Incidence and assessment of delirium following open cardiac surgery: a systematic review and meta-analysis

This systematic review and meta-analysis sought (i) to provide an overview of the incidence of delirium following open cardiac surgery and (ii) to investigate how incidences of delirium are associated with different assessment tools.

A systematic search of studies investigating delirium following open cardiac surgery was conducted in Medline (Ovid), EMBASE, PsycINFO, CiNAHL, and the Cochrane Database. Only studies with patients diagnosed or screened with a validated tool were included. Studies published from 2005–2021 were included in the meta-analysis. Of 7126 individual studies retrieved, 106 met the inclusion criteria for the meta-analysis, hereof 31% of high quality. The weighted pooled incidence of delirium following open cardiac surgery across all studies was 23% (95% CI 20–26%), however we found a considerable heterogeneity (I² = 99%), which could not be explained by subgroups or further sensitivity analyses. The most commonly applied screening tool for delirium is CAM/CAM-ICU. The lowest estimates of delirium were found by applying the Delirium Observation Scale (incidence 14%, 95% CI 8–20%), and the highest estimates in studies using ‘other’ screening tools (Organic Brain Symptom Scale, Delirium Symptom Interview) with a pooled incidence of 43% (95% CI 19–66%), however, only two studies applied these.

Delirium following open cardiac surgery remains a complication with a high incidence of overall 23%, when applying a validated tool for screening or diagnosis. Nevertheless, this systematic review and meta-analyses highlight the significant inconsistency in current evidence regarding assessment tools and regimens.

Prospero CRD42020215519.