Value of Critical Flicker Frequency and Psychometric Hepatic Encephalopathy Score in Diagnosis of Low-Grade Hepatic Encephalopathy

Abstract

Critical flicker frequency (CFF) and psychometric hepatic encephalopathy score (PHES) analyses are widely used to diagnose hepatic encephalopathy (HE), but little is known about their value in the diagnosis of low-grade HE. The diagnostic values of CFF and PHES were compared using a computerized test battery and West Haven criteria as reference. We performed CFF analysis on 559 patients with cirrhosis and 261 without (controls). Of these 820 patients, 448 were evaluated using a modified PHES system and 148 were also evaluated using the conventional PHES system. CFF distinguished between patients with overt HE and without minimal or overt HE in the entire study population with 98% sensitivity and 94% specificity and in the subgroup of patients who were evaluated by conventional PHES with 97% sensitivity and 100% specificity. Conventional PHES identified patients with overt HE with 73% sensitivity and 89% specificity. CFF distinguished between patients with and without minimal HE with only 37% sensitivity but 94% specificity (entire study population). In the subgroup of patients evaluated by conventional PHES, CFF distinguished between patients with and without minimal HE with 22% sensitivity and 100% specificity; these values were similar to those for conventional PHES (30% sensitivity and 89% specificity). The modified PHES distinguished between patients with and without minimal HE with 49% sensitivity and 74% specificity. The diagnostic agreement values between CFF and conventional or modified PHES in patients with minimal HE were only 54% or 47%, respectively. In an analysis of patients with cirrhosis and controls, CFF distinguished between patients with overt HE and without minimal or overt HE. PHES testing produced a statistically significant difference among groups, but there was considerable overlap between controls and patients with overt HE. PHES, CFF, and a combination of PHES and CFF could not reliably distinguish patients with minimal HE from controls or those with overt HE.