Vasopressin

Vasopressin is synthesised in the hypothalamus and released from the posterior pituitary in response to rising plasma osmolality or falling blood volume. V1a receptors on vascular smooth muscle are Gq-coupled, raising intracellular calcium through phospholipase C and producing vasoconstriction that is preserved even in the acidotic, catecholamine-resistant vasculature of late septic shock. V2 receptors in the renal collecting duct drive aquaporin-2 insertion and free-water reabsorption. Endogenous vasopressin is relatively deficient in prolonged septic shock, which is the rationale for low-dose replacement infusion rather than pressor-dose escalation; the VASST trial established that adding it to noradrenaline lets you use less noradrenaline.

Targets: V1a receptor, V2 receptor, V1b receptor, Vascular smooth muscle tone

• · Dosed in units, not micrograms: 0.01 to 0.03 units per minute is standard, with 0.04 units per minute as an upper limit in most protocols. Not titrated like a catecholamine; it is run at a fixed low dose and weaned last or second-to-last.

Work out your exact syringe units →

• commonDigital, mesenteric and skin ischaemia— The dose-limiting toxicity. Necrosis of fingers and toes has been reported at higher infusion rates.
• commonHyponatraemia— From V2-mediated free-water retention.
• commonBradycardia and reduced cardiac output— Afterload rises without any inotropic support.
• uncommonExtravasation tissue necrosis— Central line administration is strongly preferred.

• Not an outpatient drug under any circumstances
• Caution in known coronary, mesenteric or peripheral arterial disease
• Hypersensitivity to chlorobutanol in preserved formulations

• redundantdesmopressin — Desmopressin is the V2-selective analogue used when antidiuresis rather than vasoconstriction is wanted.
• redundantterlipressin — Terlipressin is the V1-selective prodrug used for hepatorenal syndrome and variceal bleeding.

• · Continuous arterial blood pressure
• · Serum sodium daily
• · Lactate and peripheral perfusion
• · Direct inspection of digits and skin

Prescription hospital product in the US and EU.

• Vasostrict FDA prescribing information (label)
• Russell et al. 2008 NEJM, VASST trial of vasopressin versus noradrenaline in septic shock (trial)
• Gordon et al. 2016 JAMA, VANISH trial of early vasopressin in septic shock (trial)

The reason vasopressin earns a place in vasodilatory shock has nothing to do with being a stronger pressor than noradrenaline, because it is not. It is that its receptor and second-messenger pathway are different. V1a is Gq-coupled: phospholipase C generates inositol trisphosphate and diacylglycerol, calcium comes out of the sarcoplasmic reticulum and PKC sensitises the contractile apparatus. Adrenergic receptors are Gs-coupled and are progressively desensitised and downregulated by prolonged catecholamine exposure and by the acidosis of late shock, whereas V1a signalling holds up. On top of that, endogenous vasopressin stores in the neurohypophysis are depleted after roughly the first day of septic shock, so a genuine relative deficiency develops, and the low fixed-dose infusion is replacement rather than pharmacological escalation. Vasopressin also closes ATP-sensitive potassium channels that nitric oxide and acidosis have forced open in vascular smooth muscle, which is a large part of why vasoplegia responds. The efferent arteriole of the glomerulus is more responsive than the afferent, which raises filtration fraction and is the proposed reason VANISH found less renal replacement therapy. The dose-limiting toxicity is the same mechanism unopposed: V1a receptors are dense in skin, digits and mesentery, and above about 0.04 units per minute you buy ischaemia rather than pressure.

Peripheral administration is the classic mistake: extravasation causes tissue necrosis and the drug should go through a central line wherever possible. The second is escalating past 0.04 units per minute in the belief that more pressure is coming; what arrives instead is dusky fingers, mottled skin and, in the worst cases, digital gangrene or bowel ischaemia. Somebody has to physically look at the hands, feet and skin at least once a shift, because the monitor will not tell you. The third is abrupt discontinuation, which produces rebound hypotension because endogenous stores are still depleted. The fourth is missing the hyponatraemia in a patient being flooded with hypotonic fluid. The fifth is a units-versus-millilitres error at the pump: vasopressin is dosed in units per minute while everything around it is in micrograms per kilogram per minute, and that mismatch has caused real overdoses.

1. —Start — 0.03 units per minute as a fixed infusion, added when noradrenaline is around 0.25 to 0.5 mcg/kg/min. Dosed in units, not micrograms. Most units do not titrate it at all.
2. —If used as a titratable agent — 0.01 to 0.04 units per minute is the accepted range. Above 0.04 you are buying digital, skin and mesenteric ischaemia for very little extra pressure, and most protocols cap there.
3. —Weaning — Wean the catecholamine first in most protocols, then reduce vasopressin in 0.01 unit per minute steps. Abrupt discontinuation causes rebound hypotension, which is a well-described and avoidable event.

• V1a receptor on vascular smooth muscle — High. Numeric Ki not resolved this session.

  Gq-phospholipase C signalling raising intracellular calcium and producing vasoconstriction that persists in acidotic, catecholamine-resistant vasculature. Also closes ATP-sensitive potassium channels opened by nitric oxide in vasoplegia.

• V2 receptor in renal collecting duct — High

  Aquaporin-2 insertion and free-water reabsorption, which is why hyponatraemia develops on prolonged infusion.

• V1b (V3) receptor in anterior pituitary — Present

  Potentiates CRH-driven ACTH release, contributing to a rise in endogenous cortisol during infusion.

• Oxytocin receptor — Weak cross-reactivity

  Clinically minor at pressor doses, but the structural similarity is why oxytocin has weak antidiuretic activity and vice versa.

• VASST Randomised controlled trial · 2008

  28-day mortality in septic shock, comparing low-dose vasopressin plus noradrenaline against noradrenaline alone. No overall mortality difference; the drug is catecholamine-sparing, with a signal of benefit in the less severe shock stratum.

• VANISH Randomised, factorial, double-blind · 2016

  Kidney failure-free days in septic shock. Early vasopressin did not improve the primary renal endpoint but reduced the use of renal replacement therapy.

Blood pressure begins to rise within minutes of starting the infusion and the full effect of a given rate is present within about 15 minutes. Because the half-life is 10 minutes or less, the haemodynamic effect also fades within roughly 20 minutes of stopping, which makes it easy to titrate but unforgiving of accidental interruptions. Hyponatraemia develops over days, not hours. Ischaemic complications typically declare themselves after 12 to 48 hours of higher-rate infusion.

Vasopressin is almost never used alone. The standard construction is noradrenaline as the titratable agent with vasopressin added at a fixed low dose as a catecholamine-sparing second mechanism, and in refractory vasoplegia a third mechanism, angiotensin II or methylene blue, can be layered on. Hydrocortisone is a rational partner, since relative adrenal insufficiency and vasopressin deficiency travel together and steroid plus vasopressin has been examined in post-hoc VASST analyses. Do not combine vasopressin with terlipressin, and do not combine it with desmopressin; they are the same receptor family aimed at different problems. Watch for stacking with any drug that impairs free-water excretion, because the V2 arm is already working against you.

Against noradrenaline: noradrenaline is the first-line titratable vasopressor and vasopressin is the fixed-dose adjunct. Vasopressin has no inotropic effect at all, so in a patient whose problem is cardiac output rather than tone it makes things worse by raising afterload. Against angiotensin II: a newer third mechanism for refractory vasoplegia, with the same logic of adding a non-catecholamine pathway. Against terlipressin: terlipressin is the slow-release, splanchnic-selective relative used for hepatorenal syndrome and variceal bleeding, not for shock. Against desmopressin: the same molecule stripped of its pressor arm, used for the opposite clinical purpose. And a point worth being blunt about: this drug has no home use, no enhancement use and no sensible role outside a monitored bed with an arterial line.

Not applicable as a monthly cost. This is an inpatient infusion priced per vial, and its US price history is notorious: Vasostrict went from an unapproved cheap generic to a branded product costing orders of magnitude more after FDA approval, before generic competition arrived. Figures not sourced in this session.

• ?Volume of distribution is quoted by the label as 140 mL/kg rather than in litres, so the litre field is left null; for an 80 kg adult that is roughly 11 L.
• ?Plasma protein binding is not stated in the label.
• ?VASST enrolled around 780 patients and VANISH around 400, but I did not confirm either figure, so both are null.
• ?Numeric V1a and V2 receptor affinities were not resolved this session.
• ?Whether vasopressin has any mortality benefit in any subgroup remains genuinely unresolved; the VASST less-severe-shock signal was a subgroup finding and has never been prospectively confirmed.

• Vasopressin versus norepinephrine infusion in patients with septic shock Russell JA, et al., New England Journal of Medicine, 2008 · PMID 18305265

  VASST. The trial that established catecholamine sparing without a mortality benefit, which is still the honest summary of this drug.

• Effect of Early Vasopressin vs Norepinephrine on Kidney Failure in Patients With Septic Shock: The VANISH Randomized Clinical Trial Gordon AC, et al., JAMA, 2016 · PMID 27483065

  The renal question, answered ambiguously: no change in the primary endpoint, less renal replacement therapy.

• VASOSTRICT (vasopressin) injection - FDA prescribing information Par Health USA, DailyMed

  Source of the 140 mL/kg volume of distribution, 9 to 25 mL/min/kg clearance, half-life of 10 minutes or less, the serine protease and disulfide oxido-reductase metabolism, the 6 percent unchanged urinary excretion and the nine-residue sequence.