Challenges to the Biomechanical Understanding of Pediatric Head Trauma

John  D. Lloyd, Ph.D., M.Erg.S., CPE, CBIS
Board Certified Ergonomist & Certified Brain Injury Specialist

A number of child-abuse professionals have rejected decades of biomechanical research that casts doubt on the classic model of shaken baby syndrome. Some have attempted to conduct their own studies. The results have never been published but have been offered as evidence in court. This page explains why some of the reasons her conclusions are scientifically invalid.

It has been argued that the neck of the Child Restraint Airbag Interaction (CRABI) biofidelic mannequins is too stiff and therefore not representative of an infant. To address this alleged problem, biomechanic data were collected over a series of trials in which one subject shook the Aprica 2.5 and Aprica 3.4 mannequins, developed at the Aprica Child Care Institute in Japan, where they’re used to evaluate the biomechanics of Japanese infants in car seats and strollers.

Results were presented at the Eleventh International Conference on Abusive Head Trauma/Shaken Baby Syndrome, in Atlanta, Georgia in September of 2010, and in a recorded Grand Rounds lecture at Rhode Island Hospital in July of 2010. The latter Grand Rounds talk, titled “Junk Medical Science in the Courtroom,” can be viewed at http://lifespan.mediasite.com/mediasite/Viewer/?peid=d237bed531df42e49223ccdb685c48741d

First, the Aprica 2.5 and Aprica 3.4 mannequins weigh 2.5 kg (5.5 lb) and 3.4 kg (7.5 lb), respectively. Even the heavier model weighs slightly less than an average American infant at birth, and significantly less than the older infants typically alleged to be victims of abusive shaking.

It is claimed that the issues of biofidelity of the CRABI neckform are exemplified by below, in which the mean and maximum angular displacement of the Aprica head/neck in the sagittal plane were 221 and 248 degrees, respectively.

In his own laboratory, Dr. Lloyd‘s subjects have generated a total sagittal angular displacement of 120.8 degrees during vigorous shaking of the CRABI mannequin. This figure is 17 percent less than the radiographically measured passive range of motion of a real infant’s neck, 145 degrees, as illustrated in Figure 1. As the left-hand image illustrates, flexion is limited by the chin hitting the chest and extension is limited by the internal structure of the neck.

Figure 1 – Maximum Extensions Measured in Neck of Infant Cadaver

While Dr. Lloyd’s figure falls short of the maximum possible range as measured in an infant decedent, his results are much closer to the measured maximum than is the maximum measured total for the Aprica, which exceeds the true infant range by 71%. Furthermore, it has been suggested that the majority of infant head rotational motion occurs at the more flexible atlanto-occipital (AO) joint, but these images of an infant decedent show the majority of rotational motion lower in the neck and upper thoracic spine, centering around the C7-T1 joint, which is comparable to design of the CRABI neckform.

Over a series of trials, one subject achieved much higher levels of peak angular acceleration with the Aprica than were measured in the original Duhaime research, These results suggest the probability of a grave mathematical error, as I proffer that such results grossly defy laws of physics and biomechanics.

Furthermore, if we consider Newton's 3rd law – “Actioni contrariam semper et æqualem esse reactionem”, that is “every force has an equal and opposite force”  – forces exerted by the caregiver would be equal and opposite to forces exerted on the infant. It is further reasoned that such forces are a function of the caregiver’s strength, not the mass of the infant. Hence, forces exerted on an APRICA 2.5, APRICA 3.4, CRABI 6, CRABI 12, and even NCSBS demonstration doll would be similar, ignoring physical limitations of the models. According to Newton’s 2nd law “Mutationem motus proportionalem esse vi motrici impressae, et fieri secundum lineam rectam qua vis illa imprimatur”, the common representation of which is: Force equals mass times acceleration (F=ma). Therefore, if force is a function of caregiver’s strength and would be similar between infant surrogates, which vary substantially by mass, then increases in angular accelerations would be inversely proportional to model mass. It would be reasonable to produce higher angular accelerations with smaller surrogates (e.g. APRICA 2.5 = 5.5lb) vs. larger infant representatives (e.g. CRABI12 = 22lb).

The CRABI biofidelic mannequin, which I have used in my studies, was developed to evaluate small child restraint systems in automotive crash environments. Basic anthropometry for this biofidelic mannequin was taken from the University of Michigan Transportation Research Institute Report 85-23. The Society of Automotive (SAE) Infant Dummy Task Group approved weight distribution and scaling methods for the infant mannequin, which meets anthropometry, biomechanical and instrumentation design requirements of SAE.

The CRABI biofidelic mannequin is frequently used by biomechanists to investigate non motor-vehicle impact scenarios. Dr. Van Ee et al utilized a CRABI biofidelic mannequin and electronics instrumentation (in compliance with SAE J211) to investigate a fatal fall of a 23-month-old child. Their experimental methodology and findings was published in the Proceedings of the ASME 2009 International Mechanical Engineering Congress.