Using Animal Models to Understand Traumatic Brain Injury

Traumatic brain injury (TBI) is the main cause of death in children and young adults living in the industrialized world.  Current estimates suggest that in the U.S. between 2.5 million and 6.5 million individuals are living with the consequences of TBI, much of it caused by motor vehicle accidents and for which no truly efficacious and approved therapies are currently available (Fig. 1) [1]. 

Figure 1: Incidence of selected health problem in the United States. An estimated 1.5 million Americans sustain a TBI each year, which is eight times the number of people diagnosed with breast cancer and 34 times the number of new cases of HIV or AIDS, according to the Centres for Disease Control and Prevention. (Adapted from the American Society for Neuroscience).

The pathogenesis of TBI is incompletely understood, in large part because patients often present with a complexity of lesions of varying severity and regional distribution.  Brain injuries are broadly divided into two groups; focal and diffuse. Focal injuries are characterized by contusions (i.e. damage at the site of the blow) and lacerations, often accompanied by hematoma. In contrast, diffuse injuries involve brain swelling, ischemic brain damage and diffuse axonal injury (DAI) observed in the direct vicinity and also remote from the site of impact. Research conducted on animal models of brain trauma over the past decades have generated an abundant amount of data that has helped gain an insight into the events that occur during and after injury. In this review we focus on the main models that cause TBI by applying mechanical energy to the head, skull or dura of the animal [2].

No single animal model is entirely successful in reproducing the complete spectrum of pathological changes observed after TBI, and further research is necessary to fully elucidate the acute and chronic changes that occur after TBI. These studies will be directed at clarifying and validating the present concepts, establishing new therapeutic strategies and extending the pharmacological approach towards a more effective treatment for head-injured patients. The in vivo animal model however, remains necessary to prove new concepts and make clinical trials successful and safe.

Ethical and Theoretical Considerations

Ethically responsible use of experimental animals should be of the utmost importance in the development of animal models of TBI. The research scientist must explain how the results may benefit human clinical practice. As few animals as possible should be used and special attention should be paid to minimizing pain or discomfort. It is clear that animals that endure pain or distress may provide erroneous data, particularly where subtle changes in neurotransmitter release are studied.

What do Animal Models Tell Us?

Although it may be impossible to model the entire complex symptomatic spectrum of TBI in an animal, selected symptoms may be mimicked and have some validity. It is helpful if an animal model of TBI be discussed in terms of its face, construct, aetiological and construct validity. These terms are now discussed.

  1. Face validity refers to the phenomenological similarity between the behavior exhibited by the animal and the specific conditions of the human condition, however, it is unrealistic to expect similar behaviors in rodents and humans, and the aim is to search for relevant equivalents based upon the brain regions assumed to be involved.
  2. Construct validity refers to similarity in underlying mechanisms even though the precise expression of behaviors may be different between experimental animals and humans.
  3. Aetiological validity is an extension of construct validity and refers to the degree of similarity of aetiology between the changes seen in the experimental animals and those observed in the human. TBI is complex, and animal models can be used to test hypotheses about aetiology.
  4. Predictive validity and reliability refer respectively to (a) the predictive value that observations made in animals will have for the human condition (b) to the accuracy with which both the experimental and clinical observations are made. Both predictive validity and reliability are important for assay models such as those used in the development of new therapeutics. Behavioral TBI models should at minimum have predictive validity and be reliable.  

Animal Models of Traumatic Brain Injury

Figure 2: An idealised representation demonstrating a focal brain injury resulting from an impact. A subdural haematoma is formed as a result of vascular disruption. Adapted from [3].

Animal models of TBI play an important role in the process of evaluating and understanding the complex physiologic, behavioural and histopathologic changes associated with TBI. Since human TBI is very much a heterogenous disease, no single animal model of TBI can mimic the whole spectrum of findings observed in human TBI. Rather, the concurrent use of several distinct yet complementary models (e.g. in vivo microdialysis in the rat combined with finite element modelling, is necessary to characterise the features observed upon clinical and post-mortem examination of TBI in patients and to develop novel diagnostic and therapeutic strategies for the treatment of TBI [5,6,7,8].

A number of clinical and animal studies over the past decade have now established that the principal mechanism of brain damage after head injury is due to either direct contact (i.e. impact) or acceleration/deceleration types of injury [9]. Lesions due to direct impact result from an object striking the head. In contrast, acceleration/deceleration brain injury results from unrestricted head movement in the instant after injury. It is generally agreed that the focal pathologies associated with an impact are more likely to be sustained as a result of a fall, whereas the diffuse pathologies are most commonly associated with acceleration/deceleration in motor vehicle accidents.

Figure 3: An idealised representation demonstrating diffuse brain injury resulting from an inertial force. Rapid rotational acceleration-deceleration of the head in the coronal plane (yellow arrow) results in deformation of the entire brain. The falx membrane along the sagittal midline acts as a barrier to lateral brain motion (blue arrow) creating high strain between the hemispheres. This overall mechanical deformation results in diffuse axonal injury (DAI) with prominent axonal pathology in midline structures.Adapted from [3].

Two mechanical phenomena constitute the most common causes of primary brain injury. These are (a) local skull distortion due to direct contact and propagation of stress waves through the brain from the point of impact (Fig.2 and Fig. 3) and (b) movement and distortion of brain material due to inertial or acceleration loading. Both of these phenomena occur when the head is struck by a rigid or padded object (i.e. impact loading) (Fig.4). However, only inertial effects are present when the head moves indirectly as a result of impact to another region of the body (i.e. impulsive loading). In direct head impact there is a local bending of the skull, underlying tissue strain and gross movement of the brain tissue. Impulsive loading on the other had does not create local contact effects, but rather produces a non-uniform distribution of pressure and tissue strain. Both phenomena can result in significant primary tissue injury.

Figure 4: Mechanical loading characteristics such as the magnitude of impact force and rate of head acceleration strongly contribute to the type and severity of brain injury. Contact forces create focal lesions such as skull fracture, haematomas and cerebral contusions. Acceleration effects can cause diffuse lesions such as diffuse axonal injury (DAI) as well as numerous focal injuries.

For the purposes of clarity, experimental models of TBI are defined under the following categories: direct brain deformation, impact acceleration and inertial acceleration and these are now discussed.

Experimental models that use gases, liquids and rigid-bodies as impacting media are divided between those which apply direct impact (invasively) to the brain (i.e. direct brain deformation) via a craniotomy and those that cause acceleration of the intracranial contents by impacting the exterior of the head itself (i.e. impact acceleration). Furthermore, inertial models that allow translational and/or rotational acceleration by indirect impact (i.e. inertial acceleration) have also been designed.

Experimental models can be graded on their behavioural, pathologic, physiologic and biochemical fidelity to that observed in human brain injury. To closely mimic the whole range of TBI observed in the clinical situation, an experimental model must also be capable of delivering variable degrees of brain trauma by adjusting the main mechanical parameters of the impact device (e.g. the height or mass of a free-falling weight, the depth of the traumatic impact or impact velocity, the height of the pressure impulse by 
adjusting the pendulum of the fluid percussion device or changes in the plane or velocity of the rotational forces).

Concluding Remarks and Outlook

Head injury is a spontaneous, unpredictable event and no single animal model is entirely successful in reproducing the complete spectrum of pathological changes observed after TBI in humans. A major requirement in choosing a model of TBI is to develop a system that will provide a high-level of useful and accurate impact data.In this regard, it is also imperative that an impact be applied in a repeatable, precise and quantifiable manner. Unprotected direct impact (i.e. when impact is delivered directly to the skull) in the rat results in either no change, or dramatic changes, due to the extremely steep injury tolerance curve of the rat skull. The occurrence of skull fracture from impact is not well correlated with injury in the rat model [10] and direct cranial impact methods have a high degree of variability in the response such that even a slight change in the impact parameters may change the injury outcome from minor to fatal [11]. The relative utility of models for duplicating human TBI conditions is shown in Table 1.  These animal models of TBI may be useful in the investigation of new therapeutic strategies and pharmacological testing for an effective treatment for head-injured patients.

Relative Utility of Models for Duplicating  Human TRAUMATIC BRAIN INJURY Conditions

Model

Concussion

Contusion

Axonal Injury

ASDH

Skull
Fracture

Weight Drop

+++

+

+

-

+++

Fluid Percussion

+++

+

+

-

-

Controlled Cortical Impact (CCI)

+++

++

++

-

-

Dynamic Cortical Deformation

-

-

-

-

-

Inertial

+++

6

+++

++

-

Impact
Acceleration

+++

+

++

-

+

Table 1: Relative utility of models for duplicating human traumatic brain injury conditions. ASDH = acute subdural hematoma; ICH = intracerebral hematoma. - does not duplicate the condition; + inconsistent; + duplicates to some degree; ++ greater fidelity; +++ greatest fidelity. Adapted from [9].

The translational value of animal models of TBI depends in large part on the degree to which they reproduce patterns similar to those experienced in the human.  The use of animal models also allows for a better understanding of the neurobiological mechanisms underlying therapy including the prediction of novel and better treatments for brain injury. At present, there is a major unmet need for drugs to treat the cognitive deficits in brain injury.  In this regard, the study of neurotransmitter interactions within and between brain regions can facilitate the development of novel compounds targeted to treat those cognitive deficits not limited to a single pharmacological class.  It is hoped that we are edging closer to a stage where we can predict drug efficacy in brain injured patients (against all classes of symptoms) from the efficacy seen in animal models. This in turn should greatly facilitate the discovery of improved drug treatments for TBI.

Acknowledgements

Supported by Science Foundation Ireland (SFI), Higher Education Authority (HEA), National Development Programme (NDP). Programme for Research in Third Level Institutions (PRTLI) and Enterprise Ireland (EI). Issues raised in this article are discussed in more detail in a review article [2].

Corresponding Author

Professor William T. O'Connor
Foundation Chair, Head of Teaching and Research in Physiology
Graduate Entry Medical School. 
Faculty of Education & Health Sciences,
University of Limerick,
Limerick,
Ireland

Web: www.ul.ie/medicalschool
http://inside-the-brain.com/

Web Site:  Inside the Brain

Readers may also be interested in my web site - Inside the brain - http://inside-the-brain.com/ which covers recent neuroscience findings on brain injury including some helpful tips on how to get the best from your brain. Sign-up to receive up-to-date alerts.  Subscription is free.

References

  1. Maas AI, Stocchetti N, Bullock R (2008). Moderate and severe traumatic brain injury in adults.Lancet Neurol. 8:728−741.
  2. O'Connor WT. et al. (2011). Animal models of traumatic brain injury. A critical evaluation. Pharmacol & Therapeut. 130:106–113.
  3. Smith DH, Meaney DF. (2000). Axonal damage in traumatic brain injury. The Neuroscientist.6(6):483-495.
  4. Colgan NC, Cronin MM., Gobbo OL, O’Mara SM, O’Connor WT, Gilchrist MD. (2010). Quantitative MRI analysis of brain volume changes due to controlled cortical impact. J. Neurotrauma. 7:1265-74.
  5. Smyth, A, Gilchrist MD, Kelly JA, O'Connor WT. (2005a). Perfusion with TRH into the medial prefrontal cortex following a mild local injury increases local glutamate release in the rat. Acta Neurobiol. Exp. 65(Suppl. 1–115):90.
  6. Smyth A, Gilchrist MD, O'Connor WT. (2005b). A selective depolarisation-induced increase in excitatory acid neurotransmitter release in rat medial prefrontal cortex using a microdialysis model of traumatic brain injury. In M. D. Gilchrist (Ed.), Proceedings of the IUTAM Symposium on Impact Biomechanics: from Fundamental Insights to Applications (pp. 393−404): University College Dublin 11–15th July.
  7. Smyth A, Gilchrist MD, O'Connor WT. (2005c). Increases in local glutamate, aspartate and GABA levels following traumatic brain injury to the medial prefrontal cortex. Ir. J. Med .Sci. 174(1):40.
  8. Smyth A. O'Connor WT. (2004). Increased glutamate, aspartate and GABA levels following controlled cortical impact to the medial prefrontal cortex in the rat. Int. J. Neuropsychopharmacol. 7. (S1):S474−S475.
  9. Gennarelli TA. (1994). Animate models of human head injury. J Neurotrauma. 11(4): 357-368.
  10. Lighthall JW, Dixon CE, Anderson TE. (1989). Experimental models of brain injury. J Neurotrauma. 6(2): 83-97.
  11. Nilsson P, Hillered L, Ponten U, Ungerstedt U. (1990). Changes in cortical extracellular levels of energy-related metabolites and amino acids following concussive brain injury in rats. J. Cereb.Blood Flow Met. 10:631-637.