Animal models of neuropathic pain after spinal cord injury

Emraja, Ahmed M. M. (2013) Animal models of neuropathic pain after spinal cord injury. PhD thesis, University of Glasgow.

Full text available as:
[thumbnail of 2013EmrajaPhD.pdf] PDF
Download (7MB)
Printed Thesis Information:


Approximately 70% of spinal cord injured patients suffer from pain and it is estimated that in 40-50% of these, the pain is of central neuropathic origin. This pain can be perceived to originate at, or below the level of injury and both evoked and spontaneous pain can occur. Neuropathic pain after spinal cord injury (SCI) is difficult to treat and often poorly controlled by the currently available analgesics so that development of better treatments is an important need. Current ideas about the treatment of SCI pain are that different approaches may be needed to treat the different types of pain (e.g. evoked and spontaneous, at level and below level) as they may have different mechanisms. However, this mechanistic approach to treatment is hampered by a poor understanding of the underlying mechanisms. This in turn depends on development of animal models and pain assessment techniques suitable for mechanistic studies. In this thesis, several rodent SCI models have been investigated using a range of assessment techniques some of which were developed in the course of the study.

Contusion injuries at a low thoracic level are currently the most popular model used to investigate central neuropathic pain in rodents. However this model and the assessments used with it are subject to a number of limitations. We therefore began by re-evaluating this model using a relatively severe (200 kdyn) injury since this is indicated in the literature as being necessary for the development of robust signs of neuropathic pain. We found that this model showed robust signs of tactile allodynia and thermal hyperalgesia of the forepaws and in addition by developing new tests, were able to demonstrate cold allodynia and hyperalgesia. Although the hindpaws also showed responses that would normally be interpreted as mechanical allodynia and thermal hyperalgesia, the absence of accompanying supraspinally mediated behaviours (including licking following heat stimuli) indicated that the enhanced responsiveness to these stimuli might not give rise to pain. Further investigation using operant testing supported this idea and tract tracing suggested that this may be due to substantial interruption of ascending nociceptive pathways. Testing over the back at locations confirmed electrophysiologically to involve sensory processing at, above and below the injury level supported the idea that increased sensitivity in this model developed at and above, but not below level. In addition, observations on the forepaws suggested evidence of spontaneous pain which has never been described in SCI models previously and provides an important opportunity for studying the underlying mechanisms.
Because the 200 kdyn low thoracic model proved unsuitable for the study of below level pain we next investigated whether a less severe injury at this level would provide a better model. Injuries of 150 kdyn were found to result in most of the same indicators of pain following forelimb testing as were seen following 200 kdyn injuries but all signs were less pronounced, in particular, indicators of evoked pain. Testing over the back led to increased sensitivity below level which had not been evident in the 200 kdyn model, providing an opportunity for below level testing. However, interpretation of hindpaw tests remained equivocal.

Because the low thoracic model showed features that suggested forelimb assessments were particularly useful for the assessment of above level pain of different modalities as well as spontaneous pain, we investigated the effect of moving the injury closer to the segments assessed by such tests. Injuries at the T3/T4 level were found to lead to enhancement of all of the behavioural signs seen in the 200 kdyn low thoracic injury animals, especially signs of spontaneous pain. This model may therefore be optimum for the assessment of above/at level pain.

The work presented in this thesis provides the clearest and most comprehensive data yet on the utility of models of SCI for the investigation of central neuropathic pain and represents a significant advance in the field. The finding that injuries at low thoracic levels may (depending on injury severity) be unsuitable for assessment of below level pain has implications for previous studies of the mechanisms of post SCI pain, many of which have used exclusively hindlimb assessments in these models. The hope is that an improved understanding of the models used here and an improved ability to investigate different modalities of evoked pain, and in addition spontaneous pain, will enhance the quality of future research in this area and lead to both a better understanding of central neuropathic pain mechanisms and the development of more effective analgesics for this type of pain.

Item Type: Thesis (PhD)
Qualification Level: Doctoral
Keywords: Modelling of SCI neuropathic pain, animal models of SCI pain
Subjects: R Medicine > RZ Other systems of medicine
Colleges/Schools: College of Medical Veterinary and Life Sciences > School of Psychology & Neuroscience
Supervisor's Name: Riddell, Dr. John
Date of Award: 2013
Depositing User: Mr Ahmed Mohammed Majied Emraja
Unique ID: glathesis:2013-4176
Copyright: Copyright of this thesis is held by the author.
Date Deposited: 17 Apr 2013 14:28
Last Modified: 15 Apr 2016 15:24

Actions (login required)

View Item View Item


Downloads per month over past year