Objective—To determine the effect of region of interest (ROI) setting and slice thickness on trabecular bone mineral density (BMD) measured with quantitative CT in dogs. Animals—14 healthy Beagles. Procedures—CT of the lumbar vertebrae and a quantitative CT phantom was performed. The BMD of trabecular bone was measured from L1 to L7 in 2 ways in all dogs. First, sequential 9.6-mm-thick CT images were acquired and then CT images were reconstructed into transverse CT images with slice thicknesses of 2.4, 4.8, and 9.6 mm. The obtained images were analyzed by circular ROI and trace ROI methods. Second, lumbar vertebrae were scanned with the installed quantitative CT protocol with a slice thickness of 10 mm and then the CT images were analyzed by installed automatic BMD software. Results—Interclass correlation coefficients of the automatic software (0.975 to 1.0) and the circular method (0.871 to 0.996) were high, compared with those of the trace method (0.582 to 0.996). The BMD measured with the automatic software was not significantly different from that measured with circular ROI and a slice thickness of 9.6 mm. The BMD measured by use of the circular method was not different according to slice thickness. Conclusions and Clinical Relevance—Results obtained by use of automatic software were similar to those obtained by use of more manual methods. The CT images with thinner slice thickness (2.4 and 4.8 mm) could be used in dogs of toy and small breeds to measure lumbar vertebrae BMD to reduce the limitations of the standard 10-mm slice thickness.

Objective-To determine the gross morphology of the multifidus, longus colli, and longus thoracis muscles in the cervical and cranial thoracic portions of the equine vertebral column. Sample-15 horse cadavers. Procedures-The vertebral column was removed intact from the first cervical vertebra (C1) to the seventh thoracic vertebra (T7). After removing the superficial musculature, detailed anatomic dissections of the multifidus, longus colli, and longus thoracis muscles were performed. Results-The multifidus cervicis muscle consisted of 5 bundles/level arranged in lateral, medial, and deep layers from C2 caudally into the thoracic portion of the vertebral column. Fibers in each bundle attached cranially to a spinous process then diverged laterally, attaching caudally on the dorsolateral edge of the vertebral lamina and blending into the joint capsule of an articular process articulation after crossing 1 to 4 intervertebral joints. The longus colli muscle had ventral, medial, and deep layers with 5 bundles/level from C1 to C5 that attached cranially to the ventral surface of the vertebral body, diverged laterally and crossed 1 to 4 intervertebral joints, then attached onto a vertebral transverse process as far caudally as C6. The longus thoracis muscle consisted of a single, well-defined muscle belly from C6 to T5-T6, with intermediate muscular attachments onto the ventral aspects of the vertebral bodies, the intervertebral symphyses, and the craniomedial aspects of the costovertebral joint capsules. Conclusions and Clinical Relevance-Results indicated that there were multiple, short bundles of the multifidus cervicis, multifidus thoracis, and longus colli muscles; this was consistent with a function of providing sagittal plane intersegmental vertebral column stability.