Pain-related somatosensory evoked potentials and functional brain magnetic resonance in the evaluation of neurologic recovery after cardiac arrest: a case study of three patients

Early and accurate prognostic assessment of neurological functional outcomes in comatose patients after cardiac arrest is a relevant medical, ethical, and economic issue. It has been shown that beyond the Glasgow Coma Scale (GCS), a patient's pupil light reactivity, corneal reflexes, myoclonus status epilepticus, and serum neuron-specific enolase, short latency somatosensory-evoked potentials (SEPs) (N20/P25) improve the accuracy of neurological prognosis in comatose patients after cardiac arrest [1]. SEPs have shown high sensitivity and specificity in predicting poor outcomes. Indeed, that the bilateral disappearance of cortical N20/P25 is well-established to be associated with adverse outcomes such as death or survival in a vegetative state. Nonetheless, the presence of N20/P25 may not be sensitive enough to predict a good neurological outcome [2]. In fact, only the event-related evoked potentials (i.e., mismatched negativity and novelty P300), middle latency cortical somatosensory-evoked potentials (MLCEPs), and reactivity electroencephalogram (EEG) background have been associated with a favourable neurological prognosis [3-7].In recent years, brain functional neuroimaging has been used in order to clarify the diagnosis of the vegetative state, suggesting that brain activation imaging may reflect awareness and/or cognition and provide reliable prognostic information [8,9]. However, given that this methodology is based on imagery and communication task paradigms, this approach is only feasible in the chronic phase of consciousness disorders. Moreover, studies using positron emission tomography have shown that minimally conscious state patients, compared to vegetative state patients, may show brain processing activation elicited by noxious electrical stimulation of the median nerves similar to that seen in healthy controls, suggesting a possible cortical processing of pain [10-12]. However, the intensity of electrical stimuli applied in vegetative patients (i.e., 1