Outcomes after attempted resuscitation for in-ICU cardiac arrest in septic shock: stratification by the 2026 Delphi consensus vasopressor threshold

Abstract

Introduction:

Patients with septic shock may experience in-ICU cardiac arrest while receiving vasopressors, yet clinicians lack objective survival estimates following attempted resuscitation. The 2026 SCCM/ESICM Delphi consensus proposed norepinephrine-equivalent (NEE) >0.50 μg/kg/min as one criterion within the definition of refractory septic shock.

Aim of Study:

To characterize outcomes after attempted resuscitation for in-ICU cardiac arrest in septic shock, stratified by the 2026 Delphi consensus vasopressor threshold.

Materials and Methods:

We conducted a retrospective cohort study using MIMIC-IV (2008–2022) involving adults with septic shock who received continuous vasopressors for ≥3 hours, regardless of dose, and subsequently experienced in-ICU cardiac arrest with attempted cardiopulmonary resuscitation (CPR). Patients were stratified by pre-arrest NEE dose (≤0.50 vs >0.50 μg/kg/min). The primary outcome was ICU survival; secondary outcomes were 2-hour, 24-hour, and hospital survival. Prespecified sensitivity analyses evaluated ≥1-hour and ≥6-hour vasopressor cohorts. Exploratory multivariable logistic regression adjusted for age, sex, and Charlson Comorbidity Index.

Results:

Among 94,458 ICU admissions, 564 had in-ICU cardiac arrest with attempted CPR, of whom 213 had sepsis or septic shock. Of these, 118 received continuous vasopressors for ≥3 hours and constituted the final analytic cohort (51 with pre-arrest NEE >0.50 μg/kg/min and 67 with NEE ≤0.50). Among patients with pre-arrest NEE >0.50 μg/kg/min, 2-hour survival was similar to those with NEE ≤0.50 (49.0% vs 55.2%; P=0.50), but ICU survival was lower (2/51 [3.9%; 95% CI, 0.5–13.5] vs 11/67 [16.4%; 95% CI, 8.5–27.5]; P=0.039), as were 24-hour (13.7% vs 41.8%; P=0.001) and hospital survival (1/51 [2.0%; 95% CI, 0.1–10.5] vs 9/67 [13.4%]; P=0.042). After adjustment, NEE >0.50 μg/kg/min was associated with lower ICU survival (OR 0.18; 95% CI 0.04–0.85; P=0.031). No patient receiving ≥6 hours of vasopressors with pre-arrest NEE >0.50 μg/kg/min survived to hospital discharge (0/39; 95% CI, 0–9.0%).

Conclusions:

Among patients with septic shock who underwent attempted resuscitation for in-ICU cardiac arrest, pre-arrest NEE >0.50 μg/kg/min was associated with substantially lower post-resuscitation ICU survival. Hospital survival was rare: only one patient receiving ≥3 hours of vasopressors survived, and none receiving ≥6 hours. These exploratory findings warrant validation in larger multicenter studies.