Background

Cardiac arrest in the intensive care unit (ICU-CA) is distinct from other in-hospital cardiac arrests, involving critically ill patients in monitored settings. Its prevalence and outcomes remain unclear. This systematic review and meta-analysis aimed to fill this evidence gap and identify areas for future research to improve outcomes in this patient population.

Methods

We searched MEDLINE, Embase and Scopus databases from inception until 26 November 2025 for studies reporting on ICU-CA in adults. We performed random effects meta-analyses with the generalised linear mixed model. We used the Joanna Briggs Institute Checklist to assess risk of bias and the GRADE approach to assess the certainty of evidence. The primary outcome was the prevalence of patients with ICU-CA; secondary outcomes included ICU and in-hospital mortality. We performed subgroup analyses based on geographical region (continent), study source (registry vs. non-registry), and COVID-19 status (infected vs. non-infected).

Results

We included 35 observational studies including 36 cohorts in the meta-analysis. The pooled proportion of ICU-CA was 3.23% (95% CI: 2.08–4.98, moderate certainty). The proportion of ICU-CA was significantly higher (p < 0.001) in patients with COVID-19 (10.6%, 95%-CI: 5.4–19.7) compared with non-COVID-19 cohorts (2.3%, 95%-CI: 1.5–3.5). Pooled ICU mortality was 74.0% (95%-CI: 63.7–82.1, moderate certainty) and in-hospital mortality was 82.0% (95%-CI: 75.9–86.9, high certainty). Meta-regression found that proportion of shockable rhythm was inversely associated with ICU mortality, but not in-hospital mortality.

Conclusion

Cardiac arrest in the ICU was associated with poor survival, with a higher prevalence in patients with COVID-19. Future studies should focus on peri-arrest characteristics to guide prognostication and management of individuals who experience ICU-CA.