|Year : 2018 | Volume
| Issue : 2 | Page : 91-100
Brain injury biomechanics and abusive head trauma
Rudy J Castellani1, Carl J Schmidt2
1 Center for Neuropathology, Western Michigan University Homer Stryker MD School of Medicine, Kalamazoo; Department of Pathology, University of Maryland School of Medicine, Baltimore, MD, USA
2 Department of Pathology, University of Michigan, Ann Arbor, Michigan, USA
|Date of Web Publication||29-Jun-2018|
Dr. Rudy J Castellani
Center for Neuropathology, Western Michigan University Homer Stryker MD School of Medicine, Kalamazoo, Michigan; Department of Pathology, University of Maryland School of Medicine, Baltimore, MD
Source of Support: None, Conflict of Interest: None
Contemporary biomechanical theory of traumatic brain injury has its foundation in Holbourn's thesis on shear strain and Ommaya's primate experimentation demonstrating the role of rotation in a variety of lesions including subdural hematoma (SDH) and diffuse axonal injury. Empirical human observations have since confirmed, for the most part, the early concepts. Ethical concerns regarding primate research, however, have prompted in vitro models, which in turn has led to challenges with respect to the correlation between in vitro observations and the clinical data. Despite these challenges, medicolegal proceedings may call upon biomechanical engineers to reconstruct complex injury scenarios and offer opinions on the scientific plausibility of clinical disease states, such as SDH, hemorrhagic retinopathy, and cerebral edema, associated with hypothetical or proffered action sequences during the course of an unwitnessed homicide. It is important to note, however, that in vitro models by their nature are low-evidence quality studies that attempt to advance hypotheses but do not address cause and effect. As a whole, biomechanical models, as they pertain specifically to the brain and spine, are mathematically imprecise. Often, endpoints of limited relevance are relied upon (e.g., skull fracture thresholds), which predictably overestimate the in vivo risk of significant injury. Given the increasing role of biomechanical engineering in the interpretation of fatal pediatric head trauma, a heightened awareness of the limitations warranted.
Keywords: Abusive head trauma, biomechanics, brain injury, impact, subdural hematoma
|How to cite this article:|
Castellani RJ, Schmidt CJ. Brain injury biomechanics and abusive head trauma. J Forensic Sci Med 2018;4:91-100
| Introduction|| |
As difficult as it is to relate in vitro biological chemistry to the biologic basis of disease at the whole organism level, even more challenging may be the application of mathematics and physics to trauma biology. Clinical data continue to be more meaningful. Human cadaver studies, anthropomorphic models, and finite-element analyses, for example, predict significant risks for subdural hematoma (SDH), neck injury, and skull fracture in pediatric populations with seemingly innocuous falls. Yet, SDH is absent from the epidemiological fall literature in young children, and the risk of neck injury and serious skull fracture from short falls is quite low [Table 1]. In addition, many clinical or biological factors are not adequately addressed in experimental models, including the age of the decedent; the level of bony development with respect to age; skull thickness; cranial suture thickness; the clinical condition of the decedent; the presence or absence of preexisting structural brain injury; the multiplicity and time course of inflicted traumatic events; the directionality of the blows with respect to the axis of rotation; the relationship of any impact to the center of gravity; the frequency of impact versus whiplash; the impact surface; the relative contributions of swelling, hemorrhage, apnea, seizures, ischemia, and increased intracranial pressure to the clinical decompensation; the various physical forces that differentially affect trauma to bone versus soft tissue versus gray matter versus white matter; the relative contributions of translational versus rotational trauma; and the extent of brainstem and upper cervical spinal cord damage, among many other potential factors. It is, therefore, important to conclude at the outset that threshold levels of force (or any other measured quantity) required for injury cannot be defined in mathematical terms. It is also important to point out that experimental models of any construct fail to address human components such as clinical history, changing histories, delays in seeking health care, organ system examination outside the central nervous system, and laboratory/radiology findings, for which there is abundant literature. In short, the clinical context, which is critical to the diagnosis of child abuse and the development of an appropriate differential diagnosis, is not a feature of experimental models.
|Table 1: Proposed risks for subdural hematoma, serious neck injury, and skull fracture from short falls based on biomechanical models, in comparison with the risks based on clinical studies|
Click here to view
In order to address issues that more directly relate to neurologic impairment, the below discussion focuses on the intracranial compartment. Importantly, any discussion of biomechanical factors involved in lethality is necessarily superficial and any conclusive remarks are preliminary. This is particularly true for abusive head trauma, in which cognitive, personal, and pecuniary biases are abundant, disinterested witnesses are uncommon, and directly measurable biomechanical quantities are ethically precluded.
| Physical Properties of the Brain|| |
The biomechanics of brain injury has been a subject of interest for many decades. In a seminal article in 1943, AHS Holbourn, a physicist, noted the following basic properties of the brain: (i) it has relatively uniform density; (ii) it is extremely incompressible (due to high water content), such that a force of about 10,000 tons is required to reduce the brain to one half its volume; and (iii) it has a low modulus of rigidity, i.e., the brain offers little resistance to changes in shape compared to changes in size. The low modulus of rigidity combined with extreme incompressibility, means that tissue damage, or pulling apart of constituent particles, is proportional to the shearing forces (movement of one tissue plane over another). These properties dictate that rotational forces, due to motion around an axis of rotation, are capable of producing tissue injury through shearing (such as concussion and traumatic axonal injury) with an intact skull. In contrast, translational forces, being largely compressive in nature, have little effect on brain tissue integrity in the absence of failure of the overlying bone.
In addition to producing tissue damage through shearing, rotational forces applied to the head lead to differential movements of the brain with respect to the skull and dura. This appears to underlie the contrecoup phenomenon, where contusion may occur opposite to the point of impact. More relevant to abusive head trauma is strain on bridging veins due to movement of the brain against veins fixed to the dura. In addition, whereas brain parenchymal damage (e.g., diffuse axonal injury) appears more likely with low strain rate mechanical forces (strain rate in this context being acceleration, rather than length, divided by pulse duration or time. Acceleration over time is also referred to as “jerk”) and coronal plane acceleration,, breaching of the ultimate tensile strain of bridging veins appears more likely with sagittal plane acceleration and is less related to pulse duration., The result is a complex interplay between applied forces, contact surfaces, directionality of the trauma, and primary trauma outcomes (vascular injury versus parenchymal brain injury), even as rotation appears to be the rate-limiting mechanical event for both subdural hemorrhage and diffuse axonal injury.
| Biomechanics of Subdural Hemorrhage|| |
Acute subdural hemorrhage secondary to trauma is believed to result from one of the three phenomena–direct laceration of cortical arteries and/or veins from contact or penetrating injury, cerebral contusions that result in associated bleeding into the adjacent subdural space also from contact injury (“burst lobe”), and/or bridging vessel rupture from rotational forces (often venous, but sometimes arterial). The so-called contact subdural hemorrhages are alluded to only in passing in the literature, although contact events can, on occasion, produce small subdural hemorrhages. In general, these are clinically benign, and lack the mass effect and brain injury that often occur with rotational bridging vein rupture. Direct arterial or venous laceration from contact penetrating injury is uncommon in abusive head trauma, as are cerebral contusions that communicate with the subdural space. The principle mechanism at hand, therefore, is bridging vein rupture via rotational force.
The potential for subdural hemorrhage in the absence of impact was initially described by Ommaya in the late 1960s  and was followed by primate experiments demonstrating the importance of sagittal plane rotational acceleration,, noting that substantial SDHs were only seen after the rotational component was introduced. Pure translational forces produced focal but not diffuse lesions, although Ommaya did note that local skull deformation likely contributed to the injury process based on these experimental data. The study of rhesus monkeys by Gennarelli and Thibault went so far as to say that a collision with pure translational acceleration was unable to cause bridging vessel rupture and subdural hemorrhage. The fact that SDH is essentially absent from the translational free fall literature in children supports this point. In a finite-element analysis by Huang et al., the hypothetical translational acceleration threshold of bridging vein rupture in humans was estimated at 3913 g. This value is approximately 40 times the average translational acceleration experienced during a severe impact resulting in concussion in professional athletes, emphasizing the inapplicability of translational forces to bridging vein injury in vivo.
It should also be noted that the specific mathematical quantity responsible for morbidity and mortality of head trauma is unknown. In other words, whether velocity, acceleration, strain rate, force, energy, or power, or some other physical quantity is more or less relevant to parenchymal tissue or blood vessel injury has not been explored in depth. In diffuse axonal injury, for example, it is not just the magnitude of rotational acceleration but the time interval over which the acceleration occurs that seems to influence the pathological outcome., In a head trauma study using a biofidelic model, peak rotational acceleration differed significantly as a function of the impacting surface (carpet versus concrete), while rotational velocity did not. In short, even the most sophisticated biomechanical models only superficially address intracranial injury thresholds and potential outcomes.
Given the importance of rotation in traumatic brain injury, it is interesting that the safety of automobiles and sporting equipment has historically been assessed using the “head injury criterion,” which takes into account neither rotational loads nor directional dependency,, and it is based on skull fracture thresholds (most skull fractures in children are clinically benign ). For example, biomechanical modeling of infants using anthropomorphic test dummies (ATDs) designed to test child restraint airbag interaction (CRABI) proposed a peak translational acceleration threshold of 51 g, chosen to represent a 5% risk of “significant head injury.” 83 g has been proposed as an injury threshold for older children. Mohan et al. proposed conservative tolerance limits of 200–250 g peak acceleration for children, based on computerized free fall (i.e., translational acceleration) reconstruction, despite the limited applicability of falls and translational acceleration to concussion, diffuse traumatic axonal injury, and SDH. Still others have proposed tolerance limits between 50 g and 200 g, with 50 g being the before-injury threshold and 200 g representing fatal injury threshold, again relating only indirectly to rotational acceleration.
Thompson et al. noted maximum linear acceleration of 423 g using CRABI ATD short falls onto a hard linoleum surface (all other surfaces were <200 g), raising the issue of impact surface. They also concluded that the likelihood of serious head injury was low for most surfaces, but that “serious head injury” is theoretically possible with impact on hard surfaces. Complicating this latter point, pulse duration with falls onto hard surfaces was low (mean 5.4 ms), while severity of injury tends to increase with increasing pulse duration, and increasing pulse duration decreases rotational injury thresholds., This raises the possibility that impacts of a given magnitude on soft surfaces with longer pulse durations may produce severe brain injury, possibly at a lower threshold of applied force (discussed below). Transient deformation of the skull and brain appears to play a minor role relative to rotation because of the empirical benignity of short falls, although an age-related variation in pulse duration (maturation of the sutures and skull) on impact may have some influence on bridging vein rupture thresholds.
Thresholds for bridging vein rupture
A number of studies have suggested critical thresholds for bridging vein rupture, although these are widely variable among research groups [Table 2]. The first cadaveric study was reported by Löwenhielm in 1974. Using occipital impacts and impulse durations ranging from 15 to 44 ms in cadaveric specimens, the authors proposed 4500 radians/s 2 as the critical threshold of bridging vessel rupture and SDH formation. A complementary study was performed by Depreitere et al. in 2006. This study also used occipital impacts in cadaveric specimens; however, impulse durations were shorter–lasting <10 ms, in order to model impact from falls (an issue again of marginal relevance to SDH in young children). Data obtained from this study suggested a tolerance level for SDH of approximately 10,000 radians/s 2 which decreased with longer pulse durations. A finite-element analysis by Kleiven, on the other hand, suggested a threshold for SDH of 50 kW  (the head injury power criterion was introduced in 2000 to objectively measure mechanical force normalized for unit time). This corresponded to a rotational acceleration threshold of approximately 34,000 radians/s 2 for a 5 ms' duration and an angular velocity of 85 radians/s. Huang et al. suggested a critical rotational acceleration magnitude of 71,200 radians/s 2 for bridging vein rupture, based on a three-dimensional finite-element analysis.
|Table 2: Proposed subdural hemorrhage or bridging vein rupture thresholds (rotational acceleration in radians/s2 and peak short-fall rotation in experimental models|
Click here to view
The magnitudes of proposed rotational acceleration thresholds thus differ by nearly a factor of 16 across studies. The above also presupposes uniformity of bridging veins, which is an oversimplification. Despite their importance in biomechanical theory, the properties of bridging veins with respect to microstructural anatomy and response to mechanical trauma are incompletely understood. Limited data suggest that trauma to bridging veins is nonlinear, viscoelastic, and variable as a function of development, age, and the underlying brain atrophy. Mural thickness of bridging veins in the subdural compartment varies from 5 microns to 500 microns, compared to leptomeningeal vessels that have a minimum thickness of 50 microns and more dense collagen. Different studies have not surprisingly shown widely variable experimental results with respect to a number of parameters, including stress, strain, Young's modulus of elasticity, and outer diameter.
Although some biomechanical studies have shown that trivial trauma produces mechanical forces well below thresholds required for serious injury,,,,, others suggest the possibility of substantial acceleration with minor falls. Thompson et al., for example, noted a maximum angular acceleration of 11,730 radians/s 2 with simulated horizontal falls from 61 cm, with mediolateral impact. The authors suggest that risk for concussion, and a smaller, but still possible, risk for subdural hemorrhage, is evident in short falls onto hard surfaces. Ibrahim and Margules noted peak angular accelerations in the horizontal and sagittal planes exceeding 20,000 radians/s 2 and 40,000 radians/s 2, respectively, with drops of 3 ft of an 18-month toddler model. One foot drops onto concrete were comparable to concussive blows in football impacts, and reportedly “well above the concussive range.” Drops from one and two feet onto carpet pad were within the range of angular accelerations of the boxer blows that did not result in concussion, compared to drops onto concrete which were four times larger than nonconcussive boxer blows. The authors theorized that the restrictive head motion in drops, and lack of follow-through motion compared to sporting impacts, contributed to the high values from a seemingly trivial event. In support of their data, the authors point to a previous study (unpublished dissertation) in which the majority of toddlers who experienced low height falls had “altered mental status” including “lethargy, sluggishness, unexplained irritability, or loss of consciousness.” These findings have not been replicated in clinical studies on short falls.
Sullivan et al. noted in their model with a revised biofidelic neck that about 15% of falls onto hard surfaces from 61 to 91 cm exceeded the 10,000–15,000 radians/s 2 lower threshold of subdural hemorrhage. Importantly, the authors indicated that this was at variance with the clinical data. They also commented that their infant model with an improved biofidelic neck significantly decreased the observed kinematic loads below subdural hemorrhage thresholds with most short-distance drops, counter to earlier studies, but more consistent with actual short-fall data in small children.
Clinical data appear to be more realistic. Thompson et al., for example, report two SDHs in 79 falls in children <4 years old, both of which were contact subdural hemorrhages with a benign clinical course. The majority of short-fall studies in children do not report SDH at all. Accidental SDHs in young children, in contrast, are dominated by motor vehicle accidents, other high energy rotational events, and trauma of unusual severity., The rare short fall-associated subdural hemorrhages either have associated rotational forces (walkers falling downstairs, playground equipment) or otherwise features suggestive of high-energy trauma. One study that specifically looked at head-first falls from height described no SDHs. Given the marked discrepancy between the breaching of subdural thresholds with experimental models and the rarity of subdural hemorrhage caused by trivial events in vivo, biomechanical models appear to overestimate the risk. It should also be remembered that SDH has the highest morbidity and mortality among the traumatic brain injury pathologies (including epidural hematoma, cerebral contusion, traumatic intracerebral hematoma, and diffuse axonal injury). It stands to reason that trivial household falls would rarely, if ever, produce a clinically significant SDH.
The data from Kleiven and Huang et al. noted above may be more realistic in their distinction from published concussion thresholds. For example, the threshold for concussion scaled to infant brain mass (500 g) has been proposed to be between 10,000 and 15,000 radians/s 2 based on adult primate data , and instrumented football helmet data, respectively [Figure 1]., The fact that concussion in football is relatively common and subdural hemorrhage is rare suggests a substantially higher threshold for SDH in the athlete cohort in vivo, perhaps more in line with Kleiven and Huang. On the other hand, the small absolute size of the infant brain compared to the adult brain leads to a formulaic increase in thresholds for concussion, possibly closer to the subdural hemorrhage range in young children. These uncertainties are compounded by the lack of rotational acceleration data in young children compared to helmeted athletes, although even among the latter cohort, translational acceleration (g forces) are used as a surrogate for rotational acceleration and are unable to predict concussion with any degree of accuracy. The diagnosis of concussion or mild traumatic brain injury in pediatric patients has similar problem.
|Figure 1: The graph from Duhaime et al. shows angular acceleration versus angular velocity for shakes and impacts, with injury thresholds from primate experiments scaled to 500-g brain weight. DAI: Diffuse axonal injury, SDH: Subdural hematoma. Note the marked differences on acceleration achieved with and without impact (reproduced with permission from the Journal of Neurosurgery)|
Click here to view
Repetitive head trauma
Relevant to the biomechanics of SDH and lethality of abuse is the role of multiple traumatic events and the spacing of those traumatic events over time. It is well known, for example, that abusive head trauma is often repetitive, both over an extended period of time and within the confines of a given assault. Confessed perpetrators in one study admitted to up to thirty abusive assaults. Yet, surprisingly, few studies have examined repetitive trauma. Even among the studies attempting to quantify rotational forces with “shaking,” the endpoints were individual measurements with no consideration of potential biological effects of repetition.,,, The few studies that have looked at repetition tend to focus on outcomes related to axonal pathology, rather than SDH or other manifestations of parenchymal brain injury. Bonnier et al., for example, showed that repetitive shaking of 8-day-old mouse pups for 15 s produced white matter injury. Huh et al. demonstrated that repetitive, nonconcussive trauma in immature rats exacerbated axonal injury. Raghupathi showed that traumatic axonal injury was exacerbated with repetitive injury in the neonatal pig. Coats et al. recently showed that cyclic, low-velocity head rotations in immature piglets showed modest axonal injury that significantly increased with time postinjury (P< 0.035) and showed significantly greater amounts of red cell neuronal/axonal change and extra-axial hemorrhage compared to noncyclic head rotations 24 h postinjury.
An interesting study in primates was carried out in the late 1950s and early 1960s and published in a book in 1983. The experiments involved translational deceleration with a supported torso, unsupported head, and forward impact, as a means of simulating crash landing of fighter jets immediately after take-off from an air craft carrier. The animals in this study were subjected to progressively increasing translational forces starting from as low as 10 g to as high as 163 g. A threshold for fatality was identified at about 100 g in this construct, often with atlanto-occipital dislocation and spinal cord transection. SDH was not among the lethal injuries, and no change in threshold with repeated traumas was apparent. The importance of translational force in skeletal trauma, and the marked difference in neuropathology with acceleration of the torso in this study and acceleration of the head in other primate studies,,, was also noteworthy.
The putative “second-impact” phenomenon is occasionally invoked as a means of explaining malignant brain swelling with the second of two impacts over a period of hours or days. The data are too sparse, however, to permit any conclusions. At present, only a handful of cases have been described with a verified first impact, none have meet rigorous criteria for second-impact syndrome (McCrory 2001), all have occurred in young male athletes (mostly American football, followed by boxing, hockey [one case]), and downhill skiing [one case], and all but one in the United States. No cases of second-impact syndrome have been described to date on the entire continent of Australia. As such, the null hypothesis, that second-impact syndrome is simply catastrophic brain swelling from a single-impact event, still holds. The fact that the described cases often include SDH, and that SDH is present in about 90% of catastrophic head trauma cases in sports, suggests high magnitude rotational acceleration, optimized by impact site and direction of movement (see below), as the relevant factor, rather than repetitive trauma.
A recent study by Feldman et al. demonstrated a significantly higher Glasgow Coma Score in abused children with mixed acute and chronic subdural hemorrhages compared to acute subdural hemorrhages. Since radiographic characteristics of the associated acute bleeds were otherwise similar, this could suggest a raised threshold for parenchymal brain injury in children with repeated abuse, i.e., an adaptive phenomenon. Alternatively, it could mean that hemorrhage in the setting of a preexisting injury occurs with lesser applied force. Neither are conclusive. Accurately stated, no meaningful human data exist as to whether the thresholds for diffuse brain injury or subdural hemorrhage in young children increase, decrease, or remain the same with repetitive head trauma.
Influence of impact site and direction
The tendency for subdural hemorrhage varies with impact site, with occipital impacts being more likely to produce subdural hemorrhaging, and frontal impacts being less likely., This appears to be based on the specific anatomy of the bridging veins, i.e., the angle and course of veins between the brain surface and dura in different brain regions. In a cadaver study, the ruptured bridging veins were invariably located in the rolandic or postrolandic regions, veins that form angles with the superior sagittal sinus in the sagittal plane, which have been reported to vary between 40° and 50°. These findings are in close agreement with the calculations reported by Huang et al., with forward-draining bridging veins at angles of 50° incurring the greatest strain during occipital impact. Veins draining posteriorly at angles of 135° would theoretically be most vulnerable to frontal impacts; however, such veins do not exist in humans. Kleivin also noted that a substantially lower risk of subdural hemorrhage is predicted for frontal impact compared to the other directions. In a study by Aoki and Matsuzawa, occipital impacts predominated in a series of infantile acute SDH. Interestingly, in a 20-year review of cases from Auckland, New Zealand, the presence of occipital skull fracture was relatively more common in abuse cases compared to accident cases (P< 0.03), although there was no difference in fractures overall between the groups.
In addition to impact site and magnitude of rotation, rotation direction is another variable that may influence the extent of intracranial injuries. Sagittal plane rotation was identified as important for bridging vein rupture in classic primate studies, as noted above. Margulies and Thibault noted longer durations of unconsciousness, larger decreases in cerebral blood flow, more pronounced greater behavioral changes, and more persistent axonal injury following sagittal head rotation compared to other rotational directions.,, Hirakawa et al. noted that sagittal rotation predominated in a study of 309 adult cases of chronic SDH, consistent with the experimental data. The authors also found that nine out of 27 cases of SDH due to sport accidents were caused by judo, when the cases “fell down on their back.” On the other hand, coma and diffuse axonal injury in primates were related to coronal plane head rotations; neither were observed with sagittal plane rotation. Weaver et al. reported that head rotations in the horizontal plane resulted in the largest distribution of high strains, in a finite-element analysis. In one large European cohort of youth soccer players, concussion was most likely with temporal or occipital impact. It thus appears likely that direction of head rotation is among the biomechanical factors that influences outcome in traumatic brain injuries, and that this may differ significantly for SDH versus concussion/traumatic axonal injury.
Injury to the brain in subdural hematoma
Evaluated collectively, biomechanical studies indicate that acute SDH is secondary to violent rotational force applied to the head. Since rotation may produce shearing of the underlying brain, SDH in the clinical setting may also be a surrogate for parenchymal brain injury. In adult patients, outcome in SDH is determined more by injury to the underlying brain than mass effect from the expanding hematoma. In young children, the extent of brain damage with abusive head trauma and SDH is often devastating. Traumatic axonal injury may be one substrate for encephalopathy in a subset of abusive head trauma cases with subdural hemorrhage, although this is present in a minority of cases, possibly due to differential susceptibilities to rotational plane noted above. Secondary ischemic brain injury supervenes in many cases of abusive head trauma, for reasons that are not well worked out. The so-called “big black brain” phenomenon, with asymmetric edema and swelling that is not explained by ischemia-reperfusion or diffuse axonal injury, adds to the complexity of biomechanical trauma and brain injury in vivo. Each of these factors indicate that bridging vein thresholds inadequately address brain injury in the setting of acute SDH.
| Impact and Neck Injury|| |
The absence of impact injuries in a subset of abusive head trauma homicides and the customary absence of witnesses raise several questions: (1) What is the relationship between impact injury and intracranial injury? (2) Can lethal, inflicted rotational force be generated without impact? (3) Can inflicted rotational force cause lethal intracranial injuries without lethally injuring the neck?
Relationship between impact injury and intracranial injury
If the above basic biomechanical precepts hold, impact injuries (either by a moving object striking the decedent or the moving decedent striking an object) such as soft-tissue contusions and skull fractures occur by a fundamentally different mechanism than intracranial injury. Indeed, lack of a requirement for impact for SDH has long been recognized, both clinically and experimentally., Impact injuries instead relate not only to translational force, but to the size, shape, and consistency of the impacting object. A small, pointed, dense impacting object would more likely produce impact injury than a broad, flat, and soft impacting object. The likelihood of bridging vein rupture, however, would still depend on the magnitude of rotational force much more than the translational force. In addition, as noted above, it is axiomatic that impact on soft surfaces is associated with longer pulse durations, while longer pulse durations are associated with more severe injuries and a decreased injury threshold from rotation. The susceptibility of the intracranial compartment to rotational impact on soft surfaces may thus be underappreciated. By contrast, soft-tissue injury thresholds (“evidence of impact”) are undefined because the impact surface is the limiting factor. One is left with the likelihood that levels of rotational acceleration capable of tearing bridging veins and damaging axons are achievable with impact on soft surfaces that in turn dissipate the impact energy and leave little or no external evidence of impact.
The lack of relationship between impact injury and intracranial injury in abusive head trauma is consistent with this concept. Datta et al. characterized subdural hemorrhages by magnetic resonance imaging (MRI) in 49 cases of abusive head trauma and noted impact injuries (skull fractures or scalp swelling) in only 16 cases. Of the 16 cases with evidence of impact, subdural hemorrhages were both at and away from the impact site, while the majority of their cases had more than two sites of subdural hemorrhage, indicating no predictable relationship with impact injury. As noted above, bridging veins in the rolandic convexities are relatively vulnerable, while tentorial and falcine subdural hemorrhages are common in pediatric abusive and accidental head trauma in general, regardless of the presence or extent of impact injury.
Vinchon et al. looked specifically at scalp swelling as evidence of impact injury in their prospective study and found that evidence of impact correlated with accidental injuries (P< 0.001). Circularity was avoided in this study by including only witnessed accidents versus confessed abuse. This is consistent with an earlier study suggesting that absence of impact injuries by clinical examination is a marker of inflicted abuse. Distinction may be made between clinical studies and autopsy studies, as autopsy examination permits in-depth analysis of the layers of the scalp and the periosteum of the skull. Even in the case of autopsy examination, however, abusive injury patterns are found with regularity in the absence of physical evidence of impact  (i.e., shaken baby syndrome). Definitive analysis of fatal inflicted head injury versus fatal accident is not yet available (as regards impact), although impact injuries were evident in seven of ten fatal abusive head trauma cases in study by Vinchon et al., four of which had skull fractures. In an autopsy study of nonaccidental trauma by Geddes et al., 85% had signs of impact at autopsy (45/53), either scalp bruising or skull fracture, although the extent of bruising was not analyzed. All eight decedents without impact injuries were infants. Skull fractures were present in 19 out of 53 cases, and parietal and/or occipital fractures comprised 18 out of 19, again highlighting posterior impact events in the morbidity and mortality of blunt force trauma to the head.
In short, impact injuries and intracranial injuries occur by fundamentally different biomechanical mechanisms, as evidenced by the wide variability of impact injuries vis a vis stereotypical intracranial injuries. Autopsy examination often discloses evidence of impact where clinical evidence of impact was lacking, although impact injury is not invariable, even at autopsy. Moreover, there is no predictable relationship between the extent of impact injury to soft tissues and bone and the location or extent of subdural hemorrhaging, nor is the extent of parenchymal brain injury (edema, traumatic axonal injury) predicted by the presence or extent of impact injuries.
Can lethal, inflicted rotational force be generated without impact?
Among the more influential studies attempting to address the whiplash mechanism hypothesized by Guthkelch  and Caffey  was that of Duhaime et al. in 1987. In this study, the authors combined clinicopathological analysis of abusive head trauma patients with mathematical modeling of child abuse in Just Born dolls, with and without impact. The clinical arm of the study pointed out that evidence of impact at autopsy was often seen when there was no evidence of impact clinically. Among 13 fatal cases, all had evidence of impact, of which 7 showed evidence of impact only at autopsy. Their experimental model measured translational acceleration and scaled those data to angular acceleration extrapolated from primate studies, assuming a 500-g brain. Noteworthy was the finding that impact increased the translational acceleration measurements by nearly 50 fold. Impacts on a padded surface were also tested, showing less acceleration compared to a metal bar, but still 40 fold higher than the no-impact group. The authors hypothesized that shaking alone would likely not reach the biomechanical threshold for diffuse axonal injury or subdural hemorrhage and suggested a “shaken-impact” model for abusive head trauma.
The study by Duhaime et al. makes intuitive sense, although other research groups have tested this hypothesis using other surrogates and computer models, with varying results.,, Jenny et al. recently demonstrated that rotational accelerations in the range of subdural hemorrhage thresholds were possible with manual shaking alone. At present, mathematical “proof” of shaking versus shaken-impact theory seems to depend on one's choice of experimental construct. In either case, physicians still confront abusive injury patterns – subdural and retinal hemorrhages, seizures, apnea, collapse – with no evidence of impact, no natural disease explanation, and no trauma history. More research is needed.
Can inflicted rotational force cause lethal intracranial injuries without lethally injuring the neck?
The neck failure theory, proposed initially by Bandak, suggests the following: (i) that rotational forces required for intracranial injuries exceed the structural limit of the infant neck; (ii) that lethal cervical spine or brainstem injury could occur “at levels well below those reported for shaken baby syndrome;” and (iii) that the “shaken baby syndrome diagnosis in an infant with intracerebral injuries but without cervical spine or brainstem injury is questionable.” The theory was initially based on proposed rotational thresholds for intracranial injury and the proposed resulting distraction forces on the neck that would have exceeded structural neck tolerance, although the article has been criticized for overestimating the applied forces to the neck by more than a factor of 10. This hypothesis is also at variance with clinical data,, including voluntary admissions of shaking in some cases., In addition, an intensive education program about the dangers of shaking led to a substantial decrease in infantile traumatic brain injury in one study.
Perhaps more concerning, however, was the conclusion that physicians underdiagnose alternative explanations. This conclusion requires both that spinal cord injury is routinely missed, and that apnea causes the intracranial pathology. Neither is plausible. Neuropathological assessment of abusive head trauma cases including spinal cord examination indicates that damage to the spinal cord is uncommon. In a recent retrospective study of 183 children with head trauma (inflicted or accidental) specifically examined by spinal MRI out of concern for neck injury, ligamentous injuries were common (78% of the abuse group and 46% of the accidental trauma group), but spinal cord injury was not seen. The assertion that apnea from spinal cord trauma causes intracranial pathology is also untenable, since uncomplicated hypoxia does not lead to neuronal necrosis, cytotoxic edema, or hemorrhage.,
On the other hand, Matshes et al. noted unilateral or bilateral hemorrhages in the nerve roots of C3, C4, and C5 of those infants suspected of having hyperflexion/hyperextension injuries. Given the relative difficulty in accessing the nerve roots of the upper cervical spine at autopsy, the authors suggest that such hemorrhages are more common than generally appreciated, and that anoxic encephalopathy secondary to neck injury plays a central role, somewhat in keeping with Bandak's hypothesis. Among the unanswered questions are the relationship, if any, between nerve root hemorrhage and functional disruption of both phrenic nerves, and whether nerve root hemorrhage is a surrogate for lethal neck trauma. Again problematic is that gross structural injury to the cervical spinal cord is uncommon in homicidal abusive head trauma. The issue of hypoxia as a driver of cerebral edema and subdural hemorrhage is again not valid as noted above.
The relative frequency of the so-called spinal cord injury without radiographic abnormalities in infants and young children  raises the possibility that parenchymal injury to the spinal cord may occur in the absence of bone or soft-tissue injuries. Whether more rigorous characterization of the lower brainstem and cervical spinal cord will increase the frequency of direct injury detection in these regions remains to be determined. Also of interest is the age-related cervical spine injury pattern in pediatric patients, in which the level of cervical injury descends with increasing age., The infantile upper cervical spine, and in particular C1 and C2, is clinically more vulnerable to injury compared to lower segments. This is consistent with the findings by Choudhary et al., but at variance with the Matshes et al's. study showing nerve root hemorrhages in infants limited to C3–C5.
Neck injury has proven difficult to model. Preliminary neck injury criteria have been developed for use in the automotive industry to assess the injury risk of frontal impact collisions, with no demonstrated relevance to abusive head trauma. Thompson et al. used these criteria in bed fall simulation experiments and found that short falls onto hard surfaces were associated with a 22% risk of serious neck injuries. This finding is once again contrary to the clinical reality, since serious neck injury has never been reported from bed falls or other short-distance furniture falls. In an infant lamb model of whiplash (considered similar to the infant because of large head and weak neck), widespread axonal injury, including brainstem and spinal cord, was noted, providing some experimental support for brain injury without neck failure. Finally, it should be kept in mind that models of neck trauma caused by shaking require a chest-only mechanism. The action sequences in abusive homicide are obviously more complex and heterogeneous. The frequency and specificity of neck bruises in physical abuse, for example, raise the possibility of neck support during the assault (e.g., rotating the head and torso while holding the neck), which has not been modeled to date. Rotation of the head, neck, and torso with impact on soft surfaces, i.e., a biomechanical construct that would dissipate translational force on soft tissue and bone, but leave intact rotational force on the intracranial compartment, has not been studied.
| Conclusions|| |
Limited knowledge and exceedingly varied parameters as they pertain to injury biomechanics and lethality permit only general conclusions. As regards mechanical forces applied to the head, the magnitude of rotational force appears to be relevant in producing the inflicted head injury pattern. Experimental thresholds for individual traumatic lesions vary widely, however. Noteworthy is that translational forces do not, for practical purposes, cause bridging vein rupture, which may explain why falls, including falls from height, rarely cause clinically significant subdural hemorrhage in young children. The collective data indicate that in vivo thresholds for various intracranial injuries in general, and lethality in particular, are impossible to quantitate. The effect on injury thresholds of multiple rotational events over multiple time frames, either as punctuated assaults over a protracted period or within the context of a single assault, is entirely unknown. The role of impact in abusive head trauma is unresolved, and there is no predictable relationship between impact injuries, when present, and intracranial injuries in abusive head trauma.
Any conclusions with respect to injury biomechanics of the infant neck are also preliminary. Raw data from experimental models show an overestimation of the in vivo risk, similar to biomechanical studies of bridging vein rupture. The hypothesis that the lethal distraction forces to the cervical spine are achieved by rotation of the infant chest below thresholds required for intracranial injury may be debated as a matter of theory, although the required corollary that apnea causes the intracranial pathology is not plausible or credible. Clinical studies as well as recent studies in improved biofidelic models also refute the premise that shaking is benign up to the point of neck injury. On the other hand, spinal cord injury from homicidal abuse can be both difficult to identify and difficult to exclude. More research is needed.
In short, biomechanical models are not designed to “prove” whether a proposed lethal mechanism does or does not exist in vivo. They simply explore hypotheses. While case reports emphasizing household accidents and putative mimics of abuse will continue to appear, as will refinements of biomechanical models that more closely parallel in vivo biology, clinical studies at present supersede injury biomechanics and argue against trivial accidental injury or natural disease as plausible alternatives to the abusive head trauma pattern. Extreme caution, if not skepticism, in the role of biomechanical engineering in the reconstruction of unwitnessed and complex injury scenarios, is warranted.
Financial support and sponsorship
Conflicts of interest
There are no conflicts of interest.
| References|| |
Sullivan S, Coats B, Margulies SS. Biofidelic neck influences head kinematics of parietal and occipital impacts following short falls in infants. Accid Anal Prev 2015;82:143-53.
Thompson A, Bertocci G, Pierce MC. Assessment of injury potential in pediatric bed fall experiments using an anthropomorphic test device. Accid Anal Prev 2013;50:16-24.
Weber W. Experimental studies of skull fractures in infants. Z Rechtsmed 1984;92:87-94.
Mohan D, Bowman BM, Synder RG, Foust DR. A biomechanical analysis of head impact injuries to children. J Biomech Eng 1979;101:250-60.
Chadwick DL, Bertocci G, Castillo E, Frasier L, Guenther E, Hansen K, et al.
Annual risk of death resulting from short falls among young children: Less than 1 in 1 million. Pediatrics 2008;121:1213-24.
Warrington SA, Wright CM; ALSPAC Study Team. Accidents and resulting injuries in premobile infants: Data from the ALSPAC study. Arch Dis Child 2001;85:104-7.
Feldman K. Commentary on “congenital rickets” article. Pediatr Radiol 2009;39:1127-9.
Vinchon M, de Foort-Dhellemmes S, Desurmont M, Delestret I. Confessed abuse versus witnessed accidents in infants: Comparison of clinical, radiological, and ophthalmological data in corroborated cases. Childs Nerv Syst 2010;26:637-45.
Holbourn AH. Mechanics of head injuries. Lancet 1943;242:438-41.
Dawson SL, Hirsch CS, Lucas FV, Sebek BA. The contrecoup phenomenon. Reappraisal of a classic problem. Hum Pathol 1980;11:155-66.
Adams JH, Graham DI, Murray LS, Scott G. Diffuse axonal injury due to nonmissile head injury in humans: An analysis of 45 cases. Ann Neurol 1982;12:557-63.
Gennarelli TA, Adams JH, Graham DI. Acceleration induced head injury in the monkey.I. The model, its mechanical and physiological correlates. Acta Neuropathol Suppl 1981;7:23-5.
Gennarelli TA, Thibault LE. Biomechanics of acute subdural hematoma. J Trauma 1982;22:680-6.
Lee MC, Haut RC. Insensitivity of tensile failure properties of human bridging veins to strain rate: Implications in biomechanics of subdural hematoma. J Biomech 1989;22:537-42.
Forbes JA, Zuckerman S, Abla AA, Mocco J, Bode K, Eads T, et al.
Biomechanics of subdural hemorrhage in American football: Review of the literature in response to rise in incidence. Childs Nerv Syst 2014;30:197-203.
Hymel KP, Jenny C, Block RW. Intracranial hemorrhage and rebleeding in suspected victims of abusive head trauma: Addressing the forensic controversies. Child Maltreat 2002;7:329-48.
Wilberger JE Jr., Harris M, Diamond DL. Acute subdural hematoma: Morbidity, mortality, and operative timing. J Neurosurg 1991;74:212-8.
Geddes JF, Hackshaw AK, Vowles GH, Nickols CD, Whitwell HL. Neuropathology of inflicted head injury in children. I. Patterns of brain damage. Brain 2001;124:1290-8.
Ommaya AK, Yarnell P. Subdural haematoma after whiplash injury. Lancet 1969;2:237-9.
Gennarelli TA, Thibault LE, Adams JH, Graham DI, Thompson CJ, Marcincin RP, et al.
Diffuse axonal injury and traumatic coma in the primate. Ann Neurol 1982;12:564-74.
Huang HM, Lee MC, Chiu WT, Chen CT, Lee SY. Three-dimensional finite element analysis of subdural hematoma. J Trauma 1999;47:538-44.
Gadd CW. Use of a Weighted Impulse Criterion for Estimating Injury Hazard. Proceedings 10th
Stapp Car Crash Conference SAE Paper No. 660793. SAE Paper: 164-174; 1966.
Marjoux D, Baumgartner D, Deck C, Willinger R. Head injury prediction capability of the HIC, HIP, SIMon and ULP criteria. Accid Anal Prev 2008;40:1135-48.
Bonfield CM, Naran S, Adetayo OA, Pollack IF, Losee JE. Pediatric skull fractures: The need for surgical intervention, characteristics, complications, and outcomes. J Neurosurg Pediatr 2014;14:205-11.
Thompson A, Bertocci G. Pediatric bed fall computer simulation model: Parametric sensitivity analysis. Med Eng Phys 2014;36:110-8.
Cory CZ, Jones MD, James DS, Leadbeatter S, Nokes LD. The potential and limitations of utilising head impact injury models to assess the likelihood of significant head injury in infants after a fall. Forensic Sci Int 2001;123:89-106.
Depreitere B, Van Lierde C, Sloten JV, Van Audekercke R, Van der Perre G, Plets C, et al.
Mechanics of acute subdural hematomas resulting from bridging vein rupture. J Neurosurg 2006;104:950-6.
Prange MT, Coats B, Duhaime AC, Margulies SS. Anthropomorphic simulations of falls, shakes, and inflicted impacts in infants. J Neurosurg 2003;99:143-50.
Löwenhielm P. Strain tolerance of the Vv. Cerebri sup. (Bridging Veins) calculated from head-on collision tests with cadavers. Z Rechtsmed 1974;75:131-44.
Ibrahim NG, Margulies SS. Biomechanics of the toddler head during low-height falls: An anthropomorphic dummy analysis. J Neurosurg Pediatr 2010;6:57-68.
Kleiven S. Influence of impact direction on the human head in prediction of subdural hematoma. J Neurotrauma 2003;20:365-79.
Newman JA, Shewchenko N, Welbourne E. A proposed new biomechanical head injury assessment function – The maximum power index. Stapp Car Crash J 2000;44:215-47.
Famaey N, Ying Cui Z, Umuhire Musigazi G, Ivens J, Depreitere B, Verbeken E, et al.
Structural and mechanical characterisation of bridging veins: A review. J Mech Behav Biomed Mater 2015;41:222-40.
Yamashima T, Friede RL. Why do bridging veins rupture into the virtual subdural space? J Neurol Neurosurg Psychiatry 1984;47:121-7.
Jones MD, James DS, Cory CZ, Leadbeatter S, Nokes LD. Subdural haemorrhage sustained in a baby-rocker? A biomechanical approach to causation. Forensic Sci Int 2003;131:14-21.
Jenny C, Bertocci G, Fukuda T, Rangarajan N, Shams T. Biomechanical response of the infant head to shaking – An experimental investigation. J Neurotrauma 2017;34:1579-88.
Bertocci GE, Pierce MC, Deemer E, Aguel F, Janosky JE, Vogeley E, et al.
Influence of fall height and impact surface on biomechanics of feet-first free falls in children. Injury 2004;35:417-24.
Thompson AK, Bertocci G, Rice W, Pierce MC. Pediatric short-distance household falls: Biomechanics and associated injury severity. Accid Anal Prev 2011;43:143-50.
Feldman KW, Bethel R, Shugerman RP, Grossman DC, Grady MS, Ellenbogen RG, et al.
The cause of infant and toddler subdural hemorrhage: A prospective study. Pediatrics 2001;108:636-46.
Duhaime AC, Alario AJ, Lewander WJ, Schut L, Sutton LN, Seidl TS, et al.
Head injury in very young children: Mechanisms, injury types, and ophthalmologic findings in 100 hospitalized patients younger than 2 years of age. Pediatrics 1992;90:179-85.
Ewing-Cobbs L, Kramer L, Prasad M, Canales DN, Louis PT, Fletcher JM, et al.
Neuroimaging, physical, and developmental findings after inflicted and noninflicted traumatic brain injury in young children. Pediatrics 1998;102:300-7.
Snyder RG, Foust DR, Bowman B. Study of Impact Tolerance Through Free Fall Investigation. University of Michigan Safety Research Institute. Final Report; 1977.
Gennarelli TA, Spielman GM, Langfitt TW, Gildenberg PL, Harrington T, Jane JA, et al.
Influence of the type of intracranial lesion on outcome from severe head injury. J Neurosurg 1982;56:26-32.
Duhaime AC, Gennarelli TA, Thibault LE, Bruce DA, Margulies SS, Wiser R, et al.
The shaken baby syndrome. A clinical, pathological, and biomechanical study. J Neurosurg 1987;66:409-15.
Ommaya AK, Goldsmith W, Thibault L. Biomechanics and neuropathology of adult and paediatric head injury. Br J Neurosurg 2002;16:220-42.
Broglio SP, Schnebel B, Sosnoff JJ, Shin S, Fend X, He X, et al.
Biomechanical properties of concussions in high school football. Med Sci Sports Exerc 2010;42:2064-71.
Rowson S, Brolinson G, Goforth M, Dietter D, Duma S. Linear and angular head acceleration measurements in collegiate football. J Biomech Eng 2009;131:061016.
Lieh-Lai MW, Theodorou AA, Sarnaik AP, Meert KL, Moylan PM, Canady AI, et al.
Limitations of the Glasgow Coma Scale in predicting outcome in children with traumatic brain injury. J Pediatr 1992;120:195-9.
Adamsbaum C, Grabar S, Mejean N, Rey-Salmon C. Abusive head trauma: Judicial admissions highlight violent and repetitive shaking. Pediatrics 2010;126:546-55.
Bandak FA. Shaken baby syndrome: A biomechanics analysis of injury mechanisms. Forensic Sci Int 2005;151:71-9.
Bonnier C, Mesplès B, Carpentier S, Henin D, Gressens P. Delayed white matter injury in a murine model of shaken baby syndrome. Brain Pathol 2002;12:320-8.
Huh JW, Widing AG, Raghupathi R. Basic science; repetitive mild non-contusive brain trauma in immature rats exacerbates traumatic axonal injury and axonal calpain activation: A preliminary report. J Neurotrauma 2007;24:15-27.
Raghupathi R, Margulies SS. Traumatic axonal injury after closed head injury in the neonatal pig. J Neurotrauma 2002;19:843-53.
Coats B, Binenbaum G, Smith C, Peiffer RL, Christian CW, Duhaime AC, et al.
Cyclic head rotations produce modest brain injury in infant piglets. J Neurotrauma 2017;34:235-47.
Unterharnscheidt F. Neuropathology of rhesus monkeys undergoing G impact acceleration. In: Ewing CL, Thomas DJ, Sances A, Larson SJ, editors. Impact Injury of the Head and Spine. Springfield, IL: Charles C Thomas; 1983. p. 94-176.
Adams JH, Graham DI, Gennarelli TA. Acceleration induced head injury in the monkey. II. Neuropathology. Acta Neuropathol Suppl 1981;7:26-8.
McCrory P. Does second impact syndrome exist? Clin J Sport Med 2001;11:144-9.
Feldman KW, Sugar NF, Browd SR. Initial clinical presentation of children with acute and chronic versus acute subdural hemorrhage resulting from abusive head trauma. J Neurosurg Pediatr 2015;16:177-85.
Aoki N, Masuzawa H. Infantile acute subdural hematoma. Clinical analysis of 26 cases. J Neurosurg 1984;61:273-80.
John SM, Jones P, Kelly P, Vincent A. Fatal pediatric head injuries: A 20-year review of cases through the Auckland coroner's office. Am J Forensic Med Pathol 2013;34:277-82.
Margulies SS, Thibault LE. A proposed tolerance criterion for diffuse axonal injury in man. J Biomech 1992;25:917-23.
Eucker SA, Smith C, Ralston J, Friess SH, Margulies SS. Physiological and histopathological responses following closed rotational head injury depend on direction of head motion. Exp Neurol 2011;227:79-88.
Sullivan S, Friess SH, Ralston J, Smith C, Propert KJ, Rapp PE, et al.
Behavioral deficits and axonal injury persistence after rotational head injury are direction dependent. J Neurotrauma 2013;30:538-45.
Hirakawa K, Hashizume K, Fuchinoue T, Takahashi H, Nomura K. Statistical analysis of chronic subdural hematoma in 309 adult cases. Neurol Med Chir (Tokyo) 1972;12:71-83.
Weaver AA, Danelson KA, Stitzel JD. Modeling brain injury response for rotational velocities of varying directions and magnitudes. Ann Biomed Eng 2012;40:2005-18.
Case ME, Graham MA, Handy TC, Jentzen JM, Monteleone JA; National Association of Medical Examiners Ad Hoc Committee on Shaken Baby Syndrome. Position paper on fatal abusive head injuries in infants and young children. Am J Forensic Med Pathol 2001;22:112-22.
Duhaime AC, Christian C, Moss E, Seidl T. Long-term outcome in infants with the shaking-impact syndrome. Pediatr Neurosurg 1996;24:292-8.
Ichord RN, Naim M, Pollock AN, Nance ML, Margulies SS, Christian CW, et al.
Hypoxic-ischemic injury complicates inflicted and accidental traumatic brain injury in young children: The role of diffusion-weighted imaging. J Neurotrauma 2007;24:106-18.
Duhaime AC, Durham S. Traumatic brain injury in infants: The phenomenon of subdural hemorrhage with hemispheric hypodensity (“Big black brain”). Prog Brain Res 2007;161:293-302.
Ommaya AK, Faas F, Yarnell P. Whiplash injury and brain damage: An experimental study. JAMA 1968;204:285-9.
Datta S, Stoodley N, Jayawant S, Renowden S, Kemp A. Neuroradiological aspects of subdural haemorrhages. Arch Dis Child 2005;90:947-51.
Bradford R, Choudhary AK, Dias MS. Serial neuroimaging in infants with abusive head trauma: Timing abusive injuries. J Neurosurg Pediatr 2013;12:110-9.
Vinchon M, Defoort-Dhellemmes S, Desurmont M, Dhellemmes P. Accidental and nonaccidental head injuries in infants: A prospective study. J Neurosurg 2005;102:380-4.
Gill JR, Goldfeder LB, Armbrustmacher V, Coleman A, Mena H, Hirsch CS, et al.
Fatal head injury in children younger than 2 years in New York city and an overview of the shaken baby syndrome. Arch Pathol Lab Med 2009;133:619-27.
Guthkelch AN. Infantile subdural haematoma and its relationship to whiplash injuries. Br Med J 1971;2:430-1.
Caffey J. The whiplash shaken infant syndrome: Manual shaking by the extremities with whiplash-induced intracranial and intraocular bleedings, linked with residual permanent brain damage and mental retardation. Pediatrics 1974;54:396-403.
Cory CZ, Jones BM. Can shaking alone cause fatal brain injury? A biomechanical assessment of the Duhaime shaken baby syndrome model. Med Sci Law 2003;43:317-33.
Wolfson DR, McNally DS, Clifford MJ, Vloeberghs M. Rigid-body modelling of shaken baby syndrome. Proc Inst Mech Eng H 2005;219:63-70.
Rangarajan N, Shams T. Re: Shaken baby syndrome: A biomechanics analysis of injury mechanisms. Forensic Sci Int 2006;164:280-1.
Levin AV, Magnusson MR, Rafto SE, Zimmerman RA. Shaken baby syndrome diagnosed by magnetic resonance imaging. Pediatr Emerg Care 1989;5:181-6.
American Academy of Pediatrics: Committee on Child Abuse and Neglect. Shaken baby syndrome: Rotational cranial injuries-technical report. Pediatrics 2001;108:206-10.
Starling SP, Patel S, Burke BL, Sirotnak AP, Stronks S, Rosquist P, et al.
Analysis of perpetrator admissions to inflicted traumatic brain injury in children. Arch Pediatr Adolesc Med 2004;158:454-8.
Dias MS, Smith K, DeGuehery K, Mazur P, Li V, Shaffer ML, et al.
Preventing abusive head trauma among infants and young children: A hospital-based, parent education program. Pediatrics 2005;115:e470-7.
Choudhary AK, Ishak R, Zacharia TT, Dias MS. Imaging of spinal injury in abusive head trauma: A retrospective study. Pediatr Radiol 2014;44:1130-40.
Levine S. Anoxic-ischemic encephalopathy in rats. Am J Pathol 1960;36:1-7.
Auer RN, Dunn JF, Sutherland G. Hypoxia and related conditions. In: Love S, Louis DN, Ellison D, editors. Greenfield's Neuropathology. 8th
ed. London: Edward Arnold Ltd.; 2008. p. 63-119.
Matshes EW, Evans RM, Pinckard JK, Joseph JT, Lew EO. Shaken infants die of neck trauma, not of brain trauma. Acad Forensic Pathol 2011;1:82-91.
Rozzelle CJ, Aarabi B, Dhall SS, Gelb DE, Hurlbert RJ, Ryken TC, et al.
Spinal cord injury without radiographic abnormality (SCIWORA). Neurosurgery 2013;72 Suppl 2:227-33.
Baumann F, Ernstberger T, Neumann C, Nerlich M, Schroeder GD, Vaccaro AR, et al.
Pediatric cervical spine injuries: A rare but challenging entity. J Spinal Disord Tech 2015;28:E377-84.
Knox JB, Schneider JE, Cage JM, Wimberly RL, Riccio AI. Spine trauma in very young children: A retrospective study of 206 patients presenting to a level 1 pediatric trauma center. J Pediatr Orthop 2014;34:698-702.
Finnie JW, Blumbergs PC, Manavis J, Turner RJ, Helps S, Vink R, et al.
Neuropathological changes in a lamb model of non-accidental head injury (the Shaken baby syndrome). J Clin Neurosci 2012;19:1159-64.
Pierce MC, Kaczor K, Aldridge S, O'Flynn J, Lorenz DJ. Bruising characteristics discriminating physical child abuse from accidental trauma. Pediatrics 2010;125:67-74.
[Table 1], [Table 2]