DOI: 10.1097/01.CCM.0000053668.75972.00

,  

,  

PMID: 12627030

Issn Print: 0090-3493

Publication Date: 2003/03/01


Print

Clinical neurophysiologic assessment of comatose patients *

G. Bryan Young

    loading  Checking for direct PDF access through Ovid

Excerpt

Neurologic prognostication of the comatose patient within the first few days of coma onset is a common, difficult problem. Clinical assessment primarily addresses brainstem function. The condition of the cerebral cortex and thalamus, the most vulnerable part of the brain and yet the seat of awareness and cognitive function, is clinically inaccessible. Of course, if the criteria for brain death are met, assessment need go no further. Even if brainstem function is partly compromised, we cannot reliably infer that the usually more vulnerable cerebrum is irreversibly devastated; some exceptions occur (1, 2). Most often, brainstem function is entirely preserved or recovers quickly. According to the Multi-Society Task Force on Persistent Vegetative State (PVS), the earliest one can, using clinical methods, reliably predict outcome of a state better than PVS is 3 months for nontraumatic and 6 months for traumatic coma for both adults and children (3). This is an inordinately long time to wait and has enormous implications for resources and suffering families. For reliable, early prognostication of comatose patients, we need to go beyond clinical examination to investigations that reliably access and assess thalamocortical function. Electrophysiologic techniques offer insights into this function; the article by Dr. Robinson and colleagues (4) in this issue of Critical Care Medicine is a timely review of the promise offered by somatosensory evoked potential (SEP) testing.
Dr. Robinson and colleagues conducted a systematic literature review of SEP findings and associated outcomes in hypoxic ischemic encephalopathy, intracranial hemorrhage, traumatic brain injury, and children and adolescents with various etiologies for coma. They used well-defined inclusion and exclusion criteria and examined sources of variation and potential flaws. As the authors indicated, difficulties with this endeavor include the variations in methodology of the assembled articles that met criteria, variability of testing (stimulus rates, timing) and varied follow-up times. The authors took as their end point the recovery of “wakening.” This was not defined, but clearly is more than the alerting or arousal response and return of wake–sleep cycles that characterize PVS. “Awareness” might be a better term, but this contains numerous subconcepts, including sensation, perception, links with memory, language function, motivation, judgment, and cognitive function. Unfortunately, in some patients, only some of these functions are regained—testimony to the modular organization of the brain; and the patient who appears to be “aware” may still be quite disabled (5). However, the article is valuable, especially for establishing the positive predictive values for an outcome of no better than PVS. They confirm the earlier metaanalysis of Zandbergen and colleagues (6) in hypoxic ischemic encephalopathy: SEPs have nearly 100% positive predictive value for an outcome of no better than PVS if the N20 response (initial waveform from the cortical primary sensory area with median nerve stimulation) is absent bilaterally. For intracranial hemorrhage and trauma, approximately 1% and 4%, respectively, of patients with absent responses may recover. A slightly higher percentage of children may recover. More extensive work is needed for children, who often show periods of continued improvement that are much longer than adults (7, 8).
There are several implications of the report by Dr. Robinson and colleagues (4). Great care should be taken in the timing of testing (usually not in the first 24 hrs); stimulation should be <5 Hz; it should be assured that the earlier waveforms along the somatosensory pathway are intact, e.g., the Erb’s point (brachial plexus) and posterior-column-medullary (N14) potentials. One should keep in mind that one is testing only a very limited sample of the nervous system. It is theoretically possible this pathway may be more severely affected than other brain regions, e.g.
    loading  Loading Related Articles