Medical College of Wisconsin
CTSICores SearchResearch InformaticsREDCap

Generating level-dependent models of cervical and thoracic spinal cord injury: Exploring the interplay of neuroanatomy, physiology, and function. Neurobiol Dis 2017 Sep;105:194-212



Pubmed ID




Scopus ID

2-s2.0-85020764778   26 Citations


The majority of spinal cord injuries (SCI) occur at the cervical level, which results in significant impairment. Neurologic level and severity of injury are primary endpoints in clinical trials; however, how level-specific damages relate to behavioural performance in cervical injury is incompletely understood. We hypothesized that ascending level of injury leads to worsening forelimb performance, and correlates with loss of neural tissue and muscle-specific neuron pools. A direct comparison of multiple models was made with injury realized at the C5, C6, C7 and T7 vertebral levels using clip compression with sham-operated controls. Animals were assessed for 10weeks post-injury with numerous (40) outcome measures, including: classic behavioural tests, CatWalk, non-invasive MRI, electrophysiology, histologic lesion morphometry, neuron counts, and motor compartment quantification, and multivariate statistics on the total dataset. Histologic staining and T1-weighted MR imaging revealed similar structural changes and distinct tissue loss with cystic cavitation across all injuries. Forelimb tests, including grip strength, F-WARP motor scale, Inclined Plane, and forelimb ladder walk, exhibited stratification between all groups and marked impairment with C5 and C6 injuries. Classic hindlimb tests including BBB, hindlimb ladder walk, bladder recovery, and mortality were not different between cervical and thoracic injuries. CatWalk multivariate gait analysis showed reciprocal and progressive changes forelimb and hindlimb function with ascending level of injury. Electrophysiology revealed poor forelimb axonal conduction in cervical C5 and C6 groups alone. The cervical enlargement (C5-T2) showed progressive ventral horn atrophy and loss of specific motor neuron populations with ascending injury. Multivariate statistics revealed a robust dataset, rank-order contribution of outcomes, and allowed prediction of injury level with single-level discrimination using forelimb performance and neuron counts. Level-dependent models were generated using clip-compression SCI, with marked and reliable differences in forelimb performance and specific neuron pool loss.

Author List

Wilcox JT, Satkunendrarajah K, Nasirzadeh Y, Laliberte AM, Lip A, Cadotte DW, Foltz WD, Fehlings MG


Kajana Satkunendrarajah PhD Assistant Professor in the Neurosurgery department at Medical College of Wisconsin

MESH terms used to index this publication - Major topics in bold

Caspase 3
Cervical Vertebrae
Disease Models, Animal
Evoked Potentials, Somatosensory
Exploratory Behavior
Magnetic Resonance Imaging
Motor Activity
Motor Neurons
Nerve Tissue Proteins
Psychomotor Performance
Rats, Wistar
Spinal Cord Injuries
Thoracic Vertebrae
Time Factors