ABSTRACT
Objective:To examine whether the combination of quantitative regional apparent diffusion coefficient (ADC) and amplitude-integrated electroencephalogram (aEEG) can predict the outcome of comatose patients with severe traumatic brain injury (sTBI).Methods:A prospective study was conducted. The patients with coma caused by sTBI [Glasgow coma scale (GCS) < 8] admitted to Suqian First Hospital from January 2016 to June 2019 were enrolled. All patients underwent aEEG examination and magnetic resonance imaging (MRI) scan within 1 week after emergency treatment. The ADC values of 9 regions of interest (frontal gray matter and white matter, parietal gray matter and white matter, temporal gray matter and white matter, caudate nucleus of basal ganglia, lenticular nucleus and thalamus) were measured by head MRI, and the mean ADC values of frontal lobe, parietal lobe, temporal lobe and basal ganglia were calculated respectively. According to the follow-up results after 12 months, the differences of each index between patients with poor prognosis [Glasgow outcome score (GOS) 1-2] and patients with good prognosis (GOS 3-5) were compared; the receiver operating characteristic curve (ROC curve) was drawn to evaluate the predictive ability of aEEG and ADC for the good prognosis of patients with sTBI, and the predictive value of the combination of aEEG and ADC.Results:A total of 52 patients with sTBI were enrolled, with mean age of (36.7±13.9) years old, 35 of whom were male. Within 12 months follow-up, 29 patients had achieved favorable outcomes and 23 patients had unfavorable outcome. There were 21, 17 and 14 patients with aEEG , and grade, respectively, and 19, 10 and 0 patients had good prognosis respectively. ADC values of 9 regions of interest in patients with good prognosis were significantly higher than those in patients with poor prognosis (×10 -6 mm 2/s: 924±107 vs. 531±87 in frontal gray matter, 804±95 vs. 481±74 in frontal white matter, 831±93 vs. 683±72 in temporal gray matter, 726±87 vs. 654±63 in temporal white matter, 767±79 vs. 690±75 in parietal gray matter, 716±84 vs. 642±62 in parietal white matter, 689±70 vs. 465±68 in caudate nucleus, 723±84 vs. 587±71 in lenticular nucleus, 807±79 vs. 497±67 in thalamus, all P < 0.01). ROC curve analysis showed that the area under ROC curve (AUC) of aEEG for predicting good prognosis of sTBI patients was 0.826, when the cut-off value of aEEG was < 1.5, the sensitivity was 94.7% and the specificity was 72.8%. Among the ADC value prediction abilities in the interested areas, the prediction of ADC value in frontal lobe and basal ganglia area were better than that in sTBI patients. AUC was 0.817 and 0.903 respectively. The best cut-off values were > 726×10 -6 mm 2/s and > 624×10 -6 mm 2/s respectively, the sensitivity of predicting prognosis were both 100%, and the specificity was 63.4% and 61.8%. A model combining frontal ADC and basal ganglia ADC with aEEG was 91.0% sensitive and 93.7% specific for favorable outcome of sTBI patients. Conclusion:Combination of the quantitative measurement of regional ADC and aEEG may be useful for predicting the outcome of the patients with sTBI.
ABSTRACT
Objective To assess cognitive impairment and its correlation with power of quantitative electroencephalogram (qEEG) in patients with temporal lobe epilepsy (TLE).Methods Fifty-five patients with TLE seen consecutively in Suqian First Hospital and 40 healthy controls (HC) were assessed by Mini-Mental State Examination (MMSE),Brief Cognitive Battery (BCB) and qEEG.The global interhemispheric and intrahemispheric difference values for power spectral ratios (Dv-PSR) were calculated.Cognitive functions and Dv-PSR of the TLE patients and the HC were compared,and correlation between cognitive impairment and power of qEEG was assessed using Spearman correlation analysis.The significance level was set at P≤0.05.Results Statistical analysis showed that MMSE scores did not have statistically significant difference between the TLE patients and the HC (26.9±2.4 vs 27.3±2.6,t=0.502,P=0.549).However,BCB examination showed that immediate memory,incidental memory,delayed recall,learning test,verbal fluency and recognition differed significantly between the TLE and the HC groups (7.34± 1.33 vs 8.92±1.37,6.05±1.12 vs 7.93±1.20,6.77±1.08 vs 8.19±1.14,11.87±4.47 vs 16.8±4.56,8.52±1.74 vs 9.75 ± 1.36,8.74 ± 1.19 vs 9.87 ± 1.18,respectively;t=2.916,Z=3.204,t=2.549,Z=3.937,t=1.341,t=2.791,P＜ 0.05).Interhemispheric Dv-PSR in frontal,central,temporal and parietal area was higher in the TLE group than in the HC group (0.478±0.043 vs 0.252±0.028,0.441±0.051 vs 0.306±0.039,0.394±0.027 vs 0.247± 0.018,0.511±0.036 vs 0.224±0.021,respectively;t=3.711,2.403,3.144,4.327,P＜0.05),and intrahemispheric Dv-PSR in frontal,central,parietal and occipital area (minus temporal area respectively) was also higher in the TLE group than in the HC group (0.521±0.024 vs 0.221±0.017,0.249±0.012 vs 0.167±0.008,0.187± 0.013 vs 0.104 ± 0.007,0.313 ± 0.021 vs 0.127 ± 0.009,respectively;t=4.208,3.192,2.611,3.737,P＜0.05).Spearman analysis showed positive correlations between intrahemispheric,interhemispheric Dv-PSR and several cognitive domains impairment assessed by BCB (P＜0.05).Conclusion There was mild cognitive impairment in TLE patients,which was significantly associated with Dv-PSR assessed by qEEG,suggesting that Dv-PSR measurement may be used as a marker for cognitive impairment in epilepsy.