Blog → ANDROMEDA-SHOCK-2 REDUX

ANDROMEDA-SHOCK-2 REDUX

Listen to this article
0:00 –:–

Merry Christmas!

From all of us here at JournalFeed, we wish you a very Merry Christmas. Thank you for reading, subscribing, and sharing in this journey with us. We can’t wait to serve you (and through you, your patients) even better in 2026!

Written by Samuel Rouleau

Spoon Feed

Discussion around ANDROMEDA-2 has been lively—is this a positive or negative trial? Enclosed is a more in-depth distillation of my thoughts and interpretations. So, instead of your typical “Spoon Feed,” grab your favorite pint of ice cream, sit down on the couch, and indulge in this longer-than-normal post.

Deep Dive

I understand how ANDROMEDA-2 could be considered a negative trial. Essentially, there was no difference in mortality, and the composite outcome was mainly driven by a 1-day difference in renal replacement therapy between the capillary refill time personalized hemodynamic resuscitation group (CRT-PHR) and the usual care group. Notably, the intervention arm lasted only for 6 hours, so it is remarkable to me that there was a difference, and I find it almost impossible to imagine that there was no clinical dilution to the usual care group. This was an unblinded trial, so the patients randomized to intervention were undergoing intensive monitoring, resuscitation, and hemodynamic testing by the providers. Certainly this changed the culture in participating centers and impacted the care that the “usual care” group received. Nevertheless, a relatively small difference in composite outcomes is not exciting.  

Debating whether this trial is a positive or negative trial misses the mark in my view. The bottom line is that care for septic shock has stagnated. Hopefully, we are heading towards a future of identifying clinical phenotypes (along the lines of the recent ImmunoSep trial) to tailor treatment. Until then, however, we are left in the post-Manny-Rivers world, stuck with sepsis treatment based on guidelines that call for 30 cc/kg of IV fluids, start norepinephrine, add on some vasopressin, and follow lactates. The results and framework of ANDROMEDA-1 & 2 give us individualized targets for resuscitation, in the form of CRT, that we can implement at the bedside now.

Normal Physiology–Macro- and Microcirculation are Coupled

The power of CRT is hidden in the physiology. Under normal physiology, macrocirculation (i.e. blood pressure, heart rate, stroke volume, cardiac output) is aligned with microcirculation. To review the obvious, typical blood flow out of the heart → aorta → arteries → arterioles → capillary beds → venous return system. The flow of blood across the capillary bed is also termed the “vascular waterfall.” This flow of blood across the capillaries relies on several factors (sorry to all of you who don’t want me remaining you of Guyton’s contributions to physiology). One, the critical closing pressure (i.e. the minimum pressure required by the arterial side of the capillary bed to stay open) is exceeded by the pressure of flow from the arteriole into the capillary. Two, post-capillary bed pressure on the venous side is lower to allow for the vascular waterfall to flow. This is informed by the mean systemic filling pressure on the venous side, which is typically higher than the central venous pressure (here’s a helpful schematic). As such, under normal conditions: Tissue Perfusion Pressure = MAP – Venous Back Flow (which we typically approximate as CVP, because mean systemic filling pressure is difficult to measure).

Abnormal Physiology–Macro- and Microcirculatory Uncoupling

During septic shock, capillary bed physiology can change by several mechanisms: impaired microvascular reactivity, endothelial inflammation, adrenergic tone, and venous congestion (in the over fluid resuscitated). These changes are heterogenous across the innumerable capillary beds throughout the body, and when patients are in shock, a “normal” MAP of 65 may not be adequate to overcome a deranged critical closing pressure in various capillary beds. This is macro- and microcirculation uncoupling. In this scenario: Tissue Perfusion Pressure = MAP – Critical Closing Pressure (of capillary beds, which cannot be measured directly).

Capillary Refill Provides Insight into Macro- and Microcirculation Uncoupling

Evidence suggests that CRT reflects the health of microcirculation. Without belaboring the point, CRT is associated with the pulsatility index (a.k.a. blood flow) on doppler ultrasound of the kidneys, liver, spleen, and intestines. Variation in CRT with passive leg raise was associated with improved peripheral perfusion after fluid resuscitation. And heck, I am going to stop here, because you should just read Josh Farkas’ insights on personalized hemodynamic resuscitation, especially this handy table; if he can’t convince you, then I can’t.

Where Does this Leave Us?

CRT is a valid resuscitation marker because it gives insight into when macro- and microcirculation uncoupling is occurring. This is vital, and we have all seen it in our clinical experience; the “resuscitated” patient who is in shock, has received “appropriate fluids,” and has normal macro hemodynamic parameters, but is just not getting better. Likely, this group of patients has microcirculatory dysfunction that needs more individualized and tailored assessment and treatment.

ANDROMEDA-2 provides us with a roadmap. While there is more work to be done in personalizing sepsis resuscitation, this is an important trial for resuscitation medicine. Simply looking at it as a “negative trial” is clinically nihilistic. Here are my takeaways for your next shift:

  • Continue to check CRT in sick patients in a standardized manner; use changes in CRT as a marker of response or lack of response to current treatment.
  • Make sure you have the right diagnosis (Hint: you may not have source control, or the patient doesn’t have sepsis).
  • In patients with shock, narrow pulse pressure (< 40 mm Hg) suggests impaired stroke volume and should prompt the following:
    • Evaluate cardiac function with POCUS (very narrow pulse pressure is more consistent with cardiogenic or obstructive shock).
    • Evaluate fluid status and fluid responsiveness (history, exam, ultrasound, and fluid responsiveness maneuvers).
  • As an aside, performing fluid responsiveness maneuvers relies on either an arterial line or the ability to perform LV VTI on POCUS. If consistent and accurate LV VTI is outside your POCUS wheelhouse, then you need pulse pressure variation (PPV) from an arterial line. Despite the recent EVERDAC RCT, feel free to place an arterial line in these patients with persistent shock and circulatory dysfunction. This will allow for thoughtful evaluation of fluid responsiveness (ventilator maneuvers and passive leg raise).
    • End-expiratory occlusion test: decrease in PPV or increase in CO by 10% suggests fluid responsiveness.
    • Tidal volume increase (from 6 → 8 cc/kg): increase in PPV or decrease in CO by 3.5% suggests fluid responsiveness.
    • Passive leg raise: decrease in PPV or increase in CO by 10% suggests fluid responsiveness.
  • The diastolic blood pressure push to above 50 mm Hg is reasonable to ensure adequate coronary perfusion pressure (MAP—diastolic blood pressure).
  • Lastly, ANDROMEDA-2 teaches us the need for frequent bedside assessment of patients in persistent shock. If the patient isn’t getting better, then assess, implement a change, and come back in an hour or less to assess again.
    • This is why I think the MAP push to 80–85 in Tier 2 is probably okay. It is a 1-hour trial of higher blood pressure goals when concurrent severe cardiac dysfunction has been ruled out. While OPTPRESS showed harm with this approach, patients were randomized to 72 hours of higher MAP and left there for the entire intervention period.

Sources

Leave a Reply