An in situ model of chronic cervical spinal cord injury

In vivo rodent models of cervical spinal cord injury (cSCI) are commonly used to examine impaired respiratory activity using measures of either phrenic nerve output or diaphragmatic activity. These methods allow quantification of overall deficits due to hemisection or contusion injury but hemostatic and ventilatory movement issues often confound recordings of affected neural circuitry. To address this challenge, we have developed an in situ rodent model of cSCI. This preparation uses an established method of contusion injury (Infinite Horizon pneumatic impactor – 100 kdyne at C4) in juvenile (P21) rodents and allows up to 2 weeks of recovery prior to experimental recording. The improved accessibility and elimination of hemostatic concerns facilitates recording of individual cervical and thoracic ventral roots while allowing stable monitoring of neuronal ensembles using multielectrode array technology. Additionally, this decerebrate and artificially perfused preparation removes the need for anesthesia. Initial recordings demonstrate segmental efferent impairment (C4 and C5) corresponding to injury location, with sparse inspiratory effort representing a reduction of motoneuron activity. Increased tonic and expiratory activity of inspiratory ventral roots contributing to the phrenic nerve (ipsilateral to injury) suggest dysfunction of the local network. We propose that this model is a significant improvement over in vivo paradigms for defining both the spinal circuitry and plasticity involved in recovery from cSCI. Funded by the University of Florida College of Veterinary Medicine, NIH 1R01NS080180‐01A1 and R01 NS054025.