Marion Fournely1, Yvan Petit2, Éric Wagnac2, Jérôme Laurin3, Virginie Callot4, Pierre-Jean Arnoux5. 1. Aix-Marseille Université, IFSTTAR, UMR T24, Laboratoire de Biomécanique Appliquée (LBA), Marseille, France; International Laboratory on Spine Imaging and Biomechanics (iLab-Spine), France. 2. Mechanical Engineering Department, École de technologie supérieure, Montréal, Canada; Research Center, Hôpital du Sacré-Cœur, Montréal, Canada; International Laboratory on Spine Imaging and Biomechanics (iLab-Spine), France. 3. Aix-Marseille Université, CNRS, UMR 7287, Institut des Sciences du Mouvement - Etienne Jules MAREY (ISM), Marseille, France. 4. Aix-Marseille Université, CNRS, UMR 7339, Centre de Résonance Magnétique Biologique et Médicale (CRMBM-CEMEREM), Marseille, France; International Laboratory on Spine Imaging and Biomechanics (iLab-Spine), France. 5. Aix-Marseille Université, IFSTTAR, UMR T24, Laboratoire de Biomécanique Appliquée (LBA), Marseille, France; International Laboratory on Spine Imaging and Biomechanics (iLab-Spine), France. Electronic address: pierre-jean.arnoux@ifsttar.fr.
Abstract
BACKGROUND: Animal models of spinal cord injuries aim to utilize controlled and reproducible conditions. However, a literature review reveals that mouse contusion studies using equivalent protocols may show large disparities in the observed impact force vs. cord compression relationship. The overall purpose of this study was to investigate possible sources of bias in these measurements. The specific objective was to improve spinal cord compression measurements using a video-based setup to detect the impactor-spinal cord time-to-contact. NEW METHOD: A force-controlled 30kDyn unilateral contusion at C4 vertebral level was performed in six mice with the Infinite Horizon impactor (IH). High-speed video was used to determine the time-to-contact between the impactor tip and the spinal cord and to compute the related displacement of the tip into the tissue: the spinal cord compression and the compression ratio. RESULTS & COMPARISON WITH EXISTING METHOD(S): Delayed time-to-contact detection with the IH device led to an underestimation of the cord compression. Compression values indicated by the IH were 64% lower than those based on video analysis (0.33mm vs. 0.88mm). Consequently, the mean compression ratio derived from the device was underestimated when compared to the value derived from video analysis (22% vs. 61%). CONCLUSIONS: Default time-to-contact detection from the IH led to significant errors in spinal cord compression assessment. Accordingly, this may explain some of the reported data discrepancies in the literature. The proposed setup could be implemented by users of contusion devices to improve the quantative description of the primary injury inflicted to the spinal cord.
BACKGROUND: Animal models of spinal cord injuries aim to utilize controlled and reproducible conditions. However, a literature review reveals that mouse contusion studies using equivalent protocols may show large disparities in the observed impact force vs. cord compression relationship. The overall purpose of this study was to investigate possible sources of bias in these measurements. The specific objective was to improve spinal cord compression measurements using a video-based setup to detect the impactor-spinal cord time-to-contact. NEW METHOD: A force-controlled 30kDyn unilateral contusion at C4 vertebral level was performed in six mice with the Infinite Horizon impactor (IH). High-speed video was used to determine the time-to-contact between the impactor tip and the spinal cord and to compute the related displacement of the tip into the tissue: the spinal cord compression and the compression ratio. RESULTS & COMPARISON WITH EXISTING METHOD(S): Delayed time-to-contact detection with the IH device led to an underestimation of the cord compression. Compression values indicated by the IH were 64% lower than those based on video analysis (0.33mm vs. 0.88mm). Consequently, the mean compression ratio derived from the device was underestimated when compared to the value derived from video analysis (22% vs. 61%). CONCLUSIONS: Default time-to-contact detection from the IH led to significant errors in spinal cord compression assessment. Accordingly, this may explain some of the reported data discrepancies in the literature. The proposed setup could be implemented by users of contusion devices to improve the quantative description of the primary injury inflicted to the spinal cord.