thefirst 9 months of the study in cats receiving aspirin

Figure 1 Flow diagram for animal enrollm

with 14/27 (52%) cats similarly experiencing arecurrent CATE event within the first 3 months ofthe study.

When the primary endpoint of recurrent CATEwas combined with death due to cardiac diseaseinto a composite endpoint, clopidogrel was alsoassociated with a significantly reduced likelihood(p ¼ 0.033) and longer median time to event thanaspirin [346 (95% CI 146e495) days vs. 128 (95% CI58e243) days, respectively] (Fig. 3). Death due toCHF occurred in 3/39 (8%) clopidogrel treated catsand 3/36 (8%) in aspirin treated cats. Sudden deathoccurred in 4/39 (10%) clopidogrel treated catswith three of those cats undergoing a full necropsyconfirming lack of recurrent CATE. There were 2/36 (6%) cats in the aspirin group that died sud-denly, with neither receiving a necropsy.

ent, allocation, and outcome on study.

Table 1 Demographic and echocardiographic data from cats that were enrolled onto the study.

Parameter Aspirin Clopidogrel Between groups

Age (yr) 8.5 � 3.3 7.5 � 3.6 p ¼ 0.187Weight (kg) 5.3 � 1.7 4.8 � 1.5 p ¼ 0.144Gender (MN/FN) 27/9 30/9 p ¼ 0.840Breed DSH (29), DLH (2), Abyssinian,

Bengal, Himalayan, Persian,Snowshoe

DSH (24), DLH (7), Persian (2), ASH,Birman, Himalayan, Maine Coon, Ragdoll,

SphynxCardiac disease HCM (19), RCM (7), UCM (6),

HOCM (2), DCM, congenitalHCM (27), HOCM (6), UCM (6)

Echocardiogram n ¼ 28 n ¼ 31IVSDd (mm) 6.18 � 1.45 6.63 � 1.60 p ¼ 0.257LVFWd (mm) 7.04 � 2.12 6.58 � 1.72 p ¼ 0.401LA/Ao 2.22 � 0.44 2.10 � 0.52 p ¼ 0.207LVIDd (mm) 14.21 � 2.54 15.27 � 2.78 p ¼ 0.133LVIDs (mm) 7.89 � 2.90 8.57 � 2.81 p ¼ 0.360FS (%) 44.52 � 16.67 44.52 � 11.26 p ¼ 0.998SAM (Y) 2 10 p ¼ 0.024Spontaneouscontrast (Y)

10 8 p ¼ 0.573

ASH ¼ American shorthair; DLH ¼ domestic longhair; DSH ¼ domestic shorthair; FN ¼ female neutered; FS ¼ fractional short-ening; IVSDd ¼ interventricular septal diameter at end-diastole; LA/Ao ¼ left atrial to aortic ratio; LVFWd ¼ left ventricular freewall diameter at end-diastole; LVIDd ¼ left ventricular internal diameter at end-diastole; LVIDs ¼ left ventricular internaldiameter in systole; MN ¼ male neutered; SAM ¼ systolic anterior motion of the mitral valve; Y ¼ yes; see abbreviation table forthe remainder of the abbreviations.Bold and italic data represents significantly different between the groups identified.

S312 D.F. Hogan et al.

There was no significant difference (p ¼ 0.105)in the likelihood of all-cause mortality betweenclopidogrel and aspirin treated cats; the mediansurvival time for clopidogrel was 248 days (95% CI137e431) and 128 days (95% CI 58e243) foraspirin.

There was one cat in each group that wasremoved from the study for a possible adverseeffect from study drug administration. Icterus andelevated liver enzymes developed in a clopidogreltreated cat at 40 days and persistent vomiting andhematochezia occurred in an aspirin treated cat at70 days.

Table 2 Concurrent cardiac medications before and afte

Cardiac drug Pre-CATE

Aspirin Clopid

ACE inhibitor 1Aspirin 1Atenolol 2Clopidogrel 1Digoxin 0Diltiazem 1Furosemide 0LMWH 1Spironolactone 1Multiple 2

CATE: cardiogenic arterial thromboembolism; LMWH: low molecul

Clinical pathology parameters did not differsignificantly between treatment groups at baselineor at 1 month (Table 3). Within the aspirin group,there was a significant increase in alkaline phos-phatase (ALP) and a significant decrease ingamma-glutamyl transpeptidase (GGT) at 1 monthcompared to baseline, although none of theparameters for the individual animals fell outsideof reference ranges. Total white blood cell countdecreased and serum alanine aminotransferase(ALT) increased significantly at 1 month comparedto baseline for the clopidogrel group, althoughnone were outside of the reference range. The

r enrolling embolic event.

Post-CATE

ogrel Aspirin Clopidogrel

3 18 251 36 NA2 5 70 NA 390 1 01 2 33 14 170 NA NA1 1 13 14 18

ar-weight heparin; NA: not allowable.

Figure 2 KaplaneMeier event curve for the primaryendpoint of recurrent cardiogenic arterial throm-boembolism for the aspirin (blue line) and clopidogrel(brown line) groups.

FAT CAT study S313

clopidogrel treated cat that developed icterus andelevated liver enzymes had normal ALT, ALP, andGGT values recorded 1 month after starting clopi-dogrel administration.

At enrollment, there were 3/39 (8%) cats in theclopidogrel treated group that had been previouslydiagnosed with CHF and were receiving therapywhile no cats (0%) in the aspirin group had a historyof CHF prior to the initiating CATE event (Table 2).By the end of the study, 17/39 (44%) cats in theclopidogrel group and 14/36 (39%) cats in theaspirin group had received therapy for CHF. Thedistribution of concurrent cardiac therapies wassimilar between the groups with furosemide (17/39; 44%), angiotensin-converting enzyme inhibitor(25/39; 64%), beta-blocker (7/39; 18%), calcium

Figure 3 KaplaneMeier event curve for the compositeendpoint of recurrent cardiogenic arterial throm-boembolism or cardiac death for the aspirin (blue line)and clopidogrel (brown line) groups.

channel blocker (3/39; 8%), and spironolactone (1/39; 3%) administered in the clopidogrel group whilefurosemide (14/36; 39%), angiotensin-convertingenzyme inhibitor (18/36; 50%), beta-blocker (5/36; 14%), calcium channel blocker (2/36; 6%),digoxin (1/36; 3%), and spironolactone (1/36; 3%)were administered in the aspirin group. Similarly,multiple cardiac drugs were used in 18/39 (46%) inthe clopidogrel group compared to 14/36 (39%)cats in the aspirin group.

Discussion

This is the first prospective, randomized, double-blind, positive-controlled antithrombotic trial incats with underlying cardiac disease. Clopidogreltreated cats were significantly less likely to reachthe primary endpoint of recurrent CATE(p ¼ 0.024) and had a longer median time to event(443 days vs. 192 days) than aspirin treated cats.Given the most common forms of cardiac diseaseand absence of epicardial coronary arterial diseasein cats, an antithrombotic drug would not beexpected to have a direct impact on clinical out-comes outside of recurrent thrombotic events.However, clopidogrel treated cats were also sig-nificantly less likely to reach the composite end-point of recurrent CATE or cardiac death(p ¼ 0.033) and had a longer median time to event(346 vs. 128 days) than aspirin treated cats.Approximately twice as many cats developedcongestive heart failure in the clopidogrel groupcompared to the aspirin group and this could beexplained by the fact that cats in the aspirin groupdid not live long enough to develop progressivecardiac disease while the cats receiving clopidog-rel were able to live longer, due to not having arecurrent CATE event.

Twenty-six percent (10/39) of the clopidogreltreated cats did not have a recurrent CATE event ordie from a cardiac cause with six still alive at theend of the study (400, 505, 659, 1076, 1729, and2034 days). Within the aspirin group, 11% (4/36) didnot have a recurrent CATE event or die from acardiac cause with one cat alive at the end of thestudy (715 days). The difference between groups isemphasized by the fact that 16/39 (41%) of clopi-dogrel treated cats were still on study after 1 yearcompared to 7/36 (19%) of aspirin treated cats.Published overall median survival times from ret-rospective studies of recurrent CATE with aspirintherapy range from 24 days to 149 days1,10,17 whichis similar to the median survival time in the aspiringroup within this study [128 (95% CI 58e243) days],while the median survival time for the clopidogrel

Table 3 Clinicopathological data for aspirin and clopidogrel groups at baseline and after 1 month of study drugadministration.

Aspirin Baseline 1 month Within group(baseline-1 month)

Between groups(baseline-1 month)

Hematocrit (%), n ¼ 22 35.4 � 5.0 35.4 � 5.4 p ¼ 0.937 p ¼ 0.723Platelets (�103/ml), n ¼ 14 184.8 � 107.8 174.0 � 96.5 p ¼ 0.952 p ¼ 0.645Total WBC (�103/ml), n ¼ 22 13.1 � 7.8 11.7 � 6.5 p ¼ 0.244 p ¼ 0.488Neutrophils (�103/ml), n ¼ 22 9.6 � 7.2 8.1 � 5.2 p ¼ 0.455 p ¼ 0.321ALT (IU/L), n ¼ 22 85.1 � 59.2 55.2 � 35.7 p ¼ 0.005 p ¼ 0.062ALP (IU/L), n ¼ 22 26.8 � 11.9 28.0 � 10.8 p ¼ 0.635 p ¼ 0.355GGT (IU/L), n ¼ 17 4.1 � 2.8 2.8 � 2.1 p ¼ 0.031 p ¼ 0.557Total bilirubin (mg/dl), n ¼ 19 0.3 � 0.3 0.2 � 0.2 p ¼ 0.127 p ¼ 0.496

Clopidogrel Baseline 1 month

Hematocrit (%), n ¼ 23 35.2 � 4.3 35.6 � 5.5 p ¼ 0.703Platelets (�103/ml), n ¼ 13 193.9 � 63.7 225.0 � 102.8 p ¼ 0.340Total WBC (�103/ml), n ¼ 19 11.2 � 5.8 8.9 � 3.6 p ¼ 0.014Neutrophils (�103/ml), n ¼ 19 7.6 � 4.8 5.5 � 2.7 p ¼ 0.066ALT (IU/L), n ¼ 24 76.3 � 40.3 77.1 � 62.7 p ¼ 0.989ALP (IU/L), n ¼ 24 30.0 � 13.5 34.2 � 13.4 p ¼ 0.044GGT (IU/L), n ¼ 16 3.7 � 3.7 2.9 � 3.4 p ¼ 0.188Total bilirubin (mg/dl), n ¼ 20 0.2 � 0.1 0.2 � 0.1 p ¼ 0.677

WBC ¼ white blood cells; see abbreviation table for the remainder of the abbreviations.Bold and italic data represents significantly different between the groups identified.

k Plavix� package insert, December 2013, Bristol-MeyersSquibb/Sanofi Pharmaceuticals Partnership, Bridgewater, NJ,USA.

S314 D.F. Hogan et al.

group was somewhat longer [248 (95% CI 137e431)days]. However, the cats in this study had a run-inperiod of 1e3 months which could impact thecomparison to other studies.

Both aspirin and clopidogrel were well tol-erated in this study but the relatively small pop-ulation size potentially limits the clinicalrecognition of adverse events to those that occurcommonly while more infrequent or episodicevents would only be recognized with a largerstudy population. Vomiting, hematemesis, inap-petance and other gastrointestinal signs havebeen reported in up to 22% of cats receivingaspirin at the same dose used in this study.1

Aspirin had to be discontinued in 1 cat due tovomiting and hematochezia while one additionalaspirin treated cat who died following recurrentCATE had a large gastric ulcer on necropsy.Empirical observations of vomiting and frothing atthe mouth have been reported in cats receivingclopidogrel, presumably due to the bitter taste ofthe drug. None of the cats receiving clopidogrel inthis study experienced vomiting or gastro-intestinal signs attributable to clopidogreladministration. The apparent lack of gastro-intestinal signs for both drugs is unexplained, butthese drugs were put into an empty gelatin cap-sule for administration and this could havereduced oral or gastric irritation. After 1 month ofstudy drug administration in the study reportedhere, there was a significant increase in ALP in the

aspirin group and a significant increase in ALT inthe clopidogrel group. However, none of theseindividual cats had values outside of the referencerange. One cat in the clopidogrel group experi-enced elevated liver enzymes and developedicterus after 40 days of drug administration. Therewas a significant reduction in the white blood cellcount at 1 month in the cats from the clopidogrelgroup (Table 2) but none were outside of thereference range. There was no identifiablebleeding noted in any of the cats in this studyaside from the one cat in the aspirin group withhematochezia. In the seminal CAPRIE study eval-uating clopidogrel and aspirin in humans, adversegastrointestinal events including diarrhea, indi-gestion, nausea, and vomiting were reported in19% of patients receiving clopidogrel and 21%receiving aspirin with diarrhea significantly morecommon with clopidogrel and indigestion, nausea,and vomiting significantly more common withaspirin.18 A small number of patients developedelevated liver values with the incidence sig-nificantly higher for aspirin (3.15%) than clopi-dogrel (2.97%). Thrombocytopenia andneutropenia were rare and severe neutropenia hasbeen reported in <1% of human patients takingclopidogrel in post-market monitoring.k

FAT CAT study S315

Clopidogrel was associated with a significantlyhigher frequency of dermatologic rash than aspirinbut no such occurrence was noted in the cats inthe study reported here. There was no significantdifference in all-cause bleeding between clopi-dogrel and aspirin (9.27% vs. 9.28%) although therewas a significantly higher incidence of gastro-intestinal bleeding associated with aspirin (2.66%vs. 1.99%).

Retrospective studies of aspirin therapy havereported recurrent CATE rates from 17% to75%1,10,13,17 and the recurrence rate for aspirin inthis study was 75% while the recurrence rate inclopidogrel treated cats was 53%. The overallrecurrence rate of CATE may not be the bestmarker of antithrombotic therapy in this pop-ulation of cats as complete protection againstrecurrent thrombotic events in such animals isunlikely and unrealistic. Such a marker interprets arecurrent CATE event in a cat at 2 months in thesame light as a recurrent event in another cat at 2years. For this reason, the median time to eventfor the primary endpoint of recurrent CATE mayprovide a more objective assessment of ther-apeutic success.

Anticoagulant therapy with warfarin has beenshown to be superior to aspirin or combined aspirinand clopidogrel therapy for the prevention ofcardioembolic events in humans.19 Similarly, clo-pidogrel added to aspirin is superior to aspirinalone in patients unsuitable for anticoagulanttherapy.20 Newer anticoagulant drugs have beenshown to be as effective as warfarin for car-dioembolic events with atrial fibrillation withoutthe need for frequent monitoring and the same orreduced risk of bleeding.21e23 These drugs mayprovide similar benefits in cats at risk for recurrentCATE and should be evaluated against clopidogrelin the future.

There a number of study limitations. The overallstudy population size is relatively small but thesurvival rate for CATE is poor which limitedenrollment. Despite the small study groups a sig-nificant difference was identified between thestudy groups with respect to the primary endpointalthough this did require an increase in populationsize determined through a repeat power analysisduring the study. The small population size doesimpair the ability to identify less common adverseevents from the study drugs. The echocardio-graphic exams were performed by individuals withvaried levels of expertise and precise measure-ments were not verified by a central site. How-ever, diagnosis of underlying cardiac disease wasalways made by a board-certified cardiologisteither directly from their examination or from

images provided to the principal investigator.Subsequent echocardiographic exams were notincluded in the study design, so the assessment ofprogressive cardiac disease and association withrecurrent CATE cannot be determined. Whileconcurrent cardiac therapy was attempted to bestandardized based on the underlying cardiac dis-ease, this was not always possible. Adherence to astrict schedule for blood test monitoring was notpossible in some cases and platelet counts weredifficult to reliably obtain. Platelet function wasnot assessed in this study so we cannot determineif an individual cat had appropriate plateletinhibition in response to the study drug and absentor reduced responders to both aspirin and clopi-dogrel have been reported in humans.24e27 Thedose of aspirin used in this study could be chal-lenged. We decided to use the standard dose ofaspirin as it allowed for easier creation of studydrug capsules and precluded the need to changedosing based on changing weight of the cats.Concerns have been expressed that this dose ofaspirin may inhibit endothelial cells and reducethe formation of prostacyclin, which inhibits pla-telet aggregation. However, it has been shown thatthis dose of aspirin was not associated with ahigher frequency of recurrent CATE compared to alower dose of aspirin in a retrospective study.1

Additionally, there is evidence that the dose ofaspirin used in this study does not result in plateletinhibition in the cat28 (Hogan DF, unpublisheddata) so the use of a lower dose of aspirin wouldseem unlikely to result in platelet inhibition. Theuse of a lower aspirin dose has been associatedwith reduced adverse events in cats but we did notsee a high frequency of adverse events associatedwith aspirin administration in this study. The studydrugs were put into capsules of different colorsand if an investigator had two cats on differentcolored capsules, they could assume they wereassigned to different study groups. However, thedifferent colored capsules reduced the risk of catsaccidently receiving multiple consecutive days ofaspirin and the investigators and enrolling veter-inarians were not informed during the study thatthe different colors were associated with differenttreatment groups. Lastly, this study was not aplacebo controlled trial, but the steering com-mittee had concerns regarding the ethics ofincluding a placebo arm when the prognosis ofCATE is so poor. We were also concerned that catowners and enrolling veterinarians would bereluctant to enroll cats knowing that the cat mayreceive no prevention for a recurrent CATE event.That being said, this study is a comparative studybetween clopidogrel and aspirin and the possibility

S316 D.F. Hogan et al.

that aspirin actually resulted in a worse outcomethan no treatment must be considered; resulting inclopidogrel having a neutral effect on the studyoutcome.

In conclusion, clopidogrel was superior toaspirin for the prevention of recurrent CATE andchronic clopidogrel administration was well tol-erated and bleeding was not reported. It is pos-sible that cats who survive a CATE event aresomehow different from the broader populationof cats with underlying heart disease so it isunknown if the superiority of clopidogrel overaspirin is applicable to primary prevention ofCATE. However, given the route of admin-istration, good tolerability, lack of bleeding, andconcern regarding the inadequate antiplateleteffect of aspirin in cats, clopidogrel would appearto be a reasonable clinical choice for primarythromboprophylaxis in cats determined to be atrisk for CATE.

Conflicts of interest

D.F. Hogan: prior grant support from Bristol-MeyersSquibb/Sanofi Pharmaceuticals Partnership.

There are no other reportable potential conflictsof interest related to this study.

Acknowledgments

This study was funded by the Morris Animal Foun-dation (D04FE-005, D09FE-001). Additional supportwas provided by the VIN Foundation and privatedonors.

We would like to thank Kim Sederquist for hercritical role as study coordinator and we would liketo thank Dr. Mark Rishniw for his assistance insetting up the study website.

The steering committee would like to acknowl-edge the following clinicians who enrolled catsonto the study:

Leslie Ahluwalia, Eddison Barrientos, HelenBell, Ben Beyers, Jan Bright, Sara Brown, CarolChampion, I-Ping Chan, Kevin Christiansen, TaroCuetara, Laura DeLellis, Cathy Eastman, AmaraEstrada, Scott Fausel, Nonya Fiakpui, Deb Fine,Darren Fry, Carrie Ginieczki, Jonathan Goodwin,Henry Green, Karen Hayworth, Debra Henderson,Beverly Hickman, Andonia Hsu, Herbert Hulls, KateJacob, Lauren Kappers, William Keatts, Mike Lue-thy, Aliya Magee, Rebecca Malakoff, Sue Marshall,Dawn Martin, Anne Masloski, Christopher Mid-dleton, Lindsay Norman, Stacy O’Quinn, Jeanne

Pittari, Analisa Prahl, Robert Prosek, RebeccaReardon, Erin Reed, Evelyn Richer, Yvonne Rob-erts, Carl Sammarco, Karen Sanderson, JosefSchiele, Sarah Scruggs, Eryn Shipley, Eva Sikorska,Meg Sleeper, Barb Smith, Aarthi Subram, DennisTrafny, Robert Vasilopulos, Carole Werkhoven,Elaine Wexler-Mitchell, Aaron Wey.

References1. Smith SA, Tobias AH, Jacob KA, Fine DM, Grumbles PL.

Arterial thromboembolism in cats: acute crisis in 127 cases(1992e2001) and long-term management with low-doseaspirin in 24 cases. J Vet Int Med 2003;17:73e83.

2. Green HW, Hogan DF. Suspected iatrogenic paradoxicalembolism in a cat. J Am Anim Hosp Assoc 2005;41:193e197.

3. Kim CK, Jung S, Yoon B-W. Practical issues to prevent strokeassociated with non-valvular atrial fibrillation. J Stroke2013;15:144e152.

4. Ferro JM. Cardioembolic stroke: an update. Lancet Neurol2003;2:177e188.

5. Veterinary Medical Data Base (VMDB, www.vmdb.org),1980e2003.

6. Atkins CE, Gallo AM, Kurzman ID, Cowen P. Risk factors,clinical signs, and survival in cats with a clinical diagnosis ofidiopathic hypertrophic cardiomyopathy: 74 cases(1985e1989). J Am Vet Med Assoc 1992;201:613e618.

7. Rush JE, Freeman LM, Fenollosa NK, Brown DJ. Populationand survival characteristics of cats with hypertrophic car-diomyopathy: 260 cases (1990e1999). J Am Vet Med Assoc2002;220:202e207.

8. Laste NJ, Harpster NK. A retrospective study of 100 cases offeline distal aortic thromboembolism: 1977e1993. J AmAnim Hosp Assoc 1995;31:492e500.

9. Moore KE, Morris N, Dhupa N, Murtaugh RJ, Rush JE. Ret-rospective study of streptokinase administration in 46 catswith arterial thromboembolism. J Vet Emerg Crit Care 2000;10:245e257.

10. Schoeman JP. Feline distal aortic thromboembolism: areview of 44 cases (1990e1998). J Feline Med Surg 1999;1:221e231.

11. Deedwania PC, Huang GW. Role of emerging antithrombotictherapy in the prevention of cardioembolic complications inpatients with atrial fibrillation. Am J Cardiovasc Drugs 2011;11:265e275.

12. Freeman WD, Aguilar MI. Prevention of cardioembolicstroke. Neurotherapeutics 2011;8:488e502.

13. Pion PD, Kittleson MD. Therapy for feline aortic throm-boembolism. In: Kirk RW, editor. Current veterinary therapyX. Philadelphia: WB Saunders; 1989. p. 295e302.

14. Quinn MJ, Fitzgerald DJ. Ticlopidine and clopidogrel. Cir-culation 1999;100:1667e1672.

15. Coukell AJ, Markham A. Clopidogrel. Drugs 1997;54:745e750.

16. Bhatt DL, Chew DP, Hirsch AT, Ringleb PA, Hacke W,Topol EJ. Superiority of clopidogrel versus aspirin inpatients with prior cardiac surgery. Circulation 2001;103:363e368.

17. Borgeat K, Wright J, Garrod O, Payne JR, Fuentes VL.Arterial thromboembolism in 250 cats in general practice:2004e2012. J Vet Intern Med 2014;28:102e108.

18. CAPRIE Steering Committee. A randomized, blinded, trial ofclopidogrel versus aspirin in patients at risk of ischaemicevents (CAPRIE). Lancet 1996;348:1329e1339.

FAT CAT study S317

19. The ACTIVE Writing Group. Clopidogrel plus aspirin versusoral anticoagulation for atrial fibrillation in the Atrialfibrillation Clopidogrel Trial with Irbesartan for preventionof Vascular Events (ACTIVE W): a randomized controlledtrial. Lancet 2006;367:1903e1912.

20. The ACTIVE Investigators. Effect of clopidogrel added toaspirin in patients with atrial fibrillation. N Engl J Med 2009;360:2066e2078.

21. Connolly SJ, Ezekowitz MD, Yusuf S, Eikelboom J, Oldgren J,Parekh A, Pogue J, Reilly PA, Themeles E, Varrone J,Wang S, Alings M, Xavier D, Zhu J, Diaz R, Lewis BS,Darius H, Diener HC, Joyner CD, Wallentin L, RE-LY SteeringCommittee and Investigators. Dabigatran versus warfarin inpatients with atrial fibrillation. N Engl J Med 2009;361:1139e1151.

22. Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W,Breithardt G, Halperin JL, Hankey GJ, Piccini JP, Becker RC,Nessel CC, Paolini JF, Berkowitz SD, Fox KA, Califf RM,ROCKET AF Investigators. Rivaroxaban versus warfarin innonvalvular atrial fibrillation. N Engl J Med 2011;365:883e891.

23. Granger CB, Alexander JH, McMurray JJ, Lopes RD,Hylek EM, Hanna M, Al-Khalidi HR, Ansell J, Atar D,Avezum A, Bahit MC, Diaz R, Easton JD, Ezekowitz JA,Flaker G, Garcia D, Gerales M, Gersh BJ, Golitsyn S, Goto S,Hermosillo AG, Hohnloser SH, Harowitz J, Mohan P,

Jansky P, Lewis BS, Lopez-Sendon JL, Pais P,Parkhomenko A, Verheugt FW, Zhu J, Wallentin L, ARIS-TOTLE Committees and Investigators. Apixaban versuswarfarin in patients with atrial fibrillation. N Engl J Med2011;365:981e992.

24. Krasopoulos G, Brister SJ, Beattie WS, Fraser R,Buchanan MR. Aspirin “resistance” and risk of car-diovascular morbidity: systematic review and meta-analy-sis. BMJ 2008;336:195e203.

25. Hankey GJ, Eikelboom J. Aspirin resistance. Lancet 2006;367:606e617.

26. Angiolillo DJ, Fernandez-Ortiz A, Bernardo E, Alfonso F,Macaya C, Bass TA, Costa MA. Variability in individualresponsiveness to clopidogrel: clinical implications, man-agement, and future perspectives. J Am Coll Cardiol 2007;49:1505e1516.

27. Matetzky S, Shenkman B, Guetta V, Shechter M, Beinhart R,Goldenberg I, Novikov I, Pres H, Savion N, Varon D, Hod H.Clopidogrel resistance is associated with increased risk ofrecurrent atherothrombotic events in patients with acutemyocardial infarction. Circulation 2004;109:3171e3175.

28. Cathcart CJ, Brainard BM, Reynolds LR, Al-Nadaf S,Budsberg SC. Lack of inhibitory effect of acetylsalicylic acidand meloxicam on whole blood platelet aggregation in cats.J Vet Emerg Crit Care 2012;22:99e106.

Available online at www.sciencedirect.com

ScienceDirect