Prevalence of Anticoagulant Rodenticide Exposure in Red-tailed Hawks (Buteo jamaicensis) in New York State and Diagnostic Utility of Russell Viper Venom Test for Detecting Associated Coagulopathies

Cynthia Hopf-Dennis (  crh245@cornell.edu ) Cornell University College of Veterinary Medicine https://orcid.org/0000-0001-6400-7066 Sarrah Kaye Staten Island Zoo Nicholas Hollingshead Cornell University College of Veterinary Medicine Marjory Brooks Cornell University College of Veterinary Medicine Elizabeth Bunting Cornell University College of Veterinary Medicine Noha Abou-Madi Cornell University College of Veterinary Medicine

. Anticoagulant rodenticides exert their action by preventing the hepatic recycling of vitamin K and thereby impeding the post-translational processing of coagulation factors II, VII, IX, and X that is required for procoagulant complex assembly. Following development of resistance to rst-generation toxins, such as warfarin, second-generation AR were developed (Jackson andKaukeinen 1972, Buckle et al. 1994). Second-generation ARs have been shown to have longer elimination half-lives in serum and liver of mice, are more potent, and only need to be consumed once to produce death (Vandenbroucke et al. 2008, Buckle and Prescott 2018). As an example, brodifacoum showed a plasma elimination half-life of 91.7 days and liver half-life of 307.4 days in laboratory mice (Vandenbroucke et al. 2008). Second generation rodenticides include brodifacoum, bromadiolone, difenacoum, diphacinone, and chlorophacinone. Currently brodifacoum and bromadiolone are two products that are available for use by licensed pesticide applicators only (US EPA).
Anticoagulant rodenticide applications are highly regulated by the United States Environmental Protection Agency (US EPA) but are still frequently used for pest control ( High-level exposure via ingestion of these products results in uncontrolled bleeding, hypovolemic shock, and death , DuVall et al. 1989, Murphy and Gerkin 1989. In were included in the study. Good health was ascertained based on the animal's history, physical exam, hematological and biochemical pro les. This group of birds had been maintained in captivity for more than 2 months with a controlled rodent food source. These RTHs were classi ed as "control" and assumed to have no exposure to ARs due to controlled food source and housing conditions. None of the birds in the control group died during the study period. The clinical cohort consisted of 48 free-ranging red-tailed hawks that presented to the Cornell University Janet L. Swanson Wildlife Hospital (SWH) in Ithaca, NY from June 2016 to May 2018. Due to the volume of blood required to be collected from individual birds on presentation, red-tailed hawks weighing > 0.80 kg and in condition to sustain the blood collection were enrolled in this study. Intake information including presenting complaint and any history available about each bird was collected from the members of the public presenting the birds. A complete physical examination was performed on each bird including careful assessment for any signs of hemorrhage greater than what would match the extent of trauma seen on exam. When possible, necropsies were also performed on free-ranging red-tailed hawks that either died or were euthanized during the study period. The necropsy information was detailed with gross and histopathological results.
Free-ranging RTHs were classi ed into groups based on anticoagulant testing results. They were considered "free-ranging AR exposed" if AR was detected in blood and/or liver. "Free-ranging AR not exposed" designated RTHs that had no liver or blood evidence of AR exposure, and "Free-ranging AR unknown exposure" was assigned to RTHs that were blood AR negative but were not necropsied (unknown liver exposure).

Sampling Methods
To minimize the in uence of pre-analytic processing on coagulation test results, blood collection technique was standardized and performed in the same manner for all birds. In free-ranging RTHs, blood was collected within 24 hours of admission to the hospital. Blood sampling was often performed at the rehabilitation facility or permanent housing for control birds. The birds were restrained in left lateral recumbency with the head covered loosely with a towel or hooded to decrease stress. Approximately 3.5 ml of blood was collected from the right jugular vein using a 3cc syringe and a 22-gauge needle. Gentle pressure was applied to the phlebotomy site until hemostasis was achieved and then birds were released to their carrier or holding pens. The collected blood was immediately aliquoted into (1) one tube containing one-tenth volume sodium citrate anticoagulant (total volume 0.2 mL citrate + 1.8ml blood) for whole blood AR testing, (2) an EDTA tube (0.5-1ml blood) frozen for blood AR analysis, (3) a micro-EDTA tube (0.25ml blood) for hematology analysis, and (4) a third micro-heparin tube (0.6ml blood) for biochemistry testing. The citrate and heparin tubes were spun down immediately, and the supernatant plasma collected. Citrated plasma was frozen for later analysis. Heparinized plasma was refrigerated along with whole blood EDTA for submission and analysis within 24 hours for hematology analysis and biochemistry.
If any clinical birds included in the study died spontaneously or were euthanized, 1x1 cm portions of liver tissue were collected for an anticoagulant screening panel through the Michigan Animal Health Diagnostic Laboratory. A subset of the birds was submitted for complete necropsy when necropsy time was available, and the carcasses were in adequate condition to provide good results. Complete necropsy was performed by the Cornell University Animal Health Diagnostic Laboratory on 5/26 birds that died.

Sample Analyses
Samples were submitted for complete blood count, biochemical pro le, coagulation testing (RVVT, brinogen, and PT), and anticoagulant rodenticide analysis.

NY.
Coagulation testing (PT, RVVT, brinogen) was performed at the Comparative Coagulation Lab at Cornell University, Ithaca, NY. The PT and RVVT were performed following previously described methods used to monitor anticoagulant effects of ARs in avian (bobwhite and owl) toxicity studies (Rattner et al. 2010, 2012, Rattner, Horak et. al. 2014. Brie y, the PT and RVVT tests were performed using aliquots of citrate plasma diluted in an equal volume of imidazole buffered saline (pH 7.4) and warmed to 37°C before the addition of a coagulation trigger reagent and calcium. Thromboplastin for the PT assay was prepared in-house from avian (chicken) brain tissue via acetone extraction as previously described. (Rattner et al 2010) For the PT assay, 100 uL of dilute plasma was incubated for 60 seconds, followed by the addition of 200 uL of the thromboplastin reagent containing 25 mM CaCl 2 . The RVVT assay was con gured with a commercially available venom reagent (Russell' viper venom, Sigma, St. Louis, MO). For the RVVT, 100 uL of dilute plasma was incubated with an equal volume venom reagent (prediluted 1:10 in buffered saline) for 30 seconds, followed by the addition of 100 uL 25 mM CaCl 2 . Fibrinogen concentration was measured in undilute plasma using a commercial thrombin reagent (Fibrinogen, Diagnostica Stago, Parsippany NJ) via Clauss method (Clauss 1957) calibrated to the manufacturer's human brinogen standard. A single batch of thromboplastin, and single lots of commercial reagents were used for the study. The assays were performed in automated ( brinogen) or semi-automated (PT, RVVT) coagulation instruments with mechanical endpoint detectors (ST4 and STACompact, Diagnostica Stago, Parsippany, NJ). The upper limit for clot detection for both PT and RVVT was set at 180 seconds. Samples that had not clotted were assigned a clotting time of 180 seconds.
When available, frozen whole blood and liver tissues were submitted for anticoagulant rodenticide analyses to the Michigan State Animal Health Diagnostic Laboratory. The panel included the following rst-and second-generation anticoagulants: brodifacoum, bromadiolone, chlorphacinone, difenacoum, difethialone, diphacinone, and warfarin. Lower cutoff for detection for ARs were as follows: <0.002ppm for brodifacoum, <0.07ppm for difethialone, and <0.02ppm for bromadiolone, chlorphacinone, difenacoum, diphacinone, and warfarin in whole blood or liver.

Statistical analysis
The distributions of coagulation times (PT, RVVT), and brinogen concentrations were assessed for homogeneity of variance (Leven's test) and for normality (Shapiro-Wilk test, normal probability plot, descriptive statistics) and log-transformed when appropriate. If transformations were applied, distributions of logtransformed values were reassessed for homogeneity of variance and normality. Coagulation times (PT, RVVT) and brinogen concentrations in free-ranging AR exposed, free-ranging AR not exposed, and control RTHs were compared between groups using one-way analysis of variance (ANOVA). Post hoc comparisons using the Tukey multiple comparison test were made when appropriate. The relationship between PT and RVVT was assessed using Pearson correlation and Deming regression analysis. The relationships between liver brodifacoum concentration and coagulation times (PT, RVVT) were assessed using linear regression. All statistical analyses were performed using GraphPad Prism version 7.05 (GraphPad Software, La Jolla California USA, www.graphpad.com)

Population Description, Presenting Causes and Hematology and Biochemistry Analyses
From 2016 to 2018, 72 RTHs were included in this study. The control population included 23 captive RTHs. Collected samples from these birds are summarized in Figure 1a. Blood from 20 out of the 23 control birds was analyzed for hematology and 21 were analyzed for biochemistry These ndings are summarized in Tables 1a and 1b. Two control RTHs did not have a full biochemistry and three did not have a full hematology due to insu cient sample volume. Results from blood chemistry panel and hematologic analyses in the control birds fell within normal ranges reported for this species (Species 360).
The free ranging population included 49 wild birds and the causes for presentation are summarized in Table 2. Unknown reason for presentation was the leading complaint followed by trauma (including vehicular trauma, trauma of unknown origin, and fractures of unknown origin), and lastly infectious diseases. Full serum biochemistry was obtained in 49/49 birds and full hematology was obtained in 48/49 birds. These ndings are summarized in Tables 3a and 3b. One free-ranging RTH was not included in hematology analysis due to insu cient sample volume. Available samples are summarized in Figure 1a. Hematology abnormalities included anemia, leukocytosis, heterophilia, band heterophils, monocytosis, eosinophilia, basophilia, toxic changes, reactive lymphocytes, polychromasia, and the presence of hemoparasites (leukocytozoon, hemoprodius, and plasmodium). Serum biochemistry abnormalities included hyperuricemia, hyperphosphatemia, hypoproteinemia, hyperglycemia, elevated AST, CK, lipemia, hemolysis, and icterus.

Anticoagulant Rodenticide Exposure and Clinical Signs
Twenty-three control birds were examined, and none exhibited clinicopathologic abnormalities consistent with AR toxicity during the study period. Control birds did not have blood submitted for AR analysis.
Results from anticoagulant rodenticide serum and liver residues from free-ranging RTHs are reported in Table 4. Anticoagulant rodenticide exposure was assessed in 35/49 free-ranging birds. AR exposure was assessed in blood only in 19/35 (54%) birds, in liver only in 1/35 (0.03%) birds and in both liver and blood in 15/35 (42%) birds. Liver residue was only assessed in RTHs that died or were euthanized. Eleven of sixteen (68%) livers tested for AR exposure were positive. Difethialone was found in 1/16 (6%) liver samples, and brodifacoum was detected in 10/16 (62%) liver samples. Insu cient sample sizes precluded residue blood testing in all free-ranging RTHs. Available samples were summarized in Figure 1b. Difethialone was found at a concentration of 0.18 ppm and brodifacoum concentrations ranged from 0.003-0.234 ppm. Two of 34 (6%) RTH assessed for blood rodenticide had brodifacoum in blood with measured concentrations of 0.003 and 0.006 ppm. One of these birds had overt bleeding at presentation. No other AR was detected in blood or liver analyses.

Coagulation testing
Coagulation panels including PT, RVVT, and brinogen levels were completed on 22 control birds and 31 free ranging birds. Results of these tests are summarized in Tables 4-6. One control RTH was excluded from coagulation panels due to insu cient sample volume. Six free ranging birds were excluded from this analysis due to clotting of the sample that rendered them unsuitable for coagulation testing and 12 free-ranging RTHs were not analyzed due to insu cient sample size. Of the 31 free-ranging birds for which coagulation panel results were available, RVVT and PT assay clotting times exceeded the upper limit of detection of 180 seconds for 1 bird that was excluded from statistical analysis. The samples included in coagulation test results are summarized in Figure 1c. Median (Interquartile range (IQR); range) for PT, RVVT, and brinogen in control, free-ranging RTHs with AR residues detected in the liver, free-ranging RTHs with no AR residues detected in the liver, and free-ranging RTHs with no AR analysis are summarized in Table 6.
There was no evidence of a difference in log-transformed PT between control (n=22), free-ranging AR exposed (n=9), and free-ranging AR not exposed RTHs (n=5) (ANOVA F( 2,33 ) = 0.08181, p = 0.9216). There was also no convincing evidence of a difference in log-transformed RVVT between the three groups (ANOVA F( 2,33 ) = 2.128, p = 0.1351). However, there was moderate evidence of a difference in log-transformed brinogen concentrations between the thee groups (ANOVA F( 2,33 ) = 6.17, p = 0.0053). Post hoc comparisons using the Tukey multiple comparison test indicate that log-transformed brinogen concentrations differed between the free-ranging AR exposed birds and both the free-ranging AR not exposed birds (p-value = 0.0117) and the control birds (p = 0.0129). However, there was no evidence of a difference between control birds and free-ranging AR not exposed birds (p = 0.5619) ( Figure 2).

Correlation between PT and RVVT assays
A total of 22 control birds and 30 free ranging birds were used to assess the relationship between PT and RVVT. Correlation between log-transformed PT and log-transformed RVVT was weak (Pearson's r = 0.6212). The Deming regression also showed strong proportional bias ( gure 3). The slope of the Deming regression line was 2.715 (95% CI 1.816 to 3.614) and the y-intercept was -2.441 (95% CI -3.699 to -1.183).
Brodifacoum levels and coagulation time A total of 8 free-ranging AR exposed (liver) RTHs were used to assess the relationship between brodifacoum liver residues and PT or RVVT times. There was no evidence of a linear dependence of PT on brodifacoum liver residues (Test of slope <> 0, F 1,6 =2.286, p=0.1813), nor was there evidence of a linear dependence of RVVT on brodifacoum liver residues (Test of slope <> 0, F 1,6 =0.1399, p=.7213) ( gure 4).

Necropsies
Twenty-one animals were euthanized and ve died. Full necropsy results were available for ve free ranging birds summarized in Table 7.

Discussion
Toxicity secondary to AR exposure is still widely reported in free-ranging raptors despite restrictions on sale and use of certain products and knowledge of its In 2008 the Environmental Protection Agency prohibited the sale of brodifacoum products in some retail outlets due to the exposure risks to animals and humans, but these products are still widely available online and in agricultural supply stores. Evidence of continued exposure to second generation ARs in free-ranging RTHs in New York State as detected by blood and liver analysis is presented in this study. Second generation rodenticides were the only products identi ed in hawks in this study. The second generation ARs were introduced in the United States in the 1970s in response to resistance of rodents to rst generation rodenticides like warfarin (Buckle et al. 1994). Second generation rodenticides like brodifacoum have a longer half-life in animal tissues longer than rst generation products and this increases the possibility that predators like the red-tailed hawks would ingest contaminated prey species and accumulate the toxins in their liver ( Two major clinical challenges to diagnosing coagulation de ciencies in birds is the lack of physical exam evidence and the lack of a validated patient side coagulation test. The lack of physical exam changes consistent with AR exposure was well demonstrated in this study with only one bird showing clinical signs of hemorrhage on presentation despite liver residue evidence in several birds. Hemostasis depends on the integrity of the vasculature, the health and number of the platelets, and appropriate activation of secondary coagulation pathways The clotting cascade is the enzymatic process by which secondary coagulation is achieved. The vitamin K dependent factors of this pathway rely on a vitamin K-dependent carboxylation of glutamic acid residues for them to bind Ca 2+ and to become functional. Anticoagulant rodenticides inhibit the vitamin K epoxide reductase enzyme complex that is needed to recycle vitamin K, resulting in vitamin K de ciency and the production of inactive coagulation proteins (Harvey 2012). As this is an enzymatic reaction, a certain threshold likely needs to be met before the system is overwhelmed and the body is no longer able to produce functional proteins for the clotting cascade. There were several birds that did not appear to have coagulopathies at presentation, but had measurable stores of anticoagulant in their livers. It is likely that their vitamin K production was inhibited due to this exposure but the inhibition had not yet reached the threshold needed to impair the vitamin K dependent coagulation factors in order to see overt hemorrhage. Only one bird in our study, which died shortly after presentation, was found to have overt bleeding on presentation which correlated to subcutaneous and visceral hemorrhage on necropsy and PT/RVVT values of >180 seconds, along with a liver brodifacoum liver residue of 0.234ppm. It is likely that RTHs exposed to high levels via acute secondary exposure would die from overt hemorrhage prior to arrival at a wildlife hospital.
In this study, we investigated the use of avian optimized PT and RVVT as clinically relevant diagnostic tests to establish rodenticide exposure in free ranging RTHs. Several tests are available for diagnosing AR exposure in mammals, but these tests are impractical or not applicable to avian patients. A recent study assessing the e cacy of a commercially available point of care device (Coag-Sense ® ) used to assess PT in mammals is not suitable for rapid assessment in birds of prey (Dickson 2020). Commercially available toxicological screening for blood rodenticide levels requires a large volume of blood and takes several (3)(4)(5) days before results are available. Prothrombin time (PT) and the activated prothrombin time (aPTT) are the most commonly used coagulation tests for AR exposure in mammals. As a test of the extrinsic, or tissue factor dependent, coagulation pathway, PT measures the function of factors II, VII, and X and clotting time increases if factor VII is depleted. Commercially available PT tests use mammalian thromboplastin as the test reagent. These assays produced unreliable and imprecise results in avian patients, so avian thromboplastin is recommended for PT measurements in birds (Kase, 1978, Morrisey, 2003. PT tests using avian thromboplastin have been developed for research purposes, and in rodenticide feeding trials, PT is a sensitive marker of acute AR coagulopathy in birds of prey (Rattner, et al., 2010(Rattner, et al., , 2011(Rattner, et al., , 2012. However avian-speci c PT assays are experimental and are not available to clinical practitioners.
We compared PT and RVVT in RTHs in controlled rehabilitation environments versus free-ranging RTHs presented to a wildlife hospital. The ranges reported here for PT (18-39.7sec) and RVVT (11.5-91.8sec) in control RTHs are higher than previously reported values of free-ranging RTHs admitted to another wildlife hospital (Hindmarch et al. 2019). We report poor correlation between these two tests despite previous reports that have shown prolonged PT and RVVT times with exposure of ARs in controlled environments (Rattner et al. 2011(Rattner et al. , 2012. . We also investigated the relationship between PT, RVVT, and blood or liver residues of ARs found in exposed free ranging RTHs. Feeding trials in Eastern screech owls and American kestrels report prolonged PT and RVVT within 48 and 96 hr following a 50 mg/kg dose of diphacinone and a correlation between dose consumed and liver concentrations (Rattner et al. 2011(Rattner et al. , 2012. In this study we found no signi cant relationship between liver AR residue levels and PT or RVVT times in exposed RTHs. Because these RTHs were wild there was no way to control or measure dosage or time from exposure to sampling liver or blood. The doses in oral exposure in controlled feeding trials may be much larger than those RTHs are exposed to from prey species in the wild. Hepatic tissue has been shown to be a reliable indicator of AR accumulation in birds through both wild caught and controlled studies (Thomas et  . This discrepancy in exposure compared to determined cause of death is common and was found in this study as well. Rodenticide was only detected in the blood of two birds in this study (6%) and only one bird was clinical for AR toxicosis. Exposure studies relying on blood and not tissue samples for AR exposure have reported similarly lower prevalence of ARs compared to tissue studies. Kwasnoski et al report AR exposure in 11% of serum samples collected from wild red-tailed hawks (Kwasnoski et al. 2019). This lower prevalence is likely because the half-life of these products in blood is much shorter than in tissue like the liver (Anderson et al. 2011). The blood half-life of brodifacoum, the most commonly used AR in rats, is short, and blood levels may no longer be detectable at the time of clinical presentation in many species (Erickson and Urban, 2004). In this study we report postmortem liver AR residues in The majority of the birds used as controls were housed in rehabilitation settings and included previously wild RTHs that were overwintered prior to release. When selecting our control birds, we made the assumption that they were not exposed to ARs based on the inclusion criteria of having been maintained in captivity for at least 2 months, being fed a controlled food source (laboratory bred rodents), and the lack of potential exposure to rodenticide on the property. It was stipulated that the control birds be free of signs of systemic disease as determined by a normal physical examination, and normal hematology and biochemistry results. The time frame of two months was established early in the study under the assumption that most of the rodenticide exposure would be coming from rst generation anticoagulants which have a half-life of hours to days in mammalian tissue and a short elimination time shown in experimentally exposed Screech Owls (Megascops asio) (Fisher et al. 2003, Rattner andHorack et al. 2014). However, as was found in this study, the second generation anticoagulants are still very present in wild RTHs and the limit of two months may mean that some control red-tailed hawks included in this study could have had rodenticide residue in their livers. Classi cation of RTH as exposed and unknown exposure was based on detection of ARs in blood or liver. We did not submit liver samples for AR analysis in RTH that survived and we acknowledge the fact that their negative blood AR results may be due to the short half-life of circulating AR, thus their classi cation as "unknown exposure". For this reason, these birds were included in a separate category. In retrospect, a liver biopsy to measure hepatic AR residues on all birds in this study would have improved the interpretation of the data but was beyond the limits of the project. We found that there was no statistical difference between PT and RVVT between control and wild exposed RTH.
In this study, we described the physical ndings, baseline bloodwork and necropsy ndings when applicable, for red tailed hawks admitted to a wildlife hospital that may have led to the identi cation of the cause for presentation. Trauma, including vehicular strikes and gunshot wounds, toxicities, starvation, infections, parasitic infestations, and orphaning are the most common reasons for RTH admissions to a wildlife hospital and known prior history are usually limited ( In our study, RTHs that were seen being hit by a vehicle or were found with acute trauma on the side of the road were considered cases of vehicular trauma. Some of the birds with fractures of unknown origin were found by roads but had chronic wounds so this relationship was suspected but could not be con rmed. Birds with trauma of unknown origin were found in various places including yards, elds, or buildings and the original trauma could not be con rmed as vehicular. Three birds were con rmed to have Aspergillus sp. infections. All three birds with Aspergillus sp. infections were found to have other comorbidities. One had bilateral uveitis and anemia, the second had conjunctivitis and stomatitis, and the third had central neurologic lesions and was suspected to be a case of vehicular head trauma. The condition of fourteen of the hawks was classi ed as unknown as the presentation and history did not reveal a speci c presenting cause. Infectious diseases were overrepresented in necropsies performed in this study but RTHs presenting with overt trauma such as head trauma or fractures were not submitted for full necropsy due to funding constraints. Hematologic and biochemical analysis were performed on control and wild RTHs in this study. Control birds had values within normal reference range expected for this species (Species 360). Changes to blood work found in wild RTHs were re ective of common causes of presentation previously reported for wild birds to wildlife hospitals including evidence of trauma (AST and CK elevation) evidence of infection (leukocytosis, heterophilia, band heterophils, monocytosis, eosinophilia, basophilia, toxic changes, reactive lymphocytes, polychromasia, and the presence of hemoparasites). Four of the wild RTHs were found to have PCV <=22% on presentation. Two of these were found to have AR exposure, one in blood and one in liver. One of the four birds was not tested for AR due to loss of sample. One of these AR exposed birds was presented for pelvic limb paresis as suspected hit by car trauma, and the other was presented as an unknown trauma with an elbow luxation. AR exposure should be considered a differential in anemic RTHs regardless of the presenting complaint.
The total population of wild-caught RTHs in this study were presented to the hospital by the general public, DEC o cers, or wildlife rehabilitators which meant that the number of wild caught RTHs was dependent on external sources. The birds had to weigh more than 0.80 kg and needed to be stable enough for blood collection upon admittance to the hospital. Many RTHs presented underweight and severely dehydrated, excluding them from the study. Despite our efforts to select individuals that were stable, hypovolemia may have complicated our ability to obtain proper samples for analysis. The right jugular vein was used as the preferred vessel for collecting a large volume of blood. If the bird struggled, or multiple venipuncture were required, it was very likely that pre-analytical artifact from clotting factor activation rendered the sample invalid. Eight samples were excluded in this study due to visible clots prior to PT and RVVT analysis.

Conclusion
Birds of prey are exposed to vitamin K-antagonist anticoagulant rodenticides (ARs) through consumption of intoxicated rodent prey. While acute, high-dose rodenticide exposure causes severe and typically fatal hemorrhage in raptors, the cumulative effect of repeated, low-level exposure is unknown. PT and RVVT were not found to correlate in our control group and no correlation was found between liver AR residue and either coagulation test in this study. PT and RVVT were not found to be sensitive markers of AR exposure in the RTH presented to the wildlife hospital but the widespread exposure to secondary ARs continues to be a problem in this free-ranging population. Without a sensitive test to detect early subclinical coagulopathies, AR toxicity in birds remains underdiagnosed and untreated. The data from this study can be utilized by governmental agencies to determine the impact on free ranging wildlife and help bolster efforts to restrict the use of these pesticides in the environment. This study also provides a foundation for future research on the use of PT or RVVT in a clinical setting to monitor patients being treated for AR toxicity and the possible use of liver biopsies to study the subclinical effects of AR exposure on long term tness of wild birds of prey.

Declarations
Acknowledgements: The authors would like to thank the staff of the Janet L. Swanson Wildlife Hospital as well as the staff in the Wildlife Health Lab and Comparative Coagulation Lab who assisted with obtaining and processing the samples used in this study. The authors would also like to thank Tatiana Weisbrod for her invaluable help in establishing this project.

Funding-This study was funded by the Wiederhold Foundation at the Cornell University College of Veterinary Medicine
Con icts of interest/Competing interests-The authors declare that there was no con ict of interest or competing interests.

Availability of data and material-Not applicable
Code availability-Not applicable Ethics approval-This study was approved by the Cornell University Animal Care and Use Committee Consent to participate-Not applicable Consent for publication-This paper and all its contents were review and approved for publication by all contributing authors   Table 6. Median, IQR, and range for coagulation time (sec) and brinogen (mg/dL) in control birds, wild AR exposed birds, free-ranging AR not exposed birds, and free-ranging unknown exposure birds.  West Nile virus infection with myocardial necrosis (acute and severe) and vascular necrosis-con rmed by IHC of brain (extensive), heart, kidney and serosas. Hemoparasitism, probably due to Haemoproteus sp, marked. Intestinal coccidiosis, marked, probably due to Sarcocystis sp. 47 euthanasia cerebral necrosis, focally extensive, with hemorrhage-possible larva migrans or trauma. Lung microgranuloma with nematode larvae-probable larva migrans. Pneumonia, interstitial, brinous, focally extensive (mediastinum) with associated mediastinal pleuritis and air sacculitis. Hepatitis, multifocal, mild to moderate, chronic-active (possibly also associated with larva migrans) c. Description of hematology and biochemistry results available for control and free ranging red-tailed hawks (RTHs). RVVT, PT and brinogen in control, free-ranging AR exposed, free-ranging AR not exposed, and free-ranging unknown exposure RTHs using the Tukey method for plotting whiskers and outliers (Tukey, 1977).