Prognostic value of evoked potentials and clinical grading in primary subarachnoid haemorrhage

Brainstem dysfunction in critically ill patients

The brainstem conveys sensory and motor inputs between the spinal cord and the brain, and contains nuclei of the cranial nerves. It controls the sleep-wake cycle and vital functions via the ascending reticular activating system and the autonomic nuclei, respectively. Brainstem dysfunction may lead to sensory and motor deficits, cranial nerve palsies, impairment of consciousness, dysautonomia, and respiratory failure. The brainstem is prone to various primary and secondary insults, resulting in acute or chronic dysfunction. Of particular importance for characterizing brainstem dysfunction and identifying the underlying etiology are a detailed clinical examination, MRI, neurophysiologic tests such as brainstem auditory evoked potentials, and an analysis of the cerebrospinal fluid. Detection of brainstem dysfunction is challenging but of utmost importance in comatose and deeply sedated patients both to guide therapy and to support outcome prediction. In the present review, we summarize the neuroanatomy, clinical syndromes, and diagnostic techniques of critical illness-associated brainstem dysfunction for the critical care setting.

Mismatch negativity to predict subsequent awakening in deeply sedated critically ill patients

BACKGROUND

Mismatch negativity (MMN) is the neurophysiological correlate of cognitive integration of novel stimuli. Although MMN is a well-established predictor of awakening in non-sedated comatose patients, its prognostic value in deeply sedated critically ill patients remains unknown. The aim of this prospective, observational pilot study was to investigate the prognostic value of MMN for subsequent awakening in deeply sedated critically ill patients.

METHODS

MMN was recorded in 43 deeply sedated critically ill patients on Day 3 of ICU admission using a classical 'odd-ball' paradigm that delivers rare deviant sounds in a train of frequent standard sounds. Individual visual analyses and a group level analysis of recordings were performed. MMN amplitudes were then analysed according to the neurological status (awake vs not awake) at Day 28.

RESULTS

Median (inter-quartile range) Richmond Assessment Sedation Scale (RASS) at the time of recording was -5 (range, from -5 to -4.5). Visual detection of MMN revealed a poor inter-rater agreement [kappa=0.17, 95% confidence interval (0.07-0.26)]. On Day 28, 30 (70%) patients had regained consciousness while 13 (30%) had not. Quantitative group level analysis revealed a significantly greater MMN amplitude for patients who awakened compared with those who had not [mean (standard deviation) = -0.65 (1.4) vs 0.08 (0.17) μV, respectively; P=0.003).

CONCLUSIONS

MMN can be observed in deeply sedated critically ill patients and could help predict subsequent awakening. However, visual analysis alone is unreliable and should be systematically completed with individual level statistics.

Proving cortical death after vascular coma: Evoked potentials, EEG and neuroimaging

OBJECTIVES

Several studies have shown that bilateral abolition of somatosensory evoked potentials after a nontraumatic coma has 100% specificity for nonawakening with ethical consequences for active care withdrawal. We propose to evaluate the prognostic value of bilateral abolished cortical components of SEPs in severe vascular coma.

METHODS

A total of 144 comatose patients after subarachnoid haemorrhage were evaluated by multimodal evoked potentials (EPs); 7 patients presented a bilateral abolition of somatosensory and auditory EPs. Their prognosis value was interpreted with respect to brainstem auditory EPs, EEG, and structural imaging.

RESULTS

One patient emerged from vegetative state during follow-up; 6 patients did not return to consciousness. The main neurophysiological difference was a cortical reactivity to pain preserved in the patient who returned to consciousness. This patient had focal sub-cortical lesions, which could explain the abolition of primary cortical components by a bilateral deafferentation of somatosensory and auditory pathways.

CONCLUSIONS

This is the first report of a favourable outcome after a multimodal abolition of primary cortex EPs in vascular coma. For the 3 cases of vascular coma with preserved brainstem function, EEG reactivity and cortical EPs were abolished by a diffuse ischaemia close to cerebral anoxia.

SIGNIFICANCE

The complementarity of EPs, EEG, and imaging must be emphasised if therapeutic limitations are considered to avoid over-interpretation of the prognosis value of EPs.

Early impairment of intracranial conduction time predicts mortality in deeply sedated critically ill patients: a prospective observational pilot study

Background

Somatosensory (SSEP) and brainstem auditory (BAEP) evoked potentials are neurophysiological tools which, respectively, explore the intracranial conduction time (ICCT) and the intrapontine conduction time (IPCT). The prognostic values of prolonged cerebral conduction times in deeply sedated patients have never been assessed. Sedated patients are at risk of developing new neurological complications, undetected. In this prospective observational bi-center pilot study, we investigated whether early impairment of SSEP’s ICCT and/or BAEP’s IPCT could predict in-ICU mortality or altered mental status (AMS), in deeply sedated critically ill patients.

Methods

SSEP by stimulation of the median nerve and BAEP were assessed in critically ill patients receiving deep sedation on day 3 following ICU admission. Deep sedation was defined by a Richmond Assessment sedation Scale (RASS) <−3. Mean left- and right-side ICCT and IPCT were measured for each patient. Primary and secondary outcomes were, respectively, in-ICU mortality and AMS defined as the occurrence of delirium and/or delayed awakening after discontinuation of sedation.

Results

Eighty-six patients were studied of which 49 (57%) were non-brain-injured and 37 (43%) were brain-injured. Impaired ICCT was a predictor of in-ICU mortality after adjustment on the global Sequential Organ Failure Assessment score (SOFA) [OR (95% CI) = 2.69 (1.05–6.85); p = 0.039] and on the non-neurological SOFA components [2.67 (1.05–6.81); p = 0.040]. IPCT was more frequently delayed in the subgroup of patients who developed post-sedation AMS (24%) compared those without AMS (0%). However, this difference did not reach statistical significance (p = 0.053). Impairment rates of ICCT and IPCT were not found to be significantly different between non-brain- and brain-injured subgroups of patients.

Conclusion

In critically ill patients receiving deep sedation, early ICCT impairment was associated with mortality. Somatosensory and brainstem auditory evoked potentials may be useful early warning indicators of brain dysfunction as well as prognostic markers in deeply sedated critically ill patients.

Prognostic Value of Evoked Responses and Event-Related Brain Potentials in

The behaviourally unresponsive patient, unable to exhibit the presence of cognition, constitutes a conundrum for health care specialists. Prognostic uncertainty impedes accurate management decisions and the application of ethical principles. An early, reliable prognosis is highly desirable. In this review investigations studying comatose patients with coma of different etiologies were selected. It is concluded that objective prognostication is enhanced by the use of electrophysiological tests. Persistent abnormalities of brainstem auditory evoked potentials and short-latency somatosensory evoked potentials reliably indicate the likelihood of irreversible neurological deficit or death. Meanwhile, the presence of “cognitive” event-related brain potentials (e.g., P300 and mismatch negativity) reflects the functional integrity of higher level information processing and, therefore, the likelihood of capacity for cognition. An approach that combines clinical and electrophysiological values provides optimal prediction of outcome and level of disability.

[Multimodal monitoring in neurointensive care medicine: state of the art]

The prognosis of neurointensive care patients depends largely on the occurrence of secondary ischemic/hypoxic tissue damage, which is mediated by different pathomechanisms, such as edema formation or increased intracranial pressure. Due to the cerebral damage and need for sedation as well as intubation, clinical assessment of these patients is limited. Furthermore, clinical signs of secondary damage, such as advanced herniation syndromes are often delayed and therefore mostly indicate irreversible brain damage. To adequately predict and detect secondary neuronal damage, various neuromonitoring techniques have been developed in recent years with ongoing technical refinement. These can be used for bedside and ideally continuous monitoring of various functional systems of the brain. Neuromonitoring is used to implement early therapeutic measures before irreversible brain damage has occurred, to monitor therapeutic effects, for evaluation of the prognosis and to improve the neurological outcome of patients. Different monitoring techniques are often combined in multimodal neuromonitoring. This article gives an overview of the most promising neuromonitoring techniques available.

Predictive value of somatosensory evoked potentials for awakening from coma

OBJECTIVES

A systematic review of somatosensory evoked potentials performed early after onset of coma, to predict the likelihood of nonawakening. The pooled results were evaluated for rates of awakening, confidence intervals, and the possibility of rare exceptions.

DATA SOURCES

Forty-one articles reporting somatosensory evoked potentials in comatose patients and subsequent outcomes, from 1983 to 2000.

STUDY SELECTION

Studies were included if they reported coma etiology, age group, presence or absence of somatosensory evoked potentials, and coma outcomes.

DATA EXTRACTION

We separated patients into four groups: adults with hypoxic-ischemic encephalopathy, adults with intracranial hemorrhage, adults and adolescents with traumatic brain injury, and children and adolescents with any etiologies. Somatosensory evoked potentials were categorized as normal, abnormal, or bilaterally absent. Outcomes were categorized as persistent vegetative state or death vs. awakening.

DATA SYNTHESIS

For each somatosensory evoked potential result, rates of awakening (95% confidence interval) were calculated: adult hypoxic-ischemic encephalopathy: absent 0% (0%-1%), abnormal 22% (17%-26%), normal 52% (48%-56%); adult intracranial hemorrhage: absent 1% (0%-4%), present 38% (27%-48%); adult-teen traumatic brain injury: absent 5% (2%-7%), abnormal 70% (64%-75%), normal 89% (85%-92%); child-teen: absent 7% (4%-10%), abnormal 69% (61%-77%), normal 86% (80%-92%).

CONCLUSIONS

Somatosensory evoked potential results predict the likelihood of nonawakening from coma with a high level of certainty. Adults in coma from hypoxic-ischemic encephalopathy with absent somatosensory evoked potential responses have <1% chance of awakening.

Prognostic value of SSEP in early aneurysm surgery after SAH in poor-grade patients

We evaluate the prognostic value of somatosensory evoked potentials (SSEP) in poor-grade patients after early surgery for aneurysmal subarachnoid hemorrhage compared to the Hunt and Hess (H&H) and WFNS scales. Ninety patients with angiographically proven aneurysms graded H&H IV or V were evaluated retrospectively. The aneurysms of 72 patients were clipped. In 53 out of 72 patients 147 SSEP examinations were recorded. The SSEP were classified according to the central conduction time (CCT) and the number of cortical potentials. Outcome was determined according to the Glasgow Outcome Scale. To evaluate the predictability of the SSEP to clinical grading scales receiver operating characteristic (ROC) analysis was done. The H&H scale did not demonstrate statistically significant predictability for poor-grade patients. The WFNS scale predicted the outcome for only one group (survival/death) (p = 0.035). Predictability of outcome by the SSEP was statistically confirmed. Normal CCT indicated a potential for a good recovery, but not consistently so. Bilaterally enhanced CCT was predictive of a poor outcome. Bilateral lack of cortical responses was always related to fatal outcome. ROC analysis confirmed that SSEP are superior to clinical grading scales in determining prognosis in poor-grade patients. In doubt, whether early aneurysm surgery or conservative treatment in a poor-grade patient should be done, SSEP will be helpful.

Correlation between central somatosensory conduction time, blood flow velocity, and delayed cerebral ischemia after aneurysmal subarachnoid hemorrhage

In this retrospective study of 67 aneurysmal patients, the predictive role of central conduction time (CCT) on vasospasm occurrence evaluated by means of transcranial Doppler sonography (TCD) and the correlation of CCT to blood flow velocity measured simultaneously in postoperative course were studied. Data about the clinical state of patients at the time of admission (Hunt Hess scale), severity of subarachnoidal hemorrhage on initial CT scan (Fisher grade), timing of surgery (acute or delayed), outcome (Glasgow Outcome Scale), severity of vasospasm graded by highest mean blood flow velocity (BFV) during the entire clinical course and CCT values measured at admission (preoperatively), then postoperatively (one day after surgery) and simultaneously with later TCD investigations were collected from the files. Interhemispheric difference of CCT was also calculated. The results showed that CCT at admission was not predictive for vasospasm. CCT measured either at admission or on the first postoperative day did not differ significantly in the different grades of vasospasm. Similar results were obtained in the acute and in the late operated group of patients. The results also suggest that increased CCT and interhemispheric difference at the time of admission indicate a worse prognosis, but this can be related to higher surgical risk rather than to a higher incidence of late ischemic deterioration. Simultaneous CCT and TCD examinations demonstrated that coincident and statistically significant (p < 0.01) increase of actual CCT (6.7 msec) was found only in the severe grade of vasospasm (BFV 200 cm/s). The authors discuss the role of CCT and TCD monitoring in the management of aneurysmal subarachnoid hemorrhage.

Neurological grades of patients with poor-grade subarachnoid hemorrhage improve after short-term pretreatment

OBJECTIVE

Short-term pretreatment of patients with subarachnoid hemorrhage, but without hematomas causing mass effect, who presented in poor neurological condition at admission was evaluated as a protocol for the selection of candidates for radical surgery.

METHODS

One hundred-three patients were pretreated for 12 hours with control of blood pressure and intracranial pressure, using diuretic agents and/or ventricular drainage.

RESULTS

Neurological improvement was observed for 32 of 47 patients in Grade IV at admission and 23 of 56 patients in Grade V (P < 0.01). Hydrocephalus requiring drainage was more common (P < 0.05) and the interval between onset and admission was shorter (P < 0.01) for the improved group. Clipping surgery was performed for all patients in Grade III or better and for patients in Grade IV who were less than 75 years of age and without systemic complications, i.e., 38 of 47 patients in Grade IV and 16 of 56 patients in Grade V at admission. Good outcomes (defined as moderately disabled or better on the Glasgow Outcome Scale) were achieved by 34 of 38 patients in Grade IV and 10 of 16 patients in Grade V (P < 0.01). The proportion of patients in Grade IV after pretreatment was lower for Grade IV (2 of 38 patients) than for Grade V (9 of 16 patients) (P < 0.00001). However, none of the 49 patients who underwent nonsurgical treatment achieved good outcomes.

CONCLUSION

Our protocol may be beneficial for the selection of candidates for radical surgery among patients with subarachnoid hemorrhage but without hematomas who are in poor neurological condition at admission and for the improvement of postoperative outcomes.

Serial evoked potentials and outcome in cerebrovascular critical care patients

Median-nerve evoked somatosensory evoked potentials (SEPs) and brainstem auditory evoked potentials (BAEPs), examined early in the course of patients suffering from cerebrovascular disease, correlate statistically significantly with outcome. Little is known about the changes of evoked potentials in the course of disease and their correlation to outcome. In a series of 215 patients (75 supratentorial infarctions, 36 infratentorial infarctions, 58 supratentorial hemorrhages, 18 infratentorial hemorrhages, and 28 aneurysmatic subarachnoid hemorrhages) requiring neurologic intensive care treatment, we prospectively examined the correlation between the findings of serial SEPs and BAEPs and outcome at 4 weeks. Evoked potentials were examined after admission, after 1 week, and after 2 weeks. The findings were classified in 4 categories (normal, unilateral or bilateral pathologic findings, unilaterally attenuated, and bilaterally attenuated). Clinical outcome was determined by classification according to the Glasgow Outcome Scale (death, persistent vegetative state, severely incapacitated, mildly incapacitated, and recovery). Statistical evaluation was performed using Fisher's exact test for all variables. In all subgroups, SEPs correlated statistically significantly with outcome at all three examinations. No correlation was found for BAEPs at first examination in infratentorial disease, nor at second examination in subarachnoid hemorrhages. In all other cases, SEPs and BAEPs were correlated statistically significantly with outcome at all three examination timepoints.

Auditory and somatosensory evoked potentials in coma following spontaneous cerebral hemorrhage: early prognosis and outcome

The aim of this study was to check the prognostic power of auditory brain-stem responses (ABRs) and somatosensory evoked potentials (SEPs) in coma following spontaneous cerebral hemorrhage. Seventy patients comatose following subarachnoidal or hypertensive hemorrhage were submitted to ABR and SEP recordings during the acute phase of clinical course. Twenty-one patients survived (30%), two remained vegetative (2.9%) and 47 died (68.1%). The Glasgow Coma Score (GCS) was significantly related to the outcome (P < 0.001), but showed a low sensitivity, since about 50% of patients with GCS = 5-8 died or remained vegetative. ABRs and SEPs showed a much closer correlation with outcome (P < 0.001): their combined use allowed there to be a sensitivity of 96%, a specificity of 90% and a predictive power of 96%; the relative risk of poor outcome in patients with at least one abnormal modality was equal to 223 times the one for patients with normal evoked potentials. Moreover, in surviving patients a significant relationship appeared to exist between abnormalities of SEPs during the acute phase and the severity of disability. Our results confirm the prognostic effectiveness of short latency evoked potentials in cerebral hemorrhage: they are far superior to clinical data, being able to yield a marked decrease of falsely optimistic predictions.

Contribution of initial median-nerve somatosensory evoked potentials and brainstem auditory evoked potentials to prediction of clinical outcome in cerebrovascular critical care patients: a statistical evaluation

In a prospective study of 200 patients with cerebrovascular disease (48 intracerebral hemorrhages, [ICH]; 64 subarachnoid hemorrhages, [SAH]; 48 supratentorial and 40 infratentorial ischemic strokes), we assessed the individual and combined prognostic value of median-nerve somatosensory evoked potentials (SEP) and brainstem auditory evoked potentials (BAEP) within 72 hours of admission. Clinical outcome was graded in three ranked categories according to a modified Glasgow Outcome Scale. Likewise, the initial SEP and BAEP findings were graded in a three-class score. In all groups, the SEP were significantly correlated with outcome (P < 0.01). Likewise, after partialling out the prognostic effect of SEP, the contingency between BAEP and outcome was statistically significant, except in ICH (P = 0.07). The contingencies of SEP and outcome in ICH and supratentorial infarcts were higher than the corresponding partial contingencies for BAEP, while the latter were higher in infratentorial infarction and SAH. Therefore, in all disease groups except for SAH, the multiple contingency coefficients ranging from 0.67 to 0.75 were statistically significant and greater than either simple or partial contingencies alone. The results of the two evoked potential modalities combined permit statistically significant superior prognostication in most cerebrovascular diseases when compared to those of either of the modalities alone.