The present review initially describes the rationale for the use of non-adrenergic vasopressors in the treatment of distributive shock and then provides an overview of the individual vasopressin-receptor agonists, namely arginine vasopressin, terlipressin, and selepressin. Following a brief summary of their current use in clinical practice, the present review focuses on the influence of vasopressin-receptor agonists on macro- and microvascular coupling, also referred to as hemodynamic coherence. On the basis of the current evidence from experimental and clinical studies, vasopressin-receptor agonists do not negatively influence macro- and microvascular coupling as compared to the standard therapy with norepinephrine, when used in established treatment regimes. A higher selectivity for the V 1a -receptor seems to be beneficial; however, future clinical trials are warranted to verify this assumption. Notably, the optimal treatment regime for non-adrenergic vasopressors with respect to compound, dose, and timing still needs to be defined.
Introduction
Macro- and microvascular coupling, also referred to as hemodynamic coherence , describes a physiological condition in which macrohemodynamic values (e.g. mean arterial pressure (MAP) and cardiac output) reflect microvascular perfusion. In shock states, this coupling may be disturbed, resulting in deteriorated microvascular perfusion despite optimized macrohemodynamics (hemodynamic incoherence). This correlation between macro- and microcirculation might be influenced not only by the underlying pathophysiology, but also by the respective treatment applied. In addition to fluid resuscitation, vasoactive drugs represent the mainstay of therapy in patients suffering from shock. In particular, vasopressors are used to sustain a (presumably) sufficient MAP required for the perfusion of vital organs. The catecholamine norepinephrine currently represents the standard of care . However, there is increasing evidence that adrenergic compounds cause direct organ damage and are associated with numerous negative effects on the immune, metabolic, and coagulation system, thereby negatively influencing patients’ outcome . Furthermore, in some patients, hemodynamics deteriorate despite high doses of catecholamines, a condition described as catecholamine resistance. Therefore, alternative treatment strategies with non-adrenergic vasopressors such as vasopressin-receptor agonists, methylene blue, and angiotensin II are of increasing interest in shock management. The present article will focus on the different vasopressin-receptor agonists and their influence on macro- and microcirculation in distributive shock states. In addition, potential advantages and pitfalls compared to the standard treatment with norepinephrine will be discussed.
Pathophysiology of distributive shock
Shock is defined as an inadequate oxygen supply to vital organs due to circulatory failure combined with an impaired oxygen utilization in the tissues . Although this general definition applies to all shock entities, the underlying pathophysiology differs substantially. In hypovolemic shock, for example, there is a loss of intravascular volume, whereas in cardiogenic shock, myocardial function is severely compromised. Notably, in distributive shock, oxygen delivery to vital organs is inadequate, although there is not necessarily a reduction in intravascular volume or deterioration of cardiac function. The central mechanism of distributive shock is a profound vasodilation leading to its alternative designation as vasoplegic shock. It is most frequently caused by sepsis, systemic inflammation, or the post-cardiotomy syndrome following cardiac surgery.
On a cellular level, the loss of vascular tone is caused by persistent opening of ATP-sensitive and calcium-regulated potassium channels in vascular smooth muscle cells and an excessive production of nitric oxide (NO) due to an inappropriately high activation of NO synthases . Furthermore, NO impairs the reactivity of adrenergic receptors. In parallel, high endogenous plasma levels of catecholamines may induce a downregulation and desensitization of adrenergic receptors . Consequently, supra-physiological doses of exogenous catecholamines are often required for hemodynamic support, thereby, further aggravating catecholamine resistance, a condition associated with increased mortality rates of more than 70% . These patients were the first to be treated with non-adrenergic vasopressors as a last-resort therapy. It is not surprising that, except from sporadic case reports, outcome could not be improved substantially with non-adrenergic vasopressors at this disease stage. Several studies emphasize that non-adrenergic vasopressors are more effective, if they are administered early in the time course of shock . Last but not least, an endogenous vasopressin deficiency contributes to vascular hypo-responsiveness , representing the original rationale for the use of arginine vasopressin (AVP) in septic shock.
In distributive shock, macrocirculation is typically characterized by a hyperdynamic state: severely reduced vascular resistance, low MAP, and a compensatory increase in heart rate and cardiac output. While microcirculation in septic shock mostly is initially “flow-sensitive” and is improved with optimized macrocirculation, this is not necessarily the case in the later stage . Even if the organism by its compensatory mechanisms or by therapeutic measures can increase systemic oxygen delivery, tissue oxygenation might not be improved due to dysregulations in microcirculatory perfusion. Investigation of microcirculatory changes in distributive shock suggests that there are hypo-perfused and hyper-perfused areas within the same organ, both resulting in inadequate oxygen supply .
Pathophysiology of distributive shock
Shock is defined as an inadequate oxygen supply to vital organs due to circulatory failure combined with an impaired oxygen utilization in the tissues . Although this general definition applies to all shock entities, the underlying pathophysiology differs substantially. In hypovolemic shock, for example, there is a loss of intravascular volume, whereas in cardiogenic shock, myocardial function is severely compromised. Notably, in distributive shock, oxygen delivery to vital organs is inadequate, although there is not necessarily a reduction in intravascular volume or deterioration of cardiac function. The central mechanism of distributive shock is a profound vasodilation leading to its alternative designation as vasoplegic shock. It is most frequently caused by sepsis, systemic inflammation, or the post-cardiotomy syndrome following cardiac surgery.
On a cellular level, the loss of vascular tone is caused by persistent opening of ATP-sensitive and calcium-regulated potassium channels in vascular smooth muscle cells and an excessive production of nitric oxide (NO) due to an inappropriately high activation of NO synthases . Furthermore, NO impairs the reactivity of adrenergic receptors. In parallel, high endogenous plasma levels of catecholamines may induce a downregulation and desensitization of adrenergic receptors . Consequently, supra-physiological doses of exogenous catecholamines are often required for hemodynamic support, thereby, further aggravating catecholamine resistance, a condition associated with increased mortality rates of more than 70% . These patients were the first to be treated with non-adrenergic vasopressors as a last-resort therapy. It is not surprising that, except from sporadic case reports, outcome could not be improved substantially with non-adrenergic vasopressors at this disease stage. Several studies emphasize that non-adrenergic vasopressors are more effective, if they are administered early in the time course of shock . Last but not least, an endogenous vasopressin deficiency contributes to vascular hypo-responsiveness , representing the original rationale for the use of arginine vasopressin (AVP) in septic shock.
In distributive shock, macrocirculation is typically characterized by a hyperdynamic state: severely reduced vascular resistance, low MAP, and a compensatory increase in heart rate and cardiac output. While microcirculation in septic shock mostly is initially “flow-sensitive” and is improved with optimized macrocirculation, this is not necessarily the case in the later stage . Even if the organism by its compensatory mechanisms or by therapeutic measures can increase systemic oxygen delivery, tissue oxygenation might not be improved due to dysregulations in microcirculatory perfusion. Investigation of microcirculatory changes in distributive shock suggests that there are hypo-perfused and hyper-perfused areas within the same organ, both resulting in inadequate oxygen supply .
Hemodynamic therapy in distributive shock
On the basis of the pathophysiology of distributive shock, reconstitution of the vascular tone combined with volume resuscitation is the mainstay of therapy. According to current recommendations, the treatment of septic shock and vasoplegic syndrome following cardiac surgery is usually guided by MAP, cardiac output, central venous pressure, pulmonary artery occlusion pressure, and other variables calculated from these parameters . In addition, echocardiographic assessment of cardiovascular functions and the use of dynamic preload variables are recommended . Notably, all these parameters allow only estimation of the macrocirculation, that is, pressures, flow, pump function, and oxygen delivery. It is well known, however, that even if macrocirculation is optimized, organ function might still be severely compromised . Therefore, surrogate variables of organ function and oxygenation, for example, urine output, central venous oxygen saturation (S cv O 2 ), the ratio of central venous to arterial carbon dioxide partial pressure (P v-a CO 2 ), and lactate levels, have been incorporated in recommended treatment algorithms. Although these global markers represent a major advantage in hemodynamic therapy, they do not allow direct monitoring of microcirculatory changes. Consequently, there remains a serious contradiction between our knowledge about the pathophysiology of shock (resulting in a definition referring to the microcirculatory level) and current treatment options guided by macrocirculatory variables and global surrogate markers of oxygen extraction and consumption. This fact needs to be stressed since potential advantages of non-adrenergic vasopressors may not be obvious when assessing only the macrocirculation.
The catecholamine norepinephrine is currently recommended as the standard vasopressor for the treatment of distributive shock . Norepinephrine combines a pronounced alpha-adrenergic receptor activation, resulting in increases in vascular resistance and subsequently MAP, with moderate beta-adrenergic stimulation, thereby also supporting myocardial function to maintain cardiac output. Although every clinician knows that norepinephrine works very well in most of the patients, from a pathophysiological point of view, it seems to be questionable to infuse an exogenous catecholamine to shock patients in whom endogenous catecholamine levels are increased up to 20-fold and adrenergic receptors are downregulated and desensitized . Conversely, compounds whose receptors are upregulated and may even have an increased sensitivity because of an endogenous deficiency, such as vasopressin-receptor agonists, appear more promising. In line with this theory, there is increasing evidence that adrenergic stress and increasing doses of catecholamines contribute to worse outcome and were identified as independent risk factors for increased mortality rates in several shock states . Against this background, non-adrenergic vasopressors might be advantageous.
Non-adrenergic vasopressors
Several non-adrenergic compounds have been reported to improve hemodynamics in distributive shock. Unfortunately, clinical evidence about methylene blue and angiotensin II is currently limited to case reports or small clinical studies. Therefore, these non-adrenergic compounds are not considered in the present article. For detailed information about mechanisms of action and potential therapeutic algorithms for methylene blue or angiotensin II, respective review articles are recommended . On the contrary, vasopressin-receptor agonists have been extensively investigated following the first description of a relative AVP deficiency and substitution in septic shock in 1997 . AVP is an endogenous nonapeptide hormone that regulates plasma osmolality, intravascular volume status, and arterial blood pressure. It is rapidly metabolized by aminopeptidases and has an elimination half-life of about 6 min . The vasopressin system comprises 3 different vasopressin receptor subtypes. Importantly, the desired vasoconstrictive effects of vasopressin-receptor agonists are mediated through the V 1a -receptor (V 1a R) on vascular smooth muscle cells . Further, V 2 -receptor (V 2 R) activation is responsible for water retention (renal), stimulation of coagulation , vasodilation , and promotion of endothelial leakage (endothelial cells). Consequently, the affinity of individual vasopressin-receptor agonists for these two receptors, expressed as the V 1a R:V 2 R ratio, needs to be considered when interpreting and comparing study results. AVP has a V 1a R:V 2 R ratio of 1:1, suggesting an equal effect on both receptors .
AVP was initially administered as a continuous low-dose infusion to substitute the relative endogenous deficiency, targeting plasma levels of about 100 pg/mL . Another reason for the low dose were reports about pronounced vasoconstrictive effects in the splanchnic and coronary circulation . However, experimental and clinical studies using higher doses of AVP suggested superior effects with respect to hemodynamic stabilization, norepinephrine sparing effect, and renal function, without additional side effects . These results led to a shift from hormone substitution to a titrated vasopressor therapy, similar to that with norepinephrine.
Because AVP is not available in several European countries, the synthetic vasopressin analog terlipressin, a 12-amino-acid peptide (1-triglycyl-8-ysine-vasopressin), was tested as a replacement. Its active metabolite lysine vasopressin originates after a stepwise cleavage by endopeptidases, resulting in a slow release over several hours . In addition, terlipressin has an intrinsic pharmacologic activity that is responsible for the immediate effects reported after its administration . Because of the prolonged elimination half-life of about 50 min, terlipressin was initially administered in repetitive bolus doses until it was recognized that using a continuous low-dose infusion of 2 mg/24 h was as effective and not associated with the negative effects of bolus administration , such as increased pulmonary hypertension and reduction in systemic oxygen delivery . Notably, direct comparisons between AVP (experimental: 0.5 mU/kg/min, clinical: 0.03 U/min) and terlipressin (experimental: 1 μg/kg/h, clinical: 1.3 μg/kg/h) revealed a superiority of terlipressin with respect to hemodynamic stabilization, renal function, rebound phenomena after discontinuation, and positive fluid balance . This advantage of terlipressin over AVP is probably caused by the higher selectivity for the V 1a R resulting in a V 1a R:V 2 R ratio of 2.2:1 .
Following this discovery, highly selective V 1a R agonists have been investigated. Experimental studies in multiple established sepsis models not only revealed a consistent superiority of Phe 2 -Orn 8 -Vasotocin (POV, V 1a R:V 2 R ratio of 220:1 ) and selepressin (Phe 2 ,Ile 3 ,Hgn 4 ,Orn(iPr) 8 , V 1a R:V 2 R ratio of 1107:1 ) over AVP with respect to hemodynamic stabilization, but also a pronounced reduction of cumulative fluid balance . The mechanism of action that is responsible for the latter effect of selective V 1a R-agonists is not yet fully understood. Potential explanations include reduced fluid requirements due to better hemodynamic stabilization, increased urine output, attenuation of inflammation, and vascular leakage. Selepressin represents the only highly selective V 1a R-agonist tested in clinical studies. It is a short-acting synthetic peptide analog of AVP with a V 1a R:V 2 R ratio of 1107:1 . Despite this higher selectivity, its potency is about four times less than that of AVP. Accordingly, selepressin has been titrated up to 10 pmol/kg/min in experimental studies, corresponding to 2.5 pmol/kg/min of AVP (∼1 mU/kg/min or 0.07 U/min in a patient weighing 70 kg) . In clinical trials, selepressin doses ranging from 1.2 to 5 ng/kg/min (∼1.1–4.8 pmol/kg/min) are currently being investigated ( NCT01612676 , NCT01000649 , and NCT02508649 ), with the highest dose roughly corresponding to 0.5 mU/kg/min of AVP.
Vasopressin-receptor agonists versus norepinephrine: large randomized controlled trials and meta-analyses
The randomized controlled multicenter Vasopressin and Septic Shock Trial (VASST) revealed no significant differences in serious adverse events between patients treated solely with norepinephrine or norepinephrine supplemented with AVP (0.01–0.03 U/min). Notably, there was a reduced 28-day mortality in patients with less severe septic shock, defined as a norepinephrine dose at randomization of 5–14 μg/min . Post-hoc analyses of the latter study suggested beneficial effects on kidney function and a survival benefit, if AVP was combined with corticosteroids . The second double-blind, randomized controlled trial investigated the “effect of early vasopressin versus norepinephrine on kidney failure in patients with septic shock” (VANISH) using AVP as a first-line therapy titrated up to 0.06 U/min combined with hydrocortisone or placebo . Again, there were no differences in mortality rates and serious adverse events between the groups. However, the authors pointed out that “the confidence interval included a potential clinically important benefit for vasopressin” that needs to be investigated in future trials. Notably, VANISH demonstrated that the use of AVP in doses up to 0.06 IU/min as a first-line vasopressor and replacement for norepinephrine (not only a supplementation) is feasible and safe. This strongly challenges current sepsis guidelines that explicitly exclude AVP doses higher than 0.03–0.04 IU/min and is used as a first-line vasopressor. In patients with vasoplegic shock following cardiac surgery the use of AVP (0.01–0.06 IU/min) as first-line vasopressor was even superior to norepinephrine as suggested by a significant reduction in the composite endpoint of mortality and severe complications .
Randomized controlled trials using terlipressin verify the ability to effectively stabilize hemodynamics and reduce norepinephrine requirements as compared to norepinephrine and AVP . Notably, a supplemental bolus infusion of terlipressin (1 mg) was associated with a decrease in cardiac index and a deterioration in oxygen consumption , whereas there was no deterioration in systemic oxygen delivery with a continuous low-dose infusion of 1.3 µg/kg/h terlipressin (TERLIVAP) . Notably, there were no differences between vasopressors with respect to regional hemodynamics, as suggested by blood clearance and plasma disappearance rate of indocyanine green as well as by gastric mucosal arterial carbon dioxide partial pressure difference (P g-a CO 2 ).
Results from clinical trials with selepressin in septic shock patients are not published yet. However, two phase II trials investigating doses from 1.25 to 2.5 ng/kg/min in a total of 84 patients with septic shock have been completed ( NCT01612676 and NCT01000649 ). Preliminary reports describe a superiority of the higher dose that accelerated vasopressor weaning, decreased mortality, maintained cumulative fluid balance, and decreased the need for mechanical ventilation during the first 7 days . The randomized, placebo-controlled phase II-III trial “Selepressin evaluation program for sepsis-induced shock—adaptive clinical trial” (SEPSIS-ACT) is currently recruiting patients and will focus on vasopressor and mechanical ventilator–free days, using doses from 1.7 to 5.0 ng/kg/min ( NCT02508649 ).
In addition to these clinical studies, several systemic meta-analyses on the use of AVP and terlipressin in vasodilatory shock confirm the safe use and the catecholamine sparing effect . The results with respect to mortality, however, are conflicting and seem to rely on the individual methods applied. For example, an improved survival was suggested when the trials with AVP, terlipressin, and methylene blue were pooled, but this was not the case for any of the compounds individually .
In summary, randomized clinical trials and meta-analyses agree on the safe use of vasopressin-receptor agonists and the catecholamine sparing effect. The evidence on survival benefits over norepinephrine is not conclusive yet. In this regard, it needs to be considered that the results from the recently published VANISH trial were not included in any of the discussed meta-analyses. Further studies are warranted to identify the optimal therapeutic regime for the most effective treatment with vasopressin-receptor agonists (compound, V 1a R selectivity, dose, and timing) and the patient population that will probably benefit the most (genomics, causes of vasodilatory shock, etc.).
Macro- and microvascular coupling or hemodynamic coherence
Hemodynamic therapy in distributive shock is currently mainly guided by variables characterizing macrocirculation, as previously mentioned . Volume resuscitation and vasopressors are often sufficient to achieve the treatment goals suggested for these variables, but in critically ill patients, microcirculatory changes may not infallibly correspond to these macrocirculatory effects . An improvement in macrocirculation that is not accompanied by an improved microcirculation is described as “loss of hemodynamic coherence” . Pathophysiological mechanisms include heterogeneous blood flow, reduced capillary density due to hemodilution, vasoconstriction, and tissue edema . The major threat of hemodynamic incoherence is that the medical team might rely on the therapeutic success of macrocirculatory interventions without recognizing the patients’ fundamental problem on a microcirculatory level. This is of special importance for potent vasoconstrictors such as vasopressin-receptor agonists, because main reasons for the restrictive usage include the risk for occult ischemic lesions in microvascular beds, such as the coronary, splanchnic, or peripheral circulation . The above discussed large randomized controlled trials did not evaluate the coupling of macro- and microcirculation. Therefore, the following discussion is based on experimental studies and small clinical trials listed in Tables 1 and 2 , respectively. Notably, most of these were not designed to focus on hemodynamic coherence, and thus, did not evaluate the microcirculation directly, for example through incident dark-field imaging. Consequently, studies reporting surrogate variables such as lactate levels, parameters of global oxygen transport, and other variables were also included. Accordingly, conclusions on hemodynamic coherence demonstrated in the last columns of Tables 1 and 2 mostly represent our interpretation and not that of the authors of the individual article.
| 1st author year | Design | Groups | Dose | Macrocirculation | Microcirculation | Coherence |
|---|---|---|---|---|---|---|
| He 2016 | 46 sheep septic shock | SP AVP NE | 1 pmol/kg/min 0.1 mU/kg/min 0.5 μg/kg/min titrated to MAP > 70 mmHg | SP: MAP↑, CO↑ mesenteric blood flow=, renal blood flow↑, crea-clearance ↑ | SP: DO 2 I + SvO 2 ↑ lactate↓ | + |
| Bomberg 2015 | 18 pigs on CPB | AVP | 0.006 U/kg/min titrated to BL-MAP | MAP↑ | gastric mucosal microcirculation↓ | − |
| Bomberg 2014 | 18 pigs on CPB | AVP | 0.006 U/min/kg titrated to BL-MAP | MAP↑ | rectosigmoidal mucosal blood flow↓ | − |
| Qiu 2014 | 25 rats endotoxic shock | TP NE | 8 μg/kg/h | MAP↑ (=vs. NE) | gut: MFI↑ heterogeneity index↓ | + |
| Bomberg 2013 | 24 pigs on CPB | AVP | 0.006 U/kg/min titrated to BL-MAP | MAP↑ | jejunal mucosal capillary density + tissue blood flow↑ | + |
| Rehberg 2011 | 24 sheep septic shock | AVP POV NE | 0.05 μg/kg/h 0.05 μg/kg/h | POV: NE↓, mesenteric blood flow= | POV: SvO 2 ↑, DO 2 I↑, lactate↓ | + |
| Holt 2011 | 20 piglets endotoxic shock | AVP AVP+ dobuta-mine | 0.04 U/min 10 μg/kg/min | MAP↑ CO↓ | shunting of microvasculatory flow from skin + GI to the brain, liver and kidneys; no effect of dobutamine | −/+ |
| Rehberg 2010 | 21 sheep septic shock | AVP AVP+ Levo NE | 0.5 mU/kg/min 0.2 μg/kg/min | both: NE↓, mesenteric blood flow= | AVP + Levo: SvO 2 ↑, lactate= DO 2 I + VO 2 I= | (+) |
| Maier 2009 | 24 pigs endotoxic shock | AVP+ NE NE | 57 mU/kg/h | both: MAP↑, CO↑, blood flow AMS↑ | AVP + NE: jejunal PO 2 muc↓ | − |
| Rehberg 2009 | 16 sheep septic shock | AVP TP NE | 0.5 mU/kg/min 1 μg/kg/h | both: NE↓, mesenteric blood flow= | TP: SvO 2 ↑, lactate= DO 2 I + VO 2 I= | (+) |
| Kopel 2008 | 11 rabbits endotoxic shock | AVP | 1–1000 ng bolus | MAP↑ mesenteric blood flow= >100 ng: mesenteric blood flow↓ | microcirculation gut= >100 ng: microcirculation gut ↓ | (+) − (>100 ng) |
| Hiltebrand 2007 | 32 pigs sepsis | AVP | 0.06 U/kg/h | MAP↑ blood flow AMS↓ | VO 2 + DO 2 ↓ O 2 ER↑ microcirculatory flow in the stomach, jejunum ↓, colon= P(g-a)CO 2 ↑ | − |
| Krejci 2007 | 32 pigs sepsis | AVP | 0.06 U/kg/h | portal vein↓ hepatic artery= celiac trunk↑ renal↓ | pancreas↓ kidney↓ liver= | +(neg) |
| Knotzer 2006 | 20 pigs endotoxic shock | AVP | 0.014–0.229 U/kg/h | MAP↑ | jejunal mucosal oxygen tension and supply= | (+) |
| Nakajima 2006 | 36 mice endotoxic shock | AVP NE | titrated to BL-MAP | MAP↑ | arteriolar flux, velocity and density of perfused villi= | (+) |
| Asfar 2005 | 18 pigs endotoxic shock | TP | 5–15 μg/kg/h | MAP↑, CO↓ hepatic blood flow↑ | VO 2 ↓, lactate↑ hepatic O 2 exchange + metabolism↑ | −/+ |
| Faivre 2005 | 57 rabbits endotoxic shock | AVP NE | 0.005 U/kg/min | MAP↑, CI↓ | renal medullary microcirculation↓ | – |
| Albert 2004 | 26 rabbits endotoxic shock | AVP | 1–1000 ng bolus | MAP↑ aortic flow= | kidney cortex↑ | + |
| Westphal 2004 | 15 rats septic shock | AVP | 0.006 U/min | MAP↑ | ileum mucosal blood flow↓ | − |
| Asfar 2003 | 77 rats endotoxic shock | TP | 6 μg/kg bolus | MAP↑, aortic blood flow= mesenteric blood flow= | ileal microcirculation:with fluid↑ without fluid ↓ | + (fluid) − (no fluid) |
| Westphal 2003 | 6 sheep endotoxic shock | TP | 10–40 μg/kg/h | MAP↑, CI↓ | DO 2 I + VO 2 I↓ | − |
| Westphal 2003 | 6 sheep endotoxic shock | AVP | 0.6–3.6 U/h | MAP↑, CI↓ | DO 2 I + VO 2 I↓ | − |
Related posts:
Hemodynamic coherence: Its meaning in perioperative and intensive care medicine
Haemodynamic coherence — The relevance of fluid therapy
Haemodynamic coherence in haemorrhagic shock
Haemodynamic coherence in perioperative setting
Effects of vasodilators on haemodynamic coherence
The response of the microcirculation to mechanical support of the heart in critical illness
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
