Why this poison is interesting

Hydrogen sulphide is the only poison in this library that the body manufactures deliberately and uses as a signal. The UKHSA overview records that it is produced endogenously from sulphydryl-containing amino acids, by bacteria in the gut and mouth, and "enzymatically in the brain and smooth muscle"; that it can compose "up to 10% of intestinal gasses"; and that at normal concentrations it "acts as a 'gasotransmitter'" involved in "vasodilation and vasorelaxation, and neuromodulation".¹ A molecule with its own synthetic enzymes and its own physiological role becomes, three orders of magnitude higher, a poison that stops the electron transport chain.

The second reason is the failure of the warning. The smell is famously detectable — the overview gives an odour threshold of "approximately 0.011mg/m3 (0.008ppm)" — and it disappears at exactly the wrong point: above "140mg/m3 (100 ppm) olfactory paralysis can occur", and "the loss of odour perception makes hydrogen sulphide especially dangerous".¹ The gas anaesthetises the sense that detects it. A worker who reports that the smell has gone has not moved to safety; they have crossed a threshold.

The third is the shape of the exposure. Collapse is described in case reports "following only a few breaths", an effect called knockdown, and in some cases individuals "appear to make a rapid and complete recovery where exposure is promptly terminated".¹ Both halves of that sentence matter. The first is why the casualty is found unconscious at the bottom of a tank and why the person who climbs in after them becomes the second casualty. The second is why the correct response to a knockdown is to get the patient into air rather than to assume an irrecoverable injury.

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

The toxic principle

The mechanism as the overview states it "is thought to be the direct inhibition of cellular enzymes, such as cytochrome c oxidase, which is involved in cellular oxidative processes and energy production"; the inhibition "leads to disruption of the electron transport chain and impairs oxidative metabolism, leading to anaerobic metabolism, decreased ATP production and generation of lactic acid", and "the nervous and cardiac tissues are most sensitive to the disruption of oxidative metabolism".¹

That is, to a first approximation, the cyanide page's mechanism paragraph with a different molecule in it — and the two pages are written to be read together, because the differences are more interesting than the similarity. Cyanide binds the ferric iron of cytochrome c oxidase to form "a stable but reversible complex", and the reversibility is what lets an antidote compete for the ion. Hydrogen sulphide inhibits the same enzyme and, crucially, it is cleared by its own metabolism so quickly that the competition an antidote would offer is largely redundant. The gas that inhibits harder recovers faster.

Why the brainstem, and why that is a distribution fact rather than a susceptibility argument

Most organ-selectivity claims in toxicology are arguments about vulnerability — this tissue has the highest oxygen demand, that one the least reserve. Here there is a measured distribution finding as well: hydrogen sulphide is distributed "to the brain, liver, kidney, pancreas, and small intestine", and the brainstem has the greatest net uptake following exposure.¹ The tissue that takes up most of the gas is the tissue that runs breathing. That is a far more satisfying explanation of apnoea and instantaneous collapse than a general statement about neuronal oxygen demand, and it is the same kind of explanation as paraquat's active transport into lung — selectivity achieved by where the poison goes. Established

One consequence deserves stating because it distinguishes this poisoning from every simple asphyxiant. A knockdown is not caused by a lack of oxygen in the air. Hydrogen sulphide can drop somebody at concentrations that leave the atmosphere perfectly breathable, which is why oxygen monitoring alone does not make a confined space safe, and why the collapse does not behave like the gradual impairment of carbon dioxide and the simple asphyxiants.

Toxicokinetics

This is the fastest set of kinetics in the library in both directions, and the fact that recovery can be "rapid and complete"¹ is a direct consequence of them. The gas is absorbed instantly, metabolised fast, and stored hardly at all.

Hydrogen sulphide — a poison with almost no residence time
Parameter	Therapeutic	In overdose	Why it changes
Endogenous production	From sulphydryl amino acids, gut and mouth bacteria, and enzymatically in brain and smooth muscle; up to "10% of intestinal gasses"¹	Irrelevant to the poisoning, and it explains why a disposal pathway already exists	The body already handles this molecule continuously. A poisoning is therefore an overwhelming of a working system rather than the arrival of something foreign — which is the mechanistic reason detoxification is so quick once exposure ends.
Absorption	"Exogenous hydrogen sulphide is absorbed rapidly through the respiratory tract"¹	Same. Effect within a breath or two¹	There is no absorption phase to intervene in. The only exposure variable that can be changed is the duration, which is why extrication is the whole of the early treatment.
Distribution	"Distributed throughout the body (partially dissociated as HS-)" to "the brain, liver, kidney, pancreas, and small intestine"¹	Greatest net uptake in the brainstem¹	The row that explains the syndrome. Apnoea and abrupt loss of consciousness are what you would predict from a poison that concentrates where breathing is controlled, rather than from a general cellular energy failure.
Metabolism	Three routes: "oxidation, methylation, and reaction with metalloproteins or disulphide-containing proteins". The major one is "oxidation in the liver to form thiosulphate, which is further metabolised to sulphate"¹	Same routes, saturable at extreme exposure	Every route is a detoxification route. There is no bioactivation anywhere on this page — a property shared with paraquat and nitrous oxide and with very little else in the library.
Storage	—	"Hydrogen sulphide is rapidly metabolised and excreted, therefore storage in the body is limited"¹	No reservoir, no redistribution, no relapse. Contrast the lipophilic organophosphates, which are released from fat for days. A patient removed from the atmosphere is not carrying a further dose.
Elimination	"Rapidly eliminated from the body in the form of sulphate (free sulphate or thiosulphate) in urine"; also unchanged in exhaled air, faeces and flatus¹	Unchanged	Elimination is complete and fast, and it is the treatment. The gap between this and paraquat, where excretion of the poison is irrelevant to the outcome, is the widest in the band.
The exposure axis	—	Concentration, not cumulative dose. Odour threshold "0.011mg/m3 (0.008ppm)"; olfactory paralysis above "140mg/m3 (100 ppm)"¹	A poisoning governed by a single breath at a concentration rather than by an ingested quantity. There is no level to measure and no clearance to wait for; the history is the atmosphere, not the dose.
Dialysability	—	Never assessed and not applicable. EXTRIP has published no recommendation covering hydrogen sulphide²	A gas cleared by hepatic oxidation within minutes cannot be a target for extracorporeal removal, and the patient either recovers or dies long before a circuit could run. This is an absence reflecting a question not worth asking, in the same category as nitrous oxide.

Metabolism and the metabolites

There is no toxic metabolite, and the detoxification pathway leads to a molecule that appears elsewhere in this library as an antidote. Hydrogen sulphide is oxidised to thiosulphate — the sulphur donor that cyanide's rhodanese pathway runs short of — a fact belonging to that page's source rather than to this one's. The two cellular asphyxiants are connected through the same sulphur chemistry from opposite ends.

Hydrogen sulphide — every route out is a detoxification route

Hydrogen sulphideEndogenous at trace concentration as a gasotransmitter; a poison at industrial concentration¹
Distributed, with greatest net uptake in the brainstemThe distribution step is the organ-selectivity step¹
Inhibits cytochrome c oxidaseElectron transport chain disrupted"Anaerobic metabolism, decreased ATP production and generation of lactic acid"¹
Hepatic oxidation — the major detoxification routeThiosulphate, then sulphateExcreted in urine.¹ The same thiosulphate that is given as an antidote in cyanide poisoning
Methylation, and reaction with metalloproteins or disulphide-containing proteinsMinor disposal routesNamed in the source as the other two of three pathways¹
Failure of the nervous and cardiac tissue"The nervous and cardiac tissues are most sensitive to the disruption of oxidative metabolism"¹ — respiratory paralysis, convulsions, arrhythmia¹
Recovery, if exposure endsStorage is limited and elimination is rapid¹ — hence "a rapid and complete recovery where exposure is promptly terminated"¹

Elimination and accumulation

Nothing accumulates. "Storage in the body is limited"¹ and elimination is rapid, so a patient removed from the atmosphere is no longer being poisoned — which is a genuinely unusual statement in this library and the most important practical fact on the page.

Against that sits the most hopeful sentence in Band D. Some individuals "appear to make a rapid and complete recovery where exposure is promptly terminated".¹ A poison with no storage, no metabolite and a fast endogenous disposal route offers something almost nothing else here does: an injury that stops when the exposure stops. The outcome is determined by how long the brainstem was inhibited and by whether hypoxic-ischaemic injury supervened — not by anything still circulating.

Target organs — and why those

The source names the sensitive tissues directly — "the nervous and cardiac tissues"¹ — and adds something better than a susceptibility argument for the first of them: a measured uptake distribution favouring the brainstem.¹

Brainstem

TargetRespiratory control centres

Why hereThe greatest net uptake of any tissue following exposure.¹ Selectivity here is achieved by distribution and then compounded by susceptibility: the region that takes up most of the gas is also nervous tissue, named as among the most sensitive to disrupted oxidative metabolism.¹ This is why respiratory paralysis is described in the same sentence as collapse rather than following it as a secondary consequence. Established

At the bedside"Collapse ('knockdown'), respiratory paralysis, cyanosis, convulsions, coma", and death "within minutes" at high concentration.¹ Apnoea with a preserved circulation is the pattern that makes early ventilation worth attempting.

Olfactory epithelium

TargetOlfactory neurones

Why hereA target organ whose failure is the poisoning's most dangerous feature. Above "140mg/m3 (100 ppm) olfactory paralysis can occur" and "the loss of odour perception makes hydrogen sulphide especially dangerous".¹ Animal data support a real lesion rather than mere sensory adaptation: in rats, "loss of olfactory neurones and basal cell hyperplasia" followed repeated exposure at 42 mg/m³ (30 ppm) and above.¹ The animal finding is chronic and repeated, and the human olfactory paralysis is acute — the flag is here rather than buried. Established

At the bedsideThe smell of rotten eggs at trace concentrations, then no smell at all at concentrations that cause knockdown. A worker reporting that the smell has gone should be treated as having moved into greater danger, not less.

Heart

TargetCardiac myocytes

Why hereNamed alongside nervous tissue as most sensitive to the disruption of oxidative metabolism.¹ The reasoning is the same as on the cyanide page — continuous, inflexible ATP demand with little tolerance of anaerobic metabolism — and the consequence is the same conversion of a utilisation failure into a delivery failure. Established

At the bedside"Cardiac arrhythmias"¹ and cardiovascular collapse. Once the output falls, the presentation is indistinguishable from any other arrest.

Eyes and respiratory tract mucosa

TargetSurface epithelium at lower concentrations

Why hereA separate and much less dangerous mechanism: direct irritation rather than enzyme inhibition. "Acute inhalation exposure to low concentrations... will irritate the eyes and respiratory tract, resulting in sore throat, cough and dyspnoea."¹ It matters because it defines the exposures the gas does not kill at, and it is what a chronic low-level complaint usually is. Established

At the bedsideSore throat, cough, dyspnoea and the keratoconjunctivitis classically reported in gas workers. These patients are not about to collapse — the syndromes separate by concentration, not by duration.

Timeline of effects

Hydrogen sulphide — no latent phase, and a decision made in seconds

Time

What you seeWhat is happening

Immediate, low concentrationThe warning works
What you seeThe smell of rotten eggs. Sore throat, cough, dyspnoea, eye irritation.¹
What is happeningDetectable from an odour threshold of "approximately 0.011mg/m3 (0.008ppm)".¹ Direct mucosal irritation, not enzyme inhibition — a different mechanism from the one that kills.
Above ~100 ppmThe warning stops working
What you seeThe smell disappears. The patient reports the air has cleared.
What is happening"At concentrations greater than 140mg/m3 (100 ppm) olfactory paralysis can occur. The loss of odour perception makes hydrogen sulphide especially dangerous."¹ The one protective signal is removed at the threshold where it becomes necessary.
SecondsKnockdown
What you see"Individuals abruptly collapsing and becoming unconscious following only a few breaths."¹ No warning to bystanders.
What is happeningCytochrome c oxidase inhibition throughout the body, with the greatest net uptake in the brainstem.¹ Respiratory control fails alongside consciousness rather than after it.
MinutesRespiratory and cardiac failure
What you see"Respiratory paralysis, cyanosis, convulsions, coma, cardiac arrhythmias, and death within minutes" at high concentration.¹
What is happeningThe tissues named as most sensitive to disrupted oxidative metabolism¹ fail first. From here the poisoning becomes a hypoxic-ischaemic problem, and the gas is no longer the thing to treat.
On removal to airRapid clearance
What you seeIn some cases "a rapid and complete recovery where exposure is promptly terminated".¹
What is happeningAbsorption stops immediately and hepatic oxidation to thiosulphate and sulphate proceeds fast; "storage in the body is limited".¹ Extrication is the definitive treatment, and there is nothing left circulating to reverse.
Days–weeksNeurological sequelae in survivors
What you seePersistent cognitive, memory and neurobehavioural complaints are reported after severe exposures.¹
What is happeningHypoxic-ischaemic injury sustained during the knockdown, rather than continuing action of a gas that has long gone. The same reasoning, and the same uncertainty, as the equivalent claim on the cyanide page. Inferred

What the mechanism predicts at the bedside

Do not enter the space. Knockdown occurs after a few breaths¹ and the smell has already failed as a warning above 100 ppm¹. Rescue without breathing apparatus produces further casualties, and this is the mechanism predicting it rather than a safety slogan.
Extrication is the treatment. Storage is limited and elimination is rapid¹; a patient in clean air is no longer being poisoned.
A collapsed worker in a confined space with a normal-looking atmosphere is this diagnosis until proven otherwise — hydrogen sulphide drops people at concentrations that leave the air perfectly breathable, so an oxygen meter reading normal does not exclude it.
Expect a lactic acidosis.¹ Do not expect the saturations to be reassuring — unlike cyanide, the acute syndrome here includes "respiratory paralysis" and "cyanosis" among its named features¹, and an apnoeic patient is hypoxaemic. The cellular block is the same; the reason the oximeter can look normal in cyanide poisoning does not transfer to a poison that stops the brainstem.
Apnoea may precede circulatory arrest, because the brainstem takes up more of the gas than anything else.¹ Early ventilation of an apnoeic patient with a pulse is the intervention the mechanism most supports.
'The smell went away' is a red flag in the history, not reassurance.¹
Recovery can be rapid and complete¹, so resuscitation is worth pursuing — the poison is gone within minutes of extrication.
Low-concentration exposure and knockdown are different syndromes, separated by concentration rather than duration. A patient with sore eyes and a cough is not on a trajectory to collapse.
There is no established antidote and no level to measure. The history is the atmosphere and the examination is the patient.
Dialysis has no role and has never been assessed² — the gas is cleared faster than a circuit can be assembled.

The antidote, from the poison's side

There is no established antidote, and unusually for this library that is a statement about the mechanism rather than about the evidence base. The endogenous disposal route is fast and is not substrate-limited, so the two strategies that work against cyanide — restocking a cofactor and scavenging the ion — have far less to offer here.

Removal from the atmosphere: The definitive treatment, and the only one with an unambiguous mechanistic case. Absorption is immediate¹ and so is its cessation; storage is limited and elimination rapid.¹ Everything else on this list is supportive.
Oxygen: Given, on the same reasoning as in cyanide poisoning: it does not lift the enzyme inhibition, and raising the dissolved fraction may support the uninhibited remainder while the gas is cleared. The contrast with paraquat, where oxygen is fuel for the mechanism, is worth holding in mind across the band.
Ventilation: The intervention the distribution data most support. If the brainstem has the greatest net uptake¹ and respiratory paralysis is part of the acute syndrome¹, then supporting ventilation through the clearance period addresses the proximate cause of death directly.
Sodium nitrite: Proposed on the theory that induced methaemoglobin offers ferric iron to bind sulphide, exactly as it is used against cyanide. This page makes no claim that it works. The theory is coherent; the practical objection is that a gas cleared this fast may be gone before methaemoglobin can be generated, and deliberately disabling haemoglobin in a hypoxic patient has a cost — see methaemoglobin inducers. No source cited here establishes benefit. Inferred
Thiosulfate: The argument for it is weaker here. Thiosulphate treats cyanide poisoning by restocking a rate-limiting sulphur donor — a fact sourced on the cyanide page rather than in this page's own reference. Hydrogen sulphide's own detoxification produces thiosulphate rather than consuming it¹ — there is nothing to restock.
Hyperbaric oxygen: Described in case reports. This page has no source establishing benefit and makes no claim either way.
Extracorporeal removal: Never assessed² and not applicable to a gas cleared by hepatic oxidation within minutes.

Critical appraisal

No evidence-tier downgrade appears on this page, and the specific paper that might have supplied one has now been checked. The Band D audit left open whether Milby and Baselt's Clarification of some controversial issues³ contests the olfactory-paralysis or knockdown teaching, since a reviewer declined to supply a citation it had not opened. It was retrieved for Band E and it does not: it addresses the neurotoxicity, pulmonary oedema, exposure standards and analytical pitfalls, and says nothing against either claim. It does add that some neurotoxic effects are "almost certainly a result of hypoxia secondary to H2S-induced respiratory insufficiency" rather than direct toxicity, and that there is "suggestive evidence of hyperactive airway responses in some individuals following brief H2S-induced unconsciousness (knockdown)".³ So no source found contests the cytochrome c oxidase mechanism, the brainstem uptake finding or the olfactory paralysis, and a badge would have to be manufactured to appear here — which is the failure mode this library is most prone to. (Only the abstract was retrievable; the full text is paywalled, and this page does not claim to have parsed it.)
The mechanism statement in the source is hedged and this page should not un-hedge it. The overview says the mechanism "is thought to be" direct inhibition of cellular enzymes such as cytochrome c oxidase.¹ The badge on the brainstem card is applied to the uptake finding, which is stated without hedging, not to the enzymology.
The knockdown description comes from case reports, as the source itself says.¹ Case reports establish that the phenomenon occurs and cannot establish how often, at what concentration, or what proportion recover completely. The word some in "some cases... a rapid and complete recovery" is doing real work and is quoted rather than paraphrased.
The olfactory neurone loss is a rat finding after repeated exposure¹ and the human olfactory paralysis is acute. The organ card flags this rather than presenting the animal data as the mechanism of the acute phenomenon. Describing an animal finding in human terms is a failure shape this library has recorded five times in Band C, and this is the only place on this page where the risk arises.
The argument that no antidote is needed is inference, badged as such. It follows from the disposal pathway not being substrate-limited and from the speed of clearance, and no source cited here tests any antidote against placebo.
The nitrite discussion states a theory and declines to endorse it. It is included because a reader will encounter the proposal, and it is badged inferred with an explicit statement that no source here establishes benefit.
The delayed neurological sequelae are attributed to hypoxic-ischaemic injury and badged inferred. The source lists persistent neurological effects¹ without adjudicating their mechanism.
No lethal concentration and no exposure limit is printed as a clinical threshold. The odour threshold and the olfactory paralysis concentration are printed because they describe the failure of a warning system, which is a mechanism; they are not a dose–response curve and must not be read as one.
EXTRIP's silence² is uninformative here in the same way it is for nitrous oxide and cyanide — the question is not merely unasked but incoherent on this timescale.

How this page connects to the rest of the library

It is the twin of cyanide, and neither page is complete without the other: same enzyme, same histotoxic hypoxia, and completely divergent antidote logic because one disposal pathway is substrate-limited and the other is not. It shares its organ-selectivity form with paraquat — the poison goes where the damage is, by measured distribution rather than by argued susceptibility. It is the counter-example to the simple asphyxiants on the carbon dioxide page, which drop people by displacement rather than by poisoning oxygen's use — carbon dioxide itself being that page's stated exception, and the distinction matters at a scene because the two demand different monitoring. And within Band D it is the page with no delay at all, which is what makes the band's other latent phases legible as opportunities rather than as mere curiosities.

References

1
Hydrogen sulphide: toxicological overview. UK Health Security Agency, Compendium of Chemical Hazards. (Mechanism: "Hydrogen sulphides mechanism of action is thought to be the direct inhibition of cellular enzymes, such as cytochrome c oxidase, which is involved in cellular oxidative processes and energy production. Such enzyme inhibition leads to disruption of the electron transport chain and impairs oxidative metabolism, leading to anaerobic metabolism, decreased ATP production and generation of lactic acid. The nervous and cardiac tissues are most sensitive to the disruption of oxidative metabolism." Endogenous production from sulphydryl amino acids, gut and mouth bacteria, and "enzymatically in the brain and smooth muscle"; "up to 10% of intestinal gasses"; at normal levels acts as a "gasotransmitter" regulating "vasodilation and vasorelaxation, and neuromodulation". Distribution "partially dissociated as HS-" to "the brain, liver, kidney, pancreas, and small intestine"; "the brainstem has the greatest net uptake following exposure"; "rapidly metabolised and excreted, therefore storage in the body is limited". Metabolism "via three pathways: oxidation, methylation, and reaction with metalloproteins or disulphide-containing proteins", the major route "oxidation in the liver to form thiosulphate, which is further metabolised to sulphate"; eliminated as sulphate in urine, also in exhaled air, faeces and flatus. Acute high concentration causes "collapse ('knockdown'), respiratory paralysis, cyanosis, convulsions, coma, cardiac arrhythmias, and death within minutes". Odour threshold "approximately 0.011mg/m3 (0.008ppm)"; above "140mg/m3 (100 ppm) olfactory paralysis can occur" and "the loss of odour perception makes hydrogen sulphide especially dangerous". Case reports of "individuals abruptly collapsing and becoming unconscious following only a few breaths" with, "in some cases", "a rapid and complete recovery where exposure is promptly terminated". Rat data: loss of olfactory neurones and basal cell hyperplasia after repeated exposure at 42 mg/m³ (30 ppm) and above.) gov.uk
2
EXTRIP Workgroup — published recommendations index. Extracorporeal Treatments in Poisoning Workgroup. Cited for the absence of any recommendation covering hydrogen sulphide. extrip-workgroup.org/recommendations
3
Milby TH, Baselt RC. Hydrogen sulfide poisoning: clarification of some controversial issues. American Journal of Industrial Medicine 1999;35(2):192–5. PMID 9894543. Retrieved at the Band E session to close an open thread from the Band D audit. Cited for what it does NOT contest. Note the authors are Milby and Baselt — the Band D handoff recorded a different second author. Abstract only; the full text is paywalled.

Hydrogen sulphide

At a glance