Why this poison is interesting

Calcium-channel blocker overdose is the best example in this library of a poisoning whose treatment makes no sense until you notice which tissues share the target. The L-type voltage-gated calcium channel is not a cardiac protein that happens to appear elsewhere. It is the same channel on the cardiomyocyte, on vascular smooth muscle, on the sinoatrial and atrioventricular nodes, and on the pancreatic beta cell, where calcium influx is the trigger for insulin exocytosis.

Block it everywhere at once and the result is a patient in cardiogenic and vasoplegic shock who cannot release insulin. The myocardium, which under stress shifts from fatty acids towards carbohydrate, is denied both the hormone and the substrate at precisely the moment it needs them. That is why the glucose rises, why the rise is a clue rather than a complication, and why giving very large amounts of insulin — a treatment that would be incomprehensible in any other form of shock — is a first-line recommendation.⁶

The second reason this page exists is a pharmaceutical accident rather than a pharmacological one. These drugs are overwhelmingly prescribed as modified-release preparations, and the verapamil label states that in overdose with slow-release forms the release of the active drug and the absorption in the intestine may take more than 48 hours, with lumps of incompletely dissolved tablets along the entire length of the gastrointestinal tract, which function as active drug depots.² A poisoning whose absorption phase outlasts two days invalidates every intuition about observation periods built on immediate-release drugs.

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

The toxic principle

The L-type channel opens in response to depolarisation and admits calcium. What that calcium then does depends entirely on the cell it enters, and this is the whole of calcium-channel blocker toxicology.

Cardiac myocyte: Calcium entry triggers calcium-induced calcium release from the sarcoplasmic reticulum, which is the contraction. Blockade is therefore directly negatively inotropic. Established
Sinoatrial and atrioventricular nodes: These cells have no fast sodium current worth speaking of — their upstroke is a calcium current. Blocking L-type channels here does not merely slow conduction, it removes the depolarising current that generates it. This is why AV block is so prominent, and why it appears with the cardioselective agents. Established
Vascular smooth muscle: Calcium entry drives calmodulin-dependent myosin light-chain kinase activity and therefore tone. Blockade produces vasodilatation — the therapeutic effect of the dihydropyridines and, in overdose, distributive shock. Established
Pancreatic beta cell: Glucose metabolism closes ATP-sensitive potassium channels, the cell depolarises, L-type calcium channels open, and the calcium influx triggers insulin granule exocytosis. Blocking the channel blocks the final common step of insulin secretion. Established

The second axis of variation is selectivity, and it behaves exactly as selectivity does everywhere else in this library: it is a property of the therapeutic dose and it fails at the toxic one.

Non-dihydropyridines — verapamil, diltiazem: Bind preferentially to channels in cardiac and nodal tissue. Overdose is bradycardia, AV block of every degree, Wenckebach with or without escape, sinus arrest and asystole; the verapamil labels list all of these, together with hyperglycaemia, metabolic acidosis and acute respiratory distress syndrome.¹² Established
Dihydropyridines — amlodipine, nifedipine and relatives: Vascular selectivity at therapeutic doses. The amlodipine label describes overdose as excessive peripheral vasodilatation and possibly reflex tachycardia, with marked and probably prolonged systemic hypotension.⁴ A tachycardic hypotensive patient is not thereby excluded from calcium-channel blocker poisoning — it is the expected dihydropyridine picture. In large overdose the selectivity is lost and cardiac depression appears too. Established

Toxicokinetics

Verapamil, diltiazem and amlodipine — three agents, three separate kinetic traps
Parameter	Therapeutic	In overdose	Why it changes
Absorption	Verapamil: >90% absorbed from the small intestine; SR peak at 4–5 h². Diltiazem: almost completely absorbed³. Amlodipine: peak blood levels at 6–12 h⁴	Modified-release overdose: absorption may take more than 48 hours, with undissolved tablet material acting as depots along the gut²	The most dangerous single fact on this page. The clinical course is governed by a dissolution rate rather than by a half-life. A patient who is well at six hours has not passed a meaningful test, and a patient who has stabilised may still be absorbing.
Bioavailability	Verapamil SR about 33% after a single dose, owing to extensive hepatic first pass — and about two times higher with repeated administration²	Rises further as first-pass capacity is exceeded	A high-extraction drug delivers a disproportionately greater systemic exposure as the dose rises. The ingested amount and the achieved concentration are not linearly related, which is a large part of why dose-based risk assessment is unreliable here.
First-pass metabolism	Diltiazem: extensive, and saturable at higher doses, resulting in non-linear accumulation³	Saturated	This is the library's four-pattern 'saturation' failure, printed on a UK label. Once the first-pass enzymes are saturated, each additional tablet contributes far more systemic drug than the last.
Volume of distribution	Verapamil 1.8–6.8 L/kg²; diltiazem 3–11 L/kg³; amlodipine approximately 21 L/kg⁴	Unchanged	All three are far too large for extracorporeal removal to reach a meaningful share of the body burden — the same arithmetic that defeats dialysis in digoxin poisoning.
Protein binding	Verapamil approximately 90%²; diltiazem about 80%³; amlodipine 97.5%⁴	Unchanged	High binding compounds the volume-of-distribution problem. Amlodipine's label draws the conclusion itself: since amlodipine is highly protein-bound, dialysis is not likely to be of benefit.⁴
Metabolism	Verapamil by CYP3A4, CYP1A2, CYP2C8, CYP2C9 and CYP2C18 to 12 metabolites; norverapamil has about 10–20% of the parent's activity and reaches similar steady-state concentrations². Diltiazem to desacetyldiltiazem, 25–50% as potent and present at 10–20% of parent³	Unchanged in kind; the active metabolites contribute proportionally	Neither is a bioactivation in the sense used on the paracetamol page — the metabolites are weaker than the parent. But they are not inert, and they extend the effect beyond the parent's clearance.
Clearance	Verapamil total clearance is nearly as high as hepatic blood flow, about 1 L/h/kg (range 0.7–1.3 L/h/kg)²	Falls with cardiac output	A flow-limited drug in a drug-induced low-output state. Clearance depends on how much blood the liver receives, and the poisoning reduces it. This is a genuine self-reinforcing loop and it is why verapamil poisoning can deteriorate over hours without any further absorption. Inferred
Half-life	Verapamil 3–7 h oral²; diltiazem about 4 h, extended from prolonged-release formulations³; amlodipine long, and further prolonged in hepatic impairment⁴	Prolonged; and for MR preparations the absorption phase, not the half-life, sets the duration	Quoting a half-life for a modified-release overdose is close to meaningless. The drug is still going in.
Renal impairment	No effect on verapamil pharmacokinetics — confirmed in end-stage renal failure against healthy kidneys²	Unchanged	Which removes the one argument that rescued dialysis for atenolol and sotalol. Renal impairment does not make a calcium-channel blocker accumulate, so it does not create a dialysis indication either.
Dialysability	—	No. EXTRIP recommends against ECTR in severe amlodipine, diltiazem or verapamil poisoning — a strong recommendation on very low quality evidence⁵	The labels agree independently: verapamil hydrochloride cannot be removed by haemodialysis¹², and verapamil and norverapamil are not significantly removed by hemodialysis². Amlodipine's label reaches the same conclusion from protein binding.⁴ Three separate sources, one arithmetic.

Metabolism and the metabolites

There is no bioactivation here. Every metabolite discussed below is less active than the drug that produced it — which makes this section a study in how a poisoning can be prolonged without ever being intensified by metabolism.

Verapamil — extensive metabolism that does not detoxify quickly enough to matter

Verapamil (racemic; R- and S-enantiomers)Already the toxic species. >90% absorbed from the small intestine²
Hepatic first pass — CYP3A4, CYP1A2, CYP2C8, CYP2C9, CYP2C18Removes about two-thirds of a single SR dose; saturable, and flow-limited²
Norverapamil10–20% of verapamil's activity, but steady-state concentrations similar to verapamil's² — so it is a real contributor
Eleven other metabolitesMost in trace amounts; none with appreciable activity²
Excreted drug and metabolitesAbout 50% within 24 h, 70% within five days — only 3–4% of the renally excreted drug as unchanged drug²

The practical consequence is that hepatic impairment and CYP3A4 inhibition are routes into calcium-channel blocker toxicity without any overdose at all. Verapamil's label records that its half-life is prolonged in impaired liver function through lower oral clearance and a higher volume of distribution², and diltiazem's that concentrations are higher in the elderly and in renal and hepatic insufficiency.³ Clarithromycin, itraconazole, grapefruit juice and ritonavir-boosted regimens all act at the same enzyme.

Elimination and accumulation

Three separate mechanisms extend a calcium-channel blocker poisoning beyond the point at which an equivalent immediate-release ingestion would have resolved, and they compound one another.

Continued absorption. Modified-release release and intestinal absorption may take more than 48 hours, and incompletely dissolved tablet material forms depots along the gut.² Reduced gut motility in a shocked patient prolongs contact time further.
Falling clearance. Verapamil's total clearance approaches hepatic blood flow², so the fall in cardiac output produced by the poisoning reduces the rate at which the poison is removed. Inferred
Saturated first pass. Diltiazem's presystemic metabolism is explicitly saturable, giving non-linear accumulation.³ Each additional unit of dose contributes more systemic exposure than the one before.

Amlodipine adds a fourth and quite different late problem. Its label records that non-cardiogenic pulmonary oedema has rarely been reported as a consequence of amlodipine overdose, that it may manifest with a delayed onset at 24 to 48 hours post-ingestion, and that it may require ventilatory support — adding, importantly, that early resuscitative measures (including fluid overload) to maintain perfusion and cardiac output may be precipitating factors.⁴ The instinctive treatment for vasodilatory shock is a plausible cause of the late complication.

Target organs — and why those

Sinoatrial and atrioventricular nodes

TargetL-type calcium channels providing the depolarising upstroke

Why hereNodal cells depolarise using a calcium current where working myocardium uses a sodium current. Blocking L-type channels therefore removes the node's action potential upstroke rather than merely slowing it. This is why conduction block dominates the non-dihydropyridine picture and why it is so resistant to atropine, which acts on vagal tone rather than on the current. Established

At the bedsideFirst- and second-degree block, frequently Wenckebach with or without escape rhythms, complete block with AV dissociation, sinus arrest and asystole — all listed in both verapamil labels.¹²

Ventricular myocardium

TargetL-type channels initiating calcium-induced calcium release

Why hereContraction depends on trigger calcium entering through the channel to release the much larger sarcoplasmic reticulum store. Blocking the trigger reduces the release, so the negative inotropy is direct and not compensable by sympathetic drive that has nowhere to act. Established

At the bedsideCardiogenic shock, and the signs of myocardial insufficiency for which the labels suggest inotropes and repeated calcium.¹²

Vascular smooth muscle

TargetL-type channels controlling myogenic tone

Why hereThe dihydropyridines were designed for this tissue, and their overdose is the exaggeration of their therapeutic effect: excessive peripheral vasodilatation and possibly reflex tachycardia.⁴ The non-dihydropyridines reach it too at overdose concentrations, producing a mixed cardiogenic and distributive shock. Established

At the bedsideProfound vasoplegia. The reflex tachycardia of dihydropyridine poisoning is a genuine discriminator from the bradycardia of verapamil or diltiazem — until the dose is large enough to abolish it.

Pancreatic beta cell

TargetL-type channels triggering insulin granule exocytosis

Why hereThe final step of glucose-stimulated insulin secretion is calcium influx through this channel. Block it and insulin release fails regardless of the glucose concentration, which is why the patient is hyperglycaemic rather than appropriately hyperinsulinaemic. The myocardium, which shifts towards carbohydrate utilisation under stress, is denied its substrate at the moment of greatest need. Inferred

At the bedsideHyperglycaemia, listed in both non-dihydropyridine labels¹²³. It is the mechanistic justification for high-dose insulin, and it is the cheapest discriminator from beta-blocker poisoning.

Lung

TargetNot explained by the receptor — and no label proposes a mechanism

Why hereIncluded because it is the one organ injury on this page that is not explained by the receptor. Both verapamil labels list acute respiratory distress syndrome among overdose findings¹², and amlodipine's describes delayed non-cardiogenic pulmonary oedema, noting that early fluid loading may be a precipitating factor.⁴ The mechanism is not established, and this page does not supply one: none of the three labels proposes a mechanism, so the association is reported and the explanation is simply absent. That is stated here rather than filled in.

At the bedsideDelayed onset at 24–48 hours for amlodipine, potentially requiring ventilatory support.⁴ It is a reason not to treat vasoplegia with volume alone.

Timeline of effects

Modified-release calcium-channel blocker overdose — the formulation writes the timeline

Time

What you seeWhat is happening

0–4 hThe reassuring window
What you seeFrequently well. Normal blood pressure, normal conscious level. This is the phase that gets people discharged.
What is happeningModified-release matrix has barely begun to release. A meaningful fraction of the dose is still inside intact tablet material. Nothing has been absorbed, so nothing has happened yet — and nothing about that is reassuring.
4–12 hOnset
What you seeHypotension; bradycardia and AV block with verapamil or diltiazem, tachycardia with amlodipine. Hyperglycaemia may already be present.
What is happeningSR verapamil peaks at 4–5 h even at therapeutic dose²; amlodipine peaks at 6–12 h⁴. Beta-cell channel blockade is established, so insulin release is already impaired.
12–48 hThe absorption plateau
What you seeShock that does not resolve, and may deepen despite treatment.
What is happeningRelease and absorption may continue for more than 48 hours, with undissolved tablet lumps acting as depots.² Simultaneously, falling cardiac output reduces the hepatic blood flow on which verapamil's clearance depends.² Input continues while clearance falls.
24–48 hAmlodipine's late complication
What you seeNon-cardiogenic pulmonary oedema, potentially needing ventilatory support.⁴
What is happeningRare, delayed, and the label names early fluid overload during resuscitation as a possible precipitant.⁴ A late deterioration in oxygenation in a patient who was being volume-resuscitated is not necessarily a new diagnosis.
DaysResolution
What you seeSlow recovery as the depot is finally exhausted.
What is happeningOnly once absorption ends does the half-life become the governing variable — 3–7 h for verapamil², longer for amlodipine⁴.

What the mechanism predicts at the bedside

Check the glucose, and read it as a marker of channel blockade. A rising glucose in a shocked patient on a calcium-channel blocker is not incidental hyperglycaemia of critical illness; it is the beta-cell channel being blocked.¹²³
A high glucose points away from beta-blocker and towards calcium-channel blocker; a low glucose points the other way. Neither is diagnostic, and co-ingestion is common.
A tachycardic hypotensive patient may still have taken a calcium-channel blocker — that is the expected dihydropyridine presentation.⁴ Absence of bradycardia excludes nothing.
Ask what formulation it was before deciding anything about timing. Modified-release absorption may exceed 48 hours.² An observation period designed for immediate-release drugs does not transfer.
Atropine is predictably disappointing in nodal block. The node's problem is a missing calcium current, not excess vagal tone.
Renal impairment does not create a dialysis indication here, unlike atenolol and sotalol — verapamil's kinetics are unchanged in end-stage renal failure.²
Volume resuscitation for vasoplegia is not free. Amlodipine's label names early fluid overload as a possible precipitant of the delayed non-cardiogenic pulmonary oedema.⁴
A patient improving on high-dose insulin still needs glucose and potassium watched for a long time. Glucose supplementation is likely to be required throughout therapy and for up to 24 hours after it stops, and the potassium fall represents a shift into cells rather than a loss of total body stores.⁷

The antidote, from the poison's side

Four treatments, three mechanisms, and no antidote in the strict sense — nothing here removes the drug or displaces it from the channel.

Calcium: The verapamil labels call calcium the specific antidote.¹² It is not, in the sense that Fab is for digoxin. It raises the extracellular calcium concentration and therefore the electrochemical gradient across every channel that is not blocked, forcing more current through the ones still available. It is a mass-action workaround, and it is why the effect is real but often transient and why repeated dosing is described.
High-dose insulin: Replaces a hormone the poison has stopped the pancreas from releasing, and does so at doses that also provide inotropy. Three proposed mechanisms: increased inotropy, increased intracellular glucose transport, and vascular dilatation.⁷ It is the only treatment on this list that addresses the metabolic lesion rather than the electrical one.
Catecholamines: Noradrenaline for vasoplegia, dobutamine or adrenaline where cardiogenic shock predominates.⁶ They act upstream of the block — raising cyclic AMP to phosphorylate channels that are antagonised — so their effect is limited by how many channels remain available.
Lipid emulsion: Recommended by the consensus panel for refractory shock, periarrest and cardiac arrest.⁶ The usual explanation is a lipid sink sequestering a lipophilic drug, which is plausible for these agents and is not established.
Glucagon: Raises cyclic AMP by an unblocked receptor. In the six animal calcium-channel-blocker studies Bailey identified, it appeared to increase heart rate and cardiac output and reverse second and third degree AV blocks, all at least transiently, with no effect on survival.⁸

What the expert consensus actually says

St-Onge and colleagues, for an international expert panel, recommend as first-line therapies intravenous calcium (1D), high-dose insulin therapy (1D–2D), and noradrenaline and/or adrenaline (1D); they suggest dobutamine or adrenaline where cardiogenic shock is present (2D) and atropine for symptomatic bradycardia or conduction disturbance (2D). For refractory patients they suggest incremental high-dose insulin if myocardial dysfunction is present (2D), lipid emulsion (2D), and pacing for unstable bradycardia or high-grade block without significant inotropic impairment (2D). For refractory shock or periarrest they recommend incremental high-dose insulin (1D) and lipid emulsion (1D), and suggest venoarterial ECMO where shock has a significant cardiogenic component (2D). In cardiac arrest they recommend calcium in addition to standard ALS (1D) and lipid emulsion (1D).⁶

Critical appraisal

The pulmonary oedema mechanism is not established. ARDS appears in the verapamil labels¹² and delayed non-cardiogenic pulmonary oedema in amlodipine's⁴, and the latter names fluid overload as a possible precipitant — but may be precipitating factors is the label's own hedge, and that is a precipitant rather than a mechanism. No evidence-tier badge is applied here, because there is no published account of this phenomenon for a badge to grade: the association is repeatedly reported and no mechanism has been proposed to doubt. An absent explanation is not the same as a contested one, and this page does not manufacture one in either direction.
Glucagon's calcium-channel-blocker evidence is animal only, and showed no survival effect. Bailey identified six animal studies, found transient improvements in heart rate, cardiac output and AV block, no consistent effect on mean arterial pressure and no effect on survival rate, in unblinded studies with small numbers.⁸
High-dose insulin has no controlled human trial. The supporting review is explicit: animal models showed superiority over calcium salts, glucagon, adrenaline and vasopressin, and human evidence is case reports and series.⁷ It is a first-line consensus recommendation resting on that.
Every consensus recommendation is graded D.⁶ The distinction between a strong recommendation (1) and very low quality evidence (D) is the most important thing on that page and the thing most often dropped when it is cited.
Lipid emulsion's mechanism is assumed. The lipid-sink account is plausible for drugs of this lipophilicity and is not demonstrated for these agents in humans; the consensus recommends it at the refractory and arrest end, where the alternative is nothing.
The verapamil self-limiting-clearance argument is inference. That total clearance approaches hepatic blood flow is measured and on the label²; that this produces a clinically important autocatalytic spiral in overdose is a reasonable deduction from flow-limited kinetics, not something this page can cite a study for.
The dihydropyridine/non-dihydropyridine split is a good rule that fails at the top of the dose range. Selectivity is a therapeutic-dose property. A very large amlodipine ingestion produces cardiac depression, and treating a hypotensive tachycardic patient as though the myocardium must be spared is a mistake the framework can encourage.
Nothing here is a decontamination recommendation. The verapamil label describes gastric lavage beyond 12 hours, whole-bowel measures and endoscopic removal for slow-release poisoning²; those are decisions for TOXBASE and NPIS, and they are quoted here only to show how far the absorption phase extends.

References

1
Verapamil 2.5 mg/mL Solution for Injection — Summary of Product Characteristics. electronic medicines compendium, product 979. Section 4.9 (Overdose). medicines.org.uk/emc/product/979
2
Securon SR 240 mg modified-release tablets (verapamil) — Summary of Product Characteristics. electronic medicines compendium, product 6243. Sections 4.9 and 5.2. medicines.org.uk/emc/product/6243
3
ADIZEM-XL 180 mg capsules (diltiazem) — Summary of Product Characteristics. electronic medicines compendium, product 7588. Sections 4.9 and 5.2. medicines.org.uk/emc/product/7588
4
Amlodipine 10 mg Tablets — Summary of Product Characteristics. electronic medicines compendium, product 2746. Sections 4.9 and 5.2. medicines.org.uk/emc/product/2746
5
EXTRIP Workgroup. Calcium channel blockers — recommendations. Extracorporeal Treatments in Poisoning Workgroup. extrip-workgroup.org/calcium-channel-blockers
6
St-Onge M, Anseeuw K, Cantrell FL, et al. Experts Consensus Recommendations for the Management of Calcium Channel Blocker Poisoning in Adults. Critical Care Medicine 2017 Mar;45(3):e306–e315. PMID 27749343.
7
Engebretsen KM, Kaczmarek KM, Morgan J, Holger JS. High-dose insulin therapy in beta-blocker and calcium channel-blocker poisoning. Clinical Toxicology 2011 Apr;49(4):277–83. PMID 21563902.
8
Bailey B. Glucagon in beta-blocker and calcium channel blocker overdoses: a systematic review. Journal of Toxicology — Clinical Toxicology 2003;41(5):595–602. PMID 14514004.

Calcium-channel blockers

At a glance