Skip to main content

Thank you for visiting You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Disc degeneration of the lumbar spine in relation to overweight



To study the association between overweight and lumbar disc degeneration.


Population-based 4-y follow-up magnetic resonance imaging (MRI) study.


The subjects were 129 working middle-aged men selected to the baseline magnetic resonance imaging (MRI) study from a cohort of 1832 men representing three occupations: machine drivers, construction carpenters, and office workers. The selection was based on the paticipants’age (40–45 y) and place of residence. MR images of the lumbar spines were obtained at baseline and at 4-y follow-up.


Signal intensity of the nucleus pulposus of the discs L2/L3–L4/L5 was visually assessed by two readers using the adjacent cerebrospinal fluid as an intensity reference. The weight (at age 25 and 40–45 y) and height of the subjects, history of car driving, smoking, and back injuries were assessed by questionnaire.


Multiple regression analyses allowing for occupation, history of car driving, smoking, and back injuries showed that persistent overweight (body mass index (BMI) ≥25 kg/m2 at both ages) associated strongly with an increased risk of the number of lumbar discs with decreased signal intensity of nucleus pulposus at follow-up, adjusted odds ratio (OR) being 4.3 (95% confidence intervals (95% CIs) 1.3–14.3). Overweight at young age (risk ratio (RR) 3.8; 95% CI 1.4–10.4) was a stronger predictor of an increase in the number of degenerated discs during follow-up than overweight in middle age (RR 1.3; 95% CI 0.7–2.7).


The study provides evidence that the BMI above 25 kg/m2 increases the risk of lumbar disc degeneration. Overweight at young age seems to be particularly detrimental.


Increasing obesity is a serious public health problem worldwide. Obesity has been shown to increase the risk of cardiovascular disease and diabetes,1, 2, 3 osteoarthritis,4, 5 and spine diseases.6, 7 It has been shown that the adverse effects of excess weight tend to be delayed, sometimes for 10y or longer.8

Disc degeneration of the lumbar spine is considered to be one of the underlying factors of low back pain.9, 10 Among the risk factors most commonly suspected of accelerating degeneration are age, occupational physical loading, back injury, and smoking.11

Obesity has been implicated as a risk factor of lumbar disc degeneration, but epidemiological studies have reported both positive and null associations. In a cross-sectional study of consecutive patients who underwent surgery for lumbar intervertebral disc herniation, overweight was found to be strongly over-represented.12 Obesity predicted admission to hospital for a herniated lumbar intervertebral disc in men but not in women.13 However, overweight did not have an effect on the prevalence of disc degeneration among concrete reinforcement workers and painters in a cross-sectional X-ray study.14

Symmons et al15 studied women aged 45–64 y in a 9-y follow-up with repeated lumbar radiographs, and found that increased body mass index (BMI) was a risk factor of degeneration. Parkkola et al16 found a positive association between overweight and disc degeneration of the L1 disc but not at any other level in a study of 74 healthy volunteers.

Knowledge of the risk factors of lumbar disc degeneration is mainly based on studies of lateral radiographs, in which only indirect information of advanced disc degeneration can be obtained. Magnetic resonance imaging (MRI) gives direct morphological information on the biochemical status of the intervertebral discs, not obtainable by other methods. Even early degenerative changes can be detected by MRI.

In all, there are very few MRI follow-up studies of lumbar disc degeneration.17, 18 Both studies had a relatively small sample size. No association was found between BMI and change in disc degeneration in the 5-y follow-up evaluation of 41 asymptomatic individuals.18

The aim of the current follow-up MRI study was to investigate the effect of an increased BMI on the prevalence and progression of lumbar disc degeneration among middle-age men taking other influential factors related to occupation and lifestyle into account.



The subjects of this study were employed middle-aged men, a subgroup from a cohort of 1832 men representing three occupations (machine drivers, construction carpenters, and office workers) who had participated in two previous questionnaire surveys.19 In 1991, 210 men were selected to the baseline MRI study using age (40–45 y) and place of residence (Helsinki metropolitan area) as inclusion criteria. A total of 164 (71%) persons took part in MRI examination. All of them were invited in 1995 to the follow-up study and 129 (79%) participated.

Subjects lost to follow-up

A total of 33 men were lost from follow-up, because of either nonparticipation or technical problems in imaging. Two men were excluded because of incomplete data on their weight history. The group of subjects lost from follow-up and the participants in the baseline study did not differ significantly as to the degree of disc degeneration as determined by the number of discs with decreased signal intensity of the nucleus pulposus.

Magnetic resonance imaging equipment and protocol

The same 0.1 T MRI scanner and the same surface coil were used both in the baseline study and in the follow-up study. The same dual-echo technique was used with a 2000 ms repetition time and 25 and 86 ms echo times producing partial saturation proton density and spin-echo T2-weighted images in sagittal plane. Slice thickness was 7 mm, field of view 410 mm × 410 mm and pixel size 1.6 mm × 1.6 mm. Care was taken to assure that the subjects’ position in the scanner was as closely as possible the same in the follow-up MRI as it had been in the baseline MRI. The positioning was compared with the baseline MR image, and corrected if needed, before starting the main scanning sequence.

Visual assessment of signal intensity of nucleus pulposus

Two independent readers who were blinded to occupation of the subjects separately evaluated the MR images. The images of 1991 and 1995 were independently read. It was impossible to be blinded to the year of imaging because of a change in the appearance of the hard copies. Reader one (AL) was an experienced radiologist with more than 10 y experience in MRI, and reader two (ML) was a specially trained medical doctor. The radiologist's evaluations were used in the data analyses. Image quality was not sufficient for reliable evaluation of the discs at L1/L2 level and of many at L5/S1 level and therefore these levels were excluded from the analysis, and only the three middle lumbar discs (L2/L3, L3/L4, and L4/L5) were evaluated.

The signal intensity of the nucleus pulposus was visually assessed by comparing it with that of the cerebrospinal fluid (CSF) (brighter than CSF, as bright as CSF, darker than CSF, and as dark as cortical bone of vertebra). For data analysis, the signal intensity variable was dichotomized: discs with a nucleus pulposus brighter or as bright as CSF were classified as having normal intensity, and discs with a nucleus pulposus darker than CSF were classified as having decreased signal intensity. If the disc was classified as dark in the baseline study, it was classified as dark also in the follow-up study. To agree on the classification criteria, a pilot study was performed in which 40 MR images (20 subjects) were independently evaluated by both readers. Interobserver agreement was tested. The readers thereafter jointly discussed the evaluations to reach consensus on the classification criteria. All the images were then evaluated by both readers.

The interobserver agreement of visual assessments was calculated for all the MR images evaluated (n=151). The weighted κ coefficient ranged from 0.71 to 0.87 for different disc levels. The intraobserver agreement was calculated for the evaluations of the 40 images in the pilot study and the actual evaluations of the radiologist. The weighted kappa coefficients were 0.63–0.81 for different disc levels.


The subjects were sent a questionnaire inquiring their occupational exposure, history of back accidents, car driving, and smoking. The questionnaire was checked during the interview when they came to the MR imaging. The subjects’ current weight and height, as well as their weight at the age of 25 y were recorded. BMI (weight (kg) per height squared (m2)) was calculated based on self-reported height and weight at the age of 25 y and current weight. The history of overweight was grouped into three categories: no overweight (BMI <25 kg/m2 at the age of 25 and 40–45 y), persistent overweight (BMI ≥25 kg/m2 at the age of 25 and 40–45 y) and other (BMI ≥25 kg/m2 either at age 25 or at age 40–45 y). The job title was used as a measure of occupational load. The history of accidental back injuries was classified into two categories: one or more injuries before baseline, no injuries before baseline; the history of car driving into: >15 000 km/y before baseline, and less driving. Smoking status at baseline was classified into three categories: smoker, ex-smoker (quit smoking before baseline), and non-smoker (had never smoked). The description of history of overweight and the other risk factors is presented in Table 1.

Table 1 The description and prevalence of the potential risk factors (N=129)

Statistical analysis

The association of the MRI findings with the history of overweight and the potential confounders was analyzed by multivariate modeling. The dependent variable was the number of discs with decreased signal intensity (nucleus pulposus darker than adjacent CSF). History of overweight was used as an explanatory variable, and occupation, smoking status at baseline, history of car driving, and history of back accidents were included in the modeling as potential confounders.

The effect of overweight at different age on the progression of degenerative changes was also evaluated. For this analysis, the dependent variable was the 4-y changes in the number of degenerated discs (dark nucleus pulposus) in the lumbar spine. BMI at the age of 25 and 40–45 y was used as an explanatory variable, and occupation, smoking status at baseline, history of car driving, and history of back accidents were included in the modeling as potential confounders. BMI was dichotomized for Model I: BMI < 25 kg/m2 (no overweight) and BMI ≥25 kg/m2 (overweight) and was grouped into three categories for Model II: BMI <24 kg/m2, 24 kg/m2 ≤BMI <25 kg/m2, and BMI ≥25 kg/m2. BMI below 25 kg/m2 (Model I) and BMI below 24 kg/m2 (Model II) were used as the reference category.

Logistic regression analysis was used (SAS 8.2) to estimate the odds ratios (ORs) and risk ratio (RR) and their 95% confidence intervals (CIs). The number of discs with decreased signal intensity was handled as a count term (0, 1, 2) in the analysis.


Mean age of the subjects at baseline was 44 y (s.d. 2, range 41–46 y). The prevalence of discs with a decreased signal intensity of the nucleus pulposus was 63% at baseline and 73% at follow-up. In 1991, 38 subjects (29%) had degenerative changes at two or three disc levels, in 1995 a multilevel decrease of signal intensity was present in 53 (41%) subjects. There were no statistically significant differences in the number of discs with decreased signal intensity at baseline and follow-up between the occupations (Table 2).

Table 2 Association between occupation, overweight and decreased signal intensity of the nucleus pulposus

The means of subjects’ height, weight at the age of 25 y and at the age of 40–45 y were 178.1 cm (s.d. 5.8, range 167–192 cm), 73.2 kg (s.d. 9.1, range 54–103 kg) and 81.9 kg (s.d. 13.3, range 57–130 kg), respectively. The prevalence of overweight at the age of 25 y varied from 12 to 20.5% depending on occupational group, being highest for carpenters and lowest for office workers (Table 2). Overweight at the age of 40–45 y was significantly more often observed among carpenters and machine drivers as compared with office workers (68.2 and 64.9 vs 32.0%).

Persistent overweight (BMI ≥25 kg/m2 at the age of 25 and 40–45 y) was strongly associated with disc degeneration at follow-up (adjusted OR 4.3; 95% CI 1.3–14.3). The effects of other explanatory variables were not statistically significant (Table 3).

Table 3 Effect of overweight and other risk factors on the number of lumbar discs L2/L3–L4/L5 with decreased signal intensity of nucleus pulposus at baseline and follow-up

The risk of progression of degenerative changes (4-y changes in the number of discs with decreased signal intensity of the nucleus pulposus) was statistically significantly increased by overweight at young age (adjusted RR 3.8; 95% CI 1.4–10.4; Table 4) but not by high BMI at the age of 40–45 y (adjusted RR 1.3; 95% CI 0.7–2.7).

Table 4 Effect of body mass index (BMI) at the age of 25 and 40–45 y on the 4-y changes in the number of lumbar discs with decreased signal intensity of nucleus pulposus

We also tested the effect of BMI between 24 and 25 kg/m2 on developing decreased signal intensity (Model II). The risk ratio was 3.7 for men with BMI between 24 and 24.9 kg/m2 and 4.3 for subjects with BMI of 25 kg/m2 or more as compared with the reference group with BMI less than 24 kg/m2 (Table 4).


Overweight was found to be a significant risk factor of lumbar disc degeneration as indicated by a marked reduction of the signal intensity of the nucleus pulposus (darker than adjacent CFS). The effect was stronger for overweight at young age than for overweight at the age of 40–45 y.

Decreased signal intensity of the intervertebral discs is the most common sign of disc degeneration on MRI. A recent study by Luoma et al showed that the signal intensity of the nucleus pulposus seems to be a more feasible measure of early degeneration than absolute disc height.20 Height and weight of the subject and distance from the surface coil affect the signal intensity. Since height and weight of the subjects did not change much during follow-up and the positioning was the same, their effect on the change of signal intensity between studies was minimal. The technical quality of the images at follow-up was on the average lower than that at baseline. However, only images with good or satisfactory image quality were included in the study, and only discs L2/L3–L4/L5, located in the middle of the field of view were evaluated. CSF was used as an intensity reference in both studies. Therefore, small differences in the signal-to-noise ratio between baseline and follow-up should not cause substantial differences in classifying the signal intensity.

The World Health Organization defines overweight as a BMI of 25 kg/m2 or higher.21 In an extensive population study, Heliövaara22 found that among men, the risk of herniated lumbar disc requiring hospitalization increased even when BMI was more than 22 kg/m2, and the risk increased up to BMI 29.9 kg/m2 when the RR was 3.7 (95% CI 1.7–8.0). In our study, an increased risk of lumbar disc degeneration was found already when BMI at the age of 25 y was between 24 and 24.9 kg/m2.

The mechanism by which overweight causes lumbar disc degeneration is poorly understood. The contribution of both mechanical and systemic factors is likely. Overweight may increase the mechanical load of the spine, thereby increasing the risk of degeneration and back disorders. Epidemiological studies produce evidence that obesity increases the risk of atherosclerosis and cardiovascular disease. Poirier and Eckel23 proposed that obesity may affect atherosclerosis through dyslipidemia, hypertension, glucose intolerance, inflammatory markers, and the prothrombotic state. Spinal blood circulation has been shown to affect disc degeneration in a cadaver study by Kauppila,24 in which subjects with atherosclerosis of the spinal vessels had an increased risk for disc degeneration.

It is believed that intervertebral disc degeneration is the result of enzymatic breakdown of the extracellular matrix, and possibly local inflammation.25 Recently, Das26 proposed that obesity could be an inflammatory disorder. Overweight and obese children and adults have elevated serum levels of C-reactive protein, interleukin-6, tumor necrosis factor-α, and leptin, all known markers of inflammation and closely associated with cardiovascular risk factors.26, 27 This may explain the increased risk of diabetes, ischemic heart disease, and many other chronic diseases in the obese. It can be speculated that inflammation could be the common link between obesity and disc degeneration. This hypothesis warrants further studies.

The current study is the part of the project which aimed to examine the relationship of occupation with disc degeneration and low back pain.9, 28 A marginally elevated risk of decreased signal intensity of the nucleus pulposus was found among carpenters compared with office workers. In the present analyses, occupation was a potential confounder. Overweight was more prevalent among machine drivers and construction carpenters as compared with office workers. The educational level of the men was not uniform. The office workers were more highly educated than the machine drivers and carpenters. Lifestyle factors depend on education and social class and may affect the occurrence of disc degeneration. Therefore, it was necessary to control for occupation in the analyses. It is not likely that the association of overweight with decreased signal intensity observed was due to other risk factors. The subjects’ possible recall error in reporting their weight and height tends to underestimate BMI and thence also the effect on degeneration.29

In summary, this prospective MRI study provides evidence that BMI above 25 kg/m2 increases the risk of lumbar disc degeneration, with a stronger effect of high BMI at young age than in middle age. Owing to the small size of the study sample, the effect estimates had low precision and could have occurred by chance or could have been biased. The results need to be confirmed in larger studies in different settings.


  1. 1

    Vega GL . Results of expert meeting: obesity and cardiovascular disease. Obesity, the metabolic syndrome, and cardiovascular disease. Am Heart J 2001; 142: 1108–1116.

    CAS  Article  Google Scholar 

  2. 2

    Field AE, Coakley EH, Must A, Spadano JL, Laird N, Dietz WH, Rimm E, Colditz GA . Impact of overweight on the risk of developing common chronic diseases during a 10-year period. Arch Intern Med 2001; 161: 1581–1586.

    CAS  Article  Google Scholar 

  3. 3

    Melanson KJ, McInnis KJ, Rippe JM, Blackburn G, Wilson PF . Obesity and cardiovascular disease risk: research update. Cardiol Rev 2001; 9: 202–207.

    CAS  Article  Google Scholar 

  4. 4

    Sowers M . Epidemiology of risk factors for osteoarthritis: systemic factors. Curr Opin Rheumatol 2001; 13: 447–451.

    CAS  Article  Google Scholar 

  5. 5

    Hinton R, Moody RL, Davis AW, Thomas SF . Osteoarthritis: diagnosis and therapeutic considerations. Am Fam Physician 2002; 65: 841–848.

    Google Scholar 

  6. 6

    Kostova V, Koleva M . Back disorders (low back pain, cervicobrachial and lumbosacral radicular syndromes) and some related risk factors. J Neurol Sci 2001; 192: 17–25.

    CAS  Article  Google Scholar 

  7. 7

    Fanuele JC, Abdu WA, Hanscom B, Weinstein JN . Association between obesity and functional status in patients with spine disease. Spine 2002; 27: 306–312.

    Article  Google Scholar 

  8. 8

    Kopelman PG . Obesity as a medical problem. Nature 2000; 404: 635–643.

    CAS  Article  Google Scholar 

  9. 9

    Luoma K, Riihimaki H, Luukkonen R, Raininko R, Viikari-Juntura E, Lamminen A . Low back pain in relation to lumbar disc degeneration. Spine 2000; 25: 487–492.

    CAS  Article  Google Scholar 

  10. 10

    Borenstein DG . Epidemiology, etiology, diagnostic evaluation, and treatment of low back pain. Curr Opin Rheumatol 2001; 13: 128–134.

    CAS  Article  Google Scholar 

  11. 11

    Riihimäki H, Viikari-Juntura E . Back and limb disorders. In: McDonald C, Wheatley M (eds). Epidemiology of Work Related Diseases. BMJ Books: London; 2000. pp 233–243.

    Chapter  Google Scholar 

  12. 12

    Böstman OM . Prevalence of obesity among patients admitted for elective orthopaedic surgery. Int J Obes Relat Metab Disord 1994; 18: 709–713.

    Google Scholar 

  13. 13

    Heliovaara M, Knekt P, Aromaa A . Incidence and risk factors of herniated lumbar intervertebral disc or sciatica leading to hospitalization. J Chronic Dis 1987; 40: 251–258.

    CAS  Article  Google Scholar 

  14. 14

    Riihimäki H, Mattsson T, Zitting A, Wickström G, Hanninen K, Waris P . Radiographically detectable degenerative changes of the lumbar spine among concrete reinforcement workers and house painters. Spine 1990; 15: 114–119.

    Article  Google Scholar 

  15. 15

    Symmons DPM, van Hemert AM, Vandenbroucke JP, Valkenburg HA . A longitudinal study of back pain and radiological changes in the lumbar spine of middle aged women. I Clinical findings, II Radiographic findings. Ann Rheum Dis 1991; 50: 158–166.

    CAS  Article  Google Scholar 

  16. 16

    Parkkola R, Rytökoski U, Kormano M . Magnetic resonance imaging of the discs and trunk muscles patients with chronic low back pain and healthy control subjects. Spine 1993; 18: 830–836.

    CAS  Article  Google Scholar 

  17. 17

    Erkintalo MO, Salminen JJ, Alanen AM, Paajanen HE, Kormano MJ . Development of degenerative changes in the lumbar intervertebral disk: results of a prospective MR imaging study in adolescents with and without low-back pain. Radiology 1995; 196: 529–533.

    CAS  Article  Google Scholar 

  18. 18

    Elfering A, Semmer N, Birkhofer D, Zanetti M, Hodler J, Boos N . Risk factors for lumbar disc degeneration: a 5-year prospective MRI study in asymptomatic individuals. Spine 2002; 27: 125–134.

    Article  Google Scholar 

  19. 19

    Riihimaki H, Viikari-Juntura E, Moneta G, Kuha J, Videman T, Tola S . Incidence of sciatic pain among men in machine operating, dynamic physical work, and sedentary work. Spine 1994; 19: 138–142.

    CAS  Article  Google Scholar 

  20. 20

    Luoma K, Vehmas T, Riihimaki H, Raininko R . Disc height and signal intensity of the nucleus pulposus on magnetic resonance imaging as indicators of lumbar disc degeneration. Spine 2001; 26: 680–686.

    CAS  Article  Google Scholar 

  21. 21

    World Health Organisation. Obesity: Preventing and Managing the Global Epidemic. World Health Organisation: Geneva; 1997.

  22. 22

    Heliövaara M . Body height, obesity and risk of herniated lumbar intervertebral disc. Spine 1987; 12: 469–472.

    Article  Google Scholar 

  23. 23

    Poirier P, Eckel RH . Obesity and cardiovascular disease. Curr Atheroscler Rep 2002; 4: 448–453.

    Article  Google Scholar 

  24. 24

    Kauppila LI . Ingrowth of blood vessels in disc degeneration. Angiographic and histological studies of cadaveric spines. J Bone Joint Surg Am 1995; 77: 26–31.

    CAS  Article  Google Scholar 

  25. 25

    Rannou F, Corvol MT, Hudry C, Anract P, Dumontier MF, Tsagris L, Revel M, Poiraudeau S . Sensitivity of anulus fibrosus cells to interleukin 1 beta. Comparison with articular chondrocytes. Spine 2000; 25: 17–23.

    CAS  Article  Google Scholar 

  26. 26

    Das UN . Is obesity an inflammatory condition? Nutrition 2001; 17: 953–966.

    CAS  Article  Google Scholar 

  27. 27

    Visser M, Bouter LM, McQuillan GM, Wener MH, Harris TB . Elevated C-reactive protein levels in overweight and obese adults. JAMA 1999; 282: 2131–2135.

    CAS  Article  Google Scholar 

  28. 28

    Luoma K, Riihimäki H, Raininko R, Luukkonen R, Lamminen A, Viikari-Juntura E . Lumbar disc degeneration in relation to occupation. Scand J Work Environ Health 1998; 24: 358–366.

    CAS  Article  Google Scholar 

  29. 29

    Palta M, Prineas RJ, Berman R, Hannan P . Comparison of self-reported and measured height and weight. Am J Epidemiol 1982; 115: 223–230.

    CAS  Article  Google Scholar 

Download references


This work was financially supported by the Finnish Work Environment Fund. All authors contributed to the development of the key ideas of the paper, and to the interpretation of the results.

Author information



Corresponding author

Correspondence to H Riihimäki.

Additional information

Competing interest:

None declared.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

Liuke, M., Solovieva, S., Lamminen, A. et al. Disc degeneration of the lumbar spine in relation to overweight. Int J Obes 29, 903–908 (2005).

Download citation


  • age
  • overweight
  • intervertebral disc degeneration
  • magnetic resonance imaging
  • prospective study

Further reading


Quick links