An atypical presentation of anticoagulant rodenticide toxicosis in a dog

Abstract

A 5-month-old intact female Australian shepherd dog was referred to our clinic for neurologic signs including ataxia, a head tilt, and altered mentation following consumption of an unidentified rodenticide several days prior to developing clinical signs. A provisional diagnosis of bromethalin toxicosis had been made, given the neurologic signs seen and the general increased use of bromethalin-containing rodenticide products. However, on physical examination, the dog was noted to have scleral hemorrhage and bleeding at the venipuncture sites, which was inconsistent with bromethalin toxicosis. Coagulation testing was supportive of anticoagulant rodenticide toxicosis and the rodenticide was later identified as the first-generation anticoagulant rodenticide diphacinone. The neurologic signs seen were attributed to a coagulopathy causing multifocal hemorrhage into the central nervous system. Neurologic signs rapidly resolved following treatment with a frozen plasma transfusion and vitamin K₁. This atypical presentation of an anticoagulant rodenticide toxicosis highlights the need for accurate product identification, if available, and thorough patient examination and laboratory testing. An atypical presentation of anticoagulant rodenticide toxicosis should be considered when neurologic signs are present with clinical bleeding, especially if the type of rodenticide is unknown, or even if it was not thought to have an anticoagulant as the active ingredient.

Key clinical message:

Given the change in commercially available rodenticide products, this case highlights the need for accurate product identification in cases of suspected toxicosis, and the variable clinical signs that can be seen following anticoagulant rodenticide toxicosis.

Rodenticides were among the top 10 most-reported toxin exposures in 2020 [American Society for the Prevention of Cruelty to Animals (ASPCA) New York, New York, USA; unpublished data, 2021]. Categories of common rodenticides used (Table 1) include anticoagulant rodenticides and non-anticoagulant rodenticides (1). According to the Environmental Protection Agency (EPA) ruling on rodenticide risk mitigation in 2008, rodenticide products commercially available to consumers in the United States may not contain second-generation anticoagulant rodenticides. The ruling outlines that rodenticides can contain first-generation anticoagulants or a non-anticoagulant such as bromethalin or cholecalciferol (2,3). Since this ruling, there has been a shift in the types of rodenticide toxicoses reported to the ASPCA. Between the years 2010 and 2020, there has been a reported increase in bromethalin and cholecalciferol exposures and a decrease in anticoagulant rodenticide exposures (Table 2).

Table 1.

Classification of select rodenticides.

Rodenticide	Type	Generation	Short- or long-acting	Multiple- or single-feeding	Available to individual consumers for residential use?	Antidote
Warfarin	Anticoagulant	First	Short	Multiple	Yes	Vitamin K1
Chlorophacinone	Anticoagulant	First	Long	Multiple	Yes	Vitamin K1
Diphacinone	Anticoagulant	First	Long	Multiple	Yes	Vitamin K1
Bromadiolone	Anticoagulant	Second	Long	Single	No	Vitamin K1
Difethialone	Anticoagulant	Second	Long	Single	No	Vitamin K1
Brodifacoum	Anticoagulant	Second	Long	Single	No	Vitamin K1
Difenacoum	Anticoagulant	Second	Long	Single	No	Vitamin K1
Bromethalin	Non-anticoagulant	NA	NA	Single	Yes	None
Cholecalciferol	Non-anticoagulant	NA	NA	Multiple or single	Yes	Bisphosphonates

NA — Not applicable.

Table 2.

Changes to percentages of canine rodenticide exposure cases by type over time.

Rodenticide type	2010	2020
Anticoagulant rodenticides	75.67%	37.12%
Bromethalin	19.36%	51.73%
Cholecalciferol	0.83%	9.37%

Anticoagulant rodenticide toxicosis can present with a variety of clinical signs, some of which can mimic other toxicities, depending on where the hemorrhage occurs (4,5). As the numbers of bromethalin and cholecalciferol exposures are on the rise, accurate identification of the rodenticide product is critical for timely and effective treatment and for determining the expected prognosis. Direct treatment of an anticoagulant rodenticide toxicosis is possible with the antidote vitamin K₁. Additional treatment to replace clotting factors may also be required. Many rodenticides do not have an antidote available; therefore, treatment relies on symptomatic and supportive care once an animal develops clinical toxicosis (1,6).

This report describes a dog that was presented with neurologic signs after consumption of, at the time, an unidentified rodenticide. With the increase in non-anticoagulant rodenticide exposures following the 2008 EPA ruling on rodenticide risk mitigation, the neurologic signs were initially assumed to be secondary to bromethalin toxicosis, which carries a poor prognosis once moderate-to-severe neurologic signs develop (7). Over the course of treatment, the rodenticide was identified as a first-generation, long-acting anticoagulant that resulted in presumptive multifocal central nervous system (CNS) hemorrhage leading to the atypical presentation of neurologic clinical signs. Given the change in rodenticide products available for purchase by an individual consumer for residential use in the United States and Canada, this case highlights the need for accurate product identification in cases of suspect toxicosis, and the variable clinical signs that can be seen following anticoagulant rodenticide toxicosis. Anticoagulant rodenticide toxicosis should be suspected when there is evidence of clinical bleeding concurrent with other clinical signs, including neurologic signs.

Case description

A 5-month-old intact female Australian shepherd dog weighing 18.4 kg was presented to our emergency service as a referral for suspected bromethalin rodenticide toxicosis. The dog was initially taken to a primary care veterinarian (pDVM) for lethargy and limping on the right pelvic limb. On physical examination, the referring clinician noted soft tissue swelling of the right hock, placed a compression bandage for support, and discharged the dog. No medications were given or prescribed during that visit. Overnight, the dog developed vomiting and ataxia, which prompted re-presentation to the pDVM 24 h after the initial visit. Given the progression of the clinical signs, the owners had searched their property and reported finding partially chewed blocks of a Tomcat-brand rodenticide (Bell Laboratories, Windsor, Wisconsin, USA). The active ingredient of the rodenticide was not known to the pDVM at the time of the visit because the packaging was not available. The exact time of ingestion of the rodenticide was unknown but was suspected to be several days before development of clinical signs, based on when the dog had unsupervised access to the area where the rodenticide was found. Upon repeat presentation to the pDVM (48 h after the initial onset of clinical signs), the dog was described as quiet and responsive to stimuli, but with an altered mentation. Scleral hemorrhage was noted in the left eye. A neurologic examination showed that the dog was ataxic and had occasional muscle tremors in the shoulders and hind limbs. An intravenous catheter was placed in the right cephalic vein and blood was drawn for a complete blood (cell) count (CBC) and biochemical profile. During the catheter placement, a large hematoma developed. Laboratory results showed an anemia with a hematocrit of 26% [reference range (RR): 35 to 55%] with a normal platelet count of 320 K/μL (RR: 175 to 500 K/μL) (Table 3). Hypernatremia was the only abnormality noted on the chemistry panel (Na: 158 mmol/L, RR: 141 to 150 mmol/L). Given the clinical signs, a presumptive diagnosis of bromethalin toxicosis was made and the dog was referred to the emergency service for continued treatment. Prior to referral and despite the differential diagnosis of bromethalin toxicosis, the dog was given a single, 60-milligram dose of vitamin K₁ subcutaneously (3.25 mg/kg).

Table 3.

Select hematologic and coagulation parameters.

Assessment date and time	HCT (%)	PCV (%)	TS (mg/dL)	Plt (K/μL)	PT (s)	PTT (s)
4/23 12:30	26.6	—	—	320	—	—
4/23 16:00	26	24	6.6	243	43.7	33
4/24 9:00	—	27	7.0	—	7.5	—
4/25 9:00	—	29	7.0	—	—	—

HCT — Hematocrit; PCV — Packed cell volume; Plt — Platelets; PT — Prothrombin time; PTT — Partial thromboplastin time; TS — Total solids.

Upon presentation to the referral emergency department (48 h after the initial onset of clinical signs), the animal was quiet but responsive. She had a right-sided head tilt and was falling primarily to the right. A proprioceptive ataxia with reduced-to-absent paw placement in all 4 limbs was also noted. The previously placed compression bandage on the right hind limb was removed and revealed mild edema of the right hock that was underneath the bandage. The fur over the jugular veins had been shaved; this region was edematous on palpation, with no apparent bruising. Visualization of the reported hematoma at the right cephalic catheter site was limited due to the wrap used to secure the intravenous catheter. Scleral hemorrhage was noted in the left eye; however, no further evidence of petechiae or bruising was noted (Figure 1). The rest of the physical examination was within normal limits.

Figure 1.

Photographs of a 5-month-old female Australian shepherd dog with a right-sided head tilt and scleral hemorrhage in the left eye (shown enlarged at right). No further evidence of petechiae or bruising was noted.

Initial diagnostic tests carried out upon presentation to the referral emergency department included an abdominal and thoracic point-of-care ultrasound that was unremarkable. A chemistry panel and CBC were repeated upon admission to the critical care unit. The CBC confirmed an anemia with a hematocrit of 26% (RR: 39 to 57%). Hemoglobin was also low at 8.6 g/dL (RR: 14.0 to 21.0 g/dL). The chemistry profile showed a low sodium, of 138 mmol/L (RR: 141 to 150 mmol/L) and a low potassium, of 3.8 mmol/L (RR: 3.9 to 5.3 mmol/L). Total protein (TP) at the time of admission was 6.6 mg/dL (RR: 5.6 to 8.0 mg/dL). A coagulation profile obtained 3 h after vitamin K₁ administration showed a prolonged prothrombin time (PT), at 43.7 s (RR: 7.0 to 9.4 s) and partial thromboplastin time (PTT), at 33.0 s (RR: 8.5 to 13.8 s) (Table 3).

By the time of admission to the referral emergency department, the owner was able to identify that the package of rodenticide had an active ingredient of diphacinone, a first-generation vitamin K antagonist rodenticide, which was consistent with the animal’s coagulation panel and clinical signs. The presumptive diagnosis at this time was anticoagulant rodenticide toxicosis, and the neurologic signs were suspected to be secondary to multifocal hemorrhage within the brain and spinal cord. The dog was hospitalized and treated with a transfusion of 275 mL of frozen plasma over 4 h; as well as an additional, subcutaneous injection of vitamin K₁ (Bimeda-US; Animal Health, Oakbrook Terrace, Illinois, USA), 55 mg (3 mg/kg); followed by oral vitamin K₁ (VetOne; MWI Animal Health, Boise, Idaho, USA), 50 mg (2.7 mg/kg) orally, q12h.

Twenty-four hours after admission, the PT was reassessed and had normalized (7.5 s, RR: 7.0 to 9.4 s). A recheck packed cell volume (PCV) was static at 26% (RR: 35 to 55%), with a TP of 7.0 g/dL (RR: 4.8 to 7.2 g/dL). Upon repeated physical examination, the dog was bright, alert, and responsive, with normal mentation. The right-sided head tilt was still present, but improved, and no proprioceptive deficits or ataxia were observed. Forty-eight hours after admission, the animal’s mentation continued to be normal. The right-sided head tilt was almost completely resolved and no other neurologic signs could be appreciated. The PCV and TP were reassessed and were 29% and 7.0 g/dL, respectively. The dog was discharged with oral vitamin K₁ supplementation [50 mg (2.7 mg/kg) to be given orally, q12h for 30 d, with a high-fat meal]. It was recommended to recheck her coagulation panel 2 to 3 d after discontinuing the vitamin K₁ supplementation, to ensure extending the duration of therapy was not indicated.

Discussion

All anticoagulant rodenticides have a similar mechanism of action via the inhibition of the vitamin K epoxide reductase enzyme, which converts vitamin K from an inactive form to its active form. In its active form, vitamin K carboxylates, and therefore activates, coagulation Factors II, VII, IX, and X. Inhibition of vitamin K epoxide reductase enzyme causes depletion of the active vitamin K₁, which results in the depletion of the vitamin K-dependant factors. As the factors are depleted, coagulopathies occur and can present as a variety of vague and nonspecific symptoms. Commonly, patients present with acute dyspnea due to bleeding into the thoracic cavity, petechia, ecchymosis, epistaxis, or hematemesis. Patients can also present with general lethargy, weakness, and exercise intolerance. Other clinical signs develop depending on where hemorrhage occurs. Bleeding into the joint space or muscles can result in limping and, although less common, hemorrhage into the CNS can present as ataxia and other neurologic signs (4,5,8). Treatment with vitamin K₁ results in a replenishment of vitamin K stores, allowing for activation of the vitamin K-dependent coagulation factors and cessation of bleeding.

Early anticoagulant rodenticides (warfarin, chlorophacinone, diphacinone) are first-generation anticoagulants (9). Brodifacoum, bromadiolone, difenacoum, and difethialone are second-generation anticoagulant rodenticides that are formulated to be more potent against warfarin-resistant rodents (Table 1).

In 2008, the United States EPA released a risk-mitigation ruling stating that second-generation anticoagulants are not available to general consumers but are still available for agricultural use and pest control operators. Similarly, in January 2013, Health Canada implemented restrictions on residential use of second-generation anticoagulants in an effort to reduce exposure to nontarget animals and children. There are no restrictions on the use of bromethalin or first-generation anticoagulant rodenticides (10). Due to these rulings, many commercially purchased rodenticides now contain a non-anticoagulant rodenticide, such as bromethalin or cholecalciferol, as the main active ingredient (2,3). Unfortunately, bromethalin does not have an antidote, and prognosis following high-dose bromethalin toxicosis with development of moderate-to-severe neurologic signs is grave (7). Bisphosphonates are an available antidote for cholecalciferol toxicosis. However, treatment of clinical cholecalciferol toxicosis can be prolonged, which is not feasible for all owners (11). Given the change in commercially available rodenticide products, pets that are presented for ingestion of a rodenticide with an unknown active ingredient pose a unique diagnostic and treatment challenge. Especially in cases where the active ingredient of the ingested rodenticide is unknown, atypical clinical signs can lead to a delay or inaccurate diagnosis, which may result in morbidity and mortality.

In this case, the dog presented with nonspecific lethargy and limping on a hind limb, which progressed to ataxia and a head tilt. The possible rodenticide ingestion was not reported until 24 h after the initial presentation. At that time, the active ingredient in the rodenticide was unknown. Examining the visual appearance of a rodenticide is not an accurate way to identify its type, as many products are prepared as blue-green pellets or blocks despite having different active ingredients (1,4,9). Additionally, many manufacturers produce different types of rodenticides containing different active ingredients. Given this dog’s progression of neurologic signs, the main differential diagnosis that prompted the referral was bromethalin toxicosis, and the owners were informed that the prognosis for that type of toxicosis is poor. Although some evidence of coagulopathy was present on physical examination by the pDVM, these signs were underappreciated due to the neurologic signs. The reported muscle tremors were suspected to be fasciculations from generalized weakness, as true muscle tremors are not expected with anticoagulant rodenticide toxicosis.

Rapid changes in sodium levels can cause neurologic signs. This dog was noted to have markedly different sodium levels measured at the primary care clinic versus the referral center. However, as no treatments were initiated that would cause a change in sodium, the differences between the 2 values were likely spurious.

By the time of presentation to the referral emergency department, the rodenticide was identified as the first-generation anticoagulant rodenticide diphacinone. Coagulation testing done at the referral hospital was supportive of anticoagulant rodenticide toxicosis. Anticoagulant rodenticides cause prolongation of the PT and PTT values, which progresses over time. Due to depletion of Factor VII, which has the shortest half-life of the vitamin K-dependent factors in dogs, the PT is the first to become prolonged. The PT elevations occur 24 to 36 h after ingestion of long-acting anticoagulant rodenticides. Clinical hemorrhage is thought to occur secondary to a decrease in Factor II, which occurs after the decrease in Factor VII. Therefore, clinical bleeding is not seen until 3 to 4 d after ingestion, as the half-life of Factor II is 41 h (12). By the time clinical hemorrhage is present at 3 to 4 d after ingestion, the PT value is typically “out of range” and the PTT is either markedly prolonged or also “out of range” (13). An activated clotting time (ACT) will only be prolonged once PTT is prolonged, and is therefore not as useful in diagnosing anticoagulant rodenticide toxicosis before clinical bleeding is present. In this case, both the PT and PTT were elevated (4.6 and 2.4 × above the high end of the reference interval, respectively) on presentation, which was supportive of anticoagulant rodenticide toxicosis. Administration of vitamin K₁ subcutaneously at 3 h before measurement of the coagulation times may have affected these values. One dose of 5 mg/kg vitamin K₁ has been shown to normalize PT within 1 h when given intravenously to dogs with anticoagulant rodenticide toxicosis (13). Less is known about the time to normalization of PT following subcutaneous vitamin K₁ without concurrent blood product transfusions. However, dogs that received 0.83 mg/kg of vitamin K₁ subcutaneously, q8h for 3 doses, followed by 0.83 mg/kg orally, q8h for 3 doses, had normal PT when tested at 48 h after initiation of vitamin K₁ therapy (14). The PT values before administration of vitamin K₁ were not available in this dog. Measurement of anticoagulant rodenticide levels in whole blood can be used to confirm that clinical signs are due to anticoagulant rodenticide toxicosis; however, there is a long turnaround time to obtain results for this type of test, and treatment often must be initiated on an emergent basis for patients with clinical bleeding. Therefore, in this case, with the identification of the active ingredient from the package and evidence of clinical bleeding, treatment with vitamin K₁ and frozen plasma was initiated and supported by further confirmatory testing.

The recommended dose and route of administration of vitamin K₁ for an animal with anticoagulant rodenticide exposure is variable. In those exposed to long-acting or second-generation anticoagulant rodenticides, the recommended dose of vitamin K₁ is 1.25 to 3.5 mg/kg q12h (15). Significant adverse effects of proper administration of vitamin K₁ are unlikely. Vitamin K₁ is most commonly administered orally or subcutaneously. When given orally, there is an increased bioavailability when administered with a fatty meal. The intramuscular route is not recommended due to the risk of causing further hemorrhage, whereas the intravenous route is typically avoided due to the risk of acute hypersensitivity reactions mediated by solubilizing agents (e.g., polyoxyethylated fatty acid derivatives), which are used to dissolve vitamin K into a liquid form for injection (16) Intravenous use of vitamin K₁ could be considered when using mixed micelle formulations (Vitamin K₁; Laboratoire TVM, TVM UK Animal Health, Oxfordshire, United Kingdom), as hypersensitivity reactions have not been reported with these formulations. However, the mixed micelle formulations are not readily available outside of the United Kingdom. The dog in this case received an initial dose of 3.25 mg/kg of vitamin K₁, subcutaneously, before presentation to the referral hospital. The dog then received an additional 3 mg/kg of vitamin K₁, subcutaneously, at the time of admission to the referral hospital, to bring her total daily dose of vitamin K₁ above 5 mg/kg. The subcutaneous route was chosen over oral administration for the initial dose of vitamin K₁ due to the neurologic signs. The animal was then transitioned to oral vitamin K₁ for maintenance as her neurologic signs improved. The duration of vitamin K₁ therapy is determined based on the type of rodenticide involved. Short-acting anticoagulant rodenticides, such as warfarin, are treated with an initial 2 wk of vitamin K₁ therapy; long-acting anticoagulant rodenticides are treated with 3 to 4 wk of vitamin K₁ therapy. For any type of anticoagulant rodenticide, it is important to recheck a PT value at 48 to 72 h after stopping vitamin K₁ therapy, to ensure that this value remains normal and extension of the course of vitamin K₁ therapy is not needed.

Many patients with clinical bleeding from anticoagulant rodenticides also benefit from transfusions as part of their treatment. Provision of active clotting factors through the transfusion of whole blood, fresh-frozen plasma, frozen plasma, or cryoprecipitate is recommended in patients with active, life-threatening hemorrhage. In this case, 15 mL/kg of frozen plasma was chosen to replace active clotting factors. Frozen plasma maintains adequate levels of the vitamin K-dependent clotting factors (FII, FVII, FIX, FX) (17) and is often available at a reduced cost compared to fresh-frozen plasma due to the longer shelf life and loss of the labile clotting factors that limits the utility of frozen plasma for other coagulopathies. Beyond replacement of active clotting factors, autotransfusion, packed red blood cells, or whole blood can be given to replace oxygen carrying capacity if patients have anemia. Treatment with frozen plasma to provide activated clotting factors as well as vitamin K₁ allowed for the successful treatment and rapid recovery of this dog.

Given the changes in commercially available rodenticide products, this case highlights the need for accurate product identification in cases of suspect rodenticide toxicosis, and the variable clinical signs that can be observed following anticoagulant rodenticide toxicosis. An atypical presentation of anticoagulant rodenticide toxicosis should be considered when neurologic signs are present with clinical bleeding, even if the type of rodenticide is unknown or an anticoagulant is not thought to be the active ingredient. CVJ