Reliability of EEG in brain death diagnosis: an analysis of clinical and technical factors.

内容摘要

In brain death determination (BDD), electroencephalography (EEG) is a commonly used diagnostic method in Germany for proving irreversibility of previously clinically proven loss of brain function. Despite the comprehensive availability of EEG in intensive care units, the reliability of EEG readings is often limited by biological or technical artefacts. This study aimed to determine the detection rate of electrocerebral inactivity (ECI) with a special focus on reliability and challenges of using EEG in BDD. A retrospective monocentric analysis of all BDD EEG data acquired from January 2015 to December 2025 at our tertiary care hospital was conducted. All identified BDD EEG reports were systematically analysed for their evaluability and for technical and biological artefacts. In addition, sociodemographic and patient-specific parameters that could influence the accuracy of EEG readings (e.g., craniotomies or intracranial drainages) were recorded. A total of 47 BDD EEG reports from 42 patients (women: 38.1%, mean age: 48.4 ± 14.4 years, median age: 52 years, age range: 21-78 years) were identified. In 81.0% of the patients, BDD could be confirmed using EEG. While biological artefacts did not significantly limit the evaluability of the EEG readings (p > 0.05), technical artefacts were significantly associated with a higher risk for limited EEG readability (p = 0.0309, relative risk: 4.2, 95% confidence interval: 1.3-13.8). This finding was primarily attributable to electrode artefacts (p = 0.0030), while other artefacts such as electrocardiogram or ventilation artefacts did not significantly impair readability (p > 0.05). Neither the aetiology of brain damage nor other patient-specific or sociodemographic parameters had any influence on the usability of EEG in BDD. We conclude that EEG is a reliable diagnostic tool in BDD in most cases if technically accurately recorded. Further studies are required to optimise existing technical errors in selected patients to increase the already high level of methodological reliability.