Predicting unconsciousness from a pediatric brain injury threshold

Abstract

The objective of this study was to utilize tissue deformation thresholds associated with acute axonal injury in the immature brain to predict the duration of unconsciousness. Ten anesthetized 3- to 5-day-old piglets were subjected to nonimpact axial rotations (110-260 rad/s) producing graded injury, with periods of unconsciousness from 0 to 80 min. Coronal sections of the perfusion-fixed brain were immunostained with neurofilament antibody (NF-68) and examined microscopically to identify regions of swollen axons and terminal retraction balls. Each experiment was simulated with a finite element computational model of the piglet brain and the recorded head velocity traces to estimate the local tissue deformation (strain), the strain rate and their product. Using thresholds associated with 50, 80 and 90% probability of axonal injury, white matter regions experiencing suprathreshold responses were determined and expressed as a fraction of the total white matter volume. These volume fractions were then correlated with the duration of unconsciousness, assuming a linear relationship. The thresholds for 80 and 90% probability of predicting injury were found to correlate better with injury severity than those for 50%, and the product of strain and strain rate was the best predictor of injury severity (p=0.02). Predictive capacity of the linear relationship was confirmed with additional (n=13) animal experiments. We conclude that the suprathreshold injured volume can provide a satisfactory prediction of injury severity in the immature brain.