Why this drug is interesting

Vasopressin is the vasopressor that does not use an adrenoceptor. It constricts vascular smooth muscle through the V1a receptor, which matters because the α₁ receptors that noradrenaline and metaraminol depend on become progressively less responsive in severe acidosis and prolonged shock. A drug with a different receptor is a drug that keeps working when the others have stopped.

That is an elegant argument, and it has repeatedly outrun the evidence. Vasopressin has been introduced into, and then removed from, the cardiac arrest algorithm; it has been trialled in septic shock with results that are best described as neutral; and in the UK its product licence covers neither use.

What the UK licence actually covers

Licensed dosing

Diabetes insipidus (non-nephrogenic): 5–20 units (0.25–1 mL) subcutaneously or intramuscularly every four hours
Oesophageal varices: For initial control, give intravenously — 20 units diluted in 100 mL of 5% dextrose infused over 15 minutes
Children: Not recommended under 18 years
Elderly: As for adults; no age-specific data

In practice, variceal bleeding in the UK is managed with terlipressin, a vasopressin analogue with a longer half-life and greater V1 selectivity, rather than with vasopressin itself. Vasopressin's appearance in that licensed indication is largely historical.

Septic shock — the off-label use that matters

The rationale is catecholamine sparing. Endogenous vasopressin is depleted in prolonged septic shock, and replacing it at physiological rather than pharmacological doses restores vascular tone through a receptor that remains responsive. The practical aim is to reduce the noradrenaline requirement, on the assumption that high catecholamine doses are themselves harmful.

What the trials showed

VASST randomised patients already receiving at least 5 µg/min of noradrenaline to blinded low-dose vasopressin (0.01–0.03 units/min) or blinded additional noradrenaline, and found no significant difference in 28-day mortality (35.4% versus 39.3%, P=0.26).² In the prospectively defined stratum of less severe septic shock, mortality was lower with vasopressin (26.5% versus 35.7%, P=0.05) — prospectively defined rather than post hoc, but a stratum result at a borderline p-value.
VANISH compared early vasopressin against noradrenaline in septic shock and found no improvement in kidney failure-free days, its primary outcome, and no significant mortality difference.³ Note that VANISH titrated vasopressin up to 0.06 units/min — double the fixed rate described above — so the trial evidence does not straightforwardly validate the fixed-rate convention.
The consistent finding across trials is reduced catecholamine requirement without a demonstrated mortality benefit. Whether catecholamine sparing is itself worth having remains an assumption rather than a result.

Why it left the cardiac arrest algorithm

Vasopressin was for some years offered as an alternative to adrenaline in cardiac arrest, on the reasoning that the arrested, profoundly acidotic myocardium is exactly the setting in which α receptors fail. Trials did not support it, and current UK resuscitation guidance does not recommend it — adrenaline remains the vasopressor in cardiac arrest, with the timing determined by rhythm.⁴ See adrenaline.

Pharmacology

Mechanism

V1a receptors: Vascular smooth muscle — vasoconstriction, independent of adrenoceptors and relatively preserved in acidosis
V2 receptors: Renal collecting duct — aquaporin insertion, water reabsorption; the antidiuretic effect that gives the hormone its other name
V1b receptors: Anterior pituitary — ACTH release

The V2 effect is the reason vasopressin causes water retention and hyponatraemia, and the reason the SmPC warns about water intoxication. It is also why the vasopressor use is deliberately kept at a low fixed rate — the doses needed for a large pressor effect carry substantial V2 and ischaemic consequences.

Contraindications, cautions and adverse effects

Contraindications

Hypersensitivity or anaphylaxis to vasopressin or excipients
Vascular disease, especially coronary artery disease
Patients intended to receive halogenated anaesthetic agents
Chronic nephritis with nitrogen retention, until reasonable blood nitrogen levels are attained

Cautions

Peripheral vascular disease — observe the skin carefully for signs of ischaemia
Epilepsy, migraine, asthma, heart failure, or any state in which rapid addition to extracellular water is hazardous
Monitor blood urea nitrogen in chronic nephritis

Adverse effects

Digital, cutaneous and mesenteric ischaemia — the effects that limit the dose in practice; look at the fingers and toes daily
Myocardial ischaemia and infarction
Hyponatraemia and water intoxication from the V2 effect
Abdominal cramps, nausea, vomiting, pallor
Bradycardia and reduced cardiac output at higher rates

Interactions

Carbamazepine, chlorpropamide, clofibrate, fludrocortisone, urea and tricyclic antidepressants potentiate the antidiuretic effect. Demeclocycline, noradrenaline, lithium, heparin and alcohol reduce it. Ganglion-blocking agents markedly increase sensitivity to the pressor effect.

Critical appraisal

The mechanism is better than the evidence, and this drug is a case study in the difference. Receptor-level reasoning predicted that vasopressin would work where catecholamines fail. Two large trials in septic shock found no mortality benefit, and the arrest indication was withdrawn. Mechanistic plausibility is a reason to test a drug, not a reason to believe in it.
Catecholamine sparing is an intermediate outcome dressed as a benefit. Every trial shows vasopressin reduces the noradrenaline requirement, because that is close to arithmetic. Whether reducing it improves anything for the patient has not been demonstrated.
The fixed-rate convention is unusual and poorly explained to trainees. Nearly every other infusion in the resuscitation room is titrated to effect. A drug that is deliberately not titrated invites well-meaning escalation, and the harm from that escalation is ischaemic and irreversible.
Its UK licence bears almost no relationship to its use. A drug licensed for diabetes insipidus and variceal bleeding, used almost exclusively as a second-line vasopressor in critical care, is one where the SmPC will not answer the question you have. Local critical care protocols do the real work here.

References

1
Argipressin 20 IU/ml Solution for Injection — Summary of Product Characteristics, ADVANZ Pharma. electronic Medicines Compendium. Sections 4.1–4.5, 4.8. Verified 24 Aug 2026. Compare Argipressin AOP, emc.
2
Russell JA, Walley KR, Singer J, et al. Vasopressin versus norepinephrine infusion in patients with septic shock (VASST). N Engl J Med 2008;358(9):877–87. PubMed
3
Gordon AC, Mason AJ, Thirunavukkarasu N, et al. Effect of early vasopressin vs norepinephrine on kidney failure in patients with septic shock (VANISH). JAMA 2016;316(5):509–18. PubMed
4
Resuscitation Council UK. Adult advanced life support, 2025 Resuscitation Guidelines. resus.org.uk — local copies at lifesupport.resusdoc.uk
5
Association of Anaesthetists. Quick Reference Handbook, 3-10 Local anaesthetic toxicity, version 2, June 2023. anaesthetists.org/qrh
6
Evans L, Rhodes A, Alhazzani W, et al. Surviving Sepsis Campaign: international guidelines for management of sepsis and septic shock 2021. Crit Care Med 2021;49(11):e1063–143. PubMed

Vasopressin

At a glance