Methanol

Why this poison is interesting

Put methanol and ethylene glycol side by side and almost everything matches. Both are small alcohols with minimal intrinsic toxicity. Both are activated by alcohol dehydrogenase. Both are blocked by the same antidote. Both produce a wide osmolar gap that closes as a wide anion gap opens. Both have a latent period during which the patient looks better than they are.

And one destroys the optic nerve while the other destroys the kidney.

That divergence is the most instructive thing in this library, because the two organs are chosen by two completely different kinds of mechanism. Ethylene glycol's target is decided by physical chemistry — oxalate precipitates with calcium where the filtrate is most concentrated, so the proximal tubule is where a salt comes out of solution. Methanol's target is decided by biochemistry — formate inhibits cytochrome c oxidase,¹ so the injury falls on the tissue least able to tolerate a failure of oxidative phosphorylation at that site. Same enzyme at the front of the pathway; entirely different logic at the back.

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

The toxic principle

The poison is formate, the anion of formic acid. Methanol itself is a weak CNS depressant of roughly ethanol's character and little else — which is why the first few hours look like ordinary drunkenness and are, as the review notes, initially misdiagnosed as ethanol poisoning.¹

Formate does one thing that matters, and it does it at a very specific place:

Two consequences follow, and they account for the whole clinical picture.

• A metabolic acidosis with two contributors. Blocked electron transport forces anaerobic metabolism, so formate and lactate accumulate together.¹ The anion gap is therefore not purely formate, which matters when interpreting it. Established
• Selective injury to the optic nerve and retina — see the organ section below. This is the same class of lesion as cyanide's and carbon monoxide's, and it is why methanol sits mechanistically closer to a cellular asphyxiant than to ethylene glycol despite sharing ethylene glycol's front end. Established

Toxicokinetics

Metabolism and the metabolites

Three enzymatic steps, and — unusually — both ends are drugable. Fomepizole blocks the first; folinic acid accelerates the last.

Why the bottleneck is at the end rather than the middle

Ethylene glycol accumulates its toxic metabolite behind glycolate oxidase, a step in the middle of its chain. Methanol accumulates formate behind the final step. The difference is not cosmetic: it means there is no non-toxic intermediate to shunt into, and it means the only way to speed disposal is to supply the cofactor for that last reaction. Folinic acid is not a scavenger and not a blocker; it is a substrate for a traffic jam. Inferred

What changes the answer

• Co-ingested ethanol is strongly protective, and quantifiably so: ethanol concentrations of 100–150 mg/dL inhibit ADH-mediated methanol conversion.¹ Outbreaks of methanol poisoning from adulterated spirits produce a characteristic spread of severity partly for this reason — the people who drank the most ethanol alongside it presented latest and often best. Established
• Folate status. Since disposal of formate is folate-dependent,¹ the malnourished, the alcohol-dependent and anyone with a folate-poor diet are expected to accumulate formate faster. This is the mechanistic rationale for giving folinic acid to everyone rather than only to the deficient. Inferred
• Delay to treatment, which here means delay to blocking the enzyme. Every hour without an ADH blocker is an hour of formate production that no later intervention can undo.

Elimination and accumulation

Methanol poisoning has two accumulation problems that must be tracked separately, and conflating them is the commonest analytical error in this poisoning.

Methanol accumulates in the osmolar gap

It is small, present in molar quantities and osmotically active. The gap is widest early and closes as methanol is converted. A methanol concentration and an osmolar gap are measuring the same thing. Established

Formate accumulates in the anion gap

Along with lactate, generated because inhibited electron transport forces anaerobic metabolism.¹ The gap opens late. Formate is not osmotically significant, so by the time the patient is dangerously poisoned the osmolar gap may have closed. Established

Ion trapping, with a chemical caveat

As the blood becomes acidaemic, the proportion of formic acid existing as the non-ionised acid rises, and non-ionised acid crosses membranes into tissue. This is the same ion-trap that operates in salicylate poisoning, and it means acidaemia here also relocates the poison rather than merely marking it. It is a large part of why EXTRIP treats pH ≤7.15 as a dialysis indication in its own right, independent of any concentration.² Inferred

The EXTRIP position

Two hundred and seventy-two relevant publications were identified, with publication and selection bias noted.² ECTR is recommended in severe methanol poisoning (1D), where severe means any of:

• Coma, seizures, or new vision deficits (each 1D)²
• Blood pH ≤7.15, or persistent metabolic acidosis despite adequate supportive measures and antidotes, or a serum anion gap higher than 24 mmol/L — which EXTRIP calculates here as [Na⁺] − [Cl⁻] − [HCO₃⁻], without potassium, unlike the ethylene glycol set (each 1D)²
• Serum methanol >700 mg/L (21.8 mmol/L) on fomepizole, >600 mg/L (18.7 mmol/L) on ethanol, or >500 mg/L (15.6 mmol/L) with no ADH blocker (each 1D); in the absence of a concentration, the osmol gap may be informative (1D)²
• In the context of impaired kidney function (1D)²
• Stop when methanol is <200 mg/L (6.2 mmol/L) and clinical improvement is observed (1D). ADH inhibitors are to be continued during dialysis, as well as folic acid (1D)²

Target organs — and why those

Timeline of effects

What the mechanism predicts at the bedside

Why visual symptoms are a dialysis indication and not an ophthalmic problem

Because they are a marker of the metabolite, not of the eye. New vision deficits sit alongside coma, seizures and pH ≤7.15 in EXTRIP's list of strong indications (1D),² and the logic is that visual change means formate has reached a concentration sufficient to inhibit cytochrome c oxidase in the most vulnerable tissue.¹ A patient describing a snowstorm is telling you their formate concentration. Inferred

Why the diagnosis is made on the anion gap in a late presenter

Because the osmolar gap will already have closed. The two gaps trade places across the latent period, and the patients who present late — with visual symptoms and a severe acidosis — are the ones whose methanol has been almost entirely converted. A negative osmolar gap in a patient with an unexplained high-anion-gap acidosis and visual symptoms is consistent with severe methanol poisoning, not against it. Established

Why there are two antidotes and both are given

Because the pathway can be attacked at both ends and the two interventions do different things. Fomepizole prevents formate being made and does nothing about formate already present. Folinic acid accelerates disposal of formate that already exists — the tetrahydrofolate pathway is the key route for converting formic acid to carbon dioxide and water, and folinic acid is given on the rationale that it speeds that step.¹ EXTRIP requires that both ADH inhibitors and folic acid be continued through dialysis (1D).² A patient given fomepizole alone has had the tap turned off with the bath still full. Inferred

Why the threshold for dialysis is lower without an antidote

Covered in the kinetics section, and it is the clearest example on this site of a mechanism dictating a number. The same methanol concentration means something different depending on whether it is going to become formate.² The concentration is a measure of future formate, and the antidote changes the exchange rate.

Why an outbreak is a different clinical problem from a single poisoning

Methanol poisoning tends to occur as mass poisonings, with a clear predilection for poorer communities.¹ Adulterated spirits produce clusters in which people have drunk different amounts at different times alongside different amounts of protective ethanol — so a cohort presents across the entire timeline at once, some inebriated, some in the latent interval, some blind. EXTRIP notes explicitly that the relative importance of individual indications for triaging patients when need exceeds resources is unknown.² That is an honest admission of a real and recurring problem.

The antidote, from the poison's side

Fomepizole

A competitive inhibitor of alcohol dehydrogenase, acting at the first step. It prevents rather than reverses: formate already made is untouched, and vision already lost is not recovered. Its entire benefit is in metabolite not made, which is why time to administration dominates outcome. Established

Ethanol

The same idea by competition — give the enzyme its preferred substrate. Ethanol concentrations of 100–150 mg/dL inhibit methanol conversion by ADH.¹ Workable where fomepizole is unavailable, and inferior for the same reasons as in ethylene glycol: a target concentration must be reached and held, and ethanol is itself a CNS depressant and a hypoglycaemic agent at those concentrations. EXTRIP's lower dialysis threshold on ethanol than on fomepizole encodes exactly this reduced confidence in the block.²

Folinic acid

Acts at the last step, supporting the tetrahydrofolate-dependent conversion of formic acid to carbon dioxide and water.¹ This is the only antidote in Band A that accelerates disposal of a toxic metabolite rather than preventing its formation or scavenging it. The rationale is biochemically direct; the human outcome evidence is thin, and it is given because it is plausible, cheap and harmless. Inferred

Haemodialysis

The third arm, and the only one that removes formate already distributed. EXTRIP makes intermittent haemodialysis the modality of choice (1D) and requires that both antidotes be continued through it (1D) — because both are themselves dialysed out.²

Critical appraisal

1. The published mechanism contains a chemical impossibility that has to be read past. Inferred The source review states that falling blood pH increases non-ionised methanol and thereby tissue penetration.¹ Methanol has no ionisable proton at physiological pH; formic acid does. The mechanism is real and the molecule named is wrong. It is recorded here because a mechanism library's characteristic failure is repeating a well-formed sentence without checking whether it can be true.
2. Cytochrome c oxidase inhibition is well established; the quantitative link to the human lesion is not. Established for the enzyme inhibition itself, which is demonstrated in vitro and in vivo at 5–30 nM formic acid.¹ What is inferred is that this specific mechanism, rather than the accompanying oxidative stress, cytokine response and acidaemia, is what produces the human optic neuropathy — the review itself lists those other mechanisms as also playing an important role.¹
3. Folinic acid is given on biochemical logic rather than on outcome data. Inferred The tetrahydrofolate pathway's role in formate disposal is established;¹ that supplementing it changes human outcomes is a plausible extrapolation the review describes with "it is believed that". It is cheap and harmless, so the risk–benefit is easy — but it should not be described as an antidote of the same evidential standing as fomepizole.
4. The 30–40% persistent-visual-sequelae figure is a review-level summary across heterogeneous outbreaks, not a cohort statistic from a defined population.¹ It is likely to depend heavily on time to treatment, which varies enormously between the outbreak settings the literature is drawn from. The same caution applies to the 18–44% mortality range — a range that wide is describing different health systems, not different poisons.
5. Every EXTRIP methanol recommendation is graded 1D — strong recommendation, very low quality evidence.² The workgroup identified 272 relevant publications and explicitly noted publication and selection bias. The thresholds are consensus positions built on case-level data and should be read as such, and the workgroup says as much about triage in outbreaks.
6. The comparison with ethylene glycol is a teaching device and should not be overdrawn. The two poisons share their front end genuinely, but methanol's back end is closer to that of a cellular asphyxiant — its nearest mechanistic relative in this library is not ethylene glycol at all but carbon monoxide, and cyanide when that page is written.

References

1. 1
   Liberski S, Kaluzny BJ, Kocięcki J. Methanol-induced optic neuropathy: a still-present problem. Arch Toxicol 2022;96(2):431–51. PMC8731680 Open access. Source of the cytochrome c oxidase mechanism and the 5–30 nM formic acid figure, the full metabolic pathway with ADH1b, ALDH and 10-formyltetrahydrofolate dehydrogenase, the 1-minute formaldehyde half-life, the 30–90 min peak, the 10–12 h asymptomatic period, the 12–24 h CNS and 12–48 h visual timings, the ~50% incidence of visual symptoms, the 30–40% persistent sequelae and 18–44% mortality, the putaminal necrosis and retrolaminar demyelination findings, the 100–150 mg/dL ethanol inhibition figure, and the primate susceptibility statement. It is also the source of the non-ionised-methanol wording discussed in the appraisal. Verified 31 Aug 2026 from the full text.
2. 2
   Roberts DM, Yates C, Megarbane B, Winchester JF, Maclaren R, Gosselin S, Nolin TD, Lavergne V, Hoffman RS, Ghannoum M; EXTRIP Work Group. Recommendations for the role of extracorporeal treatments in the management of acute methanol poisoning: a systematic review and consensus statement. Crit Care Med 2015;43(2):461–72. PubMed 25493973 · Recommendation set also published at extrip-workgroup.org/methanol. Source of every indication, concentration threshold, cessation criterion and grade quoted above, the 272-publication evidence base with its noted publication and selection bias, the instruction to continue ADH inhibitors and folic acid during dialysis, the anticoagulation caution, and the statement about triage in outbreaks. Verified 31 Aug 2026 from the abstract and the workgroup's own published recommendation page.
3. 3
   Kraut JA, Mullins ME. Toxic alcohols. N Engl J Med 2018;378(3):270–80. PubMed 29342392 Cited as the standard clinical review of the toxic alcohols. Paywalled; not used as the source of any figure on this page. Citation verified 31 Aug 2026.
4. 4
   TOXBASE — methanol; fomepizole. National Poisons Information Service. toxbase.org (NHS login required. NPIS: 0344 892 0111. The authoritative UK source for treatment thresholds, antidote and folinic acid dosing and dialysis criteria — and the first call in a suspected outbreak. Login-gated, so not quoted here.)