Minimal hepatic encephalopathy (MHE) is an early, clinically important stage of hepatic encephalopathy characterized by impairments in attention, executive function, visuospatial processing and motor speed. These deficits negatively affect daily functioning, increase risk of falls and hospitalizations, and raise the likelihood of progression to overt HE. The Psychometric Hepatic Encephalopathy Score (PHES) is widely used as a reference standard for diagnosing MHE but has limited sensitivity for detecting subtle cognitive and motor impairments. Prior studies reported that a proportion of patients classified as "without MHE" by PHES nonetheless display alterations in attention, coordination and motor function on more sensitive tests such as the d2, oral Symbol Digit Modalities Test (SDMT) and Stroop.
Because conventional psychometric testing is time consuming, requires trained personnel and adjustment for age and education, there is interest in objective, rapid and reproducible approaches. Abnormalities in saccadic and fixation eye movements reflect changes in cognitive processing in multiple neurological conditions. Thus, analysis of eye movement parameters using video-oculography may offer a non-invasive tool to identify early neurological alterations in cirrhotic patients undetected by PHES.
The study assessed 118 cirrhotic patients and 35 healthy controls. Among the patients, 32 were classified as having MHE by PHES and 86 as without MHE by PHES. All participants underwent a battery of psychometric assessments and eye movement testing recorded with video-oculography.
Psychometric testing included several instruments from which the most affected and discriminatory variables were identified. Eye movement protocols incorporated fixation tasks and antisaccade tasks in horizontal and vertical planes; parameters from these tasks were extracted for analysis.
A novel score to detect early cognitive impairment (early-MHE) was developed using historical cohorts comprising 519 cirrhotic patients and 74 age-matched healthy controls. The optimal set of psychometric parameters selected for the score came from the d2, Oral SDMT, and Stroop tests. The new score was applied to the 86 patients classified as without MHE by PHES to identify those with subtle cognitive deficits.
Using this approach, 46 of the 86 PHES-negative patients (53.5%) met criteria for impaired cognitive performance and were classified as early-MHE.
Eye movement recording used video-oculography to capture saccades and fixation behavior. Measured variables included performance on horizontal and vertical antisaccade tasks and fixation stability metrics. The study analyzed which specific eye movement parameters correlated with psychometric impairment and which had the greatest predictive value for detecting mild cognitive alterations (combined early-MHE and MHE groups).
Cirrhotic patients classified as early-MHE by the new psychometric score showed poorer cognitive performance compared with PHES-negative patients who did not meet the early-MHE threshold. Early-MHE patients also manifested measurable alterations in eye movement tests, particularly in antisaccade and fixation tasks. The magnitude of those abnormalities was greater in patients classified as MHE by PHES, supporting the notion of a continuum of cognitive dysfunction from subclinical changes to diagnosable MHE.
Eye movement variables derived from horizontal and vertical antisaccade tasks and from fixation testing had the highest predictive value for detecting mild cognitive impairment when early-MHE and MHE groups were combined for analysis.
When eye movement variables were grouped by task and entered into multivariate models, a model including variables from the horizontal antisaccade test provided the best predictive performance. That model achieved an area under the receiver operating characteristic curve (AUROC) of 0.803 (95% CI: 0.711–0.896; p < 0.0001). At a reported relative-risk cutoff of 65.66%, the horizontal antisaccade model reached 66.7% sensitivity and 90.6% specificity for distinguishing patients with mild cognitive alterations (early-MHE + MHE) from those without.
The findings indicate that a substantial fraction of cirrhotic patients labeled as cognitively normal by PHES nevertheless have early cognitive impairment detectable by a focused psychometric score and by objective eye movement measures. Antisaccade and fixation abnormalities were prominent in early-MHE and MHE patients, consistent with altered attention and executive control that can be reflected in saccadic control.
Eye movement analysis offers several practical advantages: it is rapid, objective, reproducible and non-invasive. The study's multivariate results suggest that horizontal antisaccade metrics in particular may serve as a useful screening tool to identify cirrhotic patients who warrant more comprehensive neuropsychological assessment or earlier clinical intervention.
Limitations and additional methodological details are provided in the full article and supporting information. The authors also report that the datasets, including eye movement variables, are available on Zenodo as indicated in the publication.
Video-oculography and analysis of eye movement parameters can detect subtle neurological alterations in cirrhotic patients who are classified as without MHE by the PHES. A psychometric score derived from d2, Oral SDMT and Stroop variables identified early-MHE in over half of PHES-negative patients in this cohort. Horizontal antisaccade variables showed the best multivariate predictive performance (AUROC 0.803) with high specificity, supporting the potential role of eye movement testing as a rapid, objective screening method to uncover early cognitive impairment in cirrhosis and facilitate timely intervention.