Incidence of delirium post cardiac surgery: Discrepancy between clinical observation, DOS scores, and single‑lead EEG

Abstract

Rational

Postoperative delirium (POD) is a common complication after cardiac surgery, associated with increased morbidity, mortality, prolonged hospitalization, and cognitive decline. Early and accurate diagnosis is crucial, but current methods like the Delirium Observation Screening (DOS) scale rely on subjective assessments. Single‑lead EEG (sl-EEG), particularly the DeltaScan Brainstate Monitor, offers a more objective approach. This study compares the incidence of delirium detected by clinical observation, DOS scores, and single‑lead EEG scores in patients undergoing aortic valve replacement (AVR) surgery.

Methods

This prospective cohort study included 50 patients aged 65 or older scheduled for AVR surgery. Delirium was assessed preoperatively and on postoperative days 1, 3, and 7 using clinical observation, DOS, and single‑lead EEG. Incidence rates were calculated, and the McNemar's Chi-squared test was used to assess differences between methods.

Results

Delirium incidence varied widely by method: 32 % by clinical assessment, 28 % by DOS, and 76 % by single‑lead EEG. Clinical assessment and DOS had an 80 % concordance, while single‑lead EEG detected significantly more cases (p < 0.001). Incidence declined across all methods over seven postoperative days.

Conclusion

Our findings reveal significant discrepancies in POD detection rates by diagnostic methods. The high sensitivity of sl-EEG suggests a risk of false positives, while clinical assessment and DOS may risk underdiagnosis, especially in hypoactive delirium. An integrated diagnostic approach combining multiple methods may improve accuracy and capture the full spectrum of delirium symptoms. Future studies should refine these tools and explore advanced technologies to develop reliable, easily deployable diagnostics for clinical practice.