Pharmacology of vasopressin and its analogues

This chapter is related to Section U2(vii) from the 2023 CICM Primary Syllabus, which expects the exam candidate to "understand the pharmacology of vasopressin and its analogues", and so is a dreary repetition of several vasopressin-related chapters from all over the CICM exam syllabus. To be fair, in the actual syllabus document vasopressin really only appears in the endocrinology section, but realistically it would be amiss to leave it out of the cardiovascular section, or to ignore it where the regulation of body water is discussed, or to gloss over it in the discussion of sepsis management. As the result, vasopressin and vasopressin-like drugs appear in several other pages, listed below for reader convenience:

Even though all this material exists somewhere in another form, it still seemed reasonable to bring it all together in this summary, considering especially it is an Official Syllabus Item. Having said this, First Part Exam endocrinology questions have never asked anything about the pharmacology or physiology of vasopressin,  but Question 22 from the second Fellowship Exam paper of 2009 somehow did, and it even included the specific phrase "vasopressin and its analogues". From this, it might appear that this syllabus item really belongs in the Part Two exam, considering especially that the CICM First Part already existed in 2009. 

  • Chemical structure: vasopressin and terlipressin are cyclic peptides; DDAVP is a nonapeptide with a deaminated cystein residue
  • Administration is mainly IV, though terlipressin can also be administered subcutaneously and and DDAVP can be used as a nasal spray or tablet
  • Absorption is minimal as these peptides are degraded by digestive enzymes
  • Distribution is largely confined to extracellular fluid, with VOD around 0.2L/kg for vasopressin and 0.3-0.6 L/kg for terlipressin and DDAVP
  • Protein binding is minimal, something like 50% for DDAVP and 30% for vasopressin
  • Metabolism and elimination is different for each analog:
    • Vasopressin is partly metabolised rapidly by hepatic sinusoidal endothelial peptidases, and 65% is excreted unchanged in the kidney, with a rather short half-life (17-35 minutes)
    • Terlipressin is slowly metabolised by endothelial peptidases and is thereby transformed into lysine vasopressin (the active daughter molecule), which means the duration of effect is up to 6 hours
    • DDAVP is immune to hepatic metabolism because of its molecular structure, and has a greatly increased half-life (perhaps 3 hours)
  • Receptor effects are mediate by V1 and V2 receptors
    • V1 receptors are Gq-coupled mediate vasoconstriction by increasing the intracellular availability of calcium through an IP3-mediated mechanism
    • V2 receptors are Gs-coupled and mediate aquaporin expression in the collecting duct by increasing intracellular cAMP
  • Pharmacodynamics vary for each drug because of different receptor selectivity and different effects that manifest at different doses:
    • Vasopressin is non-selective and causes both vasoconstrictions via V1 receptors and water conservation via V2 receptors
    • Terlipressin (i.e. lysine vasopressin) has six times greater selectivity for V1 receptors, and therefore has minimal antidiuretic effect when used in normal doses
    • DDAVP has 1500 times greater selectivity for V2 receptors, which means it has no cardiovascular effects at normal doses
    • The high V2 selectivity of DDAVP also makes it possible to use this drug in the large doses necessary to liberate vWF from vascular endothelium and to partially reverse antiplatelet drug effects, which would not be possible with vasopressin or terlipressin

Anyway: rather than to just dump all the vasopressin class content from the local CICM First Part Exam Pharmacopoeia, it felt more professional to discuss these peptides as a class, and to go through their pharmacology systematically, pointing out unusual features along the way. This has already been done by Glavaš et al (2022) so comprehensively that the amateur writer may despair, concluding nothing further could possibly be added. Wherever a reader finds an unsupported fact in the text below, it likely comes from Glavaš et al. For a more intense deep dive into vasopressin physiology the reader may also be directed to the two papers by Holmes et al (1992, Part 1 and Part 2).

Chemical structure of vasopressin and its analogues

8-arginine vasopressin is a synthetic analogue of an endogenous nonapeptide hormone where the eighth amino acid molecule is an arginine.  It is indistinguishable from the endogenous hormone which is secreted by the posterior pituitary of humans, and is only slightly different from porcine vasopressin, which contains a lysine in same position instead.

an idealised vasopressin molecule

It is a cyclical peptide: The two cysteines are joined by a disulfide bridge, and this curls the molecule into a donut.

a more realistic shape for the vasopressin molecule

Since its discovery in 1895, pituitary extract has been just that- extract of pituitary glands, fresh or dried. Specifically, the glands of oxen, it would seem. Interest in its clinical use (and, one must assume, the practical inconvenience of having to extract, dry and process industrial quantities of bovine pituitary glands) has ultimately led to the de novo synthesis of it by du Vigneaud in 1955 (he had already synthesized oxytocin by the same method, and received the Nobel Prize in Chemistry for it).

Combined with 0.5% chlorobutanol as a preservative, it comes in a 1ml ampoule of 20 "pressor units" per ml. 

What the hell is a "pressor unit", you ask?

Standards of vasopressin dose measurement: the Pressor Unit

The "pressor unit" is defined as a quantity of pressor agent which is equivalent in pressor activity to the activity of 0.5 milligram of the USP Standard Powdered Pituitary U.S.P., a USP posterior pituitary reference standard. Yes, there is a standard of pressor activity, analogous to the platinum rod used as the international standard of the metre, or to the unit standard of heparin (keeping the liquid state of refrigerated cat blood, etc.)

Like the metre, it is ancient, and like the rabbit seizure unit measurement of insulin it has a weird history. Dale and Laidlaw first described the standard in 1912, before the actions of the different posterior pituitary hormones were known. The "infundibular extracts" were tested on the uterus of the virgin guinea pig, which was found to contract with nicely predictable linear responsiveness to successively increased doses of the extract. However, subsequent modifications of the method resorted to using rat uterus, the authors complaining about the expense of obtaining guinea pigs in post-war Britain. As the difference between the two posterior pituitary hormones became apparent in coming years, and as synthetic analogues became available, the USP units became separated into the "oxytocic unit" (used to compare uterine activity) and the "vasopressor unit" (used to compare the effect on the blood pressure of the rat).

The USP standard was a homogenised extract of the posterior pituitary, weighing about 20mg, which was distributed to laboratories in the 1950s to standardise the purification of posterior pituitary hormones. The definition is used to this day, even though now there are better ways to measure vasopressin dose. Synthetic vasopressin is very pure and potent; 1 mg of pure vasopressin corresponds to about 600 pressor units.

Structure and function relationships of vasopressin analogues

Numerous variations of the basic molecular structure of vasopressin exist; or rather, vasopressin and its analogues in humans and other animals are all some variations on the same common theme which probably arose in unimaginably ancient times, likely along with such fundamental animal features as bilaterality (though it must be noted that even radially symmetrical animals like the starfish have neuropeptides structurally similar to vasopressin which do various vasopressin-like things). As most people would not have much interest in the reproductive hormones of molluscs, no further attention will be given to these substances here, considering especially that we have plenty of manmade vasopressin analogues to discuss. To tabulate a diagram from Glavaš et al (2022), their appearance in pharmacology can be chronicled as follows:

Vasopressin analogues by year of their discovery
Lypressin (porcine 8-lysine vasopressin) 1952
Argipressin (human 8-arginine vasopressin) 1953
Desmopressin (DDAVP) 1960
Felypressin 1961
Phenypressin 1962
Ornipressin 1964
Terlipressin 1965
Selepressin 2009

CICM trainees are unlikely to ever encounter the vast majority of these, and so it felt wastefully self-indulgent to examine them in any great detail. Terlipressin and desmopressin are far more likely to appear as exam questions, and in fact terlipressin has its own syllabus item in the gastroenterology section. 

Moreover, oxytocin (the other posterior pituitary hormone) is structurally very similar to vasopressin:

oxytocin molecule

The leucine at the 8th position here makes this molecule an 8-leucine oxytocin, identical to the synthetic oxytocin. Yes, there is crossreactivity. Oxytocin has a mild antidiuretic effect, and large quantities of vasopressin can stimulate uterine contractions. However, in contrast to vasopressin, a bolus of oxytocin causes hypotension. There is little agreement as to why: some have blamed the chlorobutanol preservative, others point to the release of atrial natriuretic peptide. Oxytocin receptors are also present in blood vessels, suggesting a direct role.

Terlipressin is a longer molecule:

terlipressin molecule

The lysine at the 8th position here makes this molecule an triglycine-lysine vasopressin. The glycines are enzymatically cleaved by endothelial peptidases, and the result of the breakdown is lysine-vasopressin, identical to pig vasopressin. The elimination half-life of terlipressin is thus about 50 minutes, but the effect lasts for 6 hours because of the liberation of this daughter molecule.

Desmopressin, or 1-desamino-8-d-arginine vasopressin (DDAVP), is a cysteine-deaminated version of arginine vasopressin:

desmopressin molecule

Whatever the deamination of cysteine contributes to the function of desmopressin, is an exercise left up to the reader to discover. It is thought that the substitution of a D-stereoisomer for arginine is a feature that protects the desmopressin from rapid metabolism by the various endothelial peptidases. This grants it a duration of action lasting 6 to 14 hours.

Pharmacokinetics of vasopressin and its analogues

Administration and absorption

For the majority of these substances, the IV route of administration is preferred, mainly because oral administration is completely out of the question. These defenceless peptides will be made into nutrient soup by the proteases and peptidases of the gastrointestinal tract. The only possible variation on this is desmopressin, which is available as a nasal spray, where about 10 times the normal IV dose needs to be administered for the equivalent effect. The nasal bioavailability is therefore said to be about 10-20%. Theoretically, this means terlipressin and vasopressin should also have this route available, but practically it appears that nobody has done this to achieve cardiovascular or endocrine endpoints (though people have squirted vasopressin up various noses for much weirder reasons). Oral and sublingual desmopressin tablets are available, but generate a tremendous feeling of lost value, as  more than 99% of the active ingredient is destined to be wasted. It is only possible to administer it in this way because of its extremely high potency and affinity for V2 receptors, which means it remains effective in spite of an extremely small systemic dose.

Solubility, distribution and protein binding

The pKa of vasopressin terlipressin and desmopressin (as well as the rest of these peptides) are such that these substances are sufficiently water-soluble to achieve what they need to achieve.  Remember that they are active in nanogram-per-litre concentrations, and so do not need to be available as a highly concentrated aqueous solution. They are slightly protein-bound to plasma proteins, but not enough to keep them in the circulating volume, and so the volume of distribution for these peptides resembles something between extracellular fluid and total body water (0.2-0.5L/kg). 

Metabolism and elimination

This is where the species of vasopressin analogues differ the most in their pharmacokinetic properties. 

  • Vasopressin itself is cleared rather rapidly by predominantly hepatic metabolism into inactive breakdown products, though it appears to be mediated mainly by the hepatic sinusoidal endothelium and not the hepatocytes. As a result its half life is said to be about 17-35 minutes.
  • Terlipressin is metabolised by the same hepatic endothelial enzymes into lysine vasopressin, the porcine hormone. This substance has a puzzlingly high affinity for human V1 receptors, which means the duration of effect is about 6 hours, even if the drug itself only has a half-life of 60 minutes.
  • Desmopressin contains a D-arginine in the 8th position instead of an L-arginine, which makes it essentially invisible to the normal enzymes that metabolise vasopressin. In contrast to the other agents this one is eliminated mainly by the kidneys (65%) and will accumulate in renal failure.

Apart from their duration of action, vasopressin analogs differ the most in their different affinity for vasopressin receptors, which results in completely different clinical use indications for vasopressin terlipressin and desmopressin.

Pharmacodynamics of vasopressin and its analogues

Vasopressin binds to high-affinity G-protein-coupled membrane receptors, of which there are two main classes (unimaginatively designated V1 and V2). There are also V3 receptors in the central nervous system, which are mainly involved in coordinating the secretion of corticotropin (which is often required whenever a large amount of vasopressin is required), but these are generally involved in the kind of physiological micromanagement that would be invisible to the intensivist, and so will not be discussed at any great length below. For the critical care practitioner, V1 (vasopressor) and V2 (antiduretic) receptors are by far the most important.  

V2 receptors: the antidiuretic receptors

Though violating the natural rules of numerical order, V2 receptors are a more logical place to start, as vasopressin is at a fundamental level a hormone designed to regulate body water. This receptor system is ancient, pre-dating the divergence of vertebrates and invertebrates, and as we have already mentioned virtually all animals have some sort of vasopressin-like peptides. The V2 receptor system specifically seems to have arisen in the late Silurian period, perhaps 440-410 million year ago, when our fishy ancestors transitioned to a semi-aquatic lifestyle and found themselves in need of a mechanism that could help them conserve body water by reabsorbing it from the urine. The basic structure of the receptor complex has become so essential that over about 170 million years of mammalian evolution most of the amino acid sequence has remained essentially unchanged.  Juul (2012) produces an excellent discussion of these evolutionary detailes, in case the reader has had their fill of exam-relevant material and needs to escape into a fascinating diversion. 

V2 receptors are Gs-protein coupled receptors, which means their intracellular activities are mediated mainly by the actions of cAMP.

V2 receptor intracellular signalling pathway

At the principle cells of the collecting tubule, the activated protein kinase A causes the migration of aquaporin-containing vesicles to the apical membrane of the cell. It is thought that these vesicles are constantly cycling between the cell surface and interior, and the addition of vasopressin just shifts the equilibrium in favour of luminal expression. The appearance of aquaporin proteins on the surface greatly increases the water permeability of this cell, and water flows in along an osmotic gradient supported by the high urea concentration in this part of the nephron. Obviously, if too much water were reabsorbed here, this concentration gradient would dilute down and collapse, and so to maintain an abundance of urea an active urea transporter (VRUT) is also activated by the action of vasopressin on V2 receptors.

Non-renal effects of V2 vasopressin receptor activation

It is probably worth mentioning that V2 receptors can perform other tricks, of which the most useful is probably to increase the availability of circulating Factor VIII and von Willebrand factor. This is only seen at extremely high concentrations of agonist, eg. for desmopressin a dose of around 0.4 μg/kg is required (something like 24-28 μg). The mechanism seems to be a release of these clotting factors from endothelial storage sites (Weibel-Palade bodies). This property is often turned to by intensivists who need to immediately reverse some antiplatelet drug effects.

Apart from that V2 receptors are expressed surprisingly widely, and have a host of poorly studied effects which are probably irrelevant to the ICU exam candidate. Juul et al (2014) list some of these in their excellent article. As an example, they seem to mediate some kind of nootropic effect, pulmonary vasodilation, lymphocyte activity, and various others. It is expressed in expressed in the inner ear, the exocrine pancreas, parathyroid glands, skin cells and salivary glands. 

V1 receptors: the vasopressor receptors

V1 receptors are Gq-protein coupled receptors, just like the α-1 noradrenaline receptors. They are heptaspanning membrane proteins and their activation causes smooth muscle contraction, by stimulating the release of calcium from the sarcoplasmic reticulum. This mediates the vasoconstrictor effects of vasopressin. The potency of this effect is about 100 to 1000 times greater than the same effect from catecholamines, in the sense that the concentration of vasopressin required to achieve the same clinical effect is much smaller. Moreover, both vasopressin and catecholamine receptors can be activated at the same time, which is why the vasopressin infusion can be used as a noradrenaline-sparing drug in the ICU (i.e. the same cAMP concentration can be achieved inside the cells at a lower noradrenaline dose if vasopressin does some of the work).

v1 receptor intracellular signalling pathway

The effects of V1 vasopressin receptor activation

V1 vasopressin receptors are ubiquitous, but the densest distribution is seen in the vascular smooth muscle, reflecting the role of vasopressin in circulatory regulation. You wouldn't call its role "pivotal" or "decisive", as it is clearly possible to maintain a blood pressure purely by the activity of the other neurohormonal systems, but it certainly plays its part, especially when those other systems are challenged (for example, when Carp et al (1994) inflicted T2-level epidural anaesthesia on their human volunteers, the blood pressure drop was modest until the vasopressin and angiotensin systems were both suppressed). 

The clinical effects of activating V1 receptors are well detailed in the chapters dealing with vasopressin as a vasopressor, particularly where is its used for the resuscitation of septic shock. For healthy people not in advanced stages of septic shock, vasopressin seems to perform a subtle and mainly regulatory role in the direction distribution of blood flow, and its effects on blood pressure are suitably complex and variable.  In summary, these can be listed as follows:

  • Vasoconstriction (everywhere), which is not reversed by sympathetic blockade. All vessels seem to be affected, including the splanchnic circulation, coronaries, and the brain. There are also some venoconstrictor effects, and these are minor and generally less prominent then those of catecholamines, but additive with them  (Trippodo, 1981, and Diana & Masen, 1965). The overall pattern is similar to the effects of activating the sympathetic nervous system: blood flow is redistributed away from nonessential tissues and towards vital organs.
    • Splanchnic blood flow is definitely affected, not to the extent that routine use might worry you with gut ischaemia, but definitely to the extent that portal venous pressure is reduced by the decrease in splanchnic blood flow. That is in fact the main reason for the use of terlipressin in the scenario of variceal haemorrhage.
    • Skin circulation is reduced, to the point where some authors blame vasopressin for cutaneous and digital ischaemia in septic shock patients (though realistically it may just be that they were using a lot of noradrenaline - the mean dose was about 0.67 mcg/kg/min, almost 50ml/hr of a standard 6mg/100ml dilution).
    • Renal blood flow is affected unevenly, with afferent arterioles relatively spared and efferent arterioles more constricted, which in effect produces an increase in the glomerular filtration rate and an increase in the urine output
  • Platelet aggregation – Vasopressin is also stored in platelet granules
  • Increased intestinal motility
  • Increased glycogenolysis in the hepatocytes
  • Increased ACTH release
  • Increased smooth muscle cell growth
  • Increased prostaglandin synthesis

The effect of acidosis on V1 vasopressin receptor affinity

One interesting feature of V1 receptors is their supposed increased affinity for their ligand in the face of wildly erratic biochemical conditions. It is a view generally held by intensivists (and repeated in their online rants and peer-reviewed publications) that "the effects of arginine vasopressin are preserved during hypoxia and severe acidosis", whereas the effects of catecholamines tend to decrease quite substantially with even modest changes in pH. Wherever this sort of statement is found, it tends to be accompanied by a reference to a 1992 article by Fox et al, to the extent that it seems each time anybody has claimed that vasopressin is unaffected by acidosis over the last thirty years, their article has directly or indirectly quoted this paper.

That Fox experiment was a rat study, only the first page of which appears to be available online as the result of some kind of weird decision by the Journal of Cardiovascular Pharmacology (the links of that entire issue seem to suggest that they should lead to a full text article). Without delving into the exact specifics, it appears they found that at lower pH vasoconstrictor responses to acidosis are relatively preserved for vasopressin, but not for α-1 agonists, on the basis of observing the vascular resistance of isolated rat tail arteries under laboratory conditions. Findings by Okada et al (1991) seem to contradict this somewhat - they found the vasopressor activity and receptor affinity of vasopressin does decrease with low pH, becoming approximately halved at a pH of 7.0 (which is still a lot better than catecholamines, as they are 75% less effective at a pH of 7.15). 

Human clinical data is even messier. The question of whether vasopressin will be better for severely acidotic patients than just increasing catecholamine doses is difficult to answer because many other factors influence what  and  what "better" means in any given clinical scenario. For example,  Bauer et al (2022) observed a retrospective cohort of over 1300 patients and found that those who had lower pH were still less likely to have a meaningful response to vasopressin.  Only about 33% of patients with pH less than 7.19 had any improvement in their blood pressure or catecholamine dose requirements when they were started on a vasopressin infusion. One could theorise that, no matter how much more stable the affinity of vasopressin receptors must be for their natural ligand at low pH, the circulatory systems of some septic shock patients are just fundamentally broken, and require something other than just a mindless uptitration of vasopressors to repair. 

The distribution of V1 vasopressin receptors

V1 vasopressin receptors are ubiquitous, but the densest distribution is seen in the vascular smooth muscle, reflecting the role of vasopressin in circulatory regulation. You wouldn't call its role "pivotal" or "decisive", as it is clearly possible to maintain a blood pressure purely by the activity of the other neurohormonal systems, but it certainly plays its part, especially when those other systems are challenged (for example, when Carp et al (1994) inflicted T2-level epidural anaesthesia on their human volunteers, the blood pressure drop was modest until the vasopressin and angiotensin systems were both suppressed).  For healthy people not in advanced stages of septic shock, vasopressin seems to perform a subtle and mainly regulatory role in the direction distribution of blood flow, and its effects on blood pressure are suitably complex and variable.  


Glavaš, Mladena, et al. "Vasopressin and its analogues: from natural hormones to multitasking peptides." International Journal of Molecular Sciences 23.6 (2022): 3068.

Holmes, Cheryl L., Donald W. Landry, and John T. Granton. "Science review: vasopressin and the cardiovascular system part 1–receptor physiology." Critical care 7.6 (2003): 1-8.

Holmes, Cheryl L., Donald W. Landry, and John T. Granton. "Science review: vasopressin and the cardiovascular system part 2–clinical physiology." Critical care 8.1 (2003): 1-9.

Turner, Robert A., John G. Pierce, and Vincent du Vigneaud. "The purification and the amino acid content of vasopressin preparations." Journal of Biological Chemistry 191.1 (1951): 21-28.

Oliver, George, and E. A. Schäfer. "On the Physiological Action of Extracts of Pituitary Body and certain other Glandular Organs Preliminary Communication."The Journal of physiology 18.3 (1895): 277-279.

Van Dyke, H. B., et al. "The isolation of a protein from the pars neuralis of the ox pituitary with constant oxytocic, pressor and diuresis-inhibiting activities." Journal of Pharmacology and Experimental Therapeutics 74.2 (1942): 190-209.

"Vasopressin Synthesised" Chem. Eng. News, 1956, 34 (23), p 2754

Nielsen, Aage Theil, and Christian Hamburger. "The Oxytocic and Pressor Activities of the USP Posterior Pituitary Reference Standard." Acta Pharmacologica et Toxicologica 12.2 (1956): 200-210.

Dale, Henry Hallett, and Patrick Playfair Laidlaw. "A method of standardising pituitary (infundibular) extracts." Journal of Pharmacology and Experimental Therapeutics 4.1 (1912): 75-95.

Holton, Pamela. "A modification of the method of Dale and Laidlaw for standardization of posterior pituitary extract." British journal of pharmacology and chemotherapy 3.4 (1948): 328-334.

Odekunle, Esther A., and Maurice R. Elphick. "Comparative and evolutionary physiology of vasopressin/oxytocin-type neuropeptide signaling in invertebrates." Frontiers in Endocrinology 11 (2020): 225.

Harris, A. S., et al. "Effects of concentration and volume on nasal bioavailability and biological response to desmopressin." Journal of pharmaceutical sciences 77.4 (1988): 337-339.

Juul, Kristian Vinter. "The evolutionary origin of the vasopressin/V2-type receptor/aquaporin axis and the urine-concentrating mechanism." Endocrine 42.1 (2012): 63-68.

Juul, Kristian Vinter, et al. "The physiological and pathophysiological functions of renal and extrarenal vasopressin V2 receptors." American journal of physiology-renal physiology 306.9 (2014): F931-F940.

Dünser, Martin W., et al. "Management of vasodilatory shock." Drugs 63.3 (2003): 237-256.

Mutlu, Gökhan M., and Phillip Factor. "Role of vasopressin in the management of septic shock." Intensive care medicine 30.7 (2004): 1276-1291.

Fox, Anthony W., Robert E. May, and William E. Mitch. "Comparison of peptide and nonpeptide receptor-mediated responses in rat tail artery." Journal of cardiovascular pharmacology 20.2 (1992): 282-289.

Okada, K. O. J. I., et al. "Effects of extra-and intracellular pH on vascular action of arginine vasopressin." American Journal of Physiology-Renal Physiology 260.1 (1991): F39-F45.

Bauer, Seth R., et al. "Association of Arterial pH With Hemodynamic Response to Vasopressin in Patients With Septic Shock: An Observational Cohort Study." Critical care explorations 4.2 (2022).

Carp, H., et al. "Endogenous vasopressin and renin-angiotensin systems support blood pressure after epidural block in humans." Anesthesiology 80.5 (1994): 1000-7.

Dünser, Martin W., et al. "Ischemic skin lesions as a complication of continuous vasopressin infusion in catecholamine-resistant vasodilatory shock: incidence and risk factors." Critical care medicine 31.5 (2003): 1394-1398.