Why this poison is interesting

Arsenic is the only entry in this library in which the same element poisons by two chemistries that have nothing to do with each other, and the chemistry it uses depends on which oxidation state it happens to be in. Arsenate — the pentavalent form — is a close structural analogue of phosphate and "may enter cells through phosphate transport proteins", after which it is either reduced to arsenite or "substituted for phosphate in metabolic reactions such as glycolysis".¹ Arsenite — the trivalent form — does nothing of the kind. It enters "either by diffusion or by specific transporters" and then "react[s] with sulphhydryl (-SH) groups of cellular proteins", disrupting enzymes including pyruvate dehydrogenase.¹ Traditional teaching Two entrances, two targets, one element.

The second reason is that the two states are not equally dangerous, and the difference is not a detail. The independent review that this page uses for its mechanism states that "the trivalent arsenicals, including those methylated, have more potent toxic properties than the pentavalent arsenicals".² So the body's own handling of arsenic matters: metabolism interconverts the two states,¹ and the direction of that interconversion changes the toxicity of what is present. This is one of the few poisonings in which reduction rather than oxidation is the activating step.

The third reason is the one that shaped how this page is written. The account above is standard, is taught everywhere, and is stated by both of this page's principal sources — and the peer-reviewed one says, in the same abstract in which it states it, that "the exact mechanism of the action of arsenic is not known, but several hypotheses have been proposed".² That is a citation for the doubt about a mechanism the page is simultaneously explaining, which is precisely the circumstance this library's evidence-tier system exists for. The mechanism is given here in full because it is the best available account and because it predicts the clinical picture well. It is also badged, everywhere it appears, as teaching rather than as demonstrated fact.

A poison is a drug whose kinetics have escaped its pharmacology.

The toxic principle

The two mechanisms converge on the same casualty — ATP — which is why a poison with two unrelated chemistries produces one clinical syndrome. The arsenite route disables enzymes by binding their thiols, and the enzyme named in the source is pyruvate dehydrogenase, "which is key to oxidative phosphorylation".¹ The arsenate route does not disable anything; it inserts a molecule that looks enough like phosphate to be accepted into reactions such as glycolysis, "resulting in further disruption of oxidative phosphorylation and the loss of ATP formation".¹ One route breaks the machine and the other feeds it the wrong part.

Why phosphate is the substrate worth impersonating

Phosphate is not a bystander in energy metabolism; it is the thing being moved. Every high-energy phosphate bond in glycolysis and oxidative phosphorylation is a phosphate group transferred from somewhere to somewhere else, so a molecule accepted as phosphate is accepted at the exact point where energy is captured. Arsenate esters are unstable in a way phosphate esters are not, so the reaction proceeds and the energy is simply lost rather than stored — the classical account of why the substitution wastes ATP instead of merely blocking its production. Neither cited source states it — it is standard textbook chemistry supplied here to make the account coherent, and it is flagged rather than badged as contested teaching. Inferred The distinction between blocking and wasting is the mechanistic difference between the two halves of this page, and it is the part of the account that is least directly evidenced in the sources cited here.

The clinical consequence of a lesion in ATP production is a multi-system illness in whichever tissues have the least metabolic reserve — which is why severe acute arsenic poisoning does not look like an organ-specific poisoning at all. The overview describes patients progressing "to multi-organ involvement within hours", with "rhabdomyolysis, renal failure, respiratory failure, failure of vital cardiovascular and brain functions".¹ A poisoning that disables energy production presents as everything failing at once, which is the same shape as cyanide on a slower clock.

Where this page's mechanism section comes from, and why not from the obvious place

The UKHSA toxicological overview has a section headed Mechanism of Toxicity, and it is quoted above. Both of its paragraphs are attributed, in the source, to a single reference — and that reference is NPIS TOXBASE, which is login-gated, cannot be verified from outside an NHS account, and must not be quoted in this library. The overview's account is therefore corroborated here against Hughes 2002,² an independent peer-reviewed review in Toxicology Letters which states the same two mechanisms in its own words: "inorganic arsenic in the pentavalent state may replace phosphate in several reactions" and "in the trivalent state, inorganic and organic (methylated) arsenic may react with critical thiols in proteins and inhibit their activity".² Every mechanistic claim on this page is anchored to that second source. This is the second consecutive band in which a free government document has been found carrying content that traces back to a login-gated database, and it is the first time it has affected an entire section rather than a table.

Toxicokinetics

The kinetic table matters more here than on most pages, for one reason that is easy to miss: arsenic leaves quickly. A half-life of three to five days¹ means that by the time the characteristic late findings appear, the poison itself may be undetectable. The investigation and the illness run on different clocks.

Arsenic — a poison that is gone before its signature findings arrive
Parameter	Therapeutic	In overdose	Why it changes
Oral absorption	—	"Approximately 95%" in volunteers given water with high arsenic levels; "less than 5% of an oral dose of arsenite was recovered in the faeces"¹	Near-complete. Decontamination has a narrow window and the absorbed fraction is effectively the ingested one
Inhalational absorption	—	Studies suggest 30 to 60% may be deposited in the lungs, and "the majority if not all of this fraction is then absorbed"¹	Depends on solubility, species and particle size.¹ Particles are mostly under 2 µm, which "may reach the alveoli"¹
Dermal absorption	—	"Considered low relative to the oral and inhalation routes"¹	Contact produces local injury rather than systemic poisoning — the reverse of hydrofluoric acid
Distribution	—	Widely distributed; autopsy data show "muscle, bone, the kidneys, liver, and the lungs accumulate the highest absolute amounts"¹	No single sanctuary compartment, which is part of why no chelator has an obvious target organ
Effect of oxidation state on distribution	—	"Arsenite species reach higher levels in tissues than arsenate species" in experimental animals¹	An animal finding, flagged as such. It is consistent with arsenite's greater potency² but is not human data
Metabolism	—	Two processes: redox interconversion of arsenate and arsenite, and methylation of arsenite to monomethylarsonic and dimethylarsinic acid¹	Metabolism changes the toxicity in both directions. Reduction to arsenite makes it more potent;² methylation speeds elimination¹
Excretion	—	Predominantly urinary. In one study in pregnant women exposed through drinking water, the species were 79 to 85% DMA, 8 to 16% inorganic and 5 to 6% MMA¹	The methylated species dominate what is excreted, which is why speciation matters when a urine result is interpreted
Half-life	—	3 to 5 days for ingested inorganic arsenic; "methylated compounds are excreted more rapidly"¹	The poison is cleared in under a week. The neuropathy and the nail changes are not
Blood versus urine	—	Blood clears rapidly, which limits its usefulness for detecting exposure; urine is the specimen of choice, with blood as an adjunct in suspected acute exposure. Paraphrased rather than quoted — the sentences stating this in the overview are attributed to its reference 6, which is TOXBASE	Urine is the specimen; blood is an adjunct. A normal blood arsenic days after an exposure means very little
Hair and nails	—	"Not considered useful for recent exposures, but may be used for past exposures"¹	The keratin record is a retrospective instrument, and it is the same biology that produces Mee's lines
Dialysability	—	Not addressed by EXTRIP⁴	Widely distributed into tissue and rapidly cleared renally in the patient with working kidneys — the kinetics do not suggest a role

Metabolism and the metabolites

Arsenic is one of the few poisons in this library whose metabolism runs in both directions at once, changing toxicity as it goes. The overview describes the process as "two processes; reduction and oxidation reactions that interconvert arsenate and arsenite, and methylation reactions which form mono-methylarsonic (MMA) and dimethylarsinic acid (DMA) from arsenite".¹ Reduction produces the more potent species;² methylation produces the species that leaves fastest.¹ The patient's own biochemistry is simultaneously activating and detoxifying the poison, and the balance is not something a clinician can observe or influence.

Arsenic — one element, two states, and metabolism that cuts both ways

Inorganic arsenic, ingested or inhaledNear-complete oral absorption — "approximately 95%"¹
Arsenate, As(V)Enters "through phosphate transport proteins"¹ — admitted because it is mistaken for a nutrient
Arsenite, As(III)Enters "by diffusion or by specific transporters"¹. The more potent form²
Redox interconversion — reduction of arsenate to arsenite is an activating step^1,2
Substitution for phosphate in glycolysis"Resulting in further disruption of oxidative phosphorylation and the loss of ATP formation"¹
Reaction with protein sulphydryl groups"Disruption of enzymes such as pyruvate dehydrogenase which is key to oxidative phosphorylation"¹
Methylation of arsenite to MMA and DMA¹ — the detoxifying limb
Renal excretion, half-life 3 to 5 days¹Mostly as DMA (79–85% in one study).¹ The poison leaves; the injury it started does not

Elimination and accumulation

Arsenic is the fastest-clearing poison in this band by a wide margin. Ingested inorganic arsenic has a half-life of 3 to 5 days,¹ against one to two months for mercury and ten to thirty years for lead in bone. There is no long-term reservoir comparable to the skeleton, and "the vast majority of arsenic is excreted rapidly in the urine" after inhalation.¹ In one sense that is reassuring: the patient who survives the acute illness will clear the poison within about a week without any intervention at all.

Where this latent phase sits among the others

Arsenic — a tissue declaring on its own timetable rather than the poison's — the arsenic is excreted within days, but the nail that was growing while it circulated does not show its white transverse line for several weeks

The other 26 kinds of latent phase in this library

Amphetamines and MDMA — a hormone acting normally on a kidney behaving normally, while the patient supplies the water
Antipsychotics — a physical object in the stomach — extended-release quetiapine forming a pharmacobezoar
Arsine and stibine — a red cell mass haemolysing faster than a kidney can cope with — the exposure is over, the haemolysis is silent until the urine changes colour, and the renal failure that follows is the cause of death
Beta-blockers — a repolarisation lesion waiting for an ectopic beat to fall inside it — sotalol prolongs the QT and then, for hours, nothing happens
Calcium-channel blockers — a tablet that has not yet dissolved
Carbon monoxide — an inflammatory process continuing after the poison itself has gone
Chlorine and ammonia — paraquat's cause at its shortest — the time an injured alveolar epithelium takes to leak
Cocaine — a package that has not yet failed — the calcium-channel blocker's cause moved outside the tablet, where wrapping integrity cannot be measured
Digoxin — distribution of the drug to its target — which that page argues is not a latent phase in the paracetamol sense at all
Ethylene glycol — paracetamol's cause with a different enzyme — glycolate accumulating behind glycolate oxidase while the patient looks merely drunk
GHB and GBL — not the poisoning but its withdrawal — a half-life under an hour means admission for any reason interrupts round-the-clock redosing
Hydrofluoric acid — an ion diffusing far enough to reach a nerve ending — and the thinner the solution, the further it travels before anybody feels it
Iron — a true remission — the corrosive injury settling while absorbed iron moves into the cells it will poison, so the patient improves on their way to the dangerous phase
Ketamine — nitrous oxide's cause on a longer clock — urothelial damage accumulating over months to years of repeated exposure
Lead — bone giving lead back — a store with a half-life of 10 to 30 years releasing its contents when pregnancy, lactation, menopause or osteoporosis resorbs it, so the blood concentration rises after the exposure has ended
Lithium — transport across cell membranes
Mercury — distribution on two clocks — tissue concentrations peaking within 24 hours everywhere except the brain, which is not reached until 2 to 3 days, and which then cannot let the poison out again
Methaemoglobin inducers — paracetamol's cause at its shortest — one to four hours manufacturing the toxic species
Methanol — paracetamol's cause again — formate accumulating behind a folate-dependent disposal step that primates perform poorly
Nitrous oxide — damage accumulating to a threshold
Opioids — an antidote wearing off before the poison does — renarcotisation, the only gap in this set that treatment creates rather than reveals
Organophosphate insecticides — a fat store emptying — and, separately, a second and unexplained lesion declaring itself at a neuromuscular junction the first phase had already left
Paracetamol — time spent manufacturing a toxic metabolite
Paraquat — the body responding to an injury that is already complete
Sodium-channel blockade — a gap that cannot be shortened
Thallium — hair on its own clock rather than the poison's — sensory symptoms come first and the alopecia that makes the diagnosis obvious follows them, well after the interval in which treatment is recommended

Arsenic's gap is a new kind in this set, and the mechanism is a matter of tissue turnover rather than of the poison. Every other member waits for something to happen to the poison — for it to be made, distributed, released or dissolved. Arsenic's nail sign waits for a nail to grow. The injury was inflicted at the time of exposure, to the growing matrix; it becomes visible only when that section of keratin has advanced far enough up the nail bed to be seen. The clock belongs to the tissue, not to the toxin, and it is still running long after the arsenic has been excreted.

Target organs — and why those

A poisoning that disables energy production has no single target organ, and the honest version of this section says so. What determines the organ list is not affinity but exposure and turnover: the gut meets the highest concentration, the tissues with the least metabolic reserve fail first, and the tissues that turn over fastest carry the record.

Gastrointestinal tract

TargetMucosa, at the highest concentration it will ever meet

Why hereThe organ of first contact after ingestion, and the reason acute arsenic poisoning is mistaken for gastroenteritis. The fluid loss is not incidental to the poisoning; it is severe enough to cause the shock that kills in the first phase. Established

At the bedsideFeatures develop "30 minutes to 2 hours after exposure" — "abdominal pain, vomiting and diarrhoea".¹ "In severe cases fluid loss may be excessive, causing decreased blood volume, lowered blood pressure, electrolyte imbalance, hypovolaemic shock and acute tubular necrosis."¹ Profuse diarrhoea distinguishes acute arsenic from lead and thallium, which both tend to constipate.

Peripheral nervous system

TargetLong sensory and motor axons

Why hereThe signature chronic and delayed lesion, and the one that brings patients to a clinic long after the exposure. A cell dependent on axonal transport over a metre of distance has an unusually high and unusually continuous ATP requirement, which is the standard explanation for why long axons fail first when energy metabolism is impaired — an explanation this page offers as reasoning rather than as a sourced fact. Inferred

At the bedside"Peripheral neuropathy" after both acute and chronic exposure,¹ developing after the acute illness has settled, and "severe pain" is described with it in chronic exposure.¹ The overview's own "symmetrical sensorimotor" descriptor is attributed to TOXBASE alone and is therefore not quoted here. The contrast with lead's motor-predominant picture is this page's reading rather than a sourced comparison. Inferred

Skin

TargetKeratinocytes, over months to years

Why hereNamed by the source as "reported to be sensitive indicators" of chronic exposure,¹ which makes the skin a screening organ rather than merely an affected one. Why keratinocytes in particular is not explained by the source and is not explained here. Inferred

At the bedside"palmoplantar hyperkeratosis, hyperkeratinized warts or corns and hyperpigmentation interspersed with areas of hypopigmentation".¹ After an acute exposure, "hyperkeratosis and 'rain-drop' pigmentation" may also appear.¹ Inorganic arsenic is a known human carcinogen¹ and the skin lesions are on the pathway to that.

Nails

TargetNail matrix at the moment of exposure

Why hereIncluded because it is the only physical sign in this library that is a timestamp. The matrix is injured while arsenic is circulating; the visible line is that injured keratin, seen weeks later when it has grown out. The distance of the line from the cuticle dates the exposure, which is why the finding is worth more than its appearance suggests. Established

At the bedside"Single or multiple transverse white lines on the nails ('Mee's lines') may appear several weeks after absorption."¹ Also described after arsine exposure, at two to three weeks.³ Absence proves nothing early, because the line cannot be visible until the nail has grown.

Kidney

TargetTubule, mostly through perfusion

Why hereA secondary organ in acute poisoning, injured by hypovolaemia and by rhabdomyolysis rather than by a specific nephrotoxic action.¹ It matters clinically because it is also the route of elimination — the organ that has to clear the poison is one of the organs the poison's consequences damage. Inferred

At the bedside"Acute tubular necrosis" following hypovolaemic shock,¹ and "renal failure" as part of the multi-organ picture.¹

Cardiovascular system

TargetPeripheral vasculature, chronically

Why hereA chronic-exposure phenomenon and mechanistically unexplained by the source, but distinctive enough to have acquired its own name in the populations where it was described. Inferred

At the bedside"cyanosis, Raynaud's phenomenon and in extreme cases progression to endarteritis obliterans and gangrene of the lower extremities (‘Black foot disease’)".¹ These are features of long-term exposure — typically from contaminated groundwater — rather than of an acute poisoning.

Timeline of effects

Arsenic after acute ingestion — the poison clears in days, the findings take weeks

Time

What you seeWhat is happening

30 min – 2 hGastrointestinal onset
What you see"Abdominal pain, vomiting and diarrhoea."¹ Frequently mistaken for gastroenteritis or food poisoning, particularly where several people ate together.
What is happeningDirect mucosal effect at the highest concentration the poison will reach anywhere, plus the beginning of near-complete absorption.¹
HoursShock and multi-organ failure
What you see"Decreased blood volume, lowered blood pressure, electrolyte imbalance, hypovolaemic shock and acute tubular necrosis."¹ In severe poisoning, "multi-organ involvement within hours" including "rhabdomyolysis, renal failure, respiratory failure, failure of vital cardiovascular and brain functions".¹
What is happeningFluid loss plus a failure of ATP production affecting every tissue at once. This is the phase in which chelation has to be started to be useful — efficacy "declines or disappears as the time interval between metal exposure and onset of chelation increases".⁵
DaysElimination
What you seeSurvivors improve. Urinary arsenic falls, blood arsenic becomes uninformative,¹ and the patient may appear to have recovered entirely.
What is happeningA half-life of 3 to 5 days,¹ with most of the burden excreted as methylated species. The poison is leaving while the injury it initiated is still developing — and unusually, nothing therapeutic is needed for this step to happen.
WeeksThe delayed findings
What you seeA delayed peripheral neuropathy, and "single or multiple transverse white lines on the nails ('Mee's lines')" appearing "several weeks after absorption".¹ "Skin lesions typical of chronic arsenic poisoning may also occur; hyperkeratosis and 'rain-drop' pigmentation."¹
What is happeningTwo different delays with two different causes. The neuropathy is axonal injury declaring itself over the time a long axon takes to fail. The nail line is not an evolving injury at all — it is a completed injury becoming visible as the nail grows out. "After acute exposure, death may be delayed and occur as a result of multiple organ damage."¹

What the mechanism predicts at the bedside

Severe vomiting and diarrhoea with shock, out of proportion to a presumed gastroenteritis, is the acute presentation. Onset is "30 minutes to 2 hours after exposure"¹ and the fluid loss alone can cause hypovolaemic shock and acute tubular necrosis.¹
Send urine, not blood — and send both if the exposure was acute. Arsenic clears rapidly from blood, which limits its usefulness for detecting exposure; urine reflects recent exposure and blood is worth collecting alongside it when an acute exposure is suspected. This is paraphrased and not quoted, because the overview attributes those sentences to TOXBASE alone.
Ask when the exposure was before interpreting a normal result. The half-life is 3 to 5 days.¹ A urine sample taken two weeks after a poisoning is testing for a substance that has largely gone.
For a suspected past exposure, hair and nails are the right specimens — "not considered useful for recent exposures, but may be used for past exposures".¹ This is the inverse of the usual rule and it follows directly from the kinetics.
Look at the nails, and measure from the cuticle. Mee's lines appear "several weeks after absorption",¹ and their position along the nail dates the exposure. Their absence in the first fortnight is expected and excludes nothing.
A painful, symmetrical, sensory-predominant neuropathy weeks after an unexplained severe gastrointestinal illness is the classic sequence, and the two episodes are usually not connected by the patient.
Distinguish the metals by their bowels and their nerves. Acute arsenic causes profuse diarrhoea; lead and thallium tend to cause constipation. Arsenic and thallium are described with a painful neuropathy and lead with a motor-predominant one — a comparison drawn across three pages here rather than stated by any one source.
Do not extrapolate from this page to arsine. Arsine is the same element as a gas and it is a haemolytic poison, not a thiol poison — "arsine primarily targets the erythrocyte and rapidly induces intravascular haemolysis".³ The chelators on this page are not the treatment for it.
Start chelation on clinical suspicion rather than on a confirmed concentration in a severe acute case. Prompt treatment is described as "within minutes to a few hours", and efficacy "declines or disappears" with delay.⁵

The antidote, from the poison's side

The antidote follows directly from the arsenite half of the mechanism. If arsenite poisons by "react[ing] with critical thiols in proteins",² then the antidote is a small molecule presenting two thiols of its own, competing for the arsenic before the patient's enzymes do. That is exactly what dimercaprol is, and its history is unusually literal: chelation for heavy metal intoxication "began more than 70 years ago with the development of British anti-lewisite (BAL; dimercaprol) in wartime Britain as a potential antidote [to] the arsenical warfare agent lewisite".⁵ The antidote on this page was designed against this poison, and the rest of the band inherited it.

Why there are three of them, and why two are better: DMPS (unithiol) and DMSA (succimer) are "dithiol water-soluble analogs of BAL... developed in the Soviet Union and China in the late 1950s", and all three "have remained the mainstay of chelation treatment of arsenic and mercury intoxication for more than half a century".⁵ The advantage of the newer two is stated directly: they "have a higher therapeutic index than BAL and do not redistribute arsenic or mercury to the brain".⁵ An antidote that moves a poison into the central nervous system is a real hazard rather than a theoretical one, and it is the main reason dimercaprol has been displaced where the alternatives are available.
Why the timing is unusually unforgiving: "Treatment should be initiated as rapidly as possible (within minutes to a few hours), as efficacy declines or disappears as the time interval between metal exposure and onset of chelation increases."⁵ The mechanism explains why: a chelator in blood and extracellular fluid can compete for arsenic that is still in blood and extracellular fluid. Once arsenite has bound the thiols of an intracellular enzyme, the competition has already been lost — the antidote prevents binding far better than it reverses it.
What the evidence actually supports: "Controlled animal experiments support a therapeutic role for these chelators in the prompt treatment of acute poisoning by arsenic and inorganic mercury salts."⁵ Enhanced excretion is supported by "animal experiments and in some instances human data".⁵ That is a genuine but animal-weighted evidence base for the acute indication, and it is stated here as such.
Where the evidence stops: For chronic exposure the same review is explicit that although chelation "may accelerate metal excretion and diminish metal burden in some organs, potential therapeutic efficacy in terms of decreased morbidity and mortality is largely unestablished".⁵ Traditional teaching Increasing urinary arsenic is not the same as helping the patient, and only the first has been shown for chronic intoxication. This is a treatment-efficacy downgrade under this library's stated exemption, and it is the same badge, from the same source, as the one on the mercury page.

Critical appraisal

Two evidence-tier downgrades appear on this page, and both are on its central mechanism. The citation for the doubt is unusually direct and is not an absence: Hughes states that "the exact mechanism of the action of arsenic is not known, but several hypotheses have been proposed" in the same abstract that describes the phosphate-substitution and thiol-binding accounts.² This is the lithium precedent, and it is what licenses the badge. That page's inositol/GSK-3 downgrade rests on UK labels positively asserting that the mechanism of action is not known, and the register records the reasoning explicitly: a regulator-approved document asserting the mechanism is unknown contests a claim that the mechanism is X, and that is not a bare absence. Hughes does the same thing for arsenic. The test that distinguishes this page from arsine — whose source makes a near-identical statement and which carries no badge — is what the page asserts, not what the world teaches: this page prints a detailed, load-bearing mechanistic narrative as its central explanatory frame while its own anchor declines to establish it, and the badge marks that gap. The arsine page prints the ignorance itself and reasons downstream of it, so there is nothing there to downgrade. (Two reviewers at the Band E audit split on this. One would have kept both badges and rewritten both justifications, which is what was done; one would have downgraded these two to inferred to match arsine. Both arguments are on the two pages rather than only the outcome.)
The third downgrade is on the efficacy of chelation in chronic exposure, and it is a treatment-efficacy downgrade under this library's stated exemption, citing the same sentence as the mercury page.⁵ It deliberately does not extend to acute poisoning, where the same review supports a therapeutic role.
A deliberate non-downgrade: the greater potency of the trivalent forms is badged established. It is stated flatly by Hughes² and corroborated by the overview's separate observation that arsenite reaches higher tissue concentrations,¹ and no source found contests it.
The specific claim that the overview makes about the timing of the delayed neuropathy — that it may develop up to five weeks after exposure — is deliberately not printed here, because in the source that sentence is attributed to reference 6 alone, and reference 6 is TOXBASE. The neuropathy itself is described from the sections that are independently sourced, and Mee's lines are timed from the overview's reference 17, an IPCS Poisons Information Monograph. This is the finest-grained application of the login-gated rule this library has had to make, and an auditor should check it rather than take it on trust.
A third claim was badged traditional in the first draft and was downgraded to Inferred at audit. The instability of arsenate esters — the reason substitution wastes ATP rather than merely blocking its production — has no citation at all in either source. A traditional teaching badge requires a citation for the doubt, and there is no citation of any kind here, so the badge was not available. It is now flagged in place as unsourced textbook chemistry supplied to make the account coherent. This is the same error the badge rule exists to prevent, caught on the page that argues about that rule most.
One animal finding is described and flagged in place: that arsenite reaches higher tissue concentrations than arsenate, which the overview reports from experimental animals.¹ The overview also notes that "most laboratory animal species appear to be far less sensitive to arsenic toxicity than humans",¹ which is a reason for caution about animal data on this page generally and is why the chelation evidence base is described as animal-weighted rather than as established.
A commonly quoted figure for a fatal ingested dose of inorganic arsenic appears in the overview and is deliberately not reproduced here, along with the animal median-lethal-dose ranges in the same document. Lethal-dose and comparative-lethality framing is permanently out of scope for this library. The overview also carries a UK workplace exposure limit, which is a regulatory figure rather than a clinical threshold and is not presented as one.
EXTRIP has not addressed arsenic.⁴ The absence is uninformative rather than a gap: arsenic distributes widely into tissue and is cleared renally with a half-life of days,¹ so extracorporeal removal has little to offer a patient with functioning kidneys. This is the same category of absence as iron's, not hydrofluoric acid's.

How this page connects to the rest of the library

Arsenic pairs in two directions at once. Chemically and therapeutically it is mercury's twin: the same three dithiol chelators, the same governing review,⁵ the same advantage for the water-soluble analogues, and the same badge on chronic-exposure efficacy. Mechanistically it belongs with lead and thallium in the band's pattern of molecular impersonation — and it is the most literal of the three, because arsenate is not merely admitted to the cell but built into its metabolism. Its sharpest contrast is with arsine, the same element in a different physical form, where the mechanism is haemolysis, the target is the erythrocyte, and every chelator recommended on this page is beside the point. That pairing is the reason arsine has a page of its own rather than a paragraph on this one.

Evidence tier of the mechanisms on this pageEstablishedDemonstrated in humans, or in a model that reproduces the human syndromeInferredConsistent with the biochemistry and widely accepted, but the causal step has not been shown in humansTraditional teachingTaught and repeated but not demonstrated — the source questioning it is cited

References

1
UK Health Security Agency. Arsenic: toxicological overview. Compendium of Chemical Hazards. gov.uk. Kinetics, organ effects and clinical descriptions. Its reference 6 is NPIS TOXBASE and its entire Mechanism of Toxicity section rests on that reference, so the mechanism is corroborated here from reference 2 below instead. Its reference 17, cited for Mee's lines, is an IPCS Poisons Information Monograph.
2
Hughes MF. Arsenic toxicity and potential mechanisms of action. Toxicology Letters 2002;133(1):1–16. PMID 12076506. The independent anchor for every mechanistic claim on this page: pentavalent arsenic replacing phosphate, trivalent arsenic reacting with critical protein thiols, the greater potency of the trivalent arsenicals — and the statement that the exact mechanism of action is not known, which carries this page's evidence-tier downgrades.
3
UK Health Security Agency. Arsine and stibine: toxicological overview. Compendium of Chemical Hazards. gov.uk. Cited here only for the contrast with arsine, which is a haemolytic poison rather than a thiol poison. See the arsine page.
4
EXTRIP Workgroup. Recommendations index. extrip-workgroup.org. Cited for an absence: the published set covers 22 substances and arsenic is not among them. Verified against the site's own menu, 7 September 2026.
5
Kosnett MJ. The role of chelation in the treatment of arsenic and mercury poisoning. Journal of Medical Toxicology 2013;9(4):347–54. PMID 24178900. The origin of dimercaprol as an antidote to lewisite, the development of DMPS and DMSA, their higher therapeutic index and failure to redistribute arsenic to the brain, the timing dependence of efficacy, and the statement that efficacy in chronic intoxication is largely unestablished.
6
National Poisons Information Service. TOXBASE — NPIS 0344 892 0111. The authoritative UK source for arsenic risk assessment and chelation indications. Login-gated, and deliberately not quoted anywhere on this page — including where the government overview above quotes it.

Arsenic

At a glance