Vasopressors and Inotropes

In depth

The one-page summary above is what the PDF prints. These sections go further.

Receptor physiology and second messengers

Every vasopressor ends in the same place: free calcium and myosin light-chain phosphorylation in vascular smooth muscle. α1, V1 and AT1 receptors are all Gq-coupled: phospholipase C splits PIP2 into IP3, which releases calcium from the sarcoplasmic reticulum, and DAG, which activates protein kinase C. Calcium–calmodulin activates myosin light-chain kinase and the vessel contracts; RhoA/Rho kinase adds calcium sensitization by inhibiting myosin light-chain phosphatase. Because the three receptor families are independent, a patient whose α1 receptors are saturated or downregulated can still respond to vasopressin or angiotensin II, which is the logic of multimodal vasopressor therapy.

β1 and β2 receptors are Gs-coupled: adenylyl cyclase makes cAMP, which activates protein kinase A. In the heart PKA phosphorylates L-type calcium channels (more calcium in, stronger contraction), phospholamban (faster SERCA reuptake, better relaxation) and troponin I, so β1 stimulation raises inotropy, chronotropy, dromotropy and lusitropy, and myocardial oxygen demand with them. In smooth muscle the same cAMP rise inhibits myosin light-chain kinase, so β2 stimulation relaxes arterioles and bronchi. PDE3 inhibitors raise cAMP by blocking its breakdown, which is why milrinone still works when β receptors are blocked or downregulated.

Septic vasoplegia attacks this machinery at several points. Inducible NO synthase floods smooth muscle with NO, which raises cGMP through soluble guanylate cyclase and activates myosin light-chain phosphatase. Open K-ATP channels hyperpolarize the cell and close voltage-gated calcium channels. Prolonged catecholamine exposure desensitizes adrenergic receptors (GRK phosphorylation, β-arrestin binding, internalization), relative cortisol insufficiency lowers receptor expression, and vasopressin levels fall inappropriately low once early stores are spent (Landry, Circulation 1997). Endothelial injury reduces ACE activity, so angiotensin I accumulates while angiotensin II falls. Each mechanism has a matching rescue drug (see below).

Drug by drug: kinetics, dose-response, adverse effects

Catecholamines have half-lives of about 2 minutes: they are taken back up by nerve terminals and broken down by COMT and MAO, so steady state arrives within about 10 minutes of a rate change and the effect disappears within minutes of an interruption. Phenylephrine, milrinone and levosimendan are the exceptions with longer tails.

Receptor selectivity changes with dose. Epinephrine at 0.01–0.05 mcg/kg/min is mainly β: cardiac output and HR rise while β2 vasodilation can lower diastolic pressure; α1 constriction dominates as the dose climbs. The classic dopamine bands (DA below 5, β1 5–10, α1 above 10 mcg/kg/min) overlap widely between patients, so a 'renal dose' may already be chronotropic. Norepinephrine dose is the common severity marker: around 0.1 mcg/kg/min is moderate support, and requirements above about 0.5 mcg/kg/min, and especially above 1, define refractory shock, with reported mortality of 50% or more. Vasopressin is fixed because the aim is to replace a deficient hormone; doses above about 0.04 units/min add pressure along with more digital, mesenteric and myocardial ischemia.

US labels dose norepinephrine as the base; many European and Australian trials report the tartrate salt, about twice the base number for the same drug (1 mcg base ≈ 2 mcg bitartrate).

• Norepinephrine: digital, skin and splanchnic ischemia, worse with hypovolemia; extravasation necrosis.
• Epinephrine: tachyarrhythmias and myocardial ischemia from higher O₂ demand; lactate rise from β2-driven aerobic glycolysis (type B, not hypoperfusion); hyperglycemia; hypokalemia from β2 Na⁺/K⁺-ATPase stimulation; splanchnic vasoconstriction.
• Vasopressin: patchy skin and fingertip necrosis, mesenteric ischemia, lower cardiac output and HR as afterload rises; hyponatremia is uncommon at 0.03 units/min.
• Phenylephrine: reflex bradycardia and a fall in cardiac output when the ventricle is weak; raises PVR.
• Dopamine: tachyarrhythmias, especially AF; suppresses prolactin and TSH; no renal benefit.
• Dobutamine: tachycardia, ventricular ectopy, hypotension from β2 vasodilation; tolerance after about 72 h; rare eosinophilic hypersensitivity myocarditis with long infusions.
• Milrinone: hypotension, atrial and ventricular arrhythmias, accumulation in AKI; thrombocytopenia (label 0.4%).
• Angiotensin II: arterial and venous thrombosis (label 13% vs 5% with placebo), so give VTE prophylaxis; tachycardia, peripheral ischemia, delirium.

Choosing the agent by shock type and patient

Cardiogenic shock: norepinephrine is preferred over dopamine (the SOAP II cardiogenic subgroup had higher mortality with dopamine) and over epinephrine (OptimaCC, Levy JACC 2018: refractory shock 37% vs 7% after MI; stopped early). Once MAP is adequate, an inotrope treats low output, and temporary mechanical support is the escalation, not a third or fourth pressor.

RV failure and pulmonary hypertension: the RV is perfused in systole and diastole, so its blood supply depends on the gap between aortic and RV pressure. Systemic hypotension starts a spiral of RV ischemia, dilation, septal shift and lower LV output. Norepinephrine restores the gradient; vasopressin raises SVR with little rise in PVR at low doses. Phenylephrine raises PVR and adds no inotropy. Milrinone and dobutamine add contractility and lower PVR but also lower SVR, so they usually run with a vasopressor; inhaled NO or epoprostenol lowers PVR without systemic hypotension. Avoid hypoxia, hypercapnia and acidosis, which raise PVR.

Dynamic LVOT obstruction (HOCM, systolic anterior motion after mitral repair, apical ballooning with LVOT obstruction): anything that raises contractility, speeds the heart or shrinks the LV worsens the gradient. Treat with volume, a pure afterload agent (phenylephrine or vasopressin) and rate control with a β-blocker; inotropes, dopamine, nitrates and diuretics are harmful (Sherrid, JAHA 2021). Suspect it when a patient worsens on dobutamine.

β-blocked patients: in β-blocker or calcium-channel-blocker toxicity the receptor is occupied or the channel downstream is blocked. Glucagon (its own Gs-coupled receptor; 3–5 mg IV, then infusion), high-dose insulin euglycemia therapy (1 unit/kg bolus, then 1 unit/kg/h titrated upward with dextrose and K⁺ checks), milrinone, calcium salts and high-dose catecholamines all act beyond the β receptor; VA ECMO is the rescue.

Neurogenic shock: loss of sympathetic tone gives hypotension with bradycardia, so norepinephrine (α1 plus β1) or dopamine fits better than phenylephrine. In acute spinal cord injury, AANS/CNS 2013 targeted MAP 85–90 for 7 days; AO Spine/Praxis 2024 suggests a lower limit of 75–80 and an upper limit of 90–95 for 3–7 days.

Starting, titrating and weaning

Start low (for example norepinephrine 0.05 mcg/kg/min or 5 mcg/min) and titrate in small steps, commonly 0.02–0.05 mcg/kg/min or 1–2 mcg/min every 5–10 min, since steady state takes about 5 half-lives. Titrate to the MAP target and reassess perfusion (mentation, capillary refill, mottling, urine output, lactate trend), not MAP alone.

A MAP of 65 suits most patients; the 65 trial supports 60–65 at age 65 or older, and in SEPSISPAM a target of 80–85 only reduced renal replacement in chronic hypertension.

No randomized trial defines the weaning order. Common practice removes the last-added and most toxic agent first (epinephrine), then lowers norepinephrine, and stops vasopressin last once norepinephrine is low (often below about 0.05–0.1 mcg/kg/min), because retrospective cohorts found more hypotension when vasopressin was stopped first (Bauer, J Crit Care 2010, and later cohorts). Vasopressin is then stopped outright or tapered by 0.01 units/min every 30–60 min; practice varies. Wean norepinephrine in small steps every 15–30 min while MAP holds, one drug at a time.

Peripheral administration and extravasation

SSC 2026 suggests starting peripherally rather than delaying for a central line. A 2026 meta-analysis (ZhangJian, JAMA Netw Open) found minor adverse events in 2.3% of peripheral infusions, major injury close to 0% with short peripheral catheters vs 1.4% with midlines, and a central line avoided in 59.7%. In Loubani's systematic review (J Crit Care 2015) most tissue injuries came from catheters distal to the antecubital or popliteal fossa.

• Catheter: 20 G or larger, ideally ultrasound-guided and long enough to sit well within a large upper-arm or forearm vein; not the hand, wrist or a flexion crease.
• Confirm brisk blood return before starting and at each check; dedicated line with nothing else running.
• Check the site every 1–2 h (hourly is common): pain, blanching, swelling, coolness, leaking.
• Many protocols cap dose (for example norepinephrine 0.2–0.3 mcg/kg/min) and duration (24–72 h) before a central line is required; SSC 2026 sets no limits.

1. Stop the infusion, aspirate what you can through the catheter, then remove it.
2. Restart the pressor at another site or centrally right away: the patient still needs it, and MAP falls within minutes.
3. Mark the borders, photograph and elevate the limb.
4. Catecholamine or phenylephrine: phentolamine 5–10 mg in 10–15 mL saline as multiple small subcutaneous injections across the blanched area with a fine needle, as soon as possible (label: within 12 h). Blanching should reverse quickly; watch for hypotension and tachycardia.
5. No phentolamine: terbutaline 0.5–1 mg diluted in up to 10 mL saline infiltrated subcutaneously, or 2% nitroglycerin ointment over the area (Stefanos, Pharmacotherapy 2023).
6. Warm vs cold compresses: sources disagree; follow policy.
7. Reassess hourly at first; wound care or plastic surgery for blistering or necrosis; document.

Catecholamine resistance and rescue for vasoplegia

Before adding another drug, look for a fixable reason the current ones are failing.

• Delivery: wrong drug or concentration, wrong weight, infiltrated or disconnected line, pump occlusion, empty bag, a line that was flushed.
• Measurement: damped or misleveled arterial line; compare with a cuff.
• Volume and source: ongoing bleeding, hypovolemia, uncontrolled infection.
• Chemistry: severe acidosis (pH below about 7.15–7.2 blunts catecholamine response), low ionized calcium after transfusion, relative adrenal insufficiency.
• Obstruction: tamponade, tension pneumothorax, massive PE, auto-PEEP, abdominal compartment syndrome.
• Wrong physiology: LVOT obstruction or a cardiogenic component; get an echo.

Methylene blue inhibits NO synthase and soluble guanylate cyclase. The best septic shock data are a small single-center RCT (Ibarra-Estrada, Crit Care 2023, 91 patients) in which methylene blue given early shortened time to vasopressor discontinuation by about a day and increased vasopressor-free days, with no mortality difference; most other data are post-cardiac-surgery cohorts. It inhibits MAO-A, so it can cause serotonin syndrome with SSRIs, SNRIs and linezolid (FDA warning 2011); it causes hemolysis and methemoglobinemia in G6PD deficiency, reads falsely low on pulse oximetry, turns urine blue-green and can worsen oxygenation by blunting hypoxic pulmonary vasoconstriction.

Hydroxocobalamin binds NO and hydrogen sulfide; it is used mainly for vasoplegia after cardiopulmonary bypass, especially when methylene blue is contraindicated, and the evidence is case series and retrospective cohorts. It turns skin, plasma and urine red for days, interferes with colorimetric labs and CO-oximetry, and triggers false blood-leak alarms on dialysis and CRRT machines.

Inotropes: dobutamine, milrinone and levosimendan

DOREMI (Mathew, NEJM 2021) randomized 192 cardiogenic shock patients to milrinone or dobutamine: the composite of in-hospital death, arrest, transplant or MCS, MI, stroke or new dialysis was 49% vs 54% (RR 0.90, not significant), with no difference in any component. Neither drug wins, so choose by patient: milrinone for β-blocked patients and RV failure with pulmonary hypertension; dobutamine for borderline BP, AKI or when a fast offset matters. OPTIME-CHF (Cuffe, JAMA 2002): routine milrinone in heart failure without shock caused more hypotension and atrial arrhythmias with no benefit.

Levosimendan sensitizes troponin C to calcium without raising intracellular calcium and opens K-ATP channels, so it is an inodilator that does not raise myocardial O₂ demand much. Its active metabolite keeps working for about a week after a 24 h infusion (0.05–0.2 mcg/kg/min, load often omitted because of hypotension). It is not FDA-approved. SURVIVE (2007), LeoPARDS (sepsis) and cardiac surgery trials were negative, and SSC 2026 suggests against it in septic shock.

Nursing safety: pumps, lines, labels and daily care

ISMP lists IV adrenergic agonists and inotropes as high-alert medications. Typical errors: wrong concentration, weight or units entry, a flush through the pressor line, a gap at a bag change.

• Standardize one concentration per drug where possible (for example norepinephrine 4 mg/250 mL, 16 mcg/mL; vasopressin 20 units/100 mL, 0.2 units/mL) and build them into the smart pump library with hard and soft limits. A double- or quadruple-strength bag needs its own library entry and label.
• Program in dose units (mcg/kg/min, mcg/min, units/min, ng/kg/min), never mL/h, and confirm a measured weight in kg.
• Independent double-check at start, concentration change and handoff where policy requires.
• Label each line at the bag, the pump channel and near the connection; trace every line from bag to patient at handoff.
• Dedicated lumen; never flush, bolus or push other drugs through it.
• Hang the next bag before the current one runs out and switch over on a second channel per policy; with a 2-minute half-life, a short gap drops MAP.
• Arterial line leveled at the phlebostatic axis, zeroed and checked with a fast-flush test before you titrate to it.

• Hemodynamics: MAP trend, rhythm, total norepinephrine equivalents, capillary refill, mottling, lactate.
• Skin and gut: dusky fingers, toes, ears or nose; distension, feeding intolerance or GI bleeding (mesenteric ischemia).
• Renal and labs: hourly urine output, creatinine (milrinone); glucose and K⁺ on epinephrine, Na⁺ on vasopressin, ionized calcium.
• Lines and documentation: site checks, central line need; dose in drug units, MAP target, every titration.
• Transport: full bags, charged pumps, labeled lines, transduced arterial line, push-dose pressor drawn up.

Evidence and controversies

What changed practice: SOAP II ended dopamine as a first-line pressor. VASST and VANISH showed vasopressin is a safe norepinephrine-sparing add-on but not a survival drug; VASST's less-severe stratum (norepinephrine 5–14 mcg/min, 26.5% vs 35.7%) is the reason many add it early. ATHOS-3 led to FDA approval of angiotensin II in 2017. The 65 trial and SEPSISPAM moved practice toward accepting 60–65 in older patients.

Unsettled: when to add vasopressin (SSC 2026 sets no threshold), the weaning order, whether peripheral dose and duration limits matter, whether MAP targets should follow the patient's baseline pressure, and whether methylene blue or hydroxocobalamin change outcomes rather than numbers.

Exam pearls

• α1, V1 and AT1 are Gq (IP3, calcium); β1, β2, D1 and V2 are Gs (cAMP). Milrinone raises cAMP beyond the receptor, so it works during β-blockade.
• Norepinephrine is first line in septic and most cardiogenic shock; dopamine causes more arrhythmias (SOAP II).
• Vasopressin runs at a fixed 0.03 units/min; it spares norepinephrine but did not lower mortality (VASST, VANISH).
• Epinephrine raises lactate through β2 metabolism, so a rising lactate on epinephrine does not always mean worse perfusion.
• Phenylephrine causes reflex bradycardia and is the pressor for LVOT obstruction; avoid it in RV failure.
• Angiotensin II is dosed in ng/kg/min and needs VTE prophylaxis.
• Methylene blue: serotonin syndrome with SSRIs, avoid in G6PD deficiency, falsely low SpO₂.
• Catecholamine extravasation: stop, aspirate, remove, phentolamine subcutaneously within 12 h, restart elsewhere.
• Milrinone accumulates in renal failure and drops BP; dobutamine causes tachycardia and loses effect after about 72 h.
• Never flush a pressor line; titrate to MAP and perfusion and wean one drug at a time.