(G-274) Changes in naturally behaving animals correlate with spinal cord injury pain

Introduction:: Spinal cord injuries (SCI) can degrade patient quality of life  through unpredictable disability and chronic pain. Furthermore, managing SCI-associated pain and disability is often ineffective but can require tens to hundreds of thousands of dollars annually in direct costs . Better understanding SCI pathogenesis will improve how clinicians treat SCI. Preclinical rodent models offer powerful platforms to explore SCI pathogenesis, including pain and disability, where clinical studies are unsuitable. However, quantifying relevant pain-related physiology and behavioral symptoms in rodents is challenging, with many gold standard assays (e.g. rotorod, Randall-Selitto, grip strength, Hargreaves, etc.) introducing exacerbating stress and variability in animal performances; Most animal assessments measuring rodent pain and disability require stressful human interaction, out-of-cage environments, and evoked pain ultimately confounding results. Further, the nature of current assays only captures a few minutes of animal experience infrequently, often once a week or less, thus limiting their interpretive strength. The goal of this project is to develop a low-stress, noninvasive, continuous method to assess predictive rodent homecage physiology and behaviors (eating, drinking, grooming, rearing, exploring) to better understand SCI pathogenesis.

Materials and Methods:: Adult female c57b/6 mice (n=17) were pair-housed with standard 12-hour light:dark cycles with ad libitum food and water. Homecages used clear lids and temporary insert to spatially isolate animals during video recordings while enabling visual, olfactory, and thermal interactions between cagemates. Homecages were placed in custom acoustic isolation chambers and animals were video recorded from above during the 12-hour dark cycle twice weekly before and for 10 weeks after receiving either a moderate T10 contusion spinal cord injury (n=7) or T10 laminectomy sham (n=10) procedure. Briefly, animals were prepared for aseptic surgery and deeply anesthetized. A midline skin incision, bluntly dissected muscle and fascia, and laminectomies exposed the T10 spinal cord segment. At this point, sham animal wounds were closed while T10 SCI animals received a contusion (dorsal impact, Infinite Horizon Impactor, 50 kdynes, no dwell time) prior to wound closure. All animals received post-operative pain relief, antibiotics, supplemental hydration, and bladder expression as needed. Hind paw mechanical sensitivity was assessed weekly using the gold standard Chaplan’s up-down protocol.

Videos were manually scored using ELAN software (6.3, Nijmegen: The Max Planck Institute for Psycholinguistics) to measure total time, frequency, and average duration of spontaneous resting, drinking, eating, grooming, exploring, or rearing behaviors. Data were aggregated to weekly averages per animal. ANOVAs with Tukey’s HSD posthoc tests were performed with a significance of 0.05 to identify changes from baseline to week 10. General linear regressions were performed to identify correlations between behavioral measures and hind paw mechanical sensitivity (e.g. evoked pain).



Results, Conclusions, and Discussions:: Initial findings presented here summarize one hour of manually scored video per animal at baseline and week 10 post-injury. At 10-weeks post-injury, SCI animals spent more percentage of their time eating (p < 0.047), groomed fewer times (p < 0.01), and exhibited heightened mechanical sensitivity (p< 0.025) relative to their respective baseline (Figure 1A). These behavioral changes suggest SCI pain is associated with changes in natural animal behaviors (eating and grooming). For example, grooming requires full-body movement, including twisting around to reach the hindquarters and tail, which may be intuitively impaired through disability or pain after SCI. Mechanical sensitivity correlated with the percentage of time animals spent rearing (p < 0.033, R2 = 0.20), suggesting thoracic SCI pain may compromise behaviors that put large forces on the back, such as rearing. This is reinforced with the finding that SCI animals tended to consolidate their rearing behavior into less frequent, longer bouts of rearing (n.s.) possibly because the transition motion of elevating to a rearing position may be more uncomfortable than maintaining a rearing position. Though the correlation between percentage time rearing and pain is significant, the R2 value is low indicating mechanical sensitivity explains a relatively small amount of variability in rearing behavior data and suggests other unmeasured factors may be involved. This behavioral and pain variability mimics the significant challenge clinicians face in their own SCI patient disability and pain variability. We expect additional behavior and pain relationships within the context of animal/SCI patient variability to be uncovered once the entire data set are manually scored.

These initial findings represent a promising new method to quantify SCI behavior symptoms to better understand SCI pathogenesis. Several natural animal behaviors changed after SCI and the development of pain. Measuring animals in their homecage environment without human contact reduces experimental bias (specifically environmental variability) and enables a better understanding of how variability in SCI pain and behavior may relate. Future work will be focused on clarifying these findings and identifying new behaviors to monitor for pain and discomfort, as well as developing an automated method to score these behaviors.




Acknowledgements (Optional): :

References (Optional): : National Spinal Cord Injury Statistical Center, Facts and Figures at a Glance. Birmingham, AL: University of Alabama at Birmingham, 2016.