Author information

1Department of Medicine, University of Padua, Padua, Italy.

Abstract

Electroencephalography (EEG) is useful to objectively diagnose/grade hepatic encephalopathy (HE) across its spectrum of severity. However, it requires expensive equipment, and hepato-gastroenterologists are generally unfamiliar with its acquisition/interpretation. Recent technological advances have lead to the development of low-cost, user-friendly EEG systems, allowing EEG acquisition also in settings with limited neurophysiological experience. The aim of this study was to assess the relationship between EEG parameters obtained from a Standard-EEG system and from a commercial, low-cost wireless headset (Light-EEG) in patients with cirrhosis and varying degree of HE. Seventy-two patients (58 males; 61 ± 9 years) underwent clinical evaluation, the Psychometric Hepatic Encephalopathy Score (PHES), and EEG recording with both systems. Standard, automated EEG parameters were calculated on two derivations. Strong correlations were observed between automated parameters obtained from the two EEG systems. Bland and Altman analysis indicated that the two systems provided comparable automated parameters, and agreement between classifications (normal vs. abnormal EEG) based on Standard- and Light-EEG was good (0.6< k <0.8). Automated parameters such as the mean dominant frequency (MDF) obtained from the Light-EEG correlated significantly with the Model of End-Stage Liver Disease (MELD) score (r=-0.39, p<0.05), fasting, venous ammonia levels (r=-0.41, p<0.01) and PHES performance (r=-0.49, p<0.001). Finally, significant differences in Light-EEG parameters were observed in patients with varying degrees of HE.

CONCLUSION: Reliable EEG parameters for purposes of HE diagnosing/grading can be obtained from a cheap, commercial, wireless headset. This may lead to more widespread use of this patient-independent tool both in routine liver practice and in the research setting.