Study Design: A new magnetic resonance image analysis method is proposed which is based on the definition of the borders of the vertebral bodies adjacent to the intervertebral disc and their varying relationships. The reproducibility of this method (the socalled “centroid” method) was assessed by consecutive measurements. Its potential to depict diurnal disc height variations was studied using randomized groups of volunteers.
Objective: To determine if magnetic resonance imaging can reliably measure disc height variations in the lumbar spine in vivo.
Summary of Background Data: A review of the literature indicates that noninvasive, accurate methods to study the effect of load on intervertebral discs in vivo are needed.
Methods: The reproducibility of the centroid method was assessed in 10 healthy volunteers in 2 consecutive measurements and compared to a conventional method (mean anterior and posterior disc height). To investigate the potential for the depiction of diurnal disc height variations, 10 volunteers were randomized in a study group (1 measurement in the morning, 1 measurement in the evening) and a control group (2 consecutive measurements in the morning).
Results: The centroid method allows the depiction of disc height variations as small as 0.85 mm with a 95% confidence (tolerance limits), whereas a conventional method needs variations of at least 1.66 mm. In the study (diurnal) group, the disc height decreased significantly (P < 0.0001) during the day (mean, ‐0.9 mm), while no variation (P < 0.8) was found in the control group.
Conclusions: These results indicate that the centroid method can reliably detect disc height variations in an experimental setting. The centroid method provides the potential for evaluations of the effects of various work places, work equipment, work tasks, and postures.
From the *Department of Orthopaedic Surgery, University of Bern, †M.E. Müller‐Institute for Biomechanics, University of Bern, ‡Department of Orthopaedic Surgery, McGill University, Montreal, Canada, and §Department of MR‐Spectroscopy and Methodology, University of Bern.
Presented in Part at the 40th Annual Meeting of the Orthopaedic Research Society, February 21‐24, 1994, New Orleans, Louisiana.
Supported by a grant from the Swiss National Science Foundation (Grant # 32‐28633.90) and from the AO/ASIF‐Foundation, Switzerland.
Acknowledgment date: December 23, 1994.
First revision date: April 26, 1995.
Acceptance date: July 18, 1995.
Device status category: 1.
Address reprint requests to: Norbert Boos, MD; Centre for Spinal Studies and Surgery; Queen's Medical Centre; University Hospital; Nottingham NG7 2UH; United Kingdom