The Comparison Of Prognostic Value Of40-Hz Auditory Steady-state Response And Multimodality Evoked Potentials In Patients With Large Hemispheric Infarction

BackgroundLarge hemispheric infarction is a devastating disease that is associated with development of subsequent space-occupying edema and cerebral herniation. It is associated with high mortality rate despite proper managed with conservative medical treatment even under maximum intensive care treatment. Although hemicraniectomy after large hemispheric infarction has been proved to be effective and reduce the mortality from about78%to20%. Early and accurate evaluation and prediction of prognosis enable proper medical management.Clinical neurological exams and evaluations, such as Glasgow coma scale (GCS) and National Institutes of Health Stroke Scale score (NIHSS), are still the most widely used prognostic measures applied in clinical practice. However, in clinical practice in the neuro-intensive care unit (NICU), the objectivity could be influenced by physician’s clinical experience and the application of sedative medications. It has incontestable limitations. First, there is a risk of false pessimistic predictions-even a GCS of three, which means extremely severe brain damage. It is compatible with a good outcome. Second, the accuracy of clinical parameters is limited in intubated and aphasic patients. Various imaging modalities, such as computed tomography (CT), magnetic resonance imaging (MRI), diffusion weighted magnetic resonance imaging (DWI), and (PET) combined with neuromonitoring by means of microdialysis were applied to monitor patients at risk for large hemispheric infarction, but they are relatively expensive and can’t be used as a bedside complement. Evoked potentials (EPs) test as non-invasive bedside functional exam, including brainstem auditory evoked potential (BAEP), short latency somatosensory evoked potential (SSEPs). The40Hz auditory steady-state response (40-Hz ASSR), which is primarily generated from the superposition of the major wave of the auditory brainstem response and the middle latency response, has been investigated in a number of studies. Auditory evoked potentials at40Hz have also been applied to predict prognosis in post-trauma coma patients. It is demonstrated that recording of the40-Hz auditory evoked potential is a useful method for evaluation and determination of brain death. However, little is known about the relationship between40Hz ASSR and changes in consciousness levels induced by cerebrovascular diseases. The present study was firstly designed to determine the prognostic prediction value of40-Hz ASSR in patients with large hemispheric infarction and its clinical value in the NICU.Subjects and methodsA prospective study was carried out at the NICU of Nanfang hospital, Southern Medical University of China.169patients in large hemispheric infarction admitted into NICU from2008March to2011August, Inclusion criteria included:40-Hz auditory steady-state response (40-Hz ASSR), brainstem auditory evoked potentials (BAEP) and short-latence somatosensory evoked potentials(SSEPs) could be finished within72h, all the patients would receive regular medical treatment, without the influence of sedative medications, anaesthetic agents or muscle relaxants (EPs were recorded after the last medications application at least24h). Exclusion criteria was as follows:patients died from factors of other systems but neurology system, hearing impairment, definitely diseases of peripheral nerves, lower limbs and medulla spinals and seriously failure of liver, renal, respiratory and other systems when admission.40-Hz ASSR in patients were compared with values collected in36age and gender-matched normal subjects.The following information was collected for all the patients within24h after admission:basic demographic data (age, gender), major risk factors for infarction (history of hypertension, diabetes, atrial fibrillation and past stroke), infarction side and findings of40-Hz ASSR, BAEP and SSEPs. Internal laboratory standards of BAEP and lower limb SLSEP were established in a sufficient number (>100) of healthy volunteers Latencies of evoked potentials were considered abnormal when they exceeded the2.5-fold standard deviation of established normal values. Amplitudes of evoked potentials were considered abnormal when the side-to-side difference exceeded50%compared to the unaffected contralateral response or when the amplitude was below the2.5-fold value of the established standard value.According to Cant’s grading standards, BAEP was classified into three grades as follows:grade(1) normal bilateral waves appeared and wave V was always present; grade (2) bilateral waves were abnormal but wave V was always present; grade (3) only unilateral wave V appeared or even bilateral wave V disappeared. Lower limb SLSEP were classified as follows:grade (1):bilateral P40appeared normally; grade (2) bilateral P40appeared but abnormal or unilateral P40disappeared; grade (3) bilateral P40disappeared.Patients were followed up in our outpatient neurology clinic30days and90days after stroke as part of routine follow-up. The severity of disability was assessed as standard practice. In our study, subjects were dichotomized into two groups:the survival group and death group based on the30days prognosis. The outcome for survivors was assessed at90days using Modified Rankin Scores from chart review of the outpatient or rehabilitation clinic note and categorized to favorable (modified Rankin Score0,1,2,3and4) or unfavorable (modified Rankin Score5and6) outcome. No symptoms at all, No significant disability despite symptoms, Slight disability, Moderate disability, or moderately severe disability was considered a favorable outcome, severe disability, or death an unfavorable outcome.Recording technologyWithin72h after admission40-Hz ASSR, BAEP and SSEPs were recorded using Viking Quest system evoked potential equipment (Nicolet Company, American). Platinum needle electrodes were applied to the head according to the10/20system:①40HzaSSEP:Monaural95dB pip39.1/s was applied through standard inserted phones, while the other ear with70dB white noise. Two channel referencing electrode was in ipsilateral earlobes, recording electrode was in the central position (Cz), and ground electrode was in the midforehead point (FPz). Filters were set at150-3000Hz. Analysis time was10ms, overlapping500-1000times.②BAEP:Monaural90dB clicks11.1/s, overlapping1024-2048times, ibid.③SSEPs: the central scalp recording electrodes were placed over Cz referenced to the FPz, and the peripheral recording electrodes were positioned over the fossa poplitea point (PF) referenced to anterior patella point (K). Stimulate the posterior tibias nerve until the toes move to approximately1cm. overlapping250times. Electrode impedance was maintained below5kΩ. All the tests above were operated at least twice in order to make sure the accuracy and repeatability.Statistical methods were applied for the evaluation of the obtained results: Means and standard deviations were calculated for demographic factors including age, NIHSS. Differences between groups were assessed using the Wilcoxon ram test for nonparametric data. Regard death (or unfavorable outcome) as positive occurrence, and calculate the evaluation index including sensitivity, specificity, accuracy, a=0.05was the size of test, and bilateral test was used. SPSS version13.0was applied for the statistical analysis.ResultsLaboratory standards of40-Hz ASSR were established in36healthy volunteers (19males and17females; age ranging:25to74years). None of these subjects reported history of hearing difficulties and all had BAEP of normal wave and absolute latencies. Robust40-Hz ASSR was elicited in all health subjects. The first positive peak (P1) of the response was analyzed. Then the following parameters were measured for the amplitude, the side-to-side amplitude ratio and latency were measured. The normal values for40-Hz ASSR were available.According to the normal values of40-Hz ASSR, latencies of wave P1were considered abnormal when they exceeded the2.5-fold standard deviation of established normal values. Amplitudes of wave P1were considered abnormal when the side-to-side difference exceeded41%compared to the unaffected contralateral response. A poor repeatability of the sine waves was considered abnormal.To simplify the global analysis and comparison of results, we graded40-Hz ASSR in3categories as follows:grade1:normal response, bilateral sine wave appeared; grade2:Unilateral or bilateral amplitude reduction; only the healthy lateral wave appeared and the waves didn’t repeat very well, or the lesion-side wave N1was lower than41%of the contralateral side; grade3:neither of them appeared.A total of169patients admitted into NICU of Nanfang hospital, southern medical university. Finally78patients fulfilled our criteria and were included into the final analysis. The average age of the patients were65years (age range:25-74years). The average NIHSS score were18. Fifty-two (67%) patients had left side infarction. Prognosis prediction on30days outcome:In our study,43(55%) patients survived. There was a significant difference between the survival and death groups tested by40-Hz ASSR (P=0.001). The specificity and sensitivity of40-Hz ASSR grade3for predicting the death outcome were98%and34%, respectively. The same findings were observed in the SSEPs recordings (P=0.017). However, no significant differences between the death group and survival group were observed with regard to the BAEP recordings (P>0.05).Prognosis after90days:A total of43survivors were followed up in our neurology clinic at day90from hospital admission or by telephone,1patient lost. The outcome of77patients were dichotomized to favorable (modified Rankin Score0,1,2,3,4) and unfavorable (modified Rankin Score5and6) outcomes. A total-of57patients had an unfavorable outcome. Eight patients were severely disabled. Forty-nine patients died. The clinical parameters did not differ significantly between different groups except baseline NIHSS on admission.The40-Hz ASSR differed significantly between different groups (Z=-2.296, P=0.022). Thirteen patients showed extinct response of two sides. Grade1and Grade2combined predicted a favorable outcome with a sensitivity of100%and a specificity of23%. Grade3alone predicted an unfavorable outcome with a sensitivity of23%and a specificity of100%. The SSEPs results differed statistically between the favorable and the unfavorable groups (Z=-2.874, P=0.004). Unilateral absence of SSEPs was recorded in35patients. Of these35,29had unfavorable outcome. Bilateral absence of SSEPs was seen in10patients. Of these10patients,1patient survived with moderate disability,9had unfavorable outcome.ConclusionIn our study, we found that40Hz ASSR was a relatively reliable and feasible tool in predicting the prognosis of patients with MCA infarction in the NICU. This method may also assist neurointensivists to establish early treatment strategies such as decompressive hemicraniectomy or mild hypothermia. The40-Hz ASSR test provides a convenient, efficient, stable and easy-reading bedside examination for neurointensivists. Larger prospective studies in multiple centers will be required to confirm this conclusion and to address the role of other factors that could potentially influence large hemispheric infarction outcome.