Thresholds for Electrophysiological Impairment to in Vivo White Matter Abstract

Allison C. Bain and David F. Meaney Department of Bioengineering University of Pennsylvania An in vivo, tissue-level, mechanical threshold for functional injury to CNS white matter was determined by comparing electrophysiological impairment to estimated tissue strain in an in vivo model of axonal injury. Axonal injury was produced by transiently stretching the right optic nerve of an adult male guinea pig to one of seven levels of ocular displacement (N1eve1=1 0; Ntota1=70). Functional injury was determined by the magnitude of the latency shift of the N35 peak of the visual evoked potentials (VEPs) recorded before and after stretch. A companion set of in situ experiments (NieveF 5) was used to determine the empirical relationship between ocular displacement and optic nerve stretch. Logistic regression analysis, combined with sensitivity and specificity measures and receiver operating characteristic (ROC) curves were then used to predict strain thresholds for axonal injury. From this analysis, we determined three Lagrangian strain-based thresholds for electrophysiological impairment to the optic nerve tissue. The liberal threshold intended to minimize the false positive rate was a strain of 0.28, and the conservative threshold that minimized the false negative rate was 0 . 13. The optimal threshold criteria that balanced the specificity and sensitivity measures was 0 . 18 . With this threshold data, it is now possible to predict more accurately the conditions that cause diffuse axonal injury in man. THE MOST FREQUENT TYPE OF CLOSED HEAD INJURY, diffuse axonal injury (DAI), accounts for the second largest percentage of deaths due to brain trauma (Adams, Doyle et al. 1 989). Although treatment paradigms are being developed for DAI , many areas of injury research now focus on developing the tools and techniques needed to prevent DAI . Central to understanding and evaluating preventive strategies for DAI is determining the local tolerance of the brain tissue to mechanical forces. Tissue tolerance data have become important as several finite element models emerge to transfer macroscopic head motions into estimates of the local stress and strain of the intracranial contents (Ruan, Khalil et al. 1 991 ; Mendis 1 992; Chu, Lin et al. 1 993; Bandak and Eppinger 1 994; Zhou, Khalil et al. 1 994; Ueno, Melvin et al. 1 995; Shreiber and Meaney 1 998). Coupled together, validated finite element models and IRCOBI Conference Sitges (Spain), September 1999 83 84 tissue tolerance data offer a means to identify the hazardous mechanical environments that cause DAI , and offer a complete tool to design countermeasures for reducing the incidence and morbidity of this type of brain injury. The goal of this investigation was to determine tissue-level thresholds for axonal injury using the guinea pig optic nerve stretch model (Gennarelli, Thibault et al. 1 989). In this model, the loading conditions (i.e. tensile stretch) used to produce axonal injury are simple, and are easily controlled and quantified. Also, the simple architecture of the optic nerve permits straightforward quantification of the deformation of the tissue. Thus, it is easy to compare in vivo axonal injury to an experimentally measured strain value in order to define injury tolerances for the tissue. In this study, we use the change in visual evoked potential caused by optic nerve stretch to determine a set of strain-based tissue-level thresholds that define the in vivo conditions associated with electrophysiological changes to the CNS white matter. In addition, we describe the ability of the proposed thresholds to predict morphological and functional axonal injury in practice, using measures of positive (PPV) and negative predictive value (NPV) (Einstein, Bodian et al. 1997; Shalev, Freedman et al. 1 997; Varela, Bosco Lopez Saez et al. 1 997).