Citation

Sharma K, Stevens RD. Handb. Clin. Neurol. 2017; 140: 379-395.

Affiliation

Division of Neurosciences Critical Care, Johns Hopkins University School of Medicine, Baltimore, MD, USA. Electronic address: rstevens@jhmi.edu.

Copyright

(Copyright © 2017, Elsevier Publishing)

DOI

10.1016/B978-0-444-63600-3.00021-0

PMID

28187811

Abstract

A neurocatastrophe or severe brain injury (SBI) is a central nervous system insult associated with a high likelihood of death or severe disability. While many etiologic processes may lead to SBI, the most common and best-studied clinical paradigms are traumatic brain injury and anoxic-ischemic encephalopathy following cardiac arrest. Clinical phenotypes following SBI include acute and chronic disorders of consciousness as well as a range of cognitive and behavioral impairments. A fundamental task for medical teams working in the acute phase is to estimate SBI recovery probabilities with the highest degree of accuracy possible. Predictions made on the basis of single features or variables lack discrimination and are generally supplanted by multivariable models that combine clinical, imaging, and laboratory data into tractable scoring systems. Yet existing scores fail to classify outcomes with the accuracy that would support individual patient-level decision making. Improved prognostication will likely depend on the use of molecular and imaging data that capture unique biologic features in individual patients with SBI. The integration of these additional layers of information will require iterative computational approaches.

© 2017 Elsevier B.V. All rights reserved.

Language: en

Keywords

anoxic-ischemic encephalopathy; coma; intracerebral hemorrhage; minimally conscious state; outcome prediction; prognosis; subarachnoid hemorrhage; traumatic brain injury; vegetative state