ATHOS-3

Published

Khanna A et al. "Angiotensin II for the Treatment of Vasodilatory Shock". New Engl J Med. 2017. 377:419-30.
PubMed • Full text • PDF

Clinical Question

In patients with severe vasodilatory shock requiring high-dose catecholamines, does angiotensin II result in improvement in mean arterial pressure (MAP) compared to placebo?

Bottom Line

In patients with severe vasodilatory shock (MAP 55-70 despite 0.2ug/kg/min norepinephrine or equivalent), administration of angiotensin II is associated with a 45% absolute increase in MAP response (defined as MAP increase ≥ 10mmHg or MAP > 75mmHg) when compared to placebo.

Major Points

Shock is defined as any process resulting in insufficient blood pressure to provide adequate organ perfusion. The most common form of shock is vasodilatory shock, commonly defined as shock in the setting of peripheral vasodilation and intact cardiac output. In addition to treating the underlying cause of vasodilatory shock, catecholamine (e.g., dopamine, norepinephrine, epinephrine) and non-catecholamine (e.g., vasopressin) vasopressors are agents that combat vasodilatory shock by inducing peripheral vasoconstriction. However, these agents are not uniformly effective in reversing vasodilatory shock, and their use is associated with significant side effects including limb ischemia and cardiac arrhythmia. These difficulties are reflected in the fact that patients with vasodilatory shock requiring high-dose vasopressors have a mortality rate approaching 50% at 1 month.

As a result, more effective vasopressor options are needed. Angiotensin II is a naturally occurring hormone secreted as part of the renin-angiotensin system that results in powerful systemic vasoconstriction. An early study of angiotensin II in patients with vasodilatory shock suggested favorable effects of this agent on systemic mean arterial pressure (MAP). A larger randomized controlled trial was needed to assess the efficacy of this agent in a broader spectrum of patients with severe vasodilatory shock.

The 2017 Angiotensin II for the Treatment of High-Output Shock (ATHOS-3) trial randomized 321 severe vasodilatory shock patients to either angiotensin II or placebo and assessed for a primary outcome of MAP response (defined as MAP increase ≥ 10mmHg or MAP > 75mmHg without increase in background vasopressors) at 3 hours. In ATHOS-3, 70% of patients who received angiotensin II met criteria for a MAP response (a 45% absolute increase compared to placebo). There was an associated significant reduction in catecholamine doses in patients receiving angiotensin II. At 48 hours, mean improvement in the cardiovascular Sequential Organ Function Assessment (SOFA) score, a general indicator of systemic organ function, was greater in the angiotensin II group (partially driven by lower catecholamine dosing, which is reflected in this score). Angiotensin II was well-tolerated, with a lower adverse event rate compared to placebo. As an exploratory endpoint, there was a trend toward reduced mortality with angiotensin II, with statistically non-significant 9% absolute reduction in death at 30 days.

In summary, ATHOS-3 provides fairly compelling evidence that angiotensin II is safe and effective in reducing peripheral vasodilation and improving hemodynamics in severe vasodilatory shock. Further studies are needed to determine whether the effects of angiotensin II translate into improved morbidity and mortality in this condition.

Guidelines

As of August 2017, no guidelines have been published that reflect the results of this trial.

Design

• Prospective, multi-center, double-blind, randomized controlled trial
• N=321
  • Angiotensin II (n=163)
  • Placebo (n=158)
• Setting: 75 centers in North America, Australia, Asia, and Europe
• Enrollment: May 2015 to January 2017
• Duration follow-up: 28 days
• Analysis: Modified intention-to-treat (study drug started)
• Primary outcome: MAP response at 3 hours (defined as MAP increase ≥ 10mmHg or resultant MAP > 75mmHg)

Population

Inclusion Criteria

• Age ≥ 18 years
• Catecholamine-resistant hypotension defined as > 0.2ug/kg/min of norepinephrine or equivalent for 6-48 hours to maintain MAP 55-70mmHg
• Central venous access and arterial catheter present
• Indwelling urinary catheter present
• Received at least 25mL/kg of crystalloid or colloid equivalent over the previous 24-hour period and felt by treating investigator to be adequately volume resuscitated
• Features of high-output shock defined as 1 of the following criteria:
  • Central venous O2 saturation > 70% and central venous pressure > 8mmHg
  • Cardiac index > 2.3 L/min/m2

Exclusion Criteria

• Burn covering > 20% of total body surface area
• Cardiovascular SOFA score ≤ 3
• Acute occlusive coronary syndrome requiring intervention
• On VA ECMO or previously on ECMO for less than 12 hours
• Liver failure with MELD score ≥ 30
• History of asthma with active bronchospasm requiring bronchodilation (unless intubated)
• Acute mesenteric ischemia or a history of mesenteric ischemia
• History of, presence of, or high suspicion for aortic dissection or abdominal aortic aneurysm
• Requirement for 500mg daily or more of hydrocortisone or equivalent
• Presence of Raynaud's phenomenon, systemic sclerosis, or vasospastic disease
• Expected lifespan of < 12 hours
• Active bleeding with either of:
  • anticipated need (within 48 hours of initiation of the study) for transfusion of > 4 units of packed red blood cells
  • hemoglobin < 7g/dL or any other condition that would contraindicate serial blood sampling
• Absolute neutrophil count < 1000 cells/mm3
• Known allergy to mannitol
• Currently participating in another clinical trial
• Known to be pregnant at the time of screening

Baseline Characteristics

All patients

• Demographics: age 64, male 60.7%, BMI ≥ 30 44.3%
• Shock: MAP 66.3, APACHE II score 28, ScvO2 77.0, CVP 12mmHg, CI 3.1 L/min/m2, septic 80.7%, ARDS 28.4%, vasopressin use 6 hours before randomization 69.8%, norepinephrine dose 0.34 ug/kg/min
• Labs: Albumin 2.3, MELD 22
• Medication exposures: ACE inhibitor 9.3%, ARB 6.9%

Interventions

• Randomized 1:1 to angiotensin II or matching placebo
  • Randomization stratified according to MAP (< 65 vs. ≥ 65) and APACHE II score (≤ 30, 31 to 40, or ≥ 41)
• Investigators, research personnel, patients, families, and the sponsor were all blinded to treatment assignment
• Baseline MAP established as the mean of 3 determinations performed 30, 15, and 0 minutes prior to treatment
• Study infusions initiated at a rate equivalent to 20ng of angiotensin II per kg per minute and adjusted during the first 3 hours to increase MAP to at least 75mmHg
  • Maximum study infusion rate 300ng/kg/min of angiotensin II
• During study infusion adjustment, standard-of-care vasopressors could only be adjusted for safety reasons
  • If vasopressors were adjusted, the patient was deemed a MAP non-responder
• MAP response determined with averaged triplicate MAP measurements performed at 2 hours 45 minutes, 3 hours, and 3 hours 15 minutes
• After 3 hours and 15 minutes, the study drug or placebo and other vasopressors were adjusted to maintain a target MAP 65-75 mmHg
  • Angiotensin II or placebo could be adjusted to an infusion rate equivalent to 1.25-40ng/kg/min of angiotensin II
• At hour 48, the study infusion was discontinued according to a protocol-specified taper
  • If the background vasopressor dose was subsequently increased to more than 0.1ug/kg/min of norepinephrine equivalent or the patient's condition became unstable, study drug could be resumed for up to 7 days
  • If study drug was discontinued for > 48 hours, it could not be resumed under any circumstances

Outcomes

Comparisons are angiotensin II vs. placebo

Primary Outcomes

MAP response (3 hours)

114 (69.9%) vs. 37 (23.4%) [OR 7.95, p<0.001]

Secondary Outcomes

Mean change in cardiovascular SOFA score (48 hours)

-1.75 vs. -1.28 [p=0.01]

Mean change in total SOFA score (48 hours)

1.05 vs. 1.04 [p=0.49]

Mean change in norepinephrine-equivalent dose (3 hours)

-0.03 vs. 0.03 [p<0.001]

All cause mortality (7 days)

47 (29%) vs. 55 (35%) [OR 0.78, 95% CI 0.53-1.16, p=0.22]

All cause mortality (28 days)

75 (46%) vs. 85 (54%) [OR 0.78, 95% CI 0.57-1.07, p=0.12]

Adverse Events

Adverse events

142 (87.1%) vs. 145 (91.8%)

Serious adverse events

99 (60.7%) vs. 106 (67.1%)

Criticisms

• Small study size limits power to detect differences in clinical outcomes including mortality
• Short duration of follow up limits ability to make conclusions regarding longer-term efficacy and safety of angiotensin II
• Study drug titration period may allow for inadvertent unblinding (which was not assessed) by providers due to observable response in MAP, although this is mitigated somewhat by a MAP response in 1/4 patients receiving placebo

Funding

• Study supported by La Jolla Pharmaeuticals
• Authors with multiple ties to industry