Where the Residue Goes: The Routes Rodenticide Takes Into Wildlife, and What Every Alternative Costs

1. Introduction: one problem, two literatures

Rodenticide resistance is discussed in pest management. Secondary poisoning of raptors is discussed in conservation. The two subjects have separate journals, separate conferences and largely separate readerships.

They are the same problem. The property that was engineered into the compounds to defeat resistance is the property that moves them up the food chain.

1.1 What this paper covers, and what it does not

This journal treats the resistance treadmill itself in a separate article: how anticoagulants kill, the VKORC1 mechanism in detail, the speed of the evolutionary response, and the regulatory instrument that failed to change exposure levels. Sections 3 and 4 here give only what is needed to follow the argument.

What this paper is about is the second half: the specific routes by which the compounds reach animals nobody intended to poison, what has been measured at the far end of those routes, and what every available substitute costs. The alternatives section is the part we think is least discussed.

2. The first generation

Where this starts.

Warfarin, a representative first-generation anticoagulant rodenticide, has been widely used since its introduction in the 1950s. Prolonged exposure has led to the emergence of resistant rodent populations worldwide, particularly in Europe, driven by mutations in the VKOR gene, making effective pest control increasingly difficult.⁴

2.1 Why anticoagulants took over in the first place

The class did not arrive in a vacuum. Their efficiency in dealing with rat problems became a fundamental reason for operators to switch their interest from acute poison to anticoagulants as a default rat control measure.⁷

2.2 The property that made them work

An anticoagulant kills slowly. A rat that eats a lethal dose does not die at the bait, does not associate illness with the food, and continues feeding.

This journal's article on rodent neophobia set out why that matters. A cautious animal that samples a new food and becomes acutely ill learns to avoid it, and the resulting bait shyness defeats fast-acting poisons. Delayed action is the specific answer to that behaviour.

2.3 The second advantage

Anticoagulants block a vitamin K dependent pathway, which means vitamin K is a treatment. That gives the class something almost no other rodenticide has, and §19 returns to what its absence costs.

2.4 Seventy years of selection

This journal has documented resistance developing over decades in cockroaches, bed bugs, stored product beetles and head lice. Warfarin predates all of those deployments.

A compound in continuous use against a fast-breeding target since the 1950s is about as complete a selection experiment as pest management has run, and the result is what the theory predicts.

3. How resistance works

The mechanism, which is unusually clean.

Anticoagulant rodenticides act by inhibiting the vitamin K epoxide reductase complex, impairing coagulation.⁹ The gene encoding that enzyme, VKORC1, is the primary molecular target of coumarin-based rodenticides, and point mutations in it confer resistance.²

3.1 Why that specificity matters

A rodent resistant at VKORC1 is resistant to compounds acting there and not to compounds acting elsewhere. That is the basis for the alternatives in §16, and it is why rotation advice in this class is about anticoagulant generations rather than about unrelated chemistry.

4. The mutations that have been found

The specific changes documented, which vary by species.

In one national survey, researchers detected in the Mus genus two missense mutations, Leu128Ser and Tyr139Cys, that confer resistance to anticoagulant rodenticides in house mice, plus a new missense mutation Ala72Val not previously described in a related species; and in the Rattus genus one missense mutation Leu90Ile in the roof rat and one missense mutation Ser149Ile in the Norway rat.⁸

4.1 The named resistant strain

One German field programme studied control of Y139C-resistant Norway rats specifically, genotyping the DNA of each rat to determine resistance status conferred by the VKORC1 gene.¹

Genotyping individual animals in a field trial is an unusually rigorous design, and it is what allows efficacy to be reported against a known resistance status rather than against an assumed one.

4.2 Why most field efficacy claims are weaker than this

An ordinary field trial reports that a bait reduced activity at a site. It cannot say whether the animals that survived were resistant, were never exposed, or immigrated afterwards.

Radio-tagging, carcass recovery and genotyping separate those explanations. This journal has repeatedly argued that treatment failure has several candidate causes and that field data rarely distinguishes them; here is a design that does.¹

5. Resistance is not universal

An important corrective, because the literature is dominated by places where resistance is present.

One study found that Mus musculus populations in Western Australia lack VKORC1 mutations conferring resistance to first generation anticoagulant rodenticides, with the authors drawing out implications for conservation and biosecurity.³

5.1 Why that is worth knowing

Where resistance is absent, the first generation works, and the first generation is less persistent than the second.

So a jurisdiction without documented resistance has an option that a jurisdiction with it does not, and defaulting to the most potent available compound in that setting incurs the ecological cost described in §7 to purchase nothing.

It also wastes the option. Using the most potent compound where a weaker one would work applies selection pressure at the top of the ladder, which is the fastest route to having nothing left above it. The resistance management argument and the ecological argument agree here, which is unusual enough to be worth noting.

5.2 The monitoring implication

Observed prevalence of key VKORC1 mutations informs localised resistance management and advises rotational use of different anticoagulant classes, and highlights the evolutionary pressure exerted by prolonged rodenticide deployment.²

That is an argument for testing before escalating, which is the same argument this journal made about cockroach product selection.

The difference is practical. A cockroach bioassay can be run on animals collected from the site within a week. Genotyping a rodent population is a laboratory undertaking that no ordinary service call supports, which is why resistance status in structural rodent work is almost always assumed rather than measured, and why §21.2 treats it as an open local question.

6. The engineered fix

What was done about resistance, and how.

Second-generation compounds such as difenacoum were developed through chemical modification of the warfarin scaffold to enhance lipophilicity and biological persistence.⁴ The class includes bromadiolone, brodifacoum, difenacoum, difethialone and flocoumafen, developed displaying more efficacy and potency for eradicating rodent infestations.⁸

6.1 Why persistence defeats resistance

A resistant animal survives a dose that would kill a susceptible one. A compound that stays in the body longer, and that the animal continues to encounter over more feeding visits, accumulates toward a lethal total.

Persistence is therefore not an incidental property of the second generation. It is the mechanism by which it works on animals the first generation could not kill.

7. Why that fix creates the second problem

The consequence, stated by the same source in the next breath.

Despite their effectiveness, second-generation compounds have raised serious environmental concerns due to their high bioaccumulative potential, with numerous reports of secondary poisoning in non-target wildlife, particularly raptors. They exhibit prolonged hepatic retention in predatory birds owing to their high lipophilicity.⁴

7.1 The single property

Lipophilicity and persistence are what make the compound work against a resistant rat and what make it sit in a raptor's liver.

There is no version of the molecule that has the first without the second, because they are the same property described from two directions. That is why this is not a problem that better formulation solves.

7.2 The one route to selectivity that has been proposed

There is a possible way out, and it is at the level of the target rather than the persistence. Interspecies differences in VKORC1 are likely to influence isomer-specific binding and are directly relevant to mitigating secondary poisoning in non-target wildlife, since VKORC1 sequences and structures vary among species and such differences may modulate isomer-selective ligand binding and retention patterns, with species-dependent variability potentially influencing retention in non-target wildlife exposed through secondary poisoning.⁴

7.3 What that would mean if it held

A compound whose active isomer binds tightly to the rodent enzyme and poorly to the raptor enzyme would break the link in §7.1 at the receptor rather than in the liver.

That is a research direction rather than a product, and the authors present it as warranting further validation across multiple species. We record it because it is the only proposal we located that addresses the problem at its root rather than by limiting dose or placement.

8. What is in the owls

The measured residues.

Total rodenticide concentrations in liver samples ranged from below 0.2 to 6,743 nanograms per gram dry weight, with a median of 398, and three compounds were detected: brodifacoum in 74 per cent, bromadiolone in 26 per cent and flocoumafen in 15 per cent.⁵

8.1 The spread

A range spanning four orders of magnitude, with a median well above the detection limit, describes a population in which exposure is normal and severe exposure is not rare.

The dominance of brodifacoum in that profile matters, because it is the most potent and most persistent of the group.

9. The concentration thresholds

The figures that convert residue into harm.

Brodifacoum was detected in three-fourths of the sampled owls, with 60 per cent exhibiting hepatic concentrations above 100 nanograms per gram wet weight and 40 per cent above 200, concentrations previously associated with health effects and increased mortality in predatory birds respectively.⁵

9.1 Reading those two thresholds

Six in ten birds above the level linked to health effects. Four in ten above the level linked to increased mortality.

Those are not trace detections. They are a majority of a sampled population carrying body burdens that the literature associates with harm, which is a materially different claim from residues being present.

9.2 The caution we would attach

Association of a concentration with mortality in the literature is not a demonstration that these individual birds were harmed. Sampling of dead or admitted birds can also over-represent exposed individuals, and we have not established the sampling frame.

10. Exposure without resistance

A finding that separates the two halves of this paper.

In the same study, the absence of genetically mediated resistance via exon 3 Vkorc1 mutations does not necessarily mean that the risk of exposure to rat predators is low, as 16.0 per cent of the live-trapped rats bioaccumulated high levels of rodenticide.⁵

10.1 Why this matters

A susceptible rat still carries a residue before it dies, and a predator that takes it is exposed regardless of whether resistance is present in the local population.

So secondary poisoning is not confined to places with resistance. Resistance worsens it, for the reasons in §13, and does not cause it.

10.2 The window that creates the exposure

The delayed action described in §2.2 is what opens it. An animal that dies immediately at the bait is a carcass in a station. An animal that takes several days to die is a living, slowing, increasingly catchable rodent carrying a full dose around the site.

The property that defeats bait shyness is therefore also the property that presents the residue to a predator in the most available possible form. We flag that as our reasoning, and note it is the third time in this paper that a designed advantage turns out to be the mechanism of the harm.

11. The route nobody expects

An exposure pathway we had not considered before reading this literature.

Analyses of liver tissues from piscivorous birds in one European country detected second-generation residues in nearly half of the sampled great cormorants and common mergansers, with concentrations reaching up to 35 nanograms per gram wet weight, reflecting transfer of brodifacoum, difenacoum and bromadiolone from contaminated fish in effluent-receiving waterways.²

The mechanism is documented directly: heavy rainfall provokes anticoagulant rodenticides' release from baited sewer systems and outdoor surfaces into receiving streams.³

11.1 Why this breaks the usual model

Secondary poisoning is normally understood as a predator eating a poisoned rodent. Here the compound reaches a fish-eating bird, which does not eat rodents at all.

The bait went into a sewer, rain washed it into a watercourse, it entered fish, and it reached a bird three steps removed from any rodent. That is environmental contamination rather than secondary poisoning in the classical sense.

11.2 Why this route is invisible to everyone involved

A technician baiting a sewer has no line of sight to a cormorant. There is no dead rodent, no predator at the site, and no observable event connecting the two.

Secondary poisoning through predation at least has a mechanism a practitioner can picture. This one runs through a storm event, a watercourse, a fish and a bird, and the only thing that reveals it is somebody analysing liver tissue in a laboratory hundreds of kilometres away.

11.3 The weather dependency

The trigger named is heavy rainfall.³ That makes this an episodic release rather than a steady one: bait sits in place until a storm mobilises it.

Which implies that placement timing relative to forecast rainfall is a control variable, and we are not aware of any guidance that treats it as one. We flag that as our inference.

11.4 The operational consequence

Sewer baiting and exposed outdoor placements are the practices implicated. Bait in a secured station inside a building has no equivalent route.

We would draw the same indoor-outdoor distinction here that this journal drew for pollinator exposure, and flag the parallel as ours.

12. The regulatory turning point

Direct evidence that a policy decision moved the residues.

In the United Kingdom, long-term biomonitoring of Common Buzzards has identified a regulatory turning point in 2016 when the prevalence and levels of brodifacoum in raptor tissues rose sharply following approved outdoor use, while residues of earlier compounds such as difenacoum have declined.²

12.1 Why this is unusually good evidence

A long biomonitoring series, a dated regulatory change, and a step in the measured quantity that follows it. That is about as close to a natural experiment as this field offers.

It also shows substitution rather than addition: the newly permitted compound rose while the displaced one fell, which is what you would expect if applicators switched rather than added.

That detail is worth holding onto, because it means the total quantity of anticoagulant in the environment may not have changed much while its ecological weight did. Substituting a more persistent compound for a less persistent one at the same usage rate raises the burden on predators without raising the tonnage applied.

12.2 What the declining compound tells us

Difenacoum residues fell while brodifacoum rose.² That the displaced compound's signal decayed in raptor tissue over the same period is itself informative: it shows the biomonitoring is sensitive enough to track a change in practice, in both directions.

A series that only ever rose could be explained by improving analytical methods. One that shows a substitution is measuring the thing it claims to measure.

12.3 The lesson for approvals

Permitting outdoor use of the most persistent compound in the class produced a measurable increase in raptor body burden within a monitored population.

We would say that is the clearest available demonstration that placement rules, rather than compound bans alone, determine ecological outcome. That reading is ours.

13. The feedback loop

How resistance makes the ecological problem worse rather than merely coexisting with it.

Difficulties stem from managing widespread use in agricultural and urban environments, where resistance in target species drives the continuous use of baits, exacerbating the exposure risk for non-target wildlife. Resistance leads to greater environmental contamination, as rodenticides persist in ecosystems and bioaccumulate in non-target species like birds of prey and mammals, exacerbating the risks of secondary poisoning and biomagnification in the food chain.⁶

Where resistance has diminished the effectiveness of first-generation and some second-generation compounds, control of resistant populations must rely on the use of other second-generation compounds, which are widely known to be associated with poisoning of non-target wildlife.⁸

13.1 The loop stated plainly

Resistance means bait works less well. Bait that works less well stays down longer and is replaced more often. More bait in the environment for more of the time means more exposure for everything that eats rodents or encounters the bait.

And continued exposure selects for more resistance, which returns to the start.

13.2 The historical version of the same loop

The sequence has already run once in a documented case. The first detection of wood rat resistance against warfarin was reported in the 1980s in three separate localities, and to combat this problem, more potent anticoagulants were used and proved to be effective to control warfarin-resistant rats. Over-dependence on those more potent compounds then rose to a more complex situation when two major concerns emerged.⁷

13.3 Reading that as a template

Resistance appeared, a more potent compound solved it, and the solution generated the next set of problems. That is the shape of every escalation this journal has documented, from pyrethroid resistance in bed bugs to phosphine in stored grain.

What distinguishes this case is that the second set of problems fell on species that were never the target and were never party to the decision.

14. Double mutations

Where the loop leads.

Over-intensive use could also lead to the emergence of new resistances by recombination between different genotypes and generation of populations carrying double mutations in the Vkorc1 gene, as has been observed in Europe.⁸

14.1 Why this is the concerning endpoint

A single mutation confers resistance to some compounds. Recombination producing individuals carrying two resistance mutations is a mechanism for producing animals that nothing in the class reliably kills.

This journal reported the cockroach field study that found unexpected selection for broad cross-resistance. This is the same shape of finding in a vertebrate: intensive use producing a population harder to treat than the sum of its parts.

15. What the predator's own genes show

A recent line of evidence that measures exposure from the bird's side.

Blood samples from 72 eagle owl nestlings were analysed for expression of the target gene and a related gene. The target gene was predominantly expressed, with the related gene lower, but in one-third of samples the related gene was predominant. The authors suggest that expression of the second gene acts as a compensatory mechanism, increasing in response to recent exposure, and that this pattern was evident in anthropogenic environments, such as urban areas and landfills.⁹

15.1 Why this is interesting

It is a biological marker of exposure rather than a chemical one, measured in living nestlings rather than in dead birds.

That the pattern concentrates in urban areas and landfills points at exactly the settings where structural pest control operates, which is a more direct connection to this trade than the agricultural literature provides.

It is also a reminder of where the rodents are. A landfill or a dense urban block supports a large rodent population, that population supports predators, and the baiting intensity is highest in exactly those places. The three things co-locate by design rather than by accident.

15.2 The caution

The compensatory interpretation is described by the authors as probable rather than established, and expression differences have other possible causes.⁹ We record it as a developing line of evidence.

16. The non-anticoagulant options

What else exists, and the honest accounting of it.

Alternating solutions could involve rotation between anticoagulants, but also other non-anticoagulant solutions such as alphachloralose or cholecalciferol.⁸

These have the advantage of being less ecotoxic for wildlife via ingestion of contaminated rodents because they are not persistent.⁸

16.1 The property that reverses

Non-persistence is the reason these compounds do not accumulate in predators, and it is the opposite of the property engineered into the second generation in §6.

So the ecological problem and the resistance problem pull in opposite directions on the same axis, which is why there is no compound that solves both.

17. The efficacy comparison

Whether the alternative actually works.

In a six-day laboratory feeding trial against wild-caught rats, cholecalciferol baits had the highest mortality rate at 71.39 per cent, the first-generation anticoagulant chlorophacinone recorded 74.20 per cent, and warfarin recorded the lowest at 46.07 per cent, with days to death in the range of six to eight.⁷

17.1 What that shows

Cholecalciferol performed comparably to a first-generation anticoagulant and substantially better than warfarin in that trial.

The warfarin figure is itself informative. Under 50 per cent mortality in a six-day trial against wild rats is the resistance story of §2 appearing as a number.

18. The stop-feed effect

A practical property with an unexpected use.

A study examining a low concentration of cholecalciferol in brodifacoum bait found that cholecalciferol caused a strong stop-feed effect after two days in the laboratory, and tested both baits in field trials against Y139C-resistant rats with radio-tagged animals and carcass searching.¹

The stated motivation was that second-generation compounds potentially build persistent residues in animals and accordingly pose a risk of secondary poisoning.¹

18.1 The logic of the combination

If an additive causes rats to stop feeding sooner, each animal consumes less of the persistent compound before dying, which reduces the residue available to a predator.

That is an attempt to break the §7 link by limiting dose rather than by changing the molecule, and it is the most direct engineering response to this problem that we located.

18.2 Why the design is clever

It accepts that the persistent compound is needed against resistant animals and attacks the quantity consumed instead. The rat still dies; it just carries less.

Whether that translates into lower residues in predators is the question the field component was built to answer, with radio-tagging and carcass recovery providing the route to measuring it.¹ We have the design and not the outcome, and we would not present the approach as demonstrated.

18.3 The tension inside it

A stop-feed effect that acts too strongly risks sub-lethal dosing, which is the classic route to selecting for resistance. An additive that makes animals stop eating before they have taken a lethal dose would worsen the problem in §13 while improving the one in §7.

We are raising that as a question rather than a criticism; the study is a laboratory and field evaluation and the balance between those two effects is exactly what such work exists to determine.

19. The antidote problem

The cost of the less persistent options, which is easy to overlook.

Non-anticoagulant alternatives have been implicated in cases of primary poisoning of domestic wildlife through direct ingestion of baits and lack of antidote.⁸

19.1 Why this is a serious trade-off

Anticoagulant poisoning has a treatment, because vitamin K addresses the pathway the compound blocks. A dog that eats anticoagulant bait can be treated.

A compound with no antidote offers no such route. So the option that is gentler on the raptor eating a poisoned rat is harsher on the pet that eats the bait directly.

19.2 How we would weigh it

Primary poisoning is controlled by bait placement and station security, which is within an operator's control. Secondary poisoning through a predator is not.

That asymmetry argues for the less persistent compound in a properly secured programme, and we flag it as our reasoning rather than a sourced recommendation.

19.3 The argument against our own position

It deserves stating fairly. Bait station security fails in the real world. Stations get moved, lids get left open, children and dogs are inventive, and a programme that depends on perfect containment to avoid an untreatable poisoning is depending on something no operator can guarantee.

Someone weighing it the other way would say the antidote is exactly the insurance you want against the failure mode you cannot design out. We do not think that is unreasonable, and we would not claim this is settled.

20. What follows for practice

Do not default to the most potent compound. Resistance is not present everywhere.³

Test before escalating. Prevalence of mutations informs localised management.²

Treat placement as the ecological variable. Sewer and outdoor placements are the route into watercourses.³

Expect residues even in susceptible populations. Sixteen per cent of live-trapped rats carried high levels.⁵

Remember the second generation was built to persist. That is the property, not a defect.⁴

Non-anticoagulants trade one risk for another. Less persistent, and no antidote.⁸

Exclusion avoids the choice entirely. It is the only measure in this article with no ecological cost.

21. The Manitoba position

What we can and cannot establish locally.

We located no Manitoba or prairie data on VKORC1 mutation frequency in local rodent populations, and none on rodenticide residues in local raptors. The evidence assembled here is European, Iberian, Australian, Middle Eastern and Southeast Asian.

21.1 What transfers regardless

The chemistry does. A second-generation compound is lipophilic and persistent wherever it is applied, and a rodent carrying a residue is an exposure to whatever eats it.

The rainfall and sewer route also transfers to any municipality with combined or storm sewer outfalls, which is most of them.³

21.2 The question we would want answered

Whether local rodent populations carry resistance mutations at all. If they do not, §5 says the less persistent first generation remains available, and the ecological cost of defaulting to the second is being paid for nothing.

That is a testable question and, as far as we can tell, an unanswered one here.

22. Limitations and open questions

No Canadian data. Stated in §21.

Several findings reach us through abstracts and a topic summary. The German piscivorous bird residues, the United Kingdom buzzard series and the rainfall release finding come from a review-style summary and a citation list rather than the primary papers read in full.²³

The owl residue figures are one study in one country. Sampling frame not established, as noted in §9.2.⁵

The efficacy comparison is a laboratory trial on one rat species. Wood rats in an oil palm context, which is not a structural pest setting.⁷

The gene expression work is recent and interpreted cautiously by its own authors. Stated in §15.2.⁹

The resistance treadmill is treated separately. This journal covers the mechanism, the speed of the evolutionary response and the concentration-limit instrument that failed to change exposure levels in a different article. Sections 2 to 5 here are deliberately compressed and should not be read as this journal's full treatment of them.

We have not addressed regulatory status. Which compounds are registered, at what concentrations, and with what placement restrictions differs by jurisdiction, and nothing here describes what is permitted in Manitoba. Operators should work from current label and provincial requirements.

Sections 5.1, 11.2, 12.2, 13.1, 19.2 and 21.2 are our reasoning. The default-compound argument, the indoor-outdoor parallel, the placement-rules conclusion, the loop statement, the asymmetry argument and the local testing question are ours rather than sourced positions.

Our commercial position. This company applies rodenticide. An article arguing that the potent compounds should not be the default, that placement is the ecological variable, and that exclusion avoids the question entirely is an article constraining our own practice. We think it is correct.

23. Conclusion

Warfarin has been in use since the 1950s, and prolonged exposure selected for resistance conferred by point mutations in the gene encoding its target enzyme.⁴² The response was to modify the warfarin scaffold to enhance lipophilicity and biological persistence, producing compounds that work against resistant animals and that exhibit prolonged hepatic retention in predatory birds owing to their high lipophilicity.⁴ Those are the same sentence read forwards and backwards.

In one owl population brodifacoum was present in about three quarters of birds, with 60 per cent above a concentration associated with health effects and 40 per cent above one associated with increased mortality.⁵ Residues reached fish-eating birds in another country through rain washing bait from sewers into streams.²³ Raptor brodifacoum levels rose sharply after outdoor use was approved.² And resistance keeps the bait down longer, which feeds all of it.⁶

There is no molecule that escapes this, because the property being complained about is the property that was purchased. Every substitute trades one cost for another: the less persistent compounds spare the predator and remove the antidote, an additive that limits the dose each rodent carries risks sub-lethal feeding, and a first-generation product works only where resistance has not arrived.

Which leaves placement as the variable most within an operator's control, and it is the one the routes in this paper turn on. A bait in a secured station inside a building has no path to a watercourse. The same bait in a sewer has one, and it opens the next time it rains. The only measure here with no ecological cost at all is closing the hole the rodent came through.

References

1. The stop-feed effect of cholecalciferol and the efficacy of brodifacoum combined with cholecalciferol in Y139C-resistant Norway rats (Rattus norvegicus). PubMed Central PMC9948778. Used for the statement that second-generation anticoagulant rodenticides potentially build persistent residues in animals and accordingly pose a risk of secondary poisoning; for the study design examining a low concentration of cholecalciferol in brodifacoum bait on bait consumption and control success in laboratory and field trials, with efficacy determined against resistant Y139C rats; for the finding that cholecalciferol caused a strong stop-feed effect after two days in the laboratory study; and for the field design in which bait containing brodifacoum alone or brodifacoum with cholecalciferol was applied at two sites each, infestations assessed before and after treatment, rats radio-tagged, carcasses searched for during the treatment period, and the DNA of each rat genotyped to determine resistance status conferred by the VKORC1 gene. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9948778/
2. Anticoagulant Rodenticide Effects on Wildlife Systems. Research topic summary. Used as a compilation of findings rather than a primary report. Used for the definition of VKORC1 as the gene encoding vitamin K epoxide reductase, the primary molecular target of coumarin-based rodenticides, subject to point mutations; for the statement that observed prevalence of key VKORC1 mutations informs localised resistance management, advises rotational use of different anticoagulant classes, and highlights the evolutionary pressure exerted by prolonged rodenticide deployment; for the report that analyses of liver tissues from piscivorous birds in Germany detected second-generation residues in nearly half of sampled great cormorants and common mergansers, reaching up to 35 nanograms per gram wet weight, reflecting transfer of brodifacoum, difenacoum and bromadiolone from contaminated fish in effluent-receiving waterways; and for the report that long-term biomonitoring of Common Buzzards in the United Kingdom identified a regulatory turning point in 2016 when the prevalence and levels of brodifacoum in raptor tissues rose sharply following approved outdoor use, while residues of earlier compounds such as difenacoum declined. https://www.nature.com/nature-index/topics/l4/anticoagulant-rodenticide-effects-on-wildlife-systems
3. Adverse Outcome Pathway and Risks of Anticoagulant Rodenticides to Predatory Wildlife. Environmental Science and Technology. doi:10.1021/es501740n. Cited together with its citing literature, which we have used to identify rather than to read in full. Used for the identification of Regnery and colleagues (2020), Heavy rainfall provokes anticoagulant rodenticides' release from baited sewer systems and outdoor surfaces into receiving streams, Science of the Total Environment 740, 139905; of Duncan and colleagues (2020), Mus musculus populations in Western Australia lack VKORC1 mutations conferring resistance to first generation anticoagulant rodenticides, with implications for conservation and biosecurity, PLOS ONE 15(9); and of work detecting Vkorc1 single nucleotide polymorphisms indicating the presence of anticoagulant rodenticide resistance in introduced rats in Australia, Pest Management Science 2025. https://pubs.acs.org/doi/10.1021/es501740n
4. Isomer-Dependent Pharmacokinetic Behavior and VKOR Interactions of Second-Generation Anticoagulant Rodenticides. PubMed Central PMC13164442. Used for the statements that warfarin, a representative first-generation anticoagulant rodenticide, has been widely used since its introduction in the 1950s; that prolonged exposure has led to the emergence of resistant rodent populations worldwide, particularly in Europe, driven by mutations in the VKOR gene, making effective pest control increasingly difficult; that to overcome warfarin resistance, second-generation anticoagulant rodenticides such as difenacoum were developed through chemical modification of the warfarin scaffold to enhance lipophilicity and biological persistence; that despite their effectiveness they have raised serious environmental concerns due to their high bioaccumulative potential, with numerous reports of secondary poisoning in non-target wildlife particularly raptors in Europe and North America; and that these compounds exhibit prolonged hepatic retention in predatory birds owing to their high lipophilicity. https://pmc.ncbi.nlm.nih.gov/articles/PMC13164442/
5. Toxic legacy: the hidden impact of anticoagulant rodenticides on Portuguese raptors. Science of the Total Environment. Used for the finding that total rodenticide concentrations in liver samples ranged from below 0.2 to 6,743 nanograms per gram dry weight with a median of 398, and that three rodenticides were detected, brodifacoum in 74 per cent, bromadiolone in 26 per cent and flocoumafen in 15 per cent; for the finding that brodifacoum was detected in three-fourths of sampled owls with 60 per cent exhibiting hepatic concentrations above 100 nanograms per gram wet weight and 40 per cent above 200, concentrations previously associated with health effects and increased mortality in predatory birds respectively; and for the observation that the absence of genetically mediated resistance via exon 3 Vkorc1 mutations does not necessarily mean the risk of exposure to rat predators is low, as 16.0 per cent of live-trapped rats bioaccumulated high levels. https://www.sciencedirect.com/science/article/pii/S0048969725021874
6. Differential exposure to second-generation anticoagulant rodenticides in raptors from continental and insular regions of the Iberian Peninsula. Environmental Pollution. Used for the statement that difficulties stem from managing widespread use of second-generation compounds in agricultural and urban environments, where resistance in target species drives the continuous use of baits, exacerbating exposure risk for non-target wildlife; and for the statement that widespread distribution of resistance-conferring mutations in the Vkorc1 gene among house mouse populations has severely diminished the effectiveness of first-generation and some second-generation compounds, leading to greater environmental contamination as rodenticides persist in ecosystems and bioaccumulate in non-target species such as birds of prey and mammals, exacerbating the risks of secondary poisoning and biomagnification in the food chain. https://www.sciencedirect.com/science/article/pii/S0269749124017494
7. Efficacy of cholecalciferol rodenticide to control wood rat, Rattus tiomanicus, and its secondary poisoning impact towards barn owl. PubMed Central PMC9938244. Used for the study comparing laboratory efficacy of cholecalciferol with the first-generation anticoagulants chlorophacinone and warfarin; for the result that in a six-day wild wood rat laboratory feeding trial cholecalciferol baits had the highest mortality rate at 71.39 per cent, chlorophacinone recorded 74.20 per cent and warfarin recorded the lowest at 46.07 per cent, with days to death in the range of six to eight; and for the account that the first detection of wood rat resistance against warfarin was reported in the 1980s, that more potent anticoagulants were then used and proved effective against warfarin-resistant rats, and that over-dependence on second-generation compounds subsequently raised further concerns. https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9938244/
8. Investigation of anticoagulant rodenticide resistance induced by Vkorc1 mutations in rodents in Lebanon. Scientific Reports. doi:10.1038/s41598-022-26638-5. Used for the statement that second-generation rodenticides including bromadiolone, brodifacoum, difenacoum, difethialone and flocoumafen were developed displaying more efficacy and potency for eradicating rodent infestations; for the detection in the Mus genus of the missense mutations Leu128Ser and Tyr139Cys conferring resistance in house mice and a new missense mutation Ala72Val in a related species, and in the Rattus genus of Leu90Ile in the roof rat and Ser149Ile in the Norway rat; for the statement that control of resistant mouse populations must rely on other second-generation compounds which are widely known to be associated with poisoning of non-target wildlife; for the warning that over-intensive use could lead to emergence of new resistances by recombination between genotypes and generation of populations carrying double mutations in the Vkorc1 gene as observed in Europe; and for the account of alternating solutions involving rotation between anticoagulants and non-anticoagulant options such as alphachloralose or cholecalciferol, which are less ecotoxic for wildlife via ingestion of contaminated rodents because they are not persistent, but which have been implicated in cases of primary poisoning of domestic wildlife through direct ingestion of baits and lack of antidote. https://www.nature.com/articles/s41598-022-26638-5
9. Gene expression of anticoagulant rodenticide target genes in a non-target wildlife species: the eagle owl (Bubo bubo). PubMed record 42531760. Used for the statement that anticoagulant rodenticides are globally used for rodent control and pose significant risks to non-target predators and scavengers through secondary poisoning, acting by inhibiting the vitamin K epoxide reductase complex and impairing coagulation; for the study design analysing blood samples from 72 nestlings for target gene and related gene messenger RNA levels by RT-qPCR, with canonical correlation analysis exploring relationships between environmental variables and gene expression; and for the results that the target gene was predominantly expressed while the related gene showed lower expression, that in one-third of samples the related gene was predominant, that this probably acts as a compensatory mechanism increasing in response to recent exposure, and that the pattern was evident in anthropogenic environments such as urban areas and landfills. https://pubmed.ncbi.nlm.nih.gov/42531760/