Capillary refill plus lactate sharpened the definition of refractory septic shock

Design

Exploratory secondary analysis of the ANDROMEDA-SHOCK-2 randomised trial

Population

1,363 patients with septic shock and complete 6-hour data

Primary outcome

28-day mortality by refractory septic shock definition

Effect

73.6% vs 23.7% (adjusted HR 4.68, 95% CI 3.31 to 6.64); positive LR 8.12 vs 2.71 for dose alone

This exploratory secondary analysis of the ANDROMEDA-SHOCK-2 trial asked whether adding persistent tissue hypoperfusion to vasopressor dose improves risk stratification in refractory septic shock. After 6 hours of protocolised resuscitation, refractoriness was defined as a norepinephrine-equivalent dose above 0.5 micrograms/kg/min plus abnormal capillary refill (over 3 seconds) and a non-falling lactate.

Among 1,363 patients with complete data, 188 (13.8%) had a high vasopressor dose and 47.9% of them died by 28 days. Only 53 (3.9%) met both hypoperfusion criteria in addition to the dose. Their mortality was 73.6% (39 of 53) against 23.7% for non-refractory patients (adjusted HR 4.68, 95% CI 3.31 to 6.64). With at least one hypoperfusion criterion, mortality was 55.0% vs 22.7%. The two-criterion definition had a higher positive likelihood ratio than dose alone (8.12 vs 2.71).

The authors call these findings hypothesis-generating. It is an exploratory analysis of a trial population managed to a protocol, and it describes prognosis, not what treatment to give.

• Reassess perfusion at 6 hours: capillary refill and lactate trend add to vasopressor dose.
• Persistent hypoperfusion on high-dose vasopressors marked a group with very high mortality (73.6%).
• Use the information for prognosis and family discussions, and to decide on escalation and transfer.
• This analysis does not show that any treatment improves outcomes in this group.

Why it matters

It suggests vasopressor dose alone misses patients whose tissues remain underperfused.

Don't overread it

Exploratory secondary analysis of one trial; prognosis was described but no treatment effect was tested.

The statistics, in plain English

A positive likelihood ratio of 8.12 means meeting the definition was about eight times more common among patients who died than among those who survived. With only 53 patients meeting the definition, the estimates are imprecise.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.