Measurement of three-dimensional cervical segmental kinematics: Reliability of whole vertebrae and facet-based approaches.

BACKGROUND: Previous studies have used orientation and translation of whole-vertebrae to describe three-dimensional cervical segmental kinematics. Describing kinematics using facet joint movement may be more relevant to pathology and effects of interventions but has not been investigated in the cervical spine. This study compared the reliability of two different methods (whole-vertebrae vs facet joint) to evaluate cervical kinematics. METHODS: Two healthy adults each had six cervical (C1 to T1) magnetic resonance imaging scans, two each in neutral and left and right rotation. A semi-automated method of segmentation and alignment determined the relative orientation and translation of each whole-vertebrae and translation of each facet joint. Intra-rater and inter-rater reliability was determined using limits of agreement (LOA) with 95% confidence intervals and intraclass correlation coefficients (ICC3,1 for intra- and ICC2,1 for inter-rater). RESULTS: The LOA for intra-rater evaluation of facet movement was superior to whole vertebra translation. Both methods showed excellent intra-rater ICC3,1 (0.80-0.99) and inter-rater ICC2,1 (0.79-0.85) for all variables except for Euler angle for flexion/extension which was good (0.65). Intra-and inter-rater ICCs were better for facet movement than all measures of whole of vertebrae movement except Euler angles of axial rotation where no difference was detected. CONCLUSIONS: Measurement of three-dimensional segmental kinematics using either the facet joint or the whole-vertebrae method demonstrated excellent and comparable reliability. These findings support the use of the facet joint method as an option for describing and investigating cervical segmental kinematics.