Measurements and morphometric landmarks of the human spinal cord: A cadaveric study.

The topographical neuroanatomy of the human spinal cord (SC) is currently based on the adjacent vertebrae. This morphometric study sought to develop a dataset allowing for statistical analysis of human SC segment characteristics. Overall, 32 human SCs were dissected (18 female and 14 male donors), and individual SC segments were identified. Anterior and posterior lengths, thicknesses and widths were measured by two examiners. Statistical analyses included t-tests, as well as intraclass and Pearson's correlation coefficients. The SC length was significantly shorter in females than males. The cranial (C4) and caudal (T1/T2) limits of the cervical enlargement, along with its maximal width (C6-C7), were identified by combining widths and thicknesses of the segments. The thoracic region, T2 to T12, could be identified using segments widths and thicknesses values. The length of the lumbosacral region, from segments L2 to S5, was particularly stable, independently of SC length and sex. The lumbar enlargement was characterized by a thickness increase between the segments L2 and S1 which reached its maximum at the level of L3, L4, and L5, whereas the width was not significantly increased. From the S2 to S5 segments, width and thickness were equal, with both decreasing of 1 mm per segment. The morphometrical analysis of 32 human SCs provided a dataset allowing for statistical analysis of segmental measures with significant results. A combined approach mostly using widths and thicknesses provided landmarks of potential interest for the localization of SC segments in a clinical MRI setting.