Introduction
Sixteen percent of cats in the US are affected with cardiomyopathy (CM),1 and up to 45% of cats with hypertrophic or other types of CM have evidence of arterial thromboembolism (ATE) at necropsy.2 Spontaneous echocardiographic contrast (SEC), which describes an ultrasonographic pattern thought to represent the aggregation of cellular blood components, might reflect an activated hemostatic system and can be seen in 50% of cardiomyopathic cats with or without a left atrial (LA) thrombus and in 56% of cats with ATE and LA enlargement.3 Hypercoagulability, or the tendency to form thromboemboli resulting in cardioembolic sequelae, is difficult to identify in cats prior to an ATE event.4 In a case series5 published in 2003, the median survival time for cats that experienced an acute distal aortic ATE and survived the acute episode was approximately 120 days. A subsequent report6 that evaluated the use of clopidogrel as an antiplatelet agent to treat cats after an ATE event demonstrated an improved median survival time of 248 days (95% CI, 137 to 431 days).
The improved survival time conferred by antiplatelet therapy suggests that platelet activation contributes to the formation of cardioembolic events in cats with CM. Platelets may be activated in the presence of altered blood flow, vascular shear stress, and endothelial damage within the cardiovascular system, all of which can occur in cats with CM.7 Coagulation factor activation resulting from blood stasis within the LA can also enhance platelet activation and may represent an independent trigger for clot formation.8 In patients with preexisting hypercoagulable states (eg, due to systemic inflammation), the development of altered blood flow in the LA may provide an additional impetus for clot formation.
Both antiplatelet (clopidogrel) and anticoagulant (heparins) medications have been recommended as adjunct therapies for cats with CM presumed to be at risk of thromboembolism.9,10 Historically, therapeutic recommendations have focused on antiplatelet medications because of the convenience of orally administered drugs and desire to avoid multiple daily heparin injections. Because of this bias and the unknown etiology of ATE in affected cats, it is not known whether anticoagulant drugs (ie, those specifically acting on proteins of the clotting cascade rather than platelet function) might further delay ATE recurrence compared to antiplatelet drugs or whether a combination of antiplatelet and anticoagulant medications is superior to either drug class alone.11
Rivaroxaban (RVX) is an orally administered direct inhibitor of activated Factor X (FXa)12 that does not require antithrombin for FXa inhibition, distinguishing it from heparin and related drugs. Because RVX is given orally, there is no need for owners to administer injections at home (as is necessary with heparin). Rivaroxaban is effective for the prevention of cardioembolic events (stroke) in human patients with atrial fibrillation, among other indications.13 The hemodynamic alterations noted in hearts with atrial fibrillation (eg, nonlaminar blood flow and intracardiac blood stasis) mirror some of those seen in feline CM. In cats, oral RVX reaches peak anticoagulation between 1 and 3 hours after administration,14 similar to dogs, rats,15 and humans.13,16,17 The maximum plasma drug concentration is dose dependent, and at daily oral dosages of 2.5 and 5 mg/cat, peak plasma drug concentrations are achieved that are within the range considered therapeutic in people.14 If plasma coagulation factor activation plays a major role in the formation of thrombi in cats with CM, RVX may be a more effective therapy than antiplatelet agents for the prevention of recurrent ATE.
The present study aimed to compare single-agent anticoagulation with RVX (2.5 mg, PO, q 24 h) to clopidogrel (18.75 mg, PO, q 24 h) to determine the impact of anticoagulation compared to antiplatelet therapy for the prevention of recurrent ATE in cats with CM. We hypothesized that treatment with RVX in this sample of cats would increase the time to ATE recurrence when compared to treatment with clopidogrel. We further hypothesized that long-term therapy with RVX at this dosage would be well tolerated without adverse effects in this group of cats.
Methods
This was a prospective, double-masked, randomized, multicenter clinical trial, carried out at 18 veterinary specialty hospitals in the US and the UK. All study procedures were approved by the University of Georgia Clinical Research Committee (CR-402) and other local committees as required by specific participating institutions. Case recruitment was supported via periodic targeted emails to emergency/critical care practitioners and veterinary cardiologists, advertising on clinical trials websites, outreach to support groups for owners of cats with heart disease, and through general interest advertising (eg, the Morris Animal Foundation website).
Eligible participants were client-owned cats of any age recruited from those that had recently (ie, within 1 month of potential initiation of therapy with the study drug) experienced cardiogenic ATE, survived the initial thromboembolic event, and were eating reliably and otherwise stable enough to be discharged from the hospital. Cats were additionally required to have undergone an echocardiographic examination performed by a board-certified veterinary cardiologist to confirm the presence of sufficiently severe heart disease to suggest that the thromboembolus or clinical signs of thromboembolism were likely cardiogenic. Cats were excluded if their owner was unable or unwilling to administer study medication, bring the cat for scheduled recheck appointments, or regularly complete online questionnaires. In addition, cats undergoing chronic treatment for a disease that involved medications known or suspected to impact coagulation (eg, corticosteroids) were excluded. Other exclusion criteria included evidence or history of coagulopathy, body weight < 2.5 kg, presence of comorbid conditions (eg, neoplasia) that were expected to negatively impact survival, and echocardiographic evidence of intracardiac thrombi. Cats with SEC were not excluded. Cats receiving antiplatelet or anticoagulant drugs at the time of eligibility assessment were not excluded, unless the owner or attending veterinarian was unwilling to transition the cat to one of the study drugs. Following enrollment, cats diagnosed with diseases requiring the initiation of drugs known or suspected to impact coagulation were not removed from the study.
Once a cat met study eligibility criteria, the principal investigator (BMB) reviewed the study plan by phone with the owner for the purposes of obtaining informed consent, and a plan to transition the current anticoagulant therapy to the study drug was developed with the attending veterinarian and owner. The plans were individually tailored and varied depending on the shipping schedule of the study drug, owner availability, and the need to wean current medications (vs abrupt discontinuation).
Study participants were randomized to receive either clopidogrel (18.75 mg/cat, PO, q 24 h) or RVX (2.5 mg/cat, PO, q 24 h) for 24 months or until one of the study end points (ATE recurrence or death) was met, whichever occurred first. A randomization sequence was generated prior to the start of the study by use of block randomization with block sizes of 6 and a 1:1 treatment allocation ratio. Only 1 individual had access to the randomization sequence; this person was not involved with eligibility assessment, enrollment, case management, or data analysis. All study investigators, attending cardiologists, and owners were masked to treatment assignment.
Study drugs were provided to the owner as full tablets (either 75 mg of clopidogrel or 10 mg of RVX), and owners were instructed to give one-fourth tablet by mouth every 24 hours. A commercial pill cutter was supplied with the first shipment of study medication. The drugs were provided in a prescription vial with a label that did not indicate the specific study drug. Owners were also provided with a sealed envelope containing the cat’s treatment assignment and drug safety information in the event of an emergency that required unmasking to the study drug. Standardization of pill appearance was not possible due to logistical difficulties relating to the use of compounded tablets.
The assigned study medication was mailed directly to the owner by a single study technician, who also maintained the study drug randomization assignments, as a 2-month supply. Medication shipments occurred every 2 months following the owner’s completion of an online survey designed to assess quality of life (QOL; Supplementary Table S1) and to prompt owner reporting of potential adverse events. Email reminders were automatically generated and delivered to owners 2 weeks prior to the next shipment date to prompt completion of the survey and subsequent shipping of the next supply of medication. If an owner did not complete the online survey, direct phone contact was made to ensure completion, but this did not delay shipment of the medication.
Study subjects were requested to undergo echocardiographic examinations by a board-certified veterinary cardiologist at 2, 6, 12, and 18 months after enrollment. At each of these visits, cats underwent transthoracic echocardiographic examinations performed by a board-certified veterinary cardiologist. Participating cardiologists were provided a list of requested echocardiographic measurements, calculations, and observations, focused on assessment of cardiac chamber size and function (Supplementary Table S2). After each examination, these data were uploaded by the cardiologist using a web-based portal. A gift card to defray the cost of the recheck examinations was provided directly to owners after data from each echocardiogram were received. At each recheck examination, owners also completed a questionnaire, designed to assess their ability to reliably give the study medication, specific changes to the cat’s well-being, and other signs related to illness. Completed questionnaires were scanned and emailed to the study principal investigator, who reviewed and compiled the responses. Also at each recheck visit, for cats in the US that were amenable to venipuncture, blood was collected and placed into collection tubes containing EDTA. Following centrifugation, plasma was frozen and shipped overnight to the study principal investigator, where it was stored at –80 °C until its use for analysis of plasma drug or metabolite concentrations. Plasma samples were analyzed within 12 months of collection for plasma drug concentration in cats receiving RVX (Auburn University Specialized Pharmaceutical and Experimental Center for Translational Research and Analysis) or for the SR 26334 metabolite concentration in cats receiving clopidogrel (Clinical Pharmacology Laboratory at the North Carolina State University College of Veterinary Medicine).
During the treatment period, adverse events could have been reported by owners in several ways. Direct questions regarding the incidence of hemorrhage or other concerns were part of the bimonthly web-based at-home questionnaires, as well as the questionnaires administered during scheduled in-hospital recheck visits. Owners were also provided phone and email contact information for the principal investigator and instructed to reach out directly to address any concerns. During the 2-year study period, if a cat developed suspected intolerance to the study drug, experienced an adverse event that required discontinuation of the study drug or veterinarian or owner unmasking, or the owner or attending veterinarian elected to discontinue the study drug or terminate study participation, the cat was removed from the study and no additional monitoring was performed. In addition to the individual queries on the QOL surveys, a composite activity of daily life (ADL) score was generated for each time point by summing the responses to the first 8 questions on the QOL questionnaire.18
The primary and secondary study end points were recurrence of ATE and death from any cause, respectively. In addition, the occurrence of adverse events (eg, hemorrhage) or intolerance to the study medication was recorded. Assessment of the circumstances surrounding suspected ATE recurrence was performed by the principal investigator and determined to be consistent with ATE on the basis of follow-up discussions with the owner or attending veterinarian. Similarly, if a cat died during the study period, cause of death was noted and categorized as cardiovascular or noncardiovascular, as determined by the principal investigator after discussion with the attending veterinarian or owner. Deaths due to ATE were classified as cardiac, and those attributable to CHF were generally accompanied by dyspnea, cardiac arrhythmias, hypotension, and compatible radiographic or echocardiographic evidence. If a death could not reliably be characterized as cardiogenic (ie, at least 2 of the above criteria were present), it was classified as noncardiogenic (including those that may have been sudden cardiac death but were not otherwise classifiable). For the purposes of this study, euthanasia due to owner-perceived decreased QOL (and not due to ATE or CHF) was classified as a non-CHF death, even if the decline in status was a consequence of progressive cardiac disease.
Power analysis
In a prior report,6 a significant difference in time to ATE recurrence was demonstrated between cats receiving aspirin or clopidogrel when data from 36 subjects in each treatment group were evaluated. In that report, median time to ATE recurrence in animals treated with clopidogrel or aspirin was 443 or 192 days, respectively. The present study aimed to demonstrate at least a 6-month (180-day) difference between clopidogrel- and RVX-treated cats with regard to time to ATE recurrence. To achieve this with a power of 92% and an α of .05, 35 animals in each group would be necessary. Ten additional animals were added to each group in anticipation of those that might be lost to follow-up; therefore, targeted enrollment was 45 cats per treatment group.
Statistical analysis
Statistical analyses were performed with commercially available statistical software (SAS, version 9.4; SAS Institute Inc). A significance threshold of .05 was used for all analyses. Linear mixed models (LMMs) were used to analyze all echocardiographic variables and QOL and ADL scores, and generalized LMMs (ordinal logistic GLMM) were used to analyze specific survey aspects such as appetite, litter box usage, and overall activity scores. For all questionnaire factors, answers were converted to numerical values, with 1 corresponding to the worst possible rating, 5 considered normal, and 6 representing increased activity. The model for each echocardiographic variable included fixed factors for group, time, and a group by time interaction effect as well as a baseline covariate. The model for each ADL survey score was the same except without the baseline covariate. All models had a random intercept for each cat. Conditional residual plots were examined for each LMM and indicated nonhomogeneous variance for the echocardiographic data for left ventricular outflow tract velocity and LA appendage flow velocity. These variables were log transformed, and the assumption of homogeneous variance was confirmed. Histograms and Q-Q plots of LMM conditional model residuals were examined and confirmed the assumption of normality for all end points. Simple effects of group for each time were performed, and adjacent time points within the same group were compared additionally for survey variables. Time comparison P values for each group separately were adjusted for multiple comparisons with the Hochberg method. The Satterthwaite degrees of freedom method was used.
Kaplan-Meier analysis and survival curves were constructed for overall survival time (OST), disease-specific survival (DSS), time to recurrence (TTR), disease-free survival, and disease-free disease-specific survival (DFDSS) and used to estimate median survival times and 1-year survival probabilities. Overall survival time represented the time to one of the study end points (recurrent ATE or death from any cause); cats were censored if they were alive at last follow-up. Disease-specific survival time was calculated, censoring cats that died of causes deemed unrelated to ATE or cardiac disease. All cats that experienced a recurrence of ATE were analyzed separately as a TTR analysis; all cats without a recurrence were censored at the time of last follow-up or time of death. Disease-free survival comprised all cats, censoring those that were recurrence-free and alive at last follow-up, and DFDSS represented all cats that lived or died of noncardiac causes. Log-rank tests were used to compare times between groups.
The physical and echocardiographic characteristics of each group were compared with an unpaired t test or Mann-Whitney U test as appropriate for the distribution of data. The presence or absence of adverse events was compared between groups with χ2 analysis. Other categorical comparisons analyzed with the χ2 analysis included the presence or absence of SEC, the relationship of SEC to recurrence of ATE, cardiac diagnosis, and location of ATE. Normally distributed data are described as mean ± SD, while nonparametric data are described as median (IQR).
Results
Cats were evaluated between January 2016 and May 2023 by 9 private and 9 university-based practices in the US (2016 to 2019) or in the US and the UK (2019 to 2023). All enrolled cats were randomized within 1 month of ATE occurrence. Sixty-one cats were assessed for eligibility. Of these, 45 were enrolled and randomized: 19 to the clopidogrel-treated group (CG) and 26 to the RVX-treated group (RG). Of the 16 excluded cats, the owners of 2 declined to participate, 8 were receiving dual therapy with clopidogrel and RVX and the owners did not want to transition to monotherapy, and 6 died prior to randomization (Figure 1). Baseline characteristics of all cats are presented in Table 1.
Study sample and outcome summary data for a comparative investigation of survival of cats following aortic thromboembolism and assigned to receive either clopidogrel or rivaroxaban as a single-agent anticoagulant therapy.
Citation: Journal of the American Veterinary Medical Association 2025; 10.2460/javma.24.09.0584
Baseline characteristics of 45 cats randomized to receive either clopidogrel or rivaroxaban as single-agent anticoagulant therapy following aortic thromboembolism.
| Characteristic/parameter | Treatment group | |||||
|---|---|---|---|---|---|---|
| Clopidogrel | Rivaroxaban | All cats | ||||
| No. of cats | 19 | 26 | 45 | |||
| Age (y) | 8.15 ± 4.5 | 8.6 ± 3.6 | 8.4 ± 3.9 | |||
| No. (%) sex | ||||||
| Spayed female | 3 (16) | 11 (42) | 14 (31) | |||
| Neutered male | 16 (84) | 15 (58) | 31 (69) | |||
| Body weight (kg) | 4.92 ± 1.13 | 4.42 ± 0.86 | 4.62 ± 1.0 | |||
| No. (%) cardiac phenotypic diagnosis | ||||||
| HCM | 9 (47) | 17 (65) | 26 (58) | |||
| HCM with obstruction | 3 (16) | 6 (23) | 9 (20) | |||
| Unclassified cardiomyopathy | 2 (11) | 3 (12) | 5 (11) | |||
| Restrictive cardiomyopathy | 4 (21) | — | 4 (9) | |||
| Dilated cardiomyopathy | 1 (5) | — | 1 (2) | |||
| No. (%) anatomic location(s) affected by initial ATE | ||||||
| Bilateral pelvic limb | 10 (53) | 17 (65) | 27 (60) | |||
| Unilateral pelvic limb | 4 (21) | 5 (19) | 9 (20) | |||
| Unilateral thoracic limb | 3 (16) | 4 (21) | 7 (16) | |||
| Central nervous | 1 (5) | 0 (0) | 1 (2) | |||
| Echocardiographic findings | ||||||
| No. (%) SEC present* | 9 (47) | n = 19 | 18 (69) | n = 26 | 27 (60) | n = 45 |
| LA:Ao | 2.08 ± 0.34 | n = 18 | 2.12 ± 0.41 | n = 26 | 2.11 ± 0.38 | n = 44 |
| Left atrial appendage flow velocity (m/s) | 0.19 ± 0.06 | n = 8 | 0.21 ± 0.14 | n = 21 | 0.23 ± 0.12 | n = 29 |
| LV FS (%) | 36.6 ± 14.1 | n = 18 | 42.7 ± 9.4 | n = 25 | 40.1 ± 11.9 | n = 43 |
Data are presented as number (%) or mean ± SD.
ATE = Arterial thromboembolism. HCM = Hypertrophic cardiomyopathy. LA:Ao = Left atrial-to-aortic diameter ratio. LV FS = Left Ventricular Fractional Shortening. SEC = Spontaneous echocardiographic contrast.
*Percentage of cases for which information was available.
Of the 45 randomized cats, almost all (17 of 19 [89%] clopidogrel- and 25 of 26 [96%] RVX-treated cats) met one or both study end points or completed the 2-year study period. Seventeen cats experienced ATE recurrence: 7 of 19 cats (37%) in the CG and 10 of 26 cats (39%) in the RG (P = 1.0). In these cats, median (95% CI) time to ATE recurrence was 663 days (150 to not calculable) and 513 days (242 to not calculable) for the CG and RG, respectively (P = .797).
All but 3 cats (1 in the CG and 2 in the RG) experiencing recurrent ATE were euthanized, with 2 surviving the remainder of the 2-year study period (although they were no longer included in the data analysis after the ATE) and the other lost to follow-up. Twelve cats were euthanized over the course of the study due to congestive heart failure (3 in the CG and 7 in the RG) and 1 due to progression of heart disease including atrial fibrillation (RG). Two cats (1 in the CG, 1 in the RG) died of presumed cardiac causes. Two cats were euthanized (1 in the CG, 1 in the RG) due to unrelated disease (eg, renal disease, other QOL concerns). One cat in the RG and 2 cats in the CG died at home without adequate information to determine the underlying cause. Four cats were withdrawn from the study: 2 cats initially diagnosed with hypertrophic CM that were subsequently found to have transient myocardial thickening (1 each in the RG and CG), 1 cat for noncompliance with the study protocol (RG), and 1 cat for which the attending cardiologist prescribed a second anticoagulant drug after new SEC was identified on a recheck echocardiogram (CG). Three cats in each treatment group were alive at the end of the 2-year study period without having experienced ATE recurrence (P = .697 between groups).
Adverse events that did not require discontinuation of the study drug or investigator unmasking occurred in 2 cats, both in the RG. These included transient epistaxis in one and the development of diabetes mellitus and diabetic ketoacidosis requiring hospitalization in another. A third cat in the RG experienced worsening of inflammatory bowel disease (previously responsive to dietary management and intermittent metronidazole therapy) after the 18-month study recheck examination and was prescribed budesonide therapy. This cat remained in the study for the remaining 6 months and completed the full 2 years of study. Three owners reported difficulty administering the study drug to their cats (2 in the CG, 1 in the RG), which was resolved through the use of gelatin capsules to hide the pills.
There were no changes in most echocardiographic parameters with time or between treatment groups at individual time points (Table 2; Supplementary Table S3). Exceptions to this were the LA-to-aortic root diameter ratio (LA:Ao) at the 6- and 12-month time points, for which mean LA:Ao in the CG was 2.19 ± 0.47 and 2.13 ± 0.35, respectively, compared to 2.45 ± 0.66 and 2.65 ± 0.32 in the RG (P = .041 and .025 for comparison at months 6 and 12, respectively). Although there was an apparent increase in the LA:Ao in the RG group with time, this was not significant following a Tukey correction for multiple comparisons (P = .0251). In addition, the presence of intracardiac SEC was more commonly noted in the RG compared to the CG at the 12-month time point (P = .024).
Two-dimensional LA:Ao over time in 45 cats randomized to receive either clopidogrel or rivaroxaban as single-agent anticoagulant therapy following aortic thromboembolism. There was no significant difference in mean body weight of either group over time or between individual time points.
| Variable | Month | Group | N | Mean | Median | SD | Min | Max | P valuec | Cat weight (kg) |
|---|---|---|---|---|---|---|---|---|---|---|
| LA:Ao | 0 | Clopidogrel | 18 | 2.08 | 2.07 | 0.34 | 1.58 | 2.71 | 4.79 ± 1.2 | |
| Rivaroxaban | 26 | 2.12 | 2.09 | 0.41 | 1.36 | 3.28 | 4.42 ± 0.9 | |||
| 2 | Clopidogrel | 13 | 2.08 | 2.11 | 0.3 | 1.42 | 2.55 | .115 | 4.83 ± 1.4 | |
| Rivaroxaban | 20 | 2.37 | 2.4 | 0.51 | 1.40 | 3.55 | 4.35 ± 0.9 | |||
| 6 | Clopidogrel | 9 | 2.19 | 2.14 | 0.47 | 1.46 | 3.22 | .041 | 5.10 ± 1.4 | |
| Rivaroxaban | 15 | 2.45 | 2.28 | 0.66 | 1.30 | 3.60 | 4.54 ± 0.8 | |||
| 12 | Clopidogrel | 5 | 2.13 | 2.1 | 0.35 | 1.69 | 2.52 | .025 | 5.47 ± 1.9 | |
| Rivaroxaban | 7 | 2.63 | 2.69 | 0.32 | 2.00 | 3.04 | 4.92 ± 0.6 | |||
| 18 | Clopidogrel | 1 | 2.17 | 2.17 | — | 2.17 | 2.17 | 6.2 | ||
| Rivaroxaban | 2 | 2.71 | 2.71 | 0.16 | 2.60 | 2.82 | 4.87 ± 0.7 |
Bolded P values indicate significant differences between CG and RG for that timepoint.
Concurrent therapies administered during at least a portion of the study period included furosemide in 12 of the 19 CG cats (63%) and 23 of the 26 RG cats (88%). Pimobendan was administered to 7 (37%) and 12 (46%) of the cats in the CG and RG, respectively. An angiotensin-converting enzyme inhibitor (enalapril, lisinopril, or benazepril) was administered to 5 of the 19 CG cats (26%) and 10 of the 26 RG cats (38%). Therapies that were administered to fewer than 3 cats each of either group included taurine, gabapentin, spironolactone, mirtazapine, metronidazole, diltiazem, amlodipine, and amoxicillin–clavulanic acid.
From the QOL surveys, ratings between treatment group and among study time points (months) were compared for select parameters, including appetite, ability to use the litter box, overall QOL, and the ADL score.18 There were no significant differences between treatment groups for interest/ability to eat (P = .282), litter box usage (P = .109), or overall QOL (P = .251) or ADL (P = .476) scores at any time point. There was a significant increase between the second and fourth months in the overall QOL (P = .028) and ADL (P = .004) scores in the RG, despite there being no difference in these scores between RG and CG for the same time points. There were no other significant changes in scores with time in either group for the evaluated categories. Mosaic plots illustrating the distribution for overall QOL assessment are shown in Figure 2.
Mosaic plot of overall quality of life (QOL) over time in cats receiving clopidogrel (A) or rivaroxaban (B) as single-agent anticoagulant therapy following an aortic thromboembolic event. Time is listed in months following the event, and the score is organized so that 5 represents a normal QOL, 6 an increased QOL, and 4 down to 1 an incrementally subnormal QOL. The size of the numbered area represents the number of owners of a total of 19 cats receiving clopidogrel or 26 cats receiving rivaroxaban reporting the specific score for their cats.
Citation: Journal of the American Veterinary Medical Association 2025; 10.2460/javma.24.09.0584
Overall survival was assessed from the perspective of several different end points (OST, DSS, TTR, disease-free survival, and DFDSS) as described previously. There were no significant differences in any survival comparison between groups (Table 3). Kaplan-Meier survival curves for OST and TTR are presented in Figure 3.
Median (95% CI) overall survival, disease-free survival, time to recurrence, and recurrence-free survival in 45 cats randomized to receive either clopidogrel or rivaroxaban as single-agent anticoagulant therapy following aortic thromboembolism.
| End point | No. censored | Clopidogrel | Rivaroxaban | aP value | ||
|---|---|---|---|---|---|---|
| Median (95% CI) time (d) | 1-year probability | Median (95% CI) time (d) | 1-year probability | |||
| Overall survival time | 13 | 335 (150–515) | 38%b | 296 (209–510) | 42%b | .724 |
| Disease-specific survival time | 18 | 371 (156–NC) | 52%c | 296 (210–513) | 45%c | .443 |
| Time to recurrence | 28 | 663 (150–NC) | 70%d | 513 (242–NC) | 63%d | .797 |
| Disease-free survival | 9 | 329 (150–515) | 36%e | 246 (136–476) | 37%e | .570 |
| Disease-free disease-specific survival | 14 | 335 (150–663) | 49%f | 296 (136–480) | 39%f | .332 |
NC = Not calculable (due to being outside of the range of the data).
aLog-rank test.
bProbability of surviving to 1 year.
cProbability of not dying from disease by 1 year.
dProbability of not recurring by 1 year.
eProbability of not recurring or dying by 1 year.
fProbability of not recurring or dying from disease by 1 year.
Kaplan-Meier curve representing survival probability (A) or time to recurrent aortic thromboembolism (B) in cats recovered from a prior aortic thromboembolism and assigned to receive either clopidogrel (solid blue line) or rivaroxaban (dashed red line) as a single-agent anticoagulant therapy for 2 years.
Citation: Journal of the American Veterinary Medical Association 2025; 10.2460/javma.24.09.0584
Plasma samples for measurement of plasma RVX concentrations were received from 9 cats, comprising 14 samples (eight 2-month samples, five 6-month samples, and one 12-month sample). Median RVX concentration was 23.76 ng/mL, ranging from 0.28 to 392.3 ng/mL. Five cats had samples analyzed for SR 23662 concentration (five 2-month samples, three 6-month samples), with a median concentration of 1.75 µg/mL and a range of 0.5 to 4.6 µg/mL. While some samples were collected and analyzed, the variability in time of collection from the last administered dose was so great that samples from the latter 4 years of cats were not collected nor analyzed.
Discussion
The use of RVX as a single-agent therapy in cats following recovery from cardiogenic embolism delayed recurrence of ATE for a similar time as in cats that received single-agent therapy with clopidogrel. Additionally, median time from enrollment to death from cardiac or noncardiac causes was not different between treatment groups. The time to ATE recurrence for both groups was similar to that of cats treated with single-agent clopidogrel therapy for secondary prevention in a previous study (median, 443 days; 95% CI, 185 to 990 days).6 In addition, both groups of the present study experienced appreciably longer time to ATE recurrence than cats treated with aspirin monotherapy in that same clinical trial (median, 192 days; 95% CI, 62 to 364 days).6 A more recent report11 of 32 cats receiving dual-agent therapy with clopidogrel and RVX (at the same dosages studied herein) noted that 3 cats that were ATE survivors experienced recurrence of ATE at 46, 173, and 616 days of therapy.
The current study was powered to detect a 6-month difference between treatment groups in time-to-recurrence of ATE, assuming a target enrollment of 90 cats. There is a possibility of type II error, as only half of the planned subjects were recruited. The difficulties in recruitment were multifactorial. We suspect that the persistent belief among veterinarians that cats are unable to recover from ATE and euthanasia is the only appropriate therapy likely contributed. Another component appeared to be a change in prescribing trends among some cardiologists in the latter half of the study period that favored dual therapy (clopidogrel and RVX), particularly in cats with SEC, which likely made it difficult for some owners to opt for randomization without full knowledge of the consequences of stopping one medication. Post hoc calculation of optimal sample size for the present study is inherently problematic, particularly considering the long median survival times (and wide CIs) of both treatment groups and the relatively short, fixed observation period. Nonetheless, estimates indicate that identification of a 6-month difference in recurrence between treatment groups in a 2-year observation period would be unlikely without including a substantially higher number of subjects.
The study design utilized herein required some assumptions about the dosage and the uniformity of drug dispersion within the prescribed tablets for both study drugs. First, RVX and clopidogrel dosage was standardized across all cats, regardless of body weight. While this represents a clinically relevant approach, the dosage in milligrams per kilogram varied across cats. Additionally, in keeping with current veterinary clinical usage of both drugs, owners were instructed to quarter the tablets for administration. For both drugs, the tablets were not scored and even dispersion of drug throughout the tablet was not guaranteed. The tablets were also relatively small and could have crumbled when divided. All owners were provided with a commercial pill-cutting device consisting of an immobilized razor blade and cutting surface and instructed to give the entirety of 1 tablet over 4 days to account for any variability in the uniformity of drug dispersion. This approach implied that some cats might have received a variable daily dose of study drug. The dosage of RVX was chosen on the basis of prior investigations14; however, it is possible that the drug may have been more effective if given at a higher milligram per kilogram dose or if given twice daily instead of once. Apixaban, a similar FXa-inhibiting drug, is given twice daily in humans, while the human-labeled dosage for RVX is once daily.19 Future pharmacokinetic and pharmacodynamic investigations may help to elucidate an optimal dosage for cats.
There were no significant changes in several echocardiographic parameters over time or between treatment groups in the present study, with the exception of significantly greater LA:Ao in the RG compared to the CG at 6 and 12 months. It is plausible that delay or prevention of ATE recurrence might have allowed time for the progression of underlying cardiac disease, leading to gradual increases in LA size. However, given that there were no differences in prevalence of ATE recurrence, time to ATE recurrence, or any survival time between treatment groups, differences in LA:Ao between treatment groups would not necessarily be anticipated. An association between body weight and LA:Ao in cats has been described20; however, there was no significant change in body weight in either group throughout the study. In fact, the CG cats had a greater average weight than the RG cats at the time points for which a significantly greater LA:Ao ratio was noted in RG cats (Table 3). Although there were no significant differences in the proportions of cats with specific cardiac diagnoses between groups, there were not enough patients with specific types of cardiac disease to allow stratification on the basis of diseases (eg, hypertrophic CM vs restrictive CM) that might progress at variable rates.
The presence of intracardiac SEC was more commonly noted in the RG compared to the CG, but only at the 12-month time point. It is difficult to parse the clinical significance of this single finding, as cats in both treatment groups had SEC detected at varying points of the study. Spontaneous echocardiographic contrast refers to a swirling pattern of blood flow detected as increased ultrasonic backscatter and is believed to represent the aggregation of cellular blood elements. Spontaneous echocardiographic contrast can be seen in patients with anemia, low-flow states, and turbulent blood flow and may be due to rouleaux formation or blood protein (eg, fibrinogen, immunoglobulin)–induced aggregation.21 The presence of SEC is considered an indicator of an activated hemostatic system,3 and detection of circulating biomarkers of platelet hyperactivity or procoagulant excess in cats that also have SEC may identify those cats most at risk of future ATE.
The QOL surveys indicated a relatively rapid return to normal function in the cats of this study, even in those with some residual paresis from the ATE. While there was clearly inclusion bias favoring cats that survived the initial ATE episode, the return to normal QOL argues against the clinical approach of a blanket euthanasia recommendation in all cats presenting with ATE. With appropriate pain management, supportive care, and client education regarding the chronic management of heart disease, some cats with ATE can be nursed back to health, especially if they survive the acute presentation and impacts of reperfusion injury. Notably, both groups in the current study (and the clopidogrel-treated cats in the previously cited report6) experienced a median survival time of > 1 year following the initial ATE event, which is valuable information for owners of cats with acute ATE.
An initial goal of the present study was to collect blood from enrolled cats for measurement of plasma drug concentrations, with the hopes of providing more data about the pharmacokinetics of both clopidogrel and RVX in cats with clinical heart disease receiving multiple medications. While some samples were collected and analyzed, the variability in time of collection from the last administered dose was so great that conclusions, other than the identification of the presence of plasma drug, were unable to be drawn.
As with any clinical trial, owners of study animals were assumed to have administered medication according to the instructions provided. While the plasma drug analysis did support administration in the animals from which samples were received, there were likely variations in the administration of medications (eg, missed days), and we cannot speculate about the impact of this real-world dosing. There is a theoretical concern in humans that abrupt discontinuation of RVX may result in rebound hypercoagulability,22 but this has not been studied in cats. On the bimonthly surveys, some owners did note a single missed dose, but this notation was uncommon. While we do not have evidence that any subjects went for long periods without the study drug, we cannot confirm this, as there was no requirement for owners to return unused drugs or keep medication logs during the study.
On the basis of the data generated in this investigation, the use of once-daily oral clopidogrel (18.75 mg/cat) or RVX (2.5 mg/cat) monotherapy following recovery from ATE results in similar survival times, prevalence of ATE recurrence, and time to ATE recurrence. When compared to information from a previous report, both drugs appear to be superior to single-agent aspirin therapy for secondary ATE prevention. These comparable outcomes suggest a complex mechanism of prothrombotic imbalance with activated platelets and procoagulant factors contributing to overt thrombus formation. The increasingly common clinical practice of empirical treatment with the combination of clopidogrel and RVX following ATE should be tested in controlled randomized clinical trials that specifically compare dual therapy to monotherapy with either agent alone, considering outcomes related to both recurrence and safety (eg, risk of bleeding).23
Supplementary Materials
Supplementary materials are posted online at the journal website: avmajournals.avma.org.
Acknowledgments
The authors acknowledge the owners and cats involved in this study and thank them for their conscientious correspondence and willingness to take part in this study. The authors acknowledge Dr. Deborah Keys, Kaleidoscope Statistics, for assistance with statistical analysis and Dr. Garrett Pachtinger for internet management and programming. The authors also acknowledge the following veterinarians and technicians for their assistance in case evaluation, recruitment, and data collection: Kathy Hoover, Katie Hodges, Lydia Moss, Gregg Rapoport, Erin Achilles, Michael Cocchiaro, Andrea Eriksson de Rezende, Amy Dixon-Jimenez, Luis dos Santos, Dena Fauske, Logan Funk, Teresa DeFrancesco, Arianne Fabella, Tiffany Fallaw, Trevor Gerlach, Lori Hitchcock, Terry Huh, Rebecca Kennerly, Allison Klein, Shianne Koplitz, Kendra LaFauci, Eva Larouche-Lebel, Michel Lesser, Kerry Loughran, Heidi MacLean, Eric de Madron, Nekesa Morey, Gianninne Nader, Michael Pabon, Rosie Payne, Megan Poad, Yamir Reina-Doreste, Vanessa Roman, Samantha Salmon, Gretchen Singletary, Amelia Sinkin, Julia Treseder, Carl Toborowski, Yuki Tse, Kristen Verdoorn, Rachel Van Zile, and Kelly Wiggen.
Disclosures
Dr. Brainard is a member of the JAVMA Scientific Review Board, but was not involved in the editorial evaluation of or decision to accept this article for publication.
Dr. Brainard has received research funding from or provided professional services for Companion Animal Health and Zoetis Animal Health. Dr. Coleman has received research funding from Boehringer Ingelheim. Dr. Rush has received research funding or travel support from, given sponsored lectures for, or provided professional services for Aratana Therapeutics, Boehringer Ingelheim, Ceva Animal Health, Elanco, Increvet, and Nestlé Purina PetCare. Dr. Hogan has no active conflicts; however, he has received prior funding from Bristol Myers Squibb-Sanofi and Boehringer Ingelheim. Drs. Kurosawa, Brooks, and Kraus do not report any conflicts of interest.
No AI-assisted technologies were used in the generation of this manuscript.
Funding
This investigation was funded by the Morris Animal Foundation (D16FE-015).
ORCID
B. M. Brainard https://orcid.org/0000-0003-4299-2936
References
- 1.↑
Paige CF, Abbott JA, Elvinger F, Pyle RL. Prevalence of cardiomyopathy in apparently healthy cats. J Am Vet Med Assoc. 2009;234(11):1398-1403. doi:10.2460/javma.234.11.1398
- 2.↑
Fox PR, Sisson D, Moise NS. Feline cardiomyopathy. In: Fox PR, ed. Textbook of Canine and Feline Cardiology: Principles and Clinical Practice. 2nd ed. W. B. Saunders Co; 1999:621-678.
- 3.↑
Stokol T, Brooks M, Rush JE, et al. Hypercoagulability in cats with cardiomyopathy. J Vet Intern Med. 2008;22(3):546-552. doi:10.1111/j.1939-1676.2008.0098.x
- 4.↑
Payne JR, Borgeat K, Connolly DJ, et al. Prognostic indicators in cats with hypertrophic cardiomyopathy. J Vet Intern Med. 2013;27(6):1427-1436. doi:10.1111/jvim.12215
- 5.↑
Smith SA, Tobias AH, Jacob KA, Fine DM, Grumbles PL. Arterial thromboembolism in cats: acute crisis in 127 cases (1992-2001) and long-term management with low-dose aspirin in 24 cases. J Vet Intern Med. 2003;17(1):73-83.
- 6.↑
Hogan DF, Fox PR, Jacob K, et al. Secondary prevention of cardiogenic arterial thromboembolism in the cat: the double-blind, randomized, positive-controlled feline arterial thromboembolism; clopidogrel vs. aspirin trial (FAT CAT). J Vet Cardiol. 2015;17(suppl 1):S306-S317. doi:10.1016/j.jvc.2015.10.004
- 7.↑
Tablin F, Schumacher T, Pombo M, et al. Platelet activation in cats with hypertrophic cardiomyopathy. J Vet Intern Med. 2014;28(2):411-418. doi:10.1111/jvim.12325
- 8.↑
Smith CE, Rozanski EA, Freeman LM, Brown DJ, Goodman JS, Rush JE. Use of low molecular weight heparin in cats: 57 cases (1999-2003). J Am Vet Med Assoc. 2004;225(8):1237-1241. doi:10.2460/javma.2004.225.1237
- 9.↑
Hogan DF, Andrews DA, Green HW, Talbott KK, Ward MP, Calloway BM. Antiplatelet effects and pharmacodynamics of clopidogrel in cats. J Am Vet Med Assoc. 2004;225(9):1406-1411. doi:10.2460/javma.2004.225.1406
- 10.↑
Alwood AJ, Downend AB, Brooks MB, et al. Anticoagulant effects of low-molecular-weight heparins in healthy cats. J Vet Intern Med. 2007;21(3):378-387.
- 11.↑
Lo ST, Walker AL, Georges CJ, Li RH, Stern JA. Dual therapy with clopidogrel and rivaroxaban in cats with thromboembolic disease. J Feline Med Surg. 2022;24(4):277-283. doi:10.1177/1098612X211013736
- 12.↑
Kubitza D, Becka M, Voith B, Zuehlsdorf M, Wensing G. Safety, pharmacodynamics, and pharmacokinetics of single doses of BAY 59-7939, an oral, direct factor Xa inhibitor. Clin Pharmacol Ther. 2005;78(4):412-421. doi:10.1016/j.clpt.2005.06.011
- 13.↑
Patel MR, Mahaffey KW, Garg J, et al.; ROCKET AF Investigators. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med. 2011;365(10):883-891. doi:10.1056/NEJMoa1009638
- 14.↑
Dixon-Jimenez AC, Brainard BM, Brooks MB, et al. Pharmacokinetic and pharmacodynamic evaluation of oral rivaroxaban in healthy adult cats. J Vet Emerg Crit Care (San Antonio). 2016;26(5):619-629. doi:10.1111/vec.12524
- 15.↑
Weinz C, Buetehorn U, Daehler HP, et al. Pharmacokinetics of BAY 59-7939—an oral, direct Factor Xa inhibitor—in rats and dogs. Xenobiotica. 2005;35(9):891-910. doi:10.1080/00498250500250493
- 16.↑
Kubitza D, Becka M, Wensing G, Voith B, Zuehlsdorf M. Safety, pharmacodynamics, and pharmacokinetics of BAY 59-7939—an oral, direct Factor Xa inhibitor—after multiple dosing in healthy male subjects. Eur J Clin Pharmacol. 2005;61(12):873-880. doi:10.1007/s00228-005-0043-5
- 17.↑
Mueck W, Borris LC, Dahl OE, et al. Population pharmacokinetics and pharmacodynamics of once- and twice-daily rivaroxaban for the prevention of venous thromboembolism in patients undergoing total hip replacement. Thromb Haemost. 2008;100(3):453-461.
- 18.↑
Coleman AE, DeFrancesco TC, Griffiths EH, et al. Atenolol in cats with subclinical hypertrophic cardiomyopathy: a double-blind, placebo-controlled, randomized clinical trial of effect on quality of life, activity, and cardiac biomarkers. J Vet Cardiol. 2020;30:77-91. doi:10.1016/j.jvc.2020.06.002
- 19.↑
Kreutz R, Persson PB, Kubitza D, et al. Dissociation between the pharmacokinetics and pharmacodynamics of once-daily rivaroxaban and twice-daily apixaban: a randomized crossover study. J Thromb Haemost. 2017;15(10):2017-2028. doi:10.1111/jth.13801
- 20.↑
Häggström J, Andersson ÅO, Falk T, et al. Effect of body weight on echocardiographic measurements in 19,866 pure-bred cats with or without heart disease. J Vet Intern Med. 2016;30(5):1601-1611. doi:10.1111/jvim.14569
- 21.↑
Ralph AG, Saunders AB, Hariu CD, Nabity M. Spontaneous echocardiographic contrast in three dogs. J Vet Emerg Crit Care (San Antonio). 2011;21(2):158-165. doi:10.1111/j.1476-4431.2011.00624.x
- 22.↑
Brainard BM, Buriko Y, Good J, Ralph AG, Rozanski EA. Consensus on the Rational Use of Antithrombotics in Veterinary Critical Care (CURATIVE): Domain 5—discontinuation of anticoagulant therapy in small animals. J Vet Emerg Crit Care (San Antonio). 2019;29(1):88-97. doi:10.1111/vec.12796
- 23.↑
Mansi CD, Severa JR, Wilhelm JN, et al. Dual antithrombotic therapy dose-dependently alters hemostatic plug structure and function. J Thromb Haemost. 2024;22(4):1016-1023. doi:10.1016/j.jtha.2023.12.017
