Skip to main content


Archived Comments for: Arginine vasopressin versus norepinephrine: will the stronger one win the race?

Back to article


    Barbara Friesenecker, Amy G Tsai; Judith Martini; Hanno Ulmer; Volker Wenzel; Walter R Hasibeder; Marcos Intaglietta; Martin W Duenser

    28 June 2006

    In the recent issue of Critical Care, Ertmer and colleagues [1] comment on our animal study concerning the microvascular response of arginine vasopressin (AVP) when compared to norepinephrine [2]. We appreciate the critical statements depicted in the commentary. However, we feel that there are still some misunderstandings regarding our studies and in particular in particular the importance of combining vasopressor agents in patients suffering from severe septic shock.

    Ertmer argued that no differences in mean arterial pressure were observed between groups and concluded that AVP and NE demonstrated identical vasoconstrictor effects. Aim of our study was, however, to elucidate possible differences in the vasoconstrictor activity within the microcirculatory network between both drugs, while the study was not designed to demonstrate significant differences in macrocirculatory parameters. Therefore Norepinephrine infusion was titrated to a dosage level where it exerted a similar increase in mean arterial pressure as AVP. We further do not agree with Ertmer’s suggestion that vascular resistance is predominantly regulated by small arterioles (A2-A4) as large arterioles (A0-A1) have been found to be the major regulators of peripheral vascular resistance [3, 4, 5]. For these reasons, although using the same animal model, the study by Gerstberger [6] cannot be compared to our study. The suggestion that a plasma AVP concentration of ~100 pg/mL represents the upper therapeutic range during AVP infusion in patients with hyperdynamic septic shock is not supported by any study. In contrast, most trials have reported that plasma AVP concentrations reach 150-300 pg/mL during a continuous AVP infusion at standard dosages (1-4 IU/h) [7,8]. In view of growing evidence that the contribution of AVP to improve cardiovascular function in critical illness is minor at plasma concentrations up to 50 pg/mL [9], the hypothesis becomes more and more unlikely, that a continuous AVP infusion in severe septic shock serves as a simple hormone substitution therapy. Moreover, if septic shock patients (AVP plasma concentrations, 3.1 +/- 0.4 pg/mL) should be substituted to similar plasma AVP concentrations as in patients with cardiogenic shock (22.7 +/- 2.2) [10,11], it is difficult to understand why an AVP infusion increases plasma concentrations to as high levels as 150-300 pg/mL.

    The suggestion of Ertmer, that vasopressor therapy in septic or vasodilatory shock is based on the idea to re-distribute blood flow away from non-vital organs to vital organs by causing hypoperfusion of the non-vital organs seems too mechanistic and oversimplified. It is fact, that the administration of vasoconstrictors for example in uncontrolled hemorrhagic shock aims at the suggested re-distribution of blood flow from non-vital to vital organs in order to facilitate short time survival. However, when looking closer at the pathophysiology of cardiovascular failure in septic and vasodilatory shock [12], vasoconstrictors are known to reverse pathological vasodilatation and restore vascular tone to subnormal values in order to guarantee adequate tissue perfusion pressure. In contrast to patients with uncontrolled hemorrhage [13], the aim of vasopressor therapy in septic and vasodilatory shock is to improve tissue perfusion pressure to all organs, non-vital as well as vital organs. Nonetheless, it is true that based on the heterogeneous distribution of widely dilated, normal, constricted, or obstructed regions of the microcirculation in septic shock [14], every vasopressor drug also bears the risk to induce focal tissue ischemia. Therefore, our future therapeutic goal must be to optimize vasoconstrictor therapy by its combination with either vasodilators or drugs capable of improving microcirculatory flow in order to prevent harmful effects of vasoconstriction, which might be detrimental to survival.

    We agree with Ertmer that AVP increases the risk of gut ischemia in some rat models [14], but data from other animal [15, 16] and rat models [17] show different results. Moreover, in almost all human studies in septic shock no clinical evidence so far suggests that AVP causes intestinal ischemia [18, 19, 20]. Additionally, the observation that ischemic skin lesions can be observed during AVP infusion in a significant portion of patients with advanced vasodilatory shock more likely seems to result from the underlying dysfunction of the microcirculatory system in severe multiple organ dysfunction syndrome [21] and is not dependent on the vasopressor regime used [18].

    The basis for the current concept of AVP therapy in septic and vasodilatory shock is not a question of using the stronger vasopressor drug. This has been proven to place organ perfusion at risk in one human study [22]. In the opposite, we consider AVP treatment necessary only in a minor portion of patients who cannot be adequately stabilized by conventional hemodynamic therapy (~10% of all septic and vasodilatory shock patients). The concept of a supplementary infusion of AVP in addition to standard hemodynamic treatment attempts to keep vasopressor and inotropic catecholamine dosages within a tolerable clinical range where benefits still outweigh the risks of adverse side effects. According to recent data, this critical dosage of norepinephrine seems to be 0.6 µg/kg/min [19]. It is known that inconsiderate increase of norepinephrine dosage in patients with advanced vasodilatory shock cause significant side effects, e.g. induction of tachyarrhythmias [18, 23]. It is further proven that high adrenergic stress leads to numerous adverse effects reaching from myocardial ischemia [24], myocardial DNA damage [25], induction of myocardial cell apoptosis [26] and pulmonary hypertension [27] over immunomodulation [28], coagulation activation [29], reduction of hepatosplanchnic oxygen supply [30], as well as metabolic [31] and endocrinologic derangements [32] to bone marrow suppression [33] and direct myolytic effects [34]. Thus, supplementary AVP infusion at dosages not exceeding 4 IU/h may be considered as a “bridging” technique in a small, but severely sick group of patients with advanced cardiovascular failure until hemodynamic improvement allows reduction of toxically high catecholamine dosages. Therefore, the question may not be which one of the available vasopressors will win the race, but much more how the optimal combination of differently acting drugs available in our complex pharmacological armamentarium bears the most benefit for our patients with the complex problem of cardiovascular failure in septic or vasodilatory shock by reducing side effects known from high dosage catechoalmine therapy.

    In view of this relatively novel concept, there has so far been no data from large multicenter trials to prove the effects on patient outcome. Currently, an ongoing multicenter trial in North America examines the effects of AVP (0.03 IU/min) as a supplementary vasopressor in patients with septic shock unresponsive to standard vasopressor therapy. The study investigators hypothesized that compared with norepinephrine treatment alone, supplementary infusion of AVP will increase 28-day survival from 40 to 50%. Based on the experience in 316 vasodilatory shock patients [20] treated according to this hemodynamic protocol at our institution, a significant difference in mortality was observed when compared with a historical control group. The historical control group was recruited from our institutional databank from the years 1997 and 1998, when AVP had not yet been used. Control patients were blindly matched to AVP-treated patients according to their multiple organ dysfunction syndrome scores. Mortality was significantly lower in patients receiving supplementary AVP infusion when compared with the historical control patients (50.9% (161/316) vs. 63.6% (63/99); RR, 1.35; 95% CI, 1.02-1.85; p=0.029).

    Competing interests

    None declared

  2. Arginine vasopressin versus norepinephrine: Trials will finally unveil the riddle

    Christian Ertmer, Department of Anesthesiology and Intensive Care Medicine, University of Muenster

    21 July 2006

    Christian Ertmer, Hans-Georg Bone and Martin Westphal

    In the current recent issue of Critical Care, Friesenecker and colleagues [1] critically discuss our recent commentary [2] on their study comparing the microvascular effects of arginine vasopressin (AVP) versus norepinephrine [3]. Unfortunately, the authors did not refer to our main criticism, i.e. that they drew vague conclusions by transferring findings observed in the physiologic state to septic conditions.

    Regarding their comments on the treatment of septic shock, we fully agree that the aim of vasopressor therapy is to (re)establish adequate perfusion of all vital and non-vital organs, and that combining vasopressors with vasodilator agents might be a suitable approach. In fact, the latter theory is not new, but represents a concept which our research group is working on since 2004 [4].

    Friesenecker’s thesis that plasma AVP levels as high as 150-300 pg/ml represent the therapeutical range in septic shock patients is strongly contradicted by the findings of Van Haren et al. Notably, the latter authors reported that AVP levels > 50 pg/ml proportionally correlate with increased gastric mucosal pCO2 gap [5], which in turn is associated with increased mortality [6]. Therefore, it seems questionable that the concept postulated by Friesenecker et al. [1] is in accordance with the therapeutical goal in the treatment of distributive shock.

    In our opinion, there is no doubt that the optimal hemodynamic support for septic shock patients can only be achieved with a combination therapy. However, the question if vasopressin analogues will be part of this combination can only be answered by randomized controlled trials adequately powered to determine the impact on overall outcome, such as mortality [7].


    1. Friesenecker BE, Tsai AG, Martini J, Ulmer H, Wenzel V, Hasibeder WR, Intaglietta M, Dünser MW: Arginine vasopressin versus norepinephrine: team work wins the match or there are no simple solutions for complex problems. Crit Care 2006, Comment published on 28 June 2006.

    2. Ertmer C, Bone HG, Westphal M: Arginine vasopressin versus norepinephrine: will the stronger one win the race? Crit Care 2006, 10:145, doi:10.1186/cc4942.

    3. Friesenecker BE, Tsai AG, Martini J, Ulmer H, Wenzel V, Hasibeder WR, Intaglietta M, Dünser MW: Arteriolar vasoconstrictive response: comparing the effects of arginine vasopressin and norepinephrine. Crit Care 2006, 10:R75, doi:10.1186/cc4922.

    4. Westphal M, Sielenkamper AW, Van Aken H, Stubbe HD, Daudel F, Schepers R, Schulte S, Bone HG: Dopexamine reverses the vasopressin-associated impairment in tissue oxygen supply but decreases systemic blood pressure in ovine endotoxemia. Anesth Analg 2004, 99:878-885.

    5. Van Haren FM, Rozendaal FW, van der Hoeven JG: The effect of vasopressin on gastric perfusion in catecholamine-dependent patients in septic shock. Chest 200;124:2256-2260.

    6. Maynard N, Bihari D, Beale R, Smithies M, Baldock G, Mason R, McColl I: Assessment of splanchnic oxygenation by gastric tonometry in patients with acute circulatory failure. JAMA 1993, 270:1203-1210.

    7. Russell JA, Cooper DJ, Walley KR, Holmes CL, Singer J, Hebert PC, Granton J, Mehta S, Terins T: Vasopressin and Septic Shock Trial (VASST): baseline characteristics and organ dysfunction in vasopressor dependent patients with septic shock [abstract]. Am J Resp Crit Care Med 2003, 167:A548.

    Competing interests