Vasopressin
Administration of vasopressin with or without norepinephrine during septic shock has been studied in large-scale studies, which assessed whether vasopressin administration could result in an improvement in survival and renal function.
Russel et al. in the VASST study (VAsopressin in Septic Shock Trial) [
37], compared the effect of vasopressin administration (
n = 396) versus norepinephrine administration (
n = 382) in patients with septic shock, in a randomized, double-blinded placebo-controlled study. The primary endpoint was mortality at day 28. An a priori stratification with respect to severity was integrated into the study design: severity was defined according to the norepinephrine infusion rates required to maintain arterial pressure, i.e. < vs ≥ 15 µg/min. The vasopressin infusion rate was 0.01–0.03 U/min. No differences were observed between groups in survival at day 28 and 90. Furthermore, there was no difference in organ failure occurrence between groups. Of note, safety profile was reassuring with no significant difference between groups in side effects occurrence. In contrast to the authors’ initial hypothesis that the more severe patients would benefit from the vasopressin treatment, outcome did not differ in the stratum of patients with norepinephrine doses ≥ 15 µg/min; in patients with norepinephrine infusion rates < 15 µg/min, vasopressin administration was associated with significantly improved survival at day 28 (26.5% vs 35.7%,
p = 0.05) and 90 (35.8% vs 46.1%,
p = 0.04). It remains an open question, whether this effect was due to vasopressin per se and/or the fact that in the less severe patients, a higher proportion could be completely weaned from norepinephrine support [
38]. A post hoc analysis of VASST study according to the Sepsis-3 definition confirmed this finding [
39]: while in patients, who fulfilled the sepsis-3 criteria of septic shock, i.e. arterial hypotension requiring introduction of catecholamine to maintain mean blood pressure > 65 mmHg despite adequate fluid resuscitation and hyperlactatemia > 2 mmol/L, there was no difference in mortality, survival was increased in patients with hypotension requiring vasopressors alone without hyperlactatemia.
A further post hoc analysis of the VASST trial suggested an association between the use of vasopressin and hydrocortisone: vasopressin reduced mortality in patients, who—at the discretion of the attending physician—were treated with hydrocortisone. Conversely, in patients who did not received hydrocortisone, vasopressin administration was associated with higher mortality [
40].
Another post hoc analysis of the VASST trial tried to assess the impact of vasopressin on kidney (dys)function: Gordon et al. analysed the 464 patients with acute kidney injury of VASST study cohort [
41], categorizing them according to RIFLE classification [
42]. Patients at “Risk” stage and treated with vasopressin had significantly less worsening of renal function (progression to “injury/failure” stage) than patients treated with norepinephrine.
The potential impact of hydrocortisone treatment together with a putative beneficial effect of vasopressin on AKI prompted the design of the VANISH study [
43], which compared the effects of either vasopressin or norepinephrine as the vasopressor of first choice together with hydrocortisone or vehicle using a two-by-two factorial design. No interaction was identified between vasopressin and hydrocortisone [
43]. Furthermore, neither mortality nor AKI were significantly influenced; nevertheless, the authors concluded that “
the confidence interval included a potential clinically important benefit for vasopressin”, which should be tested further in larger trials.
Clearly, both the VANISH and VASST trials were “negative” in the sense that no superiority of vasopressin could be demonstrated. Nevertheless, these studies showed that vasopressin is not inferior to norepinephrine either, and, that its administration was not associated with higher rates of adverse effects, especially ischaemic events. Moreover, these studies demonstrated that vasopressin allows at least a reduction of norepinephrine infusion rates and fastened weaning from norepinephrine. In the context of “decatecholaminization”, vasopressin use limits norepinephrine infusion rates and potentially its related side effects [
44]. Interestingly, in a retrospective study, patients with septic shock who received fixed-dose of vasopressin in association with norepinephrine had an improve prognosis when vasopressin infusion resulted in an increase in mean arterial pressure [
45].
Vasopressin analogues: terlipressin and selepressin
Synthetic analogues of vasopressin with higher vascular selectivity (V1/V2 ratio > 1) were also studied in the setting of septic shock.
In a blinded, randomized study recruiting 32 patients with septic shock, terlipressin administration in combination with norepinephrine was superior to norepinephrine alone (complication and survival at day 7), but results have to be tempered due to the small sample size [
46].
In 2018, Liu et al. reported results of a randomized multicentric double-blind study, aimed at comparing terlipressin and norepinephrine as first-line vasopressor agent in septic shock [
47]. In this study, 526 patients were analysed: 260 received up to 4 mg terlipressin per day, and 266 received norepinephrine. There was no difference in mortality at day 28 between groups (primary endpoint). Likewise, improvement and variation of SOFA score until day 7 were comparable. Terlipressin administration was associated with an increase in adverse effects (30% vs 12%,
p < 0.01). The most common was digital ischaemia (12.6% in terlipressin group versus 0.35% in norepinephrine group,
p < 0.001). 76% of adverse effects occured during the first 24 h.
A more recent synthetic V1a agonist, selepressin, was reported, in a sheep model of faecal peritonitis, to be superior to norepinephrine and vasopressin. If, in established septic shock, selepressin maintained mean arterial pressure to a similar extent as did vasopressin and norepinephrine, selepressin administration resulted in reduced lung edema [
48]. These results were in accordance with previous study, highlighting the reduced vascular leakage with selepressin administration [
49].
Recently, the SEPSISACT study [
50] compared with a randomized placebo-controlled trial the addition of selepressin to norepinephrine. This study was divided in two parts: the first was aimed at assessing the best-performing regimen of selepressin (between three dosing regimens of selepressin), and, in the second part, the authors compared this selepressin regimen to placebo. Among the 817 patients who completed the trial, no difference between the two groups on the primary composite endpoint was observed (ventilator and/or vasopressor free days within 30 days). Likewise, no difference in secondary endpoints was observed (mortality, need for renal replacement therapy and length of stay in the intensive care unit). Selepressin administration was not associated with higher rate of side effects.
A meta-analysis of 20 study including patients with septic shock suggested a potential improvement of survival with the use of vasopressin (9 studies) and its analogues (11 studies), when compared with adrenergic vasopressors (mostly norepinephrine). This meta-analysis underlined the additional risk of digital ischaemia [
51]. In a recent meta-analysis of individual patients data from randomized studies (1453 septic shock patients) [
52], no difference in mortality was found between vasopressin and norepinephrine. Furthermore, vasopressin use was not associated with more undesirable effect, but safety profiles were different: vasopressin led to more digital ischaemia, but fewer arrhythmias. This decrease in the incidence of arrhythmias was also reported in a meta-analysis dedicated to this specific question [
53]. In this analysis were included randomized controlled trials aimed at comparing the association of vasopressin with norepinephrine to catecholamines alone for patients with distributive shock.