P300 latency for the diagnosis of minimal hepatic encephalopathy: evidence that spectral EEG analysis and psychometric tests are enough

Abstract

Background: No gold standard exists to detect minimal hepatic encephalopathy. Event-related evoked potentials (P300 latency) were proposed as the best tool to assess this condition. EEG spectral analysis and psychometric evaluation are also used to assess minimal hepatic encephalopathy.

Aims: The present study aims at comparing these three techniques.

Patients: Eighty-six cirrhotic patients without overt hepatic encephalopathy were studied.

Methods: Patients underwent EEG spectral analysis, psychometric evaluation and P300. P300 latency was age-adjusted; psychometric tests were age- and education-adjusted. Values >2Z were considered to be altered. The alteration of at least two psychometric tests was considered for cognitive impairment.

Results: At least one of the three indexes was altered in 61% (CI95% = 49-71) patients; EEG spectral analysis was altered in 41% (CI95% = 30-52%) patients, psychometric performance in 34% (CI95% = 24-45%) and P300 latency in 13% (CI95% = 7-22%). P300 latency was altered only in the patients having EEG spectral analysis or psychometric alterations, but for two cases. Psychometric performance and EEG spectral analysis, but not P300 latency, were correlated with indexes of liver function.

Conclusions: P300 latency provided little additional information and was less related to liver function than EEG spectral analysis and psychometrical investigation.