Why this poison is interesting

The interesting thing about beta-blocker overdose is how little of it is caused by beta-blockade. Competitive antagonism at a G-protein-coupled receptor is, on its own, a rather forgiving poisoning: the effect is surmountable, it plateaus, and a patient with an intact myocardium tolerates a great deal of it. What kills people in beta-blocker overdose is usually a second pharmacology that the drug happens to carry — and which of the two second pharmacologies a given molecule carries is not predictable from the class name.

The UK labels state this outright without ever drawing the conclusion. Propranolol's label lists QRS complex prolongation, ventricular tachycardia, seizures, hallucinations, dilated pupils and coma.¹ Sotalol's lists QT prolongation, premature ventricular complexes, ventricular tachycardia and torsades de pointes.³ Atenolol's lists bradycardia, hypotension, acute cardiac insufficiency and bronchospasm — and nothing else.² Those are three different clinical problems requiring three different interventions, and they are all filed under one heading.

The second reason this page exists is that the EXTRIP workgroup split its recommendation by molecule rather than issuing one for the class — recommending against extracorporeal treatment for propranolol while suggesting it for atenolol and sotalol in the presence of renal impairment.⁴ That split is not a committee compromise. It follows closely from protein binding and clearance route, both of which three summaries of product characteristics supply. Those labels do not give a volume of distribution, which is the other determinant of extracorporeal removal and the one that matters most for propranolol — so the labels make the split intelligible rather than strictly derivable.

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

The toxic principle

The shared mechanism is straightforward and, by itself, survivable. Beta-1 adrenoceptors on cardiac myocytes are Gs-coupled: agonist binding activates adenylate cyclase, cyclic AMP rises, protein kinase A phosphorylates L-type calcium channels and phospholamban, and calcium entry and sarcoplasmic reticulum calcium handling both increase. Blockade removes the tonic sympathetic contribution to all of it.

The pathway all beta-blockers interrupt — and the two points at which the antidotes re-enter it

Noradrenaline at the beta-1 adrenoceptorGs-coupled
Cyclic AMPThis is the step the blockade removes
L-type calcium channel phosphorylation
Increased calcium entryInotropy and chronotropy — lost in overdose
Glucagon receptor — a different Gs-coupled receptorCyclic AMP, by a route the blockade does not occupyThe entire rationale for glucagon; the evidence is animal only⁵

Now the two second mechanisms, which are where the mortality lives.

Propranolol — sodium-channel blockade: Propranolol carries what older texts call membrane-stabilising activity: at high concentrations it blocks the fast cardiac sodium channel, exactly as a tricyclic or flecainide does. The label's listing of QRS complex prolongation alongside ventricular tachycardia, and its listing of seizures, hallucinations, dilated pupils and coma, is the clinical signature of that second target plus high lipid solubility.¹ The channel account is a reasonable reconstruction from the label's findings and from propranolol's known membrane activity rather than something demonstrated in poisoned humans. Inferred
Sotalol — potassium-channel blockade: Sotalol is a class III antiarrhythmic that happens also to be a beta-blocker. It blocks the rapid delayed-rectifier potassium current, prolonging repolarisation — which is its therapeutic mechanism, and in overdose becomes QT prolongation, early afterdepolarisations and torsades de pointes. Its label lists QT prolongation, premature ventricular complexes, ventricular tachycardia and torsades among the findings of massive overdose.³ Established
Atenolol — neither: Hydrophilic, poorly tissue-penetrating, no significant sodium- or potassium-channel action. The label's overdose section runs to bradycardia, hypotension, acute cardiac insufficiency and bronchospasm.² Atenolol is the control condition that proves the point about the other two. Established

Toxicokinetics

The three molecules are set side by side deliberately. Every difference in the table below is taken from the relevant label, and between them they predict the EXTRIP recommendation without reference to it.

Propranolol vs atenolol vs sotalol — one receptor, three kinetic profiles
Parameter	Therapeutic	In overdose	Why it changes
Absorption	Propranolol completely absorbed, peak 1–2 h fasting¹; atenolol absorption consistent but incomplete, about 40–50%, peak 2–4 h²; sotalol bioavailability >90%, peak 2.5–4 h³	Slowed by the bradycardia and hypoperfusion the drug itself causes; modified-release preparations extend it further	Sotalol's near-complete and highly reproducible absorption — the label notes very little inter-subject variability — means the ingested dose translates into concentration more predictably than for the others.
First-pass metabolism	Propranolol: the liver removes up to 90% of an oral dose¹. Atenolol: no significant hepatic metabolism². Sotalol: not metabolised³	Propranolol's first-pass capacity can be exceeded, so bioavailability rises disproportionately with dose	This is the closest beta-blockers come to a classical saturation pattern. A high-extraction drug taken in massive overdose delivers a far greater fraction of the dose to the systemic circulation than the same drug at therapeutic dose — the exposure rises faster than the ingestion. Inferred
Protein binding	Propranolol 80–95%¹ · atenolol approximately 3%² · sotalol does not bind to plasma proteins³	Unchanged	The line that most nearly predicts the EXTRIP split. Extracorporeal removal clears only the free fraction; a drug that is 95% bound offers the dialyser almost nothing, while one that is unbound offers it everything. Binding alone is not sufficient — volume of distribution is the other determinant, as the digoxin page shows, and these labels do not give one.
Tissue penetration	Propranolol widely distributed, high concentrations in brain and heart¹; atenolol penetrates tissues poorly, brain concentration low²; sotalol CSF only 10% of plasma³	Unchanged	Determines whether the poisoning is cardiac only or cardiac plus neurological — and therefore whether seizures are part of the presentation.
Elimination route	Propranolol hepatic¹; atenolol renal — the label gives no excreted fraction, stating only that the kidney is the major route and that there is no significant hepatic metabolism²; sotalol renal, 80–90% excreted unchanged³	Renal clearance falls with the hypotension the poisoning causes — a self-reinforcing loop for the two renally cleared drugs	A renally cleared drug in a shocked patient accumulates. This is why the EXTRIP suggestion for atenolol and sotalol is conditioned on kidney impairment, and why a normal creatinine early does not settle the question.
Half-life	Propranolol 3–6 h oral (about 2 h intravenous)¹; atenolol about 6 h, rising in severe renal impairment²; sotalol 10–20 h³	All prolonged in shock; sotalol's is the longest to begin with	Sotalol's combination of a long half-life and a QT effect means the arrhythmic risk period is measured in days, not hours — the observation period is set by repolarisation, not by the beta-blockade.
Glucose	Beta-2 blockade impairs glycogenolysis and gluconeogenesis and masks adrenergic warning symptoms	Hypoglycaemia, listed among the expected signs in sotalol overdose³	The direction of the glucose is a bedside discriminator. Beta-blockers tend to lower it; calcium-channel blockers raise it, because they block insulin release from the pancreatic beta cell. Two cardiodepressant poisonings that look alike on the monitor separate on the glucose. Inferred
Dialysability	—	Molecule-dependent. EXTRIP: recommends against ECTR in severe propranolol poisoning; suggests ECTR for atenolol and for sotalol with kidney impairment and refractory bradycardia and hypotension (or recurrent torsades for sotalol); no recommendation either way for either drug with normal kidney function⁴. Every arm rests on very low quality evidence; the propranolol recommendation is graded strong, the two suggestions weak	Sotalol's label states plainly that haemodialysis results in a large reduction of plasma levels.³ Where all modalities are available EXTRIP recommends intermittent haemodialysis, and cessation based on clinical improvement — notably not on the QT interval, for which it suggests against dialysing on that basis alone.⁴

Metabolism and the metabolites

For two of the three drugs this section barely exists, which is itself the finding. Atenolol undergoes no significant hepatic metabolism and sotalol is not metabolised at all.²³ Neither has an active metabolite to worry about, and neither is subject to the enzyme-inhibition interactions that complicate so much of this library.

Propranolol is the exception in every respect. It is extensively metabolised on first pass, and the label records a specific consequence: after intravenous administration the ratio of metabolites to parent drug is lower, and 4-hydroxypropranolol is not present at all.¹ That metabolite is pharmacologically active, so the same plasma concentration of propranolol represents a different total beta-blocking activity depending on the route by which it arrived.

Elimination and accumulation

The clinically important asymmetry is that the two drugs which can be dialysed are the two which are cleared by the organ most likely to fail during the poisoning. Atenolol and sotalol are renally eliminated; the shock they produce reduces renal perfusion; the drug accumulates; the shock worsens. EXTRIP's conditioning of its suggestion on kidney impairment is describing the point in that loop at which removal becomes worth the line.

Propranolol accumulates differently and less dangerously. Hepatic clearance falls with hepatic blood flow, so a shocked patient clears it more slowly, but there is no reservoir comparable to digoxin's and no renal dependence. Its problem is not accumulation but distribution — into the brain.

Target organs — and why those

Sinoatrial and atrioventricular nodes

TargetBeta-1 adrenoceptors on nodal tissue

Why hereNodal automaticity and conduction are more dependent on cyclic-AMP-driven calcium currents than ventricular contraction is, so the nodes are the first tissue to show the blockade. This is a class effect and it is the one finding common to all three labels. Established

At the bedsideBradycardia and AV block in every beta-blocker overdose regardless of molecule.¹²³ Atropine is the first move and, in a substantial overdose, rarely sufficient — it removes vagal tone but cannot restore cyclic AMP.

Ventricular myocardium — sodium channels

TargetFast cardiac sodium channel, blocked by propranolol at high concentration

Why hereThe same target as the tricyclics and class I antiarrhythmics. Slowed phase 0 depolarisation widens the QRS and sets up re-entry. The label lists QRS prolongation, ventricular tachycardia and ventricular fibrillation among propranolol's overdose findings and does not list them for atenolol.¹² Inferred

At the bedsideA wide QRS in beta-blocker overdose points to propranolol (or a co-ingestant) and is the finding that makes sodium bicarbonate relevant — treating the channel, not the receptor.

Ventricular myocardium — potassium channels

TargetRapid delayed-rectifier potassium current, blocked by sotalol

Why hereSotalol's class III action prolongs repolarisation by design. In overdose the prolongation becomes pathological, permitting early afterdepolarisations and torsades. This is not an off-target effect discovered in poisoning — it is the therapeutic mechanism at an untherapeutic concentration. Established

At the bedsideQT prolongation, PVCs and torsades in massive sotalol overdose.³ The label's suggested management of torsades — DC cardioversion, pacing, adrenaline, magnesium — is torsades management, not beta-blocker management.

Brain

TargetNot a specific receptor — a consequence of lipid solubility

Why herePropranolol reaches high concentrations in brain tissue¹; atenolol's brain concentration is low² and sotalol's CSF concentration is 10% of plasma³. The organ is targeted by physical chemistry rather than by pharmacology. Sodium-channel blockade in cortical tissue is the likeliest proximate cause of the seizures. Inferred

At the bedsideDrowsiness, confusion, seizures, hallucinations, dilated pupils and coma — all listed for propranolol and none for atenolol.¹² Seizure in a beta-blocker overdose is a strong pointer to which drug it was.

Bronchial smooth muscle

TargetBeta-2 adrenoceptors

Why hereBeta-2 blockade removes the bronchodilator tone that sympathetic activity supplies. Even so-called cardioselective agents lose their selectivity at overdose concentrations, which is why bronchospasm appears in all three labels including atenolol's. Selectivity is a dose-dependent property, and overdose is the dose at which it fails. Established

At the bedsideBronchospasm listed in all three labels.¹²³ A patient with asthma is at risk from a drug that was cardioselective at the prescribed dose.

Liver and pancreas — glucose handling

TargetBeta-2-mediated glycogenolysis and gluconeogenesis

Why hereBlocking beta-2 receptors impairs the counter-regulatory response to a falling glucose, and simultaneously masks the adrenergic symptoms that would warn of it. Hypoglycaemia is listed among the common expected signs in sotalol overdose.³ Inferred

At the bedsideCheck the glucose, and note its direction. Low points towards beta-blockade; high points towards calcium-channel blockade. It is one of the few cheap tests that discriminates between two poisonings that present identically.

Timeline of effects

Beta-blocker overdose — one shared early course, then three divergent ones

Time

What you seeWhat is happening

0–2 hAbsorption
What you seeOften asymptomatic. Immediate-release preparations are largely absorbed by two hours.
What is happeningPropranolol peaks at 1–2 h fasting¹, atenolol 2–4 h², sotalol 2.5–4 h³. Receptor occupancy is rising; nothing has failed yet.
1–6 hThe shared syndrome
What you seeBradycardia and hypotension in all three. Bronchospasm possible in all three.
What is happeningBeta-1 blockade at the nodes and in ventricular myocardium; falling cyclic AMP, falling intracellular calcium. This phase does not tell you which drug it was.
2–12 hDivergence — propranolol
What you seeQRS widening, ventricular arrhythmias, seizures, hallucinations, dilated pupils, coma.¹
What is happeningSodium-channel blockade plus high brain concentrations. The label warns that coma and unreactive pupils are unreliable prognostic indicators during resuscitation.¹
Hours to daysDivergence — sotalol
What you seeMay look haemodynamically settled while the risk continues. Torsades can occur late.
What is happeningHalf-life 10–20 h³ and a repolarisation mechanism that is independent of blood pressure. This is a genuine latent interval — the QT is prolonged and nothing has yet triggered the arrhythmia.
ProlongedDivergence — atenolol
What you seeBradycardia and hypotension, persisting; no neurological or repolarisation component.²
What is happeningRenal clearance falling with renal perfusion. This is the arm in which a dialyser genuinely helps — 3% protein binding, no significant hepatic metabolism, and the kidney as the major route of elimination.²⁴
Any timeModified-release preparations
What you seeLate deterioration in a patient who looked well.
What is happeningRelease continues long after the ingestion; the same trap described in detail on the calcium-channel blocker page, where verapamil's label warns absorption may take more than 48 hours.

What the mechanism predicts at the bedside

The first question is which molecule, not how many tablets. Propranolol, sotalol and atenolol produce three different poisonings from the same class, and the labels for each list a different set of findings.¹²³
A wide QRS means sodium-channel blockade, and points to propranolol (or to a co-ingested tricyclic or class I agent). It is treated as sodium-channel blockade — the target is the channel, not the adrenoceptor.
A long QT means sotalol and moves the problem into torsades management: magnesium, pacing, correcting potassium.³ The half-life means the risk outlasts the haemodynamic recovery.
A seizure means the drug got into the brain, which in this class means the lipophilic one.¹²³
Check the glucose and note which way it has moved. Low suggests beta-blockade; high suggests calcium-channel blockade.³ Two poisonings that look identical on the monitor separate on a bedside test.
Atropine will usually be insufficient. It removes vagal tone; it does not restore the cyclic AMP that the blockade has taken away. Atenolol's label moves straight from atropine to glucagon.²
Renal function decides whether a dialyser is worth considering, and only for atenolol and sotalol. EXTRIP makes no recommendation either way when renal function is normal — an explicit statement of uncertainty rather than a negative.⁴
Bronchospasm can occur even with a cardioselective agent, because selectivity is dose-dependent and overdose is the dose at which it fails.²
Modified-release preparations invalidate a reassuring early observation period. So does a co-ingestion that slows gastric emptying.

The antidote, from the poison's side

There is no true antidote to beta-blockade in the sense that Fab is an antidote to digoxin. Every agent used is an attempt to restore, by some other route, the intracellular signal the blockade has removed — or to treat a channel effect that has nothing to do with the receptor at all.

Atropine: Removes parasympathetic tone at the nodes. Does nothing about the missing cyclic AMP, which is why the labels position it as a first step rather than a solution.²³
Glucagon: Binds its own Gs-coupled receptor on the myocyte and raises cyclic AMP by a route the beta-blockade does not occupy — an elegant piece of pharmacological reasoning. Atenolol's label carries a glucagon regimen for beta-blocker overdose, quoted here only to show that the licensed labelling contemplates it² — the dose to use is TOXBASE's, not this page's. The evidence is discussed below and it is weaker than the elegance suggests.
High-dose insulin: Inotropy, glucose transport and vasodilatation, acting downstream of the receptor rather than at it.⁶ Requires simultaneous glucose, and the potassium fall it causes is a shift into cells rather than a loss of total body stores.⁶
Catecholamines: Compete for the occupied receptor — which works, but requires doses far above the usual, and raises systemic vascular resistance and myocardial oxygen demand in a heart that can afford neither.⁶
Sodium bicarbonate: For propranolol's QRS prolongation only. It treats the sodium channel, exactly as on the tricyclic page, and has no action on beta-adrenoceptor blockade.
Haemodialysis: For atenolol and sotalol with renal impairment, per EXTRIP; recommended against for propranolol.⁴ Intermittent haemodialysis is preferred where all modalities are available.⁴

This is worth stating carefully, because it is easy to over-read. Glucagon's mechanism is sound and it is on the UK label.² What is not established is that it changes any outcome a patient would notice — the animal data suggest a transient heart-rate effect without a blood-pressure effect, and heart rate is not the problem in a shocked patient. A treatment can be mechanistically correct, licensed, and still unproven. Glucagon is taught as the antidote for beta-blocker overdose with a confidence the underlying literature does not support, and that gap is the reason it carries a badge on this page.

Critical appraisal

Glucagon carries a traditional-teaching badge on the basis of Bailey's review, not on the basis of an absent citation. The review actively looked for human controlled evidence and found none, and reported that the animal effect on mean arterial pressure was absent and on survival unclear.⁵ That is a citation for the doubt, which is the standard this library requires before downgrading anything.
Propranolol's sodium-channel mechanism is inferred, not demonstrated here. The label records QRS prolongation and ventricular arrhythmias in overdose¹; the attribution to fast sodium-channel blockade is the standard reading of membrane-stabilising activity and is consistent with the tricyclic literature, but no human channel study is cited on this page and none should be assumed.
The glucose direction is a useful discriminator, not a rule. Hypoglycaemia is listed for sotalol³ and hyperglycaemia for verapamil and diltiazem, but neither is universal, both are modified by co-ingestants and by diabetes, and a normal glucose excludes nothing.
EXTRIP's 'no recommendation' for normal renal function is a real finding and is often misquoted as a negative. The workgroup explicitly makes no recommendation for or against ECTR in atenolol or sotalol poisoning with normal kidney function.⁴ That is a statement that the evidence does not decide the question, which is not the same as a recommendation against.
EXTRIP suggests against dialysing on the basis of the QT interval alone in sotalol poisoning.⁴ This is easy to invert in the retelling, given that QT prolongation is the feared consequence.
The three-molecule framing is a simplification. Metoprolol, carvedilol, labetalol, nebivolol and bisoprolol each sit somewhere on the lipophilicity and channel-activity spectrum, and none of them is discussed here. The framework predicts where they should fall; it does not substitute for the individual label.
Nothing on this page establishes that treating the second mechanism improves survival. Bicarbonate for a wide QRS and magnesium for torsades are extrapolations from the tricyclic and torsades literatures respectively, applied here on mechanistic grounds.

References

1
Inderal 10 mg film-coated tablets (propranolol) — Summary of Product Characteristics. electronic medicines compendium, product 12858. Sections 4.9 (Overdose) and 5.2 (Pharmacokinetic properties). medicines.org.uk/emc/product/12858
2
Atenolol 100 mg film-coated tablets — Summary of Product Characteristics. electronic medicines compendium, product 14163. Sections 4.9 and 5.2. medicines.org.uk/emc/product/14163
3
Sotalol 40 mg Tablets — Summary of Product Characteristics. electronic medicines compendium, product 7039. Sections 4.9 and 5.2. medicines.org.uk/emc/product/7039
4
EXTRIP Workgroup. ß-adrenergic antagonists (ß-blockers) — recommendations. Extracorporeal Treatments in Poisoning Workgroup. extrip-workgroup.org/beta-adrenergic-antagonists
5
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.
6
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.

Beta-blockers

At a glance