Current and future strategies aiming at reducing catecholamine exposure in septic shock

Abstract

Norepinephrine is the first-line vasopressor in septic shock, yet prolonged catecholamine exposure is associated with adverse effects that have prompted growing interest in catecholamine-sparing strategies. This review highlights current evidence on the rationale for catecholamine use, the burden of sustained adrenergic exposure, and current and emerging sparing strategies. Early norepinephrine initiation, including via peripheral access, shortens hypotension duration and reduces fluid requirements. However, catecholamine exposure can carry dose-dependent cardiac, metabolic, and immunological consequences. Perfusion-guided strategies, including individualization of blood pressure targets and titration of vasopressor use based on capillary refill time, represent the cornerstone of reduction of catecholamines. Among alternative non-adrenergic vasopressors, vasopressin reduces catecholamine exposure and the risk of atrial fibrillation, with potential renal benefits. Angiotensin II represents an option in catecholamine-refractory shock, with post-hoc evidence suggesting benefit in patients with acute kidney injury or elevated renin concentrations. Inhibition of circulating dipeptidyl peptidase 3, which degrades angiotensin II, is an emerging therapeutic strategy. Corticosteroids restore vasopressor sensitivity and accelerate catecholamine weaning. Short-acting β1-blockers have shown hemodynamic promise but inconsistent outcomes, underscoring the need for better patient selection. Methylene blue, targeting the vasodilatory nitric oxide pathway, represents another strategy. Finally, emerging immunomodulatory approaches, including extracellular histone neutralization and polymyxin B hemoperfusion in endotoxin phenotypes, aim to attenuate the dysregulated host response driving vasopressor dependency. A personalized and multimodal approach, including perfusion-guided targets, non-adrenergic vasopressors, and phenotype-based patient selection, represents the most promising strategy to reduce potential consequences of adrenergic burden while maintaining tissue perfusion.

Background

Sepsis is defined as a life-threatening organ dysfunction caused by dysregulated host responses to infection [1]. Its more severe form, septic shock, is associated with high intensive care unit (ICU) mortality rates, approaching 40% [2]. Management of septic shock is based on early diagnosis, source control with antibiotic therapy, and hemodynamic support with fluids and early vasopressors [3]. Norepinephrine is the primary vasopressor used in septic shock [4, 5]. However, catecholamines may be associated with multiple adverse effects, and elevated endogenous norepinephrine levels have long been associated with worse outcomes in septic shock [6]. Consequently, reducing catecholamine exposure has been a topic of considerable interest in recent years [7]. In this review, we provide an overview of the rationale for catecholamine use, their potential adverse events, and current and future catecholamine-sparing strategies during septic shock.

Why catecholamines remain central to septic shock resuscitation

How catecholamines restore perfusion

α-Adrenergic effects: restoration of arterial pressure

In septic shock, loss of arterial vascular tone represents the main hemodynamic abnormality leading to hypotension, with lower values of all blood pressure components consistently associated with increased ICU mortality [8]. Norepinephrine restores arterial pressure primarily through α₁-mediated vasoconstriction, resulting in a dose-dependent increase in systemic vascular resistance and mean arterial pressure (MAP) (Fig. 1) [9]. However, the classical interpretation of vasopressor effects through the derived hemodynamic variable systemic vascular resistance (SVR = (mean arterial pressure – central venous pressure) / cardiac output) incompletely reflects the physiology of tissue perfusion [10]. Indeed, vascular pressure decline is not continuous throughout the circulation, but occurs predominantly across small arteries and arterioles. In this context, critical closing pressure (Pcrit) represents the arterial pressure threshold below which vessels collapse and flow ceases despite persistence of a pressure gradient (Fig. 1) [10]. Consequently, tissue perfusion is more accurately described by the arterial pressure to Pcrit gradient (Pa—Pcrit), while preservation of microcirculatory flow additionally depends on maintenance of the vascular waterfall gradient between Pcrit and mean systemic filling pressure (Pmsf) [11]. In septic shock, loss of vasomotor tone may decrease Pcrit toward Pmsf, effectively abolishing the vascular waterfall despite preserved or elevated cardiac output [12, 13]. Under these conditions, increasing MAP alone may not restore tissue perfusion. Norepinephrine may improve perfusion only if it increases Pcrit sufficiently above Pmsf to restore the vascular waterfall while simultaneously increasing MAP above the new Pcrit, thereby generating an effective tissue perfusion pressure [11]. Conversely, venous congestion increasing Pmsf may reduce the effective Pcrit—Pmsf gradient and impair tissue perfusion despite preserved MAP (Fig. 1) [14].

Fig. 1.

Hemodynamic effects of Norepinephrine. α₁, alpha-1 adrenergic receptor; α₂, alpha-2 adrenergic receptor; β₁, beta-1 adrenergic receptor; CO, cardiac output; CVP, central venous pressure; IVC, inferior vena cava; LA, left atrium; LV, left ventricle; LVEF, left ventricular ejection fraction; MAP, mean arterial pressure; NE, norepinephrine; Pa, arterial pressure; PA, pulmonary artery; Pcrit, critical closing pressure; Pmsf, mean systemic filling pressure; RA, right atrium; RV, right ventricle; SVC, superior vena cava; SVR, systemic vascular resistance; TPP, tissue perfusion pressure; VW, vascular waterfall gradient

α and β₁-adrenergic effects: impact on cardiac output

Beyond its arterial pressor effect, norepinephrine modulates cardiac output through actions on the venous circulation and the myocardium, via activation of both α- and β1-adrenergic receptors, thereby combining vasopressor and inotropic properties.

Through α-adrenergic mediated venoconstriction, norepinephrine increases mean systemic filling pressure, which in preload-responsive septic patients translates into enhanced venous return and cardiac output [15–17]. In a physiological study including 25 patients with septic shock who had a positive passive leg raising test and a diastolic arterial pressure ≤ 40 mmHg, increasing the norepinephrine dose was associated with significant increases in preload markers, consistent with recruitment of cardiac preload reserve. A subsequent passive leg raising test induced a smaller increase in cardiac index compared with baseline, indicating a reduction in preload dependency (13 ± 8% vs. 19 ± 6%, p < 0.05) [17]. Similar preload-mediated effects on cardiac output have been reported in other clinical physiological studies [16, 18].

Although norepinephrine is expected to increase afterload through vasoconstriction, it can simultaneously increase cardiac output via β₁-adrenergic stimulation. In another physiological study performed in 38 patients with septic shock resuscitated for less than 3 h and with persistent hypotension (MAP < 65 mmHg), norepinephrine was associated with an expected rise in MAP from 56 (7) to 80 (9) mmHg. Despite this increase in arterial load, left ventricular ejection fraction (LVEF) and especially Doppler-derived systolic velocities increased, supporting a significant inotropic effect of norepinephrine in this setting [19].

Divergent effects on the microcirculation

At the microcirculatory level, septic shock is characterized by profound and early alterations that are largely independent of systemic hemodynamics. Videomicroscopy studies consistently demonstrate a marked reduction in functional capillary density and pronounced flow heterogeneity. These abnormalities are independently associated with organ failure and mortality and may persist despite correction of MAP and cardiac output [20, 21]. Importantly, patients who exhibit early recovery of microcirculatory perfusion develop less organ dysfunction, whereas persistent alterations identify a subgroup with particularly poor outcomes [22, 23].

By increasing arterial pressure, norepinephrine can recruit pressure-dependent microvascular units when MAP is critically low. In severely hypotensive septic patients, raising MAP with norepinephrine has been shown to improve skeletal muscle tissue oxygenation assessed by near-infrared spectroscopy, indicating partial restoration of microvascular perfusion in pressure-sensitive territories [24]. However, beyond this critical hypotensive range, interventional studies consistently demonstrate marked heterogeneity in microcirculatory responses. Since arterioles at the entrance of the capillary network express α1-adrenergic receptors, the effect of norepinephrine on tissue perfusion reflects a balance between restoration of perfusion pressure and α1-adrenergic mediated vasoconstriction, which may impair microvascular blood flow [7]. From a physiological perspective, effective tissue perfusion depends not only on MAP, but also on the arterial pressure-to-critical closing pressure gradient (Pa − Pcrit), while preservation of microcirculatory flow additionally requires maintenance of the vascular waterfall gradient between critical closing pressure (Pcrit) and mean systemic filling pressure (Pmsf) (Fig. 1) [11]. Consequently, norepinephrine-induced increases in MAP may fail to restore microvascular perfusion if Pcrit remains close to Pmsf or if venous congestion increases Pmsf, highlighting the loss of hemodynamic coherence between the macro- and microcirculation [14, 25–29].

Why norepinephrine remains the first-line vasopressor

Norepinephrine has long been central to septic shock management because both the severity and duration of arterial hypotension are closely associated with increased mortality [30, 31]. Its clinical importance was underscored during the 2011 norepinephrine shortage in the United States, when substitution with alternative vasopressors (mainly phenylephrine) was associated with increased in-hospital mortality (absolute risk increase of 3.7% [95% CI, 1.5%-6.0%]; adjusted odds ratio = 1.15 [95% CI, 1.01–1.30]; p = 0.03) [32].

Comparative trials have shown no clear mortality difference between catecholamines, but important differences in safety profiles. In the SOAP II trial, patients with shock defined by a MAP < 70 mmHg despite adequate fluid resuscitation (≥ 1 L of crystalloids) and signs of hypoperfusion were randomized to receive either dopamine or norepinephrine. Mortality at 28 days was similar between groups (52.5% in the dopamine group compared to 48.5% in the norepinephrine group, p = 0.10). However, arrhythmic events were significantly more frequent in patients treated with dopamine than in those receiving norepinephrine (24.1% vs. 12.4%; p < 0.001), which has largely limited the role of dopamine in septic shock [5]. Epinephrine has also been evaluated in two randomized controlled trials (RCTs). In the CAT trial, 280 patients were randomized to receive a blinded infusion of either epinephrine or norepinephrine. The primary outcome, achievement of a MAP > 70 mmHg for more than 24 h without vasopressor support, was similar in both groups. However, epinephrine was associated with more pronounced metabolic adverse effects, leading to treatment discontinuation in 12.9% of patients [33]. The CATS trial evaluated the effect of norepinephrine plus dobutamine (when needed) versus epinephrine alone on 28-day mortality in a RCT including 330 patients. No significant difference in mortality was observed between the two groups, with 40% mortality in the epinephrine group compared with 34% in the norepinephrine plus dobutamine group (p = 0.31). Rates of serious adverse events were similar between groups [34].

While epinephrine is as effective as norepinephrine-based strategies in achieving MAP targets, epinephrine is not associated with improved clinical outcomes and is consistently associated with a higher incidence of lactic acidosis and arrhythmias. Together, these data reinforce norepinephrine as the catecholamine with the most favorable tolerability and safety profile. Consequently, international guidelines recommend norepinephrine as the first-line vasopressor for septic shock [35].

Rethinking when and how norepinephrine is started

Current evidence supporting early norepinephrine initiation

Contemporary resuscitation strategies increasingly support early initiation of norepinephrine, often in parallel with initial fluid administration rather than after completion of fixed-volume fluid loading [36]. Across randomized and observational studies, early norepinephrine initiation consistently shortens the duration of hypotension, accelerates shock control, and reduces cumulative fluid administration. In the randomized CENSER trial, patients with sepsis and MAP < 65 mmHg were randomized to early norepinephrine with fluids or delayed vasopressors after ≥ 30 mL/kg of crystalloids. Early low-dose norepinephrine significantly increased the proportion of patients achieving shock control at 6 h, defined as achievement of a MAP > 65 mmHg with urine output > 0.5 mL/kg/h for two consecutive hours or a decrease in serum lactate > 10% from baseline, compared with usual care (76% vs. 48%, p < 0.001) [37]. Additionally, early norepinephrine was associated with reduced incidences of cardiogenic pulmonary edema (14.4% vs. 27.7%, p < 0.01) and new-onset arrhythmia (11% vs. 20%, p = 0.03), while no differences were observed in the 28-day mortality (15.5% vs 21.9%, p = 0.15).

In the CLOVERS trial, conducted across 60 U.S. centers, 1563 patients with sepsis-induced hypotension refractory to at least 1 L of crystalloids were randomized to a restrictive fluid strategy with vasopressors prioritized as the primary hemodynamic intervention, versus a liberal fluid strategy with vasopressor initiation deferred when possible (liberal fluid group) for a 24-h period. Death by day 90 occurred in 14.0% in the restrictive fluid group and in 14.9% in the liberal fluid group (− 0.9%, 95% CI, -4.4 to 2.6; p = 0.61). Patients in the restrictive fluid group received significantly less intravenous fluid than those in the liberal fluid group (median difference, -2134 mL; 95% CI, − 2318 to − 1949) [38]. Although this study was not designed to directly compare early versus delayed vasopressor initiation, the findings support the safety of a vasopressor-prioritized approach, allowing comparable outcomes with reduced fluid administration and no observed difference in mortality. Importantly, in a post-hoc analysis of the CLOVERS trial among patients with advanced chronic kidney disease (eGFR less than 30 mL/min/1.73m²), the restrictive strategy was associated with a lower risk of death before discharge home by day 90 (21.7% vs. 39.4%, HR 0.5, 95% CI 0.29–0.85), suggesting a potential benefit in this population [39].

These findings support early correction of vascular tone, with fluid therapy tailored using dynamic markers of preload responsiveness [40, 41], an approach now endorsed by the Surviving Sepsis Campaign [35]. Before initiating norepinephrine, a low diastolic arterial pressure, or an elevated diastolic shock index (heart rate/diastolic arterial pressure) in tachycardic patients, may help identify vasomotor tone loss and patients more likely to benefit from early norepinephrine [42, 43].

How to administer norepinephrine: the place of initial peripheral administration

The shift toward early vasopressor initiation has been facilitated by the growing use of peripheral norepinephrine, allowing immediate vasopressor delivery without delaying treatment for central venous access. In a large prospective protocolized cohort, peripheral norepinephrine avoided central venous catheter placement for vasopressor administration in 51.6% of patients, corresponding to a median avoidance of one central-line day per patient. Peripheral norepinephrine was administered under a strict safety protocol with a maximum dose of 15 µg/min. Among 635 patients, peripheral norepinephrine was infused for a median duration of 5.8 [2.0–19.7] hrs. Although extravasation events were reported (75.8 per 1,000 PIVC-days), significant injury was rare, with no cases of tissue necrosis or need for surgical intervention [44]. Other institutional and emergency department studies similarly report a reduction in early central venous catheter utilization when peripheral norepinephrine is used as the initial route [45, 46].

A secondary analysis of the CLOVERS trial showed that vasopressors were initiated peripherally in 84.2% of eligible patients. Peripheral versus central initiation was associated with comparable 90-day mortality (26.1% vs. 37.0%; adjusted odds ratio, 0.67; 95% CI, 0.39–1.16). Complications related to peripheral vasopressor administration were rare and mild (3 of 490 patients), with no cases of tissue necrosis, whereas central venous catheter-related complications occurred in 12 of 322 patients within the first 72 h [47].

The burden of adrenergic support

Cardiac stress and septic cardiomyopathy

Catecholamines are essential to restore perfusion pressure in septic shock, yet their myocardial effects are time-dependent and non-neutral. Norepinephrine may transiently improve cardiac performance during early resuscitation, but prolonged adrenergic stimulation increases myocardial stress and may contribute to cardiac dysfunction [48]. In septic shock, left ventricular systolic dysfunction is common and may remain concealed by profound vasodilation. Indeed, as LVEF is a load-dependent marker of cardiac contractility, significant systolic myocardial dysfunction might be masked by the reduced afterload associated with systemic vasodilation. Restoration of arterial tone and thus LV afterload with norepinephrine can unmask global left ventricular hypokinesia [49]. Sepsis-induced cardiomyopathy is multifactorial, involving intrinsic mechanisms (PAMPs/DAMPs, inflammatory cytokines, nitric oxide excess, programmed cell death, calcium overload, mitochondrial dysfunction) and extrinsic factors such as altered loading conditions [50]. In an observational study of 20 patients who died from septic shock, catecholamines were suggested to further exacerbate myocardial injury through dose- and duration-dependent mechanisms, including mononuclear inflammatory cell infiltration, interstitial edema, contraction band necrosis, and subsequent fibrosis [51]. In addition, catecholamines may directly induce reversible myocardial dysfunction, including stress-induced (Takotsubo) cardiomyopathy [52, 53].

Myocardial injury has also been suggested to be aggravated by norepinephrine exposure. In an experimental model of septic shock, norepinephrine was associated with greater myocardial tissue inflammation and more severe myocardial injury compared with angiotensin II [54]. Similar findings were reported in a non-randomized clinical study comparing angiotensin II and norepinephrine as first-line vasopressor therapy in vasodilatory shock, in which markers of myocardial injury were higher in the norepinephrine-treated group [55].

Taken together, these data raise the hypothesis of a catecholamine-related myocardial toxicity, whose clinical consequences in critically ill patients remain incompletely characterized.

Arrhythmic consequences of adrenergic support

Catecholamine therapy during septic shock is associated with a substantial burden of cardiac arrhythmias, largely driven by β₁-adrenergic stimulation. In the SOAP II trial, arrhythmic events, predominantly atrial fibrillation (AF), occurred in 12.4% of patients treated with norepinephrine, compared with 24.1% in those receiving dopamine (p < 0.001), highlighting important differences in pro-arrhythmic risk among catecholamines.

In a large observational cohort study including 1,782 patients with sepsis, AF was a frequent complication, occurring in nearly one quarter of patients. The cumulative incidence of new-onset AF increased markedly with sepsis severity, reaching 10% in sepsis, 22% in severe sepsis, and up to 40% in septic shock. New-onset AF was independently associated with prolonged ICU stay and increased mortality, including a more than twofold higher risk of death when accounting for discharge as a competing event [56]. The contribution of catecholamine exposure to the burden of AF during septic shock has been highlighted in a meta-analysis including more than 3,000 patients. In this analysis, the use of vasopressin, and the consequent reduction in catecholamine exposure, was associated with a significantly lower risk of AF (relative risk 0.77; 95% CI, 0.67–0.88) without a clear mortality benefit [57]. Additionally, increased catecholamine exposure driven by higher MAP targets in the SEPSISPAM trial, comparing a MAP target of 80–85 mmHg with 65–70 mmHg, was associated with a higher incidence of AF, occurring in 6.7% of patients in the high-target group versus 2.8% in the low-target group (p = 0.02) [58].

Metabolic burden of catecholamines

Adrenergic stimulation increases metabolic flux by mobilizing energy substrates, promoting insulin resistance, increased hepatic glucose production, and lipolysis, leading to elevated free fatty acids [59, 60]. Stress hyperglycemia is largely adaptive, reflecting increased substrate availability during acute illness [61]. Although these substrates contribute to myocardial energy supply, their accumulation has been associated with pro-inflammatory and cardiotoxic effects in preclinical models [62, 63].

Catecholamines accelerate aerobic glycolysis, leading to increased lactate production independent of tissue hypoxia. This effect is particularly marked with epinephrine, without superior hemodynamic benefit compared with norepinephrine-based strategies [33, 34]. When metabolic demand exceeds mitochondrial capacity, this state may be associated with acidosis, potentially impairing myocardial contractility and vasopressor responsiveness [64, 65].

Adrenergic activation also promotes proteolysis and contributes to ICU-acquired weakness through multiple mechanisms, including mitochondrial and bioenergetic impairment, inflammatory signaling, and cellular alterations [66]. In an observational study, vasopressor exposure in mechanically ventilated patients has been independently associated with ICU-acquired weakness, although confounding factors are likely [67].

Adrenergic modulation of the immune response

Catecholamines exert broad immunomodulatory effects during septic shock, predominantly characterized in vitro and in animals’ models, with uncertain clinical translatability. Immune cells are influenced by both circulating catecholamines and sympathetic innervation of lymphoid organs [68]. Through β₂-adrenergic pathways, catecholamines reduce the release of pro-inflammatory cytokines (TNF-α, IL-1β, IL-6) while increasing IL-10, shifting the immune response toward an anti-inflammatory profile [69, 70].

Adrenergic stimulation also impairs key innate immune functions. High catecholaminergic exposure reduces neutrophil and macrophage phagocytosis and oxidative burst [71]. It also impairs adaptive immunity by inhibiting lymphocyte proliferation, promoting Th2 polarization, and reducing CD4 + T-cell responses while favoring regulatory T-cell expansion [72–74]. In parallel, catecholamines enhance bacterial growth and virulence through increased iron availability, biofilm formation, and activation of virulence-related genes [75–77].

These immunosuppressive and pro-pathogenic effects may be clinically relevant but remain unproven. Elevated catecholamine concentrations correlate with deeper immune paralysis, greater infection risk, and higher mortality [78]. Additionally, in a murine cecal ligation and puncture model, norepinephrine infusion was associated with a significantly higher bacterial load in the spleen and liver compared with PBS-treated animals, whereas no such increase was observed with vasopressin. A similar trend toward higher bacterial load in the blood was observed with norepinephrine, supporting a potential clinical impact on host defense [79].

Together, these data highlight that catecholamine exposure may contribute to cardiac, metabolic, and immune adverse effects (Table 1). In addition, an exponentially increasing relative risk of mortality has been observed with higher doses of norepinephrine, supporting a strategy of using norepinephrine at the lowest effective dose and considering a multimodal vasopressor approach [80–82] Fig. 2.

Table 1.

Summary of catecholamine-sparing strategies in septic shock

Intervention	Mechanism of action	Key evidence		SSC 2026
I. Optimization of hemodynamic targets and perfusion-guided resuscitation
MAP individualization	Titration of vasopressor to individual perfusion rather than a fixed MAP threshold	SEPSISPAM (multicenter RCT): no mortality benefit 65 trial (multicenter RCT): no mortality benefit, vasopressor-sparing positive OPTPRESS (multicenter RCT): possible harm with higher MAP target	Initial MAP target of 65 mmHg over higher MAP targets Strong recommendation Moderate certainty
Perfusion-guided resuscitation (CRT)	Vasopressor and fluid titration guided by capillary refill time	ANDROMEDA-SHOCK (multicenter RCT): no mortality benefit, vasopressor and fluid-sparing positive Zampieri et al. (Bayesian reanalysis): hypothesis-generating ANDROMEDA-SHOCK 2 (multicenter RCT): positive win-ratio outcome	Using CRT to guide resuscitation Conditional recommendation Low certainty
II. Multimodal vasopressor strategies
Vasopressin	V1a receptor agonism Non-adrenergic vasoconstriction	VASST (multicenter RCT): no mortality benefit; positive in low-NE subgroup VANISH (multicenter RCT): no mortality benefit, positive on RRT-free days VANCS II (multicenter RCT): no mortality benefit	Add vasopressin on escalating dose of norepinephrine Conditional recommendation Moderate certainty
Angiotensin II	AT-1R agonism Restore vascular tone via renin-angiotensin system	ATHOS-3 (multicenter RCT): no mortality benefit, vasopressor-sparing positive Bellomo et al. – renin subgroup (post-hoc analysis): improved survival in high renin patients Tumlin et al. – RRT subgroup (post-hoc analysis): improved survival in patients under RRT at drug initiation	Using norepinephrine as first line agent over angiotensin II Strong recommendation Very low certainty
cDPP3 inhibition	cDPP3 cleaves angiotensin II cDPP3 inhibition preserves RAAS signaling	Garcia et al. (preclinical (swine model)), reduce vasopressor requirements	Not addressed in SSC 2026
III. Adjunctive therapies
Corticosteroids	Restoration of vasopressor sensitivity. Upregulation of adrenergic receptors. Anti-inflammatory	APROCCHSS (multicenter RCT): mortality benefit, vasopressor-sparing positive	Using IV corticosteroids in septic shock Conditional recommendation Low certainty
Beta-adrenergic pathway	Beta1-selective blockade Reduce tachycardia and myocardial O2 consumption	Morelli et al. (single-center RCT): mortality benefit—limited generalizability STRESS-L (multicenter RCT): possible harm—stopped early LANDI-SEP (multicenter RCT): no mortality benefit Sato et al. (meta-analysis): no mortality benefit McChesney et al. (meta-analysis): mortality benefit—low certainty	Against using beta-blockers as a treatment for septic shock Conditional recommendation Very low certainty
Methylene blue	Inhibition of NO-mediated vasoplegia	Ibarra-Estrada et al. (single-center RCT): no mortality benefit, vasopressor-sparing positive	Insufficient evidence to make a recommendation
Extracellular histone neutralization	Neutralization of DAMPS	Garcia et al. (preclinical study in sheep): reduce vasopressor requirements Bellomo et al. (safety study)	Not addressed in SSC 2026
PMX hemoperfusion (polymyxin B)	Endotoxin adsorption; Reducing LPS-driven vasopressor dependency in endotoxin-defined phenotypes	EUPHRATES (multicenter RCT): no mortality benefit in unselected patients TIGRIS (multicenter RCT): high probability of mortality benefit in biomarker-selected population	Against using blood purification therapies Conditional recommendation Very low certainty

Bold indicates key clinical studies and recommendationsBold italics indicate the strength of recommendation andcertainty of evidence according to the Surviving Sepsis Campaign (SSC) 2026 guidelines

Fig. 2.

Beneficial and adverse effects of catecholamine

Toward catecholamine-sparing strategies

Avoiding unnecessary vasopressor exposure

Reducing blood pressure targets

In septic shock, outcomes appear to be driven primarily by avoiding prolonged severe hypotension. Sustained hypotension is strongly and independently associated with mortality in distributive shock. In a large cohort of more than 5,000 patients, both the severity and duration of hypotension were directly related to ICU mortality, with progressively stronger effects as MAP decreased below 65, 60, and 55 mmHg [83].

In contrast, targeting higher MAP levels has not improved outcomes. In the SEPSISPAM trial, Asfar et al. compared a higher MAP target of 80–85 mmHg with a standard target of 65–70 mmHg and found no difference in 28- or 90-day mortality [58]. Post-hoc analyses further suggested that increasing MAP does not necessarily translate into improved tissue perfusion. Higher MAP targets did not modify the trajectory of lactate levels or mottling score. Moreover, persistent mottling for ≥ 6 h remained strongly associated with mortality regardless of the achieved MAP [84].

More recent data suggest that tolerating lower MAP levels in selected patients may safely reduce catecholamine exposure. An individual patient-data meta-analysis of RCTs comparing higher MAP targets (≈75–85 mmHg) with lower targets (≈60–70 mmHg) in vasodilatory shock found no survival benefit associated with higher targets (28-day mortality OR 1.15, 95% CI 0.87–1.52) and no interaction with chronic hypertension [85]. In a multicenter RCT in older critically ill patients, a vasopressor-sparing strategy tolerating lower pressures achieved similar 90-day mortality while reducing vasopressor exposure [86].

A recent multicenter RCT conducted in Japan included 518 older patients (≥ 65 years) with septic shock and was terminated early following an interim analysis suggesting increased 90-day mortality in the high-MAP target group (80–85 mmHg) compared with the standard target group (65–70 mmHg). Mortality reached 39.3% in the high-target group compared with 28.6% in the standard target group (risk difference 10.7%; 95% CI, 2.6–18.9). Patients in the standard target group also had a higher number of catecholamine-free days at day 28. Notably, no increase in arrhythmic events was observed in the high-target group [87].

Together, these data support the non-inferiority of lower MAP targets in selected patients, without demonstrating superiority over standard thresholds. An initial MAP around 65 mmHg remains reasonable, but lower targets may be acceptable in the absence of clinical signs of tissue hypoperfusion [7]. Additionally, patients with pre-existing hypertension were suggested in the SEPSISPAM trial to benefit from higher blood pressure targets, a strategy that was incorporated into the personalized hemodynamic protocol of the ANDROMEDA-SHOCK 2 trial [58, 88]. In this study, a MAP test was performed in patients with chronic hypertension by transiently increasing norepinephrine to achieve a MAP of 80–85 mmHg for 1 h, while assessing capillary refill time [88].

Optimization of cardiac output and perfusion-guided resuscitation

Once a minimal perfusion pressure has been achieved, resuscitation should primarily aim at restoring adequate tissue perfusion rather than achieving predefined macrocirculatory targets such as a specific MAP or cardiac output [89, 90].

In recent years, hemodynamic resuscitation strategies integrating peripheral perfusion markers and fluid responsiveness assessment have been developed. The ANDROMEDA-SHOCK trial randomized 424 patients with septic shock to a resuscitation strategy guided either by normalization of capillary refill time (CRT), assessed every 30 min, or by lactate clearance, defined as a ≥ 20% decrease every 2 h during an 8-h intervention period. Both groups were managed using a standardized protocol after an initial resuscitation to maintain a MAP ≥ 65 mmHg. Fluid responsiveness assessment was systematically performed as the first step, with repeated 500 mL crystalloid boluses administered to responders until the predefined target was achieved or until fluid unresponsiveness or central venous pressure limits were reached. 28-day mortality did not differ significantly between the CRT-guided group and the lactate-guided group (34.9% vs. 43.4%; hazard ratio, 0.75 [95% CI, 0.55 to 1.02]; p = 0.06). However, the CRT-guided strategy was associated with lower fluid administration during the first 8 h of resuscitation (− 408 (95% CI − 705 to − 110) mL, p = 0.01) and with lower SOFA score at 72 h (− 1.00 (95% CI, − 1.97 to − 0.02); p = 0.045) [91]. A Bayesian reanalysis estimated a high posterior probability of mortality reduction with the CRT-guided approach [92]. However, this analysis remains exploratory and no adequately powered confirmatory trial has demonstrated a mortality benefit to date.

This perfusion-guided resuscitation strategy was recently evaluated in a large multicenter international RCT assessing a personalized hemodynamic resuscitation protocol [88]. The intervention relied on the integrated assessment of pulse pressure, diastolic arterial pressure, fluid responsiveness, and bedside echocardiography to guide the use of fluids, vasopressors, and inotropes, with CRT as the primary perfusion target. The primary endpoint, a hierarchical composite of mortality, duration of organ support, and length of hospital stay analyzed using the win ratio method, favored the intervention group, with a win ratio of 1.16 (95% CI, 1.02–1.33; p = 0.04), mainly driven by the duration of organ support (26.4% vs 21.1%). Among the 1,501 patients with septic shock included in the trial, this strategy was not associated with a difference in 90-day all-cause mortality (32.1% vs. 33.2%; HR 0.93, 95% CI 0.78–1.11). However, the personalized strategy was associated with a reduction in vasopressor exposure, with shorter duration of vasopressor support within 28 days (mean difference -0.95 days; 3.9 ± 4.5 vs. 4.7 ± 5.5). Taken together, these data support the integration of peripheral perfusion markers into individualized hemodynamic management, potentially allowing a reduction in fluids and catecholamine exposure.

Multimodal vasopressor approach

Vasopressin

Vasopressin deficiency: rationale for low-dose use in septic shock

A relative vasopressin deficiency has been observed in approximately one-third of patients with septic shock [93]. Although the primary outcome of the VASST trial, which compared vasopressin plus norepinephrine to norepinephrine alone, was negative, pre-specified subgroup analyses showed improved survival in patients receiving low-dose norepinephrine (< 15 µg/min) at the time of second vasopressor initiation [94]. This observation was further supported by an observational study of 1,610 patients with septic shock, which found a 20.7% increase in in-hospital mortality for every 10 µg/min increase in norepinephrine-equivalent dose prior to vasopressin initiation [95]. Similarly, the VANISH and VANCS II trials did not demonstrate a mortality benefit of vasopressin use, although VANISH suggested reduced renal replacement therapy use [96, 97].

Recent Surviving Sepsis Campaign guidelines recommend adding vasopressin (fixed dose 0.03 U/min) to reduce catecholamine exposure, typically when norepinephrine requirements reach approximately 0.25 µg/kg/min [35]. However, previous RCTs have not consistently excluded patients at risk of vasopressin-related adverse effects, such as those with left ventricular systolic dysfunction, which may have diluted potential benefits in selected populations [98]. As an example, in a post hoc analysis of the VASST trial among 241 patients managed with a pulmonary artery catheter, despite no difference in cardiac index between groups, inotropic agents were used more frequently in the vasopressin group than in the norepinephrine group, possibly reflecting the loss of the inotropic effects of catecholamines [99].

Optimizing vasopressin therapy with artificial intelligence

Recently, in a study involving over 14,000 critically ill patients with septic shock across 232 hospitals, a reinforcement learning model, developed to optimize vasopressor management, recommended initiating vasopressin more frequently, earlier, at lower norepinephrine doses, and in patients with lower organ failure scores compared to standard clinical practice [100]. Notably, patients whose vasopressin treatment closely aligned with the model’s recommendations had significantly lower in-hospital mortality, with an adjusted odds ratio of 0.81 [95% CI, 0.73–0.91]) [100]. However, these observational findings require prospective validation before clinical implementation.

Vasopressin use guided by clinical phenotype

In a global survey involving 1,919 intensivists from 124 countries, reported indications for vasopressin use included vasoplegic shock without reduced left ventricular systolic function (51%), and septic shock with high cardiac output and low systemic vascular resistance (49%) [101]. These results reflect a growing interest among ICU clinicians in tailoring vasopressor therapy to specific hemodynamic profiles and clinical phenotypes, rather than relying solely on blood pressure targets. Catecholamine-sparing strategies may be particularly relevant in patients with AF and preserved cardiac function. Indeed, a systematic review and meta-analysis of 23 trials including over 3,000 patients with distributive shock, in which vasopressin use was associated with a significantly lower risk of AF compared to catecholamines alone (relative risk, 0.77 [95% CI, 0.67–0.88]) [57].

Within this framework, septic shock is associated with a relative vasopressin deficiency, and vasopressin supplementation may be effective in reducing norepinephrine exposure in patients with profound vasoplegia and preserved cardiac output. By restoring vascular tone through non-adrenergic pathways, vasopressin may thereby limit catecholamine-related adverse effects, particularly in patients with high norepinephrine concentration [102].

Modulation of the renin angiotensin system

Pathophysiology and rationale for angiotensin II use

Recent advances have been made in understanding disturbances in the renin-angiotensin system (RAAS) during septic shock [103, 104]. The main effector of the classical RAAS pathway is angiotensin II, which acts through the angiotensin II receptor 1 (AT-1R), inducing effects such as vasoconstriction, pro-inflammatory responses, fibrosis, and the release of vasopressin and aldosterone [103]. Renin is secreted by the juxtaglomerular apparatus in response to hypoperfusion or sympathetic activation, leading to the conversion of angiotensinogen, produced by the liver, into angiotensin I, which is then cleaved into angiotensin II by the angiotensin-converting enzyme (ACE) [105].

During sepsis and vasodilatory shock, RAAS dysregulation has been described in a subset of patients with low angiotensinogen concentrations and increased renin concentrations [106, 107]. Increased renin concentrations have been correlated with an increased angiotensin I/angiotensin II ratio, which in turn has been associated with higher mortality [107, 108]. This increase in the angiotensin I/angiotensin II ratio may be explained by decreased ACE activity, increased ACE2 activity, or the degradation of angiotensin II by dipeptidyl peptidase 3 [109–112].

Taken together, these findings suggest in some patients a functional defect in angiotensin II signaling, supporting the use of exogenous angiotensin II administration. This is further supported by a defect at the tissue level, as sepsis-associated acute kidney injury (AKI) is associated with a decrease in tissular AT-1R and can be prevented by angiotensin II infusion in pre-clinical models [113]. However, the specific phenotype of patients most likely to benefit from angiotensin II therapy remains incompletely characterized at the present time, although elevated renin concentrations may represent a promising enrichment biomarker [107].

Angiotensin II reduces catecholamine exposure in vasodilatory shock

Angiotensin II has been tested in a RCT in patients with catecholamine-resistant vasodilatory shock and was shown to increase MAP 3 h after drug initiation and reduce the norepinephrine equivalent dose [114]. Post-hoc analysis of this trial suggested an increased rate of renal replacement therapy (RRT) liberation and reduced 28-day mortality in patients with AKI requiring RRT [115]. Another post-hoc study reported that in patients with baseline renin concentrations above the median of the population, angiotensin II was associated with a significant reduction in 28-day mortality compared to placebo [107].

Altogether, the use of angiotensin II during septic shock is supported by a pathophysiological background suggesting a defect in angiotensin II signaling, and by post-hoc analyses of a RCT that show improved outcomes in selected patients. These include patients with severe AKI requiring RRT or not, and patients with high renin concentrations, suggesting the need for a tailored approach to vasopressor use based on clinical or biomarker phenotypes [116–118]. These findings, derived from post-hoc analyses, are hypothesis-generating and require prospective validation before informing routine clinical practice. However, in the recent 2026 SSC guidelines, angiotensin II is neither recommended as first-line vasopressor, nor incorporated into the standard escalation strategy.

Inhibition of dipeptidyl peptidase 3

Dipeptidyl peptidase 3 (DPP3) is a ubiquitously expressed metallopeptidase involved in the defenses against oxidative stress and highly conserved across species. DPP3 can be released into the bloodstream during cellular injury or death and contributes to the degradation of components of the RAAS [119]. Circulating DPP3 (cDPP3) concentrations have been shown to be a biomarker of AKI within the first 7 days, with high concentrations associated with increased use of RRT and higher 28- and 90-day mortality in a multinational cohort of 585 patients with severe sepsis and septic shock [112]. Beyond its value as a biomarker, cDPP3 may also play a direct pathophysiological role, as it degrades circulating angiotensin II [110]. Pharmacological inhibition of cDPP3 with a humanized monoclonal antibody, procizumab (invobenitug), is currently under development as a therapeutic strategy in shock (NCT06832722). In a swine model of septic shock, where cDPP3 inhibition was associated with reduced catecholamine requirements, improved fluid balance, and preservation of the angiotensin I/angiotensin II ratio [109]. This approach remains experimental, with human data currently pending. The future strategy of cDPP3 inhibition could range from a biomarker-guided therapeutic approach, in which inhibition is reserved for patients with high cDPP3 concentrations, to its broad use in patients with circulatory failure, with or without concurrent angiotensin II therapy.

Beyond perfusion & vasopressors: modulating vasoplegia with adjunctive therapies

Corticosteroids

Corticosteroids have become a cornerstone in the management of septic shock in recent years, although RCTs and meta-analyses have reported conflicting results. In a multicenter, double-blind, RCT including 1,241 patients with septic shock, treatment with hydrocortisone plus fludrocortisone for 7 days, compared with placebo, was associated with a significant reduction in mortality [120]. 90-day mortality was 43.0% (264 of 614 patients) in the hydrocortisone-fludrocortisone group versus 49.1% (308 of 627 patients) in the placebo group (p = 0.03), corresponding to a relative risk of 0.88 (95% confidence interval, 0.78–0.99). Importantly, catecholamine exposure was also reduced, with the number of vasopressor-free days up to day 28 being significantly higher in the hydrocortisone-fludrocortisone group than in the placebo group (17 vs. 15 days, p < 0.001).

Modulating Adrenergic Responsiveness

Excessive adrenergic stimulation in septic shock promotes tachycardia, increases myocardial oxygen consumption, and impairs cardiovascular efficiency [121]. Short-acting β1-selective blockers have therefore been investigated to modulate this response. In a single-center RCT (n = 154), esmolol decreased norepinephrine requirements and was associated with lower 28-day mortality (49.4% vs 80.5%; HR 0.39, 95% CI 0.26–0.59), although an unusually high control-group mortality and very large volume of resuscitation limit the generalizability of these findings [122]. These promising results were not, however, replicated in larger multicenter trials. The STRESS-L trial (n = 126) evaluated landiolol, with a shorter half-life and greater β1-selectivity, in patients with established septic shock, tachycardia, and prolonged norepinephrine support (≥ 0.1 μg/kg/min for > 24 h). It was stopped prematurely due to a signal of possible harm. Landiolol did not reduce organ failure assessed by mean SOFA score over 14 days (8.8 vs 8.1, p = 0.24), and 28-day mortality was numerically higher in the landiolol group (37.1% vs 25.4%, p = 0.16) [123]. The LANDI-SEP RCT randomized 196 patients with septic shock and persistent tachycardia in 20 European sites. The combined primary endpoint, predefined heart rate target (80–94) maintained 24 h without increasing vasopressor requirements was achieved more frequently than standard care (39.8% vs 23.5%), but without any significant difference in 28-day mortality [124]. A post-hoc subgroup analysis raised concern that tachycardia may reflect a compensatory mechanism in some patients, with a numerically higher mortality in the landiolol group among patients with sinus tachycardia (47.2% vs 39.7%, p = 0.013), contrasting with a possible benefit in those with atrial fibrillation [124]. The most recent meta-analyses remain divided. One analysis of 8 RCTs (n = 885) found no significant mortality benefit (RR 0.84, 95% CI 0.68–1.02) [125], while a broader analysis of 12 RCTs (n = 1,170) suggested a possible reduction in 28-day mortality (RR 0.76, 95% CI 0.62–0.93) and in new-onset tachyarrhythmias, though with low-to-moderate certainty of evidence [126].

Overall, β-blockade may reduce adrenergic burden and improve hemodynamic efficiency in selected patients, but its effect on outcomes remains inconsistent. Routine use cannot be recommended, and administration should be limited to hemodynamically stabilized patients pending better identification of those most likely to benefit [127]. The ongoing HyperBetashock trial is evaluating landiolol in septic shock patients with persistent tachycardia and hypercontractility. By targeting heart rate reduction with a short-acting β1-blocker, it is expected to clarify the role of β-blockade in a selected hyperadrenergic phenotype (NCT04748796).

Targeting the nitric oxide pathway: methylene blue

In response to inflammation, the inducible isoform of nitric oxide synthase (iNOS) is activated, leading to increased nitric oxide (NO) production from arginine. The vascular effects of NO are primarily mediated by the activation of soluble guanylate cyclase (sGC), which stimulates the synthesis of cyclic guanosine monophosphate (cGMP), ultimately resulting in the reduction of vascular tone [128]. Consequently, several strategies have been developed to mitigate the harmful hemodynamic consequences of excessive NO release.

One such strategy involved the administration of L-NMMA, a non-selective nitric oxide synthase inhibitor, tested in a RCT in septic shock. However, patients randomized to receive L-NMMA experienced a significantly higher mortality rate compared to those receiving placebo (50.9% vs 49%, p < 0.001) [129]. This excess mortality has been attributed to the broad inhibition of NO, which may have disrupted several physiological processes involved in host defense and organ function, such as the maintenance of microcirculatory perfusion, the NO-dependent antimicrobial activity of immune cells, and the neutralization of reactive oxygen species [130].

An alternative strategy has involved the use of methylene blue (MB), an endothelial and iNOS inhibitor, as well as an inhibitor of sGC, which can attenuate the vasodilatory effects of NO [131]. In a recent single-center RCT involving 91 patients with septic shock, patients were randomized to receive either 100 mg of MB diluted in 500 ml of 0.9% sodium chloride over 6 h once daily for three consecutive days, or placebo. MB significantly reduced the time to vasopressor discontinuation (69 h [IQR 59–83] in the MB group vs 94 h [IQR 74–141] in the placebo group; p < 0.001). Patients treated with methylene blue also had a significantly lower cumulative fluid balance by day 4, a shorter ICU stay, and a trend toward reduced mortality (33% vs 46%; hazard ratio 0.76 [95% CI, 0.55–1.05]; p = 0.23). The only notable adverse effect was a transient green discoloration of the urine, with no other significant safety concerns reported [132]. However, given the monocentric design of the trial, the potential impact of urine discoloration on blinding integrity, as well as prior deleterious findings associated with NO inhibition, these findings require confirmation in larger, multicenter RCTs before being implemented in clinical practice.

Emerging immunomodulatory approaches

Another strategy currently under development focuses on modulating the innate immune response by targeting damage-associated molecular patterns (DAMPs). Among these, extracellular histones released during cellular injury or NETosis act as potent DAMPs and contribute to inflammation, endothelial dysfunction, and organ injury [133–135]. In a large experimental ovine model of septic shock, treatment with the histone-neutralizing compound mCBS (sodium-β-O-Methyl cellobioside sulfate) reduced vasopressor requirements to maintain MAP, decreased arterial lactate concentrations, attenuated the increase in circulating interleukin-6, and reduced the severity of AKI [136]. A pilot study has evaluated the safety, tolerability, and pharmacokinetics of mCBS (STC3141), administered over 72 h in 26 critically ill patients with sepsis, supporting the feasibility of translating DAMP-targeted therapies into the clinical setting [137].

Extracorporeal blood purification techniques, particularly hemoadsorption, have also been proposed as adjunctive strategies to modulate the dysregulated host response and potentially reduce vasopressor requirements [138]. Polymyxin B hemoperfusion has been evaluated in two RCTs, in which no survival benefit or improvement in organ failure was observed [139, 140]. Nevertheless, several authors have suggested that patients with a high endotoxin burden, as assessed by the endotoxin activity assay (EAA), may represent a subgroup more likely to benefit from PMX therapy [138]. In a post hoc analysis of the EUPHRATES trial restricted to patients with moderate endotoxemia (EAA 0.60–0.89), a range considered amenable to endotoxin removal, 26.1% of the patients in the PMX group died compared with 36.8% in the sham group (risk difference 10.7%; odds ratio 0.52; 95% confidence interval 0.27–0.99; p = 0.047). PMX treatment was also associated with greater improvements in MAP (median 8 mmHg [-0.5, 19.5] vs. 4 mmHg [-4.0, 11]; p = 0.04) [141]. The TIGRIS trial prospectively evaluated the addition of PMX hemoperfusion to standard care in 157 patients with septic shock and moderate endotoxemia (EAA 0.60–0.89), randomized 2:1 across 19 U.S. centers. Using a Bayesian framework analysis, 28-day mortality was 39% in the PMX group versus 45% in the control group (APACHE-II adjusted OR 0.67, 95% CrI 0.39–1.08; posterior probability of benefit 95.3%). However, no significant differences were observed in mean arterial pressure, vasopressor dose, or duration of vasopressor therapy between groups [142] Fig. 3.

Fig. 3.

Catecholamine-sparing strategies in septic shock. AF, atrial fibrillation; AKI, acute kidney injury; Ang, angiotensin; AP, arterial pressure; cDPP3, circulating dipeptidyl peptidase 3; CO, cardiac output; CRT, capillary refill time; DAMPs, damage-associated molecular patterns; EAA, endotoxin activity assay; HC, hydrocortisone; HR, heart rate; MAP, mean arterial pressure; NE, norepinephrine; NO, nitric oxide; PMX, polymyxin B; PP, pulse pressure; RAAS, renin–angiotensin–aldosterone system; RRT, renal replacement therapy; sGC, soluble guanylate cyclase; SSC, Surviving Sepsis Campaign; SV, stroke volume

Conclusion

Catecholamines, and norepinephrine in particular, remain the cornerstone of septic shock resuscitation because of their rapid and proven efficacy in restoring perfusion pressure. However, their use is associated with relevant cardiac, metabolic, and immunological adverse effects that should not be overlooked. Reducing catecholamine exposure through individualized hemodynamic targets, multimodal vasopressor strategies, and adjunctive therapies is a promising perspective, but requires further dedicated studies.

Abbreviations

AF

Atrial fibrillation

ACE

Angiotensin-converting enzyme

AKI

Acute kidney injury

Ang

Angiotensin

AP

Arterial pressure

APACHE-II

Acute physiology and chronic health evaluation II

ARDS

Acute respiratory distress syndrome

AT1R

Angiotensin II type 1 receptor

AVP

Arginine vasopressin

cDPP3

Circulating dipeptidyl peptidase 3

CI

Confidence interval

CO

Cardiac output

CRT

Capillary refill time

CVP

Central venous pressure

DAMPs

Damage-associated molecular patterns

DAP

Diastolic arterial pressure

DPP3

Dipeptidyl peptidase 3

EAA

Endotoxin activity assay

eGFR

Estimated glomerular filtration rate

GMP

Guanosine monophosphate

HC

Hydrocortisone

HR

Heart rate

ICU

Intensive care unit

IL

Interleukin

iNOS

Inducible nitric oxide synthase

IVC

Inferior vena cava

LA

Left atrium

LV

Left ventricle

LVEF

Left ventricular ejection fraction

MAP

Mean arterial pressure

MB

Methylene blue

MSFP

Mean systemic filling pressure

NE

Norepinephrine

NETosis

Neutrophil extracellular trap formation

NO

Nitric oxide

PAMPs

Pathogen-associated molecular patterns

Pa

Arterial pressure

PBS

Phosphate-buffered saline

Pcrit

Critical closing pressure

PIVC

Peripheral intravenous catheter

Pmsf

Mean systemic filling pressure

PMX

Polymyxin B hemoperfusion

PP

Pulse pressure

RA

Right atrium

RAAS

Renin–angiotensin–aldosterone system

RCT

Randomized controlled trial

RRT

Renal replacement therapy

RV

Right ventricle

sGC

Soluble guanylate cyclase

SOFA

Sequential organ failure assessment

SSC

Surviving Sepsis Campaign

SV

Stroke volume

SVC

Superior vena cava

SVR

Systemic vascular resistance

TNF-α

Tumor necrosis factor-alpha

V1aR

Vasopressin type 1a receptor

Author contributions

A. D. and B. G. drafted the manuscript. All authors critically revised the manuscript and approved the final version.

Funding

No external funding was received for this work.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable.

Consent for publication

Not applicable.

Competing interests

B. G. has received honoraria for presentations and participation in meetings related to DPP3, angiotensin II, and vasopressin, outside the scope of this work. S. P. received payments for lectures from Viatris and AOP-Health. The other authors declare no competing interests.