Use of vasopressors and inotropes

INTRODUCTION — 

Vasopressors are a powerful class of drugs that induce vasoconstriction and thereby elevate mean arterial pressure (MAP). Vasopressors differ from inotropes, which increase cardiac contractility; however, many drugs have both vasopressor and inotropic effects. Although many vasopressors have been used since the 1940s, few controlled clinical trials have directly compared these agents or documented improved outcomes due to their use [1]. Thus, the manner in which these agents are commonly used largely reflects expert opinion, animal data, and the use of surrogate end points, such as tissue oxygenation, as a proxy for decreased morbidity and mortality.

Basic adrenergic receptor physiology and the principles, complications, and controversies surrounding use of vasopressors and inotropes for treatment of shock are presented here. Issues related to the differential diagnosis of shock and the use of vasopressors in patients with septic shock are discussed separately. (See "Definition, classification, etiology, and pathophysiology of shock in adults" and "Evaluation and management of suspected sepsis and septic shock in adults".)

PHYSIOLOGIC MECHANISMS OF VASOCONSTRICTION — 

The main categories of adrenergic receptors relevant to vasopressor activity are the alpha-1, beta-1, and beta-2 adrenergic receptors, as well as the dopamine receptors [2,3].

Alpha adrenergic — Activation of alpha-1 adrenergic receptors, located in vascular walls, induces significant vasoconstriction. Alpha-1 adrenergic receptors are also present in the heart and can increase the duration of contraction without increased chronotropy. However, the clinical significance of this phenomenon is unclear [4].

Beta adrenergic — Beta-1 adrenergic receptors are most common in the heart and mediate increases in inotropy and chronotropy with minimal vasoconstriction. Stimulation of beta-2 adrenergic receptors in blood vessels induces vasodilation.

Dopamine — Dopamine receptors are present in the kidney, splanchnic (mesenteric), coronary, and cerebral vascular beds; stimulation of these receptors leads to vasodilation. A second subtype of dopamine receptors causes vasoconstriction by inducing norepinephrine release.

Calcium sensitizers — Some agents increase the sensitivity of the myocardial contractile apparatus to calcium, causing an increase in myofilament tension development and myocardial contractility (eg, pimobendan, levosimendan). These agents have additional pharmacologic properties, such as phosphodiesterase inhibition, which may increase inotropy and vasodilation and contribute significantly to their clinical profile, the details of which are discussed separately. (See "Intraoperative use of vasoactive agents", section on 'Inodilators'.)

Angiotensin — Angiotensin receptors (AT₁ and AT₂) are G-coupled protein receptors with angiotensin II as their ligand. Angiotensin II is a vasoconstrictor that is part of the renin-aldosterone-angiotensin (RAAS) system. When receptors are stimulated, cytosolic calcium concentration increases to mediate vasoconstrictive effects as well as aldosterone and vasopressin secretion [5].

PRINCIPLES — 

Hypotension may result from hypovolemia (eg, exsanguination), pump failure (eg, severe medically refractory heart failure or shock complicating myocardial infarction), or a pathologic maldistribution of blood flow (eg, septic shock, anaphylaxis). (See "Definition, classification, etiology, and pathophysiology of shock in adults" and "Inotropic agents in heart failure with reduced ejection fraction".)

Vasopressors are indicated for a decrease of >30 mmHg from baseline systolic blood pressure or a mean arterial pressure (MAP) <60 mmHg when either condition results in end-organ dysfunction due to hypoperfusion [6]. Hypovolemia should be corrected prior to the institution of vasopressor therapy [7]. (See "Treatment of severe hypovolemia or hypovolemic shock in adults".)

The rational use of vasopressors and inotropes is guided by three fundamental concepts:

●One drug, many receptors – A given drug often has multiple effects because of its actions upon more than one receptor. As an example, dobutamine increases cardiac output by beta-1 adrenergic receptor activation; however, it also acts upon beta-2 adrenergic receptors and thus induces vasodilation and can cause hypotension.

●Dose-response curve – Many agents have dose-response curves, such that the primary adrenergic receptor subtype activated by the drug is dose dependent. As an example, dopamine stimulates beta-1 adrenergic receptors at doses of 2 to 10 mcg/kg per minute and alpha-adrenergic receptors when doses exceed 10 mcg/kg per minute.

●Direct versus reflex actions – A given agent can affect MAP both by direct actions on adrenergic receptors and by reflex actions triggered by the pharmacologic response. Norepinephrine-induced beta-1 adrenergic stimulation alone normally would cause tachycardia. However, the elevated MAP from norepinephrine's alpha-adrenergic receptor-induced vasoconstriction results in a reflex decrease in heart rate. The net result may be a stable or slightly reduced heart rate when the drug is used.

ADMINISTRATION AND VASCULAR ACCESS

Options for initiating vasopressor therapy — Many options are available for intravenous access including peripheral, midline, and central venous access. All vascular access devices carry some risk of infection, thrombosis, phlebitis, and vascular injury [8].

Concern about the potential for extravasation and tissue injury with administration of vasopressors through peripheral intravenous catheters (PIVs) has led historically to vasopressor administration via central venous catheters (CVCs). However, CVCs are associated with high rates of major complications, including immediate insertion-related complications (eg, pneumothorax, arterial cannulation), central line-associated bloodstream infections (CLABSIs), and deep vein thrombosis (DVT) [9]. A systematic review and meta-analysis of 130 studies with 214,325 catheters found that the rate of major complications from placement of a CVC for three days is approximately 3 percent [10]. Since many patients requiring vasopressors for septic and other types of shock do not require high doses or long duration of therapy [11], CVC placement for all vasopressor administration may introduce unnecessary risk for a substantial proportion of patients. Moreover, CVC placement requires time and expertise, which may delay vasopressor initiation [12]. Amid evolving practice guidelines that favor smaller-volume resuscitation, evidence suggests that earlier vasopressor initiation may improve outcomes in septic shock [13]. These factors have contributed to more widespread acceptance of vasopressor administration via PIVs or midline intravascular catheters. Guidelines suggest initiating vasopressors peripherally rather than delaying initiation until central access is obtained [14].

It is possible that patients receiving vasopressors before intensive care unit (ICU) admission have longer durations of vasopressor therapy or higher maximum doses.

Data supporting the peripheral delivery of vasopressors include the following:

●In a randomized study of adults with sepsis and hypotension, median time from emergency department arrival to norepinephrine administration was significantly shorter (93 versus 192 minutes) in patients receiving early norepinephrine by PIV compared with standard access (usually CVC) [15].

●Similarly, another trial found that peripheral initiation was associated with faster time to vasopressor delivery compared with central delivery (median 4.2 versus 6.3 hours) [16]. Complication rates were similar.

●Several systematic reviews and meta-analyses have reported pooled incidence of extravasation below 4 percent, with no episodes of limb ischemia or tissue necrosis [17-20]. Despite these data, in our experience, limb ischemia or tissue necrosis can occur in isolated cases.

●Data suggest that PIVs are significantly less likely to have infectious complications than CVCs or peripherally inserted central catheters (PICCs) [21-23], and although extravasation and phlebitis are relatively common, complications associated with PIV use are usually self-limited, even when using vasopressors, particularly when protocols are in place to ensure appropriate monitoring of PIV sites by nurses.

The duration of safe use for PIV administration of vasopressors varies. In an international survey of ICUs, PIVs were used to deliver norepinephrine in 86 percent of institutions [24] for a median of 24 hours (range is up to 72 hours) [25]. In one study, 50 percent of patients never required CVC insertion after implementation of a PIV norepinephrine delivery protocol [25], suggesting that peripheral administration of vasopressors is safe.

Despite increasing evidence that peripheral lines are a safe option, hospital policies remain mixed with regard to criteria for peripheral vasopressor initiation and often include recommendations for limited duration of peripheral administration [26]. In our academic tertiary care centers, standard practice for initial vasopressor delivery includes the following:

●For patients with a CVC, PICC, or midline catheter in situ, our practice is to use these catheters for vasopressor delivery.

●For patients without central venous or midline access, our practice is to start vasopressors peripherally to restore mean arterial pressure (MAP) [14] and continue PIV administration for up to 72 hours, up to a maximum dose of 0.3 mcg/kg per minute (approximately 20 mcg/minute) norepinephrine equivalent. We do not deliver vasopressin via PIV since there is no local antidote. (See 'Management of extravasation' below.)

●Patients with difficult vascular access and immediate need for vasopressors may benefit from intraosseous catheter insertion. (See "Intraosseous infusion".)

Vascular access for patients requiring high-dose therapy or vasopressin — PICCs and long PIVs (ie, midline catheters) are other vascular access options with growing use in the inpatient setting [27-29]. PICCs are convenient in that they can be placed at the bedside to provide extended venous access and may be associated with a lower risk of CLABSI compared with CVCs [30]. However, the popularity of PICCs has led to their use even in patients with reliable peripheral access and no need for centrally administered medication, and their long-term use can be associated with the same infectious and thrombotic complications as CVCs [30,31].

Midline catheters are considered peripheral venous catheters, inserted near the antecubital fossa and terminating in larger peripheral veins. With dwell times of two to four weeks, they offer the option for longer infusions and are associated with a lower risk of decannulation and extravasation compared with traditional PIVs [32] and lower risk of CLABSI than PICCs [30,33] or CVCs [34]. A study of 248 patients requiring prolonged vasopressors (average duration eight days) confirmed the efficacy of this approach, noting only one extravasation event with no incidents of tissue injury or limb ischemia [32]. Approximately half of patients received vasopressin via midline catheters. Another retrospective review of 203 patients receiving vasopressors through midline catheters for an average of 32 hours found no incidents of extravasation [35].

Several studies have compared PICCs with midline catheters. One retrospective study reported a lower frequency of catheter-related complications in patients with midline catheters used for vasopressor administration compared with PICCs (5 versus 13 percent) [36]. Another retrospective study found that midline catheters were associated with a lower risk of CLABSI when compared with PICCs (0.4 versus 1.6 percent) and occlusion (2 versus 7 percent), but the risk of DVT was similar between catheter types [33]. In time to event models, the risk of DVT events was lower in patients who received PICCs compared with midline catheters (hazard ratio 0.53, 95% CI 0.38-0.74) [36].

Further details regarding device and site selection are provided separately. (See "Central venous access: Device and site selection in adults".)

OTHER PRACTICAL ISSUES — 

Use of vasopressors and inotropic agents requires attention to a number of issues.

Volume resuscitation — Adequate intravascular volume resuscitation, when time permits, is typical prior to the initiation of vasopressors since vasopressors are most effective when volume has been repleted [37]. However, the timing of administration of vasopressors in hemorrhagic shock is controversial and variable. Further details are provided separately. (See "Initial management of moderate to severe hemorrhage in the adult trauma patient", section on 'Role of vasoactive medications'.)

Fluids may be withheld in patients with significant pulmonary edema due to acute respiratory distress syndrome or heart failure. In patients with a pulmonary artery catheter, pulmonary capillary wedge pressures of 18 to 24 mmHg are recommended for cardiogenic shock [38] and 12 to 14 mmHg for septic or hypovolemic shock [39]. (See "Pulmonary artery catheterization: Interpretation of hemodynamic values and waveforms in adults".)

Selection and titration — Choice of an initial agent should be based upon the suspected underlying etiology of shock (eg, dobutamine in cases of cardiac failure without significant hypotension, epinephrine for anaphylactic shock). The dose should be titrated up to achieve effective blood pressure or end-organ perfusion as evidenced by such criteria as urine output or mentation. If maximal doses of a first agent are inadequate, then a second drug should be added to the first. In situations where this is ineffective, such as refractory septic shock, anecdotal reports describe adding a third agent, although no controlled trials have demonstrated the utility of this approach.

Tachyphylaxis — Responsiveness to these drugs can decrease over time due to tachyphylaxis. Doses must be constantly titrated to adjust for this phenomenon and for changes in the patient's clinical condition [40,41].

Hemodynamic effects — Mean arterial pressure (MAP) is influenced by systemic vascular resistance (SVR) and cardiac output (CO). In situations such as cardiogenic shock, elevating SVR increases afterload and the work of an already failing heart, thus potentially lowering CO. Some authors recommend keeping the SVR approximately 700 to 1000 dynes/-sec/cm⁵ to avoid excessive afterload and to minimize complications from profound vasoconstriction (calculator 1) [42]. However, there is no consensus regarding an ideal target cardiac index. Studies that have attempted to maintain a supraphysiologic cardiac index of >4 to 4.5 L/minute per m² have not shown consistent benefit [43,44]. (See "Oxygen delivery and consumption".)

Subcutaneous delivery of medications — Critically ill patients often receive subcutaneously injected medications, such as heparin and insulin. The bioavailability of these medications can be reduced during treatment with vasopressors due to cutaneous vasoconstriction.

This was demonstrated in a study that monitored plasma factor Xa levels in three groups of hospitalized patients following the initiation of prophylactic low molecular weight (LMW) heparin [45]. Patients who required vasopressor support (dopamine >10 mcg/kg per minute, norepinephrine >0.25 mcg/kg per minute, or phenylephrine >2 mcg/kg per minute) had decreased factor Xa activity compared with both intensive care unit (ICU) patients who did not require vasopressors and routine postoperative control patients. The clinical significance of the decrease in plasma factor Xa levels was not determined.

The authors of the study suggested that patients might need higher doses of LMW heparin to attain adequate thrombosis prophylaxis. Another approach is to change subcutaneous medications to an intravenous form whenever a patient is receiving vasopressor therapy.

Frequent re-evaluation — Critically ill patients may undergo a second hemodynamic insult, which necessitates a change in vasopressor or inotrope management. The dose of a given agent should not simply be increased because of persistent or worsening hypotension without reconsideration of the patient's clinical situation and the appropriateness of the current strategy.

Acidosis — Data suggest that vasopressors including vasopressin and catecholamine vasopressors are less effective in patients with severe acidosis [46]. However, data are inconsistent [47]. In addition, the ideal cutoff for the administration of bicarbonate is unclear.

ADRENERGIC AGENTS — 

Adrenergic agents, such as phenylephrine, norepinephrine, dopamine, and dobutamine, are the most commonly used vasopressor and inotropic drugs in critically ill patients (table 1). These agents manifest different receptor selectivity and clinical effects (table 2).

Norepinephrine — Norepinephrine (Levophed [brand name]) acts on both alpha-1 and beta-1 adrenergic receptors, thus producing potent vasoconstriction as well as a modest increase in cardiac output (CO) [7]. A reflex bradycardia usually occurs in response to the increased mean arterial pressure (MAP), such that the mild chronotropic effect is canceled out and the heart rate remains unchanged or even decreases slightly. Norepinephrine is the preferred vasopressor for the treatment of septic shock. (See 'Choice of agent in septic shock' below and "Evaluation and management of suspected sepsis and septic shock in adults".)

Phenylephrine — Phenylephrine (Neo-Synephrine [brand name]) has purely alpha-adrenergic agonist activity and therefore results in vasoconstriction with minimal cardiac inotropy or chronotropy. MAP is augmented by raising systemic vascular resistance (SVR) [48]. The drug is useful in the setting of hypotension with an SVR <700 dynes/-sec/cm⁵ (eg, hyperdynamic sepsis, neurologic disorders, anesthesia-induced hypotension). A potential disadvantage of phenylephrine is that it may decrease stroke volume, so it is reserved for patients in whom norepinephrine is contraindicated due to arrhythmias or who have failed other therapies.

Although SVR elevation increases cardiac afterload, most studies document that CO is either maintained or actually increased among patients without pre-existing cardiac dysfunction [4,49]. The drug is contraindicated if the SVR is >1200 dynes/-sec/cm⁵.

Epinephrine — Epinephrine (Adrenalin [brand name]) has potent beta-1 adrenergic receptor activity and moderate beta-2 and alpha-1 adrenergic receptor effects. Clinically, low doses of epinephrine increase CO because of the beta-1 adrenergic receptor inotropic and chronotropic effects, while the alpha-adrenergic receptor-induced vasoconstriction is often offset by the beta-2 adrenergic receptor vasodilation. The result is an increased CO, with decreased SVR and variable effects on the MAP [3]. However, at higher epinephrine doses, the alpha-adrenergic receptor effect predominates, producing increased SVR in addition to an increased CO. Epinephrine is most often used for the treatment of anaphylaxis as a second-line agent (after norepinephrine) in septic shock and for management of hypotension following coronary artery bypass grafting.

Other disadvantages of epinephrine include dysrhythmias (due to beta-1 adrenergic receptor stimulation) and splanchnic vasoconstriction. The degree of splanchnic vasoconstriction appears to be greater with epinephrine than with equipotent doses of norepinephrine or dopamine in patients with severe shock [50], although the clinical importance of this is unclear.

Ephedrine — Similar to epinephrine, ephedrine acts primarily on alpha- and beta-adrenergic receptors but with less potency. It also has an effect by leading to the release of endogenous norepinephrine. Ephedrine is rarely used except in the setting of post-anesthesia-induced hypotension.

Dopamine — Dopamine has a variety of effects depending upon the dose range administered. It is most often used as a second-line alternative to norepinephrine in patients with absolute or relative bradycardia and a low risk of tachyarrhythmias. Weight-based administration of dopamine can achieve quite different serum drug concentrations in different individuals [51], but the following provides an approximate description of effects:

●At doses of 1 to 2 mcg/kg per minute, dopamine acts predominantly on dopamine-1 receptors in the kidney, mesenteric, cerebral, and coronary beds, resulting in selective vasodilation. Some reports suggest that dopamine increases urine output by augmenting kidney blood flow and glomerular filtration rate and natriuresis by inhibiting aldosterone and kidney tubular sodium transport [52-54]. These effects may be blunted by haloperidol and other butyrophenones [54]. However, the clinical significance of these phenomena is unclear, and some patients may develop hypotension at these low doses [55]. (See "Renal actions of dopamine".)

●At 5 to 10 mcg/kg per minute, dopamine also stimulates beta-1 adrenergic receptors and increases CO, predominantly by increasing stroke volume with variable effects on heart rate [56]. Doses between 2 and 5 mcg/kg per minute have variable effects on hemodynamics in individual patients: vasodilation is often balanced by increased stroke volume, producing little net effect upon systemic blood pressure. Some mild alpha-adrenergic receptor activation increases SVR, and the sum of these effects is an increase in MAP.

●At doses >10 mcg/kg per minute, the predominant effect of dopamine is to stimulate alpha-adrenergic receptors and produce vasoconstriction with an increased SVR [56,57]. However, the overall alpha-adrenergic receptor effect of dopamine is weaker than that of norepinephrine, and the beta-1 adrenergic receptor stimulation of dopamine at doses >2 mcg/kg per minute can result in dose-limiting dysrhythmias.

In practical terms, the dose-dependent effects of dopamine mean that changing the dose of the drug is akin to switching vasopressors. Conversely, simply increasing the dose of dopamine without being cognizant of the different receptor populations activated can cause untoward results.

The usual dose range for dopamine is 2 to 20 mcg/kg per minute, although doses as high as 130 mcg/kg per minute have been employed [58]. When used for cardiac failure, dopamine should be started at 2 mcg/kg per minute and then titrated to a desired physiologic effect rather than depending on the predicted pharmacologic ranges described above.

Dobutamine — Dobutamine is not a vasopressor but rather is an inotrope that causes vasodilation. Dobutamine's predominant beta-1 adrenergic receptor effect increases inotropy and chronotropy and reduces left ventricular filling pressure. In patients with heart failure, this results in a reduction in cardiac sympathetic activity [59]. However, minimal alpha- and beta-2 adrenergic receptor effects result in overall vasodilation, complemented by reflex vasodilation to the increased CO. The net effect is increased CO, with decreased SVR with or without a small reduction in blood pressure.

Dobutamine is most frequently used in severe, medically refractory heart failure and cardiogenic shock and should not be routinely used in sepsis because of the risk of hypotension. Dobutamine does not selectively vasodilate the kidney vascular bed as dopamine does at low doses. (See "Inotropic agents in heart failure with reduced ejection fraction".)

Isoproterenol — Isoproterenol also is primarily an inotropic and chronotropic agent rather than a vasopressor. It acts upon beta-1 adrenergic receptors and unlike dobutamine, has a prominent chronotropic effect. The drug's high affinity for the beta-2 adrenergic receptor causes vasodilation and a decrease in MAP. Therefore, its utility in hypotensive patients is limited to situations in which hypotension results from bradycardia.

Midodrine — Midodrine is an enteric agent that forms an active metabolite, desglymidodrine, which is an alpha-1 agonist. It has potential use for hypotension in patients with cirrhosis, hepatorenal syndrome orthostasis, and vasovagal syncope. In the intensive care unit (ICU) it has been proposed as a vasopressor sparing agent. However, data are limited. One meta-analysis identified seven randomized trials and 10 observational studies investigating adjunctive midodrine as a vasopressor-sparing agent in patients with shock [60]. Midodrine may reduce ICU length of stay (mean difference [MD] 1.01 days less, 95% CI 2.23 days less to 0.22 days more) and intravenous vasopressor support duration (MD 19.66 hours less, 95% CI 37.87-1.46 hours less), although sample size was small and data were imprecise, limiting confidence in the outcomes. While bradycardia was the most commonly reported adverse event, there did not appear to be a difference with control groups. Further study is needed before midodrine can be used routinely for this indication.

Contraindications and interactions — Several conditions or medications require avoidance of specific agents:

●In patients with cardiogenic shock, norepinephrine is preferred over dopamine as the first-line vasopressor because a subgroup analysis from a randomized trial found that patients with cardiogenic shock who received dopamine had a higher mortality than those who received norepinephrine [61]. In addition, dysrhythmias were more common in the dopamine group.

●Patients with pheochromocytoma are at risk of excessive autonomic stimulation from adrenergic vasopressors.

●Dobutamine is contraindicated in the setting of idiopathic hypertrophic subaortic stenosis.

●Patients receiving monoamine oxidase inhibitors are extremely sensitive to vasopressors and therefore require much lower doses.

VASOPRESSIN AND ANALOGS — 

Vasopressin (antidiuretic hormone) is used in the management of arginine vasopressin disorders (formerly known as diabetes insipidus) and esophageal variceal bleeding; however, it may also be helpful in the management of vasodilatory shock (table 1). Although its precise role in vasodilatory shock remains to be defined, it is primarily used as a second-line agent in refractory vasodilatory shock, particularly septic shock or anaphylaxis that is unresponsive to epinephrine [62-68]. It is also used occasionally to reduce the dose of the first-line agent. Terlipressin, a vasopressin analog, has also been assessed in patients with vasodilatory shock [69-74].

The effects of vasopressin and terlipressin in vasodilatory shock (mostly septic shock) were evaluated in a systematic review that identified 10 relevant randomized trials (1134 patients) [75]. A meta-analysis of six of the trials (512 patients) compared vasopressin or terlipressin with placebo or supportive care. There was no significant improvement in short-term mortality among patients who received either vasopressin or terlipressin (40.2 versus 42.9 percent; relative risk 0.91, 95% CI 0.79-1.05). However, patients who received vasopressin or terlipressin required less norepinephrine. A second randomized trial compared vasopressin with norepinephrine in 409 patients with septic shock. Although vasopressin did not improve mortality or the number of kidney failure-free days, it may have been associated with a reduction in the rate of kidney failure requiring kidney replacement therapy (25 versus 35 percent) [76]. Further studies are needed before vasopressin can replace norepinephrine as the first-choice agent for those with septic shock. (See 'Choice of agent in septic shock' below.)

The effects of vasopressin are dose dependent. A randomized trial compared two doses of vasopressin (0.0333 versus 0.067 international units/minute) in 50 patients with vasodilatory shock who required vasopressin as a second pressor agent [77]. The higher dose was more effective at increasing the blood pressure without increasing the frequency of adverse effects in these patients. However, doses of vasopressin above 0.04 units/minute have been associated with coronary and mesenteric ischemia and skin necrosis in some studies [78-81], although some of these studies were in animals and necrosis in humans may also have been due to coexisting conditions (eg, disseminated intravascular coagulation). However, doses higher than the therapeutic range (0.04 units/minute) are generally avoided for this reason unless an adequate mean arterial pressure (MAP) cannot be attained with other vasopressor agents. When weaning, we generally attempt to wean off vasopressin before norepinephrine.

Based upon clinical experience, hypotension appears to be common following withdrawal of vasopressin. To avoid this, the dose can be slowly tapered by 0.01 units/minute every 30 to 60 minutes [82,83]. Data describing use of midodrine to facilitate weaning off intravenous vasopressin are conflicting [84,85].

Other potential adverse effects of vasopressin include hyponatremia, pulmonary vasoconstriction, and arginine vasopressin disorders [14,78-81]. Terlipressin appears to have a similar side effect profile to vasopressin. In a meta-analysis of four trials (431 patients) that was conducted as part of the systematic review described above, there was no significant difference in the frequency of adverse events among patients who received either vasopressin or terlipressin (10.6 versus 11.8 percent; relative risk 0.9, 95% CI 0.49-1.67) [75].

NONADRENERGIC AGENTS — 

A number of agents produce vasoconstriction or inotropy through nonadrenergic mechanisms, including phosphodiesterase (PDE) inhibitors and nitric oxide synthase (NOS) inhibitors, calcium sensitizers, or angiotensin II.

PDE inhibitors — Phosphodiesterase (PDE) inhibitors, such as inamrinone (formerly known as amrinone) and milrinone, are nonadrenergic drugs with inotropic and vasodilatory actions. In many ways, their effects are similar to those of dobutamine but with a lower incidence of dysrhythmias. PDE inhibitors most often are used to treat patients with impaired cardiac function and medically refractory heart failure, but their vasodilatory properties limit their use in hypotensive patients [56]. Inamrinone is no longer available in North America and most other countries, so the only intravenous PDE inhibitor used in the United States is milrinone. (See "Inotropic agents in heart failure with reduced ejection fraction".)

NOS inhibitors — Nitric oxide overproduction appears to play a major role in vasodilation induced by sepsis (see "Pathophysiology of sepsis"). Studies of nitric oxide synthase (NOS) inhibitors, such as N-monomethyl-L-arginine (L-NMMA) in sepsis, demonstrate a dose-dependent increase in systemic vascular resistance (SVR) [86]. However, cardiac index and heart rate decrease, even when patients are treated concomitantly with norepinephrine or epinephrine. The increase in SVR tends to be offset by the drop in cardiac index, such that mean arterial pressure (MAP) is only minimally augmented. The clinical utility of this class of drugs remains unproven.

Calcium sensitizers — Several agents increase myocardial contractility (eg, pimobendan, levosimendan), but conclusive evidence of improved outcomes with their use is lacking [87,88].

Angiotensin II — Preliminary trials have reported an adequate vasopressor effect when synthetic angiotensin II is exogenously administered for vasodilatory shock (eg, septic shock) [89,90]. Best supporting its role as a vasopressor agent is a randomized trial of 344 patients with vasodilatory shock (80 percent had sepsis and patients had a normal cardiac output) already receiving high-dose norepinephrine or a similar vasopressor equivalent [91]. When compared with placebo, at three hours, angiotensin II resulted in a greater proportion of patients achieving a ≥10 mmHg increase in the baseline MAP or an increase in the MAP to ≥75 mmHg (70 versus 23 percent). There was no effect on mortality or on the sequential organ failure assessment score. Serious adverse events were no different among the groups. Angiotensin II resulted in less background vasopressor use and higher heart rates but did not result in life-threatening tachyarrhythmias.

In another retrospective observational study of 50 patients with cardiac or respiratory failure requiring treatment with mechanical circulatory support devices who had vasodilatory shock, angiotensin II increased the MAP and allowed a reduction in the dose of other vasopressors without altering cardiac index or mean pulmonary artery pressure [92].

Further trials will be required to compare angiotensin II with other vasopressor agents and to examine its effects in populations other than those with vasodilatory shock before it can be routinely used as a second-line agent for the treatment of shock.

COMPLICATIONS — 

Vasopressors and inotropic agents have the potential to cause a number of significant complications, including hypoperfusion, dysrhythmias, myocardial ischemia, local effects, and hyperglycemia. In addition, a number of drug interactions exist.

Management of extravasation — Peripheral extravasation of vasopressors into the surrounding connective tissue can lead to excessive local vasoconstriction with subsequent skin necrosis. If infiltration occurs, local treatment with the alpha-adrenergic antagonist phentolamine (5 to 10 mg in 10 mL of normal saline) injected subcutaneously (either using the catheter through which vasopressors were infusing if not yet removed from the patient or through separate subcutaneous injection) can minimize local vasoconstriction [93]. Phentolamine will not be effective for extravasation of vasopressin or other nonadrenergic agents since these medications act via nonadrenergic receptors. Nitroglycerin (2.5 cm of 2 percent paste/ointment for 12 hours) is an alternative that can be applied to the area of extravasation if phentolamine is not available or ineffective.

Hypoperfusion — Excessive vasoconstriction in response to hypotension and vasopressors can produce inadequate perfusion of the extremities, mesenteric organs, or kidneys. Excessive vasoconstriction with inadequate perfusion, usually with a systemic vascular resistance (SVR) >1300 dynes/-sec/cm⁵, commonly occurs in the setting of inadequate cardiac output or inadequate volume resuscitation.

The initial findings are dusky skin changes at the tips of the fingers and/or toes, which may progress to frank necrosis with autoamputation of the digits. Compromise of the kidney vascular bed may produce kidney function impairment and oliguria, while patients with underlying peripheral artery disease may develop acute limb ischemia.

Inadequate mesenteric perfusion increases the risk of gastritis, shock liver, intestinal ischemia, or translocation of gut flora with resultant bacteremia. Despite these concerns, maintenance of mean arterial pressure (MAP) with vasopressors appears more effective in maintaining kidney and mesenteric blood flow than allowing the MAP to drop, and maintenance of MAP with vasopressors may be life-saving despite evidence of localized hypoperfusion [48,94].

Dysrhythmias — Many vasopressors and inotropes exert powerful chronotropic effects via stimulation of beta-1 adrenergic receptors. This increases the risk of sinus tachycardia (most common), atrial fibrillation (potentially with increased atrioventricular nodal [A-V] conduction and therefore an increased ventricular response), reentrant atrioventricular node tachycardia, or ventricular tachyarrhythmias.

Adequate volume loading may minimize the frequency or severity of dysrhythmias. Despite this, dysrhythmias often limit the dose and necessitate switching to another agent with less prominent beta-1 effects. The degree to which the agent affects the frequency of dysrhythmias was illustrated by a randomized trial of 1679 patients with shock [61]. Dysrhythmias were significantly more common among patients who received dopamine than among those who received norepinephrine (24.1 versus 12.4 percent).

Myocardial ischemia — The chronotropic and inotropic effects of beta-adrenergic receptor stimulation can increase myocardial oxygen consumption. While there is usually coronary vasodilation in response to vasopressors [95], perfusion may still be inadequate to accommodate the increased myocardial oxygen demand. Daily electrocardiograms on patients treated with vasopressors or inotropes may screen for occult ischemia, and excessive tachycardia should be avoided because of impaired diastolic filling of the coronary arteries.

Hyperglycemia — Hyperglycemia may occur due to the inhibition of insulin secretion. The magnitude of hyperglycemia generally is minor and is more pronounced with norepinephrine and epinephrine than dopamine [53]. Monitoring of blood glucose while on vasopressors can prevent complications of untreated hyperglycemia.

CONTROVERSIES — 

Several controversies exist regarding the use of vasopressors and inotropic agents in critically ill patients. Most stem from the relative paucity of large-scale studies comparing similar patient populations treated with different regimens. The development of clear definitions for the systemic inflammatory response syndrome, sepsis, and septic shock is a step forward toward comparative trials among standardized patient populations. (See "Sepsis syndromes in adults: Epidemiology, definitions, clinical presentation, diagnosis, and prognosis".)

Choice of agent in septic shock — The optimal agent in patients with septic shock is unknown, and practice varies considerably among experts. However, based upon meta-analyses of small randomized trials and observational studies, a paradigm shift in practice has occurred such that most experts prefer to avoid dopamine in this population and favor norepinephrine as the first-choice agent, the details of which are discussed separately. (See "Evaluation and management of suspected sepsis and septic shock in adults", section on 'Vasopressors'.)

"Renal dose" dopamine — Dopamine selectively increases kidney blood flow when administered to normal volunteers at 1 to 3 mcg/kg per minute [96,97]. Animal studies also suggest that low-dose dopamine in the setting of vasopressor-dependent sepsis helps preserve kidney blood flow [98]. (See "Renal actions of dopamine" and "Possible prevention and therapy of ischemic acute tubular necrosis".)

However, a beneficial effect of low or "renal dose" dopamine is less proven in human patients with sepsis or other critical illness. Critically ill patients who do not have evidence of kidney function impairment or decreased urine output will develop a diuresis in response to dopamine at 2 to 3 mcg/kg per minute, with variable effects on creatinine clearance, but the benefit of this diuresis is questionable [40,55]. The intervention is not entirely benign, because hypotension and tachycardia may ensue. One small study demonstrated that the addition of low-dose dopamine to patients receiving other vasopressors increases splanchnic blood flow but does not alter other indices of mesenteric perfusion, such as gastric intramucosal pH (pHi) [99].

There are no data to support the routine use of low-dose dopamine to prevent or treat acute kidney failure or mesenteric ischemia. In general, the most effective means of protecting the kidneys in the setting of septic shock appears to be the maintenance of mean arterial pressure (MAP) >60 mmHg while attempting to avoid excessive vasoconstriction (ie, the systemic vascular resistance should not exceed 1300 dynes/-sec/cm⁵) [37,42,100,101].

Optimal dose — Optimal dosing is poorly defined. Most experts use a physiologic target such as MAP and signs of adequate organ perfusion (eg, urine output, capillary refill time, mental status). Adequate targets are discussed separately. (See "Evaluation and management of suspected sepsis and septic shock in adults", section on 'Monitor response'.)

Although norepinephrine doses of up to 100 mcg/minute in adults are typically used in refractory shock and as salvage therapy in combination with a second vasopressor agent, there is no well-established maximum dose. Several studies have suggested improved tissue perfusion when higher doses of norepinephrine (up to 350 mcg/minute) are used [42,101], but no survival benefit of high-dose norepinephrine has been conclusively proven.

Supranormal cardiac index — Elevation of the cardiac index with inotropic agents to supranormal values (ie, >4.5 L/minute per m²) potentially increases oxygen delivery to peripheral tissues. In theory, increased oxygen delivery may prevent tissue hypoxia and improve outcomes, and initial studies appeared to support this hypothesis [102-104]. However, later larger trials showed that goal-oriented hemodynamic therapy to increase either cardiac index to >4.5 L/minute per m² or oxygen delivery to >600 to 650 mL/minute per m² with volume expansion or dobutamine resulted in either no improvement or worsened morbidity or mortality [43,44,105]. Therefore, the routine administration of vasopressors or inotropes to improve cardiac output or oxygen delivery to supranormal levels is not advocated. (See "Oxygen delivery and consumption".)

The American Thoracic Society (ATS) statement on the detection, correction, and prevention of tissue hypoxia, as well as other ATS guidelines can be accessed through the ATS website.

SOCIETY GUIDELINE LINKS — 

Links to society and government-sponsored guidelines from selected countries and regions around the world are provided separately. (See "Society guideline links: Sepsis in children and adults".)

SUMMARY AND RECOMMENDATIONS

●Definition – Vasopressors are a powerful class of drugs that induce vasoconstriction and elevate mean arterial pressure (MAP). (See 'Introduction' above.)

●Physiology – Alpha-1 adrenergic receptors induce vasoconstriction while beta-1 receptors induce inotropy plus chronotropy, and beta-2 receptors induce vasodilation (table 2). One subtype of dopamine receptor induces norepinephrine release with subsequent vasoconstriction, although many dopamine receptors induce vasodilation. Some agents increase the sensitivity of the myocardial contractile apparatus to calcium while others cause direct vasoconstriction via angiotensin II receptor stimulation. (See 'Physiologic mechanisms of vasoconstriction' above.)

●Vasopressor administration – Delivering vasopressors peripherally or through midline catheters is reasonable at centers with policies in place and appropriate monitoring protocols. Our practice is to deliver up to 0.3 mcg/kg per minute (approximately 20 mcg/minute) norepinephrine or equivalent through a peripheral intravenous catheter (PIV) for up to 72 hours. Midline catheters can be used for higher vasopressor doses. We manage extravasation with phentolamine and avoid PIV delivery of vasopressin since there is no antidote for local extravasation. (See 'Administration and vascular access' above.)

●Principles – Vasopressors are indicated for a MAP <60 mmHg or a decrease of systolic blood pressure that exceeds 30 mmHg from baseline when either condition results in end-organ dysfunction due to hypoperfusion. (See 'Principles' above.)

●Other practical issues – Hypovolemia should be corrected prior to the institution of vasopressor therapy for maximum efficacy. Patients should be reevaluated frequently once vasopressor therapy has been initiated. Common issues that arise include tachyphylaxis, which may require dose titration, and additional hemodynamic insults, which should be recognized and managed. (See 'Other practical issues' above.)

●Choosing an agent – Choice of an initial agent should be based upon the suspected underlying etiology of shock (eg, dobutamine for cardiogenic shock without significant hypotension, norepinephrine for septic and cardiogenic shock with hypotension, epinephrine for anaphylactic shock) (table 1). (See 'Other practical issues' above and "Treatment and prognosis of cardiogenic shock complicating acute myocardial infarction", section on 'Hypotension'.)

●Complications – Complications of vasopressor therapy include hypoperfusion (particularly affecting the extremities, mesentery, or kidneys), dysrhythmias, myocardial ischemia, peripheral extravasation with skin necrosis, and hyperglycemia. (See 'Complications' above.)