Chemical Name: 3-[3-(4'-bromo-[1,1'-biphenyl]-4-yl)-1,2,3,4-tetrahydro-1- naphthalenyl]-4-hydroxy-2H-1-benzopyran-2-one.
Synonyms: Brodifacoum is the approved common name. Talon® and PESTOFF® are trade names
Brodifacoum is one of the most widely used rodenticides worldwide. It has been used in New Zealand to control possums since the early 1990s. On islands, aerial application techniques are used. On mainland New Zealand it is used in cereal baits in bait stations. In January 2000 the Department of Conservation announced plans to reduce the field use of brodifacoum on the mainland.
It is essential that wildlife or livestock do not gain access to areas where brodifacoum is being used. Brodifacoum can persist (>1 year) in the liver and kidneys of sub-lethally poisoned wildlife or livestock. Hence it is important that the risk of contamination of wildlife or livestock is recognised and the product is used carefully to minimise non-target contamination.##
2.1.1 Physical and chemical properties
The empirical formula for brodifacoum is C31H23BrO3 and the molecular weight is 523.4. It is an off-white to fawn-coloured odourless powder with a melting point of 228–232ºC. It is of very low solubility in water (less than 10 mg/L at 20ºC and pH 7). Brodifacoum is slightly soluble in alcohols and benzene, and soluble in acetone. It is stable at room temperature. Commercial concentrated solutions of brodifacoum are available for bait manufacturers. 2.1.2 Historical development and use
Brodifacoum is a synthetic compound that was developed a few decades ago. It is structurally related to a naturally occurring coumarin that causes haemorrhagic syndrome in cattle eating improperly cured or mouldy sweet clover. The rodenticidal properties of brodifacoum were described in the early 1970s. It is a very potent anticoagulant active against rats and mice, including strains resistant to warfarin and other anticoagulants (Rennison & Hadler 1975). A single ingestion of 1 mg/kg is usually sufficient to kill. In New Zealand it is used principally to control possums and rats, though it has also been used for rabbits (Williams et al. 1986a,b). In January 2000 the Department of Conservation took steps to reduce the mainland field use of brodifacoum because of concerns relating to contamination of birds and game (Eason et al. 1999c). Because of the tendancy for uncontrolled exposure of non-targets through secondary poisoning (Eason et al. 1999c), the suggested practice of secondary poisoning of stoats (Brown et al. 1998) is not recommended, particularly in areas where game may be hunted for human consumption.
Brodifacoum has been used successfully in recent rodent eradication programmes on New Zealand’s offshore islands to protect populations of endangered indigenous birds (Taylor & Thomas 1989, 1993; Buckle & Fenn 1992; Robertson et al. 1993; Towns et al. 1993). In addition to its use to control and eradicate rats, brodifacoum has been successfully used in ground-laid baits or in baits placed in bait stations to eradicate rabbits (Merton 1987; Towns et al. 1993), to control wallabies (D. Moore pers. comm.) and brushtail possums (Eason et al. 1993b). Field use of brodifacoum-containing baits for rabbit or wallaby control has been discontinued in New Zealand.
Cereal baits (Talon®, Pestoff®) containing brodifacoum are used for rodent and possum control. For possum control the baits are best used to further reduce low possum numbers following use of fast-acting poisons (such as cyanide, 1080, or cholecalciferol) for the initial population reduction. The slow action of this poison overcomes the problems associated with bait shyness in areas where possum control has been sustained for many years.
2.1.3 Fate in the environment
Brodifacoum is most unlikely to be found in water even after aerial application of baits for rodent control on offshore islands. Brodifacoum is not mobile in soil and is extremely insoluble in water (<10 mg/L water at pH 7). When baits disintegrate, brodifacoum will be likely to remain in the soil, where it will be slowly degraded by soil micro-organisms. The half-life in soil varies from 12 to 25 weeks depending on the soil type. Microbial degradation will be dependent on climatic factors such as temperature, and the presence of species able to degrade brodifacoum. In leaching studies, 2% of brodifacoum added to soil leached more than 2 cm in four soil types tested (World Health Organisation 1995).
Since brodifacoum remains absorbed in soil when baits disintegrate, only the erosion of soil itself would see any brodifacoum reaching water, and even then brodifacoum would be likely to remain bound to organic material and settle out in the sediment. If baits were sown directly into streams or rivers, localised short-term contamination might occur.
2.1.4 Toxicology and pathology
Onset of symptoms The latent period between the time of ingestion and the onset of clinical signs varies considerably and in possums may take as long as 1–4 weeks (Littin et al. 2000). In rats the onset of symptoms and death usually occur within a week. Clinical signs reflect some manifestations of haemorrhage. Onset of signs may occur suddenly; this is especially true when haemorrhage of the cerebral vasculature or pericardial sac occurs. Clinical signs commonly include anaemia and weakness. Haemorrhaging may be visible around the nose, mouth, eyes, and anus of mammals. When pulmonary haemorrhage has occurred, blood-tinged froth may be visible around the nose and mouth. Swollen, tender joints are common and if haemorrhage involves the brain or central nervous system, ataxia or convulsions can occur. Poisoned animals die of multiple causes related to anaemia or hypovolemic shock. Possums respond significantly more slowly with onset of toxicosis occurring between 2 and 3 weeks after dosing. In general, possums appear to be less sensitive to anticoagulants, which may be due to species differences in the ability to metabolise xenobiotics (Olkowski et al. 1998) or difference in the half-lives of vitamin K-dependent clotting factors, or vitamin K epoxide reductase receptor binding.
Mode of action
Brodifacoum, like other anticoagulant toxicants, acts by interfering with the normal synthesis of vitamin K-dependent clotting factors in the liver of vertebrates (Hadler & Shadbolt 1975). In the liver cells the biologically inactive vitamin K1-2,3 epoxide is reduced by a microsomal enzyme into biologically active vitamin K, which is essential for the synthesis of prothrombin and other clotting factors (VII, IX, and X). Brodifacoum antagonism of the enzyme vitamin K1-epoxide reductase in the liver causes a gradual depletion of the active form of the vitamin, and consequently of vitamin K-dependent clotting factors, which results in an increase in blood-clotting time until the point where no clotting occurs.
The greater potency of second-generation anticoagulants such as brodifacoum compared to first-generation anticoagulants such as warfarin and pindone is likely to be related to their greater affinity for vitamin K-epoxide reductase and subsequent accumulation and persistence in the liver and kidneys after absorption (Huckle et al. 1988). Anticoagulants share this common binding site, but the second-generation anticoagulants have a greater binding affinity than the first-generation compounds (Parmar et al. 1987). All tissues that contain vitamin K-epoxide reductase (e.g. liver, kidney, and pancreas) are target organs for accumulating these toxicants.
Pathology and regulatory toxicology
Generalised haemorrhage is frequently evident at post-mortem. Areas commonly affected are the thoracic cavity, subcutaneous tissue, stomach, and intestine. The heart is sometimes rounded and flaccid with subepicardial and subendocardial haemorrhages. Histomorphological analysis of the liver may reveal centrilobular necrosis as a result of anaemia and hypoxia. In possums, post-mortem findings range from mild to moderate haemorrhage in some limbs and in the gastrointestinal tract, to extensive haemorrhage throughout the body and major organs.
Brodifacoum is a slight skin irritant and a mild eye irritant in the rabbit. Various in vitro and in vivo studies (including the Salmonella reverse mutation assay, the forward mutation assay using mouse lymphoma cells, and the micronucleus test in mice) have been undertaken to assess the genotoxic potential of brodifacoum. No mutagenic activity was detected. Brodifacoum, when given by oral gavage to female rats at daily dose levels of 0.001, 0.01, or 0.02 mg/kg body weight during days 6–15 of pregnancy, caused no evidence of adverse developmental effects on the foetuses. Higher daily doses (above 0.05 mg/kg) caused an anticoagulant effect in the dams, which resulted in a high incidence of abortion.
Pregnant female rabbits were given oral gavage doses of 0.001, 0.002, or 0.005 mg brodifacoum/kg body weight per day from days 6–18 of pregnancy. At the highest dose level a high proportion of maternal deaths occurred as a result of haemorrhage. Although the survivors showed signs of haemorrhage, there were no effects on the developing foetuses.
On the basis of these studies, brodifacoum can be classified as non-mutagenic and lacking in tetratogenic potential. In a 5-day study in rats, a no-observed-effect level for brodifacoum was 0.02 mg/kg/day (WHO 1995).
Fate in Animals Absorption, metabolism and excretion of brodifacoum compared with other anticoagulant toxicants 3
Brodifacoum is absorbed through the gastrointestinal tract. It can also be absorbed through the skin (Table 9).
Table 9. Acute toxicity (LD50 mg/kg) of brodifacoum in rats (Hone & Mulligan 1982)
Species Route LD50 (mg/kg)
Rat (oral) 0.27
Rat (dermal) 50.00
After absorption, high concentrations in the liver are rapidly established and remain relatively constant. Disappearance from serum is slow with a half-life in rats of 156 hours or longer. The slow disappearance from the plasma and liver and the large liver:serum ratio probably contribute to the higher toxicity of brodifacoum when compared with warfarin or pindone (Bachmann & Sullivan 1983). It is apparent that a proportion of any ingested dose of brodifacoum bound in the liver, kidney, or pancreas remains in a stable form for some time and is only very slowly excreted.
In contrast to brodifacoum, warfarin will undergo relatively extensive metabolism. The metabolites will be more polar (water soluble) than the parent compounds and therefore more readily excreted in the urine.
Brodifacoum, like other second-generation metabolites, is not readily metabolised and the major route of excretion of unbound compound is through the faeces. Enterohepatic recirculation, the process that allows drugs and pesticides that have been absorbed to return to the gastrointestinal tract from the liver via the biliary tract, undoubtedly plays an important role.
Tables 10 and 11 present comparative data on the persistence of anticoagulants. Kelly & O’Malley (1979) reported the mean half-life for disappearance of warfarin from the plasma of human volunteers given a single oral dose of 0.5–100 mg/kg body weight varied from 24 to 58 hours. No dose-level effect on half-life was apparent even over this large range of doses. Second-generation anticoagulants are much more slowly cleared from the bloodstream.
In a comparative study in rabbits Breckenridge et al. (1985) reported plasma elimination half-lives of 5.6 hours for warfarin, 83.1 hours for difenacoum, and 60.8 hours for brodifacoum. There are very limited data on the influence of dosage on elimination. However, in the case of bromadiolone, a dose-dependent increase in plasma elimination half-life from 25.7 to 57.5 hours was reported after the oral dosage was increased from 0.8 to 3 mg/kg in rats (Kamil 1987).
Woody et al. (1992) observed an elimination half-life for brodifacoum in serum of 6 4 days in four dogs. The plasma half-life of brodifacoum determined in three patients with severe bleeding disorders was found to be approximately 16–36 days (Weitzel et al. 1990).
There are very limited data on the persistence of warfarin or pindone in the liver of animals. Two studies in non-rodent species indicate comparatively rapid clearance from the liver. Warfarin concentrations declined in pigs to very low concentrations after approximately 30 days, and concentrations were declining in those that received a lethal dose and those that survived (O’Brien et al. 1987).
In sheep receiving sub-lethal doses of bromadiolone (2 mg/kg), flocoumafen (0.2 mg/kg), and pindone (10 mg/kg), bromadiolone was detectable in the liver for 256 days and flocoumafen for 128 days. In contrast pindone was undetectable in the liver after 16 days (Nelson & Hickling 1994). Diphacinone, which is a close relative to pindone, appears to have a hepatic persistence profile more akin to that of second-generation anticoagulants. In cattle receiving a single injection of 1 mg/kg, almost constant residue concentrations were found in liver and kidney, 30, 60, and 90 days after dosing (Bullard et al. 1976). It is noteworthy that in persistence studies, and in risk assessment, limited consideration has been given to organs other than the liver. This is surprising considering that quite high concentrations of anticoagulants are found in the kidneys and lungs relative to other tissues some time after dosing.
Table 10. Persistence of first-generation anticoagulants
Blood t½ †
Liver retention ‡
Rat , Rat
Breckenridge et al. 1985
Eason et al. 1999b
O’Reilly et al. 1963
O’Brien et al. 1987
Nelson & Hickling 1994
Parmar et al. 1987
Bullard et al. 1976
† t½ for plasma or liver is the elimination half-life. It is convention to report the elimination t½ (-phase) rather than the -phase.
‡ Liver retention is expressed as the time period for which residues are reported to persist in the liver unless the value is preceded by t½. Plasma is t½ unless otherwise specified.
Table 11. Persistence of second-generation anticoagulants
Blood t½ (hours)†
(except where specified)
Liver retention‡ (days)
Parmar et al. 1987
Breckenridge et al. 1985
Parmar et al. 1987
Nelson & Hickling 1994
Huckle et al. 1989 Nelson & Hickling 1994
Huckle & Warburton 1989
Newton et al. 1990
Veenstra et al. 1991
Blood t½ (hours)†
(except where specified)
Liver retention‡ (days)
Bachmann & Sullivan 1983
Parmar et al. 1987
Breckenridge et al. 1985
Woody et al. 1992
Robben et al. 1998
Eason et al. 1996c,d
Laas et al. 1985
Weitzel et al. 1990
t½ 108 §
Lechevin & Poche 1988
Robben et al. 1998
† t½ for plasma or liver is the elimination half-life. It is standard convention to report the elimination t½ (-phase) rather than the -phase.
‡ Liver retention is expressed as the time period for which residues are reported to persist in the liver unless the value is preceded by t½. Plasma is t½ unless otherwise specified.
§ The half-life hepatic elimination for difethialone reported by Lechevin & Poche (1988) is unusually short for a second-generation anticoagulant, which suggests that difethialone may be unique. Brodifacoum was detected in the liver of sheep 128 days after oral administration (0.2 and 2.0 mg/kg body weight) in concentrations of 0.64 and 1.07 mg/kg dry weight (equivalent to 0.22 and 0.36 mg/kg wet weight), respectively. The peak levels which occurred at 2 days in the high-dose group and at 8 days in the low-dose group, were 6.50 and 1.87 mg/kg dry weight (2.21 and 0.64 mg/kg wet weight), respectively (Laas et al. 1985).
Parmar et al. (1987) found that elimination of radio-labelled brodifacoum, bromadiolone, and difenacoum from rat liver was biphasic, consisting of a rapid initial phase lasting from days 2 to 8 after dosing and a slower terminal phase when the elimination half-lives were 130, 170, and 120 days, respectively. Elimination of coumatetralyl was more rapid, with a half-life of 55 days.
Similar results for difenacoum were found by Bratt (1987). After a single oral 14C-difenacoum dose of 1.2 mg/kg body weight, the highest concentration of radioactivity (41.5% of the dose) was found in the rat liver 24 hours after dosing. The elimination from the liver was biphasic. The half-life of elimination of the radioactivity during the first rapid phase was 3 days, and for the slower phase was 118 days. A similar biphasic elimination was also apparent in the kidney. In the pancreas the concentration declined more slowly than in any of the other tissues (182 days). The parent compound was the major component in the liver 24 hours after dosing (42%).
Unchanged flocoumafen comprised the major proportion of the hepatic radioactivity in rats and was eliminated with a half-life of 220 days (Huckle et al. 1989). Veenstra et al. (1991) found retention of about 8% of an administered flocoumafen dose of 0.4 mg/kg in the liver of beagle dogs 300 days after dosing. Despite the more rapid metabolism of flocoumafen in Japanese quail, a proportion of the administered dose is retained in the liver, with a elimination half-life of 155 days after oral dosing (Huckle & Warburton 1989).
There are limited data on the persistence of anticoagulants in New Zealand native species. In a study in weta, brodifacoum persisted for approximately 1 week after dosing (Morgan et al. 1996a).
Species variation in response to brodifacoum
For second-generation anticoagulants like brodifacoum only a single dose is needed to induce death, if sufficient toxicant is ingested, and brodifacoum is extremely toxic in a number of animal species. The toxicity of brodifacoum varies between mammal species (Table 12) and bird species (Table 13).
In most mammals LD50 values are 1 mg/kg or less. Some higher values are reported in sheep and dogs, but there is considerable variability in these reports (LD50 in sheep 5–25 mg/kg and in dogs 0.25–3.56 mg/kg).
It has been suggested that anticoagulants are unlikely to affect invertebrates, which have different blood-clotting systems from vertebrates (Shirer 1992) and a New Zealand-based study has shown that brodifacoum lacks insecticidal properties in weta (Morgan et al. 1996a).
Table 12. Acute oral toxicity (LD50mg/kg) of brodifacoum for mammal species (Godfrey 1985; Eason et al. 1994a, Eason & Spurr 1995)
Species LD50 (mg/kg)
Bennett’s wallaby 1.3
Table 13: Acute oral toxicity (LD50mg/kg) of brodifacoum for bird species (Godfrey 1985)
Bird species LD50 (mg/kg)
Southern black-backed gull <0.75†
Canada goose <0.75†
Hedge sparrow >3.0‡
California quail 3.3
Mallard duck 4.6
Black-billed gull <5.0†
House sparrow >6.0‡
Australasian harrier 10.0
Ring-necked pheasant 10.0
Paradise shelduck >20.0‡
† Lowest dose tested ‡ Highest dose tested
Small birds such as silvereyes, sparrows, blackbirds, and California quail are considered more resistant to brodifacoum than some larger birds such as southern black-backed gulls, Canada geese, and pukeko (Godfrey 1985). However, some large birds, including Australasian harriers, ring-necked pheasants, and paradise shelducks, are also relatively resistant.
There are limited data on the aquatic toxicology of brodifacoum. In the unlikely event of a significant amount of brodifacoum bait being applied directly to a small stream, poisoning of aquatic invertebrates and fish could result. The EC50 from Daphnia magna (first instar) was 1.0 mg/kg after 24 hours of exposure and 0.34 mg/kg after 48 hours using 50 ppm pelleted baits. The LC50 (24 hours) for rainbow trout is 0.155 mg/L. The LC50s (96 hours) for rainbow trout and bluegill are 0.05 and 0.165 mg/L, respectively (World Health Organisation 1995.
2.1.5 Diagnosis and treatment of anticoagulant poisoning
Diagnosis of non-target poisoning in domestic animals
Diagnosis of anticoagulant toxicosis is based on exposure history, clinical signs, response to treatment, laboratory analyses, and in lethal cases, lesions. Differential diagnoses vary with the species involved, and include other causes of coagulopathy (clotting disorders) such as autoimmune thrombocytopenia (reduced platelet numbers), liver disease, and hereditary clotting factor deficiencies like Von Willebrand’s disease or Haemophilia A (Beasley et al. 1997d).
Although in some cases signs have been observed within 24 hours of ingestion, there is usually a lag period of 3–5 days between exposure and the onset of clinical signs of anticoagulant toxicosis. This delayed onset represents the time required to deplete hepatic stores of vitamin K, and reduce preformed vitamin K-dependent, clotting factor concentration in the plasma to the point of functional deficiency. Initial clinical signs of anticoagulant poisoning are usually characterised by depression/lethargy and anorexia, followed shortly by anaemia with pale mucous membranes, dyspnoea, exercise intolerance, and haemorrhaging from numerous sites, as evidenced by haematemesis (vomiting blood), epistaxis (blood from the nose), haemoptysis (bronchial or pulmonary bleeding), melaena (‘tarry’ faeces), and haematomas in various locations. Periarticular or intraarticular haemorrhage causing swollen joints and lameness is especially common in pigs, and abortion induced by placental haemorrhaging has been reported in cattle. Convulsions indicate bleeding into the central nervous system. Animals experiencing prolonged toxicosis may be icteric (jaundiced). Similar clinical signs occur in humans and include haematuria, bleeding gums, and easy or spontaneous bruising (Park et al. 1986).
As blood loss continues, cardiac murmurs, irregular heart beat, weak peripheral pulses, ataxia, recumbency, and coma will be observed. Death due to hypoxia and hypovolemic shock may occur from 48 hours to several weeks after exposure. Animals may occasionally be found dead with no premonitory signs, especially if severe haemorrhage occurs in the cerebral vasculature, pericardial sac, abdominal cavity, mediastinum, or thorax (Murphy & Gerken 1989; Felice & Murphy 1995).
Laboratory evaluation of suspect anticoagulant exposures in domestic animals includes measurement of packed cell volume (haematocrit), clotting parameters, and residue analysis.
The activity of vitamin K-dependent clotting factors (II, VII, IX, and X) is commonly measured using a suite of tests, including prothrombin time (PT), activated coagulation time (ACT), and activated partial thromboplastin time (APTT). Abnormal prolongation of PT is usually the earliest indicator of anticoagulant-induced coagulopathies, due to the involvement of factor VII in the coagulation pathway assessed by this clotting parameter. Factor VII has the shortest half-life of the vitamin K-dependent factors (6.2 hours in dogs), and is therefore the first to be depleted in plasma (Murphy & Gerken 1989). Elevations of PT from 2–6 times normal may occur within 24–48 hours of ingestion of a toxic dose. This is followed several hours later by elevation in APTT to 2–4 times normal values in cases of significant exposure. In general, changes in clotting parameter times are suggestive of anticoagulant exposure if they are prolonged beyond 25% of normal values. Assessment of coagulation parameters requires a sample of fresh, non-haemolysed blood collected in a sodium citrate (Blue Top) tube, stored at 4C, and submitted immediately. The diagnostic laboratory may require submission of a parallel sample from a ‘normal’, unexposed animal of the same species to serve as a control.
The onset and severity of clinical signs of anticoagulant toxicosis are usually linked with declines in packed cell volume, except in cases of massive, acute haemorrhage. Therefore, regular assessment of this end point is a useful tool to determine the appropriate course of treatment and to monitor progress.
Suspect anticoagulant exposures can often be confirmed by laboratory identification of toxicant residues in vomitus (only in cases of very recent ingestion, prior to the onset of clinical signs) or tissue. The antemortem sample of choice is whole blood or serum (residues are protein-bound), while liver is the best post-mortem sample. Blood samples should be stored at 4C. Liver specimens should be wrapped in foil, sealed in plastic, and shipped frozen.
Response to treatment
Anticoagulant-induced coagulopathies (clotting disorders) can be distinguished from other types of coagulopathies by clinical response to treatment with the specific antidote, vitamin K1. Tests used to assess coagulation time should indicate significant improvement in clotting ability within 12–24 hours of initiation of treatment, and should return to normal within 36–48 hours (Murphy 1999).
Post-mortem lesions resulting from anticoagulant rodenticide exposure are characterised grossly by generalised haemorrhage, especially in the thoracic or abdominal cavities, mediastinal space, periarticular tissues, subcutaneous tissues, subdural space, and gastrointestinal tract. Sudden deaths are often marked by massive haemothorax, haemopericardium, and pulmonary oedema or haemorrhage. The heart is often flaccid, with subepicardial and subendocardial ecchymoses. Centrilobular hepatic necrosis secondary to anaemia and hepatocellular hypoxia may be observed histologically (Osweiler 1996b; Beasley et al. 1997d).
Treatment of anticoagulant toxicosis in domestic animals
Companion animals usually present with signs of haemorrhage or anaemia, or with a history of recent ingestion of anticoagulant bait but no clinical effects. In the latter cases, either the dose ingested is insufficient to cause significant inhibition of vitamin K-dependent clotting factor production, or insufficient time has elapsed to deplete pre-exposure plasma clotting factor concentrations to the point of deficiency. Because treatment of anticoagulant toxicosis can be expensive (especially with large dogs exposed to second-generation products requiring prolonged therapy), animals presenting with a history of exposure but no clinical signs should be assessed carefully before treatment is initiated.
Therapeutic goals for veterinarians in the treatment of anticoagulant poisoning are (1) to decrease toxicant absorption; (2) to correct low haematocrit and/or hypovolemia; and (3) to correct clotting factor deficiencies. Recommendations for the treatment of anticoagulant toxicosis in companion animals are as follows (Mount & Feldman 1983; Murphy & Gerken 1989; Osweiler1996b; Beasley et al. 1997d):
Animal is presented asymptomatic, within several hours of suspected/confirmed oral exposure:
Induce emesis with household salt solution or washing soda crystals (if <3 hours) or perform gastric lavage.
Administer activated charcoal (1–2 g/kg) with a saline cathartic (magnesium sulphate at 250 mg/kg in 5–10 times as much water).
Decision to initiate vitamin K therapy depends on potential exposure dose and effectiveness of emesis. If suspected dose ingested is low (< 10% of LD50), may elect to decontaminate and release with instructions to monitor for signs of haemorrhage. If potential exposure is more significant, measure PT at 24, 48 and 72 hours after ingestion. If results are normal, defer treatment but monitor for 10–30 days (depending on the compound involved).
Animal is presented with signs of haemorrhage and/or anaemia:
Animals with a packed cell volume of <15% with severe bleeding, or with associated complications of anaemia, require clotting factor replacement immediately (Murphy 1999). Correct low haematocrit and/or hypovolemia, and provide clotting factors with IV transfusion of fresh whole blood or plasma at 10–20 mL/kg. The initial 25% of the volume is given relatively rapidly, and the remainder by slow drip.
Handle affected animals with care. Sedate as needed (avoid protein-binding drugs like promazine that may displace toxicant residues and exacerbate signs). Maintain core body temperature. Oxygen may be beneficial with severe dyspnoea. Give replacement IV fluids only after clotting factors are on board.
Initiate antidotal therapy using vitamin K1 (phytonadione, phylloquinone), which is the most effective form available. Vitamin K3 is not recommended. Hepatic bioavailability of oral vitamin K1 is greater than the parenteral form, so this route should be used unless contraindicated (e.g. in cases of vomiting, gastrointestinal haemorrhage, or concurrent administration of activated charcoal). If parenteral administration is required on initial presentation, approved vitamin K1 formulations can be given IV (slowly, over 15–20 minutes, with a small-bore needle), although this route is associated with frequent anaphylactoid reactions. The subcutaneous route is safer, but absorption is slow in dehydrated animals.
Recommended doses of vitamin K1 for companion animals range from 1 mg/kg (once a day) for first-generation products such as warfarin and pindone, to 2.5 mg/kg for bromadiolone, and 2.5–5.0 mg/kg for potent, second-generation anticoagulants such as brodifacoum and diphacinone. Specific doses are not available for flocoumafen and coumatetralyl, but a starting dose of 2.5 mg/kg is reasonable (Felice & Murphy 1995). Oral bioavailability is enhanced by concurrent feeding of a small fatty meal.
Oral vitamin K1 therapy must be maintained for as long as the toxicant is active in inhibiting vitamin K epoxide recycling. Recommendations for duration of treatment in companion animals range from 7–14 days for warfarin and pindone, to 21 days for bromadiolone, and 30 days for brodifacoum and diphacinone (Felice & Murphy 1995; Beasley et al. 1997d). In all cases, premature termination of treatment should be avoided, and prothrombin time should be measured 5–7 days after the end of the treatment period.
Avoid protein-bound drugs, elective surgery, strenuous exercise, and large volumes of fatty food during the convalescent period. Previously exposed animals may be more sensitive to subsequent anticoagulant exposure for weeks to months after recovery, due to biologically active residues in the liver.
2.1.6 Non-target effects
Brodifacoum has the potential to cause both primary and secondary poisoning of non-target species. However, as with the other vertebrate pesticides, the adverse effects of brodifacoum on wildlife are dependent more on how baits are used and the behaviour of non-target species than susceptibility of individual species to the toxin. Baits in bait stations are less accessible to non-target species than baits on the ground. Secondary poisoning of birds is likely where target species (e.g. rabbits and rats) are a major constituent of the diet (e.g. brown skua and harriers). Despite these distinctions, a wide range of small and large birds have been found dead from primary or secondary poisoning after field use of brodifacoum in New Zealand: saddlebacks, blackbirds, chaffinches, house sparrows, hedge sparrows, silvereyes, song thrushes, paradise shelducks, Australian magpies, robins, western weka, Stewart Island weka and brown skuas (Towns et al. 1993; Williams et al. 1986a,b; Taylor & Thomas 1993; Taylor 1984; D. Brown, pers. comm.; L. Chadderton, pers. comm.).
These findings suggest that the reported differences in sensitivity (from published LD50 values; see Table 13) may be either inaccurate or irrelevant predictors of susceptibility to brodifacoum, since species such as house sparrows, silvereyes, and paradise shelducks are reported to be moderately resistant.
The impacts of brodifacoum-poisoning operations on populations of non-target species that might have eaten baits have been monitored in several studies. Numbers of three indigenous bird species (western weka, Stewart Island weka, and pukeko) have been severely reduced in poison areas. For example, the entire western weka population on Tawhitinui Island was exterminated by direct consumption of Talon® 50WB intended for ship rats, which they obtained by reaching into bait stations, by eating baits dropped by rats, and by eating dead or dying rats (Taylor 1984). About 80–90% of the Stewart Island weka on Ulva Island were similarly killed by Talon® 50WB intended for Norway rats (L. Chadderton, pers. comm.), and 98% of the western weka on Inner Chetwode Island were killed after the aerial distribution of Talon® 7-20 (Wanganui No.7 cereal baits with 20 ppm brodifacoum) intended for kiore (D. Brown pers. comm.). More than 90% of pukeko on Tiritiri Matangi Island were killed after aerial distribution of Talon® 20P for eradication of kiore (C.R. Veitch pers. comm.). Some introduced ground-feeding bird species such as brown quail, blackbirds, house sparrows, and common mynahs on Tiritiri Matangi Island were also decimated (C.R. Veitch pers. comm.). However, despite deaths of some individuals, populations of other bird species have been less affected. For example, on Stanley Island 41 of 43 banded North Island saddlebacks were still alive more than 1 month after aerial distribution of Talon® 20P (Towns et al. 1993). On Red Mercury Island, all nine little spotted kiwi with radio transmitters were still alive 1 month after aerial distribution of Talon® 20P (Robertson et al. 1993). On Tiritiri Matangi Island, little spotted kiwi, North Island saddlebacks, and North Island robin populations were not detrimentally affected by aerial distribution of Talon® 20P (C.R. Veitch pers. comm.). The South Island robin population on Breaksea Island was not detrimentally affected by the use of Talon® 50WB in bait stations (Taylor & Thomas 1993), and all banded South Island robins on Inner Chetwode Island are thought to have survived aerial distribution of Talon® 20P (D. Brown pers. comm.). Brodifacoum residues have been detected in dead birds after aerial application of baits for rodent eradication (Morgan et al. 1996a), but the extent of wildlife contamination and impact after continued use has not been comprehensively studied.
Sub-lethal doses of brodifacoum have caused abortions and reduced lambing rates in sheep (Godfrey 1985), and concerns have been expressed about the adverse effects of small doses of anticoagulants on tawny owls (Townsend et al. 1981). However, there are no publications that elucidate any potential long-term effects of low-level brodifacoum exposure in birds.
There are no published LD50 data on the direct acute toxicity of brodifacoum to New Zealand bats. However, data from other anticoagulants suggest they may be susceptible if they were to consume the toxin.
There are no published LD50 data on the acute toxicity of brodifacoum to reptiles or amphibians. However, reptiles, at least, are known to be susceptible to brodifacoum. Telfair’s skinks (Leiolopisma telfairii) were found dead after eating rain-softened Talon® 20P used for rabbit eradication on Round Island, Mauritius, and post-mortem analyses revealed brodifacoum concentrations of 0.6 mg/kg in samples of liver (Merton 1987). Skink numbers have increased markedly since the removal of rabbits (North et al. 1994). In New Zealand, lizard numbers increased after use of Talon® 20P to eradicate rabbits and rats on Stanley Island (Towns et al. 1993) and rats on Tiritiri Matangi Island (C.R. Veitch pers. comm.).
Invertebrates have been seen eating baits containing brodifacoum, and residues of brodifacoum have been found in beetles (Coleoptera) collected from bait stations containing Talon® 50WB intended for rats on Stewart Island (G.R.G. Wright unpubl. data). It is considered that invertebrates are unlikely to be directly killed by brodifacoum (Shirer 1992; Morgan et al. 1996a). However, a number of unpublished observations suggest that brodifacoum may be toxic to molluscs (D. Merton pers. comm.). Contaminated invertebrates may pose a risk of secondary poisoning to insectivorous vertebrates. However, recent studies have shown that brodifacoum does not persist in weta. If there is a similar lack of persistence in other invertebrates, then the risk of secondary poisoning via invertebrates would be short-lived. However, at this time the persistence of brodifacoum in molluscs has not been elucidated. Molluscs have a hepato-pancreas; in mammals anticoagulant rodenticides binds to vitamin K epoxide reductase in both the pancreas and the liver. It is therefore conceivable that the hepato-pancreas is a target organ in molluscs.
The risk of secondary poisoning to non-target species is far greater from second-generation anticoagulants such as brodifacoum than from first-generation anticoagulants such as warfarin, because second-generation compounds are not substantially metabolised and excreted before death. For example, five out of six owls died after feeding on rats killed by brodifacoum for 8–11 days (Mendenhall & Pank 1980).
The only confirmed report of secondary poisoning of insectivorous birds with brodifacoum was in a zoo, where avocets, rufous-throated ant pittas, golden plovers, honey creepers, finches, thrushes, warblers, and crakes died in an aviary after feeding on pavement ants and cockroaches that had eaten brodifacoum baits (Godfrey 1985). However, the potential for invertebrates to ‘carry’ poison to birds has been suggested (Stephenson et al. 1999).
In New Zealand, predator and scavenger populations have been monitored during five brodifacoum-poisoning operations. Comparable numbers of brown skuas and New Zealand falcons, the main avian predators at risk, were seen before and after use of Talon® 50WB in bait stations for eradication of Norway rats on Hawea Island (Taylor & Thomas 1989). There was no evidence of New Zealand falcons or moreporks being killed by use of Talon® 50WB in bait stations for eradication of Norway rats on Breaksea Island (Taylor & Thomas 1993). There was no evidence of a detrimental effect on populations of moreporks on Stanley Island (Towns et al. 1993) or Red Mercury Island (Robertson et al. 1993) after aerial distribution of Talon® 20P for eradication of kiore. Moreporks and Australasian harriers on Tiritiri Matangi Island decreased after aerial distribution of Talon® 20P, but it is not known whether this was induced by poisoning (C.R. Veitch pers. comm.) or the removal of their major food item, rats.
The perceived hazards of secondary poisoning to non-target wildlife have restricted second-generation anticoagulants such as brodifacoum from being registered for field use in the USA (Colvin et al. 1991). The detection of brodifacoum residues in a range of wildlife including native birds such as kiwi (Apteryx spp.) (Robertson et al. 1993), raises serious concerns about the long-term effects of broad-scale field use of brodifacoum in New Zealand. This is compounded by the recent detection of residues in a wide range of species: weka, morepork, Australian harrier, pukeko, grey duck, mallard, black-backed gull, robin, saddleback, chaffinch, mynah, magpie, and blackbird (Murphy et al. 1998; Dowding et al. 1999; G.R.G. Wright, pers. comm.). Of far less concern was the detection of brodifacoum in cats and stoats, introduced species regarded as pests and largely responsible for the decline of native birds, such as kiwi. Nevertheless, because of the potential for uncontrolled contamination of wildlife (demonstrated by field survey data) broad-scale field use of brodifacoum in New Zealand (Eason et al. 1999c, 2000) is currently being restricted by the Department of Conservation.
Recent surveys of wildlife have indicated that extensive contamination has occurred where there has been sustained use of brodifacoum. Samples of liver were collected from feral pigs, feral red deer, feral cats, stoats, and weka that were shot, or trapped, except for one feral pig and six weka found dead (Eason & Murphy 2000). All the animals were killed in areas where brodifacoum was currently in use for possum and rat control. In all cases the method of application of baits followed label instruction and bait stations were used. Fourteen out of 35 pigs (40%) contained no residues. The remaining 21 pigs, including one which was found dead, contained residues of brodifacoum at concentrations ranging from 0.007 to 1.78 mg/kg. The pig found dead contained the highest liver residue. Eleven of 33 feral deer (33%) were contaminated but the concentration did not exceed 0.03 mg/kg. Clearly, in the case of deer, the most likely route of ingestion of brodifacoum is by feeding on baits that were not adequately contained in bait stations. (Possums are known to spill significant amounts of baits when feeding). This being the case, it seems probable that at least some of the pigs may have ingested bait in the same way as deer, compounded by some ingestion of brodifacoum-poisoned target species. Fifty-seven out of 71 cats (80%), and 98 out of 115 stoats (85%) contained residues. Concentrations in cats ranged from 0.078 to 1.84 mg/kg and in stoats from 0.008 to 1.32 mg/kg. Six weka were found dead and contained residues of between 0.11 and 2.3 mg/kg. The other 12 weka were trapped; four contained no residues, and eight (67%) contained between 0.01 and 0.95 mg/kg.
These recently acquired residue results reinforce earlier recommendations that pigs and possums should not be hunted for human consumption, from areas where baits containing brodifacoum have been used for possum control, for at least 9 months after the application of the baits (Eason et al. 1996d).
In summary, indigenous New Zealand birds most at risk from feeding directly on cereal-based baits containing brodifacoum are those species that are naturally inquisitive and have an omnivorous diet (e.g. weka, pukeko, brown skua, and kea). The risk of secondary poisoning is probably greatest for predatory and scavenging birds (especially the weka, brown skua, Australasian harrier, morepork, and southern black-backed gull) that feed on target species (e.g. live or dead rats, rabbits, and possums). Recently published surveys by Department of Conservation and Landcare Research staff clearly demonstrate widespread wildlife contamination that extends to native birds as well as game species (Murphy et al. 1999; Gillies & Pierce 1999; Dowding et al. 1999; Meenken et al. 1999; Eason et al. 1999c; Robertson et al. 1999a; Stephenson et al. 1999). This pattern is mirrored overseas where there is field use of second-generation anticoagulants (Young & de Lai 1997; Shore et al. 1999; Stone et al. 1999).
The risks of non-target mortality and contamination after pest control must be carefully balanced against the benefits. The eradication of rabbits using brodifacoum on Round Island, Mauritius, in 1986 illustrates this most clearly. Telfair’s skinks and other lizards on the island were considered at risk from poisoning by eating poisoned insects and/or bait and some were killed (Merton 1987). Three years after eradication of the rabbits there has been a dramatic regeneration of vegetation and marked increases in the numbers of lizards, including Telfair’s skink (North et al. 1994). In New Zealand, the benefits of using brodifacoum (or related compounds) to eradicate rats and/or rabbits from offshore islands are also becoming apparent. For example, eradication of rats from Korapuki Island (using bromadiolone, a second-generation anticoagulant related to brodifacoum) in 1986 resulted in a 10-fold increase in lizard numbers in 3 years (Towns 1991) and a 30-fold increase in 6 years (Towns 1994). Similarly, in 1996, the successful removal of rats from Kapiti Island has resulted in a significantly improved survival rate for stitchbirds and saddlebacks, and benefits to other taxa are expected (Empson & Miskelly 1999). However, on mainland sites where the persistence of brodifacoum raises concerns about the possible transfer of this compound through the food chain to humans, dogs, or wildlife, a precautionary approach is recommended. Because of this, the use of this poison has been under review. Nevertheless, its total removal from mainland use leaves a significant gap in the armoury of the conservationist, pest controller involved in endangered species protection (Stephenson 2000).
Generally available and no licence required.
High risk of secondary poisoning of non-target species
Effective against possums that have developed poison/bait shyness and
Effective for rodent control
Persistent (>9 months) in liver of vertebrates (can enter food chain and put meat for human consumption at risk)
Although an antidote (vitamin K) is available, long-term treatment is needed
Expensive compared to 1080 or cyanide
Possums can eat excessive amounts of bait (increase costs)
Possums take 2–4 weeks to die
Brodifacoum is a synthetic pesticide that was developed approximately 20 years ago.
Brodifacoum is not readily soluble. It binds strongly to soil and is slowly degraded. It is most unlikely to significantly contaminate waterways unless large amounts of baits enter streams.
It is a potent anticoagulant, which acts by interfering with the synthesis of vitamin K-dependent clotting factors. Brodifacoum is toxic to mammals, birds, and reptiles.
Brodifacoum is extremely persistent in the livers of lethally poisoned, and to a lesser extent the meat of sub-lethally poisoned, animals, which heightens the risk of secondary poisoning of non-target species.
Livestock must not be allowed access to brodifacoum baits as residues may persist in survivors of a sub-lethal dose for >9 months.
Non-target effects on individual birds of a number of species have occurred after brodifacoum use for rodent control.
Adverse effects on individual populations of a number of species of birds have been observed after brodifacoum use for rodent control. However, short-term losses are likely to be superseded by long-term gains once predators have been removed.