BACKGROUND: Electroencephalography (EEG) is widely used in the monitoring of critically ill comatose patients, but its interpretation is not straightforward. The aim of this study was to evaluate whether there is a correlation between EEG background pattern/reactivity to stimuli and automated pupillometry in critically ill patients. METHODS: Prospective assessment of pupillary changes to light stimulation was obtained using an automated pupillometry (NeuroLight Algiscan, ID-MED, Marseille, France) in 60 adult patients monitored with continuous EEG. The degree of encephalopathy and EEG reactivity were scored by 3 independent neurophysiologists blinded to the patient's history. The median values of baseline pupil size, pupillary constriction, constriction velocity, and latency were collected for both eyes. To assess sensitivity and specificity, we calculated areas under the receiver-operating characteristic curve. RESULTS: The degree of encephalopathy assessed by EEG was categorized as mild (42%), moderate (37%), severe (10%) or suppression-burst/suppression (12%); a total of 47/60 EEG recordings were classified as "reactive." There was a significant difference in pupillary size, constriction rate, and constriction velocity, but not latency, among the different EEG categories of encephalopathy. Similarly, reactive EEG tracings were associated with greater pupil size, pupillary constriction rate, and constriction velocity compared with nonreactive recordings; there were no significant differences in latency. Pupillary constriction rate values had an area under the curve of 0.83 to predict the presence of severe encephalopathy or suppression-burst/suppression, with a pupillary constriction rate of < 20% having a sensitivity of 85% and a specificity of 79%. CONCLUSIONS: Automated pupillometry can contribute to the assessment of cerebral dysfunction in critically ill patients.
BACKGROUND: Electroencephalography (EEG) is widely used in the monitoring of critically ill comatose patients, but its interpretation is not straightforward. The aim of this study was to evaluate whether there is a correlation between EEG background pattern/reactivity to stimuli and automated pupillometry in critically ill patients. METHODS: Prospective assessment of pupillary changes to light stimulation was obtained using an automated pupillometry (NeuroLight Algiscan, ID-MED, Marseille, France) in 60 adult patients monitored with continuous EEG. The degree of encephalopathy and EEG reactivity were scored by 3 independent neurophysiologists blinded to the patient's history. The median values of baseline pupil size, pupillary constriction, constriction velocity, and latency were collected for both eyes. To assess sensitivity and specificity, we calculated areas under the receiver-operating characteristic curve. RESULTS: The degree of encephalopathy assessed by EEG was categorized as mild (42%), moderate (37%), severe (10%) or suppression-burst/suppression (12%); a total of 47/60 EEG recordings were classified as "reactive." There was a significant difference in pupillary size, constriction rate, and constriction velocity, but not latency, among the different EEG categories of encephalopathy. Similarly, reactive EEG tracings were associated with greater pupil size, pupillary constriction rate, and constriction velocity compared with nonreactive recordings; there were no significant differences in latency. Pupillary constriction rate values had an area under the curve of 0.83 to predict the presence of severe encephalopathy or suppression-burst/suppression, with a pupillary constriction rate of < 20% having a sensitivity of 85% and a specificity of 79%. CONCLUSIONS: Automated pupillometry can contribute to the assessment of cerebral dysfunction in critically ill patients.
Authors: Lorenzo Peluso; Lorenzo Ferlini; Marta Talamonti; Narcisse Ndieugnou Djangang; Elisa Gouvea Bogossian; Marco Menozzi; Filippo Annoni; Elisabetta Macchini; Benjamin Legros; Paolo Severgnini; Jacques Creteur; Mauro Oddo; Jean-Louis Vincent; Nicolas Gaspard; Fabio Silvio Taccone Journal: Front Neurol Date: 2022-03-21 Impact factor: 4.003