[Predictive value of mismatch negativity and P3a combined with electroencephalogram reactivity for the prognosis of comatose patients after severe brain injury].

ABSTRACT

Objective: To explore the clinical value of mismatch negativity and P3a combined with electroencephalogram (EEG) reactivity to predict the prognosis of patients after severe brain injury. Methods: The clinical data of patients with severe brain injury who were admitted to the neurosurgical intensive care unit of Xiangya Hospital of Central South University from October 2019 to July 2020 were retrospectively analyzed. All patients underwent evaluation of auditory mismatch negativity (MMN), P3a, and EEG reactivity (EEG-R) within 28 days after the onset of coma. Patients were divided into two groups using the 3-month Glasgow Outcome Scale (GOS) after coma onset, a GOS score of 3-5 was defined as a favorable outcome, and GOS grades 1-2 were defined as an unfavorable outcome. The correlation between clinical indicators and prognosis was analyzed, and the predictive values of statistically significant indicators and the cut-off values were determined using the receiver operating characteristic (ROC) curve. Results: A total of 48 patients were enrolled in the study, including 35 males and 13 females (age range18-68 years old). Twenty-nine of the patients had a favorable outcome and 19 had an unfavorable outcome. The Glasgow Coma Scale (GCS), EEG-R, absolute amplitude of MMN at Fz (FzMMNA), and amplitude of P3a at Cz (CzP3aA) were significantly correlated with the prognosis of comatose patients (P<0.001). Multivariate logistic regression analysis revealed that only EEG-R, FzMMNA, and CzP3aA were independent predictors for the prognosis of comatose patients after severe brain injury (all P<0.05), with the area under the curve (AUC) of 0.757 (0.613-0.900), 0.912 (0.830-0.994) and 0.887 (0.793-0.981), respectively. The combination of FzMMNA and CzP3aA and the combinationof EEG-R, FzMMNA and CzP3aA increased the value of AUC to 0.942 (0.879-1.000) and 0.964 (0.920-1.000), respectively. Moreover, a cut-off value of 1.27 µV and 2.64 µV for FzMMNA and CzP3aA, respectively, yielded the best sensitivity and specificity for the prognosis prediction of patients with severe brain injury [FzMMNA 89.66%(26/29) and 84.21%(16/19); CzP3aA82.76%(24/29) and 84.21%(16/19)]. Conclusion: This study indicates that the combination of EEG-R, FzMMNA, and CzP3aA may serve as a favorable prognostic indicator for comatose patients after severe brain injury.