Jonathan L. Sakai, B.S., Lars G. Gilbertson, Ph.D., Louis E. DeFrate, B.S.,
Seong-Hwan Moon, M.D., William F. Donaldson III, M.D.,
James D. Kang, M.D., Savio L-Y. Woo, Ph.D.
Musculoskeletal Research Center
The University of Pittsburgh
Pittsburgh, Pennsylvania 15213

Delineation of in-vivo cervical spine kinematics can provide clinically valuable benchmarks for quantifying biomechanical effects of aging, degeneration, injury, surgery, rehabilitation, and other agents. Active kinematic assessments typically involve the measurement of overall cervical spine mobility as a volunteer/patient actively performs selected spinal movements. A disadvantage of active assessments is their dependence on the ability and willingness of the volunteer or patient to control his overall movements—mandating development of methods for enabling improved control.

The magnetic tracking/virtual reality (VR) based system has recently been developed for measurement, description, and control of movements of the head relative to the torso. The objective of this study was to test the efficacy of VR-assisted visual feedback of overall cervical spine kinematics for improving control of voluntary cervical spine movements in the assessment of the active kinematics of normal volunteers and spinal surgery patients.

The magnetic tracking/VR-based system consists of a magnetic tracking system, head mounted display, and programmable visual interface, operating under the control of a personal computer. The magnetic tracking system Flock of Birds^TM (Ascension Technologies, Inc., Burlington, VT 05402) was used to measure the translational and rotational movements of the head and torso within a global coordinate system defined by the position and orientation of a stationary transmitter placed behind a seated subject. Ten healthy male volunteers (with no history of spinal problems), and three cervical fusion patients performed maximal cervical spine active range-of-motion (AROM) maneuvers in flexion/extension, axial rotation, and lateral bending, both without and with VR-assisted visual feedback of their overall rotations. The primary and secondary rotations at the extremes of primary movement were analyzed to determine normative baseline values (controls) and effect of fusion surgery (patients).

For the control group, the mean overall cervical spine primary rotations at the extremes of primary movement for AROM maneuvers with VR feedback were statistically no different from those without VR feedback—supporting use of VR feedback for acquiring physiologic range-of-motion data (Table 1).

Figure 1 illustrates the improved quality of spinal movement control achieved with VR feedback by the 10 control subjects during lateral bending. Secondary rotations with VR feedback spanned a much narrower range than those without VR feedback—indicating that the subjects were successful in reducing secondary rotations, resulting in a "purer" overall rotation—thereby improving the consistency of the AROM maneuver.

Table 2 compiles the ranges-of-motion (ROM) of the three cervical spine fusion patients, achieved with VR feedback. Patient 1 had received a one-level (C5-6) fusion surgery, while patients 2 and 3 received two-level (C5-7) surgeries. Although Patient 1 had one more mobile level (C6-7) than the other patients, range-of-motion was much less than attained by the two-level fusion patients—possibly due to presence of pain in Patient 1.

Comparison of data from Tables 1 and 2 reveal that the ROM of the patients was less than that of the control subjects in each of the primary rotational degrees-of-freedom studied. Considering only the two-level fusion patients (who were pain-free at the time of the kinematic assessment), the decreased range of motion of Patient 2 (with respect to control group) was consistent with the loss of mobility of the C5-6 and C6-7 segments. The ROM of Patient 3, however, was considerably less than expected based solely on loss of mobility of C5-7—pointing to the presence of other factors.

In summary, a magnetic tracking/VR-based system for comprehensive kinematic assessment of the cervical spine has been developed and tested on 10 asymptomatic volunteers and 3 spinal fusion patients, and has been found to aid subjects in control of active range-of-motion maneuvers. Our long-term objective is to explore the potential use of this system as a basis for standardized, more objective testing of normal volunteers and clinical patients.

Table 1. Mean (± SD) range of motion of control subjects, without and with VR feedback (n=10). Primary Movement Primary Rotation (degrees) Without VR With VR Flexion 62.9 ± 8.5 64.0 ± 10.1 Extension -57.4 ± 11.9 -58.3 ± 12.9 Left AR 70.4 ± 6.0 72.9 ± 7.0 Right AR -72.1 ± 6.1 -74.2 ± 6.6 Left LB -37.8 ± 9.0 -38.2 ± 8.7 Right LB 38.8 ± 9.3 39.4 ± 10.0

Figure 1. Box plot showing range and distribution of primary and secondary rotations during left lateral bending AROM maneuvers by the 10 control subjects, without and with VR feedback.

Table 2. Overall cervical spine rotations for three fusion patients. Primary Movement Rotation (deg) Patient 1* Patient 2** Patient 3** FE 32.8 109.6 64.7 AR 59.3 125.4 93.5 LB 28.3 64.8 36.0 FE = flexion/extension, AR = axial rotation, LB = lateral bending. * 1-level fusion; ** 2-level fusion

Supported by the Whitaker Foundation.