Open Access Highly Accessed Research article

Are the determinants of vertebral endplate changes and severe disc degeneration in the lumbar spine the same? A magnetic resonance imaging study in middle-aged male workers

Mari Kuisma1*, Jaro Karppinen278, Marianne Haapea1, Jaakko Niinimäki1, Risto Ojala1, Markku Heliövaara3, Raija Korpelainen49, Kaisu Kaikkonen49, Simo Taimela5, Antero Natri6 and Osmo Tervonen1

Author Affiliations

1 Department of Diagnostic Radiology, Oulu University Hospital, Oulu, Finland

2 Department of Physical and Rehabilitation Medicine, Oulu University Hospital, Oulu, Finland

3 Public Health Institute, Helsinki, Finland

4 Department of Sports Medicine, Oulu Deaconess Institute, Oulu, Finland

5 Department of Public Health, University of Helsinki, Finland

6 Department of Orthopaedics, Tampere University Hospital, Tampere, Finland

7 Musculoskeletal Centre, Finnish Institute of Occupational Health, Oulu, Finland

8 ORTON Orthopedic Hospital, Helsinki, Finland

9 Department of Public Health and General Practice, University of Oulu, Finland

For all author emails, please log on.

BMC Musculoskeletal Disorders 2008, 9:51  doi:10.1186/1471-2474-9-51

Published: 16 April 2008



Modic changes are bone marrow lesions visible in magnetic resonance imaging (MRI), and they are assumed to be associated with symptomatic intervertebral disc disease, especially changes located at L5-S1. Only limited information exists about the determinants of Modic changes. The objective of this study was to evaluate the determinants of vertebral endplate (Modic) changes, and whether they are similar for Modic changes and severe disc degeneration focusing on L5-S1 level.


228 middle-aged male workers (159 train engineers and 69 sedentary factory workers) from northern Finland underwent sagittal T1- and T2-weighted MRI. Modic changes and disc degeneration were analyzed from the scans. The participants responded to a questionnaire including items of occupational history and lifestyle factors. Logistic regression analysis was used to evaluate the associations between selected determinants (age, lifetime exercise, weight-related factors, fat percentage, smoking, alcohol use, lifetime whole-body vibration) and Modic type I and II changes, and severe disc degeneration (= grade V on Pfirrmann's classification).


The prevalences of the Modic changes and severe disc degeneration were similar in the occupational groups. Age was significantly associated with all degenerative changes. In the age-adjusted analyses, only weight-related determinants (BMI, waist circumference) were associated with type II changes. Exposure to whole-body vibration, besides age, was the only significant determinant for severe disc degeneration. In the multivariate model, BMI was associated with type II changes at L5-S1 (OR 2.75 per one SD = 3 unit increment in BMI), and vibration exposure with severe disc degeneration at L5-S1 (OR 1.08 per one SD = 11-year increment in vibration exposure).


Besides age, weight-related factors seem important in the pathogenesis of Modic changes, whereas whole-body vibration was the only significant determinant of severe disc degeneration.