Susan S. Margulies

A proposed tolerance criterion for diffuse axonal injury in man.

The head injury criterion (HIC) is currently the government-accepted head injury indicator. The HIC is not injury-specific, does not relate to injury severity, nor does it take into account variations in the brain mass or load direction. This report focuses on one type of inertial brain injury, diffuse axonal injury (DAI), and utilizes animal studies, physical model experiments, and analytical model simulations to determine the kinematics of DAI in the subhuman primate and to scale these results to man. A human injury tolerance for moderate to severe DAI, which includes the influences of rotational loads and brain mass, is proposed.

Physical model simulations of brain injury in the primate

Diffuse brain injuries resulting from non-impact rotational acceleration are investigated with the aid of physical models of the skull-brain structure. These models provide a unique insight into the relationship between the kinematics of head motion and the associated deformation of the surrogate brain material. Human and baboon skulls filled with optically transparent surrogate brain tissue are subjected to lateral rotations like those shown to produce diffuse injury to the deep white matter in the brain of the baboon. High-speed cinematography captures the deformations of the grids embedded within the surrogate brain tissue during the applied load. The overall deformation pattern is compared to the pathological portrait of diffuse brain injury as determined from animal studies and autopsy reports. Shear strain and pathology spatial distributions mirror each other. Load levels and resulting surrogate brain tissue deformations are related from one species to the other. Increased primate brain mass magnified the strain amplified without significantly altering the spatial distribution. An empirically-derived value for a critical shear strain associated with the onset of severe diffuse axonal injury in primates is determined, assuming constitutive similarity between baboon and human brain tissue. The primate skull physical model data and the critical shear strain associated with the threshold for severe diffuse axonal injury were used to scale data obtained from previous studies to man, and thus derive a diffuse axonal injury tolerance for rotational acceleration for humans.

Age-dependent material properties of the porcine cerebrum: effect on pediatric inertial head injury criteria

During growth and development, the immature central nervous system undergoes rapid alterations in constituents and structure. We hypothesize that these alterations are accompanied by changes in the mechanical properties of brain tissue which, in turn, influence the response of the brain to traumatic inertial loads. Samples of frontal cerebrum from neonatal (2-3 days) and adult pigs were harvested and tested within 3 h post-mortem. The complex shear modulus of the samples was measured in a custom-designed oscillatory shear testing device at engineering shear strain amplitudes of 2.5% or 5% from 20-200 Hz, at 25 degrees C and 100% humidity. In this range, the elastic and viscous components of the complex shear modulus increased significantly with the development of the cerebral region of the brain. Using an idealized model of the developing head, the age-dependent material properties of brain tissue were shown to affect the mechanical response of the brain to inertial loading. This study is a first step toward developing head injury tolerance criteria specifically for the pediatric population.

Infant skull and suture properties: measurements and implications for mechanisms of pediatric brain injury.

The mechanical properties of the adult human skull are well documented, but little information is available for the infant skull. To determine the age-dependent changes in skull properties, we tested human and porcine infant cranial bone in three-point bending. The measurement of elastic modulus in the human and porcine infant cranial bone agrees with and extends previous published data [McPherson, G. K., and Kriewall, T. J. (1980), J. Biomech., 13, pp. 9-16] for human infant cranial bone. After confirming that the porcine and human cranial bone properties were comparable, additional tensile and three-point bending studies were conducted on porcine cranial bone and suture. Comparisons of the porcine infant data with previously published adult human data demonstrate that the elastic modulus, ultimate stress, and energy absorbed to failure increase, and the ultimate strain decreases with age for cranial bone. Likewise, we conclude that the elastic modulus, ultimate stress, and energy absorbed to failure increase with age for sutures. We constructed two finite element models of an idealized one-month old infant head, one with pediatric and the other adult skull properties, and subjected them to impact loading to investigate the contribution of the cranial bone properties on the intracranial tissue deformation pattern. The computational simulations demonstrate that the comparatively compliant skull and membranous suture properties of the infant brain case are associated with large cranial shape changes, and a more diffuse pattern of brain distortion than when the skull takes on adult properties. These studies are a fundamental initial step in predicting the unique mechanical response of the pediatric skull to traumatic loads associated with head injury and, thus, for defining head injury thresholds for children.

Equibiaxial deformation-induced injury of alveolar epithelial cells in vitro

Deformation of the alveolar epithelial basement membrane with lung inflation has been implicated in blood-gas barrier breakdown during the development of ventilator-induced lung injury. To determine the vulnerability of alveolar epithelial cells to deformation-induced injury, we developed a cell-stretching device that subjects cells to cyclic, equibiaxial strains. Alveolar epithelial type II cells from primary culture were tested 1 and 5 days after seeding, during which time the cells underwent major morphological and phenotypic changes. Cells were subjected to changes in surface area of 12, 24, 37, and 50%, which corresponded to lung inflation of ∼60, 80, 100, and >100% of total lung capacity. Deformation-induced injury of alveolar epithelial cells, assessed with a fluorescent cell viability assay, increased with deformation magnitude and decreased with time elapsed after seeding. In cells stretched after 1 day in culture, the percentage of dead cells after a single deformation ranged from 0.5 to 72% over the range of deformations used. In cells stretched at 5 days, the percentage of dead cells ranged from 0 to 9% when exposed to identical deformation protocols. These results suggest that morphological and phenotypic changes with time in culture fundamentally change the vulnerability of alveolar epithelial cells to deformation.

Material characterization of the brainstem from oscillatory shear tests

Traumatic damage to the brainstem occurs frequently when the brain skull complex experiences injurious loading especially during those traumatic situations that produce diffuse axonal injury (DAI). DAI has been shown to be dependent on load direction and correlated with regional tissue deformation in response to rotational inertial loads. Possible mechanisms for the selective vulnerability of the brainstem are (1) the geometry of the central nervous system is responsible for producing high tissue strains in these regions, (2) regional differences in overall material stiffness result in larger deformations at these sites, and (3) the anisotropic mechanical properties of these regions lead to a sensitivity to the rotational load direction and magnitude. This paper investigates the latter two hypotheses by performing oscillatory shear tests on adult porcine brainstem in three mutually perpendicular directions. The complex shear moduli were calculated over a range of frequencies (20-200 Hz), for three levels of peak engineering strain (2.5%, 5.0%, and 7.5%). The directional data demonstrated that the brainstem exhibits significant transversely isotropic behavior. Both components of the complex modulus in which the axonal fibers are oriented parallel to the plane of shear but transverse to the shear direction were significantly higher than those of the other two, mutually indistinguishable test cases across the range of strains tested. By comparison with similar tests on cerebral tissue, these data demonstrated that the brainstem displays a stiffer biomechanical response. These differences were present for both components of the complex shear modulus and were greater as the magnitude of the applied strain increased. The regional stiffness and anisotropic response of the brainstem coupled with its location as a narrow bridge between CNS regions interact to result in the selective vulnerability of this region in rotational loading.

In Vivo Muscle Stiffening Under Bone Compression Promotes Deep Pressure Sores

Pressure sores (PS) in deep muscles are potentially fatal and are considered one of the most costly complications in spinal cord injury patients. We hypothesize that continuous compression of the longissimus and gluteus muscles by the sacral and ischial bones during wheelchair sitting increases muscle stiffness around the bone-muscle interface over time, thereby causing muscles to bear intensified stresses in relentlessly widening regions, in a positive-feedback injury spiral. In this study, we measured long-term shear moduli of muscle tissue in vivo in rats after applying compression (35 KPa or 70 KPa for 1/4-2 h, N = 32), and evaluated tissue viability in matched groups (using phosphotungstic acid hematoxylin histology, N = 10). We found significant (1.8-fold to 3.3-fold, p < 0.05) stiffening of muscle tissue in vivo in muscles subjected to 35 KPa for 30 min or over, and in muscles subjected to 70 KPa for 15 min or over. By incorporating this effect into a finite element (FE) model of the buttocks of a wheelchair user we identified a mechanical stress wave which spreads from the bone-muscle interface outward through longissimus muscle tissue. After 4 h of FE simulated motionlessness, 50%-60% of the cross section of the longissimus was exposed to compressive stresses of 35 KPa or over (shown to induce cell death in rat muscle within 15 min). During these 4 h, the mean compressive stress across the transverse cross section of the longissimus increased by 30%-40%. The identification of the stiffening-stress-cell-death injury spiral developing during the initial 30 min of motionless sitting provides new mechanistic insight into deep PS formation and calls for reevaluation of the 1 h repositioning cycle recommended by the U.S. Department of Health.

Traumatic Axonal Injury is Exacerbated following Repetitive Closed Head Injury in the Neonatal Pig

Inflicted brain injury is associated with widespread traumatic axonal injury (TAI) and subdural hematoma and is the leading cause of death in infants and children. Anesthetized 3-5-day-old piglets were subjected to either a single (n = 5) or double (n = 6, 15 min apart) rapid (<15 msec), non-impact, axial rotations of the head. Peak rotational velocities (averaging 172 rad/sec for single and 138 rad/sec for double loads) were lower than those utilized to induce severe injuries (240-260 rad/sec; Raghupathi and Margulies, 2002). At 6 h post-injury, brains were evaluated for the presence TAI using immunohistochemistry for the 200-kDa neurofilament protein (NF200). Accumulation of NF200 was observed in both contiguous (swellings) and in disconnected axons (axon bulbs) predominantly in central deep and peripheral subcortical white matter regions in the frontal, temporal, and parietal lobes of all injured piglets. Although the density of injured axons did not significantly increase after two rotational loads, the distribution of injured axons shifted from a few foci (2.2 +/- 2.3 per animal) with 1-2 swellings/bulbs following a single rotation to significantly more foci (14.7 +/- 11.9), and additional foci (2.5 +/- 1.9) containing 3 or more axon swellings/bulbs following two rotational loads. The density and distribution of injured axons following a single mild rotation were significantly reduced compared with those obtained previously following a single more severe rotational load. Collectively, these data are indicative of the graded response of the immature brain to rotational load magnitude, and importantly, the vulnerability to repeated, mild, non-impact loading conditions.

Diffuse optical monitoring of hemodynamic changes in piglet brain with closed head injury.

We used a nonimpact inertial rotational model of a closed head injury in neonatal piglets to simulate the conditions following traumatic brain injury in infants. Diffuse optical techniques, including diffuse reflectance spectroscopy and diffuse correlation spectroscopy (DCS), were used to measure cerebral blood oxygenation and blood flow continuously and noninvasively before injury and up to 6 h after the injury. The DCS measurements of relative cerebral blood flow were validated against the fluorescent microsphere method. A strong linear correlation was observed between the two techniques (R=0.89, p<0.00001). Injury-induced cerebral hemodynamic changes were quantified, and significant changes were found in oxy- and deoxy-hemoglobin concentrations, total hemoglobin concentration, blood oxygen saturation, and cerebral blood flow after the injury. The diffuse optical measurements were robust and also correlated well with recordings of vital physiological parameters over the 6-h monitoring period, such as mean arterial blood pressure, arterial oxygen saturation, and heart rate. Finally, the diffuse optical techniques demonstrated sensitivity to dynamic physiological events, such as apnea, cardiac arrest, and hypertonic saline infusion. In total, the investigation corraborates potential of the optical methods for bedside monitoring of pediatric and adult human patients in the neurointensive care unit.

Traumatic Axonal Injury after Closed Head Injury in the Neonatal Pig

Closed head injury is the leading cause of morbidity and mortality in infants and children, and results in pathologies such as diffuse axonal injury (DAI) and subarachnoid hematoma (SAH). To better understand the mechanical environment associated with closed head injury in the pediatric population, animal models that include salient features of human infant brain must be utilized. Based on detailed information regarding the parallels between brain development in the pig and the human, the 3-5-day-old piglet was used to represent the infant at less than 3 months of age. Anesthetized piglets (n = 7) were subjected to rapid, inertial (nonimpact) rotation of the head about its axial plane and sacrificed at 6 h postinjury. Immediately following injury, five of seven piglets were apneic, with an absence of pupillary and pain reflexes. All piglets exhibited severe coma immediately postinjury, but recovered by sacrifice time. Blood was present on the surface of the frontal lobes, cerebellum, and brainstem, and subarachnoid hemorrhage was evident in the frontal cortex. In six of seven brain-injured piglets, accumulation of the 68-kDa neurofilament protein was evident in contiguous axons (swollen) and occasionally in disconnected axons (axonal bulbs), suggestive of traumatic axonal injury (TAI). Mapping of the regional pattern of TAI revealed injured axons predominantly in central and peripheral white matter tracts in the frontal and temporal lobes and in the midbrain. The number of injured axons was equivalent in both hemispheres, and did not correlate to the load applied to the head. Together, these data demonstrate that rapid rotation of the piglet head without impact results in SAH and TAI, similar to that observed in children following severe brain trauma.