| ObejectiveTo study the predictive value of quantitative electroencephalography (QEEG) on prognosis of acute conscious disorders with supratentorial lesion. Compare with the raw electroencephalography(EEG), to explore the accuracy and feasibility of the indicators such as amplitude integrated electroencephalography (aEEG),spectral edge frequence (SEF) and total power(TP) of the predictive value in patients with supratentorial lesion. To assist the physician assess of the coma patient's condition more accurately.Methods129 comatose patients who were admitted in nero-intensive care unit (NICU) of Nanfang Hospital from January 2008 to December 2010 were included. We monitor all the patients' cerebral function including neurological examination,raw EEG and QEEG when they were admit in the NICU.The Nicolet One Monitor was used to record the QEEG signal and the raw EEG.The record last no less than 30 minutes, meantime use the sonic stimulation to stimulate the patients(The raw EEG electrodes were placed according to the international EEG 10-20 classification,and the QEEG is recorded from one pair of parietally placed electrodes P3 and P4).The digital electroencephalogrphy signal is amplified and passed through an asymmetrical band-pass filter that strongly prefers higher frequencies over lower ones and suppresses activity below 2 Hz and above 15 Hz in order to minimize artifacts from such sources as sweating, movement, muscle activity and electrical interference. Additional processing includes semilogarithmic amplitude presentation, rectification, smoothing and considerable time compression. And the signal is displayed on a semilogarithmic scale at slow speed (6 cm/h) at the cot side. Thus a full minute of EEG is represented by only a single millimeter of aEEG display. The aEEG signals include the upper,lower margins and bandwidth. Through the use of computers, the continuous frequency information obtained through FFT can be analyzed by combining data into specific bands according to frequency range. And finally the total power(TP) and spectral edge frequence(SEF) were obtained.Cases admitted criteria:①comatose patients.②Supratentorial lesion, include the focal supratentorial lesion and the diffuse lesion.③First evaluation performed in 3 days after onset④All the patients were treated with conservative medical treatments. Cases excluded criteria:①conscious patients.②to be used with antiepileptic drugs(AEDs) or sedatives.③ther factors such as shock, hypothermia, ect that may affecting the cerebral function.④The subtentorial lesions.⑤The insults include supratentorial and subtentorial lesions.⑥Obvious interfering artifact.⑦Patient died of none central nervous system causes.The raw EEG is classificated according to the background activity and the reactivity of the EEG signals.The Synek and Young grading scales were used to classificated the raw EEG. The Synek classification scale is displayed as follow:Grade 1:Abnormality is characterized by the presence of predominated regular alpha activity,with some scattered activity in the theta frequence range.Grade 2: Abnormality is associated with dominant activity in the theta frequence range,with some alpha and delta waves. Grade 3:Abnormality is characterized by dominant wildspread delta activity regular or irregular,with little activity in other frequence range; or spindle coma. Grade 4:Abnormality is characterized by frequent isoelectric intervals of variable duration, always last more than Is, with bursts of other activities(burst suppression pattern); or alpha coma;or theta coma;or very small amplitude (less than 20μv)activity,mainly with in the delta frequence range. Grade 5: Abnormality is typified by isoelectrc EEG(the EEG amplitude is less than 2μv). The Synek classification scale is as follow:Grade 1:Abnormality is characterized by the presence of predominated delta/theta activity, the propotion is more than 50%(not theta coma). Grade 2:Abnormality is characterized by triphasic waves. Grade 3: Abnormality is characterized by burst suppression pattern,with or without epliptiform discharges consisting of poly spikes or clusters of sharp waves. Grade 4:Abnormality is characterized by alpha coma;or theta coma;or spindle coma. Grade 5:Abnormality is typified by epliptiform discharges consisting of polyspikes or clusters of sharp waves (not the burst suppression pattern). Grade6:Abnormality is typified by the electrocerebral inactivity (the EEG amplitude is less than 20μv).Glasgow coma scale will be preformed in all patients before the raw EEG and the QEEG monitoring.The early prognosis included two groups consist of survival and death (containing brain death) when they discharged from the NICU.3 months prognosis after onset include good outcome and poor outcome.The modified rankin scale (mRS) is used to assess the 3 months prognosis. The modified rankin scale (mRS) is as followed:0) No symptoms.1) No significant disability. Able to carry out all usual activities,despite some symptoms.2) Slight disability. Able to look after own affairs without assistance,but unable to carry out all previous activities.3) Moderate disability. Requires some help, but able to walk unassisted.4) Moderately severe disability. Unable to attend to own bodily needs without assistance,and unable to walk unassisted.5) Severe disability.Requires constant nursing care and attention, bedridden, incontinent.6) Dead(inculde brain death).According to the mRS,the 3 months progonosis were divided into two group,the one was the good fuctional outcome(mRS 0-3), the oter was unfavorable outcome(mRS 4-6).Follow up method:all the patients were followed by phone.129 cases amdited in this study.39 patients were dead,90 patients were followed by phone.And 9 cases lost to follow up in 3 month later.SPSS 13.0 was used as statistic software.The continuous data were compared using the t test. The discrete data were analyzed by theχ2 test or the Fisher exact text. Theχ2 test or the Fisher exact text is used to analyze the relationship of Synek classification and Young classification between the group of death and survival,and the group of good outcome and bad outcome.The t test was used to analyse the relationship of aEEG, bandwidth,SEF95%,TP,GCS score between the group of death and survival,and the group of good outcome and bad outcome. Multivarite logistic regression was used to evaluated the progonostic value of aEEG, SEF95%, TP and GCS score.And finally the area under the receiver operating characteristic curve (AUC) was used to assess the selected variables to distinguish between death or alive and unfavorable vs favorable outcome.P<0.05 was considered statistically significant.ResultsClinical information of all patients:129 patients were admitted into this study. 39 patients were death, and 90 patients were followed by phone.and 9 cases lost to follow up in 3 month later.The t test and theχ2 test indicated no significant difference between the group of death and survival,and the group of good outcome and bad outcome in baseline demographic data.1)The early progonosis:There was significant difference of the Synek grading (χ2=27.392, P=.000) and the Young classification (χ2=28.736, P=.000) between the death group and survival group. The aEEG upper margin,aEEG lower margin, bandwidth,SEF95%,TP,GCS score between the group of death and survival were analyzed by t test. And there was significant difference between the two group of aEEG upper margin (t=4.099, P=0.000), aEEG lower margin (t=4.567, P=0.000), bandwidth(t=3.270, P=0.001),SEF95%(t=4.189, P=0.000),TP(t=3.296, P=0.001) and GCS score (t=5.618, P=0.000).Then,the correlation between early prognosis (the dependent variable) and aEEG upper margin,aEEG lower margin bandwidth,SEF95%,TP,GCS score, age,sexuality and past medical histories (independent variables)were analyzed by multivarite Logistic regression method. There was statistical significance in Logistic regression equation.And the aEEG upper margin (χ2=5.955,P=0.015),aEEG lower margin (χ2=5.157,P=0.023,SEF95%(χ2 =7.267, P=0.007), GCS score (χ2=4.367, P=0.037) were significantly incorporated into the model. And the lower of the aEEG upper margin,aEEG lower margin, SEF95%, GCS score, the higher risk of death.Finally,According to the ROC curve analysis, the sensitivity of aEEG upper margin was 59%, the sensitivity of aEEG lower margin was 51.3%,the sensitivity of SEF95% was 74.4%,and GCS score was 71.8% 2) 3 months progonosis:The Synek grading (χ2=23.250, P=000) between the death and survival were analyzed by Fisher exact text test,and there was significant difference between the two group. And the Young classification(χ2=22.961, P=.000) was also significant difference between the death and survival.The aEEG upper margin, aEEG lower margin, bandwidth,SEF95%,TP,GCS score between the group of death and survival were analyzed by t test. And there was significant difference between the two group of aEEG upper margin (t=3.807, P=0.000), aEEG lower margin (t=4.201, P=0.000), bandwidth (t=3.079, P=0.003),SEF95%(t=3.223, P=0.002),TP (t=2.424, P=0.017) and GCS score (t=7.085, P=0.000).Finally,the correlation between early prognosis (the dependent variable) and aEEG upper margin,aEEG lower margin, bandwidth,SEF95%,TP,GCS score, age,sexuality and past medical histories (independent variables)were analyzed by multivarite Logistic regression method. There was statistical significance in Logistic regression equation. And the SEF95%(χ2=4.252, P=0.039), GCS score (χ2=23.240, P=0.000) were significantly incorporated into the model.And The lower the GCSs and the SEF95%, the higher the risk of bad neurological outcome. According to the ROC curve analysis, the sensitivity of SEF95% was 75%,and GCS score was 62.5%Conclusions:1. With the univariate analysis of the raw EEG of the consicious disorders with supratentorial lesions,the Synek grading and the Young classification of the death group and survival group,and of the good outcome and poor outcome was significant difference. The higher the grade,the great danger of bad neurological outcome.2.With the univariate analysis of the QEEG of the comatose patients with supratentorial lesions of the early prognosis and 3 month progonosis,the aEEG upper margin,aEEG lower margin, bandwidth,SEF95%,TP,GCS score of the two group was significant difference. And the lower the aEEG upper margin,aEEG lower margin, bandwidth,SEF95%,TP,GCS score, it is more danger of death or bad neurological outcome.3.With the multivariate Logistic regression analysis of the raw EEG and QEEG in the early prognosis,the aEEG,SEF95%,GCS score were well related to the outcome of the unconscious patients with the supratentorial lesions.And the lower the aEEG upper margin,aEEG lower margin, SEF95% and GCS score, the higher risk of death.4.With the multivariate Logistic regression analysis of the raw EEG and QEEG in the 3 months prognosis,the SEF95%, GCS score were well related to the outcome of the unconscious patients with the supratentorial lesions.And the lower the GCSs and the SEF95%, the higher the risk of poor neurological outcome.5.The QEEG is feasible to monitor and evaluate adult comatose patients' cerebral function who with the supratentorial lesion. Compare to the raw EEG, the QEEG is more closely related to and more convenient to predict the progonosis of the comatose paients. The combination of EEG monitoring (include raw EEG and QEEG) and neurologic exam have a higher predictive capacity to the progonosis of the patients with supratentorial lesion,thus make sure the physicians provide early and appropriate neuroprotective measures to the patients. |
PDF Full Text Request