Background: The diagnosis of covert hepatic encephalopathy (CHE) by means of portosystemic encephalopathy syndrome (PSE) test is costly and therefore infrequently performed.
Aim: To determine the ability of critical flicker frequency (CFF) alone or in combination with laboratory findings, as an initial test to pre-select which patients should undergo further testing for the diagnosis of covert hepatic encephalopathy.
Methods: This single-centre study included all patients with cirrhosis who underwent PSE and CFF in 2011. CHE was defined by abnormal PSE test. Logistic regression analysis was performed to identify predictors of CHE. ROC curves were used to identify cut-offs of these independent predictors.
Results: One hundred and seventeen patients were included. Seventy (60%) had CHE with a higher MELD [16 (IQR 13-21); P = 0.001] and lower CFF [38 Hz (IQR 36-41) P = 0.0011]. On multivariate analyses, CFF [OR 0.83 (95% CI 0.74-0.94)] and MELD [OR 1.13 (95% CI 1.04-1.22)] were identified as independent predictors of CHE. Sensitivity and specificity of a CFF cut-off of 43 Hz was 93.5% and 42.9%, and for a MELD cut-off of 24, it was 97.5% and 32.8% respectively. Most patients with a MELD-Score <24 and a CFF >43 Hz did not have CHE (78%) and with a MELD-Score >24 and CFF <43 Hz most patients had CHE (85%). Therefore, 27% of patients could avoid further testing with a diagnostic accuracy of 81%.
Conclusion: The combination of MELD-score and critical flicker frequency may be used as a first diagnostic step to filter patients, in whom further covert hepatic encephalopathy testing could be avoided.
© 2016 John Wiley & Sons Ltd.