Sage Journals: Discover world-class research

Abstract

The most important acute neurological diseases seen at neurosurgery departments are traumatic brain injuries (TBI) and subarachnoid hemorrhages (SAH). In both diseases the pathophysiological sequela are complex and have not been fully understood up to now, and rodent models using rats and mice are most suitable for the investigation of the pathophysiological details. In both models, surgery is performed under anesthesia, followed by assessment of their functional outcome and behavioral testing before brain tissue analysis after euthanasia. Postoperative analgesia is mandatory, and supplementary care is highly recommended for refinement purposes. Pain and stress assessment is mainly based on clinical and behavioral signs, and further research is needed to improve the evaluation of severity in these models.

Keywords

pain rodent severity subarachnoid hemorrhage traumatic brain injury

Besides spinal injury, the most important acute neurological diseases that neurosurgery departments have to deal with are traumatic brain injuries (TBI) and hemorrhaging into the subarachnoid space from acute ruptures of cerebral aneurysms (aneurysm-induced subarachnoid hemorrhages, SAH). In both diseases the pathophysiological sequela are complex and have not been fully understood up to now.^1–3 For targeted therapy we have to remedy this lack of knowledge by thoroughly investigating the pathophysiological cascades using well characterized animal models that mimick the human diseases in total or in well-defined separate components. The most commonly used species for TBI and SAH research are small rodents such as rats and mice,⁴ with rats being used more often in SAH studies than mice (evaluation up to 2007⁵). We will therefore focus our review on the severity assessment and scoring for TBI and SAH models in rats and mice.

Animal models of TBI and SAH

Both diseases are characterized by an initial impact to the cerebral tissue due to a physical insult (TBI) or a massive bleeding from a ruptured artery (SAH), and the development of secondary damage which evolves over hours and days or even weeks after the initial insult, leading to the growth of a lesion. Depending on the area of the damaged tissue in the brain, specific functional impairment occurs. To evaluate the degree of cognitive, sensory or motor deficits, a battery of functional tests are available which have been established in small rodents. By using the modified neurological severity score, the grade of impairment can be quantitatively assessed. Both basic reflexes (pinna reflex and corneal reflex) and somatomotor functions (paw flexion, head support, spontaneous and initiated locomotion), as well as postural responses and startle responses are usually tested in mice and rats.^6,7 In addition to these simple tests, more specific assessments are possible, though these are highly demanding of time and experience (a review and descriptions of these tests are beyond the scope of this paper but are provided in detail elsewhere^8–13). As a final outcome measure, the brains are usually analyzed by histomorphological and molecular biological methods in detail after sacrifice of the animals.

Induction of TBI

TBI is the leading cause of mortality and disability in the younger population (up to the age of 45) worldwide.¹⁴ Within the pathophysiological course of TBI, we can differentiate between the primary damage, which rapidly evolves as a direct result of the physical impact on brain tissue, and secondary damage, which enlarges the initial injury to the tissue.¹⁵

Several injury models have been used in TBI research, and detailed descriptions of the models most often applied in rodents and other species have recently been published.^4,8,16 Based on the information provided, we have summarized important aspects of the most commonly used TBI models in Table 1a.^17–22
Table 1a.
Most commonly used traumatic brain injury (TBI) models in rats and mice.

TBI model Established for Main procedure, way of impact Reproducibility/ controllability of impact and tissue deformation Focal injury Diffuse/ axonal injury Assessment according to a) Etiology (construct validity) b) Biological markers (face validity) c) Treatment response (predictive validity)

Advantage Disadvantage

Fluid percussion injury model (FPI)^17,18 Rat, mouse Craniotomy, dura mater intact, fluid wave Moderate reproducibility: control of parameters of initiation of fluid wave, variability by location of craniotomy and dispersion of fluid wave ✓ ✓ a) Brain dislocation comparable to the situation after falls in humans b) Focal cortical contusion with axonal injury, acute apnea, acute and chronic seizures, functional deficit/behavioral impairment c) Good treatment response a) Craniotomy b) No regular brainstem damage – lack of unconsciousness, mortality variable (low to high) depending on the degree of pressure pulse, rarely monitoring of vital physiological parameters performed c) Translation into clinical trials failed

Controlled cortical impact (CCI)^19,20 Rat, mouse Craniotomy, dura mater intact, pneumatic device High reproducibility: control of time of compression, design and velocity of impactor, depth of impact ✓ – (only minor) b) Cortical contusion, lesion enlargement over days, functional deficit/behavioral impairment c) Good treatment response a) Craniotomy, only focal impact b) No axonal injury, large contusion injury, no brainstem damage – lack of unconsciousness, very low mortality, rarely monitoring of vital physiological parameters performed c) Translation into clinical trials failed

Weight drop impact acceleration model (WDIA) (‘Marmarou model)^21,22 Rat, (mouse only as repeat mild model) Skull intact, weight drop on metal disk fixed to the skull, head not fixed and lying on a cushion with spring force Moderate reproducibility: control of weight drop, acceleration/ deceleration of head not controllable – ✓ a) Strong diffuse impact b) Brainstem damage – prolonged unconsciousness, higher mortality, functional deficit/behavioral impairment b) No focal lesion or cell death c) Model rarely used in pharmacological evaluation

Induction of SAH

In men, aneurysmal SAH is accompanied by a high risk of mortality immediately after or within the first few hours after the initial bleeding. The early injury is mainly caused by a dramatic transient increase in intracranial pressure (ICP) followed by global cerebral ischemia,^23,24 and by the impact of the blood clot (inducing inflammation, vascular endothelial cell damage, etc.).⁹ Up to 30% of patients who survive the initial event develop delayed neurological deficits caused by delayed ischemic events.²⁵ The early brain injury is not only responsible for the early symptoms of the disease but is also the starting point for the delayed deterioration.^26,27 Two main approaches have been used in SAH model experiments using mice and rats which are compiled in Table 1b.^5,9,28–35
Table 1b.
Most commonly used subarachnoid hemorrhage (SAH) models in rats and mice.

SAH model Established for Main procedure Reproducibility/ controllability of bleeding severity/ amount of blood Early injury Delayed vasospasm/ delayed neurological deficits Assessment according to a) Etiology (construct validity) b) Biological markers (face validity) c) Treatment response (predictive validity)

Advantage Disadvantage

Endovascular perforation model^28–32
Rat, mouse
Small filament or wire introduced into the ICA (via the CCA or ECA) and advanced intracranially to perforate the Circle of Willis at the region of the anterior circulation at the bifurcation of the anterior cerebral artery and the middle cerebral artery
Moderate reproducibility: only by design of the filament tip/wire size, variability concerning exact perforation location and amount of blood pouring out
✓
Vasospasm of large arteries 3–7 days after SAH/delayed neurological deficit or late deterioration not systematically reported
a) Closely resembles the situation of spontaneous rupture of a cerebral aneurysm, distribution of blood in the subarachnoid space b) Large and rapid increase in ICP with transient global ischemia; early constriction of microcirculation; inflammation, vascular endothelial cell damage; BBB damage, parenchymal cell death, functional deficit/behavioral impairment, mortality comparable to human situation c) Good treatment response
a) Endovascular manipulation by the device c) Translation into clinical trials failed

Blood injection model in cisterna magna or prechiasmatic cistern (only once or repeated)^{9,28,29,33–35} Rat, mouse Small craniotomy, puncture of the cisterna magna or needle placed in the prechiasmatic cistern Moderate to high reproducibility: control of amount of blood injected and injection pressure/velocity, variability concerning the distribution of blood within the subarachnoid space ✓ Vasospasm of large arteries 3–7 days after SAH/delayed neurological deficit or late deterioration not systematically reported a) Distribution of blood in the subarachnoid space b) Large and rapid increase in ICP with transient global ischemia; early constriction of microcirculation; inflammation, vascular endothelial cell damage; BBB damage, parenchymal cell death, functional deficit/behavioral impairment, mortality depends on amount of blood injected – in more severe models comparable to human situation c) Good treatment response a) Craniotomy, no rupture of a cerebral vessel – blood artificially reaches subarachnoid space by cisternal injection c) Translation into clinical trials failed

BBB: blood–brain barrier; CCA: common carotid artery; ECA: external carotid artery; ICA: internal carotid artery; ICP: intracranial pressure.

Additional procedure (both TBI and SAH)

Despite TBI or SAH in humans occurring while they are awake, anesthesia is obligatory when performing TBI or SAH surgery in animals.

In both models, during the initial surgery (SAH) and/or at different time points after the insult (SAH, TBI), it is general practice to measure the ICP as an important parameter of disease severity.³⁶ This can be achieved by insertion of a cannula in the cisterna magna connected to a pressure transducer, or by using fiber-optic-based systems, where a small pressure transducer is placed between the skull and the dura mater. In both cases, a small craniotomy has to be performed.

Anesthesia and surgical procedures may induce hypothermia in rodents, not only during the surgery but also within hours after awakening. However normothermia has to be ensured, since hypothermia provides neuroprotection,^37,38 and may also impair animal welfare.³⁹

Reasons for choosing the rodent model for TBI and SAH

Most TBI and SAH animal models were first established in larger animal species, however in recent decades rodent models have most often been used. Besides the advantages of low cost (purchase and housing), easy handling and high level of standardization, rodents provide the possibility of advanced assessment of functional outcome and analysis of the underlying complex molecular cascades of cellular damage in the tissue after euthanasia. In addition genetically-modified mouse models can advance our understanding of molecular pathways.⁴⁰ As suggested in Tables 1a and 1b, the validity in all three common categories (construct, face and predictive) is not perfect for either TBI or SAH. This is not surprising since TBI in humans is a highly heterogeneous and complicated disease. The same is true for SAH, where size and location of the aneurysm, and the amount of blood and pressure under which it is released during rupture, depend on each individual situation. In addition, comorbidities (such as systemic hypertension or diabetes) make the situation even more complex. It is therefore not possible to model all aspects of the diseases in a single animal model,^4,16 and the selection of the model has to be based on a clear definition of the respective aim to be followed and hypothesis to be proven. Accordingly rodent models for TBI and SAH are of extraordinary high value, and have remained indispensable to advancement of our yet incomplete understanding of the pathophysiological cascades in TBI and SAH, by elucidating important parts of the complex puzzle.

Current knowledge shows that the pathophysiological events in TBI and SAH are substantially comparable between small rodents and humans; yet the human brain is structurally and functionally more complex, with significant differences in the white-to-grey matter ratio. Therefore, in order to close the translational gap for preclinical drug development, studies in larger animal species with brains more closely resembling those of humans are recommended in order to make comprehensive use of the findings on pathophysiological cascades achieved in rodent studies.

Severity of the model

As stated in the ‘Working Document on a Severity Assessment Framework’,⁴¹ a standard procedure should be followed for severity assessment, based on the thorough consideration of each step within the model procedure, to result in an assessment tailored for each respective species and strain, and for the individual laboratory situation. Therefore we can only outline the most basic considerations here and not refer to details within the models and species.

In both TBI and SAH, ICP increase occurs acutely or develops within hours. To measure ICP, which is highly recommended, a small craniotomy has to be performed. In addition, the skin and the underlying soft tissue at the head and/or the neck have to be cut. The brain itself has no pain receptors, but the neurocranium with its overlying periost and the meninges that are mainly affected during physical impact, and/or the elevated ICP, are highly algesic. The level of pain induced by the increased ICP correlates with the degree of the physical impact in TBI (which can be standardized) or the amount of blood in SAH (which can be standardized only in the blood injection model). After awaking, animals develop neurological deficits, which may reduce their ability to take food and water (with consequent risk of weight loss and dehydration) and inhibit normal behavior within their social environment. Assessment of functional deficits will probably also be associated with stress for the animals affected by the disease, particularly if they are not used to the tests. In addition, rodents are by nature prey animals, and as such fugitive,⁴² and it is not known whether being unable to escape due to motor and/or cognitive deficits itself induces stress in rats and mice. Post-surgical housing is typically individual housing, so as to avoid wound manipulation by cage mates; but isolation can also be stressful for social species.^43,44

Mortality in the controlled cortical impact (CCI)–TBI model is usually low (<10%, while controlled by the degree of impact) because no damage to the brainstem occurs. Mortality in fluid percussion injury (FPI) models depends strongly on the size of the pressure pulse, with mild to moderately severe impact resulting in mortality rates of 10–22% in rats¹⁷ and zero mortality in mice.¹⁸ In the ‘Marmarou’ (weight drop impact acceleration [WDIA]) model, the originally described mortality of up to 60% in spontaneously breathing rats is reduced to <10% by intubation and mechanically-assisted ventilation, which is recommended as the standard procedure for this model.²² Very recently, the WDIA rat model has been redesigned for application in mice as a repeat mild TBI model, with zero mortality.⁴⁵

In SAH, mortality depends strongly on the model. It can reach up to 60% in the endovascular perforation model in rats,^9,46 while most of the animals usually die immediately after the perforation (while still under anesthesia) or within the first few hours after the bleeding,⁴⁶ which resembles the early mortality rate in human patients. Recently, a modification of the endovascular perforation model in rats has been described, using a tungsten wire instead of a polypropylene filament, which induces a vessel perforation with a smaller diameter, leading to reduced mortality rates of around 20%.^47–49 In mice, the endovascular perforation model is accompanied by a mortality rate of around 30%.^5,30 In the blood injection models, mortality depends on the amount of blood injected, and usually remains ≤10%.⁵

In summary, besides ensuring a thorough assessment of pain and distress, the definition of humane endpoints is especially necessary when using the more clinically related models of TBI and SAH. Overall, based on the cumulative consideration of the above-mentioned aspects, severity is usually graded as moderate for TBI (depending on the degree of physical impact) as well as for SAH when using the modified versions of the models with reduced mortality rates, and when implementing all the refinement opportunities mentioned in the text and table 3 below.

Assessment of welfare

Expected signs of pain, stress, disease, constraint or signs of reduced welfare

In rats and mice, pain and suffering cannot be easily detected. Generally, the burden induced by the pre- and postoperative handling, the surgery and the local effect of the lesion (most important of which is the increase in ICP) on the one hand, and the impairment of normal behavior caused by the lesion-induced functional deficit on the other hand, cannot be distinguished by common measures of reduced welfare. Therefore, up to now, signs of reduced well-being in TBI and SAH studies have had to be considered as unspecific as well as model-specific. Rats and mice are social animals, and reduced welfare rapidly leads to reduced spontaneous activity and isolation of the animals from their cage mates, which can be easily detected by observing the animals in their home cages. Furthermore, reduced activity can be tested by stimulating movement with a gentle push. Unusual body posture (crooked back) or an ungroomed appearance indicate that the animal is reluctant to move. Following TBI and SAH, any disregard of grooming (resulting in porphyrin deposits around the eyes and nose in rats) may be a sign of discomfort as well as the result of functional impairment of the paws or a deficit in motor coordination. Unusually aggressive behavior is another, yet more unspecific, sign of reduced well-being or stress and pain. Another important marker of constraint, and possibly the most important and scalable one, is reduced body weight, which is measured by weighing the animals daily. This indicates reduced food and water intake, which is caused by both unspecific illness as well as by functional impairment. More model-specific symptoms are the development of convulsions and epileptic seizures, which add to the burden for the animals. Isolation and sitting with the head towards the corner of the cage may be signs of head pain caused by increased ICP.

All these signs can be monitored by clinical examination. Routine home cage observation is therefore obligatory for welfare assessment, which however requires thorough knowledge of the normal appearance and behavior of the respective species. Strain-dependent characteristics also have to be considered. Besides body weight, these measures are not scalable and are, at least to some extent, observer-dependent.

This basic clinical observation and scoring system has been developed according to the standard procedure defined in the ‘Working Document on a Severity Assessment Framework’,⁴¹ and is frequently used and accepted as a routine procedure for experimental studies of TBI and SAH. The main disadvantage of clinical scoring systems is the lack of scalability and the necessity for the observer to be sufficiently experienced; and it is not clear to what extent minor stress and pain or even subtle deviations from well-being can be detected using these tests. In addition, illness-related deviation from normal behavior cannot be discriminated from functional impairment-induced deficits, and it is not known whether (and if so, to what extent) the sole occurrence of functional impairment affects the subjective well-being of the animals. Further research is needed on this issue.

The same is true for more advanced tests assessing spontaneous locomotion (for example from home cage video recording over longer time periods or performance of home cage wheel running), which are increasingly being used for assessing pain in rodent models.⁵⁰ Overall, when applying tests requiring normal sensorimotor and cognitive functions, it must be thoroughly investigated whether abnormal test results are indeed induced by pain and stress, or rather by cognitive, sensory or motor impairment induced by the brain lesion following TBI or SAH, which could be independent of pain and stress.

Score sheet

In the forefront of an experiment a score sheet based on expected signs of reduced well-being has to be developed. In Table 2, a typical score sheet is provided, which has been developed for SAH or TBI studies and is relevant to all the main procedures (see above). It is equally applicable for rats and mice, however, strain differences have to be considered, mainly concerning the overall level of normal activity and aggression.
Table 2.
Score sheet for traumatic brain injury (TBI) and subarachnoid hemorrhage (SAH).

Parameter Description Points

Loss of body weight 0–5% 0

5–10% 1

11–15% 5

16–<20% 10

20+% 30

General condition, physical appearance Well groomed; clean fur; body openings clean; clear, shiny eyes 0

Coat slightly unkempt; chromodacryorrhea (‘red tears’) 1

Coat shaggy and dirty; signs of dehydration (no relief of skin fold); moderate piloerection 5

Coat unkempt; marked piloerection 10

Breathing Normal breathing 0

Tachypnea 5

Dyspnea 10

Spontaneous behavior Normal locomotion, social contact 0

Reluctance to move, slightly abnormal gait 5

Lethargy, apathy, separation from cage mates, markedly abnormal gait 10

Significant mobility problems, intermittent immobility <12 h 20

Immobility >12 h 30

Reaction to handling Normal curiosity, alert 0

Tense and nervous on handling 5

Markedly distressed on handling, e.g. shaking, vocalizing, aggressive 10

Wound healing, condition of suture Wound clear and inconspicuous, no signs of infection, all sutures in place 0

Suture insufficiency, wound dehiscence without signs of infection 5

Swelling, redness as first signs of infection 10

Massive infection, swelling, redness, purulent discharge 25

Simple neurologic evaluation No observable deficit 0

Forelimb flexion 1

Decreased resistance to lateral push (and forelimb flexion) without circling 5

Decreased resistance to lateral push (and forelimb flexion) with circling 10

Seizure (status epilepticus), coma, complete paralysis 30

Actions: Score 0–9: frequency of monitoring as planned, in case of suture insufficiency: suture revision Score 10–20: provide supplementary care (e.g. extra fluids, wet mash), check analgesia protocol, increase monitoring frequency, in case of signs of wound infection: consult veterinarian Score 21–29: review progress with veterinarian Score ≥30: implement humane endpoint.

Assessment of the ‘General condition, physical appearance/Breathing/Spontaneous behavior/Reaction to handling/Wound healing, condition of suture’ topics and of body weight is standard practice for an experienced staff member familiar with normal rodent behavior.

The simple neurological evaluation can be briefly explained. It serves as a rough measure of the lesion-induced functional deficit, which may lead to impaired food and water intake and to deficits in grooming and activity, and therefore adds to the summed score. To evaluate forelimb flexion, the animals are carefully lifted by the tail with support of the lower body and advanced to the surface of a table. Healthy animals symmetrically stretch their limbs towards the ground, and animals with deficits in motor function and coordination of the limbs show asymmetric stretching or failure of stretching. Asymmetric limb function also leads to circular locomotion instead of straight movement.

Humane endpoints

In TBI and SAH models, pain or distress are unavoidable, therefore humane endpoints must be obligatorily defined by an upper limit of maximally tolerable score points. Close-mesh monitoring of the animals throughout the experiment is mandatory for checking if the endpoint is reached, and reaching the humane endpoint score in TBI and SAH studies leads to euthanasia of the animal.

In TBI and SAH models, the most likely problems expected to lead to endpoint scoring are:
weight loss;

pain due to headache (craniotomy/ICP increase) or surgical manipulation;

infection of surgical wound;

immobility.

Thus, the most important symptoms, any one of which can by itself lead to euthanasia, are:
weight loss exceeding >20%;

extreme pain which cannot be relieved by analgesics;

immobility >12 h (assessed by close-mesh observation in the home cage);

status epilepticus, coma, complete paralysis.

In addition, a combination of several signs of pain and stress may also lead to euthanasia, any one of which on its own may not require euthanasia.

Refinement opportunities

According to the ‘ARRIVE – Animals in Research: Reporting In Vivo Experiments’ guidelines,⁵¹ all relevant refinement strategies should be described in the Methods sections when publishing results. In addition it should be kept in mind that every strategy to reduce (avoidable) stress in animal experiments is not only a matter of animal welfare but also leads to more stable and reliable results.

Within recent years, several strategies for refinement have been developed. Model-specific refinements aimed at reducing mortality rates have been described above. The general refinement of the effective training of surgeons should be ensured. Further opportunities for refinement have been identified in the areas of anesthesia and postoperative analgesia, care during surgery, and postoperative care and handling,^22,30,52,53 which are summarized in Table 3.
Table 3.
Refinement opportunities.

Subject Description Refinement To be considered

Anesthesia during surgery Most commonly used anesthesia: Isoflurane⁵² + fentanyl or buprenorphine; Ketamine + xylazine; Midazolam + medetomidine + fentanyl⁵³ Isoflurane alone has only little or no analgesia effect → add analgesic (fentanyl,⁵⁴ buprenorphine,⁵⁵ ketamine or at least local anesthetics⁵⁶) Effect of anesthetics on brain function: neuroprotection by reduction of spontaneous and evoked activity (all anesthetics) and/or increase of cerebral blood flow (isoflurane), or inhibition of glutamate receptors (ketamine) (see review⁵²)

Support of vital function during surgery Anesthesia incidents: Apnea in more severe SAH models and TBI models with brainstem damage (WDIA); Reduced breathing – hypoxia; Anesthesia-induced hypothermia Monitor vital parameters by pulse-oximetry and end expiratory–CO₂ monitor and adjust anesthesia level accordingly;³⁰ Intubation and mechanical ventilation;^22,30 Add oxygen to inspiration – air; Monitor body temperature and use a servo-controlled heating pad Use non-invasive approaches to measure vital parameters; Do not apply too low concentrations of anesthetics; Do not apply to high levels of oxygen; Check proper function of heating pad to avoid overheating

Postoperative analgesia Pain reducing strategies are obligatory and highly recommended Opioids more appropriate than NSAIDs and therefore recommended for postoperative pain reduction; sustained–release opioids preferable due to avoidance of frequent handling for injection Choice of the appropriate analgesic not easy: inflammation is part of the pathophysiological cascade in TBI and SAH, and NSAIDs are therefore not suitable due to their COX-2 inhibition-mediated anti-inflammatory effect; opioids can modify the functional outcome⁵²

Postoperative care – body temperature Hypothermia occurs during and following anesthesia and due to reduced activity, but has to be avoided not only during but also after surgery due to its neuroprotective effect and for reasons of animal welfare Keeping the animals in small heating cabinets with controlled inside temperature for several hours after finishing the surgery as best choice for body temperature control in small rodents Check proper function and establish correct heating temperature before first use; Red-light warming lamps not recommended/should be avoided due to inaccurate setting of the temperature and due to high temperature gradient

Postoperative care – food and water intake Animals often are too weak after surgery for appropriate food and water intake, animals might avoid chewing hard rodent chow due to headache; Straightening up to the water bottle might be hindered by pain at surgical wound and by neurological impairment Offering water-gel pads and mashed food⁵⁷ in a Petri-dish on the floor to encourage water and food intake; Subcutaneous injections of saline or saline enriched with carbohydrates before awakening and daily during recovery phase⁵⁸ Regular and frequent change of dish with mashed food due to contamination with bedding and to avoid mold formation; Subcutaneous injections may induce stress

Handling during lab routine and functional testing Handling during housing care and laboratory routine⁵⁹ and during functional tests may induce stress, which should be minimized Preoperative habituation of the animals to handling and to the tests conditions⁶⁰ Stress reducing effect of habituation to handling is less clear in mice⁶¹

SAH: subarachnoid hemorrhage, TBI: traumatic brain injury, WDIA: weight drop impact acceleration, NSAIDs: non-steroidal anti-inflammatory drugs, COX-2: cyclooxygenase isoform 2.

Recommendations and further 3R research

As described above, the evaluation of pain and stress following TBI and SAH is not easy and up to now has mainly been based on the observance of behavioral and clinical signs of reduced welfare. More advanced evaluation is not commonly performed, but further research is highly recommended for enhancing our knowledge of suffering by the models, and for refining the parameters monitored within the score sheet applied here.

By thoroughly assessing facial expressions in rats and mice, the recently developed ‘grimace scale’ can in general be used to quantify pain.^62,63 Whether it can be applied to TBI and SAH has yet to be proven.⁶⁴ Measurement of hormonal parameters such as prolactin and corticosterone in the blood can be useful for stress assessment.^65,66 However it has to be considered that these hormones underlie circadian variations and show a high variance of basal levels, and sampling itself can be stressful. For reliable assessments, basal preoperative values have to be compared with postoperative values, and blood sampling has to be conducted under similar circumstances (time of day, experimenter and sampling method). It has to be investigated whether analysis of corticosterone content in feces may be a useful alternative to assessment in the blood. This provides an index of stress reaction over several hours, in contrast to blood sampling which provides an index of stress at one brief moment in time.⁶⁷ Monitoring of ultrasonic vocalization may also be considered. While experiencing fear or anxiety, adult rats emit alarm calls at a frequency of 22 kHz.⁶⁸ Yet in a recent study the suitability of using 22 kHz vocalization for pain assessment has been questioned.⁶⁹

Compiling standard operating procedures for the most common models of TBI and SAH with implementation of the above-mentioned refinement strategies, as is already available for the focal cerebral ischemia model of filament-induced middle cerebral artery occlusion in the mouse,⁵⁷ would further help to enhance standardization, reproducibility, and animal welfare in these models.

TBI model	Established for	Main procedure, way of impact	Reproducibility/ controllability of impact and tissue deformation	Focal injury	Diffuse/ axonal injury	Assessment according to a) Etiology (construct validity) b) Biological markers (face validity) c) Treatment response (predictive validity)
Fluid percussion injury model (FPI)^17,18	Rat, mouse	Craniotomy, dura mater intact, fluid wave	Moderate reproducibility: control of parameters of initiation of fluid wave, variability by location of craniotomy and dispersion of fluid wave	✓	✓	a) Brain dislocation comparable to the situation after falls in humans b) Focal cortical contusion with axonal injury, acute apnea, acute and chronic seizures, functional deficit/behavioral impairment c) Good treatment response	a) Craniotomy b) No regular brainstem damage – lack of unconsciousness, mortality variable (low to high) depending on the degree of pressure pulse, rarely monitoring of vital physiological parameters performed c) Translation into clinical trials failed
Controlled cortical impact (CCI)^19,20	Rat, mouse	Craniotomy, dura mater intact, pneumatic device	High reproducibility: control of time of compression, design and velocity of impactor, depth of impact	✓	– (only minor)	b) Cortical contusion, lesion enlargement over days, functional deficit/behavioral impairment c) Good treatment response	a) Craniotomy, only focal impact b) No axonal injury, large contusion injury, no brainstem damage – lack of unconsciousness, very low mortality, rarely monitoring of vital physiological parameters performed c) Translation into clinical trials failed
Weight drop impact acceleration model (WDIA) (‘Marmarou model)^21,22	Rat, (mouse only as repeat mild model)	Skull intact, weight drop on metal disk fixed to the skull, head not fixed and lying on a cushion with spring force	Moderate reproducibility: control of weight drop, acceleration/ deceleration of head not controllable	–	✓	a) Strong diffuse impact b) Brainstem damage – prolonged unconsciousness, higher mortality, functional deficit/behavioral impairment	b) No focal lesion or cell death c) Model rarely used in pharmacological evaluation

SAH model	Established for	Main procedure	Reproducibility/ controllability of bleeding severity/ amount of blood	Early injury	Delayed vasospasm/ delayed neurological deficits	Assessment according to a) Etiology (construct validity) b) Biological markers (face validity) c) Treatment response (predictive validity)
Endovascular perforation model^28–32	Rat, mouse	Small filament or wire introduced into the ICA (via the CCA or ECA) and advanced intracranially to perforate the Circle of Willis at the region of the anterior circulation at the bifurcation of the anterior cerebral artery and the middle cerebral artery	Moderate reproducibility: only by design of the filament tip/wire size, variability concerning exact perforation location and amount of blood pouring out	✓	Vasospasm of large arteries 3–7 days after SAH/delayed neurological deficit or late deterioration not systematically reported	a) Closely resembles the situation of spontaneous rupture of a cerebral aneurysm, distribution of blood in the subarachnoid space b) Large and rapid increase in ICP with transient global ischemia; early constriction of microcirculation; inflammation, vascular endothelial cell damage; BBB damage, parenchymal cell death, functional deficit/behavioral impairment, mortality comparable to human situation c) Good treatment response	a) Endovascular manipulation by the device c) Translation into clinical trials failed
Blood injection model in cisterna magna or prechiasmatic cistern (only once or repeated)^{9,28,29,33–35}	Rat, mouse	Small craniotomy, puncture of the cisterna magna or needle placed in the prechiasmatic cistern	Moderate to high reproducibility: control of amount of blood injected and injection pressure/velocity, variability concerning the distribution of blood within the subarachnoid space	✓	Vasospasm of large arteries 3–7 days after SAH/delayed neurological deficit or late deterioration not systematically reported	a) Distribution of blood in the subarachnoid space b) Large and rapid increase in ICP with transient global ischemia; early constriction of microcirculation; inflammation, vascular endothelial cell damage; BBB damage, parenchymal cell death, functional deficit/behavioral impairment, mortality depends on amount of blood injected – in more severe models comparable to human situation c) Good treatment response	a) Craniotomy, no rupture of a cerebral vessel – blood artificially reaches subarachnoid space by cisternal injection c) Translation into clinical trials failed

Parameter	Description	Points
Loss of body weight	0–5%	0
	5–10%	1
	11–15%	5
	16–<20%	10
	20+%	30
General condition, physical appearance	Well groomed; clean fur; body openings clean; clear, shiny eyes	0
	Coat slightly unkempt; chromodacryorrhea (‘red tears’)	1
	Coat shaggy and dirty; signs of dehydration (no relief of skin fold); moderate piloerection	5
	Coat unkempt; marked piloerection	10
Breathing	Normal breathing	0
	Tachypnea	5
	Dyspnea	10
Spontaneous behavior	Normal locomotion, social contact	0
	Reluctance to move, slightly abnormal gait	5
	Lethargy, apathy, separation from cage mates, markedly abnormal gait	10
	Significant mobility problems, intermittent immobility <12 h	20
	Immobility >12 h	30
Reaction to handling	Normal curiosity, alert	0
	Tense and nervous on handling	5
	Markedly distressed on handling, e.g. shaking, vocalizing, aggressive	10
Wound healing, condition of suture	Wound clear and inconspicuous, no signs of infection, all sutures in place	0
	Suture insufficiency, wound dehiscence without signs of infection	5
	Swelling, redness as first signs of infection	10
	Massive infection, swelling, redness, purulent discharge	25
Simple neurologic evaluation	No observable deficit	0
	Forelimb flexion	1
	Decreased resistance to lateral push (and forelimb flexion) without circling	5
	Decreased resistance to lateral push (and forelimb flexion) with circling	10
	Seizure (status epilepticus), coma, complete paralysis	30

Subject	Description	Refinement	To be considered
Anesthesia during surgery	Most commonly used anesthesia: Isoflurane⁵² + fentanyl or buprenorphine; Ketamine + xylazine; Midazolam + medetomidine + fentanyl⁵³	Isoflurane alone has only little or no analgesia effect → add analgesic (fentanyl,⁵⁴ buprenorphine,⁵⁵ ketamine or at least local anesthetics⁵⁶)	Effect of anesthetics on brain function: neuroprotection by reduction of spontaneous and evoked activity (all anesthetics) and/or increase of cerebral blood flow (isoflurane), or inhibition of glutamate receptors (ketamine) (see review⁵²)
Support of vital function during surgery	Anesthesia incidents: Apnea in more severe SAH models and TBI models with brainstem damage (WDIA); Reduced breathing – hypoxia; Anesthesia-induced hypothermia	Monitor vital parameters by pulse-oximetry and end expiratory–CO₂ monitor and adjust anesthesia level accordingly;³⁰ Intubation and mechanical ventilation;^22,30 Add oxygen to inspiration – air; Monitor body temperature and use a servo-controlled heating pad	Use non-invasive approaches to measure vital parameters; Do not apply too low concentrations of anesthetics; Do not apply to high levels of oxygen; Check proper function of heating pad to avoid overheating
Postoperative analgesia	Pain reducing strategies are obligatory and highly recommended	Opioids more appropriate than NSAIDs and therefore recommended for postoperative pain reduction; sustained–release opioids preferable due to avoidance of frequent handling for injection	Choice of the appropriate analgesic not easy: inflammation is part of the pathophysiological cascade in TBI and SAH, and NSAIDs are therefore not suitable due to their COX-2 inhibition-mediated anti-inflammatory effect; opioids can modify the functional outcome⁵²
Postoperative care – body temperature	Hypothermia occurs during and following anesthesia and due to reduced activity, but has to be avoided not only during but also after surgery due to its neuroprotective effect and for reasons of animal welfare	Keeping the animals in small heating cabinets with controlled inside temperature for several hours after finishing the surgery as best choice for body temperature control in small rodents	Check proper function and establish correct heating temperature before first use; Red-light warming lamps not recommended/should be avoided due to inaccurate setting of the temperature and due to high temperature gradient
Postoperative care – food and water intake	Animals often are too weak after surgery for appropriate food and water intake, animals might avoid chewing hard rodent chow due to headache; Straightening up to the water bottle might be hindered by pain at surgical wound and by neurological impairment	Offering water-gel pads and mashed food⁵⁷ in a Petri-dish on the floor to encourage water and food intake; Subcutaneous injections of saline or saline enriched with carbohydrates before awakening and daily during recovery phase⁵⁸	Regular and frequent change of dish with mashed food due to contamination with bedding and to avoid mold formation; Subcutaneous injections may induce stress
Handling during lab routine and functional testing	Handling during housing care and laboratory routine⁵⁹ and during functional tests may induce stress, which should be minimized	Preoperative habituation of the animals to handling and to the tests conditions⁶⁰	Stress reducing effect of habituation to handling is less clear in mice⁶¹

Footnotes

Declaration of conflicting interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Pearn

Niesman

Egawa

. Pathophysiology associated with traumatic brain injury: current treatments and potential novel therapeutics. Cell Mol Neurobiol. Epub ahead of print xx month July 2016. DOI: 10.1007/s10571-016-0400-1.

Sehba

Friedrich

. Early events after aneurysmal subarachnoid hemorrhage. Acta Neurochir Suppl 2015; 120: 23–28.

Kooijman

Nijboer

van Velthoven

Kavelaars

Kesecioglu

Heijnen

. The rodent endovascular puncture model of subarachnoid hemorrhage: mechanisms of brain damage and therapeutic strategies. J Neuroinflammation 2014; 11: 2–2.

Marklund

Hillered

. Animal modelling of traumatic brain injury in preclinical drug development: where do we go from here? Br J Pharmacol 2011; 164: 1207–1229.

Marbacher

Fandino

Kitchen

. Standard intracranial in vivo animal models of delayed cerebral vasospasm. Br J Neurosurg 2010; 24: 415–434.

Chen

Sanberg

. Intravenous administration of human umbilical cord blood reduces behavioral deficits after stroke in rats. Stroke 2001; 32: 2682–2688.

Thal

Sporer

Klopotowski

. Brain edema formation and neurological impairment after subarachnoid hemorrhage in rats. Laboratory investigation. J Neurosurg 2009; 111: 988–994.

Gold

Lopez-Velazquez

. Functional assessment of long-term deficits in rodent models of traumatic brain injury. Regen Med 2013; 8: 483–516.

Jeon

Sabri

. Neurological and neurobehavioral assessment of experimental subarachnoid hemorrhage. BMC Neurosci 2009; 10: 103–103.

10.

Schallert

. Behavioral tests for preclinical intervention assessment. NeuroRx 2006; 3: 497–504.

11.

Zorner

Filli

Starkey

. Profiling locomotor recovery: comprehensive quantification of impairments after CNS damage in rodents. Nat Meth 2010; 7: 701–708.

12.

Morris

. Developments of a water-maze procedure for studying spatial learning in the rat. J Neurosci Methods 1984; 11: 47–60.

13.

Paul

Magda

Abel

. Spatial memory: theoretical basis and comparative review on experimental methods in rodents. Behav Brain Res 2009; 203: 151–164.

14.

Langlois

Marr

Mitchko

Johnson

. Tracking the silent epidemic and educating the public: CDC’s traumatic brain injury-associated activities under the TBI Act of 1996 and the Children’s Health Act of 2000. J Head Trauma Rehabil 2005; 20: 196–204.

15.

McIntosh

Smith

Meaney

Kotapka

Gennarelli

Graham

. Neuropathological sequelae of traumatic brain injury: relationship to neurochemical and biomechanical mechanisms. Lab Invest 1996; 74: 315–342.

16.

Morganti-Kossmann

Yan

Bye

. Animal models of traumatic brain injury: is there an optimal model to reproduce human brain injury in the laboratory? Injury 2010; 41(Suppl. 1): S10–S13.

17.

Dixon

Lyeth

Povlishock

. A fluid percussion model of experimental brain injury in the rat. J Neurosurg 1987; 67: 110–119.

18.

Carbonell

Maris

McCall

Grady

. Adaptation of the fluid percussion injury model to the mouse. J Neurotrauma 1998; 15: 217–229.

19.

Dixon

Clifton

Lighthall

Yaghmai

Hayes

. A controlled cortical impact model of traumatic brain injury in the rat. J Neurosci Methods 1991; 39: 253–262.

20.

Hannay

Feldman

Phan

. Validation of a controlled cortical impact model of head injury in mice. J Neurotrauma 1999; 16: 1103–1114.

21.

Foda

Marmarou

. A new model of diffuse brain injury in rats. Part II: Morphological characterization. J Neurosurg 1994; 80: 301–313.

22.

Marmarou

Foda

van den Brink

Campbell

Kita

Demetriadou

. A new model of diffuse brain injury in rats. Part I: Pathophysiology and biomechanics. J Neurosurg 1994; 80: 291–300.

23.

Bederson

Levy

Ding

. Acute vasoconstriction after subarachnoid hemorrhage. Neurosurgery 1998; 42: 352–360. discussion 60–62.

24.

Broderick

Brott

Duldner

Tomsick

Leach

. Initial and recurrent bleeding are the major causes of death following subarachnoid hemorrhage. Stroke 1994; 25: 1342–1347.

25.

Macdonald

. Delayed neurological deterioration after subarachnoid haemorrhage. Nat Rev Neurol 2014; 10: 44–58.

26.

Fujii

Yan

Rolland

Soejima

Caner

Zhang

. Early brain injury, an evolving frontier in subarachnoid hemorrhage research. Transl Stroke Res 2013; 4: 432–446.

27.

Terpolilli

Brem

Buhler

Plesnila

. Are we barking up the wrong vessels? Cerebral microcirculation after subarachnoid hemorrhage. Stroke 2015; 46: 3014–3019.

28.

Megyesi

Vollrath

Cook

Findlay

. In vivo animal models of cerebral vasospasm: a review. Neurosurgery 2000; 46: 448–460; discussion 60–61.

29.

Zoerle

Ilodigwe

Wan

. Pharmacologic reduction of angiographic vasospasm in experimental subarachnoid hemorrhage: systematic review and meta-analysis. J Cereb Blood Flow Metab 2012; 32: 1645–1658.

30.

Feiler

Friedrich

Scholler

Thal

Plesnila

. Standardized induction of subarachnoid hemorrhage in mice by intracranial pressure monitoring. J Neurosci Methods 2010; 190: 164–170.

31.

Kamii

Kato

Kinouchi

. Amelioration of vasospasm after subarachnoid hemorrhage in transgenic mice overexpressing CuZn–superoxide dismutase. Stroke 1999; 30: 867–871; discussion 72.

32.

Bederson

Germano

Guarino

. Cortical blood flow and cerebral perfusion pressure in a new noncraniotomy model of subarachnoid hemorrhage in the rat. Stroke 1995; 26: 1086–1091; discussion 91–92.

33.

Solomon

Antunes

Chen

Bland

Chien

. Decrease in cerebral blood flow in rats after experimental subarachnoid hemorrhage: a new animal model. Stroke 1985; 16: 58–64.

34.

Kamp

Dibue

Sommer

Steiger

Schneider

Hanggi

. Evaluation of a murine single-blood-injection SAH model. PLoS One 2014; 9: e114946–e114946.

35.

Prunell

Mathiesen

Svendgaard

. A new experimental model in rats for study of the pathophysiology of subarachnoid hemorrhage. Neuroreport 2002; 13: 2553–2556.

36.

Friess

Lapidus

Brody

. Decompressive craniectomy reduces white matter injury after controlled cortical impact in mice. J Neurotrauma 2015; 32: 791–800.

37.

Clifton

Allen

Barrodale

. A phase II study of moderate hypothermia in severe brain injury. J Neurotrauma 1993; 10: 263–271. discussion 73.

38.

Torok

Klopotowski

Trabold

Thal

Plesnila

Scholler

. Mild hypothermia (33 degrees C) reduces intracranial hypertension and improves functional outcome after subarachnoid hemorrhage in rats. Neurosurgery 2009; 65: 352–359. discussion 9.

39.

Donnerer

Lembeck

. Different control of the adrenocorticotropin–corticosterone response and of prolactin secretion during cold stress, anesthesia, surgery, and nicotine injection in the rat: involvement of capsaicin-sensitive sensory neurons. Endocrinology 1990; 126: 921–926.

40.

Xiong

Mahmood

Chopp

. Animal models of traumatic brain injury. Nat Rev Neurosci 2013; 14: 128–142.

41.

National Competent Authorities for the implementation of Directive 2010/63/EU on the protection of animals used for scientific purposes. Working document on a severity assessment framework, http://ec.europa.eu/environment/chemicals/lab_animals/pdf/Endorsed_Severity_Assessment.pdf (2012, accessed 19 October 2016).

42.

Jennings

Batchelor

Brain

. Refining rodent husbandry: the mouse. Report of the Rodent Refinement Working Party. Lab Anim 1998; 32: 233–259.

43.

Spani

Arras

Konig

Rulicke

. Higher heart rate of laboratory mice housed individually vs in pairs. Lab Anim 2003; 37: 54–62.

44.

Wallace

Han

Graham

. CREB regulation of nucleus accumbens excitability mediates social isolation-induced behavioral deficits. Nat Neurosci 2009; 12: 200–209.

45.

Nguyen

Lehar

. Repetitive mild traumatic brain injury with impact acceleration in the mouse: multifocal axonopathy, neuroinflammation, and neurodegeneration in the visual system. Exp Neurol 2016; 275: 436–449.

46.

Lee

Sagher

Keep

Hua

. Comparison of experimental rat models of early brain injury after subarachnoid hemorrhage. Neurosurgery 2009; 65: 331–343. discussion 43.

47.

Park

Meno

Witt

. Subarachnoid hemorrhage model in the rat: modification of the endovascular filament model. J Neurosci Methods 2008; 172: 195–200.

48.

Greenhalgh

Rothwell

Allan

. An endovascular perforation model of subarachnoid haemorrhage in rat produces heterogeneous infarcts that increase with blood load. Transl Stroke Res 2012; 3: 164–172.

49.

Höllig

Weinandy

Nolte

Clusmann

Rossaint

Coburn

. Experimental subarachnoid hemorrhage in rats: comparison of two endovascular perforation techniques with respect to success rate, confounding pathologies and early hippocampal tissue lesion pattern. PLoS One 2015; 10: e0123398–e0123398.

50.

Kandasamy

Calsbeek

Morgan

. Home cage wheel running is an objective and clinically relevant method to assess inflammatory pain in male and female rats. J Neurosci Methods 2016; 263: 115–122.

51.

Kilkenny

Browne

Cuthill

Emerson

Altman

. Improving bioscience research reporting: the ARRIVE guidelines for reporting animal research. PLoS Biol 2010; 8: e1000412–e1000412.

52.

Rowe

Harrison

Thomas

Pauly

Adelson

Lifshitz

. Using anesthetics and analgesics in experimental traumatic brain injury. Lab Anim (NY) 2013; 42: 286–291.

53.

Hockel

Trabold

Scholler

Torok

Plesnila

. Impact of anesthesia on pathophysiology and mortality following subarachnoid hemorrhage in rats. Exp Transl Stroke Med 2012; 4: 5–5.

54.

Flierl

Stahel

Beauchamp

Morgan

Smith

Shohami

. Mouse closed head injury model induced by a weight-drop device. Nat Protoc 2009; 4: 1328–1337.

55.

Kooijman

Nijboer

van Velthoven

. Long-term functional consequences and ongoing cerebral inflammation after subarachnoid hemorrhage in the rat. PLoS One 2014; 9: e90584–e90584.

56.

Albert-Weissenberger

Varrallyay

Raslan

Kleinschnitz

Siren

. An experimental protocol for mimicking pathomechanisms of traumatic brain injury in mice. Exp Transl Stroke Med 2012; 4: 1–1.

57.

Dirnagl U and members of the MCAO–SOP Group. Standard operating procedures (SOP) in experimental stroke research: SOP for middle cerebral artery occlusion in the mouse. Nature Precedings 2009: http://dx.doi.org/10.1038/npre.2012.3492.3.

58.

Milner

Holtzman

Friess

. Endovascular perforation subarachnoid hemorrhage fails to cause Morris water maze deficits in the mouse. J Cereb Blood Flow Metab 2014; 34. 10.1038/jcbfm.2014.108.

59.

Balcombe

Barnard

Sandusky

. Laboratory routines cause animal stress. Contemp Top Lab Anim Sci 2004; 43: 42–51.

60.

Dobrakovova

Kvetnansky

Oprsalova

Jezova

. Specificity of the effect of repeated handling on sympathetic–adrenomedullary and pituitary–adrenocortical activity in rats. Psychoneuroendocrinology 1993; 18: 163–174.

61.

Meijer

Sommer

Spruijt

van Zutphen

Baumans

. Influence of environmental enrichment and handling on the acute stress response in individually housed mice. Lab Anim 2007; 41: 161–173.

62.

Sotocinal

Sorge

Zaloum

. The rat grimace scale: a partially automated method for quantifying pain in the laboratory rat via facial expressions. Mol Pain 2011; 7: 55–55.

63.

Langford

Bailey

Chanda

. Coding of facial expressions of pain in the laboratory mouse. Nat Methods 2010; 7: 447–449.

64.

Miller

Leach

. The mouse grimace scale: a clinically useful tool? PLoS One 2015; 10: e0136000–e0136000.

65.

Seggie

Brown

. Stress response patterns of plasma corticosterone, prolactin, and growth hormone in the rat, following handling or exposure to novel environment. Can J Physiol Pharmacol 1975; 53: 629–637.

66.

Kant

Eggleston

Landman-Roberts

Kenion

Driver

Meyerhoff

. Habituation to repeated stress is stressor specific. Pharmacol Biochem Behav 1985; 22: 631–634.

67.

Ward

Blanchard

Bolivar

. Recognition and alleviation of distress in laboratory animals, Washington, DC: The National Academies Press, 2008, pp. 132–132.

68.

Sanchez

. Stress-induced vocalisation in adult animals. A valid model of anxiety? Eur J Pharmacol 2003; 463: 133–143.

69.

Wallace

Norbury

Rice

. Ultrasound vocalisation by rodents does not correlate with behavioural measures of persistent pain. Eur J Pain 2005; 9: 445–452.

Severity assessment and scoring for neurosurgical models in rodents

Abstract

Keywords

Animal models of TBI and SAH

Induction of TBI

Induction of SAH

Additional procedure (both TBI and SAH)

Reasons for choosing the rodent model for TBI and SAH

Severity of the model

Assessment of welfare

Expected signs of pain, stress, disease, constraint or signs of reduced welfare

Score sheet

Humane endpoints

Refinement opportunities

Recommendations and further 3R research

Footnotes

Declaration of conflicting interests

Funding

References