Incidence and prognostic impact of delirium after cardiac arrest: a systematic review and meta-analysis

Delirium is common after ROSC, yet reported incidence varies widely across studies that differ in instruments, assessment windows, and post-arrest care. We aimed to quantify the incidence of delirium among adult post-ROSC ICU survivors and evaluate its association with short-term mortality.

The review was reported according to PRISMA guidelines and registered on PROSPERO (CRD420251160097). PubMed, Embase, Scopus, and Web of Science were searched from 2000 to 30 Sep 2025 without language restrictions. Eligible studies enrolled adults (≥18 years) with ROSC admitted to ICU and reported ICU-period delirium using validated tools (CAM-ICU, ICDSC, DSM-based diagnosis) or structured screening scales. Proportions were pooled on the logit scale using random-effects models with Hartung-Knapp adjustment. Prespecified sensitivity analyses included validated tools only, exclusion of non-validated instruments, and binomial-normal GLMM.

Eleven studies (n = 1799) met inclusion criteria. The pooled ICU-period incidence of delirium was 0.43 (95 % CI 0.16-0.75) with a wide prediction interval (≈0.01-0.98), reflecting substantial heterogeneity largely driven by assessment instrument and ICU ascertainment window rather than patient factors. Findings were robust in binomial-normal GLMM analyses and in influence/sensitivity analyses. Delirium was consistently associated with greater resource use (longer ICU and hospital length of stay) and, in the two post-ROSC cohorts reporting adjusted models, signaled higher short-term mortality; larger CICU/AMI cohorts provided concordant mortality signals.

Delirium affects roughly two in five ICU survivors after ROSC and its measured incidence depends strongly on how and when it is assessed. Standardizing to validated tools and a common ICU window could narrow heterogeneity and improve comparability. These results support routine delirium screening and prevention as core elements of post-arrest care.